303 104 284MB
English Pages 800 [821] Year 2022
f
SPRINGER PUBLISHING ■*’5
\
i
H.
i
-●
t
■>
I
‘■k
/
■
>V-'
' ^>7 'Vf
'.●
y
i -te"' ●-● '4
11
,r
■7
%
a
5V .-
:●
/' s./'*
«L
'' I",4,
\
i«
●v< I
A
elopmental research. Orienta-Konsultit.
Engestrom, Y. (2001). Expansive learning at work; Toward an activity theoretical reconceptualization. journal of Education and Work, 14, 133-156. https: / /doi.org/10.1080/13639080123238 Englebardt, S. P, & Nelson, R. (2002). Healthcare infonnatics: An interdisciplinary approach. Gale, Cengage. Fawcett, J. (1984). Analysis and ez’aluation of conceptual models of nursing. F. A. Davis. Fernandez-Luque, L., & Imran, M, (2018). Humanitarian health computing using artificial intelligence and social media: A narrative literature review. International
Journal of Medici Informatics, 114, 136-
142. https://doi.Org/10.1016/j.ijmedinf.2018.01.015
Geels, F. W. (2007). Feelings of discontent and the promise of middle range theory for STS; Examples from technology dynamics. Science, Technology, & Human Values, 32(6), 627-651. http.s://doi .org /10.1177 / 0162243907303597
71
72
; INTRODUCTION
Goossen, W. (2000). Nursing informatics research. Nurse Researcher, 8{2), 42-54. https://pdfs .Semanticscholar.org/0927/732bab6a00d6aac43e7f6adl9dd73916f6ab.p df Graves, J. R., & Corcoran, S. (1989). The study of nursing informatics. Journal of Nursing Scholarship, 21 (4), 227-231. https: / / doi.org /10.1111 /j.l547-5069.1989.tb00148.x Greenhalgh, X, & Stones, R. (2010). Theorising big IT programmes in healthcare; Strong structuration theory meets Actor-Network Theory. Social Science & Medicine, 70(9), 1285-1294. https://doi .org/10.1016/j.socscimed.2009.12.034 Grinspun, D., Virani, X, & Bajnok, I. (2001). Nursing best practice guidelines: The RNAO (Registered Nurses Association of Ontario) project. Hospital Quarterly, 5(2), 56-60. https://doi.org/10.12927/ hcq..16690 Guyatt, G., & Drummond, R. (2002). Users' guides to the medical literature: A manual for evidence-based
clinical practice (1st ed.). American Medical Association. Harrison, M. I., Koppel, R., & Bar-Lev, S. (2007). Unintended consequences of information technologies in health care: An interactive sociotechnical analysis. Journal of the American Medical Informatics
Association, 14(5), 542-549. https://doi.org/10.1197/jamia.M2384 Harrison, R. L., & Lyerla, R (2012). Using nursing clinical decision support systems to achieve meaningful use. CIN: Computers, Informatics, Nursing, 30(7), 380-385. https://doi.org/10.1097/ NCN.0b013e31823eb813
Hasu, M. (2000). Constructing clinical use: An activity-theoreti cal perspective on implementing new technology. Technology Analysis & Strategic Management, 12(3), 369-382. https://doi
.org/10.1080/09537320050130606 Holden, R. J., & Karsh, B. (2010). The technology acceptance model: Its past and its future in health care. Journal of Biomedical Informatics, 43(1), 159-172. https://doi.Org/10.1016/j.jbi.2009.07.002 Holden, R. J., Brown, R. L., Scanlon, M. C., & Karsh, B.-T. (2012). Modeling nurses' acceptance of bar coded medication administration technology at a pediatric hospital. Journal of the American Medical Injvrmatics Association, 19(6), 1050-1058. https://doi.org/10.11 36/amiajnl-2011-000754 Hook, J., Pearlstein,J., Samarth,A., & Cusack, C. (2008). Using barcode medication administration to improve quality and safety. https://digital.ahrq.gov/sites/default/files /docs/page/09-0023-EF_bcma_0.pdf Hurtado, M. R, Swift, E. K., & Corrigan, J. M. (2001). Crossing the quality chasm: A new health system for the
21st centu^ (No. 1OM2001). Institute of Medicine.
Kannampallil, X G., Schauer, G. F., Cohen, X, & Patel, V. L. (2011). Considering complexity in healthcare systems. Journal of Biomedical Informatics, 44(6), 943-947. https: / Zdoi.org/ 10.1016/j.jbi.2011.06.006 Koppel, R., Metlay, J. R, Cohen, A., Abaluck, B., Localio, R. A., Kimmel, S. E., & Strom, B. L. (2005).
Role of computerized physician order entry systems in facilitating medication errors. Journal of the American Medical Association, 293(10), 1197-1203. https://doi.or g/10.1001/jama.293.10.1197
Koppel, R., Wettemeck, X, Telles, J. L., & Karsh, B.-T. (2008). Workarounds to barcode medication administration systems: Their occurrences, causes, and threats to patient safety. Journal of the American
Medical Informatics Association, 15(4), 408-423. https://doi.org /10.1197/jamia.M2616 Korpelainen, E., & Kira, M. (2013). Systems approad\ for analysing problems in IT system adoption at work. Behaviour & Information Technology, 32(3), 247-262. https://doi.org/10.1080/0144929X.2011.624638 Langley, G. J., Moen, R., Nolan, K. M., Nolan, X W., Norman, L., & Provost, L. .P (2009). The improvement guide: A practical approach to enhancing organizational performance (2nd ed.). Jossey-Bass. Latour, B. (1991). Xechnology is society made durable. In J. Law (Ed.), A sociology of monsters: Essays on power, technology and do7nination (pp. 103-131). Routledge. Latour, B. (2005). Reassembling the social: An introduction to Actor-Network-Theory. Oxford. Lau, A. Y, & Staccini, .P (2019). Artificial intelligence in health; New opportunities, challenges, and practical implications: Findings from the Yearbook 2019 section on education and consumer health informatics. Yearbook of medical informatics, 28(1), 174-178. https:/ /doi.org/10.1055/s-0039-1677935 Law, J., & Hassard, J. (Eds.). (1999). Actor Nehvork Theory and after. Blackwell.
Lewin, R. (1992). Complexity: Life at the edge of chaos. Maxwell Macmillan Canada. Lomov, B. (1981). The problem of activity in psychology. Psikhologicheskii Zhurnal, 2(5), 3-22. Lorenz, E. N. (1972, December). Predictability: Does the flap of a butterfly's 7vings in Brazil set off a tornado in Texas? Address presented at the meeting of the American Association for the Advancement of Science, Boston, MA.
Matney, S., Brewster, P. ]., Sward, K. A., Cloyes, K. G., & Staggers, N. (2011). Philosophical approaches
to the nursing informatics data-information-knowledge-wisdom framework. Adi’ances in Nursing
Science, 34(1), 6-18. https: / /doi.org/10.1097/ANS.Ob013e318207 1813 Matney, S. A., Avant, K., Clark, L., & Staggers, N. (2020). Development of a theory of Wisdom-inAction for clinical nursing. Advances in Nursing Science, 43(1), 28-41. https://doi.orgho.1097/ ANS.0000000000000304
Matney, S. A., Avant, K., & Staggers, N. (2015). Toward an understanding of wisdom in nursing. Online Journal of Issues in Nursing, 21(1). https://doi.org/10.3912/OJI N.Vol21No01PPT02 Matney, S. A., Staggers, N., & Clark, L. (2016). Nurses' wisdom in action in the emergency department. Global Qualitative Nursing Research, 3. https:/ /doi.org/10.1177 /2333393616650081
3: SCIENTIFIC AND THEORETICAL FOUNDATIONS FOR IMPROVING HEALTHCARE
Meeks, D. W., Takian, A., Sittig, D. E, Singh, H., & Barber, N. (2014). Exploring the sociotechnical intersection of patient safety and electronic health record implementation. Journal of the American Medical lnfi>rmatics Association, 22(el), e28-e34. https: / /doi.org/10.1136/amiajnl-2013-001762 Murphy, J. I. (2013), Using plan-do-study-act to transform a simulation center. Clinical Simulation in Nursing, 9{7),e257-e264. https://doi.Org/10.1016/j.ecns.2012.03. 002 Nardi, B. (1996). Activity theory and human-computer interaction. In B. Nardi (Ed.), Context and consciousness: Activity theory and human-computer interaction (pp. 7-16). MIT Press. Nelson, R, (2002). Major theories supporting health care informatics. In S. Englebardt & R. Nelson (Eds.), Health care informatics: An interdisciplinary approach. Mosby. Nelson, R. (2018). Informatics: Evolution of the Nelson data, information, knowledge and wisdom model; Part 1. Online journal of Issues in Nursing, 23(3). https://doi.org/10.3912/OJIN Novak, L. L., Holden, R. ],, Anders, S. H., Hong, J. Y., & Karsh, B. T. (2013). Using a sociotechnical framework to understand adaptations in health IT implementation. International journal of Medical Informatics, 82(12), e331-e344. https://doi.org/10.1016/j.ijmedi nf.2013.01.009 Pabst, M. K„ Scherubel, J. C., & Minnick, A. R {1996). The impact of computerized documentation on nurses' use of time. Computers in Nursing, 24(1), 25-30. Paley, J. (2007). Complex adaptive systems and nursing. Nursing Inquiry, 24(3), 233-242. https:/ /doi .org/10.1111 /j.l440-1800,2007.00359.x Pepito, J. A„ & Locsin, R. (2019). Can nurses remain relevant in a technologically advanced future? International journal of Nursing Sciences, 6(1), 106-110. https://doi.org/10.1016/j.ijn.ss.2018.09.013 Pinch, T. J., & Bijker, W. E. (1984). The social construction of facts and artefacts; Or how the sociology of science and the sociology of technology might benefit each other. Social Studies of Science, 14,399^41. https://doi.Org/10.1177/030631284014003004 Plsek, P. E., & Greenhalgh, T. (2001). The challenge of complexity in health care. British Medical journal, 323(7313), 625-628. https://doi.org/10.1136/bmj.323.7313.625 Rahimi, B., Tlmpka, T, Vimarlund, V., Uppugunduri, S., & Svensson, M. (2009). Organization-wide adoption of computerized provider order entry systems: A study based on diffusion of innovations theory. BioMedicalCentral: Medical Informatics and Decision Making, 9(52). https: / /doi.org/10.1186/1472-6947-9-52 Ronquillo, C., Currie, L. M., & Rodney, P. (2016). The evolution of data-information-knowledgewisdom in nursing informatics. Advances in Nursing Science, 39(1), E1-E18. https://doi.org/10.1097/ ANS.0000000000000107
Samuels-Dennis, J., & Cameron, C. (2013), Theoretical framework. In G. LoBiondo-Wood, J. Haber, C.
Cameron, & M. Singh (Eds.), Nursing research in Canada: Methods, critical appraisal, and utilization (pp. 27-47). Elsevier.
Sittig, D. E, & Singh, H. (2010). A new sociotechnical model for studying health information technology in complex adaptive healthcare systems. Quality and Safety in Health Care, 29(Suppl. 3), i68-i74. https; / / doi.org/10.1136/qshc.2010.042085 Sleutel, M., & Guinn, M. (1999). As good as it gets? Going online with a clinical information system. Computers in Nursing, 17(4), 181-185. Staggers, N., Gassert, C, A., & Curran, C. (2001). Informatics competencies for nurses at four levels of practice, journal of Nursing Education, 40(7), 303-316. http:/ /tigercompetencies.pbworks.com/f/ Informatics+comptetencies+four+levels+of-H practice.pdf Staggers, N., & Parks, .P L. (1993). Description and initial applications of the Staggers & Parks NurseComputer Interaction Framework. Computers in Nursing, 22(6), 282-290. Sugimori, Y, Kusunoki, K., Cho, E, & Uchikawa, S. (1977). Toyota production system and Kanban system: Materialization of just-in-time and respect-for-human system. International journal of Production Research, 25(6), 553-564. https:/ /doi.org/10.1080/00 207547708943149 Thompson, C., Snyder-Halpern, R., & Staggers, N, (1999). Analysis, processes, and techniques; Case study. Computers in Nursing, 27(3), 203-206. Timmons, S. (2003). Nurses resisting information technology. Nursing Inquiry, 20(4), 257-269. https:/ / doi.org/10.1046/j.l440-1800.2003.00177.x Trist, E. (1981). The ewhition ofsocio-technical systems: A conceptual frameioork and an action research program
(Occasional paper 2). Ontario Ministry of Labour/Ontario Quality of Working Life Centre. Turley, J. .P (1996). Toward a model for nursing informatics. Image, 28(4), 309-313. https://doi .org/10.1111/j.l547-5069.1996.tb00379.x
van Onzenoort, H. A., van ee Plas, A., Kessels, A. G., Veldhorst-Janssen, N. M., van der Kuy, .P-H.
M., & Neef, C. (2008). Factors influencing bar-code verification by nurses during medication administration in a Dutch hospital. American Journal of Health-System Pharmacy, 65(7), 644-648.
https: / /doi.org/10.2146/ajhp070368 Varpio, L., Hall, P., Lingard, L., & Schryer, C. F. (2008), Interprofessional communication and medical error: A reframing of research questions and approaches. Academic Medicine: journal of the Association of American Medical Colleges, 83(10 Suppl.), S76-S81. https: / / doi.org/10.1097/ACM.0b013e318183e67b
73
74
I: INTRODUCTION
Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. (2003). User acceptance of information
technology: Toward a unified view. Management Information Systems Quarterly, 27(3), 425-478. https: / / doi.org /10.2307/30036540 Wilson, M. (2002). Making nursing visible? Gender, technology and the care plan as script. Information Technology & People, 15(2), 139-158. https://doi.org/10.1108/095 93840210430570 Wilson, T., & Holt, T. (1996). Complexity and clinical complexity. British Medical Journal, 323(7314), 685688. https://doi.org/10.1136/bmj.323.7314.685 Woolgar, S. (1991). The turn to technology in social studies of science. Science, Technology, & Human Values, 16{l), 20-50. https://doi.org/10.1177/016224399101600102 Zhang, H., Cocosila, M., & Archer, N. (2010). Factors of adoption of mobile information technology by homecare nurses: A technology acceptance model 2 approach. CIN: Computers, Informatics, Nursing, 28(1), 49-56. https://doi.Org/10.1097/NCN.0b013e3181c0474a
National Healthoe/eTrahsfQrmatibn^ and InformationTechnology LIZ JOHNSON, SUSAN MCBRIDE, DAVID BERGMAN, AND MARI TIETZE
\
%
OBJECTIVES ●
Review historical and recent efforts to expand the access and use of health information
technology (HIT) in the United States. Discuss how various programs are designed to be layered to support one another, particularly about HIT adoption and use. ● Describe the elements of law, policy, and regulation creation and the potential advocacy role of an advanced practice nurse (APN). ● Describe the important roles that APNs play in interprofession al teams within national HIT ●
initiatives.
●
Review best practices used in the nation to examine case studies on how organizations can successfullyimplementstrategies to fully realize the national aims. CONTENTS
INTRODUCTION
76
RAPIDLY EXPANDED HIT INFRASTRUCTURETO PROMOTE HEALTHCARE REFORM Health Information Exchange
80
Regional Extension Center Programs
81
Remaining HITECH Act Programs Under the ONC Workforce Development
82
86
FROM BUILDINGTO USINGTHE IT INFRASTRUCTURE
86
LINKTOTHE ACA, PROMOTING INTEROPERABILITY AND 21^' CENTURY CURES ACT UNDERSTANDING FEDERAL LAW MAKING ANDTHE ROLE OF EXPERT ADVOCACY SUMMARY
93
EXERCISES AND QUESTIONS FOR CONSIDERATION REFERENCES
94
77
94
88 91
76
: INTRODUCTION
INTRODUCTION
Electronic health records (EHRs) have been around in some form since the 1970s, but prior to the passage of the Health Information Technology for Economic and Clinical
Health (HITECH) Act in 2009, their penetration rate had been relatively modest. As late as 2009, fewer than 15% of nonfederal acute care hospitals had implemented even basic EHRs (Charles et al., 2014; Jha et al., 2009). In the past 10 years, the nationwide EHR penetration rate improved substantially. By 2017,80% of providers and 96% of hospitals had adopted a federally certified EHR system. In particular, the EHR penetration rate has markedly increased since 2010, when the implementation of the HITECH Act began (Office of the National Coordinator for Health Information Technology [ONC], 2016). The HITECH Act funded $35 billion by paying incentives for providers using certified EHRs to meet specific criteria described as Meaningful Use (MU). It is critical to note that these incentives were paid to healthcare providers, including hospitals, physicians, and critical access facilities, omitting the variety of long-term providers. Although this incentive method led to impressive implementation rates, the.se improvements in EHR adoption present challenges with optimizing certified technology within institutions and across regions, states, and the nation. These challenges are intensified due to the omission of incentives for the long-term care providers who often complete tlie continuum of care either before or post-acute care discharge. The adoption of EHR technology is only one step toward the effective use of health information technology (HIT), in which APNs will play a sigiiificant role. Given this massive update in HIT infrastmcture across the United States, now the goal is to fully implement our national healthcare strategy. There are several important components to this strategy, including goals to; (a) advance the accessibility, interoperability, and usability of electronic health information and EHRs; (b) support seaire, standards-based application programming interface (APIs) for the healthcare consumer; and (3) improve the HIT infrastructure to combat opioid epidemic and other substance use disorders (Department of Health and Human Services [DHHS], 2021). The HITECH Act was a critical, landscape-changing event in adopting and using HIT around the country. The programs created through the HITECH Act were each designed to address discrete challenges related to EHR adoption and use. When considered as a
whole, these various efforts were not merely a hodgepodge of interventions but also a sophisticated, coordinated effort designed to address major shortcomings in the HIT ecosystem. Some interventions targeted adoption challenges, others targeted workforce needs, and still others were intended to bolster exchange capabilities. By addressing each of the.se major shortcomings, the HITECH Act's primary goal was to establish programs to improve healthcare quality, safety, and efficiency by promoting HIT, including EHRs and private and secure electronic health information exchange (ONC, 2021a).
While HIT expansion and adoption have been successful (Jha, 2013), EHR expansion
for the sake of meeting federal regulations (MU Reporting requirements) was never the goal. The EHR expansion and subsequent adoption are the critical foundation for building an infrastmcture that will support the universal exchange of data and
the subsequent data analysis to understand the relationship between treatments and associated outcomes. It will become possible to modify and improve current therapeutics toward better outcomes. This sets the stage for the Patient Protection and Affordable Care Act (ACA) goals along with the 21st Century Cures Act. Nonetheless, there are still major challenges around the HIT infrastmcture itself. Interoperability standards for some aspects of care data are still under discussion. For example, technology platforms like EHRs and health information exchanges (HIEs) continue to lack plug-and-play functionality that would lower implementation costs and encourage data exchange. Many HIEs are based on limited datasets. More recent
4: NATIONAL HEALTHCARE TRANSFORMATION AND INFORMATION TECHNOLOGY
developments, including private exchange networks and consortia like the Common Well Health Alliance and Carequality, are harder to characterize as entirely positive or negative. Even so, with the continuing nationwide adoption of HIT, enough infrastructure exists in enough communities to begin driving more significant changes in care outcomes, APNs—particularly those trained in informatics—are well-positioned to support the realization of HIT to improve quality of care and lower costs. This chapter reviews the major programmatic components of the HITECH Act and, subsequently, the 21st Century Cures Act which offers a unique opportunity for the APN to engage in advocacy efforts. The HITECH Act and the 21st Century Cures Act (Cures Act) provide for the foundational components of a modern HIT ecosystem. A significant focus also includes how policies are aligned with federal mandates and how they drive substantial changes in the coordination and delivery of healthcare and healthcare data. Additionally, this chapter describes ways that the HIT ecosystem is and can be leveraged to support quality outcomes with the explicit engagement of APNs. Finally, the chapter concludes by describing the role of APNs in advocating for policies and laws that allow for easier, timely, and complete patient health data exchange. For an APN, the goal is to leverage new laws and policies aimed at updated care practices for improved outcomes at the point of care and ensuring patients are engaged in their care. The federal effort to promote HIT began in 2004 when President George W. Bush created the ONC by executive order. Until the passage of the HITECH Act in 2009, the ONC served primarily as a convener and sought to build consensus on the development of data and technical standards that the HIT sector could deploy. When created, the goal of the ONC was to promote full adoption of EHRs for the entire country by 2014. The task was substantial. In 2004, just 20% of office-based physicians were using even a mdimentary EHR system, and fewer than 10% of hospitals were using even computerized provider order entry (CPOE; Ash & Bates, 2005). By 2009, some progress had been made regarding adoption of EHRs. At that time, approximately 48% of primary care practices were using some form of an EHR, but fewer than half of those used an EHR with functions that today would be considered fundamental (Hsiao & Hing, 2014). Furthermore, just 12% of hospitals reported using EHR systems (Charles et al., 2014). If the ONC was going to help realize the promise of HIT, something would need to change. Amid the economic turmoil that occurred at the end of 2008 and into 2009, Congress
passed and President Barack Obama signed into law the American Reinvestment and Recovery Act, or ARRA. Approximately 50 of the 400 pages comprising ARRA were dedicated to a separately named act, called the HITECH Act. As early as March 2009, the HITECH Act was described as reflecting the conviction that "electronic information systems are essential to improving the health and healthcare of American.s" (Blumenthal, 2009, p. 1477). In this way, the HITECH Act was strategically linked to efforts that reached fruition a
year later in March 2010 with the passage of the ACA. As shown later in this chapter, the two laws are inextricably linked in critical operational ways. Shortly after Blumenthal became the third national coordinator for HIT, the HITECH Act was established as a "substantial
down payment on the financial and human resources needed to wire the U.S. healthcare system" (Merrill, 2009, p. 1)—activities that would prove crucial to healthcare reform. RAPIDLY EXPANDED HIT INFRASTRUCTURE TO PROMOTE HEALTHCARE REFORM
The HITECH Act included funding provisions both for mandatory spending through the Centers for Medicare & Medicaid Services (CMS) and programmatic spending for HIT
77
78
:
INTRODUCTION
adoption support through the ONC for a total of six major initiatives. The mandatory spending initiative was the EHR Incentive Program, administered by the CMS, whereas the programmatic spending largely went through the ONC with billions of dollars allocated through federal budgetary processes to support these programs. Individually, each of these initiatives were designed to address different ways of supporting widespread adoption and use of HIT. Additionally, these different initiatives were intended to build on and support one another.
The HITECH Act created a mandatory spending program—the EHR Incentive Program—through the CMS. Based on the Centers for Disease Control and Prevention (CDC) estimates, the EHR incentive program was expected to result in direct outlays of $26.8 billion through 2014. The HITECH Act also provided the statutory authority to create key programs, administered by the ONC, to support HIT expansion. These programs included the Regional Extension Center (REC) program, the state HIE program, and the Beacon program, among others. The HITECH Act included several federally funded programs which are listed in the following:
■ The EHR incentive program: A 5-year program of increasing complexity to encourage providers and hospitals to adopt and meaningfully use EHRs. This program has been renamed the Promoting Interoperability Program. The focus continues to encourage the use of EHR technologies with an increasing focus on the exchange of data. ■ The EHR certification program: A national standard of functionalities that providers and hospitals could reference to ensure that their EHR could support MU. In 2021, the use of certified EHRs is still required to avoid Medicare payment impacts. For calendar year 2021, in order to be considered a meaningful user and avoid a downward payment adjustment, eligible hospitals and Critical Access Hospitals (CAHs) may use (a) existing 2015 Edition certification criteria, b) the 2015 Edition Cures Update criteria, or (c) a combination of the two in order to meet the certified EHR technology (CEHRT) definition, as finalized in the calendar year 2021 Physician Fee Schedule (PFS) final rule (85 FR 84818 through 84828; CMS, 2021). ■ The state HIE program: A program for states to build a nationwide technological infrastructure that supports the .secure exchange of clinical content between relevant care providers. In 2020, the DHHS continued to award state HIEs for clear missions for improving interoperability.
■ REC program: A program to provide technical assistance to primary care providers in small and safety-net practices to facilitate selection, adoption, and use of EHRs; the REC program was explicitly intended to help providers qualify for the EHR incentive program. ■ Beacon commimity program: Beacon grants went to communities already relatively advanced in adoption and use of EHRs. These large grants of approximately $17 million were designed to help communities more explicitly connect the use of HIT—EHRs, HIEs, and other emerging forms of technology like short message service (SMS) messaging— to improvements in community health outcomes as demonstrated by standardized quality mea.sures (ONC, 2009). The Beacon program has continued to grant monies for a variety of incentives using technologies to advance the quality of care.
■ Workforce development: This included community-college curriculum and universitybased training. As of survey results in 2019, these programs continue to produce certified individuals.
4: NATIONAL HEALTHCARE TRANSFORMATION AND INFORMATION TECHNOLOGY
■ Research development was the Strategic Health IT Advanced Research Projects (SHARP): SHARP were awarded to four university centers to spur technological
innovation regarding the development of EHR technology. Areas of research included:
● Security and technology
● Usability and alignment of technology to physician cognition and decision making ● EHR information architecture
● Integration and utilization of EHR data for quality-improvemen t purposes Ultimately, the ONC was allocated just over $2 billion for programs (HealthIT Dashboard, 2013) and the EHR incentive program paid out over $37 billion in EHR incentives.
As these programs developed, it also became clear that there were crucial ways of leveraging this burgeoning infrastmcture to achieve additional health-related goals and objectives. Program initiatives, such as accountable care organizations (ACOs) embedded in the ACA, required more robust technological infrastaicture and subsequent policy documents to develop their potential fully. One key policy docLiment was the National Quality Strategy (NQS), which laid out plans to link these separate initiatives in ways that were explicitly tied to improvements in community health, quality of care, and reduced cost.
Only one of these programs—the EHR incentive program—was not housed entirely within the ONC. Even so, the ONC's policy committee, a group of stakeholders meeting under the Federal Advisory Committee Act (FACA), was charged with developing the initial framework and recommendations defining what it meant to achieve MU of an EHR. This recommendation was officially submitted to the CMS, where it became the basis for formal rules operationalizing the definition of MU for the EHR incentive
program. Among the concepts realized through further definition were eligibility criteria, incentive amounts, program duration, and the formal MU attestation process, in addition to the MU criteria.
In early 2009, perhaps the biggest challenge faced by the HITECH Act and all its nascent programs was that, until then, the ONC had been a policy-focused organization with little in the way of operational capacity to administer formal grant programs. In 2008, for example, the ONC had an annual budget of $60.5 million (Agency for Healthcare Research and Quality [AHRQ], 2010). With the passage of the HITECH Act, suddenly, the
ONC was responsible for administering grant programs totaling $2 billion, more than a 33-fold increase to establish significant HIT infrastructure. In 2020, the budget was $60 million and was focused on continuing efforts to create interoperability of health data
sources for patients and providers (DHHS, 2021). In 2015 over four in five of all non-federal acute
care hospitals had adopted a basic
EHR with clinician notes and 96% of non-federal acute care hospitals have possession of
EHR certified by the DHHS. This percentage has held through 2020. Although the CMS broadly administered the EHR incentive program, it was also partially implemented through state Medicaid offices. “Eligible providers"—healthcare practitioners in various categories—qualified for these funds by meeting a billing threshold for Medicare or Medicaid claims. In addition, nurse practitioners (NPs) were eligible for Medicaid incentives, and under specific circumstances, physician a.ssistants (PAs) could also qualify. Hospitals, meanwhile, were eligible to qualify for both Medicare an
79
80
I: INTRODUCTION
and Medicaid through a complex formula that includes Medicare and Medicaid billings and total inpatient days, among other elements. The impact of the expenditures dramatically impacted the implementation of both EHRs and the supporting infrastructure. In 2018, the original MU program was renamed Promoting Interoperability (ONC, 2019b). Promoting Interoperability was an intentional move to align the programs beyond the existing requirements of MU to a new phase of EHR reporting measures with an increased focus on interoperability and improving patient access to health information. In 2021, hospitals and providers continued to be required to report on measures. The incentives have been awarded and now reporting is aligned with avoidance of penalties for failure to comply. Later in this chapter, the relationship between Promoting Interoperability and the 21st Cures Act will be discussed.
Health Information Exchange In addition to creating the EHR incentive program, other programs, were administered directly by the ONC. The two programs with the broadest impact were the REC program and the state HIE program. Figure 4.1 reflects the five ONC programs in addition to the EHR incentive program that the CMS administers. These programs can be compared metaphorically to laying the road for success as noted in Figure 4.1. For example, the HIE is "the road," and the EHR incentive program is the financing for "the car." These programs were strategically designed to build off of and support one another. To create a nationwide network of HIEs, the ONC
administered an application-only program to states or their designated entities to build HIE capabilities in their states. When the HITECH Act was passed in 2009,
there were a handful of exchanges in operation around the country, each operating with different levels of success. Some were specific to a single health system, whereas others were broadly supported in a community.
States were required to develop and submit a formal application to qualify for the funds. As initially designed, the ONC allocated $564 million for the state HIE program,
which was distributed to states based on covered lives. As a result, smaller states received
much smaller amounts than larger states. At the same time, states had a wide degree of latitude to implement HIE models that could leverage local infrastmcture and address local needs. The ONC's primary concern was developing a functional, sustainable infrastaicture that was capable of supporting MU. As originally conceived, the ONC intended to create the capability for all local HIEs to roll up to a nationwide HIE capable of referencing local clinical content. However, early in 2011, this goal was delayed in favor of the use of what became known as the Nationwide Health Information Network Direct (NwHIN Direct), or simply Direct. The Direct protocol functions much like a secure email system that enables clinical content
to be transmitted electronically in a way that is also consistent with privacy obligations enforced under the Health Information Portability and Privacy Act (HIPAA). Toencourage
the development of the Direct capability, the ONC required states to support the Direct protocol. Additionally, the ONC required EHRs to support the Direct functionality to
become certified.
As of 2020, Direct was available around the country but had not been widely adopted as an approach to facilitate the exchange of health data. It is primarily used for the exchange of data at admission and discharge. Although the one-time exchange of data is important for care coordination purposes, it does not support the ability of care providers to more effectively monitor and improve quality aligned with value-based
4: NATIONAL HEALTHCARE TRANSFORMATION AND INFORMATION TECHNOLOGY
81
FIGURE 4.1 Health InformationTechnology for Economic and Clinical Health (HITECH) Act Initiatives.
Spj
in
■ newer ■
● SHARP grants
● Metaphor: Designing a new engine, aerodynamic research
better tools Build workforce with technical skills to maintain the HIT Infrastructure
● Community college- and university-based training programs
● Metaphor: Automobile mechanics, road maintenance
/ / /
Demonstrate how using HIT throughout a community can drive quality improvement
● Beacon programs ● Metaphor: Learning how to navigate
/
Provide technical support to Uiose who use HIT in clinical settings
Provide financial incentives for purchasing tools that use the Hrr
Build the basic infrastructure to support the exchange of clinical information
● REC programs
● Metaphor: Driving instructions
● EHR incentive program
● Metaphor: Financing to buy a car
● Health Information Exchanges
● Metaphor: The road
Note: The HITECH Act initiatives were explicitly designed to build off and support one another.They were designed to provide foundational support for health information technology (HIT) and develop new and innovative approaches to guide long-term development. EHR, electronic health record: REC, regional extension center; SHARR Strategic Health IT Advanced Research Projects,
contracting. In September 2020, the DHHS funded new HIE activities in support of data exchange in public health emergencies. The COVID pandemic was recognized as a public emergency and there was a critical need to support the exchange of public health data. ONC granted five states a total of $2.5 million from the Coronavirus Aid, Relief, and Economic Seairity Acts (CARES Act) signed by President Trump. The five HIEs included Georgia, Arizona, Pennsylvania, Kansas, and Texas, who each have a 2-year agreement (ONC, 2021c). Regional Extension Center Programs If the HIEs and EHRs were the technological backbone of the HITECH Act, the REC
program created the "boots on the ground" to support EHR adoption. Loosely based on the extension programs in agriculture and manufacturing, the REC program was intended to support small physician practices that adopt and use EHRs. Just a few m onths after HITECH was passed in February 2009, the ONC released the initial FOA for the REC program, allocating $774 million to create REC programs around the country.
82
:
INTRODUCTION
As initially conceived, the REC program had a collective goal of supporting 100,000 primary care providers around the country in achieving stage 1 of MU of EHRs. By September 2010, every area of the United States, including Puerto Rico and Guam, was covered by one of the 62 RECs, each of which was responsible for supporting a defined number of primary care providers. The specific target number was primarily based on the total estimated number of primary care providers in the REC's region, with the minimum being 1,000 priority primary care providers (PPCPs). Although a few RECs were targeting as many as 8,000 PPCPs, the average state's PPCP target size was approximately 1,350. There were 62 REC programs around the country, covering every state and territory. Some populous states—California, Florida, New York, and Texas— were served by multiple RECs. Less populous states (Maine and North Dakota, Montana, and Wyoming) were sometimes combined and served by a single REC.
For each eligible provider, a REC was allocated $5,000 to support each eligible provider and up to $'750,000 to support building their operational infrastructure. For a REC with a target of 1,000 PPCPs and 2-year core support of $750,000, this meant a budget of $6.5 million ($1.5 million for core support, $5 million for Direct). The REC programs began with significant differences in size, scope, and initial challenges. Significantly, about 80% of REC programs were projects of three major types of parent organizations: quality-improvement organizations (QlOs), universities, and health center controlled networks (HCCNs). The remaining 20% of organizations with REC programs represented a wide range of parent organizations. These differences, in turn, meant that REC programs started with different levels of native expertise and with various levels of operational capacity. Despite some early challenges, the REC program was a striking success. By the end of January 2016, RECs had enrolled nearly 140,000 providers, had supported more than 136,000 in achieving go-live status, and supported 112,804 providers in attesting to MU (ONC, 2017).
In the long term, the RECs were expected to build a sustainable financial model to continue operations in the post-HlTECH funding world. However, in reality, as ACArelated funding wound down in 2014 and later, some REC programs were eventually spun-off as independent organizations, sold, or ceased operations. Many, however, began to seek new, emerging sources of funding. For example, around the time that REC funding was drying up, the Centers for Medicare and Medicaid Innovation (CMMI)
created the Transforming Clinical Practice Initiative (TCPl). According to the CMS (n.d.), the TCIP program was intended “to support more than 140,000 clinician practices... in
sharing, adapting and further developing their comprehensive quality improvement strategies" (para. 1). These practice transformation efforts were also aligned with the emerging payment models that were expected to reinforce the need for strong HIT systems.
Remaining HITECH Act Programs Under the OMC Beacon Programs Leading the Way
If the REC program was designed to help small clinician practices with the basic adoption and use of EHRs, the Beacon program was designed to demonstrate the kinds of clinical quality improvements that are possible in communities with more robust EHR adopti on. In early 2010, the ONC made available $250 million for the Beacon community program, a cooperative agreement program for communities with at least 30% EHR adoption among ambulatory care providers.
4: NATIONAL HEALTHCARE TRANSFORMATION AND INFORMATION TECHNOLOGY
Ultimately, 17 communities were designated Beacon communities and were awarded between $12 million and $16 million each. Communities were charged with a three-part aim:
1. Improve population health 2. Test innovative approaches 3. Build HIT infrastructure and capacity to exchange clinical information (Rein, 2012) Functionally, over the 3 years of the cooperative agreement, Beacon communities were put in the position of demonstrating how HIT can be used to drive measurable improvements in health. Each community was required to identify a set of validated clinical quality measures against which they would evaluate their performance. Some chose measures related to vaccination rates, whereas others looked at diabetes
management. The programs were given substantial latitude to identify germane measures to their community and use the federal dollars to address barriers or technological innovations that would enable the achievement of goals. Although there were some similarities among communities, as a general rule, each Beacon community was a unique program with unique goals, addressing challenges that were specific to each community (ONC, 2009). Research andTechnology Development SHARP Grants
In addition to programs building the HIT infrastructure through HIEs and EHR implementations, the ONC also funded a grant program that supported innovations to advance existing technology. The SHARP grants were awarded to universities or research institutions and were designed to support innovative research that would address critical areas of EHR functionality. As previously noted, SHARP is an acronym for Strategic Health IT Advanced Research Projects (ONC, 2012). Under this program, four grants were awarded to address target issues: privacy and security, physician cognition and decision-making, health application design, and use of EHR data. There were four target areas for the SHARP grants, and each awardee was responsible
for a single area of innovation and research. Unlike other ONC grants and cooperative agreements, the SHARP grants were not expected to impact HIT deployment directly. Instead, they informed the broader milieu of HIT, and their impact was felt more obliquely in the application of their conceptual findings. These grants have directly and indirectly produced scores of academic papers and presentations on topics that are both narrow and broad.
SHARP'S PRIVACY AND SECURITY The University of Illinois at Urbana—Champaign (UIUC) was awarded the SHARP-S grant to advance the privacy and security associated with HIT. Specifically, the grant focused on privacy and security for EHRs, HIEs, and telemed icine. Additionally, through the course of the grant, UIUC began to focus on the connec tivity between health-related sensors and devices, particularly implantable devices like insulin pumps.
In addition to exploring the governance around accessing health data—for example, the process through which identity becomes authenticated for access purposes this SHARP-S grant explores some of the technological aspects of identity authentication. SHARP-C PHYSICIAN COGNITION. The University of Texas at Houston (UT Houston) was
awarded the SHARP-C grant to explore the relationship between the presentation of
83
84
; INTRODUCTION
information in an EHR user interface and the impact of that presentation on physician decision-making. This project defined four focus areas: ■ Work-centered design-of-care process improvements in HIT, which focus on EHR usability and workflows. This work resulted in a framework for EHR usability. That framework is called TURF, an acronym for Task, User, Representation, and Function. Through the work of Zand and Waifi, these four components were identified as a set of measures that could determine whether or not end users could use an
EHR to perform required tasks. Equally as important, the framework allows the measurement of redesign of the EHR and whether that redesign is an improvement. The University of Houston is moving this framework into the EHR usability lab. As an APN, use of this framework to support clinical end-users is critical. Ensuring the EHRs are designed by and for the clinicians is the first step to fully achieving interoperability, patient engagement, and improved outcomes (Zhang & Walji, 2011). (For further information on the model, see TURF application developed out of SHARP-C research in Chapter 10.)
■ Cognitive foundations for decision-making implications for decision support ■ Automated model-based clinical summarization of key patient data ■ Cognitive information design and visualization: enhancing accessibility and understanding of patient data (ONC, 2012) SMART APPLICATION. Harvard University was awarded a SHARP grant to develop and deploy a modular, interoperable health data infrastructure known as the SMART platform. The SMART platform—substitutable medical apps and reusable technolo
gy—represents an effort to apply a smartphone "app" store functionality to EHRs. As currently designed, it is primarily a tool used to view data contained in another EHR,
with limited capability to write information into a patient record; however, this is in tended to evolve. The SMART tool sits conceptually on top of the EHR, where it reads and presents patient record data in ways that are significant to the provider. As a result of this grant, today the SMART platform is composed of open standards, open-source tools for developers building apps and a publicly accessible app gallery. To date, dozens of clinical applications have been built on this platform, and SMART applications are being used to provide clinical care at healthcare institutions, including
Boston Children's Hospital and Duke Medicine. The project is run out of not-for-profit institutions, Boston Children's Hospital Computational Health Informatics Program and the Harvard Medical School Department for Biomedical Informatics.
SHARP; SECONDARY USE OF EHR DATA, The SHARP grant, awarded to the Mayo Clinic Col lege of Medicine of Rochester, Minnesota, is designed to address the uses of data that are now available via EHRs. There are six target areas for this grant: ■ Two areas (clinical data normalization and natural language processing) are focused
on preparing data to support deeper analysis of content.
■ The remaining four areas focus on ways of applying data in different situations. For example, one target area—phenotyping—is designed to support the identification of patients with a host of clinical characteristics. This capability is particularly important for streamlining the process of identifying candidates for clinical trials and involves
4: NATIONAL HEALTHCARE TRANSFORMATION AND INFORMATION TECHNOLOGY
85
reviewing or summarizing a potentially wide range of clinical information. Another target area is focused more broadly on data quality when looking for patients with specific criteria. Still, another explores ways of calculating clinical quality measures associated with MU stage 2. In addition to the SHARP grantees named earlier, a grantee of the National Institutes of Health (NIH) was designed as an affiliate program, given that it has made use of the same goals. This program, called MD SHARP, was designed to develop further plug-andplay functionality for medical devices. This grant award was made to the Medical Device Plug-and-Play (MD PnP) interoperability program based at the Center for Integration of Medicine and Innovative Technology (CIMIT) and the Massachusetts General Hospital
(part of the Partners Healthcare System). This program was intended to support the development of standards that industry can adopt and a supportive eco.system of tools that were ready for deployment. Although the SHARP grants together consisted of only $60 million of the ARRA funding, they provided vital funding to further theconceptual and practical understanding of how HIT can be used and applied in various settings. Figure 4.2 reflects the SMART timeline and trajectory from 2009 to 2020.
FIGURE 4.2 SMART Timeline andTrajectory From 2009 to 2020. 2019 A Decade of SMART
CMS contains to SMART
Buk Data specs Microsoft lairKhes SMART on FHIH API in Its
2017 2009
Azure product Proposed rules from CMS &0NC specify SMART as
Altscripts and EPIC
HUU: No Small Cfiange lor the Health Moemalioo
Bxnom/'IntiotlueestheAPI White Paper:'...
Fostering Development of an
by White House. First
Ctiallenge.gov, promoted
'Utt. iTDotHealth meetfog government, vendors, academia
developers sandtMxes
2015 EHRcertTicalfon and
2OII SMART Apps Contest on
'IPhone-lke' Platform...'
launch SMART on FHIR
2015
health-related apps
challenge SMART Sandbox launched
-
FToject Argonaut meaningful use 3 final rules , commits to COS-HooKs require patient access via Implementation API
2013
SM/^T Team, ONC &
FHiR Genomics
Launch of SMART App GalieiY
SMART team joins
Launch o1 SMART CDS
FHIR development
Hooks, a decisfon support Carin Alliance focuses
Release of SMART on
effort
API b
Implemert 2l8t Century Cures .
HLTtaurvai the SMART
SOoudvefiflOfsatWhite House commit to SMART
and FHlRopen APIs
aATFHIR bulk data pro^
, on Implementing
^Kjlfoatlpn
the universal
' ? I'
Veterans accessing nealh
recads using Apple's
_
_ SMA^en FHiR_^HeaitnApp aws f--
t --
(ii
n
Go glc
. I
Release of SMART on FHIR
Ptecisfon Medicine
Apple adds SMART
SMART Team
BPCentlles: Rrst SMART
specificatfon
Mtiattve Sync tor Science
support to Its Heahh
Draft SMART API released
App In production at Boston Children's Hospital
Launtii of SMART
uses SMART App
App lor patient access ID
$15M SHARP pant to me
(RDF data models) Ctvdiac Risk app becomesfirst SMART app 2010
2012
'
Advisory Committee with
21sl Century Cures Act
diverse stakeholders committed to SMART on
mokesAPIsa
/
FHIR ecosystem
certtlied HIT.
vU / forButton 2.0 using SMART API access to
requirement for
VCMS ) CMS launches Blue
Incorporation language,
Medicare claims
\ for healUtcare
Cerner launches SMAI
p Argonaut project launched
on FHIR Developers
Dept ol Veterans Affairs to Incorporate SMART
JASON Task Force report
recommends a public API
to Implant SMART In EHRs
ML
medical data
Final rule k’om ONC stectfles SMART as the universal apps
2014
form the SMART team#
Sandbox
2016
J APItolmpiement21st
' Century Cures
SMART Markers Framework lor Patient Generated Data
2020
on FHIR support Into next-generatfon EHR platlonn CMS pifols SMART bulk data FLAT FHIR export SMART on FHIR
SMART®
published as an HL7 standard
O Copyrliyit 2020 Computational Health fotormatics Progrvn, Boston Children's Hospital.
2018
Source; Computational Health Informatics Program, Boston Children's Hospital, https://smarthealthit.org/.
86
l! INTRODUCTION
Workforce Development Passed during the worst financial crisis since the Great Depression, the HITECH Act also contained programs to support workforce development, making explicit the connection between advancing the HIT infrastructure and the development of new jobs to stimulate economic growth. There were a total of four grants for workforce development collectively worth $116 million. Together, these four programs were designed to create a mini-ecosystem that was intended to support the workforce needs of the HIT sector.
The largest award—of $68 million—went to five consortia of community colleges to implement short, non-degree training programs. These programs were intended to serve 10,500 individuals in all 50 states with training in six different HIT areas. Yet, as of October 2013, these programs had served more than 19,000 individuals (National Opinion Research Center, 2014). The curriculum for the community-college program was part of a separate but related award for $10 million and included a component related to the dissemination of curriculum materials to community colleges outside of the participating consortia. The ONC also made substantial awards totaling $38 million for a university-based training program. Tlrese awards went to nine colleges or universities to expand or create training programs requiring more substantial technical skills, including those related to health information management, public health, and privacy and security. As of October 2013, more than 1,600 individuals had received a master's degree or certificate of advanced study. The final component of this initiative was awarded to the Northern Virginia Community College to develop a competency exam. Although this certification was available to anyone, it was primarily geared toward those individuals who had completed
the community-college training programs. The success of these programs was assessed in 2019 and indicates that the ongoing certifications produced viable Health IT entry employees. The ONC infographic noted in Figure 4.3 reflects the overall success of the workforce development programs. As of August 2019, the results from the University-Based Training Program were: 64% of students were employed in HIT 6 months after program completion, and 89% were employed 6 months after program completion. From the Community College Consortia Program, 68% of students were employed in HIT or had HIT-related responsibilities 6 months after program completion; 74% were employed 6 months after program completion. As a microcosm of the broader HITECH Act, these programs demonstrate how the separate initiatives were intended to support the broader development of the HIT ecosystem. These programs were collectively intended to train both higher-skill and lower-skill individuals and deliver them and their skills to a population—clinics, hospitals, and private-sector companies in HIT—rapidly ramping up to deploy services. FROM BUILDING TO USING THE IT INFRASTRUCTURE
When looking at this collection of programs funded through the HITECH Act, it is tempting to see them as separate and distinct programs, each of which serves a separate constituency. However, these are best thought of as a cohort of interlocking programs,
each of which was designed to support the others. Each program was designed to compensate for different challenges, any one of which could derail or slow broad adoption efforts.
Before the HITECH Act, the HIT sector was largely stagnant with providers reluctant to adopt EHRs because few providers could exchange clinical content, and there was little clinical content to exchange because few providers were using EHRs. To address this
4: NATIONAL HEALTHCARE TRANSFORMATION AND INFORMATION TECHNOLOGY
FIGURE 4.3 Workforce Development Results in Certified Individuals Available in Healthcare.
ONCs HITECH Workforce Development Program University*Based Training Program
TT CmptOYrncnl M
Pro^nni CpmpltHon
Consortia Program
1.701
{mploymcnl
20 4'nmpot«-n(n
.SllldMilA
lO.OOO'f
Crnfflovinant il
SMcmhtAflx
Development
15K73:i
IT
Triiiiifd
PTOinjn Complttlon
Curriculum
Community College
!
Progttm Complclion
eMonihi Atut
ProfTim Comulttor
70%
HITPro Exam Two-thkdi of stud end
Close to two-thirds of students were
employed in heoltti IT 6 months after
wttt tmpleytd in htalth
64
procrnn completion
E&
IT 01 tia4 htalth it
ttlated mpenzibilititi
fl.'.a00+
68m
rieinnirt .ln health information. Additionally, since it was noted that evidence from stage 1 data for MU of an EHR indicated that most patients did not realize they could request digital data, the following requirements were included in the stage 2 standards: ■ More than 5% of patients must send secure messages to their provider, and more than 5% must have access to their online health data.
■ Electronic exchange providers must send a summary of care records for transitions of care and referrals, 10% of which must be sent electronically (CMS, 2014). This addition increases the involvement of information
meet MU requirements for patient engagement/activation.
technology in the effort to
99
100
: INTRODUCTION
As part of its efforts to improve the health of U.S. citizens, the federal government has provided reports that define the status of health and the associated evidence-based approaches to prevent poor health outcomes. One such report is the National Quality Strategy (U.S. Department of Health & Human Services [DHHS], 2012). This report is provided each year to the U.S. Congress as a means to define the status of health improvement efforts throughout the country. Of interest to consumers/patients is that one of the six major components of the strategy is "engaging individuals and families in their care" (DHHS, 2012, p. 1). An example is the Flex Medicare Beneficiary Quality Improvement Program, which is based on the assertion that high-quality care is not only safe but also timely, accessible, and consistent with individual and family preferences and values. Individuals are said to stay healthier when they and their families actively engage in their care, understand their options, and make choices that work for their lifestyles. This program provides technical assistance and national benchmarks to participating hospitals to improve healthcare outcomes in person-centered care (Federal Office of Rural Health Policy, 2014). The three-phase project emphasizes person-centered care by focusing on improving healthcare services, processes, and administration. The National Prevention Strategy (National Prevention Council, 2014) is a companion report to the National Quality Strategy report that focuses on endorsing preventive healthcare tactics aimed at the top U.S. health concerns. As noted in Chapter 2, there are four major components of the strategy; "empowered people" is one of the components. Decision-making is a complex process influenced by personal, cultural, social, economic, and environmental factors, including individuals' abilities to meet their daily needs, the opinions and behaviors of their peers, and their own knowledge and motivation. The goal of having a nation of empowered people is guided by four key recommendations, all of which can be facilitated by information technology (National Prevention Council, 2014). They are as follows: ■ Provide people with tools and information to make healthy choices.
■ Promote positive social interactions and support healthy decision-making. ■ Engage and empower people and communities to plan and implement prevention policies and programs.
■ Improve education and employment opportunities. These recommendations (National Prevention Council, 2014) are supported by descriptions of specific activities that numerous stakeholders can accomplish in this process of preventive health improvement for our nation. Those listed for healthcare systems, insurers, and clinicians are as follows:
■ Use proven methods of checking and confirming patient understanding of health promotion and disease prevention (e.g., teach-back method).
■ Involve consumers in planning, developing, implementing, disseminating, and evaluating health and safety information. ■ Use alternative communication methods and tools (e.g., mobile phone applications, personal health records [PHRs], and credible health websites) to support
more
traditional written and oral communication.
■ Refer patients to adult education and English-language instruction programs to help
enhance the understanding of health promotion and disea.se prevention messages (National Prevention Council, 2014, p. 2).
As a result of the 2010 enacted Patient Protection and Affordable Care Act (ACA),
thousands of additional U.S. citizens and their family members are expected to enter
5: CONSUMER ENGAGEMENT/ACTIVATION ENHANCED BY TECHNOLOGY
the U.S. healthcare system. This is the environment in which the National Quality Strategy and the National Prevention Strategy initiatives exist (Berwick & Hackbarth, 2012; Keehan et al., 2011; US Department of Health and Human Services, 2012). The ACA expands health insurance coverage in three ways: (a) by subsidizing private plans offered through the health insurance marketplaces, (b) by substantially increasing eligibility for Medicaid, and (c) by banning insurance practices that penalized people with even minor health problems (Collins et al., 2014). Preliminary findings indicated that the uninsured rate for the 19 to 64 age group declined from 20% in the period July to September 2013 to 15% in the period April to June 2014, which means that there were an estimated 9.5 million fewer uninsured adults and that the uninsured rate fell
significantly for people with low and moderate incomes and Latinx (Collins et al., 2014). The ACA, by itself, includes provisions for cutting payments and raising revenues that will achieve about $670 billion of gross savings for the CMS, according to Berwick and Hackbarth (2012). It is suggested that patient engagement/activation can mitigate some of the waste related to failures of care delivery, failures of care coordination, and overtreatment (Berwick & Hackbarth, 2012; O'Kane et al, 2012). However, according to
the Institute for Healthcare Improvement (IHl), accountable care organizations (ACOs) are also a "step in the right direction" (Torres & Loehrer, 2014, p. 62). Broadly speaking, ACOs (led by hospitals, health systems, physician groups, or other entities) are charged with providing coordinated, high-quality care to assigned beneficiaries while also meeting
quality metrics and financial targets. By doing this, the ACO can contribute to the savings generated by the program (Torres & Loehrer, 2014), thereby serving as a motivator for optimal healthcare delivery. Commonly, the use of information technology to support such improvement efforts is key to their success (Hibbard & Greene, 2013; Woods, 2016). AF4Q is the Robert Wood Johnson Foundation's signature effort to lift the overall quality of healthcare in 16 targeted communities across America (Robert Wood Johnson Foundation, 2014). Summary reports distill some of the key lessons learned by these regional alliances of providers, patients, and payers, indicating that most successful programs have (a) encouraged collaboration among patients, (b) made physician practices transparent, (c) involved patients in quality-improvement efforts, and (d) begun engaging patients to influence healthcare systems or policy formation (Robert Wood Johnson Foundation, 2014). PATIENT CARE-RELATED ENGAGEMENT APPROACHES
Organizing Frameworks Although several organizing frameworks exist to gui de the engagement of patients in their healthcare delivery, those associated with the major U.S. payer, the federal government, have the most at stake. As such, studies supporting frameworks by the Agency for Healthcare Research and Quality (AHRQ) of the National Institutes of Health (NIH) from the DHHS tend to be grounded in rigorous research such as randomized controlled trials (see James, 2013, for a summary). One such framework is created by Carman et al. (2013), titled the Patient Engagement in Health and Health Care Framework. According to the research studies supporting the Patient Engagement in Health and Health Care Framework (Figure 5.1), patient and family engagement offers a promising pathway toward better quality healthcare, more efficient care, and improved population health. Because definitions of patient engagement and conceptions of how it works vary, the framework first presents the different forms of engagement, ranging from consultation to partnership. Next, it presents the levels at which patient engagement can occur across the healthcare system, from the direct care setting to incorporating patient engagement into organizational design, governance, and policymaking. The factors that
101
102
: INTRODUCTION
FIGURE 5.1 Patient and Family Engagement Framework.
CONTINUUM OF ENGAGEMENT
Levels of
engagement
Consultation
Involvement
Partnerships and shared leadership
1
Treatment decisions Patients receive
Direct care
information about
a diagnosis
Organizational design and governance
Organization surveys patients
are made based
asked about their
on patients' preferences, medical evidence, and clinical Judgment
preferences in the treatment plan
about their care
Hospitals Involve patients as advisers or advisory council
experiences
members
Public agency conducts focus
Policy making
Patients are
groups with patients to ask opinions about a healdrcare issue
Patients' recomendations
about research
Patients co-lead
hospital safety and qualityimprovement committees
Patients have equal representation on agency committee
priorities are used by public agency to make funding
that makes decisions
decisions
programs
t
I
about how to allocate resources to health
J
Factors influencing engagement; ● Patient (beliefs about patient role, health literacy, education) ● Organization (policies and practices, culture) ● Society (social norms, regulations, policy)
Source: Carman, K. L„ Dardess, .P, Maurer, M., Sofaer, S., Adams, K„ Bechtel, C., & Sweeney, J. (2013). Patient and family engagement; A framework for understanding the elements and developing interventions and policies. Health Affairs, 32{2], 223-231. http://content,hea lthaffairs.org/content/32/2/223.full.
influence whether and to what extent engagement occurs are included. The framework explores the implications for the development of interventions and policies that support patient and family engagement and offers a research agenda to investigate how such engagement leads to improved outcomes (Carman et al, 2013). An expansion on the
Carmen et al. (2013) framework for patient engagement is the model published by the Healthcare Information and Management Systems Society (HIMSS) Center for Patientand Family-Centered Care (HIMSS, 2014). The HIMSS model includes MU categories that support provider efforts to meet the EHR federal requirements for patient engagement
(see Appendix 5.1 for details of the diagram). It guides healthcare organizations in
5: CONSUMER ENGAGEMENT/ACTIVATION ENHANCED BY TECHNOLOGY
developing and strengthening their patient engagement strategies through the use of organizations of all sizes and in all stages of implementation of their patient engagement strategies. This framework can help organizations navigate the path toward more efficient and effective models of care that treat patients as partners instead of just customers (HIMSS, 2014).
eHealth tools and resources and assists healthcare
Patient and Family Needs
As per the National Prevention Strategy, people's basic healthcare needs should be met regardless of age, gender, race, or socioeconomic status (National Prevention Council, 2014). Evidence-based needs were prioritized in the National Prevention Strategy report in 2014. The report indicated that people should: (a) live free of tobacco, (b) prevent drug abuse and excessive alcohol use, (c) eat healthfully, (d) be physically active, (e) avoid injury and violence, (f) be proactive about their reproductive and sexual health, and (g) pursue mental and emotional well-being. The report goes on to explain the roles of government and other stakeholders in achieving each of these aforementioned parameters and realizing the importance of patient and family engagement. Language clarity and health literacy are important to patient and family healthcare needs. Some groups are more likely than others to have limited health literacy. Certain populations are most likely to experience limited health literacy: ■ Adults older than 65 years ■ Racial and ethnic groups other than Caucasians ■ Recent refugees and immigrants
■ People with less than a high school degree or the general equivalency diploma (GED) ■ People with incomes at or below the poverty level ■ Non-native speakers of English (Office of Disease Prevention and Health Promotion, 2010)
The report calls for health literacy to be a national priority effort and sets guidelines and standards for improvement.
DNA analysis (especially in Scandinavian countries) is becoming more common in guiding healthcare decisions. DNA analysis comprises efforts to educate patients and
prevent adverse health conditions. This analysis has been embraced by consumers/ patients to better manage their future healthcare decisions. Several companies, such as 23andMe (2014), provide a reading of 23 chromosomes for an individual. This company and other such organizations are covered in more detail in Chapter 25. Websites, such as blogs, have evolved to share healthcare experiences, including information about health conditions and associated providers (medical professionals
and healthcare facilities). Patients Like Me (www.patientslikeme .com) and the Invisible Disabilities Association (www.invisibledisabilities.org) are two such sites that have proven
popular and helpful to consumers/patients. A site dedicated to the use of technology that supports self-management of health is Health Tech and You (www.healthtechandyou.com). It is one of many sites that provides content on the Internet of Things (loT) and health care. The loT website is where numerous fitness apps and wearable devices are featured
as patient engagement products providing patient-generated data to providers and appropriate caregivers. This engagement of healthcare team members further supports positive patient engagement outcomes (Reed et al., 2019). Figure 5.2 is a schematic of the transparent transfer of information between healthcare providers A and B (Cameron, 2021). Transparent transfer of data between providers grew exponentially during the
i03
104
I: INTRODUCTION
FIGURE 5.2 Schematic ofTransparentTransfer of Patient-Generated Data Between Providers.
PHR
9,
9
User
HI PA A
Cloud
HI PA A
i ●&
* 9
n
9
EMR
ET
*
EMR
w
11
Healthcare Provider A
Healthcare Provider B
■
m
M
Source: https://www.computer.org/publications/tech-news/trends/h ealthtech-innovations [Cameron, 2021,]
periods when the coronavims pandemic was at its peak (Koonin et al., 2020) and also supported the trend for increased patient engagement (Zimiles, 2020) PROVIDER APPROACHES/COMPETENCIES
Interprofessionalism for Patient Engagement In support of efforts for patient engagement, studies have indicated that interprofessional approaches to care delivery are most successful. Interprofession al education (IPE) collaboratives Irave convened where the definition
and associated competencies for
healthcare professionals are outlined (Interprofessional Education Collaborative Expert
Panel, 2011; World Health Organization, 2010). Note that the 2011 document has been updated with a 2016 version (Interprofessional Education Collaborative, 2016). A key component of IPE/collaboration is that professionals are all focused on the same goals for the patient and are aaitely aware of each other's roles. According to the expert panel report and understanding of professional roles, information technology is an
equally important component of this approach. As noted in Chapter 2, the depiction of interprofessional teamwork as defined by the lOM IPE core competencies expert panel illustrates the interdependencies of the core interprofessional team competencies such as (a) utilizing informatics,b (2) providing patient-centered care, (c) applying quality improvement, and (d) employing evidence-based practices (Figure 5.3; Interprofessional Education Collaborative Expert Panel, 2011).
5: CONSUMER ENGAGEMENT/ACTIVATION ENHANCED BY TECHNOLOGY
FIGURE 5.3 InterprofessionalTeamwork and Institute of Medicine Core Competencies.
Source: Interprofessional Education Collaborative Expert Panel, (2011). Core competencies for interprofessional collaborative practice: Report of an expert panel (No. ICEP-2011). Interprofessional Education Collaborative, https://www.aacom.org/docs/default source/insideome/ccrpt05-10-11. pdf?sfvrsn=77937f97_2.
Patient engagement, also known as patient activation, improves patient outcomes and decreases the cost of care delivery (Hibbard & Greene, 2013). Patients make many choices in their day-to-day lives that have major implications for their health and their need for care. Patients with a chronic disease must often follow complex treatment
regimens, monitor their conditions, make lifestyle changes, and decide when they need to seek professional care versus handling a problem independently. Patient activation is: ■ Understanding that one must take charge of one's health and that actions determine health outcomes
■ A process of gaining skills, knowledge, and behaviors to manage health ■ Having the confidence to make needed changes (Hibbard & Greene, 2013) Hibbard and colleagues have developed a 13-item self-report survey that measures patient activation levels. Based on results, patients fall into one of four categories, indicating a low to high level of patient activation. Each level has an associated healthcare coaching approach and content. Together these two components of patient care delivery have resulted in overall medical costs that were 5.3% lower than fees charged for those receiving only the usual support. They also had 12.5% fewer hospital admissions and 20.9% fewer preference-sensitive heart surgeries (Carman et al, 2013). Provider-Based Models
Several provider-based models exist that can guide patient engagement efforts by their focus on patients and their families. Denham describes one such model in the Famil}/ Health Model (Denham, 2003). The social constmction of family health has contextual,
functional, and structural dimensions. Examples of contextual aspects of the Family Health Model are the famili/ members, member traits, community context, resources, and
105
106
I: INTRODUCTION
threats, for example (Denham, 2003, p. 9). Similarly, examples of functional aspects of the model are (a) developing persons, (b) developing family, (c) member relationships, and (d) core processes (p. 9). Finally, examples of structural aspects of the model are routine type, routine characteristics, routine meaning, and routine participa7^ts (p. 9). Collectively, these aspects of the model provide for the overall perspective of a family and suggestions for considering the family when dealing with a given patient. The Betty Neuman Model of health is closely aligned with family-centered care (www.nursingtheories.weebly.com/betty-neuman.html). The Neuman Systems Model views the client as an open system that responds to stressors in the environment. The client variables are physiological, psychological, sociocultural, developmental, and spiritual. The client system consists of a basic or core stnicture that is protected by lines of resistance. Subsequently, the human being is viewed as an open system that interacts with both internal and external environmental forces or stressors. The human is in
constant change, moving toward a dynamic state of system stability or illness of varying
degrees within a given environment that includes the family. Family systems theory (FST) is derived from thebroader framework of general systems theory (Bowen, 1978, p. 153). According to systems theory, a system is defined as a whole with interrelated parts, in which the whole is more than the sum of its parts. People are viewed as part of their environment rather than separate from it and characterized by patterns of emotional interactions carried from generation to generation. The Institute for Patient- and Family-Centered Care (IPFCC; www.ipfcc.org), a nonprofit organization founded in 1992, represents an approach to the planning, delivery, and evaluation of healthcare that is grounded in mutually beneficial partnerships among healthcare providers, patients, and families. It redefines the relationships in healthcare in that it recognizes the vital role that families play in ensuring the health and well-being of infants, children, adolescents, and family members of all ages. It acknowledges that emotional, social, and developmental support are integral components of healthcare. It promotes the health and well-being of individuals and families and restores dignity and control to them. Patient- and family-centered care is an approach to healthcare that shapes policies, programs, facility design, and day-to-day staff interactions. It leads to better health outcomes, wiser allocation of resources, and greater patient and family satisfaction. One popular tool from the IPFCC is the hospital readiness assessment document (www.ipfcc. org/resources/downloads-tools.html). The 24-page document, which was created in partnership with the American Hospital As.sociation (AHA), includes an extensive survey of questions to rate a given hospital's patient and family centeredness characteristics. Based on these results, the document includes guidelines for beginning an IPFCC project.
Patient-Engaging Provider Competencies The ability to engage patients and their families in achieving effective healthcare delivery practices is a matter of growing concern for those involved in outcomes management and financing of such care delivery, for example, as noted in the CMS MU standard 2, where metrics for provider reimbursement are based on patient engagement levels. This notion was initially introduced in the 2003 lOM report titled Health Professions EducatioJi: A Bridge to Quality (Greiner & Knebel, 2003), in which educators and accreditation, licensing, and certification organizations were charged with the mandate that students and working professionals develop and maintain proficiency in core areas of patientcentered interprofessional competencies. One example of a profession that has embraced the notion of interprofessional competencies is nutrition and dietetics. In fact, Ayres and colleagues (2012) have described
5: CONSUMER ENGAGEMENT/ACTIVATION ENHANCED BY TECHNOLOGY
a process through which multiple disciplines harmonize on defined interprofessional collaborative competencies with a focus on information technology application. As background, the authors indicated that healthcare had entered the digital age as a result of HITECH and MU. The 2011 Dietetics Workforce Demand
Study conducted a future scan
of trends and issues that will shape dietetics practice in the future (Rhea & Betties, 2012). A consistent theme of this study was technology driving change for all areas of practice
and the potential for practitioners to embrace technology and new forms of information management to remain competitive in the marketplace. In support of this trend, a threeround online Delphi study was conducted among nutrition and dietetics professionals. In round three, there were almost 100 participants, all comprising educators, clinical/ community, informatics, or foodservice professionals. Using the Nancy Staggers model of information technology (Staggers et al., 2002), competencies by the level of practice were categorized. The study provided a summary of the 216 competencies by category and by level of practice. Then a range from "novice" to "informatics expert" was identified. This range of competencies was explored and quantified in regard to computer skills, informatics knowledge, and informatics skill. This master list of competencies by level of practice is available at www.eatrightpro.org. It is assumed that competencies at the lower levels of practice apply to higher levels of practice.
The rigorous work depicted through this study has provided a list of interprofessional informatics competencies on which healthcare professionals could potentially align. The categories, terminologies, and associated nomenclature should be adopted by multiple health professions to enhance healthcare delivery collaboration. Precision Medicine for Consumers
The healthcare delivery one-size-fits-all approach to treating patients is being replaced with more aistomized options. Precision medicine is a personalized approach to medicine that focuses on individuals and the unique needs of each family member (Slabodkin, 2017). Precision medicine impacts care delivery in several therapeutic areas such as oncology, metabolic is.sues (diabetes), neuroscience, respiratory conditions, and infectious diseases. However, delivery of treatments via the precision medicine approach will require collecting far more data than arecurrently collected by clinicians. Sophisticated analytic tools will be needed to glean meaningful insights from this data collection. New competencies are needed for healthcare professions tasked with analyzing and interpreting the results of such data collection (Slabodkin, 2017). Informatics for Integrating Biology and the Bedside (i2b2) is one example of an NIH-funded National Center for Biomedical Computing (NCBC) based at Partners Healthcare System in Boston, Massachusetts. Established in 2004 in response to an
NIH Roadmap Initiative, RFA, this NCBC is one of four national centers awarded in this first competition (www.bisti.nih.gov/); currently, there are seven NCBCs. One of 12 specific initiatives in the New Pathways to Discovery Cluster, the NCBCs will initiate the development of a national computational infrastructure for biomedical computing. The NCBCs and related ROls constitute the National Program of Excellence in Biomedical Computing. This kind of initiative allows patients to locate other patients like themselves and the medications they are receiving, their blood cell characteristics, and their prognosis (National Center for Biomedical Computing, 2018). The ability to sustain such programs has been challenged in the past (Wilcox & Murphy, 2014), but the value of i2B2 programs for consumer-based value is now evident (Slabodkin, 2017). Nurses and informaticists must be prepared to meet the associated technical competencies needed.
107
108
I: INTRODUCTION
Citizen Science and Community Engagement One foiTn of consumer engagement is to collect data from consumers for research purposes. Citizen science, for example, encourages members of the public to voluntarily participate in the scientific process (U.S. General Services Administration, 2019). Whether by asking questions, making observations, conducting experiments, collecting data, or developing low-cost technologies and open-source code, members of the public can advance scientific knowledge and benefit society. Through crozodsourcing, known as an open call for voluntary assistance from a large group of individuals, consumers can tackle complex challenges by conducting research at large geographic scales and over long periods of time, in ways that professional scientists working alone cannot easily duplicate. Typical citizen science projects involve management of pancreas diseases, Parkinson's, COVID-19 vaccine locations, 3-D medical equipment, gaming for a cause such as AIDS, and immunity passport benefits (The Medical Futurist, 2021).
The field of Citizen Science has a long history (primarily related to environmental science) and is now increasing with great potential to impact health and wellness. Historically, citizen science has been focused on earth sciences, ecology, and goes as far back as the identification of bird species with John Audubon's "Birds of American" collection in 1826 (Kimura & Kinchy, 2019). Movements like the Quantified Self, the passage of the U.S. Crowd Sourcing Citizen Science Act in 2017, and the development of new computational and sensing technology have opened the door for citizen participation in health-related research (Wang et al., 2020). Examples of some of the latest citizen science initiatives include a project recently launched by the American Lung Association's COVlD-19 Citizen Science research initiative (American Lung Association, 2021) and a similar one designed in Kerala, India (Ulahannan et al., 2020), designed to "collect and deposit data in a structured format... for visualizing the outbreak trend and describing demographic characteristics of affected individuals," (Vermicelli et al., 2021, p. 184). Beyond the "All of Us" initiative of data donation related to genomics, individuals are frequently involved in the design and development of the research project(s) in
methodologies known as Community-Based Participatory Research, where the "citizens" are involved in the development of the scientific project. Early projects emerged from "citizens" concerned about the impact of industry on the environment (including impact on air quality and water quality) with impact on health. Increasing focus on the social determinants of health (SDOH) is an opportunity for citizen science projects, as citizens (not patients) are interested in helping to identify the research questions and help recruit community-based participants (U.S. General Services Administration, 2019; Wilson
Center and FedCCS Steering Community, 2020).
Finally, technological advances are setting the stage for increased opportunities and participation (community-driven) citizen sciences efforts. For example, mobile health advances in measuring physical activity, fitness, and even atrial fibrillation and diabetesrelated sensors now provide opportunities for "embracing the reality that most health takes place outside the hospital and clinics . . . occurring on the other 362 days per year when a clinician does not see people" (Cook et al,, 2016). These mobile devices and sensors will increasingly be linked to personal health records (PHRs), which will not only provide interim data to assist in the management of chronic conditions but also contribute to community-based data to inform new evidence for practice. Growing examples include data donation re: chronic disease self-management with increasing opportunity to explore the SDOH impact on health. Finally, other important opportunities for citizen science are emerging with gaming use (to measure eye movement and identify issues related to brain health). Given the disproportionate impact of the
5: CONSUMER ENGAGEMENT/ACTIVATION ENHANCED BY TECHNOLOGY
COVID-19 illnesses and deaths re: SDOH and pre-existing conditions, Citizen Science will be an important area of science for the futLire. At this time, steps for the design of citizen science are available from the NIH, and there is also an emerging Citizen Scientist Curriculum at the University of Florida available at Clinical and Translational Science
Institute (2021). This helps bridge the gap between researchers and community members with a curriculum designed for teaching citizen science at the University of Florida (Clinical and Translational Science Institute, 2021). Mobile health advances in citizen science projects are emerging. Steps to implement citizen science projects are identified in detail at (U.S. General Services Administration, 2019), an official government website designed to accelerate the use of crowdsourcing and citizen science across the U.S. government. Collective Community Impact Consumers can benefit from collective community impact initiatives because they
ensure that the resources to improve the social and health status are provided in the most effective way possible. The Community Impact Forum (a popular portal for community impact) has indicated that "too many organizations are working in isolation from one another." Collective impact (Cl) provides a structured framework for groups to achieve social change (e.g., definition of the problem, shared vision, common agenda, and agreement on shared measurements to track progress (Collective Impact Forum, 2016, para. 1). ■ Establishes shared measurements; that means agreeing to track progress in the same way, which allows for continuous improvement (Collective Impact Forum, 2016, para. 1). Cl as an approach has been shown to lead to sustained improvements and is gaining use worldwide in working on community development concerns. This volume examines the application of Cl to a variety of development issues and offers insights into approaches that seem to have generated positive and sustainable outcomes (Walzer et ah, 2016). One example of this approach is by a local hospital system and the local office of the DHHS. In this example of community impact, the hospital system and DHHS partner to provide fresh fruits and vegetables every second and fourth Tuesday of the month to improve the health of the community members (Crutchfield & Hayes, 2018). The collective community impact approach and involved information technology professionals are important to consumers as an effective means to engage them in their health promotion efforts. TECHNOLOGY AND HEALTH COACHING FOR HEALTH BEHAVIOR CHANGE
The Health Research Institute (HRI) 2014 report estimated that the $2.8 trillion U.S. healthcare industry is being upended by companies attuned to the needs and desires of
empowered consumers (PricewaterhouseCoopers, 2014a). These new entrants from the retail, technology, telecommunications, consumer products, and automotive industries are centered on consumer transparency, convenience, and prevention. For example, $267 billion is estimated for their fitness and wellness market. Using a device attached to a phone, 46.9% of customers surveyed have checked for an ear infection at home, and 38.6% have had a live visit with a physician via their smartphone and a webcam (Pricewaterhouse Coopers, 2014a). Another HRI study indicated that 43% of consumers surveyed prefer an online healthcare website with different options and different prices so they can optimally compare the overall value (Pricewaterhouse Coopers, 2014b).
109
f 10
: INTRODUCTION
Technology and Patient-Generated Health Information Trends
Patients also have made contributions to the empowered consumer movement. They may contribute through patient-generated health information (PGHI). As part of a national eHealth collaborative, PGHI was evaluated by technical experts and summarized in a December 2013 report (National eHealth Collaborative, 2013). The general approach would be to have the PGHI entered into the health system electronically. Examples of medication-related PGHI items are medication (history, current), medication adherence (including over-the-counter medication), and medication reactions/symptom reporting. Although concerns existed on both the patient and the provider side, the main emphasis for success is that expectations be managed well. Providers and patients must have a shared understanding of what information would be most valuable, how data should be shared, and what will happen after they share the data (National eHealth Collaborative, 2013). Based on the.se findings for consumer spending and potential PGHI, information technology has been identified to enhance, expedite, and optimally support the patient information content for patient healthcare decision-making. Patient engagement starts with giving patients the tools they need to understand what makes them sick, how to stay healthy, and what to do if their conditions worsen. It means motivating and empowering patients to work with clinicians—to be active participants in their care by asking questions, knowing their medications and medical history, bringing friends or relatives to appointments for support, and learning about care that may be unnecessary. It can also mean giving them a seat at the table to improve the care that hospitals and doctors' offices provide (Robert Wood Johnson Foundation, 2014). Technology-based PHR (Agarwal et al., 2013; Lau et al., 2013) and mobile healthcare are important components of the consumer/patient engagement effort. Because of their complexity, these are addressed in detail in other chapters. However, as mentioned, during the coronavims pandemic, the use of mobile applications proliferated (Zimiles, 2020). This and the rising ownership of smartphones and tablets, or apps, are all promising tools for engaging patients in their healthcare (Singh et al, 2016). Therefore, a framework for evaluating mobile apps based on their patient engagement functionality is warranted
(Singh et al., 2016) and Figure 5.4 illustrates one propo.sed with increasingly robust levels
of patient engagement with their chronic health conditions. Using this pyramid to assess
the level of engagement for various healthcare apps, those most applicable for a given patient can be recommended with confidence.
The app called "7 Cups of Tea" anonymously connects users with trained "active listening" for emotional .support, counseling, and therapy to help with depression, anxiety, and stress (Singh et al., 2016, p. 4). Two clinicians assessed the app ba.sed on the "levels of engagement with healthcare framework" illustrated in Figure 5.4. Although this app did not provide a wide array of functionality, the clinicians both indicated that it addressed specific engagement needs of patients and was worth recommending. Other apps could be similarly as.sessed. For example, T2 Mood Tracker by the National Center for Telehealth and Technology helps users monitor their emotional health, including anxiety, depression, head injury, stress, posttraumatic stress disorder, and general well being (Singh et al., 2016, p. 4). It was initially designed for the U.S. Department of Defense for "general wellness" tracking. When asses.sed by two clinicians on the engagement healthcare framework (Figure 5.4), it was found to be rated highly for identifying wellness status but rated very low for patient engagement and for "providing guidance." For example, when the patient exhibited a rating of severe depression, and no guidance was provided, this was considered a potential patient safety issue.
In healthcare, treatments should be tailored to the patient and pre.scribed with an understanding of its benefits and risks. This is also true of apps, which appeal to different audiences by offering various functionalities (Singh et al., 2016). Using the level
5: CONSUMER ENGAGEMENT/ACTIVATION
ENHANCED BY TECHNOLOGY
FIGURE 5.4 Strategies to Engage/Activate Patients Using Mobile Applications Based on Level of Engagement With Healthcare.
:
of Engagement With Health Care framework (Figure 5.4) that considers the engagement, quality, and safety of mobile apps, clinicians can identify tmstworthy apps that serve the engagement of high-need, high-cost populations. Health and Wellness Coaching, Mobile Devices, and Health Behavior Change Much of this person/patient-centered app development is parallel to the growing health and wellness industry, where in addition to personal counseling, the customization of apps to the individual, and the teaching of how to best use these apps is undertaken. Health and wellness coaches are said to:
"Partner with clients seeking to enhance their well-being through self-directed, lasting changes, aligned with their values and in the course of their zoork, health and zoellness coaches display an unconditional positive regard for their clients and a belief in their capacity for change, honoring the fact that each client is an expert on their own life, zohile ensuring that all interactions are respectful and non-judgmental." (NCCHWC, 2017, p. 1)
The National Board for Health & Wellness Coaching (NBHWC) is overseen by a board of directors and supported by committees composed of volunteers and staff members who seek to advance the profession of health and wellness coaching by supporting the national certification exam (National Board for Health & Wellness Coaching, 2021). One
111
112
: INTRODUCTION
of the areas where health and wellness coaches have been most efficient and effective is
in supporting person /patient use of complex mobile devices to monitor health metrics. Since a significant percentage of individuals in the United States and globally are coping with and trying to manage a chronic condition, using mobile health devices and also working with health and wellness coaches hold tremendous promise. This approach not only improves the health and well-being of many adults, adolescents, and even some children, it helps reduce costs at the individual level, the community level, and at the organization level (Jonk et al, 2015). With the FDA approval of mHealth devices such as the Freestyle Libre and Dexcom 6 for diabetes monitoring, individual persons/patients are already participating in self-monitoring, using these devices for "just-in-time" evidence supporting health-behavior change. Many of these individuals are working with nationally certified health and wellness coaches in a variety of settings. In terms of cost, health and wellness coaching has been found to decrease the cost of chronic condition management (Jonk et al., 2015). A key development for associated reimbursement purposes is that the American Medical Association (AMA) has approved New Category III CPT Codes for health coaching services reimbursement (AMA, 2019). The Category III Health and WellBeing Coaching Codes include:
■ 0591T health and well-being coaching face-to-face; individual, initial assessment ■ 0592T individual, follow-up session, at least 30 minutes
■ 0593T group (two or more individuals), at least 30 minutes
Another service provided by health and wellness professionals appears to be much needed for healthcare professionals: "resilience-coaching." Resilience coaching assists individuals, groups, teams, and organizations deal with stress experienced as an
empathetic caregiver and particularly intense stress related to COVID-19 and the related social challenges (HintonWalker, 2020). This COVID-19 burnout has been negatively impacting sleep, stress, and relationships, thereby increasing the need and use of wellness coaching (HintonWalker et al., 2020). Resilience coaching focuses the person on mastering skills to transform negative thinking and to calm the stress response. Thus, health and wellness coaching is also valuable for healthcare professionals in these stressful times. HEALTHCARE COSTS AND TECHNOLOGY
Some suggest that the current approach to healthcare reform and, ultimately, healthcare cost reduction is not on the right track because of financial misalignment (Mechanic, 2008). However, others indicate that with a strong focus on infrastructure and technology, the needed transitions can be achieved (Feldman et al., 2005; Murtaugh et al., 2005). In the fall of 2013, a group from the Consumers Union and the Robert Wood Johnson Foundation met to explore potential solutions to the rising cost of healthcare (Consumers
Union, 2014). It was noted that healthcare spending consumes more than $1 of every $6 we earn, but many consumers are often unaware of the real cost of healthcare. Individuals, families, employers feel the impact, and those crafting state and federal budgets in that
rising healthcare costs undermine wage growth. Between 1999 and 2009, almost all increases in compensation have taken the form of paying rising health premiums, and almost none have been allocated to increasing the take-home paycheck. Rising healthcare costs force trade-offs in our national and local government budget priorities, reducing the money available for education and other important programs. The report indicated that our current path is unsustainable, and it is evident that good quality healthcare can
be delivered for less money (Consumers Union, 2014). An extensive effort by the lOM to determine how our nation can provide high-quality
care at low cost was described in a 450-page report (Sniith et al., 2012). Essentially, the
5: CONSUMER ENGAGEMENT/ACTIVATION ENHANCED BY TECHNOLOGY
committee members from the payer and provider organizations across the nation believed that achieving a learning healthcare system aligned to promote continuous improvement in care is necessary. Four fundamental characteristics of such a system were identified: ■ Science and informatics: (a) real-time access to knowledge and (b) digital capture of the care experience ■ Patient-clinician partnerships: engaged, empowered patients
■ Incentives: (a) incentives aligned for value and (b) full transparency ■ Culture: (a) leadership-instilled ailture of learning and (b) supportive system competencies (Smith et al., 2012, p. S-11) Recommendations, followed by specific strategies for stakeholders, are described in the full report (Smith et al, 2012). There are three major categories of the committee's recommendations: (a) fundamental elements, (b) care improvement targets, and (c) supportive policy environment. Of significance to information technology professionals is the fact that "digital infrastructure" and "data utility" (p. S-20) are the two components of the fundamental elements category. CASE STUDY
Consider the following:
Aims and objectives: To evaluate the outcome of coherent nursing practice in the form of a partnership that addresses the complexity of living with chronic obstructive pulmonary disease (CORD). Background: CORD is a wide-ranging and progressive chronic disease that requires the relentless attention of the persons having the disease and family involvement (Ingadottir & Jonsdottir, 2010). Rarticular consideration is required in healthcare for those with an advanced and complicated stage of the disease. Please respond to the following questions:
1, How would you use the Patient Engagement in Health and Health Care Framework to plan a patient engagement approach for the patient and family? 2. What type of information technology would you advise to support the care delivery?
SUMMARY
Characteristics of patient engagement/activation are becoming more well known to patients and providers. This is especially noted surrounding the healthcare delivery trends associated with the 2020 COVID-19 pandemic. Much of this is also fueled by the government initiatives in which value-based reimbursement focuses on patient engagement/activation. The use of organizational frameworks, coupled with family involvement, helps foster a patient's involvement in their care. The community science initiative provides significant support for patient engagement as it focuses on community involvement. Also of note is that providers should have the interprofessional skills and knowledge necessary to support patient engagement. Technology support for patient engagement and associated cost reduction is discussed in detail in Chapter 15. Advanced practice RNs and other advanced interprofessionals would benefit from these skills.
I 13
I 14
: INTRODUCTION
END-OF-CHAPTER
RESOURCES
EXERCISES AND QUESTIONS FOR CONSIDERATION
1. As an information technology professional, your role in support of patient engagement is guided by the five components of interprofessional teamwork and lOM core competencies described in Chapter 2. Using these components, describe how you and your colleagues reflect each component. 2. There are four specific activities identified as supportive of preventive healthcare (National Prevention Council, 2014). Select one of these four activities and describe a situation in
which you have conducted such an activity, a. Use proven methods of checking and confirming patient understanding of health promotion and disease prevention (e.g., teach-back method),
b. Involve consumers in planning, developing, implementing, di.sseminating, and evaluating health and safety information, c. Use alternative communication methods and tools (e.g., mobile phone applications, PHRs, and credible health websites) to support more traditional written and oral communication,
d. Refer patients to adult education and English-language instruction programs to help
enhance the understanding of health promotion and disease prevention messages.
3. Ayres et al. (2012) used a Delphi technique to gain consensus among a variety of healthcare professionals. Review this master list of competencies provided by the level of practice that is available at www.eatright.org. a. Identify where you would categorize your informatics competencies, b. Identify other competencies that could be added to address consumer engagement and self-reported health data. 4. According to Hibbard and Greene (2013), patient activation is
a. Understanding that one must take charge of one's health and that actions determine health outcomes,
b. A process of gaining skills, knowledge, and behaviors to manage health,
c. Having the confidence to make needed changes (Hibbard & Greene, 2013). Explain which of these three characteristics umild be the most difficult for \joii to implement and lohi/.
5. Consider content covered concerning policy changes in the United States and the strategic plan needed to improve care and drive down cost through health information technology. Give one example of an information technology policy change that has supported patient engagement.
ADDITIONAL RESOURCES
sniwiirutiKKi^.
CONNECT
A robust set of instructor resources designed to supplement this text is located at http://connect.sprlngerpub.com/content/book/978-0-8261-8526-6 . Qualifying
instructors may request access by emailing textbook@springerpub. com.
5: CONSUMER ENGAGEMENT/ACTIVATION ENHANCED BY TECHNOLOGY
REFERENCES
Agarwal, R., Anderson, C., Zarate, J., & Ward, C. (2013), If we offer it, will they accept? Factors affecting patient use intentions of personal health records and secure messaging. Journal of Medica.l Internet Research, 25(2), e43. https;/ /doi.org/10.2196/jmir.2243 Aligning Forces for Quality, National Program Office. (2015). Aligning forces pr qualiti/: Improving health and healthcare in communities across America, http: / / forces4quality.org/about-us Alston, C., Paget, L., Haivorson, G., Novelli, B., Guest, J., McCabe, .P, Hoffman, K., Koepke, C., Simon, M., Sutton, S., Okun, S., Wicks, .P, Undem, T., Rohrbach, V., & Von Kohorn, I. (2012). Communicating uhth patients on health care evidence (Discussion paper). Institute of Medicine. American Lung Association. (2021). The COVID-19 citizen science study. Lung.org. https://www.lung .org / lung-health-diseases / lung-disease-lookup / covid-19/ action-initiative/ covid-citizen-science -study American Medical Association. (2019). American medical association appraises new category III CPT codes for
coaching. AMA.org. https; / /www.24-7pres.srelease.com/ press-release/466893 /american-medical -association-approves-new-category-iii-cpt-codes-for-coaching Ayres, E. J., Greer-Carney, J., Fatzinger McShane, P. E., Miller, A., & Turner, P. (2012). Nutrition informatics competencies across all levels of practice; A national Delphi study. Journal of the Academy of Nutrition and Dietetics, 222(12), 2042-2053. nttps:/ / doi.org/10.1016/j.j and.2012.09.025
Berwick, D. M., & Hackbarth, A. D. (2012). Eliminating waste in U.S. health care. Journal of the American Medical Association, 307(14), 1513-1516. https://doi.org/10.1001 /jama.2012.362 Bowen, M. (1978). Family therapy in clinical practice. Jason Aronson. Carman, K. L., Dardess, ,P, Maurer, M,, Sofaer, S., Adams, K., Bechtel, C., & Sweeney, J. (2013). Patient and family engagement: A framework for understanding the elements and developing interventions and policies. Health Affairs, 32(2), 223-231, http://content.hea lthaffairs.Org/content/32/2/223.full Centers for Medicare & Medicaid Services. (2014, November). Stage 2: Meaningfid use. https://www .cms.gov/Regulations-and-Guidance/Legislation/EHRIncentive-Progr ams/downloads/Stage2 _EPCore_2_ePrescribingmdf Collective Impact Forum. (2016). Wltat is collective impact? http://-collectiveimpactforum.org/what -collective-impact
Collins, S. R., Rasmussen, .P W., & Doty, M. M. (2014). Tracking trends in health system performance: Gaining ground: Americans' health insurance coverage and access to care after the Affordable Care Act's first open enrollment period (Brief). The Commonwealth Fund. Committee on Quality of Healthcare in America. (2001). Crossing the quality chasm: A neio health system
far the 21st century. National Academies Press, http:/ /www.nap.edu/catalog,php?record_id=10027
Consumers Union. (2014). Addressing rising health care costs workshop summary: A pioneering meeting of
advocates seeking to address health care costs (No. RWJF410914). Consumer Union-Robert Wood John Foundation.
Clinical and Translational Science Institute. (2021). Citizen Scientists Program. CTSI.UFL.edu. https://
www.ctsi.ufl.edu/about/ctsi-programs/about-the-citizen-scientist -program/ Crutchfield, R., & Hayes, A, (2018). Dallas county health and human services partners with Baylor
Scott and White and wellness center to promote consumption of fresh fruits and z^egetables. http://
www.dallascounty.org/ department/hhs/ media / pressreleases/2018 /documents/DCHHS
_BSWHWCPartnershipPRESSRELEASE.PDF Denham, S. (2003). Family health: A framework for nursing. F. A. Davis. Federal Office of Rural Health Policy. (2014). FLEX Medicare beneficiary quality improvement project
(No. FLEX2014). http://www.ruralcenter.org/sites/default/files/MBQIP%200verview%20for%20
CAHs.pdf Feldman, .P H., Murtaugh, C. M,, Pezzin, L. E., McDonald, M. V, & Peng, T. R. (2005). Just-in-time evidence-based e-mail "reminders" in home health care: Impact on patient outcomes. Health Sendees Research, 40(3), 865-885. https:/ /doi.org/10.1111/j,1475-6773,2 005.00389.x Greiner, A. C., & Knebel, E. (Eds.). (2003). Health professions education: A bridge to quality. The National Academies.
Healthcare Information and Management Systems Society. (2014). HIMSS center afr patient- and family -centered care, http://www.himss.org/library/NEHC " ibard, J., H., & Greene, J. (2013). , . What the evidence shows about patient activation: Better health outcomes and care experiences; fewer data on costs. Health Affairs, 32(2), 207-214. https://doi .org/10.1377/hlthaff.2012,1061 HintonWalker, P. (2020). Resilience: Anecessary requirement for health, well-being and survival for caring professionals. Journal of Health and Human Experience, 6(1), 21-36. https:/ /jhhe.sempervifoundation .org/pdfs/v6nl / 02%20Hinton%20Walker.pdf HintonWalker, P, Grabbe, L., Blueford, B., & Remsey, G. (2020). Exploring the phenomenon and perspectives of empathy and compassion in the caring professions and enhancing resiliency: Part II. Journal of Health and Human Experience, 5(1), 53-74.
Ingadottir, T. S., & Jonsdottir, H. (2010). Partnership-based nursing practice for people with chronic obstructive pulmonary disease and their families: Influences on health-related quality of life and hospital admissions. Journal of Clinical Nursing, 19(19), 2795-2805. https;//doi.org/10.1111/j.l365 -2702.2010.03303.x
115
116
: INTRODUCTION
Interprofessional Education Collaborative Expert Panel. (2011). Core competencies for interprofessional collalforatnv practice: Report of an expert panel (No. ICEP-2011). Interprofessional Education
Collaborative, https:/ / nebula. wsimg.com/3ee8a4b5b5f7ab794c742 bl4601d5f23?AccessKeyId =DC06780E69ED19E2B3A5&disposition=0&aiioworigin=l
Interprofessional Education Collaborative. (2016). Core competencies for interprofessional collaborative practice: 2016 update. (Report No. IPE2016). Washington, DC: Interprofessional Education Collaborative, ittps:/ / hsc.unm.edu/ipe/resources/ipec-2016-core-competencies. pdf James, J. T. (2007). A sea of broken hearts: Patient rights in a dangerous, profit-driven health care spstem. AuthorHouse.
James, J. T. (2013). A new, evidence-based estimate of patient harms associated with hospital care, journal of Patient Safety, 9(3), 122-128. https:/ /doi.org/10.1097/PTS.0 b013e3182948a69 Jonk, Y., Lawson, K., O'Connor, H., Riise, K., Eisenberg, D., Dowd, B., & Kreitzer, M. (2015). How effective is health coaching in reducing health services expenditures? Medical Care, 53(2), 133-140.
https://doi.ore/10.1097/MLR.0000000000000287 Keehan, S. P., Sisko, A. M., Truffer, C. J., Poisal, J. A., Cuckler, G. A., Madison, A. J., Lizonitz, J. M.,
& Smith, S. D, (2011). National health spending projections through 2020: Economic recovery and
reform drive faster spending growth. Health Affairs, 30(8), 1594-1605. https://doi.ore/10.1377/ hlthaff.2011.0662
Kennedy, A. D., Sculpher, M. J., Coulter, A., Dwyer, N., Rees, M., Abrams, K. R., Horsley, S., Cowley, D., Kidson, C., Kirwin, C., Naish, C., & Stirrat, G. (2002). Effects of decision aids for menorrhagia on treatment choices, health outcomes, and costs: A randomized controlled trial. Journal of the American Medical Association, 288(21), 2701-2708. https://doi.org/10.1001 /jama.288.21.2701 Kimura, A. H., & Kinchy, A. (2019). Scienceby the People: Participation, Power, and the Politics ofEnvironmental Knowledge. Rutgers University Press. Koonin, L. M., Hoots, B., Tsang, C. A., Leroy, Z., Farris, K., Jolly, T., Antall, .P, McCabe, B., Zelis, C. B. R., Tong, I., & Harris, A. M. (2020). Trends in the use of telehealth during the emergence of the COVID-19 pandemic - United States, January-March 2020. MMWR. Morbidity and Mortality Weekly Report, 69(43), 1595-1599. https://doi.org/10.15585/mmwr.mm6943a 3 Lau, A. Y. S., Dunn, A. G., Mortimer, N., Gallagher, A., Proudfoot,J., Andrews,A., Liaw, S.-T, Crimmins, J., Arguel, A., & Coiera, E. (2013). Social and self-reflectiveuse of a web-based personally controlled health management system. Journal of Medical Internet Research, 15(9), e211. https: / /doi.org/10.2196 / jmir.2682 Mechanic, D. (2008). Barriers to eliminating waste in U.S. health care. Journal of Health Services Research & Policy, 23(2), 57-58. https: / / doi.org/10.1258/jhsrp.2008.00 7165 Murtaugh, C. M., Pezzin, L. E., McDonald, M. V., Feldman, P H., & Peng, T. R. (2005). Just-in-time evidence-based e-mail "reminders" in home health care;
Impact on nurse practices. Health Sendees
Research, 40(3), 849-864. https://doi.Org/10.llll/j.1475-6773.20 05.00388.x National Board for Health & Wellness Coaching. (2021). NBHWC: Who we are. NBHWC.org. https;// nbhwc.org / about / National Center for Biomedical Computing. (2018). Informatics for integrating biology and the bedside (i2b2). https: / /www.i2b2.org; https: / /www.i2b2.org/about/ National eHealth Collaborative. (2013). Patient generated health information: Technical expert panel final report December 2013 (No. Grant#7U24AE000006-02). Author, https://www.healthit.gov/sites/ default/files/pghi_tep_finalreportl21713.pdf National Prevention Council. (2014). 2014 annual status report (No. NPC2014). U.S. Department of Health and Human Services, Office of the Surgeon General, https: / /www.surgeongeneral.gov/priorities/ prevention/2014-npc-status-report.pdf
NCCHWC. (2017). About: National consortium for credentialing health & wellness coaches. NCCHWC.org.
https://www.ncchwc.org/about/ O'Connor, A. M., Llewellyn-Thomas, H. A., & Flood, A. B. (2004). Modifying unwarranted variations in health care: Shared decision making using patient decision aids. Health Affairs, (Suppl. Variation), VAR63-72. https://doi.org/10.1377/hlthaff.var.63 O'Connor, K. (2007). Toward the "tipping point": A new coalition of groups is working quietly to reform U.S. health care. Health Progress, 88(3), 32-34. https://www.cha usa.org/docs/default-source/health -progress/toward-the-tipping-point-pdf.pdf?sfvrsn=0
Office of Disease Prevention and Health Promotion.
(2010). National action plan to improve health literacy
(No. USDHHS2010). U.S. Department of Health and Human Services, http://www.health.gov/ communication/hlactionplan
O'Kane, M., Buto, K., Alteras, T., Baicker, K., Fifield, J., Giffin, R., Hansen, J., & Saunders, R. (2012).
Demanding value from our health care: Moth’aling patient action to reduce waste in health care. (No. Demand2012). Institute of Medicine. https://nam.edu/-wp-content/ -uploads/2015/06/VSRT
-DemandingValue.pdf PricewaterhouseCoopers. (2014a). Healthcare’s nexv entrants: Who will be the industry's amazon.com? (No. PwC-HRI2014). Author.
PricewaterhouseCoopers. (2014b). Medical cost trend: Behind the numbers 2015 (No. PwC-2014).Author.
5; CONSUMER ENGAGEMENT/ACTIVATION ENHANCED BY TECHNOLOGY
Ratliff, R., Webb, K., & Safavi, K. (2014). Insight driven health: Consumers with chronic conditions believe the abiliti/ to access electronic medical records ouhveighs concern of privacy invasion (Survey No. 1). Accenture
https: / / www.accenture.com/tOOOlOlOlTOOOOOO
w
/au-en/_acnmedia / PDF-3/ Accenture
-Consumers-with-Chronic-Conditions-Electronic-Medical-Records-v2 .pdf
Reed, M. E., Huang, J., Brand, R. ]., Neugebauer, R,, Graetz, I., Hsu, Ballard, D. W., & Grant, R. (2019). Patients with complex chronic conditions; Health care use and clinical events associated with access
to a patient portal. PloS ONE, 14(6), e0217636. https: / / doi.org/10.1371/journal.pone.0217636
Rhea, M., & Betties, C. (2012). Future changes driving dietetics workforce supply and demand: Future scan 2012-2022. journal of the Academy of Nutrition and Dietetics, 122(3 Suppl.), S10-S24. https: / /doi .org/10.1016/j.jand.2011.12.008 Robert Wood Johnson Foundation.(2014). What we're learning: Engagingpatients improves health and health care (No. RWJF411217).Author. Schoen, C., Guterman, S., Shih, A., Lau, J., Kasimow, S., Gauthier, A., & Davis, K. (2007). Bending the amv: Options for achieinng saznngs and improznng z’alue in US. health spending (No. CWF2007). Commonwealth Fund, https://www.commonwealthfund.org/publications /fund-reports/2007/ dec/bending-curve-options-achieving-savings-and-improving-value- us Slabodkin, G. (2017, December 27). Precision medicine requires more data, sophisticated analytics.
Health Data Management (Online), https;//search.proquest.com/doc view/1980704576
Smith, M., Saunders, R., Stuckhardt, L., & McGinnis, M. J. (Eds.). (2012). Best care at lozoer cost: The path to
continuously learnmg health care in America (1st ed.). National Academies Press. Stacey, D., Legare, F., Col, N. F., Bennett, C. L., Barry, M. J,, Eden, K. B., Holmes-Rovner, M., LlewellynThomas, H. A., Lyddiatt, A., Thomson, R., Trevena, L., & Wu, J. H. (2014), Decision aids for people facing health treatment or screening decisions. Cochrane Database of Systematic Reznezos, 1, CD001431, https://doi.org/10.1002/14651858.CD001431.pub4
Staggers, N., Gassert, C. A., & Curran, C. (2002). ADelphi study to determine informatics competencies for nurses at four levels of practice. Nursing Research, 52(6), 383-390. https://doi.org/10.1097/00006199 -200211000-00006
Torres, T., & Loehrer, S. (2014). ACOs: A step in the right direction. Healthcare Executwe, 29(4), 62-65. http://www.ihi.org/resources/Pages/Publications/ACOsStepinRightD irection.aspx 23andMe. (2014). Find out zohat your DNA says about you and your/am/Zy. http; / / www.23andMe.com
The Medical Futurist. (2021). Top 6 crowdsourcing examples in digital health. The Medical Futurist. https://medicalfuturist.com/crowdsourcing-in-digital-nealth/ Ulahannan, ]. P, Narayanan, N., Thalhath, N., Prabhakaran, P, Chaliyeduth, S., Suresh, S. P, Mohammed, M., Rajeevan, E., Joseph, S., Balakrishnan, A., Uthaman, J., Karingamadathil, M,, Thomas, S. T,
Sureshkumar, U., Balan, S., & Vellichirammal, N. N. (2020). A citizen science initiative for open data and visualization of COVID-19 outbreak in Kerala, India. Journal of the American Medical Informatics
Association: jAMIA, 27(12), 1913-1920. https://doi.org/10.1093/j amia/ocaa203
U.S. Department of Health & Human Services. (2012). National strategy for quality improvement in health care (Congressional Report No. DHHS-2012). Agency for Healthcare Research and Quality. U.S. Department of Health & Human Services. (2012). Patient protection and affordable care act: The health
care lazo and you. Healthcare.gov. http://www.healthcare.gov/law /index.ntml
U.S. General Services Administration. (2019). CitizenScience.goz K An official website of the United States
Government - about. CitizenScience.gov. https://www.citizenscien ce.gov/about/#
Vermicelli, S,, Cricelli, L,, & Grimaldi, M. (2021). How can crowdsourcing help tackle the COVID-19
pandemic? An explorative overview of innovative collaborative practices. R&D Management, 51 (2), 183-194. https:/ /doi.org/10.1111 /radm.12443
Walzer, N., Weaver, L., & McGuire, C. (2016). Collective impact approaches and community development issues. Community Dei’elopinent, 47(2), 156-166. https: / /doi.org/10.1080/15575330.2015.1133686 Wang, C., Han, L., Stein, G., Day, S., Bien-Gund, C., Mathews, A., Ong, J. J., Zhao, P, Wei, S., Walker, J., Chou, R., Lee, A., Chen, A., Bayus, B., & Tucker, J. D. (2020). Crowdsourcing in health and medical
research: A systematic review. Infectious Diseases of Poverty, 9(1), 8. https://doi.org/10.1186/s40249 -020-0622-9
Wilcox, A., & Murphy, S. N. (2014). Mission and sustainability of informatics for integrating biology and the bedside (i2b2). eGEMs (Generating Ezndence & Methods to Improve Patient Outcomes), 2(2), 1074.
https://doi.org/10.13063/2327-9214.1074 Wilson Center and FedCCS Steering Community. (2020). Federal community of practice for crozodsourcin and citizen science. (Report No. 2020). U.S. General Services Administration: Wilson Center and FedCC Steering Community. https://www.citizenscience.gov/assets/FedCCS .pdf Woods, D. (2016, April 6). TIte patient as a consumer: hworporating technology to improz’e the patient experience. Health l.T. & CIO Report, HIT, 1-1.
World Health Organization. (2010). Framezoork for action on interprofessional education & collaboratwe practice (No. WHO/HRH/HPN/10.3). Author. Zimiles, A. (2020, July 8). Four new statistics that prove that telemedicine isn't just a pandemic fad: Why telemedicine will largely dictate the future of healthcare. Medical Economics, 2020,1-1. https: / / www.medicaleconomics.com/view/four-new-statistics-that-prove-tha t-telemedicine-isn-t-just-a -pandemic-fad
117
00
APPENDIX 5.1 HIMSS Patient Engagement Framework With Meaningful Use Categories
H;mss
02014 Healthcare Irttormation and Martagement Systems Society (HIMSS) ehe
Tire Patient Engagement Eramework is iKCtned under a Creative Convrions
PATIENT ENGAGEMENT FRAMEWORK '
FOUNDATION
fee
Attribution Noncommercial NoDutvt i 0 Ursited Suies License
X
z 73
Engage Me + [jf Information and Way-finding ● Maps and directions
Information and Way-Finding ●
Mobile
● Services directory
o Nearest healthcare services
● Physician directory
o Symptom checker
e-Tools
●
● Healthy eating tracking
●
HIPAA
● Insurance
● ●
Advance directives Informed consent
● fitness trackirtg
progress arxf heaTth milestones on social media
● Prescribed medication ● Procedure/lreatment
● Care plan management
■ Patient profile ● Register or pay a bill
●
Schedule a clinic
appointment
Care Instructions
●
Reminders 0 Medication
t-C/ + * + 9 + i6
^l^nformation. Way-finding, and Analytics/Ouality
IniormatioiLVi^-Finding andAnalytics/Quality
c n
"
● Patient-specific predictive modeling
● Care comparison for providers,
H
treatments, and medications
-
● Patient accountability scores
o
2
u Preventive services
o Follow-up appointments
● Record correction requests ● Advance directives (scanned)
Ft
● Wellness plan ● Coordination of care across systems Integrated Forms: EKR ●
● Immunization (public health)
Patient-Specific Education ■
Matenals in
● Corsditiors-specific
Spanish and the top S rtational languages
self-management tools
●
Publish and subscribe
● Shared decision
> Home monitoring
making
s Devices
« Preferencesensitive care o Informed
● Patient-generated data in EHR 0 Questionnaires
o Tele-medkine ● Directives 0 Advatice
choke/consent ●
Adherence
Interoperable Records
● E-referral coordination between providers ● Ambulatory and hospital records integration * Images and video in EHR ● Commercial labs, radiology, medications
● Chronic care self-nusnagement ● Remirtders for dally care Patient Access and Use
Care Team-Generated Data
* Shared care plans a Episodic
● Team outcomes 3 Adherence
o Chronic 0 End of life
3 Costs
3 Quality
o Physician orders for life-sustaining treatment
reporting o Medications o Self-care 3 Wellness
exchange (HIE)
Patient-Specific Education ● Care planning
● Patient-set privacy controls Patient-Generated Data
* Seif-management diaries
● integrated with health information
collaborative care records) irt
● Publish/subscribe for complete record > Distribution of record among care team ● Patient-granted permissions
o Summaryofeare
Patient-Generated Data
o Health hirtory p Demographics
Integrated Forms: EHR
(replaced ^ interoperable
Patient Access
● EMR integrated with patient PHR
Pro-viiil
e
Clinical trial records
Patient Access: Records
n
● e-Visits as pan of ongoing care
● Advance care planning
accountable care
● Care experience surveys ● Symptom assessnwnts
0 Corxveniertce
e-Vfsrts and e-Teols
e-Tools
5
● Guides to understanding
● Transmit patient record electronically ● Copy the patient or a healthcare designee when sharing electronic record
o Costs
° Quality
Patient-Specific Education ● Materials in Spanish
Patient Access: Records ● View electronic health record ● Download electronic health record
● Secure messaging
flf t-
● Patient-specific quality indicators
Integrated Forms: EHR
Patient-Specific {ducation ●
● Online nurse
● Virtual coaching
Interactive Forms: Online
Paiiem-Specific Education ● Tests
● Quality and safety reports on providers and healthcare organizations ● Patient ratings of providea hospitals and
a
Intolerances
3 Allergies 0 0
Values Preferences
Interoperable Records ● Integrated with clinical trial records ● Integrated with public health reporting ● Integrated with claims and
Interoperable Records ● Integrated with long-term post-acute care records
adminisirative data Collaborative Cate ●
Acute
● Long-term
Collaborative Care
● Primary care
● Chiropractic
● Specialty
● Dentistry
● ●
Alternative medicine Home
post-acute care
Corruminity Support ● Onlinecommunity support forums and resources for all care team members
Aligned. Emerging Meaningful Use
Aligned: Meaningful Use 1
Aligned: Meaningful Use 2
o
9
Partner With Me
e-Tools
● Opiiontoshare
● Email customer service ● Refill a prescription
● Care plan
Infomution. Way-finding and Quality
V
e-Tools
● Pregnancytracking
Forms: Printable
t- #
other heatth^re oeganizations
● Health ertcyclopedia ♦ Wellness guidance Preverwion
Empower Me
Aligned: Meaningful Use 3
HIMSS, Healthcare Information Management and Systems Society; HiPAA, Health Insurance Portability and Accountability Act.
a Caregivers 3 Family
® Clergy 3 CouRselirtg
3 Friends
3 Servkes
Aliepted: Meaningful Use i*
O
POINT-OF-CARE TECHNOLOGY
Computers in SUSAN MCBRIDE AND RICHARD E. GILDER
OBJECTIVES ●
Discuss the basics of computer technology related to hardware, software, and networking.
●
Describe hardware specifications and criteria to implement health information technology.
Examine the ergonomic requirements needed when implementing health information technology (HIT) for nurses and other healthcare professionals. ● Identify and describe various programminglanguagesutilized in the healthcare setting. ●
●
Analyze various types of software utilized in healthcare settings relevant to HIT, including functionality, usability, human factors considerations, configuration languages, modules versus full platforms, and other important considerations when purchasing software applications.
CONTENTS INTRODUCTION
122
BACKGROUND
122
Historical Perspective
122
The Role of Computers in Healthcare REVIEW OFTHE BASICS
123
127
Bits, Bytes, and How They All FitTogether Key Terms in HIT
127
143
CASE STUDY 1: LESSONS LEARNED: ERGONOMIC-HUMAN FACTORS IN BARCODE SCANNING
144
CASE STUDY 2: EFFECTIVE SELECTION AND DEPLOYMENT OF AN ELECTRONIC HEALTH RECORD IN A RURAL LOCAL PUBLIC HEALTH DEPARTMENT 146 SUMMARY
148
EXERCISES AND QUESTIONS FOR CONSIDERATION REFERENCES
148
149
The contributions of Deb McCullough to this chapter in previous editions of this book are ncbmoicdged.
122
I: POINT-OF-CARE TECHNOLOGY
INTRODUCTION
The technology underlying health information systems is an important aspect that advanced practice nurses (APNs) have to understand, as they are the ones who are to lead teams to adopt and implement health information technology (HIT) systems. This
chapter focuses on the "bits and bytes" underlying the commonly used systems in the clinical setting. Software selection, configuration, hardware specifications, programming languages, and other technology specifics important to the selection of systems are discus.sed. Additionally, we cover the ergonomics and usability of systems and how to configure systems that work well for clinicians. BACKGROUND
In healthcare, we are currently in the very early stages of experimentation with many
different concepts in the search for an engine that will enable the powered flight of a fully automated electronic health record (EHR). As with the high mortality rates associated with early experiments in aerospace technology, vast portions of the general body of scientific knowledge have been discovered relating to technological advancements in aerospace while we balance safety with innovative technological advancement. The internal engines in the computer are electronic in nature and similar to an automobile. The automated EHR can be considered an EHR "machine"
with vast potential for good or bad outcomes, depending on its design and use. The by how well we engineer safety
cumulative effect of that outcome can be determined
into its fundamental designs and how well we educate, train, and license operators of the EHR machine. How well we regulate, mandate, and support the interactive behavior of the operators and what the operators do with the machine will determine the profitability and viability of the EHR.
Historical Perspective Computers in healthcare have great potential to solve, and have in some cases already solved, some of the most complex and difficult problems that have always challenged healthcare providers. Healthcare providers over time have discovered that documenting and archiving of their observations of the healthcare delivery process at the point of care, in a manner that was valid, accurate, immediately retrievable, and accurately reproducible, was a critical advancement in healthcare. Standardization of common
terms used to describe and articulate observations related to healthcare facilitated the
storage, retrieval, and communication of healthcare experience, knowledge, wisdom,
and skills, spanning time, cultures, and geography. Nevertheless, if transferred through oral tradition, inscribed on clay tablets, woven into intricate tapestries of knots, or written in glyphs, the intentional collection, archive, and transfer of knowledge have been identified as critical healthcare functions for centuries. The intrinsic value of discovered
knowledge, and especially knowledge of things that ease .suffering, promote health, and significantly assist healing from disease and injury, is not new. It became an expected professional behavior that contributed to that shared understanding through continuous
maintenance of this pooled reservoir of information, which was also a hallmark of the scientific culture of healthcare. The ability to share information beyond the immediate point-of-care delivery enabled great discovery and widespread implementation and adoption of successful healthcare interventions. The peer-reviewed publication process present in general science likewise forms an integral part of the structure of modern
6: COMPUTERS IN HEALTHCARE
healthcare with the ability to support continued scientific discovery and disseminate that knowledge.
The Role of Computers in Healthcare Communication of information is the essence of the role of computers in healthcare and
the primary function of any EHR or machine intelligence. In the following sections on hardware, software, and ergonomics, the computer’s role that the healthcare machine plays in communicating healthcare information among healthcare providers, and how that role impacts the delivery of healthcare at the point of care and beyond is explored. The impact of HIT is growing and accelerating perfectly in line with Moore's Law. Moore's Law, originating in the 1970s, postulates that computing power will double every 2 years {Moore, n.d.). The fundamental purpose of the EHR is to facilitate clinical communication.
Fundamentals of Communication
Successful communication requires that a minimum amount of understandable signal be present in the communication. Discrimination of the correct information in the signal, even when embedded and surrounded by noise, is a minimum characteristic of the true and accurate transfer of knowledge when communication occurs. Every medium that one human has ever used to communicate with another human, from hand gesture, eye contact, facial expression, spoken words, symbols scribed in sand or on rock, or even written by hand on parchment, is subject to noise. Ink can become smudged, symbols chiseled into rock can become eroded, even facial expressions and nonverbal gestures can be misinterpreted entirely. The.se examples illustrate the concept of signal versus noise. If gold is the money of kings and silver is the money of gentlemen, then signal (as
opposed to noise) is the money of researchers who discover vast riches of information in the signal stream that is generated in huge quantities of validated health care data before during and after the process of delivering care. Healthcare information is valuable when it is valid, resulting directly from the process of care delivery. The cleaner and less distorted the signal, the more valuable it is. With every communication medium, successful communication of the intended
signal is always at risk of misinterpretation or outright loss due to noise and distortion. A distorted signal is not neces.sarily noise. It can be a valid signal, but distortion introduces uncertainty, requiring verification and clarification. Actual noise, however, is an artifact like the static produced by lightning from a nearby thunderstorm during an AM radio broadcast. The radio signal may be di.storted. The signal goes up and down in Doppler tone due to the shifting frequency of the broadcast, but the information remains constant. If one changes the dial to accommodate the shift, the static simply overshadows the signal (i.e., a proportion of signal to noise is weaker). Distortion and noise are two distinct concepts, both of critical importance in informatics science.
When the ratio of signal to noise is very high, successful communication is very high; when the ratio of signal to noise is very low, successful communication is very low. The accurate transfer of a streaming flow of information from a healthcare provider at the point of care to an EHR in a handoff to the machine, and a return handoff from the machine to the healthcare provider engaged in providing point-of-care healthcare, requires an optimal signal-to-noise ratio with as little distortion as feasible. As a "machine," the computer has great potential to deliver even larger and more complex streams of flowing information in packets routed to wide area networks (WANs)
123
124
II: POINT-OF-CARE TECHNOLOGY
that potentially encompass the entire world. The potential to do that far more rapidly than the human recipient can assimilate and process, to stay afloat on the stream (rather than drown in it), poses problems that are addressed by the cognitive sciences that form part of the fundamental foundation of informatics science. Consider the following scenario: attempting to carry on two telephone conversations with two different people simultaneously. This could be done by holding two telephone handsets, one to each ear,
and talking to both parties simultaneously by rotating the mouthpiece up over the head while talking on the left phone or the right phone. By doing this an individual attempts to listen to the feedback in both earpieces of the handsets at the same time. This scenario represents the overwhelming complexity of being overloaded by incoming signals. Sound confusing? In this scenario, what would be the potential for misunderstanding or
mishandling critical pieces of information? While a bit silly to consider, this scenario is an example of just one characteristic behavior of the EHRand the capacity to overwhelm
the end user with too much information or too many alerts when using the EHR during clinical care delivery. This is particularly relevant when we consider other point-of-care devices coupled with the EHRs compounding the signal-to-noise ratio. A military aircraft is quite capable of operating in a very different manner from what the human body is capable of interpreting and proce.ssing. It can execute turns so rapidly and at such high speeds that even though the aircraft remains intact, and the wings do not tear themselves off during the maneuver, the human pilot—if exposed to a lethal level of artificial gravitational forces from centrifugal motion—would not live through the maneuver even though the aircraft would. Of course, safeguards are built into the aircraft to prevent this from happening accidentally. In contrast, the EHR is still at an early stage of development, similar to the first flight at Kitty Hawk, despite significant technological advances in hardware, software, and universalized interface standards.
There have been significant improvements in the EHR that have taken place over the past 10 years. Yet, each improvement introduces new unintended consequences that
have even greater potential for harm, more rapidly and on a wider scale than without them, if ignored. Not all "improvements" turn out to be "good" either. The marketplace, liability, and the law of consequences are what determine utilization and adoption. These improvements, the current state of EHRs, and challenges with new development will be discussed fully in Chapter 7.
When discussing computers in healthcare, an important consideration is that we need to be certain that we do not overwhelm the clinical
teams using the EHR to the
point that preventable lethal results occur unintentionally. In life in general, accidental lethal results are unavoidable. Mitigation of lethality is the focus of prevention, and that is a key component of patient safety that should be embedded as firmware in all healthcare informatics systems, regardless of application such as the EHR, Radiology, Lab, or Pharmacy information system. (Firmware is a level of software computing that
controls the computer and will be discussed and differentiated later in the chapter.) We have a very long way to go to make the EHR as safe and effective as even the common aspirin tablet is today. Aspirin is perhaps one of the deadliest poisons commonly found in any home, yet it is one of the most beneficial medications to keep in store as a critical item in every home. In the future, beyond doubt, there unll be an EHR that enjoys a similar long historical record of being safe and effective when used as directed, but the exact date of that future day and time remains elusive at present. As of 2021, many great strides toward universal EHR standardized improvements in patient safety have been made. These standardized universal improvements are in areas such as engineering design, legislative guidelines with damages and punitive damages, increases
6: COMPUTERS IN HEALTHCARE
in specific accountability for preventable errors, and an increase in public awareness and expectations. One final historical comparison worth considering is the stethoscope and the EHR. In the 1800s, physicians reportedly considered the stethoscope to be an annoying piece of technology with little use in medical practice. The quote here sounds very similar to what we hear clinicians say about current EHRs, yet this quote relates to the stethoscope. "That it will ever come into general use, notwithstanding its value, I am extremely doubtful; because its beneficial application requires much time and gives a good deal of trouble both to the patient and the practitioj-ier; and because its whole hue and character is foreign, and opposed to all our habits and associations." (Forbes, 1823, pp. 13-14) It has been observed that if one were to automate a
bad process, then one would
have a bad process that runs automatically. This speaks to the importance of qualityimprovement tools, such as workflow redesign, and a focus on improved processes being so crucial to the refinement of computers in healthcare. We cannot simply automate a paper-based process with potential flaws, but we must redesign, rethink, and improve processes as we automate them with computers in healthcare. This is perhaps the most important cautionary aspect of computers in healthcare. No doubt the technological advances that made air travel possible have changed the world in ways that none of the first aviators or aircraft inventors could have ever imagined. On the other hand, traveling at such great speed has also resulted in huge disasters and loss of life and property. International airline flight has led to the potential spread of global pandemics in a much faster and irreversible manner than would have ever been possible if we had to simply walk from one place to another, or pull a cart behind an animal, or even sail the high seas across continents. Through trial and error, experiment and observation, air travel has become progressively safe and highly reliable. Many versions of EHRs and the machines used to manufacture and distribute them are currently being subjected to the same scientific methods that eventually resulted in modern commercial air travel becoming a common universally available service. The same classic method of posing research questions and formulating hypotheses designed to objectively and mathematically test the hypotheses through trial and error is being employed. Studies are being conducted that evaluate EHRs and the effects on patient populations' mortality rates, readmission rates, and quality of life. Reporting of the findings to a jury of peers through professional journals and publications is no different in informatics science than any other branch of science. What the aerospace sciences have done for air travel, healthcare informatics sciences are in the process of doing for
healthcare delivery. Yet, in comparison, healthcare informatics is in the very early stages of scientific discovery. Healthcare Informatics—An Evolving Science
Healthcare informatics science constitutes a bridge between the highly technical
computer, data, mathematical, and communication sciences, and the human aspects
of cognitive, biological, medicine, nursing, and other supporting healthcare sciences. Healthcare informatics science plays a critical role wherever computers are involved in the process of delivering healthcare.
Healthcare informatics has evolved and is now rapidly expanding into many
specialized roles required by various healthcare service lines. Medical and surgical doctors have developed expertise as medical and surgical informatics scientists, as have nurses.
125
126
il: POINT-OF-CARE TECHNOLOGY
pharmacists, and other healthcare professionals. For example, perioperative surgical nursing was one of the first among the nursing profession specialty organizations to invent, develop, and adopt a standardized codified vocabulary to incorporate it into the EHR as a method of standardized documentation. These developments have contributed to an ability to better address in perioperative clinical settings research, patient safety, and quality improvement (Kleinbeck, 1999). In the background of this discussion of computers in healthcare, consider that if communication is the essential function of the EHR
''machine," then the machine plays a very strong hand in the game of automated clinical decision support (CDS). The logic modules and computational algorithms that operationalize healthcare logic into a cascade of triggers, warnings, messages, and alarms are the nature of most communication constituting CDS. The differentiation of clinical informatics science is being driven by the healthcare service line. The specific needs of a service line require that the communication input and output that are mediated by the machine are clinically meaningful and useful
to the healthcare provider who must act upon it. Messages that are clinically meaningful and useful to nurses may not be as useful to pathologists, radiologists, or pharmacists. On the other hand, some of the clinically meaningful information to nursing may be of critical value to other service lines and vice versa. Building the internal mapping between critical data elements that are of universal use to every end user (including the patient) of the EHR machine's output is a major challenge for clinical healthcare informatics
science. The map must be wide enough to account for rare and isolated expressions of data elements as a unique field having only one value in the database. Yet, it must be sufficiently deep enough to accommodate multiple expressions of the same data element value no matter how many records deep it reairs. The resulting map is what makes it possible for programmatic algorithms, logic, and rules to be developed that apply to all healthcare delivery regardless of point of care. This ideal type of map is also sufficiently flexible to provide automated decision support in a context of care that occurs very rarely,
but that may result in catastrophic failure if no supplemental support is provided at all to move the clinician's tacit knowledge into action. The ideal automated CDS system will enable anyone who interacts with it to make competent decisions, but will never replace the tacit "expert" knowledge, wisdom, and insight that require many long years
of experience in professional practice to attain. Asa final thought to consider, how do we resolve what can best be described as the "Ansatz conflict"?
The Ansatz, as used in phy.sics and mathematics, refers to that initial hunch, educated guess, guesstimate, or "best bet, given the odds" that precedes the research question. It is an assumption or a guess that works well to .support a given conclusion, with the given facts available at the time the conclusion is drawn.
It is a conclusion, not a theory. It is
the precursor to legitimate research questions. The human cognitive ability to accurately form vague hunches rapidly from intuitive insights based on an internal body of implicit knowledge that has been neiirologically stored over a lifetime of active and passive learning is a definitive a.spect of humanity. Formalization of human intuition, insight, and codification of logic in mathematical systems of reasoning have resulted in the ability to document and reproduce the steps
required to navigate the problem space. In the process of exhaustively enumerating
and analyzing the many paths that lead to a definitive solution set, it has become clear that some of the solution sets are paths that are unique, rare, much shorter, and just as accurate and valid as those that are much longer. Testing hunches before acting on hunches in life-or-death decisions with uncertain outcomes through the proxy of hypothesis testing became less of a gamble on pure chance when the outcomes of the hypothesis tests were collectively applied to the decision as a weight on the
6: COMPUTERS IN HEALTHCARE
hunch. Methods of assigning a numerical value to uncertainty, as a way to predict the otherwise unknowable outcomes of high-stakes games, were the first stirrings of the concept of probability. Probability can be considered the formal mathematical definition
for hypothesis tests, with numerical thresholds of acceptance or rejection that would yield a simple "yes" (or "no") decision. This same approach is used with computers in healthcare to utilize codified logic to assign a probability to uncertainty exhaustively. With little doubt, the healthcare industry is impacted by the advent and advances with this new age of automated decision support. It is very well understood that CDS is mechanically providing a "best guess" that the clinician con.siders when making a decision. The "Ansatz conflict" arises when the suggested action from the CDS "best guess—Ansatz" conflicts with the Ansatz arising from the experienced clinician's gestalt tacit knowledge. There is a genuine risk that the experienced clinician, whose Ansatz is orders of magnitude more accurate and appropriate than the computer CDS Ansatz, will suppress and censor their own best "conclusion" (which results in a clinical decision), in favor of an automated decision. This decision often turns out to be only second best when examined in retrospect (not
always, but often) by a jury of experienced peers. Until CDS is capable of exceeding the judgment rendered by a jury of experienced peers in retrospect, widespread use and
implementation of CDS in any point-of-care technology must necessarily proceed with great caution to ensure the highest standards of safety for patient and provider alike. The true test of how well any given CDS system or algorithm works is found in the false positive rate and false-negative rate (sensitivity and specificity) of its performance after the fact. We examine in more detail in Chapter 19 the best practices for implementing and
designing CDS systems to address some of these challenges. Before we fully redesign processes related to the use of computers in healthcare, it is important to understand the fundamentals on which these machines process and communicate information. In the following section, we review tho.se basics. REVIEW OFTHE BASICS
The basics of computer technology encompass fundamental building blocks. We examine some of the basic definitions of how computers function and the origin of some of the terms we use in computer science to describe technology, including the terms "bits," "bytes," "hardware," and "software," and conclude with connectivity considerations.
Bits, Bytes, and How They All Fit Together In the early 1980s, there was a new technical jargon, a new and special language that was an anthropological consequence of the human-machine interaction around computers. The term "computerese" appears in the title of an article written at that time by Lietzke (1982) to provide a glossary of the new words and terms that were being used to conceptually define unique attributes of computer hardware, software, human-computer interactions, and the common usage of these terms in a ubiquitous jargon. Lietzke clarified and stabilized the modern primary definition, whereas Tukey (Shannon, 1948) coined the first use of the term as "professional slang" (jargon). Tukey's usage was an idiomatic and linguistic drift at the time of first use, to describe two distinct concepts of "binary" and "digit," "bit-digit or bit." Tukey's concept of "binary digit" (bit) was indeed the root concept. Lietzke finalized the concept that data can be "counted" representationally by the number of "binary digits" in the data. Lietzke added the concept of "unit of measure' to Tukey's existing "binary digit." After that clarification in 1984, the use of
127
128
II: POiNT-OF-CARE TECHNOLOGY
"bit" to acairately describe computational architecture as a standard unit of measure (8 bit, 16 bit, 32 bit, 64 bit, 128 bit, 256 bit, chip sets, etc.) cemented the definition as a unit of measure (Dickey, 1984). The bit has only two possible values, and the two values it is allowed to have are the "digits" zero (0) or one (1), hence the term "binary" in the phrase "binary digit." If 1 is always taken to mean "on” (as opposed to "off"), and if 0 is always taken to mean "off" (as opposed to "on"), then the bit is universally understood to represent mutually exclusive values. The switch is either in the "on position" or in the "off position." Although in human experience, it is quite possible to be completely drenched from a water hose when the valve is "half open," this cannot be the case with allowing this
behavior of bits in a computer. This capability would have prevented the next phase of growth and development of computational algorithms based on the ability to form bits into special aggregations of eight bits, which form the set of 256 unique combinations of bits, each known as a "byte." In a 1958 article published in the American Mathematical Monthly, Tukey is attributed as the first to define the programs that electronic calculators ran on, describing them as "software." Tukey differentiated the code from the "hardware," including the tubes, transistors, wires, tapes, and the like (Leonhardt, 2000). The intentionally misspelled term "byte" was invented by Werner Buchholz in 1956, spelled with a "y" to prevent confusion of "bite" with "bit" (Buchholz & Berner, 1962). The byte is the basic unit of information, roughly corresponding with letters and symbols in an alphabet. It can be used to form the written codex of a phonetic language with unique codes that can be assigned to retrievable computer data archives known as the "memory." With its 256 possible unique symbolic alphabetic values formed by the natural frequency of 2 to the eighth power (28) possible combinations of zeros or ones in any of the eight allowed positions, the byte forms the computational alphabet of machine language. The difference between the base-10 system of measurement and the base-2 system of measurement has historically always been a source of confusion because of the similarity between the two. The byte also serves as a scale measure of data size, often classified by orders of magnitude in units of exponential power. The binary (base-2) prefix terms (kilo, mega, giga, etc.) form the basic unit of measure for data file size and are exactly the same as the decimal (base-10) prefix names already in use for labeling decimal quanta. Binary has its own special quanta names based on the fundamental radix (ba.se) of 1,024 bytes of data information. The byte, therefore, represents alphanumeric symbols and text characters, the smallest standard unit of data information that can be archived, transmitted, or manipulated, and that can be reliably retrieved from memory, regardless of the media used for memory archive. At the smallest level, it takes only one bit of data to discriminate true (1) and false (0). In the world of data storage and retrieval, it requires two bytes (16 bits) to define one word composed of alphanumeric text symbols. At that rate of exchange, one kilobyte of data (1,024 bytes) can contain 512 words (Yuri, n.d.). The comparisons of decimal to binary prefix terms are illustrated in Table 6.1.
The Joint Electron Device Engineering Council (JEDEC) and the International Electrotechnical Commission (lEC) set standard definitions that are shown alongside the decimal definitions by their order of exponential magnitude in Table 6.1. The lEC is one of the organizations recognized and entmsted by the World Trade Organization (WTO) to monitor the national and regional organizations that agree to use the lEC's international standards (lEC, 2014). The JEDEC is an independent .semiconductor engineering trade association that sets standards for terms, definitions, letter symbols for microcomputers, microprocessors, and memory-integrated circuits. The purpose of the standard is to
TABLE 6.1 Comparison of Decimal to Binary PrefixTerms DECIMAL
EXPONENTIAL ORDER OF MAGNITUDE
RADIX
BYTES
SCIENTIFIC NOTATION
SYMBOL
NAME
BASE
1
1000
1,000
1.000E+03
kB
kilobyte
2
1000
1,000,000
1.000E+06
MB
megabyte
GB
gigabyte
3
1000
1,000,000,000
1.000E+09
4
1000
1,000,000,000,000
1.000E+12
TB
terabyte
5
1000
1,000,000,000,000,000
1.000E+15
PB
petabyte exabyte
6
1000
1,000,000,000,000,000,000
1.000E+18
EB
7
1000
1,000,000,000,000,000,000,000
1.000E+21
ZB
zettabyte
8
1000
1,000,000,000,000,000,000,000,000
1.000E+24
YB
yottabyte
EXPONENTIAL ORDER OF
BINARY lEC NAME
SCIENTIFIC
JEDEC*
JEDEC
lEC**
NOTATION
SYMBOL
NAME
SYMBOL
1,024
1.024E+03
KB
kilobyte
KiB
kibibyte
1,048,576
1.049E+06
MB
megabyte
MiB
mebibyte
GB
gigabyte
GiB
gibibyte
RADIX BASE
BYTES
1
1024
2
1024
MAGNITUDE
3
1024
1,073,741,824
1.074E+09
4
1024
1,099,511,627,776
1.100E+12
TiB
tebibyte
5
1024
1,125,899,906,842,620
1.126E+15
PiB
pebibyte
6
1024
1,152,921,504,606,850,000
1.153E+18
EiB
exbibyte
7
1024
1,180,591,620,717,410,000.000
1.181E+21
ZiB
zebibyte
8
1024
1,208,925,819,614,630,000,000,000
1.209E+24
YiB
yobibyte
o-
n
O *D
c m
z
*JEDEC {Joint Electron Device Engineering Council) is an independent semiconductor engineering trade organization and standardization body. lEC (International Electrotechnical Commission) is an international standards organization that prepares and publishes international standards for all electrotechnology." **
I
> I
n > 73
130
II: POINT-OF-CARE TECHNOLOGY
promote the uniform use of symbols, abbreviations, terms, and definitions throughout the semiconductor industry (JEDEC, 2002). Table 6.1 illustrates the dramatic difference in the maximum-size increase in bytes at each stepwise order of magnitude increase, due to the difference in the radix (base) size being acted on by the order (exponent) of magnitude. The standard radix for binary data is 1024. Hardware Considerations
Hardware includes many devicesin the healthcare industry, including system peripherals and aspects of the computer that make up the system's physical components. Figure 6.1 reflects a visual of a typical hardware setup for a personal computing (PC) system. Within the computer, there are important elements, including the motherboard, central processing unit (CPU), random access memory (RAM), power supply, video card, hard disk drive (HDD), solid-state drive (SSD), optical drive (e.g., BD [Blu-ray disc] / DVD /CD drive), and card reader (SD/SDHC, CF, etc.). Note that most computers no longer come
with disk drives or card readers. Table 6.2 defines each of these elements and describes
their basic function. When we purchase any PC for use in healthcare, we have to pay special attention to the components of the system, particularly concerning the processing speed, storage size, and the type of graphic interfaces needed based on the requirement of the system. We also have to consider aspects of how and when that piece of equipment
will be used for patient care.
Hardware Selection and Specifications
Decision-making regarding the right hardware for implementation involves the interprofessional team and should include management, users, and systems analysts in the information technology (IT) sector. The vendors provide information on their specific hardware components. Still, the systems analyst within the IT department frequently FIGURE 6.1 Basic Hardware Configurations.
o
6: COMPUTERS IN HEALTHCARE
131
TABLE 6.2 Internal Components of the Computer COMPONENT
DEFINITION
FUNCTION
Motherboard
The backbone of the computer
Connects all of the parts of the computer together
Central
Often thought of as the "brains" of the computer
Responsible for interpreting and executing most of the commands from the computer's hardware and
processing unit (CPU)
software
Graphical processing unit (GPU)
Parallel processing unit that can be configured as a dedicated math processor (instead of exclusively graphics)
Systems with this additional GPU mathcoprocessor hardware are considered "embarrassingly parallel" and significantly faster with added expense
Random access
The working memory of the
memory (RAM)
computer
Allows a computer to work with more information at the same time in active memory processing
Power supply
A converter that supplies the power to the machine
Used to convert the power provided from the outlet into usable power forthe many parts inside the computer case
Video card
Graphics adapter or expansion card
Hard disk drive
(HDD)
Data storage device and an electromechanical magnetic disk drive
Allows the computer to send graphical information to a video display device such as a monitor,TV, or projector The HDD is the main, and usually largest, data storage hardware device in a computer where the operating system, software, and most files are stored
Solid state
drive (SSD)
Optical drive (e.g., Blueray/ DVD/CD drive)
Data storage device; no moving (mechanical) components
Storage device that is typically more resistant to physical shock, runs silently, and has lower access time and less latency, but more expensive than HDD
Optical storage devices
Optical drives retrieve and/or store data on optical disks like CDs, DVDs, and Blu-ray disks (BDs)
recommends hardware options for the team to consider and provides information on which the team can base an informed purchasing decision. Depending on the varied
options, hardware will typically have disadvantages and advantages, which should be considered before purchase. The workload for the system, the purpose of the hardware, and end-user interface needs are criteria to be considered, and, most important in the clinical setting, the answer to the question "How and where will that hardware be used to deliver and support patient care?" is very crucial. The end-user and workflow considerations for the hardware components are critical as well. We cover workflow
redesign in Chapter 9. Figure 6.2 is a schematic that reflects the interaction points for the sociotechnical interaction of the end-user, the application layer, the operating system, and the hardware. In this chapter, we focus on some of the issues and considerations that are relevant to healthcare concerning these interaction points. Human-computer interaction (HCI) is considered a subcategory within the larger field of human factors science. Human Factors and System Ergonomics
Human factors is a term defined as "the field of study focused on understanding human elements of systems, in which 'systems' may be defined as software, medical devices, computer technology, and organizations" (Kushniuk & Borycki, 2013, p. 368). Figure 6.2 reflects these interaction points of the end user (human) and the software and hardware within the healthcare organization. Ergonomics is defined by Wilson (2013) using a
132
II: POINT-OF-CARE TECHNOLOGY
FIGURE 6.2 Sociotechnical
Interaction: Hardware, Software, and Human Factors.
systematic approach, which the authors believe is particularly relevant to healthcare and HCI. Wilson defines ergonomics in terms of human factors as follows: "Ergonomics/ human factors is, above anything else, a systems discipline and profession, applying a systems philosophy and systems approach" (Wilson, 2013, p. 5). Other aspects that also apply to the field of human factors include the field of cognitive science and how the end user interfaces and interacts with the software, equipment, and the graphic user interface (GUI). In addition, the field of ergonomics and human factors science involves
a workflow within the system, which is covered in Chapter 9. Hardware and Infection-Control Issues
One crucial factor that is unique to healthcare relates to infection control and the possible transmission of microbes via the hardware within the clinical setting. This specific concern is an excellent example of how HCI, human factors, and the environment impact patient safety and quality. As the number of devices increases, the possibility for infection to be transmitted through direct patient contact and use of the hardware for patient care compounds. Neel and Sittig (2002) reviewed the literature to determine possible links to nosocomial infections transmitted through the colonization of microbes on hardware acting as a vector for the infections. These findings indicate how healthcare professionals can be cognizant of the mode of transmission to design interventions to potentially prevent infection when viewing the computer hardware as a potential vector for infection. The review of the literature indicates that there is significant opportunity for some serious transmission of infection, and of particular concern are the findings relating to methicillin-resistant Staphylococcus aureus (MRSA). MRSA was directly correlated to the keyboard, with MRSA colonization and MRSA infection in at least two intensive care unit (ICU) patients using pulse-field gel electrophoresis to deternrine that the isolates were of the same genus and species. Neel and Sittig indicate that there are steps that can be taken to prevent the transmission of infection by keeping in mind the route of transmission. They recommend knowing the risk factors that predispose a patient to serious infection, following strict handwashing protocols, using gloves as appropriate, and working with infection-control professionals to develop and adhere to policies for cleaning and decontamination of all hardware. Finally, these two researchers advise us
6: COMPUTERS IN HEALTHCARE
to think in terms of the interaction among the patient, hardware, and clinician, and how that piece of equipment is used within the workflow to identify instances that might transmit infection given the sociotechnical interaction reflected in Figure 6.2 and the chain of infection depicted in Figure 6.3. FIGURE 6.3 Steps Potentially Leading to Infection and Basic Infection-Control Interventions Used to Decrease the Risk of Infection.
colonization —> infection
Source: Neel, A. N., & Sittig, D, F (2002). Basic microbiologic and infection control information to reduce the potential transmission of pathogens to patients via computer hardware. Journal of the American Medical Informatics Association, 9(5), 500-508. https://doi.org/10.1197/jamia .M1082. Reprinted with permission from Oxford University Press.
The issues reflected in this case relate largely to the ergonomic issues involved with comfort, ease of use, and practical functionality when it comes to the effects of hardware, software, and peripherals on workplace health. Detrimental workplace ergonomic designs characteristically exhibit an absence of the comfortable and healthy attributes of the high-quality ergonomic designs that take human factors into account. The assessment of ergonomic quality surrounding the HCI is a core clinical healthcare informatics skill, especially in the modern clinical work environment that relies on the computer-enabled clinical informatics world of CDS.
Hardware Specifications
Performance of the system and configuration of hardware are also important points for consideration, including factors such as the time required per transaction as well as data input and output by the end user; how the hardware should be configured given its purpose is yet another point for consideration. The volunre of information that will be managed, how much data can be processed before the system has reached maximum capacity, and its use in the current state with the anticipation of future growth should be taken into consideration. These types of decision points are discussed in more depth in Chapter 8 with respect to hardware (Kendall & Kendall, 2014).
133
134
II: POINT-OF-CARE TECHNOLOGY
The
considerations
noted
here
take
into
account
hardware
specifications— specifications that typically include parameters such as processing speeds, memory requirements (RAM and HDD requirements), necessary interface equipment, and operating system requirements for mnning the clinical software. Table 6.3 presents a sample of what a technical specification for a clinical information system setup might look like for a clinical environment application server. TABLE 6.3 Sample Recommended Specifications for a Clinical Information Server PROCESSOR
2 X Six Core Xeon E5-2620 or higher
MEMORY
8 GB RAM
NETWORK INTERFACE CARD
1 Gbit
PRIMARY HARD DRIVE(S)
2 X 300 GB SAS in RAID 1
REPOSITORY HARD DRIVES
6 X 300 GB SAS or more in RAiD 5
OPERATING SYSTEM
Microsoft Windows 2008 R2 (64-bit) Standard Edition SP1
DATABASE SOFTWARE
Microsoft SQL 2008
SQL, structured query language. Source: Philips. (2014). Hardware specifications. https://www.ph ilips.ca/healthcare/product/HCNOCTN198/ intellispace-cardiovascular.
Software Considerations
Software is a vast topic that has many options available, and the number of available options depends on the category. Those considerations may also take into account proprietary software or open-source options in the form of free ware. This section discusses the different types of software and the considerations for selecting the correct kind of software.
Programming Language Classifications
Programming is a mechanism for transforming information into a computer in the form of machine code, which instructs the computer to do some type of task. According to Ogala (2020), programming languages are a set of English-like instructions that include sets of rules, referred to as "syntax" for putting the instructions together to create what programmers refer to as "commands."
A programming language is a formal computer language, or a "constructed language" designed to communicate instructions to a computer. Programming languages can be used to create software programs, to control the behavior of a machine or to express algorithms. A programming language can be classified into different levels. Machine code programming language is a collection of binary digits or bits that the computer reads and interprets, sometimes referred to as object code. Machine language is the only language a computer is capable of understanding. Machine code is the primary language for the computer and consists of binary Os and Is, and it is considered "first-generation" (IGL) machine language. It is also referred to as "low language."
The second-generation level (2GL) is a step higher and constitutes assembly languages that use reserved words and symbols that have special and unique meanings. It is considered a low-level language similar to machine language but uses symbolic
6: COMPUTERS IN HEALTHCARE
operation code to represent the machine operation code. The assembly code is specific to the machines, including the computer and CPU (Janssen, 2014c). Third-generation languages (3GL) were intended to be easier to use, and higher-level languages provided a programmer-friendly language. 3GL works by taking high-level source code containing easy-to-read syntax that is later converted into a lower-level language. This lower-level language can be recognized and run by a specific CPU. Some examples of this type of code include FORTRAN, BASIC, Pascal, and the C-family (C, C+, and C++; Janssen, 2014d).
Fourth-generation programming languages (4GL) are more in line with the "human language" and therefore easier to work with than 3GL. 4GL is considered a domainspecific and high-productivity language and includes database queries and report generators, GUI creators, database programming, and scripts. Many of the 4GL are dataoriented and use structured query language (SQL) developed by IBM and also adopted by the American National Standards Institute (ANSI; see Chapter 7 on data standards; Janssen, 2014b).
Fifth-generation languages (5GL) utilize visual tools to support programming. One such frequently used language is Visual Ba.sic. Further, some consider 5GL to be a type of constraint logic or problem solving-based programming. PROLOG is a programming language that fits into this description (Janssen, 2014a). Fifth generation (5GL) languages are based upon coded syntax requiring internal mathematical and logical problem solving as part of the program constraints. This type of programming language reduces problem solutions to an automated stored procedure within the software. One example that is a 5GL language is a "macro." For example, a macro is a programmed script that the end-user triggers to run at a specified time. Other examples of 5GL programming may include messages, alerts, and automatic email reminders. Advanced Programming Languages and Advances IN Big Data Computing:The Black Box
Advances in programming languages to higher generations enable computer power and advanced combinations of codes, algorithms, natural language processing, and lower-level programming languages to create new and emerging technologies. These higher-level programming languages are taking healthcare into advanced computing methods that enable emerging technologies such as advanced analytics, robotics, and nanotechnology powered by machine learning, neural networks, and what some call "the black box." Rasouli et al. (2020) describe advanced programming languages, including 5GL and 6GL programming as "the black box." They define this phenomenon of a black box because it is difficult to fully understand what goes into the programming that results in the answer. With advanced analytics, these programming languages are associated with software that is used to analyze massive amounts of healthcare data and predictive analytics. Artificial intelligence (AI) and robotics will be discussed in Chapter 26, big data and advanced analytic methods will be discussed further in Chapter 27, but the focus in this chapter is foundational to understand how computer-programming languages in healthcare can enable new and emerging technologies.
In computer programming, a 6GPL is a higher-level programming language with
extreme abstraction from the hardware. There are intricacies with 6GLbecause a command
interpreter must analyze a set of human-readable instructions. These 6GL programming languages may be domain specific or general purpose and often apply natural language processing in order to function. Nils John Nilsson (2013), a founding researcher in the
135
136
II: POINT-OF-CARE TECHNOLOGY
field of AI, describes AI as the ability to support a learning machine or any device for v/hich its actions are influenced by previous experiences. Tliese 6GL languages require a "command interpreter." They are not necessarily "installed" or even hard wired into the local machine hardware that the end-user is using to execute the language. When considering how 6GL languages work we can think in terms of the role of a human acting as a translator listening to two people that do not speak the same language (one English, the other ancient Egyptian hieroglyphics), interpreting and communicating accurately the conversational verbal content between the two (or more) parties and in such a manner that ensures that the semiotic communication triangle is completed. This scenario represents what a command interpreter (and a human interpreter) does. The net results are that all parties involved in the communication triangle clearly understand the information streamed and transferred or propagated back and forth. One hallmark of 6GL is that verbal speech recognition (spoken commands) is a feature of the interface. Most modern cell phones can interpret verbally communicated keywords to execute menu commands such as dialing or performing an internet search query. 6GL can include optical recognition of hand gestures, facial expressions, or kinetic input (sweep, pinch, spread, swipe, etc.). Speech to text, scanning a QR code, optical character recognition (OCR) all rely on a command interpreter to translate the external input (using sophisticated decompiler data dictionaries) into a very low-level language that the CPU can execute to produce the expected output action from the very high-level user input command. As of this writing (June 2021), 7GPL programming languages are ambiguously defined. A world-wide unconstrained (i.e., adult filters, etc. OFF) Google search for the exact term "seventh generation computer languages," yields a few non-specific results whereas the term "sixth generation computer languages" yields specific 6GPL results that are too numerous to count. A 10-year search of PubMed abstracts of all published
literature returns "No results found" for the specific hexadecimal string text search terms "GPL7,",or "...seventh-generation programming language...." Hence, it is asserted that at the time of this literature review (June 2, 2021) these terms are not airrently defined in the published National Library of Medicine literature of record. Knowledge representation (KR), is the codification of human cognition and sentience into an automated symbolic, glyphic and reasoning command interpreter that allows an AI agent and a human being agent to interact with each other and possibly learn from each other's simultaneous duplex KR transactions. The potential for learning (with learning defined as the acquisition of new information signal knowledge) is bi-directional and mutually inclusive as well as didactic and dynamic. Probably the best example of a ubiquitous and annoying software application type that represents seventh generation languages (as yet ambiguously defined) are the infamous "chat bots." Robotic automated telephone calls, using synthesized speech, routinely call personal cell, business, and home telephone numbers, and attempt to engage with the person who answers the call using highly sophisticated multiband acoustic vox humana digital surrogate voice to flawlessly imitate pleasant human speech, and have a conversation that results in sales of
a product. In live synchronous chat rooms on web-portals ("chat rooms") fully automated robotic AI entities routinely engage with chat room participants in long synchronous (instant back-and-forth dialogue) text messaging conversations. These conversations may even be intentionally slowed down and include typographical errors in order to perfectly imitate human keyboard interface behaviors, without the real carbon-based life forms (humans) ever becoming aware that they have been typing and messaging back and forth with an AI. Occasionally, the AI responds in a repetitive manner or in a way that reveals that it is not a human being. This highest level of computer programming
6: COMPUTERS IN HEALTHCARE
137
language that allows programmers to write and compile applications like this, depends upon Natural Language Processing taken to a level that involves Sentiment Analysis, Semantic Network Analysis, and Natural Language Understanding (NLU) in an attempt to simulate real human cognition and sentient interaction. In other words, 7GPL algorithms are quite capable of flawlessly passing the "Turing Test" of sentient cognition that is indistinguishable as artificial even by a well-informed and skeptical human being. Keywords, phrases, sentiments, emotions, feelings, language syntax and grammar, idioms and expressions, context-dependent interpretation, spelling, declension, person-number, active-middle-passive voice, nominative, acaisative, dative and genitive cases, past, present, Riture, perfect, and pluperfect tense, and even paradigms of language meanings in common usage have been codified into the 7GPL algorithms in such applications that are apparently already in common worldwide use. Types of Software
Software can be categorized as (a) system software that is used to start and i*un the computer, (b) application software that generally has a purpose or function specific to its use (e.g., accounting/financial applications), or (c) programming tools that are used to compile programs and link or translate computer program source code and libraries that belong to either the system software or the application. Figure 6.2 demonstrates how the software relates to the end user and to the various
types of software. Table 6.4 features
the kinds of system software used across industries. Systems software is related to what the software does within the computer system to support the use of the computer. For example, the device-driver software operates and manages all devices attached to the
computer. Table 6.5 lists the application software typically categorized by intended use, such as business software used to manage admissions, discharge, and transactions for patients in hospitals and healthcare systems. This is not an exhaustive list but instead gives the reader an idea of how software fits within the categories for either systems or applications. Review the table and see whether you can think of other software that might fit into some of these categories, given the definitions and descriptions. TABLE 6.4 Types of Internal Components of Computer Software TYPE
DESCRIPTION
Operating system
The software that is responsible for the direct control and management of the hardware and running application software
Open source
Free source code access licensed for use by an open community of developers and end users; proprietary software is owned and distributed for commercial use
software
Boot loader or
bootstrap
The small program that loads and executes the command to "boot up" the computer; the program is stored in the RAM
Device drivers
A program that operates the various devices on the computer, such as printers and peripherals;the driver provides software interface to the hardware device
Firm ware
Controls the devices typically seen in items such as mobile phones and digital cameras
GUI
A graphics display with user-friendly point-and-click capability that allows the end user to interact with the computer through the mouse and touch pad
Middle ware
Software that resides as an interface between the operating system and the
applications that allow developers to control input/output devices, also referred to as "software glue" Utility software
Software that helps analyze, configure, optimize, or maintain the computer
GUI. graphical user Interface; RAM, random access memory.
138
II: POINT-OF-CARE TECHNOLOGY
TABLE 6.5 Types of Application Software TYPE
DESCRIPTION
Business software
Used by and for specific business functions in healthcare; e.g., admission, discharge, and transactions software components frequently embedded in the EHR as a module component seamlessly interfaced in the background so that data are passively collected as fast as entered or become available
Communications or
messaging software
Used to exchange files and messages between systems remotely; healthcare systems require encryption of data to meet HIPAA regulatory requirements when using communications or messaging-type software
Data-management
Source software with the primary function of managing a database in a particular
software
structure, usually relational or object oriented
Graphics software
A type of software that allows the end user to manipulate graphic images on the computer; usually has the capability to import and export graphics file formats
Simulation software
Software that allows the end user to model real phenomena with a set of mathematical formulas used in healthcare professional training to simulate events rather than have students practice on patients
Gaming or video
Software that uses interaction with a user interface to generate visual feedback on a video device
software
Spreadsheet software
Software that allows data to be analyzed in a tabular format with data organized in rows and columns that can be manipulated by formulas
Word processors
Software that performs processing of text (words) to compose, edit, format, or print written material
Workflow software
Software used to reflect a process or steps within a process that provides functionality to create workflows with a diagram-based graphical designer
Presentation
Software used to create slide presentations that allow typesetting and graphical design to create a professional-looking presentation quickly
approach software
EHR, electronic health record; HIPAA, Health Information Portability and Accountability Act.
Software Selection for an EHR
The EHR is perhaps the most important criterion for software selection in tlie current setup of healthcare organizations, particularly given the emphasis in the United States on creating a national health information network across the country with full interoperability among care settings. The selection of EHR software emphasizes the importance of strategic decisions that align with the organization's overall improvement and business plans.
The basis on which EHRs are selected for a hospital or provider is very specific to that organization, and as such should be both a cost and a quality decision based on the needs of the organization and the end users within the organization. Involving the end user in selecting EHR software is critical to the .success of adoption, implementation, and effective meaningful use (MU) of the EHR software. An entire chapter is dedicated to this topic; however, we note a few essential considerations on EHRs here. Hartley and Jones (2005) outline 12 essential steps that should be completed before purchasing an EHR. These steps are as follows: Step 1: Establish the budget. Step 2: Establish the right team.
Step 3: Engage the team, but be clear on the decision-making lead. Step 4: Prioritize requirements.
Step 5: Assign fact-finding duties and responsibilities to the team members.
6: COMPUTERS IN HEALTHCARE
Step Step Step Step Step Step Step
6: 7; 8: 9: 10: 11: 12:
Develop the request for proposal (RFP). Develop a scorecard to rate the products. Schedule on-site demonstrations. Determine return on investment (ROI). Negotiate the contract. Agree on the purchase plan. Ask for help, if needed. (Hartley & Jones, 2005)
The Office of the National Coordinator (ONC) provides excellent resources to providers, including various tools used to help make software selection easier for organizations with minimal resources. These tools highlight the steps, noted by Hartley and Jones (2005), that providers should follow when selecting and purchasing an EHR. These elements include the following: 1. Contracting templates 2. Demonstration of the software and scenario-based
evaluation of the product, complete with scenarios for hospitals, clinics, and federally qualified healthcare clinics
3. Reference checks on the software with other providers with instructions on how and what to ask
4. RFP templates 5. Evaluation tools based on measurable criteria important to the organization (score card templates) 6. MU comparisons among product tools
7. Pricing and ROI calculators (Hartley & Jones, 2005) The ONC also (2014) provides tools on its website to guide a provider through the process of EHR selection and considerations. In addition, elements, such as an RFP, are discussed in full in Chapter 8. When selecting software for any purpose, attention should be paid to important criteria such as price, function, and end-user requirements; it is also crucial to include a very structured methodical approach to decision-making by involving the entire team in the decision-making process. The ONC, through tools developed primarily by the Regional Extension Centers (RECs), provides organizations with many of these types of tools to help follow this structured approach. Case Study 2 located at the end of this chapter is entitled, "Effective Selection and Deployment of an Electronic Health Record in a Rural Local Public Health Department." It demonstrates how these steps and tools were put to use by a rural health clinic. Networking, Connectivity, and Configuration of Hardware
Networking and connectivity of systems in healthcare comprise diverse areas ranging from radiology medical imaging departments to specialty care units, business administration, and information services. Networking expanded in the 1990s, as the cost of data storage (hardware) and communications technologies became available at lower costs. Healthcare is also diverse in terms of the size and scope of the organizations and communities served from small rural facilities to major networks of healthcare systems distributed across the United States and in some cases internationally. As such, we have a vast array
139
140
II: POINT-OF-CARE TECHNOLOGY
of requirements within the industry to address connecting, networking, and configuring the connectivity of technology to support the healthcare industry. Additionally, emerging trends in cloud computing and mobile technology create expanded capacity; yet we also need to incorporate older architecture as we move into the future. In the following section, we cover older technologies as well as some of the newer configurations for connectivity and remote access. Networking and Communications
Typical networking relates to connections between and among two or more computers. Local area networks (LANs) involve computers and printers connected by wire or radio frequency. In contrast, a WAN is a network that uses high-speed, long-distance communication networks or satellites to connect computers over greater distances than LANs. Ethernet connections are a means of communicating using an architecture that relies on a detection and avoidance protocol for routing data that permits more extensive networks and faster data transmission. The internet is defined asa global computer network that consists of interconnected networks of computers on which websites, resources, and data archives are located. The internet facilitates the communication and exchange of information using standardized protocols of communication among devices. The internet was originally created in 1969 during the Cold War and used by the Department of Defense. Today, the internet is used worldwide by networks of computers now maintained primarily by service providers. Some important internet terms are listed here: ■ FTP is an abbreviation for "file transfer protocol," which is a common means for transferring files within the internet from one computer to another.
■ TCP / IP is an acronym for transmission control protocol / internet protocol. TCP and IP are two protocols developed by the U.S. military for the purpose of allowing computers to communicate over long-distance networks. TCP is a verification mechanism for data packets, whereas IP relates to the actual data packets between
the given nodes of the internet.
■ HTML, or hypertext markup language, is the programming language that most webpages are based on that control the display of information via the web browser. ■ HTTP is an acronym for hypertext transfer protocol and is the means by which data is transferred via the web.
■ HTTPS indicates that the website uses a secure socket
layer (SSL) for security purposes. This is an important consideration for the protection of healthcare information, or other information such as your banking or credit card information. You can determine whether a website is secure by viewing the URL (uniform resource locator) in the address field of your web browser. If the web address starts with "https://" you can be assured that the website has been secured using SSL (TechTerms.com, 2014).
Networks have several different types of physical layouts or typology. These configurations include tree network, star network, ring network, and bus network (Figure 6.4). Network typologies have pros and cons depending on the configuration. Bus configurations are dependent on the total length of the network and the distance the computers are spaced within the network. Total distance, number of computers, and spacing are relevant to the efficiency with a bus configuration. It is confined and cannot be expanded as fully as the other options available. Historically, this was the type of network used mainly on a copper wire and was limited by the distance or size of the network, or the length of the cable between computers, which impacts timing and efficiency.
6; COMPUTERS IN HEALTHCARE
FIGURE 6.4 NetworkTypologies.
Bus network typology
Tree network typology
141
142
II: POINT-OF-CARE TECHNOLOGY
This type of network is an inexpensive setup but obviously has limitations and is seldom used. Star networks are typically connected via a switch or hub, with a limited number of computers on the network. With the tree, one builds off the switch and connects a switch to switch, and this configuration works using the internet. Expansion is available with the tree network typology. Ring networks are set up in a circular configuration with the signals transmitting around the ring until the envelope containing the data, or package of information, finds the designated address. This configuration is set up circularly and it can be difficult to add a computer; when one computer goes down, the entire network goes down, but it is easy to identify the location of the failure in the network (Zanbergen, n.d.). Table 6.6 describes these topologies and their advantages and disadvantages. TABLE 6.6 NetworkTypologies Advantages and Disadvantages TYPE
DESCRIPTION
ADVANTAGES
DISADVANTAGES
Star network
Computers or other devices are
Easy to install and manage
Bottlenecks can occur
connected to a central hub, also
because data pass through the hub
referred to as a switch
Ring
Computers or other devices are connected to one another in the shape of a circular closed loop such that devices connect directly to two other units (one on either side)
Offers high bandwidth and span distances, relatively easy to install, easy to locate points of failure
If one computer goes
Bus
Computers or other devices are
Inexpensive and easy to
Not as stable as
network
connected to a central cable referred to
install
other configurations
Improves network scalability
Expensive to configure and
network
down, the entire network is down
as a bus or backbone Tree network
A tree typology combines characteristics of linear bus and star
topologies consisting of groups of starconfigured workstations connected to a
maintain
linear bus backbone cable
Source: Mitchell, B. (2018). Introduction to computer network topology. Lifewire. https://www,lifewire.com/computernetwork-topology-817884.
Communication Protocols
Communication protocols include TCP/IP, FTP, and simple mail transfer protocol. An FTP is a standard network protocol used to transfer computer files over a network, such
as the internet, and was defined earlier in the Networking and Communications section. This communication protocol has become the industry standard for interconnecting computer hosts, networks, and the internet. One of the most widespread communication protocols today is the use of wireless internet. Protocols, such as Bluetooth, IEEE802.il,
and wireless application protocols, are expanding the growth along with the availability of mobile devices such as our cell phones. Mobile computing is expanding in the
healthcare industry and thus is a very important computing protocol for consideration, which is discussed in more detail in Chapter 16. Cloud Computing
Cloud computing is also referred to as "software as a .service." Although we could have classified this under the software subheading, we elected to place the di.scussion in the hardware section. Cloud computing is often selected because organizations do not want to host their servers, invest in the infrastiucture (hardware), and deal with the security requirements that are becoming very significant (see Chapters 14 and 29 for further details).
6: COMPUTERS IN HEALTHCARE
143
Cloud computing is a solution hosted typically by a company providing its software on a server that can be accessed from the internet; it is considered a new way of accessing services
but not necessarily a new technology (Kuo, 2011). The service model of cloud computing and software as a service is noted as one of the top emerging trends in the Gartner 2014 report titled The Top 10 Strategic Technology Trends for 2014 (Cearley, 2013/2014). Client-Server Architecture
A client-server setup involves different configuration considerations as well, including
loaded to the desktop versus a thick client with software and perhaps some processing applications on the desktop. "Thin" whether the client is a thin client with no software
indicates there is very little computing power on the desktop, whereas "thick" indicates software and computing power to drive the software are loaded on the desktop. Many software applications that we purchase in healthcare have configurations similar to this. Many of our EHRs run on thin client configurations, whereas many of our analytic tools have processing and software running off of the desktop, but we access data off of a server hosted by the organization in a data warehouse setup. Key Terms in HIT This section discusses key terms relevant to today's HIT revolution under the Health Information Technology for Economic and Clinical Health (HITECH) Act, including terms defined in Table 6.7. Common terms frequently used in the healthcare setting are often confused by the average clinician. There are differences in definitions of an EHR versus an electronic medical record (EMR) that are often mistaken. According to the ONC, the EMR is defined as an electronic record of health-related information on an individual that can
be created, gathered, managed, and consulted by authorized clinicians and staff zvithin one healthcare organization. However, an EHR is defined as an electronic record of health-related information on an individual that conforms to nationally recognized interoperability standards and that can be created, managed, and consulted by authorized clinicians and staff across more than one healthcare organization. A third type of record important to and
controlled by the healthcare consumer is the personal health record (PHR). The PHR
is defined as an electronic record of health-related information on an individual that
conforms to nationally recognized interoperability standards and that can be drawn from
multiple sources while being managed, shared, and controlled by the individual.
TABLE 6.7 Definitions of Key Health InformationTechnologyTerms EMR
EHR
PHR
An electronic record of health-related information on an individual that can
An electronic record of health-related information on an individual that
An electronic record of healthrelated information on an individual
conforms to nationally recognized
be created, gathered, managed, and consulted by authorized clinicians
interoperability standards and that can be created, managed, and consulted by authorized clinicians
that conforms to nationally recognized interoperability
and staff within one
and staff across more than one
from multiple sources while being managed, shared, and controlled by
healthcare organization
healthcare organization
the individual
standards and that can be drawn
EHR, electronic health record; EMR, electronic medical record; PHR, personal health record.
Source: Reproduced from The National Alliance for Health InformationTechnoiogy. (2008). Report to the Office of the National Coordinator for Health Information Technology on defining key health information technology terms (No. NAHIT-2008). U.S. Department of Health and Human Services, http://www.hitechanswers.net/wp-content/ uploads/2013/05/NAHIT-Definitions2008. pdf.
144
li: POINT-OF-CARE TECHNOLOGY
CASE STUDY 1
Lessons Learned: Ergonomic-Human Factors in Barcode Scanning
Recently, a healthcare system evaluated its knowledge-based medication administration process, which utilized barcode scanning. The process was implemented to provide support for a higher standard ofnursing care, with inherently fewer medication errors. Incorporating computer-assisted clinical decision support (CDS) in real time and at the point-of-care delivery required barcode scanning of the patient's armband and each individual unit-dose medication labeled with a barcode. During the process of performing the nursing care tasks of medication administration to multiple patients on a medical-surgical hospital floor, several distinct human factors and ergonomic problems were identified. The type of computer hardware "Workstations on Wheels" (WOWs) or "Computers on Wheels" (COWs) exhibited limited mobility when pushed on carpets, because of the selection of smaller wheels on the cart upon which the workstation was mounted. Network issues of crowding around Wi-Fi hotspots resulted in significant reduction in access speeds. With no external and extra batteries that could be charged separately, batteries of WOWs required frequent recharging; they were anchored to stationary charging locations, essentially removing them from service until charged. Lack of standardized maintenance and internal cleaning of cooling fan and airflow vents led to overheated/broken computers and WOWs. Log-in configurations added many time-consuming layers. These layers resulted in excessive logon-logoff interruptions required to navigate among multiple secure systems and data silos during the process of assembling and transporting the medications that had to be administered.This process included transport from the satellite pharmacy storage bins to the point of care at the patient bedside, where barcode scanning would be carried out during the process of administering routine medications to the patient. It was discovered that some computer workstations were capable of performing at much higher speeds, and these became favored workstations that were in constant crowded use. In contrast, the other slower workstations were never being used.
Reporting of slow or broken hardware was compounded by lack of a designated technical support field agent with the primary job responsibility of maintaining the optimal function of all workstations on the medical-surgical unit. Staff education was needed to establish when, where, and howto report suboptimal performance, broken equipment, or unreliable Wi-Fi or network access issues. During the implementation phase of barcode scanning knowledge-based medication administration, it became apparentthat having aglobalstrategy for device installation configuration is extremely important for successful implementation. Each point-of-care location where any human-computer interface is required to deliver care requires individual assessment and customized configuration. The assessment needs to occur before purchasing hardware, software, and peripherals (scanners, sensors, etc.) and must necessarily include testing all reasonably anticipated human-computer end-user workflows and tasks that are expected to be performed in the context of actual point of care. Identification, classification, and categorization of different tasks and workflows associated with different user groups and service lines were critical factors for success and were found to be best determined largely by the workflow and task being performed in the context
6: COMPUTERS IN HEALTHCARE
of the actual point-of-care delivery. The anticipation of utilization barriers and the consideration of time of day as a barrier resulting from peak network bandwidth congestion were also important success strategies. Careful observation of unit layout and workflow that incorporated frontline staff and rehearsal of dry-run scenarios based on proposed CDS workflows were of key importance. Time, motion, and efficiency videography studies of the entire stepwise CDS process involving frontline staff acting as patients and caregivers were performed over several weeks. Analyses of the videographic journals in time-lapse compression, transforming hours of observations into a few intense minutes, revealed numerous barriers, bottlenecks, and inefficiencies present in the actual CDS process. Issues identified were hardware placement given the physical layout of the actual pointof-care delivery room.The computer workstation and the barcodes being scanned were placed in physically separate locations that were convenient for the cabling and the barcode, but awkward and inconvenient for the end-user nurse.The nurse
was required to negotiate barriers in order to deliver care in the room. Workstation behavior required nurse-console interactions with keyboard and mouse at almost every step in the CDS barcode scanning workflow procedure.The resulting humancomputer interface behavior required the nurse to physically walk back and forth along the three sides of the room.The patient's bed and armband with the barcode to be scanned were located on one side of the room, the console with keyboard and mouse was located on the opposite side of the room, and the unit-dosed barcoded medications were located on a third side of the room on the only available shelf workspace at the foot of the patient bed.The walk-around distance among the three sides of the room, plus having to walk back and forth between the computer console and the barcode on the packaged medications arrayed side by side (they were checked off of a hand-written pick list with the handheld laser scanner), added significant time to the administration of alt patient medications on the unit.The significant time added during time-and-motion efficiency recordings did not include interruptions to the process from overhead pages, telephone calls, family members, or requests for immediate attention and help from fellow staff or physician demands.The timecompression analyses performed were therefore conservative underestimations of normal variations in the actual workflow.
Questions to consider:
1. What hardware and peripherals might have served the organization better than WOWs, wall-mounted units, kiosks, or desk workstations?
a. Secure wireless-enabled personal electronic devices (tablet. Nook, Surface, etc.) with optical, near-field, QR-code, or barcode scanning capabilities b. Secure voice-recognition "speech to text" interface and wearable augmented reality interface ("Google Glass," Virtual Reality 3 sight-sound-touch input/ output), wearables, smart-vests, and smart-shells 2. How might they have configured the hardware differently? a. Incorporating frontline staff and end-user feedback to guide device selection
b. Identifying end-user group service line-specific tasks and their unique workflow dynamic requirements
3. What are the implications to patient safety and quality?
145
146
II: POINT-OF-CARE TECHNOLOGY
There are many important objective lessons to be learned from this case study, including hardware placement, human factors considerations, and implications for CDS. Points to consider related to the case are as follows:
■ Successful implementation of CDS is associated with dramatic reduction in adverse events due to accidental human factor errors.
■ Poor implementation of CDS is associated with an increase in accidental adverse events because of staff frustration and resorting to "work-around," "cutting corners," and abandonment of automation altogether, just to deliver actual care.
■ Poor implementation of CDS is a financially unsustainable and risky business model. Damages and punitive damages awarded in settlements related to actual or perceived accidental injuries attributable to CDS implementation require mitigation of risk strategies to avoid loss. Accelerated turnover of frustrated, highly mobile nursing and physician staff to competitor healthcare provider organizations is also a low-quality outcome of poor CDS implementation. ■ The quality of CDS implementation is a direct reflection of workplace environment health, and this in turn synergistically affects patient safety culture
as well as the overall quality of care ultimately provided.
CASE STUDY 2
Effective Selection and Deployment of an Electronic Health Record in a Rural Local Public Health Department Contributed by Deb McCullough, DN,P RN, FNP Andrews County Health Department
/ssues:Asmall rural local health department (LHD) implemented an electronic health record (EHR) using a step-by-step implementation guide (Hartley & Jones, 2005). The project incorporated national standards and requirements to meet stage 1 meaningful use criteria under the HITECH Act (discussed in detail in Chapter 1).
Description: The LHD formed a seven-member interprofessional team with defined roles and responsibilities to prepare for EHR implementation. The team collaborated with the WestTexas HIT REC; attended a regional conference on EHR meaningful use, clinical decision-making (CDM), quality, and safety; reviewed the LHD's services, payer and funding sources, current charting and billing systems, and perceived benefits of EHR implementation, and developed
the vision, goals, and criteriafortheproject'ssuccess. After EHR vendor training,
the LHD went live in September 2011 employing a big-bang approach. Step-by-Step Implementation Process:The LHD followed a step-by-step guide for medical practice EHR implementation. Step 1 consisted of learning the basics of EHRs with support of the REC. During step 2, the staff conducted workflow
analyses, compared paper administrative and patient care workflows to EHR workflows, participated in vendor demonstrations, and identified EHR
champions. Additional considerations were rural connectivity issues and
6: COMPUTERS IN HEALTHCARE
whether to host the server on site or in an internet cloud-based service offered
by EHR vendors. Rural providers have specific issues related to internet connectivity that must betaken into account. During step 3, the LHD determined the appropriate EHR based on the budget and workflow needs and connectivity requirements and purchased the EHR.The LHD selected eClinIcalWorks as its vendor because of the vaccine inventory-and-management component and the ability to meet the LHD's clinical needs; they elected to host It on-site due to connectivity constraints within the rural community. The fourth step addressed the EHR implementation phase: changing from paper medical records to an EHR.
Pre- and Post-Implementation Workflow Analyses and Redesign: The staff reviewed multiple processes and conducted workflow analyses and redesign. In July 2011, stakeholders from the LHD andTexas Department of State Health Services (DSHS) met to review Texas Health Steps (THS) Medicaid forms and documentation requirements. Before the EHR, the LHD staff completed 20 forms during an initial THS visit. The LHD seized the opportunity to redesign and streamlineTHS documentation to avoid transitioning a dysfunctional paper process to an electronic format. After EHR implementation, the staff struggled with the cumbersome and difficult-to-navigate THS documentation structure within the EHR.To improve the workflow processes, decrease staff frustration, increase consistency, and meet the documentation requirements, the LHD director/system administrator built the streamlined DSHS forms into the EHR. The forms included the history of children younger than 5 years, history of children of all ages, interval history, tuberculosis questionnaire, physical exam, and health education/anticipatory guidance. The director added age-specific immunization andTHS order sets and CDS alerts.
Lessons Learned: A proactive multidisciplinary team approach is essential for implementing an EHR in a small LHD. A hosted EHR has advantages for a small LHD. Collaboration and consultation with a regional REC, local public health, and state funding sources are essential to provide expertise and Implement a
meaningful EHR applicable to the setting.The EHR can impact prevention and wellness across the life span by providing accessible preventive health services documentation and CDS; paper charting was reduced by 98%. Recommendations: Successful EHR implementation and use includes strong leadership and visions, policies on key issues, goal setting, planning, and communication. Critical factors, particularly when rural facilities are resource constrained, involve utilizing the support of the REC to provide a step-by-step process, selecting a vendor by matching the capabilities of the EHR to the staff's requirements, as well as perceived benefits. Also critical was redesigning workflow with an improvement focus and ensuring flexibility and capacity for
creating documentation components within the EHR. Consider the following questions:
■ Which tools do you believe the clinic used to help with this process?
■ How do you see that the REC might have helped with this process, given what you read in earlier chapters and this chapter on software selection considerations? ■ What hardware considerations do you think this organization considered?
147
\48
II: POINT-OF-CARE TECHNOLOGY
■ Why would cost be a critical issue for this organization, and what other considerations might they have given their public health purpose? ■ How do you think rural providers and clinics present unique challenges when implementing EHRs?
SUMMARY
We have discussed background information, including the history of computers in healthcare and the important role that computers have played and will play in the future. Healthcare informatics is an evolving science that involves the overlap between computer science and the use of clinical information to inform patient care, quality, patient safety, and population health. In addition, we have discussed hardware and software configuration considerations, examined a case study relating to ergonomics and human factors that impact patient care when not addressed strategically, and related additional information concerning software selection in niral settings. We have also discussed programming languages, types of software, and EHR selection, as well as the importance of understanding hardware, software, networking, and the specifications of technical requirements that impact the EHR selection decision.
END
1
CHAPTER
RESOURCES
EXERCISES AND QUESTIONS FOR CONSIDERATION
Evaluate the physical environment of IT in a hospital or clinical setting. Consider the HCI as you observe clinicians using the EHR in the clinical setting. Your assessment should include the following tasks: 1. Identify the location of the EHR on the unit and observe clinicians using the system. 2. Obtain input from the nursing staff who use the technology in the chosen setting. 3. Assess the site for the following:
a. The layout of the nursing unit
b. The location of technology in relation to patient care
areas
c. The type of equipment available
d. The structure, size, and function of existing furniture, flooring, equipment, and so forth e. Nursing staff input on functionality and challenges
4. Analyze the impact of the existing ergonomic design on nursing work, patient safety, and quality of care.
5. Describe the challenges in ergonomic design. 6. Develop a plan to solve identified challenges. 7. Create an electronic visual representation of the layout of the area and document issues noted and a plan of action.
6: COMPUTERS IN HEALTHCARE
ADDITIONAL RESOURCES
SPIIK(tl PUBLISHIIK
CONNECT
A robust set of instructor resources designed to supplement this text is located at http;//connect.springerpub.com/content/book/978-0~8261-8526-6 . Qualifying instructors may request access by emailing textbook@springerpub. com.
REFERENCES Buchholz, W., & Berner, R. W. (1962). Natural data units. In W. Buchholz (Ed.), Planning a computer
system: Project stretch (1st ed., pp. 33-40), McGraw-Hill. Cearley, D. W. (2014). Gartner report: The top 10 strategic technology trends for 2014. Forbes, http;// www.scribd.com/doc/220080178/Gartner-The-Top-10-Strategic-Techno logy-Trends-for-2014 Dickey, J. R (1984). Custom CMOS architecture for a handheld computer. Herolett-Packard Journal: Technical Information from the Laboratories of the Heivlett-Packard Company, 35(7), 14-17. Forbes, J. (1823). A treatise on the diseases of the chest by R. T. H. Laennec [Ren^-Theophile-Hyacinthe Laennec] (1st American Edition). James Webster.
Hartley, C. R, & Jones, E. D. (2005). EHR implementation: A step by step guide for the medical practice. AMA Press.
IBM. (2020, July 2). Natural Language Processing (NLP), https://www.ibm.com/cloud/learn/natural -language-processing
International Electrotechnical Commission. (2014). International standards (IS), http://www.iec.ch/
standardsdev/publications/is.htm Janssen, C. (2014a), Fifth generation (programming) language—5GL. http://www.techopedia.com/ definition/24309/fifth-generation-programming-language-5gl Janssen, C. (2014b). Fourth generation (programming) language—4GL. http://www.techopedia,com/ definition / 24308 / fourth-generation-programming-language-4gl Janssen, C. (2014c). Second generation (programming) language—2GL. http://www.techopedia.com/ definition/24305 / second-generation-programming-language-2gl Jans.sen, C. (2014d), Third generation (programmmg) language—3GL. http://www.techopedia.com/ definition/24307/third-generation-programming-language-3gl Joint Electron Device Engineering Council. (2002). Terms, definitions, and letter symbols for microcomputers, microprocessors, and memory integrated circuits (No. JESDIOOB.OI). JEDEC Solid State Technology Association.
Kendall, K., & Kendall, J. (2014). Project management. In Systems analysis and design (9th ed., p. 57). Pearson.
Kleinbeck, S. V. M. (1999). Development of the perioperative nursing data set. AORN Journal, 70(1), 15-28. https://doi.org/10.1016/S0001-2092(06)61851-6 Kuo, A. M. (2011), Opportunities and challenges of cloud computing to improve health care services. Journal of Medical Internet Research, 13(3), e67. https: / / doi.org/10.2196/jmir.1867 Kushniuk, A., & Borycki, E, (2013). Chapter 16: Human factors in healthcare IT. In K. M. McCormick & B. Gugerty (Eds.), Healthcare information technology exam guide (pp. 367-390). McGraw-Hill. Leonhardt, D. (2000, July 28). John Tukey, 85, statistician; coined the word "software." New York Times. http: / /www.nytimes.com/2000/07/28/us/john-tukey-85-statisticia n-coined-the-word-software .html
Lietzke, E. T. (1982) Wading through computerese. Medical Group Manage, 29(6), 32-38, 42-45. Mitchell, B. (2018). Introduction to computer network topology. Lifewire. https://www.lifewire.com/ computer-network-topology-817884 Moore, G. E. (n.d.). Moore's law. http: / /www.mooreslaw.org The National Alliance for Health Information Technology. (2008). Report to the Office of the National Coordinator for Health Information Technology on defining key health information technology terms (No. NAHIT-2008). U.S. Department of Health & Human Services, http://www.hitechanswers.net/wp -content / uploads / 2013 / 05 / NAHIT-Definilions2008.pdf Neel, A. N., & Sittig, D. F. (2002). Basic microbiologic and infection control information to reduce the potential transmission of pathogens to patients via computer hardware. Journal of the American
Medical Informatics Association, 9(5), 500-508. https://doi.org/ 10.1197/jamia.M1082
Nilsson, N. (2013). The quest for artificial intelligence. Cambridge University Press, Online resource. Office of the National Coordinator for Health Information Technology. (2014). How to implement EHRs:
Step 3—Selector upgrade to a certified EHR. https: / /www.healthit.gov/taxonomy/term/771 Ogala, J. O. N. Y. A. R. I. N. (2020). Comparative analysis of CCC and java programming languages. Global Scientific Journals, https://doi.org/10.11216/gsj.2020.05 .40055
149
150
II: POINT-OF-CARE TECHNOLOGY
Philips. (2014). Hardware specifications, https://www.philips.ca /healthcare/product/HCNOCTN198/ intellispace-cardiovascular
Rasouli, P., & Yu, I. C. (2020, July 19-24). EXPLAN: Explaining Black-box Classifiers using Adaptive Neighborhood Generation. Paper presented at the 2020 International Joint Conference on Neural Networks (IJCNN).
Shannon, C. E. (1948). A mathematical theory of communication. Bell System Technical Journal, 27(3-4), 379^23, 623-656.
TechTerms.com. (2014). Definition: Secure sockets layer (SSL), http:/ /www.techterms.com/definition/ssl William, C. (2012). Collins Ejiglish dictionary—Complete & unabridged: Internet (10th ed.). HarperCollins.
http: / / dictionary.reference.com /browse/ internet Wilson, J. R. (2013). Fundamentals of systems ergonomics/human factors. Applied Ergonomics, 45(1), 5-13. https://doi.org/10.1016/j.apergo.2013.03.021 Yuri, K. (n.d.). Units of information and data storage, http://www.translatorscafe.com/cafe/EN/units -converter/data-storage/10-5/kilobyte-word Zanbergen, P. (n.d.). How star, bus, ring & mesh topology connect computer netioorks in organizations. http: / /education-portal.com/academy / lesson / how-star-topolo gy-connects-computer-networks-in -organize tions.html#lesson
Electronic Hea
Poi nt-of-Ca re Tech no logy MARYB£TH MITCHELL AND SUSAN MCBRIDE
OBJECTIVES ●
Discuss the electronic health record (EHR) evolution within hospitals, including federal initiatives, healthcare system impact, and clinical rationale.
●
Describe the benefits of EHRs and point-of-care (PoC) devices and the challenges with rapid adoption and implementation under the HITECH Act resulting in usability, dissatisfaction, burnout, and distress of clinicians.
●
Explain the impact of integration on the EHR, including managing disparate systems, PoC devices, and device integration.
●
Explore methods for optimization to increase adoption and utilization and understand how nursing informatics (Nl) supports the sustainability of the EHR.
●
Examine the importance of governance, ongoing maintenance, support, and training to optimizing EHRs and other PoCs devices. CONTENTS
INTRODUCTION
152
History of Electronic Health Records in Hospitals
153
Drivers Behind Nationwide Adoption of Electronic Health Records
154
ELECTRONIC HEALTH RECORD, POINT-OF-CARE DEVICE INTEGRATION AND INTEROPERABILITY
155
Interoperability and integration of the Electronic Health Record Interoperability and Standards
156
156
Integration of Disparate Systems
156
Integration andTypes of Point-of-Care Devices
157
Challenges with Point-of-Care Device Integration HIT SAFETY ANDTHE SAFER GUIDES
159
164
INFORMATICS NURSE SPECIALISTS, INFORMATICS NURSE, AND SUPERUSER ROLES
165
Nursing Informatics' Role
165
Impact of Nursing Informatics Superusers'Role
166
165
152
li: POINT-OF-CARE TECHNOLOGY
BENEFITS OFTHE ELECTRONIC HEALTH RECORD FOR IMPROVING SAFETY AND QUALITY
166
Features of Certified Electronic Health RecordsThat Promote Patient Safety and Quality
Care Coordination Supported by Electronic Health Records Electronic Health Records and Predictive Analytics
167
168
168
ELECTRONIC HEALTH RECORDS' NEGATIVE IMPACT ON CLINICIANS AND A NATIONAL AGENDATOADDRESSTHE ISSUES
168
Usability Issues with Electronic Health Records
168
Triple Aim Moves to a Quadruple Aim to Address Clinician Well-Being Human Factors Science, Usability, and the Electronic Health Record Research and Initiatives to Address Clinician Dissatisfaction
168 169
169
CASE STUDY: DEVICE INTEGRATION IMPROVING PATIENT SAFETY SUMMARY REFERENCES
171
171 172
INTRODUCTION
Since the 1960s, having a comprehensive electronic health record (EHR) bringing together a person's health history in a cohesive and systematic longitudinal record, has been a national goal. The desire to provide access to healthcare resources, continuity of care for all individuals, and holistic management of a person's quality of life have influenced the rapid development of EHRs over several decades in ways never imagined. This chapter describes the history and impact of EHR adoption and implementation on the healthcare community, including healthcare consumers. It explains the need to focus attention on usability, interoperability, and optimization of EHRs and point-of-care (PoC) technologies. To fully understand this development of EHRs over the past four decades, one only has to start with the present. In the healthcare environment of today, healthcare begins with an individual engaging in healthy behaviors. The person accesses their health information via a web-based portal on a mobile device that integrates, manages, and reports all relevant health information, including those from various medical devices and mobile applications the consumer uses. Information from wearable devices that provide vital signs and activity routinely downloads and aggregates data for healthcare consumers. Lab data (such as blood sugars, urine tests) can be remotely monitored and treatments provided via telehealth platforms. Health history data are available and continually updated as situations change, and health problems are added and resolved. Every encounter with the healthcare system results in information within a patient record, whether it be an office visit to a provider, a hospital encounter, or interaction with other community services, including emergency medical services, health centers, and long-term care facilities. This information is now accessible by patients, families, and healthcare providers. In addition, this information is organized and defined within a framework of rules and tools that provide reports, reminders, alerts of potential or real problems, treatment options, intervention.s, and outcomes. Patient education is given to the patient, auto assigned, or sought out by the patient based on documented activities and medical history. The patient's response to education, treatment, compliance, and other patient activities are recorded and considered in the total health picture. Over time, a longitudinal view of a person's health emerges. This information can be compared with that of other patients in other places and systems to provide information about populations with similar health patterns that can guide research and best practices and
7: ELECTRONIC HEALTH RECORDS AND POINT-OF-CARE TECHNOLOGY
manage thehealth of the single patient across theirlife span. This type of health information technology (HIT) is what many across the United States are experiencing—both clinicians and healthcare consumers of services. Yet, this adoption of HIT occurred over several decades with many positive results and many opportunities for improvement. This chapter will review the historical evolution of EHRs, examine ways to optimize EHRs with PoC devices, explore usability challenges and the national imperative to address the challenges. PoC devices are HIT that encompass the devices and systems that support clinicians in their daily activities of patient monitoring and electronic documentation of clinical
care and health. PoC technologies can be used at the bedside in acute care hospitals to support care delivery, as well as giving providers' and nurses' use of real-time data access via mobile technologies in the EHR (Addondata.com, 2017). History of Electronic Health Records in Hospitals EHRs have become increasingly prevalent since the 1990s because of the national initiatives taken to promote EHR adoption. However, as far back as the mid-1960s, it is believed that approximately 73 hospitals had some type of health record in an electronic format. Often, universities and corporations worked together to develop these early EHRs, such as the University of Utah working with 3M or Harvard working with Massachusetts General, to create their various health record components. These were not full EHRs, but rather specific programs that provided functionality in limited ways (Atherton, 2011). However, in the 1970s, the federal government implemented an EHR system within the Department of Veterans Affairs called the De-centralized Hospital Computer Program, which was used nationwide. In addition, the Department of Defense also implemented the composite healthcare system (CHCS) to serve as the patient record for all military personnel worldwide (National Center for Research Resources [NCRR], 2006). By the 1980s, more work was done to develop and increase the use of EHRs in medicine, as their use became better defined and a greater potential for improving health was recognized. The National Academy of Medicine (NAM; previously the Institute of Medicine) launched a study in the mid-1980s on the potential of EHRs to improve patient clinical care (Dick & Steen, 1991). This study, the Computer-Based Patient Record, originally published in 1991 and again in 1997 with revisions, was the first to call for the widespread implementation of EHRs to provide timely, accurate health data and to improve the quality of care while reducing costs (Dick et al., 1997). In 1987, an EHR standards-developing organization called Health Level Seven International (HL7, n.d.) was formed to develop standardization around EHRs, thus recognizing the growing industry of health records and the need for integration and communication across platforms and organizations. HL7 has become the standard in 55 countries to exchange, integrate, share, and retrieve health information. However, a newer, faster, and more robust standard is emerging, called FHIR (Fast Healthcare Interoperability Resources). FHIR allows for a more open, interoperable, and efficient approach to interfacing multiple systems (Office of the National Coordinator for Health Information Technology [ONC], 2016). This standard will be discussed more fully in Chapter 12. With the increasing awareness of the potential of EHRs, along with the development of standards, and the increase in companies developing clinical application.s, organizations started moving toward some type of electronic systems within hospitals. Lab systems were often early clinical systems used within hospitals, as were patient registration systems, but wide adoption of complete, integrated EHRs did not occur until the 2000s.
153
154
II: POINT-OF-CARE TECHNOLOGY
In 1999, the NAM issued the landmark report To Err Is Human: Building a Safer
Health S\/steni. It ascertained that tens of thousands of Americans die every year due to
potentially preventable medical errors. In 2001, the NAM issued another report, Crossmg
the Quality Chasm: A New Health System for the 21st Century, in which the need for high-
quality, evidence-based, standardized healthcare for all was emphasized. These reports
explained that part of the reason for the schisms in care across the country was lack of technological infrastructure, uncoordinated care, and lack of organization in a difficult system to navigate and manage across multiple encounters and interactions with healthcare providers. NAM believed that a fundamental component needed to address the problems plaguing the healthcare system in America was to advance technology. NAM indicated that technology, including the internet, "holds enormous potential for transforming the healthcare delivery system, which today remains relatively untouched by the revolution that has swept nearly every other aspect of society" (lOM, 2001, p. 15). In addition, the report emphasized the need for a "national commitment" to building an information infrastmcture to support healthcare delivery, consumer health, quality measurement, improvement, public accountability, clinical and health services research, and clinical education. The goal was to eliminate most handwritten clinical data by 2010. Thus began the drive to improve the quality of care by adopting EHRs and implementing national initiatives to promote EHRs (lOM, 2001). Drivers Behind Nationwide Adoption of Electronic Health Records
Even though the two landmark NAM reports promoted EHRs to improve patient quality and safety, and even though there were more sophisticated and developed systems, the adoption of fully integrated EHRs remained low through the firstpart of the 21st century. Hospitals were starting to purchase and implement pieces of EHRs, such as radiology systems, operating room (OR) systems, or even barcoding of supplies or medications. However, a fully integrated EHR was still not implemented in most hospital systems. In 2009, a study found that only 9% of hospitals surveyed had basic or comprehensive EHR systems. Further, only 17% of hospitals had computerized provider order entry (CPOE) implemented for medications (Jha et al., 2009); thus, the Health Information Technology for Economic and Clinical Health (HITECH) Act was born.
In 2004, President George W. Bush highlighted the need for hospitals and healthcare providers to implement and adopt EHRs. President Bu.sh called for full adoption of EHRs within 10 years by stating in his State of the Union Address, "By computerizing health records, we can avoid dangerous medical mistakes, reduce costs, and improve care" (The White House, 2004). To support this effort, President Bush created the Office of the National Coordinator (ONC) within the Department of Health and Human Services. The ONC was to oversee and manage the adoption of EHRs by developing standards for interoperability, providing for certification of EHRs tliat met national standards, and developing a national infrastmcture to support health information exchanges (HIEs). As the advancement of EHRs continued within hospital and physician practices, there came an ever-growing foais on the use of technology in healthcare. In 2009, President Barack Obama made an unprecedented move and unofficial "mandate" to use EHRs by all healthcare providers and hospitals. The mandate was considered unofficial because it was financial incentives followed by financial penalties or disincentives within a legislative process. The economic stimulus bill of 2009, the American Recovery and Reinvestment Act (ARRA), also included $29 billion to adopt and utilize EHRs. The ONC and the Centers for Medicare & Medicaid Services (CMS)
7: ELECTRONIC HEALTH RECORDS AND POINT-OF-CARE TECHNOLOGY
would support and manage the HITECH Act through the administration of financial incentives for organizations that demonstrated "meaningful use" (MU) of EHRs to improve patient care and clinical outcomes. The current term for the MU is Promoting Interoperability Program (PIP); however, the program was originally titled by CMS and the ONC, the MU Program under the HITECH Act. The HITECH Act represents one of the most significant federal investments ever made in HIT, demonstrating a broad consensus and commitment to fully realizing the potential of EHRs to transform the healthcare system (Blumenthal, 2011). The "MU program," as it was known, was overseen by the ONC and administered by the CMS. As discussed in Chapters 1 and 4, this program provided extensive guidelines for providers and hospitals that defined the criteria that demonstrated MU of a certified EHR with three stages of maturity. These criteria defined thresholds that had to be met to demonstrate how organizations use their health information, ranging from recording patient information as structured data to exchanging summary care records. By 2017, 96% of all hospitals and 80% of all providers had adopted a certified EHR (ONC, 2017)—a very rapid and massive adoption occurred over a relatively short period of time driven by financial incentives and federal requirements. As discussed
in Chapters 1 and 4 the MU program has transitioned to the PIP. The focus has moved from adoption and implementation to more fully interoperable systems internally, regionally, nationally, and internationally. However, many of the lessons learned with the initial adoption and implementation of EHRs across the United States under the prior MU program continue to be relevant to optimizing and integrating new devices and application program interfaces (APIs) under the PIP. ELECTRONIC HEALTH RECORD, POINT-OF-CARE DEVICE INTEGRATION AND INTEROPERABILITY
There have been three primary ways for organizations to make the EHR adoption decision: a single-vendor decision, best of breed, or a combination of both. The ONC (2018) has created a Health IT Playbook that reviews how to approach EHR implementation and adoption. Some organizations, unhappy with their initial EHR selection, may be considering replacing the current EHR with an EHR that is more suitable for the organization. See further resources for purchase and adoption at the following website: https: / / www.healthit.gov/playbook/electronic-health-records/. In most organizations, implementing an EHR has been accomplished with the current focus on improvements, optimization, and a return on investment (ROI)—but this is an iterative process that evolves and matures over time. Considerations for improvement strategies and optimization of the EHR present additional decisions and strategies, The phases of Systems Development Life Cycle (SDLC) to be covered thoroughly in the Chapter 8 are still relevant, but not necessarily in the context of a new system, but improvements and upgrades to the adopted EHR. Today, the adoption of EHRs has become routine, given that most organizations and providers have an EHR. Organizations are not in the process of implementing new systems as much as they are working to meet the demands of the clinicians and patients through optimization and additional tools to help gain efficiencies and standardization. This transition from implementation to optimization has helped further the acceptance and adoption of these systems. It will continue to improve patient care and quality outcomes, particularly with expanded interoperability, new devices, and APIs.
155
156
H: POINT-OF-CARE TECHNOLOGY
Interoperability and Integration of the Electronic Health Records A successful EHR is dependent on its ability to integrate into a cohesive clinician experience and reflect an accurate patient record. Most EHRs today, under certifications for EHRs under the HITECH Act, can reveal a patient's journey across the continuum, from various encounters, from a physician's office visit to a hospital encounter, even to the patient's home. Population health and the CMS PIP have increased the need for interoperability. The following section addresses various types of interoperability often seen within the EHR that impact the user experience of the EHR and the opportunity to decrease redundancy in data entry.
Interoperability and Standards Interoperability describes the extent to which systems and devices can exchange data and interpret that shared data. For two systems to be interoperable, they must exchange data and subsequently present those data to be understood by a user (Health Information and Management Systems Society [HIMSS], 2013). Interoperable systems allow data to be shared and exchanged across systems, within an organization, and outside the organization. Through interoperability, systems can synergistically enhance health information management and reduce redundancy or data collection, allowing for data aggregation for analytics and reporting. Interoperability among systems makes it possible to have different applications within an organization and enables sharing of data between organizations, providers, or other stakeholders. Interoperability is becoming more prominent and more robust in its use due to the advancement of standards, such as FHIR, an API, which easily links data among multiple sources. Technical standards are now being developed across various healthcare applications, such as device standards, system standards, terminology standards, and reporting standards. Certified EHRs must comply with interoperability standards and provide various data standards to transfer data among systems, regions, nationally
and even internationally. As the prevalence of EHRs is expanding nationally and internationally, the.se standards are becoming more mature and detailed in scope and definition, allowing for greater interoperability. These types of technology and data standards will be covered extensively in Chapter 12. Many current EHR vendors are working to develop APIs to bring their content within the context of the patient record and improve the overall patient experience. This approach to EHR optimization is covered in Chapters 5 (Consumer Engagement) and 15 (Personal Health Records and Patient Portals).
Integration of Disparate Systems Many EHR vendors are making functionality more inclusive to accommodate all aspects of the patient experience. However, it is still complex for single EHR vendors to have expertise in all types of systems, especially in niche systems specifically designed for a single function, such as Labor and Delivery, Surgery, or Cardiology. Specialty departmental systems are often installed and implemented at different times over many years, so the functionality of the EHR is constantly in flux. Therefore, many organizations still have disparate systems in use, often with outdated systems and systems using newer technology. These systems are often referred to as "legacy systems." Legacy systems are older technology that are maintained because they continue to generate value for the organization. The u.se of multiple legacy systems can create challenges within the overall EHR integration strategy. For example, a hospital may have had an OR system in place
7: ELECTRONIC HEALTH RECORDS AND POINT-OF-CARE TECHNOLOGY
for many years (legacy system) when implementing a new EHR. The OR system with older technology may not be able to integrate into the new EHR Rilly.
Another example is a hospital with an EHR vendor but chooses to implement a different system specialized only for obstetrics. As a result, it more fully meets the needs of obstetrical clinical documentation and reporting requirements. However, the clinician still has specific functions that must be doaimented in the EHR for the cohesiveness of the patient record. Procedural areas, such as surgery, lab, radiology, or cardiology, often have very specialized systems that need to be integrated into the overall EHR. Therefore, interoperability and integrating these disparate systems are crucial to having a fully integrated EHR. This integration continues to be a work in progress. Most organizations take a very strategic approach in managing the integration of disparate systems and their older legacy systems. Key factors that impact these strategic decisions may include the following:
■ Safety and Quality: Is there a safety or quality benefit of a system that manages only one aspect of care, and is that benefit quantifiable? ■ Functionality: Are there specific functions required, which are not met within one .system and require a different system?
■ Ejficiencies: How well do the data integrate into a seamless system? Is there redundancy of data and /or data entry redundancy? ■ Ease of Use: How easy and straightforward is it for the clinician to use and interact with the system because of the specificity of the system's department-specific functionality?
■ Costs: How justified are the replacement costs of legacy systems and costs of management of disparate systems in terms of technical resources? ■ Return on Investment: Is there more value in u.sing a separate system, and can that value be quantified in either direct cost savings or cost avoidance?
Regardless of the type of system used, most organizations have some component of integrating various systems. As a result, the management of this integration requires resources, technical expertise, and ongoing monitoring to maintain an efficient user experience, and safe clinical workflows.
Integration andTypes of Point-of-Care Devices Devices and the ability of devices to manage multiple applications are becoming more of a concern, and the device management aspects of PoC technologies are critical to patient safety and quality. Devices often provide better clinical mobility for moving the technology to the PoC. Examples of device integration and the benefits of these devices that will be covered include PoC testing devices, bar codes medication administration (BCMA) devices, smart IV pumps, smart beds, and communication technologies. While the types of devices covered are not meant to be all-inclusive, we cover some of these
devices to emphasize both the benefit and the challenges with integration. Indeed, more mobility is key to users, but it is often difficult to manage everything from doaimentation to review on a single mobile device. Point-of-CareTesting Devices
PoC testing allows for testing and diagnosis at the patient's side. It can be conducted anywhere the patient is, such as the home, physician's office, ambulance, or hospital
157
158
I: POINT-OF-CARE TECHNOLOGY
bedside (National Institutes of Health [NIH], 2010). This technology allows for quick,
on-the-spot testing with immediately available results. Additionally, these results can be downloaded directly into the EHR through interface engines. Interface engines can transmit data in several ways. They can transmit data one way from another data system to the EHR or transmit bilaterally to and from the EHR. These interfaces are critical for clinical information, such as the immediate availability of PoC testing results. This information can decrease the risk of error in manually entered results, and the results are immediately available to caregivers for making treatment decisions. However, a poorly designed or suboptimal interface can create patient safety issues. For example, suppose an interface engine transmits data to the EHR from the PoC device but fails to transmit a change in device use from one patient to the next bilaterally. A serious patient safety error could occur from the wrong results on the wrong patient in that case. Many innovative devices are emerging. These devices include abilities to monitor and maintain health and well-being. These devices include fitness-measuring devices, scales, biometric devices, and FDA-approved medical devices, such as insulin pumps, pacemakers, defibrillators, and so forth, which can be interfaced with EHRs patient portals. These technologies are discussed fully in Chapter 15, which covers patient portals and personal health records. The potential for advancement in PoC devices is one of the most rapidly growing areas in the healthcare industry, with tremendous potential for improvement in patient safety, quality, and population health. One of the most common types of PoC testing is blood glucose monitoring. Integration with the EHR occurs when the clinician verifies the patient by scanning the patient's armband with the glucose reader and then performing the glucose test. The result is then uploaded directly into the EHR for that patient. This upload is done either through wireless upload, with immediate results on the glucose levels, or by synching the glucose meter with a docking station. The data are sent to the EHR via a secure router. This type of PoC testing is expanding to include iStat for various electrolyte tests obtained at the bedside and tests for laboring patients such as protein and glucose. The benefits of PoC testing are primarily the immediate availability of the results and a decrease in transcription errors of the value through manual entries. In addition, there is less likelihood of documenting the lab result on the wrong patient. On the downside, these results need to be verified by the clinician and often by the lab and can be problematic. For example, a lab value is accepted into the record that was not obtained accurately, inaccurate results obtained because of end-user error collecting the specimen, or a problem with the device. In addition, the EHR also provides clinical decision support (CDS) rules and alerts related to the results as they are verified and entered into the record. An example of these types of CDS rules is an alert for increasing insulin due to a high-glucose reading. However, these devices are not intended to replace the clinician's clinical judgment but provide an adjunct to support the clinician in ease of use, manual-error reduction, and efficiency. Many PoC systems require FDA approval. For example, the barcode reader interacts with the patient, and the FDA regulates all medical devices that interact with patients. EHR vendors are starting to obtain FDA approval for this type of technology (FDA, 2014). Other examples of patient PoC devices that require FDA approval are fetal monitors, blood glucose monitor.s, blood pressure monitors, and other vital sign monitors, in addition to devices used for barcode blood administration. FDA approval has been a challenge for EHR vendors because the requirements for FDA approval are specific to the devices that interact with the patient. These vendors tend to be software developers, not hardware systems manufacturers with expertise in developing for FDA approval. However, as customer demand for PoC has increased, the trend now is for the software
vendors to seek FDA approval for their PoC technology.
7: ELECTRONIC HEALTH RECORDS AND POINT-OF-CARE TECHNOLOGY
Barcode Administration Medication Administration
In the hospital, one of the most common PoC technologies utilized is BCMA. In BCMA, medication is prepared and delivered to the patient by scanning the patient's armband and the medication. An electronic process occurs that verifies the five rights of medication administration using a barcode reader. The reader verifies the patient information and the medication information; the system then proceeds to double-check for any dmgdrug interactions and allergies and then verifies that this is the right patient and the right time as the patient is scanned with the barcode reader. BCMA has significantly reduced medication errors, with studies indicating ranges from 20% to 50% improvement in error rates (Poon et al., 2010). Barcoded blood administration is also becoming increasingly
common within the EHR. This process is similar to BCMA, except that with blood administration, the specific unit of blood is associated with the patient via the product identification and the blood administration time and type.
IV Smart Pumps
IV smart pumps at the acute care hospital bedside are becoming common devices that nurses manage daily. There are many challenges with IV smart pump integration, including proper patient identification, new interfaces to and from the EHR to manage medications, dosages, times of administration, as well as CDS for clincians. CDS for smart pumps creates many advantages for accurate IV medication administration but can also create alert fatigue if clinicians are over alerted by IV pump CDS alerts (Harding, 2013). Communication Technologies
Communication technologies provide many benefits for bedside acute care nurses. Examples of these types of PoC communication devices include alerts from the EHR going to the nurses' smartphone device, nurse call systems going to the phone, and secure texting to facilitate provider-nurse communications. Studies indicate that these smart communication devices have been shown to improve clinician satisfaction, perception of efficacy, and clinical workflow (Przybylo et al., 2014).
Challenges With PoInt-of-Care Device Integration Mobile Access
Many PoC or integrated devices are mobile and require local area wireless technology, often referred to as "Wi-Fi." These devices access data via the wireless local area network
(LAN) to pull the data from the device to the patient's record. As mobility increases in the organization, Wi-Fi access needs also increase. Management of "dead" spots, where the Wi-Fi is inaccessible, slow, or drops the connection, is often a challenge and frustration for staff and, as a result, decreases efficiency. The infrastructure must be continuously updated, expanded, and maintained to ensure proper connectivity for all mobile devices using wireless integration of PoC data. Data Reliability and Validity
As with PoC testing, the clinician's competence to obtain the test and validate the data must be ensured; when data are electronically submitted to the EHR, the validity and verification of the data are critical. Validity checks are often built into the systems. To ensure accuracy, the clinicians must follow the PoC device's designed workflows
159
160
II: POiNT-OF-CARE TECHNOLOGY
according to the designed protocol for use. The management of data reliability and validity is often the responsibility of the department that "owns" the data, such as the lab being responsible for PoC blood sugar results or the blood bank being accountable for quality control of blood bag accuracy. In manual systems, quality control was part of the
workliow between the clinician and the host system. In
an electronic environment, it can
be challenging to maintain the quality and integrity of the data coming into and from the PoC .system, and these electronic systems do not mean that clinicians should not continue to institute clinically responsible validity checks. This data validity and reliability check is particularly critical when older legacy systems are integrated with new technology. Management of Rules and Alerts
CDS, through the use of mles and alerts for data obtained outside the defined parametens, is one of the safety advantages of the EHR. CDS can be used within the EHR but also has considerations with device and data integration. If mles and alerts are too frequent or do not require some type of response, it can be easy for the clinician to ignore them or fail to address the alert clinically. Defining the mles and alerts so that they are presented only when there is a tme safety concern is challenging with most systems. Nursing informaticists are vital in helping to define the parameters for these rules and alerts based on the clinical evidence. For example, a nurse scans the patient and the medication for administration and receives an alert that it is too early to give the medication. The nurse elects to give the medication despite the alert, overriding the alert. Sometimes these actions may be appropriate based on the nurse's clinical judgment, but there also can be safety concerns. Knowing when to put a required hard-stop in the system is an important consideration. A hard stop is an alert that the clinician receives that warns them against moving forward with a task until specific requirements are met. For example, a nurse navigates to a medication documentation screen and receives a hard-stop alert that this is the wrong patient for the dmg. In certain circumstances, clinicians can select an override for an alert, such as giving a medication early or late, and these are decisions that clinicians best make. Ongoing evaluation of rules and alerts to monitor how they are used, how many times they fire,
and whether the alerts are acted on correctly or are consistently overridden are critical
measures for evaluating the effectiveness of the CDS system. The maintenance of alerts and rules aligned with clinical guidelines is also best managed by clinical informaticists.
It is important that pharmacy staff have input into the system and pharmacy leadership is involved in designing alerts and rules that involve medication administration. These efforts are the role of the interprofessional teams designing the CDS systems for organizations. CDS strategies are further covered in Chapter 19. Overreliance onTechnology
The safety functions of the EHR are well documented (Sittig & Ash, 2007), and especially within PoC and device integration, safety features can be dramatic. It is not uncommon to see reductions in medication errors of over 50% with BCMA (DeYoung et al., 2009). Along with this comes the expectation by clinicians that these systems are fail-safe, and it may be easy for clinicians to miss an alert or not follow the correct workflow
because they believe the system will prevent them from making an error. Staff education must emphasize that these systems are only a tool, an adjunct to care; it is essential to keep this in mind when using PoC technology and the EHR, just like any other piece of equipment.
7s ELECTRONIC HEALTH RECORDS AND POINT-OF-CARE TECHNOLOGY
Manual Association of Devices to the Patient
One of the challenges regarding device integration is identifying the device v^^ith the
patient. Often, devices such as ventilators, critical care monitors, blood pressure machines, and anesthesia machines must be linked to the right bed or patient. The clinician may have to select the device name to associate it to the patient within the EHR. Challenges arise when patients move, such as transferring to another level of care or when the patient is discharged. If the device is not removed from that patient or bed, then the data from the device could be downloaded to the wrong patient's EHR. With many devices, the technology will automatically associate the device to the patient. However, in some systems today, the clinician must perform a step to associate and disassociate the patient to and from the device and ensure that the data coming from the device into the record are being recorded for the correct patient. Evaluating Validity of the data and Clinicians Acting on Results
There are also challenges with evaluating data validity, results reporting, and clinicians appropriately acting upon the results reporting. For example, unlicensed personnel may obtain the data recorded in the EHR, and communication of the values to the nurse or
provider may be slow or may not occur at all. An example might be when a nursing technician obtains the reading of a PoC blood glucose, and the value is directly integrated into the EHR. If the value is high or low, the technician may not notify the nurse directly, assuming that the nurse will see the value in the record. Or nurses may accept values coming from monitors, such as vital signs or other cardiac devices, and if not properly reviewed and acted upon, they could potentially accept values that are not true. Examples might include a high heart rate caused by movement, or they could accept data without noting that action should be taken, such as when low blood pressure is recorded. It is easy to get the data into the EHR with PoC technology and device integration; often, data may not be acted upon, analyzed, or communicated. HIT is not a replacement for critical thinking and professional nursing practice. These PoC devices integrated with the EHR can provide increased efficiencies for the clinician and increased patient safety when used correctly. However, the systems require updated policies, procedures, and workflows to help the clinician use PoC testing and device management accurately, respond to niles and alerts appropriately, and follow the defined best procedures. Nursing informaticists are critical in helping define the best practices, providing education and training to staff on their importance, and ensuring nursing policies are adapted to reflect the requirements related to the technology and the most up-to-date evidence. Sustainability and Change Management
EHRs are typically implemented over the course of years, although we have seen many organizations begin implementing the big-bang approach, defined as rapidly adopting an implementation strategy into an organization in as few as 3 to 6 months. Additionally, given the multiple PoC devices that continue to improve and expand with options to integrate into the EHR, the re.sult is that the "complete" system does not exist. The implementation of the EHR is never complete. There are always additional
projects, new functionality, new technology, changes in regulatory requirements, and
user workflows that require ongoing management to sustain the overall patient record. Although projects have definite start and end times, the EHR is a series of projects that must be assessed and managed through the life cycle of the EHR (see Chapter
I6i
162
II: POINT-OF-CARE TECHNOLOGY
8 for SDLC specifics). This section addresses some of the components of ongoing
management of the EHR or sustainability. Change management is the process that organizations use to manage the ongoing requirements and development of their systems. In the EHR, change management may consist of the changes to the various applications or systems that comprise the EHR. It may also be changed workflows or policies and procedures that affect the technology. Organizations develop change management processes to accommodate the ongoing sustainability and Riture development of their systems; this process may be based on specific models of change management or internally developed. Change management typically follows a basic life-cycle process of assessing the problem, or area to be changed, designing the solution or the change needed, training and implementing the change, and then evaluating the impact and success of the change. However, we cover key areas that are often a high priority in change management: maintenance, regulatory changes, patient safety, end-user requests, and strategic planning. Maintenance Requests
Maintenance change requests are a part of the normal organization processes that continue to evolve as part of everyday practice. Maintenance may include adding lab tests to an order panel and changes in medication dosage recommendations from the FDA. Maintenance requests are often the most prevalent but do not take the tremendous resources that the build and training phases of implementation do. They are often
considered the day-to-day business of managing the
system.
Regulatory Changes
There are also required changes that are governed by regulatory requirements that need to be incorporated into practice, which may have a significant impacton the resources needed to create and manage these changes, as well as to educate and incorporate these changes into practice. There may be new changes from The Joint Commission that impact the EHR and clinical practice, as well as ongoing recommendations for coremeasures. Surgical Care Improvement Project (SCIP) measures, MU, and other regulatory requirements. Often, the implementation of these changes has time requirements; managing these changes effectively and efficiently creates additional challenges to staff. Examples of regulatory
changes might include implementing functionality to accommodate medication reconciliation, managing immunization reporting requirements, or improving the ability to provide electronic patient discharge instructions. An important regulatory requirement change for EHRs is the requirement to report electronic clinical quality measures, or eCQMs. These requirements have significant implications for clinical documentation and will be covered in Chapter 23 extensively. Patient Safety Changes
Patient safety is certainly the top priority within the organization and managing the EHR to maintain and support patient safety is a critical function for informaticists. Changes may be needed to help promote patient safety or prevent a technical functionality or workflow that can impact patient safety. Change requests that have a patient safety impact are often given the highest priority in terms of available resources to manage the
change. An example of this type of patient safety issue is removing dmgs from order sets
7: ELECTRONIC HEALTH RECORDS AND POINT-OF-CARE TECHNOLOGY
that have been determined to be unsafe. Often safety issues are managed by introducing
alerts or stops in the system, which require the user to interact with the system in a very specific manner to stop an error from occurring. User Requests
Specific end-user requests comprise a significant number of calls for updates to the system. End-users are important stakeholders to ensure that clinicians and staff are satisfied with the system's usability. Examples of these common requests are additional selection options in a structured list of assessment criteria or interventions, order sets developed, and clinical flowsheets moved, added, or deleted. Users may have very good knowledge of the system's functionality and present excellent ideas. However, they may also request changes that would negatively impact the system and directly conflict with other requirements or system functionality. Processes should be in place to manage user requests and ensure that they meet the organization's needs and enhance the workflows and functionality of the system. Often, user requests may actually conflict with each other or have a downstream impact on some other workflow. Prioritization and use of decision trees as to how the organization makes these decisions are important considerations when managing scarce resources in HIT staff. Governance
In order to have a successful change management and sustainability plan, a strong governance model is fundamental. Governance consists of the leadership and organizational structures and processes that ensure that the organization's information technology (IT) sustains and extends to the organization's strategies and objectives, Governance programs can help define the prioritization of IT projects, how changes are identified and managed, and the approval process for requests. Governance programs provide the structure around the change process, including the organizational goals. Important questions for organizations to consider related to governance are as follows: (a) To what extent do the end-user clinical staff provide inputs or contribute to decisions about changes? (b) How are changes communicated to staff? (c) How often are changes put in place? and (d) How are the resources to make the changes allocated? A governance program defines all these requirements and provides methods for how requests are entered, reviewed, approved (or denied), prioritized, implemented, communicated, and put into production. Although developing a governance program .seems straightforward, it is often complicated to create and manage .such a program. Staff need a way to make requests and have those requests approved promptly. Often, the requests far exceed the available resources to meet the demand, and it is difficult to know which requests should move forward. Also, having end-user input is important to any governance program. However, it is often challenging to define user groups and maintain diem over time, so getting consensus on requests from users may be difficult or slow. Nur.sing informaticists have a key role to play in helping manage the governance process. They are uniquely positioned to facilitate the management of the clinical need for requests, manage end-user expectations, assist IT with translating requests into functional requirements, and manage training and communication to staff who are affected. A governance program generally evolves through many iterations, and nursing informaticists are a crucial part of any governance discussion and process to achieve success.
163
164
il: POINT-OF-CARE TECHNOLOGY
Downtime Contingency Planning for the Electronic Health Records
The unavailability of the EHR can represent a potential patient safety event, disruption of patient care, or negative impact on the clinician's work. Organizations must have a plan in place to manage both documentation and access to patient information for clinical decision-making in the event the system is unavailable, resulting in a system "downtime." An organization's policies and practices for managing continued patient care and business practices when systems are unavailable is the business continuity plan (BCP). Organizations must have BCPs in place as a component of the Health Insurance Portability and Accountability Act (HIPAA) security rule, and staff should be aware of the practices outlined in the contingency plan to maintain safe patient care should they face a system downtime (Ruano, 2003). There are several reasons why the EHR may be unavailable. Often systems require downtime for maintenance, upgrades, or other planned events. These planned downtimes are coordinated between the technical and clinical resources to manage the downtime best. This approach is typically coordinated during the night shift when there is the least impact on patient care and busy daytime activities. These types of downtimes are communicated, and staff should be aware of maintaining ongoing access to information during the downtime. An unplanned downtime may occur when there is any interruption in access to the system. This downtime could be due to an application failure, such as the application itself failing for some reason. In this case, other systems may be operational, but one or more systems are offline. Another type of downtime is a network failure, where the systems may be working, but there is no access to the applications by the clinicians due to the network failure. Many systems may be impacted in this type of downtime, including all computers and other applications that am on the hospital network. Organizational policies should guide staff on how to respond and manage when the system is down. This approach may include transitioning to "downtime procedures"; for example, how to view historical data, print patient information reports, access paper-based forms for documenting, and downtime recovery. EHRs have functionality that allows the user to either view data during downtime or print out reports with patient data up until the time of the downtime. It is up to the organization to define the practices around this functionality. In addition to knowing when to move to downtime procedures and what must be entered into the EHR after the downtime, it is also critical to understand how to manage other procedures that a system downtime may impact. BCMA or other device integration and data transfer to other systems must be acknowledged and managed during downtime. Strategic Development
Finally, and perhaps most important, changes need to align with the organization's ongoing vision, mission, and future strategic goals. Strategic plans should include the
impact on the EHR, and future development and additional implementations should be considered in the ongoing management of the EHR. In addition, planning for regular upgrades and enhancements to the system should be part of the strategic plan budgeted and planned for on an ongoing basis. HIT SAFETY AND THE SAFER GUIDES
Safe HIT has several criteria to help minimize the potential for a system failure or downtime of the EHR resulting in risk to patient safety. The criteria are presented as checklists geared for self-assessment. The recommended practices are outlined in the Contingency Planner Safety Assurance Factors for the EHR Resilience (SAFER) guide.
7: ELECTRONIC HEALTH RECORDS AND POINT-OF-CARE TECHNOLOGY
This guide is designed for self-assessment in nine areas to support organizations in safer use of EHRs. One of these nine areas is contingency planning. There are worksheets for each of the nine areas that begin with a checklist for recommended practices. We
encourage the reader to visit the ONC's website and walk through all of the SAFER guide principles, complete with interactive references and supporting materials, and guide the end-user through how to assess an organization (ONC, 2014). The SAFER guide breaks the assessment into three phases. The first phase relates to hardware, equipment, paper backup, data and software backups, policies, and procedures for patient identification. Phase 2 of the assessment relates to using HIT safely, including staff training on downtime and recovery procedures, communications strategies, policies, and procedures on the
continuity of safe patient care; the user interface of the maintained backup or read-only EHR is clearly differentiated. The third phase is focused on the monitoring and testing
approach used to prevent and manage EHR downtime. The authors suggest that the reader review the SAFER guide on the HealthIT.gov website for detailed information and tools on how to address patient safety and health IT through this guided process (www. healthit.gov / buzz-blog / electronic-heal th-and-medical-record s / safer-guides-optimizesafety). Informatics Nurse Specialists (INSs) and NI nurses often initiate or lead teams to address the use of the SAFER guides, and many of the methods noted to address EHR optimization with new APIs and device integrations. INFORMATICS NURSE SPECIALISTS, INFORMATICS NURSE, AND SUPERUSER ROLES
Nursing Informatics' Role The nursing informatics (NI) role is critical to success because the NI content expert understands and can assess the clinical aspects of how the EHR and PoC technologies affect the clinicians' and the patients' experience. The INS subject matter expert understands both the workflows and the system functionality and is uniquely positioned to lead teams by providing expertise on solving problems, establishing best practices, and adapting policies and procedures to support the defined workflows of the EHR and other points of care technologies. These specialty functions are critical to optimizing the EHR in the clinical setting and aligning with the NI role and competencies discussed in Chapter 2. Additionally, core competencies needed to optimize EHRs and other PoC devices and APIs are fundamental to the scope of practice for the NI specialist. For example, SDLC (discussed extensively in Chapter 8), project management, data management, and analytics, computer science, human factors science, and nursing science underpin the NI specialist's practice. The INS supports teams and the clinical setting to improve the way technologies work for clinicians.
Impact of Nursing Informatics NI has a significant impact on all aspects of the EHR. From project planning to implementation to sustainability, the NI is critical to the organization's ability to manage its EHRs. Nursing informaticists in hospitals have the primary responsibility of working with clinical staff and promoting the use of the EHR in clinical practice, helping ensure proper workflows, adoption, and identification of optimization opportunities. They must understand both the clinical side of patient care and the EHR to promote and support patient care while also understanding the IT needs and requirements for ensuring a solid EHR development and building while managing IT resource availability and change management processes. Over time, the NI can establish an identity, not just
165
166
II: POINT-OF-CARE TECHNOLOGY
within nursing and within IT, but as a unique profession and contributor to a body of knowledge used to advance technology and the EHR for clinicians and improve patient outcomes. Another vital role for many organizations to optimize and improve HIT for clinicians is the superuser. Superusers' Role Superusers are designated to be a resource to assist end-users within their units or departments. Supemsers arecritical to thesuccessof an EHR implementation. They provide staff support and are the first-line resource to assist end-users with fundamental questions about workflow and functionality. Superusers usually volunteer to be superusers or are
asked by their managers. Supemsers are individuals with a strong interest in the EHR and technology. They often matriculate into NI roles within organizations. Supemsers
are experts in the system's correct workflows and functionality, are given time to stay
current with changes and provide ongoing support to staff. Supemsers usually provide support to a group of staff or users, either within a department or within a unit, and serve
as the first line of support or the "go-to" person for all staff. In addition, it is important for supemsers to meet as a group routinely and stay abreast of issues or concerns rai.sed so they can address them. They also are typically the first ones to get information on fixes, approved workarounds, and policy changes, so they can communicate them back to their staff. The supemser is the pivotal point person during the go-live and should have no other responsibilities than providing support and keeping staff updated with information on key changes needed to move forward. Over time, supemsers stay engaged and committed to the EHR and get used to thinking about expanded use of technology, such as incorporating devices such as IV pump integration into the EHR platform. Supemsers continue to grow in their expertise of the system and remain in the .support role for the unit or department for all matters concerning the EHR. Often, supemsers are responsible for communicating changes and ongoing reinforcement of workflows, policies, and EHR functionality. In addition, they may provide input on needed changes, optimization requests, and staff perceptions of changes. Supemsers and the nursing informaticists can ensure the ongoing adoption, optimization, and support of the EHR. Because supemsers stay engaged even after implementing the EHR and continue to support both the staff and the efforts of EHR optimization and changes with growth in the expertise of the EHR, many supemsers are nurses that move on to pursue professional NI roles. BENEFITS OFTHE ELECTRONIC HEALTH RECORDS FOR IMPROVING SAFETY AND QUALITY
Although federal programs of MU and PIP have been key drivers of EHR adoption, many of the benefits of the EHR with volumes of research published to support these benefits have been a factor. As organizations deploy these systems and are fully integrated and adopted throughout an organization, the benefits are realized by providers, nurses, other clinicians, and even patients and families. Studies indicate that nurses experiencing even early stages of the former MU program measures for adoption of EHRs are more likely to be satisfied with their EHRs due to many of the benefits experienced with more mature and usable EHRs (McBride et al., 2017). Some of the expected benefits with EHRs are the
following:
■ Improve the quality of patient care ■ Increase patient/consumer participation in care
7: ELECTRONIC HEALTH RECORDS AND POINT-OF-CARE TECHNOLOGY
■ Improve the accuracy of diagnoses and health outcomes ■ Improve care coordination ■ Increase efficiencies and provide cost savings
■ Provide predictive and prescriptive tools to provide awareness and alert changes in patient condition or health risks Features of Certified Electronic Health Records That Promote Patient
Safety and Quality A certified EHR keeps a record of a patient's medications or allergies, automatically checks for problems whenever a new medication is prescribed and alerts the clinician to potential conflicts. Information gathered by a primary care provider and recorded in an EHR tells a clinician in the emergency department about a patient's li fe-threatening allergy. The emergency staff can adjust care appropriately, even if the patient is unconscious. EHRs can expose potential safety problems when they occur, helping providers avoid more severe consequences for patients and leading to better patient outcomes. EHRs can help providers quickly and systematically identify and correct operational problems. Identifying such problems in a historical paper-based setting is much more complex and correcting them could have taken years. The quality of patient care is positively impacted by using CDS consisting of rules and alerts to manage the breadth of information within the EHR. CDS may be dmg-allergy checking or advisories for the management of certain conditions. CDS tools within the EHR are ubiquitous now and drive much of the patient's care by presenting information in an organized way to facilitate clinical decision-making. Indeed, the legibility of the record, real-time access from virtually any location, and integration of data from other systems promote quality initiatives within the organization. Reporting is also enhanced as more robust patient information is more readily available through various reporting tools. Also, CPOE and BCMA have removed many of the process and human factor effects of patient care, leading to a reduction in errors and improved quality (Appari et al., 2012). The additional integration of medical devices such as physiologic monitors,
infusion pumps, beds, and scales now add to the ease with which information is fully integrated into a single healthcare record. The proliferation of these devices and the ease with which they can be integrated has helped increase safety and provide critical information to clinicians.
The use of EHRs has also increased patient and family engagement and participation in care. These types of tools were covered in Chapter 5 with the chapter Rilly focused on consumer engagement. Through online tools and portals, patients have access to their health information. They can message providers and other members of the healthcare team, request medication refills, schedule appointments, and view lab results, medication lists, and educational materials. All these initiatives result in a more positive patient
experience and EHRs can medical errors, comprehensive
a patient informed in their health management process. improve the ability of a provider to diagnose disease and reduce thus improving patient outcomes. The EHR provides an accurate and picture of the patient and allows for faster diagnosis and treatment through timely access to information and aggregation of key data elements to drive care decisions (Jamoom et al, 2011). EHRs do not simply contain or transmit information; they "compute" it. This computation means that EHRs manipulate the information to improve outcomes in several critical functional requirements. We review them to highlight some of these features of certified EHRs under the federal guidelines and standards.
167
168
II: POINT-OF-CARE TECHNOLOGY
Care Coordination Supported by Electronic Health Records EHRs can help provide care coordination across an encounter of care or the entire continuum of care. Multiple providers and clinicians have access to the same information simultaneously and can work together, providing documentation on a variety of problems, interventions, and outcomes. Care can be passed from one provider to another or from one episode of care to another, with all providers having the same access to the same information. Alerts can also notify and manage care transitions and manage medications, problems, and treatments across multiple care settings. Provider offices can be notified when patients enter or leave a care setting; care summaries can easily be provided among physician practice settings, hospital setting.s, or long-term care or home-care settings. Patients can be identified when they return to a care setting, and notifications sent to alert providers of a patient's change in status. The po.ssibilities for care management of patients across the entire continuum of care are changing the face of healthcare, improving patient outcomes, and reducing costs (Bell & Thornton, 2011).
Electronic Health Records and Predictive Analytics There is also a move to utilize more predictive and cognitive analytic tools within the EHR with advanced algorithms. These algorithmic tools look for information within the EHR to implement rules based on evidence-based protocols and present the clinician with aggregated information that signals a specific condition. For example, predictive analytic tools to detect patient deterioration and sepsis are based on changes in vital signs, laboratory values, diagnosis, or symptoms associated with sepsis. These predictive tools are transforming how caregivers respond to the information documented, and the EHR is becoming more of a tool to support clinical decisions rather than a repository of health information.
ELECTRONIC HEALTH RECORDS' NEGATIVE IMPACT ON CLINICIANS AND A NATIONAL AGENDA TO ADDRESS THE ISSUES
While there are many positive benefits with EHRs and PoC devices, there are also perhaps equally as many emerging issues. The following section addresses the downside to EHRs and its impact on clinicians' well-being and satisfaction, and disaisses research addressing a national imperative to address usability with EHRs for nurses, burden of documentation with all clinicians, and a national agenda to address the challenges. Usability Issues With Electronic Health Records
Many studies are indicating providers and nurses report usability issues with EHRs (Kutney-Leeetal., 2019;Staggersetal, 2018). The burden of documentation, dissatisfaction, and burnout is often associated with the burden of documentation (Bodenheimer &
Sinsky, 2014; McBride et al., 2017; Moy et al., 2021). The NAM calls for action to address the alarming rise of burnout among clinicians naming inadequate technology usability as a significant contributor (Kutney-Lee et al., 2019).
Triple Aim Moves to a Quadruple Aim to Address Clinician Well-Being For over 14 years, the national focus has been on a Triple Aim to improve health, quality, and reduce healthcare costs in the United States. In 2007 the Institute for
7: ELECTRONIC HEALTH RECORDS AND POINT-OF-CARE TECHNOLOGY
Healthcare Improvement (IHI.org) developed the Triple Aim with the primary focus of improving patient care. The Triple Aim, adopted by our national quality strategy, provides a framework for improving population health, the patient experience, and reducing healthcare costs per capita (Bachynsky, 2020). However, with the focus on advancing the nation's health, the health of the nation's providers has also come to the forefront. According to Brodenheimer and Sinsky (2014), healthcare workers frequently experience burnout and dissatisfaction in the workplace. For many healthcare professionals, one of the main issues contributing to burnout is the amount of clerical documentation needed in a client's EHR.
For example, Adler-Milstein et
al. (2020) state, "EHRs have been cited as a contributor to clinician burnout, and self-
reported data suggest a relationship between EHR use and burnout" (p. 531). As such, there is a national focus on reducing documentation burden of EHRs. While benefits are present with EHRs, the negatives that have resulted from rapid adoption must also be addressed.
Human Factors Science, Usability, and the Electronic Health Record Human factors science is a scientific practice discipline that studies the use of humans
and how they interact with elements within a system (HFES.org, 2021). Research and improvement strategies examining human factors science can improve understanding of how clinicians use and interact with HIT. Human factors science includes usability of the system. Usability involves design, interaction, and evaluation of clinicians' use of HIT including the EHR and other PoC devices. HIMSS defines usability of the EHR as "the effectiveness, efficiency and satisfaction with which specific users can achieve a specific set of tasks in a particular environment" (HIMSS, 2021). Very simply stated, usability is about improving the user's experience. Studies indicate that the current certified EHRs have usability issues that result in clinician dissatisfaction. Clinician's report stressful work environments, burnout due to burden of documentation and patient safety challenges from EHR vendor products that are suboptimal in design, development, or implementation (Aiken et al, 2011; Dyrbye et al, 2017; McBride et al, 2017; Poghosyan et al, 2017). According to Staggers et al. (2018), there is an imperative to address usability of EHRs for nurses.
Research and Initiatives to Address Clinician Dissatisfaction
The massive and rapid adoption of EHRs has burdened clinicians due to design, configuration, and implementation issues. These challenges result in clinician dissatisfaction due to poor usability, clinical workflow challenges, and documentation burden (Gettinger & Zayas-Caban, 2021). Research and national initiatives are focused on addressing these challenges. For example, Sieja et al (2019), investigated the use of a novel clinic-focused Sprint process (an intensive team-based intervention) to optimize the EHR efficiency and improve end-user satisfaction. Sprints were used as a quality improvement (QI) strategy with three primary components: (a) training clinicians to use existing EHR features more efficiently, (b) redesigning the multidisciplinary clinical workflow within the clinic, and (c) building new specialty specific EHR tools. This team utilized
agile methods as the guiding strategy. Agile is a method for rapid cycle improvements focused on the voice of the customer (Dennis et al, 2019, p. 47). This research team reported positive results that improved teamwork, reduced EHR burden, and increased clinicians' satisfaction with their EHR.
169
170
II: POINT-OF-CARE TECHNOLOGY
This project is notable for optimization as this team used QI methods coupled with system analysis and design often associated with information systems development. Agile is an example of these types of information systems development methods, whereas workflow redesign is not only systems development but also a method for improving the quality of care (see Chapters 9 and 21 for more information on workflow redesign and QI tools).
Other researchers investigating EHR burden of documentation that resulted in stress and burnout examined design and use factors to determine how to optimize and improve the EHR within primary care clinics. This team found that EHR design and use factors influence clinician stress and burnout. Other challenges explained more of the variance with clinicians. These factors included chaotic clinical atmospheres and workload controls. Once again, this study would point to using QI methods to improve the EHR for clinicians (Kroth et al., 2019).
Further, to address these challenges, the National Library of Medicine funded an initiative focused on reducing doaimentation burden 75% by 2025 for U.S. clinicians. The initiative "25 by 5: Symposium to Reduce Documentation Burden on U.S. Clinicians
by 75% by 2025" had the following specific goals: 1. Engage a diverse group of key stakeholders and leaders focused on reducing the documentation burden. Stakeholders included federal agency representatives from the Centers for Medicare and Medicaid Services, Centers for Disease Control and Prevention, and the ONC.
2. Assess the likely potential for burden reduction within categories of documentation burden, including identifying "low hanging fruit" for "quick wins" without adversely impacting quality or access to care. 3. Establish approaches for immediate (less than 3 months) and short-term (6 months) reduction in clinical documentation burden.
4. Generate approaches to longer-term (10 years) elimination of clinical documentation burden (Rossetti & Rosenbloom, 2021).
With this initiative, there are also principles of engagement. The first principle is to leverage technology and existing data inputs where appropriate. The second is no erosion of care standards. The third is to maximize
the clarity of proposed rules to minimize misinterpretation by health systems and providers. The fourth is no wholesale shifting of work from one clinician to another, seeking to eliminate unnecessary documentation (Rossetti & Rosenbloom, 2021). As a result of the symposium, a report is being written summarizing major findings and recommendations due out in late 2021. (The website for the initiative for more details is https;//www.dbmi.columbia .edu/23x5/.)
While the United States has come a long way in adopting and implementing EHRs, work remains to be done to fully optimize and connect EHRs across regions, states, nationally, and internationally. HIEs have not been covered in this chapter but will be fully covered in Chapter 14 regarding how HIEs play into the overall strategy of U.S. interoperability for full optimization and value for the Health IT infrastructure currently in place. We will now apply many lessons learned in this chapter to a case study.
7: ELECTRONIC HEALTH RECORDS AND POINT-OF-CARE TECHNOLOGY
CASE STUDY
Device Integration Improving Patient Safety
A large healthcare system with a neonatal ICU (NICU) has experienced a serious patient safety error with the wrong mother's breast milk given to a neonate. The patient safety and quality improvement (Ql) department evaluates the incident, evaluates studies that address safe practices with NICU, and plans a Ql project to address the event. In examining best practices nationally, they discover that bar code medication administration (BCMA) is being applied to administration of mothers expressed breast milk.The Ql department solicits the expertise of the Chief Nursing Informatics Qfficer (CNIO).The CNIO assigns an informatics nurse specialist to the project to implement device integration at the point of care (PoC) to prevent any future errors from occurring and follow national best practices for using BCMA to administer mothers expressed breast milk to NICU infants. Given the BCMA NICU Device Integration project noted herein, consider the following questions: 1. How will PoC device integration prevent future errors from occurring with the wrong breast milk given to the wrong baby? 2. Why will human factors science considerations for this project team?
and
adoption
science
be
important
3. Why did the quality department call In the CNIO to help with this project? 4. How will the project team need to consider integrating the data to and from the EHR through an interface for PoC BCMA, and why is this important?
SUMMARY
This chapter has covered essential aspects of EHRs in the current clinical setting, particularly related to PoC device integration and optimization. We have briefly explored the history and current drivers of our HIT push for a fully connected health information infrastmcture throughout the United States. Benefits of the EHR for improving patient safety and quality have been discussed, particularly related to certified technology under the HITECH Act. We have paid significant attention to ways to adopt, implement, and maintain an EHR and the need for an ongoing optimization with new device and API
integrations. In addition, we have emphasized the importance of PoC technology that interfaces with the EHR and focused on important aspects of adoption, implementation,
and maintenance that organizations should consider as improvements are made to the system. This chapter aligns with best practices for SDLC and considers important safety factors such as downtime planning. Finally, we close with a case study examining BCMA for mothers' breast milk considering many of the lessons learned for a safe, effective approach to PoC device integration with the EHR.
171
172
II: POINT-OF-CARE TECHNOLOGY
REFERENCES
AddonData.com. (2017). What is Point of Care (PoC) technolog}/? https: / / www.addondata.com / 2017/10 / what-is-point-of-care-poc-technology/ Adler-Milstein,
Zhao, W., Willard-Grace, R., Knox, M., & Grumbach, K. (2020). Electronic health
records and burnout: Time spent on the electronic health record after hours and message volume associated with exhaustion but not with cynicism among primary care clinicians. Journal of the American Medical Informatics Association, 27(4), 531-538. https://doi.org/10.1093/jamia/ocz220 Aiken, L. H., Sloane, D. M., Clarke, S., Poghosyan, L., Cho, E,, You, L., Finlayson, M., Kanai-Pak, M., & Aungsuroch, Y. (2011). Importance of work environments on hospital outcomes in nine countries. International JournalJbr Qualifi/in Health Care, 23(4), 357-364. https://doi.org/10.1093/intqhc/mzr022 Appari, A., Carian, E. K., Johnson, M. E., & Anthony, D. L. (2012). Medication administration quality and health information technology: A national study of US hospitals. Journal of the American Informatics Association, 79(3), 360-367. https:/ /doi.org/10.1136/amiajnl-20 11-000289 Atherton, J. (2011). Development of the electronic health record. American Medical Association Journal of Ethics: Virtual Mentor, 73(3), 186-189. http://virtualmentor.ama -assn.org/2011/03/mhstl-1103.html Bachynsky, N. (2020). Implications for policy: The triple aim, quadruple aim, and interprofessional collaboration. Nursing Forum (Hillsdale), 55(1), 54-64. https:/ /doi.ore/10.1111 /nuf.12382 Bell, B., & Thornton, K. (2011). From promise to reality: Achieving the value of an EHR. Healthcare Financial Management, 65(2), 50-56. Blumenthal, D. (2011). Wiring the health system: Origins and provisions of a new federal program. New England Journal of Medicine, 365(24), 2323-2329. https://doi.org /10.1056/NEJMsrlll0507 Bodenheimer, T., & Sinsky, C. (2014). From triple to quadruple aim: Care of the patient requires care of the provider. Annals of Family Medicine, 72(6), 573-576. https://doi.org/10.1370/afm.1713 Dennis, A., Wixom, B., & Roth, R. (2019). Systems analysis and design. John Wiley & Sons, Inc. DeYoung, J. L., VanderKooi, M. E., & Barletta, J. F. (2009). Effect of bar-code-assisted medication administration on medication error rates in an adult
medical intensive care unit. American Journal of Health-System Pharmacy, 66(12), 1110-1115. https:/ /doi.org/10.2146/ajhp080355 Dick, R., & Steen, E. B. (1991). The computer-based patient record. National AcademiesPress. Dick, R. S., Steen, E. B., & Detmer, D. E. (Eds.). (1997). The computer-based patient record: An essential technology for health care (Rev. ed.). National Academies Press. Dyrbye, L. N., Shanafelt, X D., Sinsky, C. A., Cipriano, P. F, Bhatt, J., Ommaya, A., West, C. P., & Meyers, D. (2017). Burnout among health care professionals: A call to explore and address this underrecognized threat to safe, high-quality care. NAM Perspectives, 7(7), 1-11. https://doi.org/10.31478/201707b Gettinger, A., & Zayas-Cab^n, T. (2021). HITECH to 21st century cures: Clinician burden and evolving health IT policy. Journal of the American Medical Informatics Association, 28, 1022-1025. https://doi. org/10.1093 / jamia / ocaa330 Harding, A. D. (2013). Intravenous smart pumps. Journal of infusion nursing: The official publication of the Infusion Nurses Society, 36(3), 191-194. https://doi.org/10.1097 /NAN.0b013e318288alcf Health Information and Management Systems Society. (2013). Wljat is interoperability? http:// WWW .himss.org/library/interoperability-standards/what-is-interopera bility Health Information and Management Systems Society. (2021). What is user experience in healthcare IT?
https://www.himss.org/resources/what-user-experience-healthcare- it Health Level Seven International, (n.d.). About Health Lei'el Seven International, http://www.hl7.org/ about / index. cfm?ref=nav
HFES.org. (2021). What is human factors and ergonomics. https://www.hfes.org/About-HFES/What-is
-Human-Factors-and-Ergonomics Jamoom, E., Beatty, R, Bercovitz, A., Woodwell, D., Palso, K., & Rechtsteiner, E. (2011). Physician adoption of electronic health record systems: United States, 2011 (NCHS data brief, no. 98). National Center for Health Statistics.
Jha, A. K., DesRoches, C. M., Campbell, E. G., Donelan, K., Rao, S. R., Ferris, T. G., & Blumenthal, D.
(2009). Use of electronic health records in U.S. hospitals. New England Journal of Medicine, 360(16),
1628-1638. https:/ /doi.org/10.1056/NEJMsa0900592 Kroth, P. J., Morioka-Douglas, N., Veres, S., Babbott, S., Poplau, S., Qeadan, F, Parshall, C., Corrigan, K., & Linzer, M. (2019, August 2). Association of electronic health record design and use factors with clinician stress and burnout. JAMA Network Open, 2(8), el99609. https://doi.org/10.1001/
jamanetworkopen.2019.9609 Kutney-Lee, A., Sloane, D.M., Bowles, K. H., Burns, L. R.,& Aiken, L. H. (2019, January). Electronic health record adoption and nurse reports of usability and quality of care: The role of work environment. Applied Clinical Informatics, 7(7(1), 129-139. https://doi.org/1 0.1055/s-0039-1678551 McBride, S., Tietze, M., Hanley, M. A., & Thomas, L. (2017, January). Statewide study to assess nurses' experiences with meaningful use-based electronic health records. CIN Computers Informatics Nursing,
35(1), 18-28. https://doi.Org/10.1097/cin.0000000000000290
7: ELECTRONIC HEALTH RECORDS AND POINT-OF-CARE TECHNOLOGY
Moy, A. J., Schwartz, J. M., Chen, R., Sadri, S., Lucas, E., Cato, K. D., & Rossetti, S. C. (2021). Measurement of clinical documentation burden among physicians and nurses using electronic health records: A scoping review, journal of the American Medical Informatics Association, 28{5), 998-1008. https://doi ●org/10.1093/jamia/ocaa325 National Center for Research Resources. (2006). Electronic health records ozvnne^o (No. NCRR-2006). National Institutes of Health, National Center for Research Resources, http://www.himss.org/
ResourceLibrary / ResourceDetail.aspx?ItemNumber=10878 National Institutes of Health, (2010), Point of care diagnostic testing (Fact Sheet No. NIH-2010). Author. Office of the National Coordinator for Health Information Technology. (2014). Safety assurance factors fi>r EHR resilience (SAFER): Self assessment contingency planning (No. SAFER-012014). https:/ / www .healthit.gOv/sites/defauIt/files/safer/pdfs/safer_contingencyplanning_sg003_form_0.pdf Office of the National Coordinator for Health Information Technology. (2016). Investing in the future: Neiv market-ready, user-friendly health technology app and infrastructure support, https: / / www.healthit. gov/buzz-blog/interoperability/investing'in-the-future-new-marke t-ready-user-friendly-health -technology-app-and-infrastructure-support/ Office of the National Coordinator for Health Information Technology. (2017). Health IT dashboard. https: / /dashboard.healthit.gov/quickstats/quickstats.php Office of the National Coordinator for Health Information Technology. (2018). Health IT playback. Section 1: Electronic health records, https: / / www.healthit.gov/playbo ok/electronic-health-records/
Poghosyan, L., Liu, J., Shang, J., & D'Aunno, T. (2017). Practice environments and job satisfaction and turnover intentions of nurse practitioners: Implications for primary care workforce capacity. Health Care Management Reiueui, 42(2), 162-171. https; / /doi.org/10.1097/HMR,0000000000000094
Poon, E. G., Keohane, C. A., Yoon, C. S., Ditmore, M., Bane, A., Levtzion-Korach, O., Moniz, T.,
Rothschild, J. M., Kachalia, A. B., Hayes, J., Churchill, W. W., Lipsitz, S., Whittemore, A. D., Bates, D. W., & Gandhi, T. K. (2010). Effect of bar-code technology on the safety of medication administration. New England Journal of Medicine, 362(18), 1698-1707, https;//doi.org/10.1056/NEJMsa0907115 Przybylo, J. A., Wang, A., Loftus, R, Evans, K. H., Chu, I„ & Shieh, L. (2014). Smarter hospital communication: Secure smartphone text messaging improves provider satisfaction and perception of efficacy, workflow, journal of Hospital Medicine, 9(9), 573-578. https: / /doi.org/10.1002/jhm.2228 Rossetti, S.,&Rosenbloom,T, (2021). 25 by 5: Symposium to reduce documentation burden on US Clinicians by
75% by 2025. https://www.dbmi.columbia.edu/wp-content/uploads/2021/01/25x5-Symposium
-Intro.pdf Ruano, M. (2003). Understanding HIPAA's role in business continuity, disaster recovery. Confidence, 12(10), 3. http://library.ahima.org/xpedio/groups/public/documen ts/ahima/bok3_005209.hcsp? dDocName=bok3_005209
Sieja, A., Markley, K., Pell, J., Gonzalez, C., Redig, B., Kneeland, R, & Lin, C. T. (2019, May). Optimization sprints: Improving clinician satisfaction and teamwork by rapidly reducing electronic health record
burden. Mayo Clinic Proceedings, 94(5), 793-802. https://doi.Org /10.1016/j.mayocp.2018.08.036 Sittig D. F., & Ash, J. S. (2007). Clinical information systems: Overcoming adverse consequences. Jones & Bartlett.
Staggers, N., Elias, B. L,, Makar, E., & Alexander, G. L. (2018, April). The imperative of solving nurses' usability problems with health information technology. The journal of Nursing Administration, 48(4), 191-196. https://doi.Org/10.1097/nna.0000000000000598 U.S. Food and Drug Administration. (2014). Medical devices: Products and medical procedures, http;// www.fda.gov / MedicalDevices / ProductsandMedicalProcedures / default.htm The White House. (2004). Promoting innoi’ation and coinpetitii’eness: President Bush's technology agenda. http: / /georgewbush-whitehouse.archives.gov/infocus/technology
173
Systems Develo^rnent Liffe..Cycle ano Project Management to Optimize Technology SUSAN MCBRIDE, SUSAN K. NEWBOLD, AND DAVID FULTON
OBJECTIVES ●
Review the systems development life cycle (SDLC) as it relates to product development and system implementation and compare and contrast differences in development versus system implementation.
●
Analyze the SDLC framework and identify each phase and the components of each phase using a four-cycle approach.
●
Understand important tools and competencies necessary for informatics professionals to master SDLC.
●
Examine case studies using SDLC, aligning SDLC with meaningful use (MU) guidelines, incentives, and the certification program for electronic health records (EHRs).
●
Examine a case study using best practices for SDLC to demonstrate the use of the technique.
●
Discuss new methods available to design systems in rapid cycles and to use object-oriented methods for design.
●
Discuss the historical perspective of MU in terms of SDLC, comparing and contrasting the rollout of EHRs under MU federal guidelines, product development, and implementation as a case study examining the SDLC framework.
●
Compare and contrast SDLC to Professional Project Management as outlined Project Management Institute (PMI). CONTENTS
INTRODUCTION
177
SYSTEMS DEVELOPMENT LIFE CYCLE
PLANNING PHASE OF SDLC
System-Planning Goals System-Planning Tools Feasibility Studies Project Charter
177
178
179 180
181
181
Informatics' Roles in the Planning Phase of SDLC System-Planning Outputs
182
Implementation Committees andTeams
182
181
176
Ih POINT-OF-CARE TECHNOLOGY
ANALYSIS PHASE OF SDLC
System-Analysis Goals
182
183
System-AnalysisTools
183
Request for Information and Request for Proposal ScoringTools for RFis and RFPs Gap Analysis
184
184
184
System-Analysis Outputs
184
Informatics' Roles in the Systems Development Life Cycle Analysis Phase DESIGN PHASE OF SDLC
System-Design Goals
185
185
185
System-DesignTools
186
System Configuration in Design Phase Situational Analysis in Design Phase
186 187
Project ManagementTools in the Design Phase
187
Gantt Charts in Systems Development Life Cycle
187
Work Breakdown Structure and RACI Diagrams EHR Project Plan Samples
188
188
System-Analysis Outputs
190
Informatics' Roles in the SDLC Design Phase
191
Design Implications Under the Disability Accommodations and Americans with Disabilities Act Requirements 192 METHODS AND STRATEGIES FOR DESIGNING AND DEVELOPING SYSTEMS Waterfall Development
193
Rapid Application Development Agile Development
193
195
Testing Phase of Systems Development Life Cycle SystemsTesting Goals and Considerations System-Testing Tools
193
195
195
196
System-Testing Outputs
196
informatics' Roles and Skills in SystemTesting
196
System Implementation, Evaluation, and Support Phase Implementation and Support Goals
196
196
System Implementation, Evaluation, and SupportTools and Strategies
Three approaches to Implementation; Phased, "Big Bang" or Parallel Managing Expectations and Issues with Go-Live
Implementation, Support, and Evaluation Output
197 197
197
197
EDUCATION ANDTRAINING CONSIDERATIONS FOR SYSTEM IMPLEMENTATION Informatics' Roles in Implementation, Evaluation, and Support Phase of Systems Development Life Cycle 199 Metrics for System Evaluation System Maintenance
200
200
198
8: SYSTEMS DEVELOPMENT LIFE CYCLE AND PROJECT MANAGEMENT COMPARE AND CONTRAST A PROFESSIONAL PROJECT MANAGEMENT FRAMEWORK 200
WITH SDLC
Systems Development Life Cycle as a Component of the Informatics Certification Exams Comparison of Phases for SDLC and Project Management Within PMBOK
201
Phases of SDLC Mapped to PMBOK Five Process Groups andTen Knowledge Domains OPTIMIZING EHRs FOR INTEROPERABILITY IN OBSTETRICS CASE STUDY SUMMARY
201
204
204 206
EXERCISES AND QUESTIONS FOR CONSIDERATION REFERENCES
201
206
207
INTRODUCTION
The systems development life cycle (SDLC) is a standardized approach used to develop and implement information technology (IT). This framework is often used across industries to structure best practices for IT development and deployment. SDLC is a phased approach used to analyze and design information systems that is broken into distinct phases (Kendall & Kendall, 2014, p. 4). Health information technology (HIT) projects can be deployed haphazardly or can follow a structured and methodical approach, such as with the SDLC. The SDLC phases are essentially complementary with professional project management phases, requiring planning, analysis, design, implementation, and evaluation. All phases of SDLC are similar whether developing a product (as you would if you were a vendor) or implementing a commercial off-the-shelf (COTS) product requiring customization—both paths cycle through all phases of SDLC. This chapter outlines the SDLC framework, discusses each phase, and compares and contrasts development compared to implementing a product off the shelf. The chapter also examines SDLC compared to the Professional Project Management framework, examining similarities and differences. Finally, the chapter concludes with a case study that examines SDLC in a clinical use case for
promoting interoperability of EHRs under the Promoting Interoperability Program (PIP) instituted as a follow-up to the Health Information Technology for Economic and Clinical Health (HITECH) Act MU Program. SYSTEMS DEVELOPMENT LIFE CYCLE
SDLC is a methodology used to describe the process of building information systems. This approach offers a road map for developing information systems in a very deliberate, structured, and methodological way. The SDLC is also used in software development and is a process of creating or altering information systems. Figure 8.1 reflects the phases of the SDLC. We examine each of these phases and discuss the importance of these steps in deploying information systems. There are some differences in developing products and information systems compared to implementing new systems or upgrading systems. We discuss those differences within the chapter and point out the differences and similarities in each SDLC stage.
SDLC is a process used by a systems analyst or software engineer to develop an information system that includes extensive planning and analysis that informs the development and evaluation strategies for the information system requirements. One of the most important aspects of SDLC is aligning the intended cu.stomer's needs with the
177
178
II: POINT-OF-CARE TECHNOLOGY
FIGURE 8.1 Phases of Systems Development Life Cycle (SDLC),
Planning
Implementation, evaluation, maintenance, & support
Analysis
Systems development life cycle (SDLC)
\ System & functionality testing
1 Design & Implementation
deliverable; without aligning the intended purpose and need of the end-user, the project will fail. Successful projects meet or exceed customer expectations. Successful projects frequently require adaptation particularly as new information is gathered and some adaptive approaches rely on rapid SDLC to be more adaptive and responsive to evolving customer expectations. Meeting customer expectations is one of the primary goals of SDLC. To meet customer expectations, informatics teams need to consider human factors science in human-
Secures
3
workstation and leaves room
Record and
etiart changes
chief comi^aitrt
in vital ^gns
record
Records history,
(/]
MU ObiecUvt:
Enters vitals &
past mechcai, social, family, substance use
(smoking history),
MU Objectives:
r
Maintain active medication 6
medication
alleigyllst
Verifies & records
allergies & current medications
etc.
MtiODjective: Record smoking status for patients 13 years old or older
Performs chart review before
Enters the room,
entering exam
greets patient, and logs onto
room
workstation
Consults witii
patient and records HPi
Documents review
Performs physical exam
o( systems & physical exam imo EHR
Updates problem
list & triggers CDS rules If needed
A
MU Objectives: Maintain problem list ol current and
active diagnoses & 03
Provides patient
>
with instructions/ materials
o Q.
Places orders
Assigns Level of Service tLOS)
Imptement relevant COS rules
as necessary
(see Orders workfkw)
I
Closes the erxxxmler^ EHR
Note; Swimlane diagram noting patient swimlane, nurse/support, and provider for an office visit workflow,
5. Finally/ observe the process several times in the practice setting to validate that the workflow diagram represents all steps, roles, and responsibilities within the process. Frequently, end-users do not always realize what "everyone" does in a process. Observation helps verify what is occurring and not the perception of what the end user believes is happening. Again, issues in the process and inefficiencies are often identified through observation. 6. If the observer notes changes, they should be noted on the diagram and a third version of the diagram prepared. 7. Return to the end-user and use the new diagram as an educational opportunity
to inform all end-users of what is happening with respect to the workflow and technology depicted. Use the diagram to identify opportunities for optimizing the technology, improving quality of care, and/or improving efficiencies (eliminating and streamlining the process). Ambulatory Workflows
Ambulatory settings include clinics for specialty and primary care or may also include more complex multispecialty clinics and federally qualified healthcare clinics. The following is a description of typical workflows in the ambulatory areas that a provider will need to address when attempting to adopt and implement an EHR for a clinic setting:
219
220
I: POINT-OF-CARE TECHNOLOGY
■ Patient check-in ■ Office visit
■ Appointment scheduling ■ E-prescribing ■ Lab orders and results
■ Referral generation
■ Office discharge Acute Care Workflows
Acute care has a plethora of workflow processes that might benefit from qualityimprovement methods and workflow redesign techniques. We highlight a few of the common workflows that presented historical challenges with respect to EHRs and achieving MU in early stages where workflow redesign methods provided effective methods to successful deployment of EHRs. These workflows include admissions processing, medication management, medication reconciliation, computerized provider
order entry (CPOE), and patient discharge process. More recently, workflow redesign is being deployed as a technique for engaging clinicians in effectively capturing the right data/information, at the right time, for the right patient, in order to track adherence to evidence-based protocols and subsequent electronic clinical quality measures (eCQMs) as process and outcome measures. eCQMs and challenges associated with capturing valid data are discussed further in Chapter 23. The EHR can be used effectively to appropriately doaiment and subsequently trigger a clinical decision support (CDS) alert that would stimulate clinicians to respond to a protocol such as the Centers for Disease Control and Prevention (CDC) Guidelines for COVID-19
along with addressing workflow redesign in the process (CDC, 2021). CDS is a computerized mechanism based on rules coded within the system to digger alerts to clinicians on such things as dmg-daig interactions, overdosage, diagnosis and management, or activating a team to act on an ii-ifectious disease appropriately (Hoelscher & McBride, 2020; Mills, 2019). CDS is discussed extensively in Chapter 19. The result of these types of rapidly evolving public health crises and the importance of valid and reliable data for electronic quality measures have several federal agencies working to standardize an approach to ^rther automate workflow redesign. The goal is to automate treatment protocols and clinical standards through digitalizing clinical guidelines and alerts for best practice. National repositories are being developed to
advance this automated capability of HIT workflow redesign (Lomoton et al., 2020). NEW ADVANCES
Advances in modeling and analysis for healthcare are rapidly evolving due to work to capture valid and reliable electronic clinical data to improve the quality and cost of care.
Additionally, public health concerns are important consideration.s, such as emergent infectious diseases including challenges with Ebola, Zika vims, and more recently the pandemic associated with COVID-19 (Hoelscher & McBride, 2020; McBride & Tietze, 2014). Butler et al. (2014) describe methods to adopt standards such that workflow
redesign strategies can be more readily shared acro.ss healthcare organizations. Butler's research team effectively mapped the process within the ambulatory clinic to the level of detail that allowed complete automation of the workflow to improve the return to follow-up rates in a busy ambulatory clinic. These methods are emerging new standards likely to rapidly evolve with advancing HITs.
9: WORKFLOW REDESIGN IN A QUALITY-IMPROVEMENT MODALITY
Butler's team developed a Modeling and Analysis Toolsuite for Healthcare (MATH),
a method to model and analyze with automated approaches (tools) to make measurable improvements to clinical workflow through a predictable, integral system. The methods developed synchronize the flow of health information and the workflow of clinical care. Butler's team notes "when the flow of information matches
better workflow, significant
gains in quality and efficiency can be achieved" (Butler et al., 2014, p. 1). Further they also note that when the data or information flow contradicts or creates barriers to care,
it can rearrange clinical workflow by accident rather than by design. Nurses often encounter this type of interferences with technology and workaround the system. CASE STUDIES
The following case studies highlight the need for workflow redesign.The case can be utilized to think through how you might create a project charter for Improvement and construct a workflow redesign. Case; The clinical setting is a small community internal medicine practice that includes eight providers—four physicians, two nurse practitioners (NPs), and two
physician assistants (PAs). The practice is pushing hard to get to MU under the Medicare incentive program because the incentive dollars are significant to offset their EHR adoption and implementation expenses.Their largest patient population has Medicare and private insurance.The e-prescribing rates are very low at less than 10%. Provider interviews within the assessment process used to examine "as is" workflow status include the following with respect to findings:
Providers have sent prescriptions that have not been routed to the pharmacy and they do not understand why. Several of the providers refuse to use the EHR for e-prescribing and would rather handwrite prescriptions and give them to the patient. The senior physician is frustrated by the system and now refuses to use it, stating, "I will retire before I will use this system—it doesn't work and I will not use it until you get it fixedi" Other providers are not having as many issues with the system, but believe the system isn't working quite right and do not know whether the problem stems from the pharmacy, EHR, or the way they are using the system.The PAs and the NPs are not having as many issues as the physicians with transmittal of prescriptions. In addition, you have a serious incident your team is asked to address related to e-prescribing by a physician inexperienced with the protocols for a particular condition. Prescribing incident: Anew physician who has recently completed residency and joined the clinic is being oriented to the new EHR within the practice.This provider is very computer savvy and takes pride in EHR competency.This provider came from a highly automated practice and in one of the areas of the country most advanced in EHRs and health information exchange.The provider is overconfident with the use of the EHR and dismisses any assistance with training, stating, "I fully understand how to use an EHR and don't need your help In training—I've got this covered!" This physician had an incident that was reported to your team for review: Dr. Rookie has received a new patient, a Hispanic female with post-myocardial infarction by 6 months with subsequent congestive heart failure. She is 57 years of age, 225 lb, and 5 ft 5 inches. Dr. Rookie e-prescribed metoprolol 100 mg twice a day. After rethinking, the dosage was changed to metoprolol 50 mg twice a day. Dr. Rookie assumed the second order would override the original order and, therefore, did not cancel the original prescription, The pharmacy processed
221
222
II; POINT-OF-CARE TECHNOLOGY
and delivered both prescriptions to the patient.This resulted in two prescriptions for metoprolol with the patient dosage of 150 mg twice daily. The patient took both prescriptions for the medication for several days before presenting to the emergency department (ED) with severe hypotensive bradycardia. In the ED triage, the nurse received a brown bag of medications and noted two different prescriptions. She asked the patient whether she had been taking both prescriptions; she confirmed she had taken both for several days. The ED physician subsequently accessed the Surescripts network through the hospital EHR, confirmed the medication prescription history from the local retail pharmacy, and called Dr. Rookie about the double prescription. Dr. Rookie investigated the issue later that week to determine why the system failed expectations. When it was discovered that the system did not have a built-in ability to override the initial order on the same prescription, the physician became defensive and openly critical and reported to your team, "Get this thing fixed—the other system we used in my residency program would have caught this double prescription.This should never have happened!"
E-Prescribing Workflow Analysis
The multidisciplinary team has requested assistance from your local regional extension center (REC).The regional coordinator for the center (RCC) recommends a process redesign approach to e-prescribing.The RCC has recently come into the clinic and with your team's help has created a process redesign for e-prescribing. Outlined here is a workflow template based on e-prescribing during an office
visit (see Figure 9.5). Your team noted the following differences between the recommended workflow and the current practice:
1. When the end-user enters a special character into the e-prescribing field, the EHR does not recognize it. This results in the prescriptions not making it to the pharmacy.This issue frequently occurs when the prescription is written both in English and in Spanish for Hispanic patients. 2. Providers do not always double-check protocols or drug benefit information.
3. When a prescription follows an electronic pathway, the prescription is frequently not received at the pharmacy and consequently is not filled. 4. There is no clinical decision support (CDS) rule or EHR process to alert providers to block duplicate or revised orders. 5. Your team is not confident that the issue noted in Case 1, the presence of special
characters in the e-prescribing field, is the only reason for the lack of receipt by the pharmacy. However, it is the only issue that has been identified thus far.The team suspects other end-user issues, particularly given the various provider behavioral responses to the e-prescribing process. 6. Your team has preliminarily investigated issues and believes that there may be workflow issues at one large retail pharmacy within a superstore. Further investigation reveals that the practice sends a significant amount of business to this pharmacy. The large retail pharmacy has reported potential workflow issues that most likely affect the practice's e-prescribing process.
9: WORKFLOW REDESIGN IN A QUALITY-IMPROVEMENT MODALITY
FIGURE 9.5 Visio workflow diagram of an office visit e-prescribing process. I
c
£
^tl«nt afrVes
fetleni drops-oti
)
tntervlews patlenft
tx(s) at pharmac] or sends mall iXdK.
>
\
MU OeiecUva:
Checks patient's
Imglemsfll dmg-
Rx benellt
tonnulaiy
Information
chKka
Notifies padGPt that Rx(s| filled & ready for pi^-up
4~
ftx(s) filled
Receives Ri(s)
1. One of the smaller pharmacies owned by a prominent family within the community only receives electronic faxes. Roughly 30% of the prescriptions are routed to this pharmacy.
8. Using a multidisciplinary approach, your team's overarching goal is to develop a corrective action plan so that the practice achieves a rate of e-prescribing of 40% or better.
SUMMARY
This chapter has covered thebasics of workflow redesign and has placed this important tool within a framework for quality improvement, as well as public health crises. We have described fundamental tools such as the project charter as an important mechanism for defining what you intend to do to improve the process and how you will measure improvement and key components relevant to the redesign. Prior to beginning a workflow redesign project with any major impact to a practice setting, a project charter should be outlined. The chapter has also provided an important
overview of best practices for workflow redesign and has identified key areas where workflow maps should be created in order to achieve optimization of HIT in both acute care and ambulatory care settings. Further, we have described rapidly evolving development related to advanced uses of workflow redesign strategies to
automate processes and address global public health concerns such as the COVID-19 pandemic.
223
224
: POINT-OF-CARE TECHNOLOGY
END
1
CHAPTER
RESOURCES
EXERCISES AND QUESTIONS FOR CONSIDERATION
1. What are all the major issues noted in the practice setting? Is Dr. Rookie's recent incident and the workflow analysis relevant to successfully achieving a safe and effective use of an EHR?
2. Prioritize the issues and create a corrective plan of action using the project charter template within the chapter.
3. Define metrics within the project charter that will measure success before and after the implementation of the redesign.
4. Within the plan, provide the team's recommended approach as to how to further investigate unexplained issues and how to incorporate community partners in the plan. 5. Discuss how the corrective plan will be monitored and evaluated. 6. What metrics are at play in this scenario in addition to the e-prescribing metric noted in the case scenario?
7. Describe how a multidisciplinary team approach could improve care within the clinic and further issues with e-prescribing. ADDITIONAL RESOURCES
iPtlKtEIPUBLlSKIKi.
CONNECT
A robust set of instructor resources designed to supplement this text is located at http://connect.sprjngerpub.com/content/book/978-0-8261-8526-6 . Qualifying instructors may request access by emailing textbook@springerpub. com.
REFERENCES
Agency for Healthcare Research and Quality. (2017). What is workflow? https://www.healthit.ahrq. gov/health-it-tools-and-resources/evaluation-resources/workflow- assessment-health-it-toolkit/ workflow
Agency for Healthcare Research and Quality. (2019). Module 5: Mapping and redesigning workflow, https: / / www.ahrq.gov/ncepcr/tools/pf-handbook/mod5.htmI
Butler, K., Bahrami, A., Schroder, K., Braxton, M., Lyon, L., & Haselkom, M. (2014). Advances in workflow modeling for Health IT: The Modeling & Analysis Toolsuite for Healthcare (MATH). In J. Zhang & M. Walji (Eds.), Better EHR: Usabiliti/, workfloio and cognitive support in electronic health records.
The University of Texas Health Science Center at Houston (UTHealth).
Centers for Disease Control and Prevention. (2021). Informationfor healthcareprofessionals about Coronainriis (COVlD-19). https://www.cdc.gov/coronavirus/2019-nCoV/hcp/index. html Glasgow, J. M., Scott-Caziewell, J. R., & Kaboli, P. J. (2010). Guiding inpatient quality improvement: A systematic review of Lean and Six Sigma. Joint Commission Journal on Quality and Patient Safety, 36(12),
533-540. https: / /doi.org/10.1016/sl553-7250(l0)36081 -8
Hoelscher, D., & McBride, S. (2020, October). Usability and the rapid deployable infectious disease decision support system. Computers, Informatics, Nursing: CIN, 38(10), 490^99. https://doi
.org/10.1097/cin.0000000000000654 Jones, S. S., Koppel, R., Ridgely, M. S., Palen, T. E., Wu, S., & Harrison, M. I. (2011). Guide to reducing unintended consequences of electronic health records (No. HHSA290200600017I). Agency for Healthcare Research and Quality, https:/ /www.healthit.gov/unintended-conse quences/
9: WORKFLOW REDESIGN IN A QUALITY-IMPROVEMENT MODALITY
Keith, B., Bahrami, A., Braxton, M., Lyon, L., & Haselkorn, M. (2014). Chapter 11: Modeling for health IT modeling & analysis tool suite for healthcare (MATH). In J. Zhang, & M. Walji (Eds.),
Better EHR: Usability, worl^ow, and cognitive support in electronic health records (pp. 159-186). National Center for Cognitive Informatics and Decision Making in Healthcare, https://doi.org/ 10.13140/2.1.1921.1841
Lomotan, E. A., Meadows, G., Michaels, M., Michel, J. J., & Miller, K. (2020, January), To share is human!
Advancing evidence into practice through a National Repository of Interoperable Clinical Decision Support, Applied Clinical Informatics, 21(1), 112-121. https:/ /doi,org/10.1055/s-0040-1701253 Lucidchart, (2021), https:/ /www.lucidchart.com/pages/ Malane, E. A,, Richardson, C,, & Burke, K. G, (2019, November/December). A novel approach to electronic nursing documentation education: ambassador of learning program, journal for Nurses in Professional Development, 35(6), 324-329. https:/ /doi.org/10.10 97/nnd,0000000000000587 McBride, S., & Tietze, M. (2014). Nurse informaticists address Texas Ebola case, EHR design questions, https: / / ajnoffthecharts.com/nurse-informaticists-address-texas-ebola-cas e-ehr-design-questions McGrath, S. R, Perreard, I. M., Garland, M. D., Converse, K. A., & Mackenzie, T. A. (2019, March).
Improving patient safety and clinician workflow in the general care setting with enhanced surveillance monitoring. IEEE journal of Biomedical and Health Informatics, 23(2), 857-866. https: / / doi.
org/10.1109/jbhi.2018.2834863 Microsoft Corporation. (2019). Microsoft Visio Professional 2019. Author. Mills, S. (2019, Junuary). Electronic health records and use of clinical decision support. Critical Care
Nursing Clinics of North America, 32(2), 125-131. https://doi.Org/10.1016/j.cnc.2019.02.006 Technology, (2019). What is workflozo redesign? Why is it important? https: / Zwww.healthit.gov/faq/what-workflo w-redesign-why-it-important
Office of the National Coordinator for Health Information
225
226
II: POINT-OF-CARE TECHNOLOGY
APPENDIX 9.1 MICROSOFT VISIO PROFESSIONAL 2019 FUNCTIONS
Launch Visio, select Flowcharts from the templates that are available within Visio and select the type of flowchart you wish to create, such as the Basic Flowchart or Cross-Functional. We recommend Cross-Functional because this chart template has the swimlanes that allow you to distinguish roles. © New P
O
♦ S
m-m
● ^
*
You will have options to select Metric Units or U.S. Units, followed by selection of horizontal or vertical swimlanes. We typically use horizontal but it is your choice. II
a
s
Note the icons depicting shapes on the right that constitute best practice in terms of reflecting a start/end, process or decision step (these three are essential to most clinical workflow diagrams).
Drag and drop shapes onto the template, click "Connector," and drag another shape onto the template. Shapes reflected are ovals for start of process and rectangles for a process step.
9: WORKFLOW REDESIGN IN A QUALITY-IMPROVEMENT MODALITY
If you wish to create a cross-functional chart with swimlanes, click on "Function and add roles and responsibilities of the individual and add additional swimlanes by clicking on "Swimlane" and dragging it onto the template, where you wish to add the ff
additional lane. 0i
Site
II Owl
In)
Nursing
●v'T
Neighb^hood^
/
;
V-'' ■
/ .
O Community
■ ● Nonprofit organizations.
/,
-k'T"
horfies
W
' ./Ors^nizations '.
/
centers
I
-
/
/
Hospitajs ●\
\
Honie'heaith ■
/
\
\
' ■ ' Drug ^ .■ treatment
● Public health''■'
/
'
/agency'.
Laboratoh^
\ Doctors \
\
Law^
Faith institution
\
enforcement CHCs
/
0
●
\
’
;
● :0 Rre
Tribal health
Employers
Mental^health
Transit
Elected officials Corrections
Source: Centers for Disease Control and Prevention. (2010a). National public health performance standards. http://www.cdc.gov/nphpsp/theinstruments.html.
The challenge in developing a plan is the uniqueness of every community concerning things such as the prevalence of disease, a community's layout, city ordinances, county
laws, human capital, existing prevention efforts, and leadership support. All of these factors must be taken into account as a community considers a plan for health improvement, along with HIT, HIE, the penetration of EHRs in the community, and the availability of clinical data to inform the plan. Although the plurality of variables that need to be accounted for is challenging, the development and execution of CHNAs and community health improvement plans can be done with a very systematic approach to examining information within the community.
Community Asset Mapping and Geo-Mapping Methods for Assessment Community assessments should function as a tool that helps rebuild communities. In Building Communities from the Inside Out, John McKnight and John P. Kretzmann (1993) propose that communities should not be developed based on their deficiencies but on their strengths. The combination of community strengths and assets is a critical component of a community assessment, with identification of strengths and assets used to build on within each community. This helps identify community competencies and organizations
that can help improve community health. It also brings community ownership to the issues. Assets to be mapped are resources such as libraries, parks, recreation centers, businesses, churches, block clubs, cultural group.s, associations, and schools. Looking even deeper into community competencies, one finds individual and household capacities that exist in each neighborhood. It may be challenging to identify
13: PUBLIC HEALTH DATA TO SUPPORT HEALTHY COMMUNITIES
these assets. However, these are the assets that change communities (Asset-Based Community Development Institute, n.d.-a). An example is the Neighborhood Health Status Improvement project by Deborah Puntenney (ABCD Institute, n,d.-b). The
approach is a place-based strategy and is designed by the residents of the community with a grassroots orientation. It includes mapping local health assets, mobilizing residents and associations, and leveraging the resources within the community to implement the plan. Figure 13.4 reflects a community assessment map completed in Texas (Edwards et al, 2012). This is an example of community assets by mapping the location of hospitals and charitable care clinics, as well as other community resources that impact health. These types of maps provide a powerful visual representation of data that can focus communities on the available resources demonstrating strengths, as well as areas that reflect the lack of available resources
within the community demonstrating
weaknesses.
Data-Analytic and Statistical Approaches A community health assessment uniquely blends different data types that relate to the health status of individuals, communities, and populations. The community assessment methodology is an epidemiologically based process that is used for identifying populations with a predisposition to poor health. The goal of a community assessment is to locate communities that have common characteristics.
It must be statistically valid,
nonjudgmental, and specific to geographic locations and health criteria. The assessment process described hereblends four different data-analytic strategies to identify community needs.
Geo-mapping methods that reflect disease prevalence and comorbidities in the population are also a powerful illustration of where at-risk populations may reside. Figure 13.5 reflects the analysis of trends related to COVID-19 and its geospatial distribution. The geospatial distribution of COVID-19 cases by ZIP Code in Dallas County, Texas, indicates where clusters of cases are occurring and interventions should be targeted (McBride, 2005, 2006 and Dallas County HHS, 2018).
The first data-analysis method begins with the compilation of secondary data. These data are demographic variables, birth statistics, leading causes of death, access to primary care, social factors such as food insecurities, policies and programs, physical environment, and health behaviors. Statistical models can be based on ranking methods related to the adherence to policies and programs, health factors and health outcomes, a hierarchical cluster analysis that groups similar communities, or a method that compares communities to state and national benchmarks. All these statistical and analytic methods can be used to help identify community needs in an objective, nonjudgmental datadriven process.
The second analytic method is based on the inpatient utilization patterns of each identified community. The utilization patterns are viewed by different product lines
(e.g., cardiology, obstetrics, oncology), diagnosis-related groups (DRGs), or primary diagnosis. Emergency department (ED) utilization for nonemergency care can also be part of this second study of utilization patterns. This type of analysis generates common disease factors based on utilization patterns that can inform assessment and planning for community intervention. For exanrple, if one's top diagnoses for a given community are cardiovascular disease and diabetes, one's plans should be directed at those high-risk populations. If one has high rates of mental health-associated admissions, consideration should address the community needs related to mental health. The third analytic method is primary data collection. These types of data can come from more than one source. Examples of primary data are survey data, preferably
315
3i6
II: POINT-OF-CARE TECHNOLOGY
FIGURE 13.4 A Map of Community Assets.
Source; Edwards, J„ Pickens, S.. Schultz, L., Erickson, N„ & Dykstra, D, (2012). HorizonstThe Dallas
County community health needs assessment. https://www.researchga te.net/profile/Jennifer _Edwards23/publication/320962033_Horizons_Dallas_County„Communit y_Health_Needs_Assessment/ links/5a04c6e4458515eddb80bb4b/Horizons-Dallas-County-Community- Heaith-Needs-Assessment.pdf.
13; PUBLIC HEALTH DATA TO SUPPORT HEALTHY COMMUNITIES
FIGURE 13.5 Geo-Mapping of Coronavirus Active Cases by ZIP Code Dallas County, Texas, 2020.
Source: Dallas County Health and Human Services and Parkland Health and Hospital System. https://www .dallascounty.org/Assets/uploads/docs/covid-19/hhs-zipcodes/DC-C OVID-19-ZipCode-History-Ongoing.pdf.
gathered by telephone; focus group information; key informant interviews; and/or a community priority-setting process.
The fourth type of analytic method involves capitalizing on new data sources from HIEs or Regional Data Initiatives. These initiatives involve collecting electronic clinical and administrative data to improve collaboration in the region on patient safety, quality, and population health initiatives. HIEs that deploy data warehouse or centralized data repository models generate data and analytics in the region for that purpose and are excellent sources of information for community health assessment and planning (Glaser, 2006).
The Importance of Data The basis for a successful community assessment is contingent on local data and criteria
to aid in the solution of regional problems that affect the health status of a community. The data should be of high quality and sourced from publicly available and /or privately held data (in the case of the HIE or regional collaborative), or the data might be purchased from a privately held third party. Data needed for valid assessment involve population statistics, economic variables, birth and birth-related data, mortality and morbidity data,
the Agency for Healthcare Research and Quality Prevention Quality Indicators (AHRQ-PQIs). access to care, and other health indicators such as
317
318
II: POINT-OF-CARE TECHNOLOGY
Population Variables Population density is a core element needed to reflect the community and may be noted as the total number of individuals living in a specific area per square mile. Many conditions produce impacts on community healthcare related to density. An example of this type of health impact is the contagion of communicable diseases occurring at higher population densities. With lower population densities, the availability of medical care tends to decrease. Other factors that sliould be noted in the community assessment include age,
race, and ethnicity. Age groups are important because children are more susceptible to communicable disease and injury, the elderly to chronic and degenerative disease, and young adults to injuries. An example of the importance of race and ethnicity can be seen by the higher prevalence rates of diabetes in the Hispanic population. Concerning racial
groups, American Indians and Alaska Natives are more than twice as likely to develop diabetes as are White Americans (Johns Hopkins Medicine, 2014). It is also evident in the
higher cases of COVID-19 in communities ba.sed on research conducted by the Kaiser Family Foundation (Artiga et al., 2020) Economics—Income
It is also important to consider economic data on the community. The effect of income can determine housing conditions, nutritional status, social standing, social ties, education, access to health services, and other social and health problems or social determinants of health. According to research from the California Endowment, "Your ZIP Code shouldn't predict how long you live" (California Endowment, 2015; Davis et al., 2010).
The authors recommend a YouTube video developed by the RWJF (www.youtube. com/watch?v=yd8DTJ6ntrA) with a powerful message related to ZIP Codes and life expectancy. Review the short video and consider the following questions:
1. Why should ZIP Codes dictate life expectancy? 2. How can methods described in this chapter address some of these community-based issues?
3. How might analytic methods such as geo-mapping, as described previously, reflect trends that might dictate community-based improvements? Birth and Birth-Related Information
Birth rates and neonatal mortality are specific indicators that reflect growth in the community, as well as potential health risks. For example, neonatal mortality is correlated with low birth weight with a direct impact on an infant's ability to survive and develop. Maternal factors, such as a mother's age and educational levels (available in birth certificate data), have been shown to affect the health status of both mother and baby (Office of Adolescent Health, 2014).
Mortality and Morbidity —Death-Rate Variables Age-adjusted death rates for leading causes of death are standard metrics in a health assessment, and these can be compared over time and among geographic areas. Mortality and morbidity (disease prevalence) are important factors to be considered when assessing a community's health status. What is the burden of disease? How could mortality rates for various comorbid conditions be compared with state and national rates? It is
13: PUBLIC HEALTH DATA TO SUPPORT HEALTHY COMMUNITIES
important to examine these questions in data analysis and reports within a CHNA. The AHRQ Inpatient Quality Indicators for mortality and utilization can be used to examine morbidity and mortality for a region. A list of these indicators is noted in Table 13.3. These indicators are sensitive not only to morbidity within the community but also to the quality of healthcare services provided in the community. They are frequently used by states to report quality measures to the public {AHRQ, 2014a). Figure 13.6 presents an example of this type of quality analysis using the congestive heart failure (CHF) riskadjusted mortality rate showing a trend over time. This indicator reflects a downward trend in mortality rates for this community (McBride, 2007). Public health data can be used to indicate the prevalence of communicable or vector-borne diseases. Figure 13.7 reflects the prevalence of human cases of the West Nile virus in 2015 for Dallas, Texas, and the surrounding area, a virus spread by mosquitos. This analysis provides valuable information to public health professionals and planners to determine the need for public health interventions. Access to Primary Care In conjunction with preventive services, access to primary care has been shown to reduce the early onset of disease and death (Bauer et al., 2014; Haughton & Stang, 2012; Nicholas & Hall, 2011; Starfield et al, 2005; Stevens et al., 2014). The AHRQ has developed an
algorithm for determining preventable hospitalizations (Artiga et al., 2020). Applying these data to local geography can help in identifying communities where access to care may be influencing the utilization patterns measured by the indicators. The Prevention Quality Indicators (PQIs) are a set of measures that can be used with hospital inpatient discharge data to identify the quality of care for "ambulatory care sensitive conditions." These are conditions for which good outpatient care can potentially prevent the need for hospitalization or for which early intervention can prevent complications or more severe disease. The PQIs are population-based and are risk-adjusted for covariates such as age, gender, and risk. Table 13.4 reflects the AHRQ-PQI measures (AHRQ, 2014b). The New York University (NYU) Center for Health and Public Service Research has developed an algorithm to help classify ED utilization. The algorithm was developed with the advice of a panel of ED and primary care physicians, and it is based on an examination of a sample of almost 6,000 full ED records. Data abstracted from these records included the initial complaint, presenting symptoms, vital signs, medical history, age, gender, diagnoses, procedures performed, and resources used in the ED. Based on this information, each case is classified into one of the following categories:
■ Nonemergent: The patient's initial complaint, presenting symptoms, vital signs, medical history, and age indicated that immediate medical care was not required within 12 hours.
■ Emergentlprimarif treatable care: Based on the information in the record, treatment was required within 12 hours, but care could have been provided effectively and safely in a primary care setting. The complaint did not require continuous observation, and no procedures were performed, or resources used that are not available in a primary care setting (e.g., CT scan or certain lab tests). ■ EmergentlED care needed—preventablelavoidable: ED care was required based on the complaint or procedures performed /resources used, but the emergent nature of the condition was potentially preventable/avoidable if timely and effective ambulatory care had been received during the episode of illness (e.g., the flare-ups of asthma, diabetes, congestive heart failure).
319
320
II: POINT-OF-CARE TECHNOLOGY
TABLE 13.3 AHRQ Quality Indicators by Condition of Procedure Name* MORTALITY RATES FOR CONDITIONS
Acute myocardial infarction Acute myocardial infarction without transfer Congestive heart failure Gastrointestinal hemorrhage Hip fracture Pneumonia Acute stroke
MORTALITY RATES FOR PROCEDURES
Abdominal aortic aneurysm repair Coronary artery bypass graft Craniotomy Esophageal resection Hip replacement Pancreatic resection
Percutaneous transluminal coronary angioplasty Carotid endarterectomy HOSPITAL-LEVEL PROCEDURE UTILIZATION RATES
Cesarean section delivery
Primary cesarean delivery Uncomplicated vaginal birth after cesarean delivery (VBAC)
Total rate for vaginal birth after cesarean delivery {VBAC} Incidental appendectomy in the elderly Bilateral cardiac catheterization
Laparoscopic cholecystectomy AREA-LEVEL UTILIZATION RATES (E.G., COUNTY, STATE) Coronary artery bypass graft Hysterectomy Laminectomy or spinal fusion
Percutaneous transluminal coronary angioplasty VOLUME OF PROCEDURES
Abdominal aortic aneurysm repair
Carotid endarterectomy Coronary artery bypass graft Esophageal resection Pancreatic resection
Percutaneous transluminal coronary angioplasty ●These are measures reflected in rates that have detailed definitions of numerators, denominators, and
inclusion and exclusion criteria developed by the Agency for Healthcare Research and Quality (AHRQ) noted populations.
Source: Agency for Healthcare Research and Quality. (2014a|. Inpatient quality indicators overview. http://www.qualityindicators.ahrq.gov/modules/iqi_resources.aspx .
on
13; PUBLIC HEALTH DATA TO SUPPORT HEALTHY COMMUNITIES
FIGURE 13.6 Congestive Heart Failure Mortality Rates—An example of Agency for Healthcare Research and Quality (AHRQ) quality indicators used in community assessments reflecting a trend line with reduced overall rates of mortality.
FIGURE 13.7 A Profile of the Dallas,Texas, Area of the United States With West Nile Virus in 2015.
WNF, West Nile Fever; WNND, West Nile neuroinvasive disease. Source: Dallas County Health and Human Services. [2016]. 2015 profile of human West Nile virus in Dallas
County. https://www.da1lascounty.org/department/hhs/documents/WN V_2015_2015Profile_Dallas.pdf.
321
322
II: POINT-OF-CARE TECHNOLOGY
TABLE 13.4 Agency for Healthcare Research and Quality (AHRQ) Prevention Quality Indicators (PQIs) by Condition CONDITION NAME
Bacterial pneumonia Dehydration Urinary tract infections
Perforated appendix Low birth weight Angina without procedure Congestive heart failure Hypertension
Chronic obstructive pulmonary disease Uncontrolled diabetes
Diabetes, short-term complications Diabetes, long-term complications Lower-extremity amputations among patients with diabetes Source; Agency for Healthcare Research and Quality. {2014b), Prevention quality indicators overview, http:// www.qualityindicators.ahrq.gov/modules/pqi_resources.aspx.
■ Emergent: ED care needed—not preventable/avoidable: ED care was required and ambulatory care treatment could not have prevented the condition (e.g., trauma, appendicitis, myocardial infarction; Ghandhi, & Sabik, 2014). ■ Special Populations: Many urban and rural areas have special at-risk populations.
Many of these groups have been marginalized because of their environment or their lack of representation in health data collected and acted upon within their community. Some of these special populations include homeless, incarcerated, behavioral health, lesbian, gay, bisexual, or transgender (LGBTQ) communities (Parkland, 2019, p. 108).
Pulling It All Together The final summary of data analysis that informs the community assessment is a complete view of the community from the data and information available. Pulling it all together in a complete picture involves using multiple data sources related to both primary data
(collected by the community for a specific purpose) and secondary data (collected for a
different purpose but used secondarily to inform assessment). It is helpful to have a point
of reference in a checklist to ensure all sources of information are covered.
Triangulation of Data
The types of public domain data and indicators highlighted earlier can be used in
combination with primary data collected in the community to triangulate the information using a method that effectively informs a community health intervention program. An example of this type of triangulation might be a community needs index, preventable hospitalizations, and avoidable ED visits. Figure 13.8 reflects such an analysis done by Parkland Health and Hospital System (Parkland Hospital System, 2008).
13: PUBLIC HEALTH DATA TO SUPPORT HEALTHY COMMUNITIES
FIGURE 13.8 Triangulation of Data Indicating Areas of Need.
CNI, community need index; ED, emergency department; PQI, Prevention Quality Indicator. Source: McBride. S. (2005). Combining theAHRQ indicator sets to assess the health of communities: Powerful information for planning purposes (1st ed., pp. 20-28). Dallas-Fort Worth Hospital Council Data Initiative. http://www.qualityindicators.ahrq.gov/Downloads/Reso urces/Presentations/2005/2005AHRQQl _McBride_Combining_Sets_to_Assess_Hea!th_oLCommunities.ppt.
One of the strengths of a needs assessment is that the data can provide objectivity when statistically valid and can also provide a complete analysis of the community. Such methods can include using national benchmarks such as Healthy People 2030 goals, American health rankings, or state health rankings. It is important to include comparisons to similar communities on relevant health status indicators, comparing trends over time, or a hierarchical cluster analysis that clusters similar communities together (DHHS, n.d.).
Mapping is a valuable process used to help find like communities or combine contiguous counties (side by side), creating an expanded community approach. This
approach can be used to compare and contrast counties and cities within the larger community. Figure 13.9 reflects a community comparison to other service areas in the
region and past performance on health risk behaviors. This visual also depicts the power of analytic tools for community assessment.
Priority Setting Based on the Data Analysis It is not possible to effectively improve the health of the community by imposing solutions from outside. Stakeholders must be involved and committed to the strategies. Public health professionals and other content experts have unique knowledge that assists communities in assessment, planning, intervention, and evaluation of community health initiatives, but they do not always have firsthand knowledge of the community. There is greater success in improving community health when the community establishes the priorities and claims ownership of its strengths and weaknesses (Brown et al., 2012; Foster-
323
324
II: POINT-OF-CARE TECHNOLOGY
FIGURE 13.9 Community Comparative Analysis and Analytic Score Card, Confidence Interval.
Source: Parkland Hospital System. (2015). Community health assessment: Cedar Hill service area. Author, https://www.pa rklandhospital.com/Uploads/Public/Documents/PDFs/ Health-Dashboard/Cedar-Hill-SA-2015.pdf.
Fishman et al., 2001; Roussos & Fawcett, 2000; Zakocs & Edwards, 2006). The community priority setting process focuses on areas of the most concern to the residents of targeted communities. Although data analysis and analytic reporting can inform these decisions, the community must decide what it believes are the top priorities to be addressed. For the
Dallas County CHNA, data were triangulated from the regional primary and secondary data analyses. Also, focus group data, primary informant surveys, and the consensus of the planning committees from each community helped create the final plan. Figure 13.10 reflects a dashboard of the Dallas County assessment. Using these types of data to inform the community, all stakeholders involved in the process organized and approved the priorities.
Special Considerations for Rural and Small Communities When assessing rural and small communities, there is often a lack of reliable and valid
data at the ZIP Code level. Available public domain data often mask data with small cell sizes in rural and small communities to protect the confidentiality of individuals within the community. When examining data within these communities, assessment and planning frequently rely on data collected from primary sources within the community or compare data at higher aggregate levels, such as the entire county, to examine patterns and trends. However, when data are used in regional and country aggregation, it is
I 3: PUBLIC HEALTH DATA TO SUPPORT HEALTHY COMMUNITIES
FIGURE 13.10 Dashboard for Dallas County Assessment onTriangulated Primary and
Secondary Data Within the Community Assessment. Note: AW data are from the years 2009 to 2012; years available vary by topic. VLBW, very low birthweight;YPLL, years of potential life lost before age 65. South Dallas Southeast Dallas Desoto Lancaster
Southwest Dallas Cedar Hill
Stemmons Corridor
Wilmer Hutchins Seagoville Grand Prairie Northeast Dallas
Irving
Each vertical line
represents 100%
North Dallas Northwest Dallas
difference from the
Dallas County average
Outer Northwest
4
More unfavorable
More favorable
Percentage difference from the Dallas County average ■ ■ ■ 1 j-
Mortality Socioeconomic Access to clinical care Preventable hospitalizations
■ High-risk sexual behavior ■ Very low birth weight (VLBW) Mobidity ■ Years of potential life lost before age 65 (YPLL) ■ Violence and injury Infant mortality
important to use additional data collected directly from the community to determine whether identified patterns and trends are relevant to the community. INTERVENTION AND EVALUATION OFTHE COMMUNITY HEALTH IMPROVEMENT PLAN
The goals and objectives of a community health improvement plan must be pragmatically determined ba.sed on current reality and resources—financial human capital and structure—and they must also be realistic. Outcomes are built into the community health improvement plan for evaluation purposes. Outcome metrics address environmental policy and systems changes necessary to sustain population health programs and are considered manageable priorities based on community resources. Thus, the inclusion of outcome metrics provides a measure of overall program effectiveness during the evaluation phase (CDC, 1999). A stakeholder's alignment in the development of goals and objectives for the community health improvement plan should be based on the data analysis and tools used to assess the community. Objectives include action steps or evidence-based population strategies aligned with current best practices in public health to address environmental and behavioral conditions contributing to disease in the community. Table 13.5 provides a tool that can be used for this purpose in the form of a community needs assessment.
¥
325
326
li: POINT-OF-CARE TECHNOLOGY
TABLE 13.5 Needs Assessment Checklist forVariables
to Be Considered in a Community
Needs Assessment
HEALTH RISK VARIABLES
INPATIENT DISCHARGES PER 1,000 POPULATION
POPULATION VARIABLES
■ ■ ■
Discharges per 1,000 population for each service area or the county as a whole (excluding
Population Total population density Population by age groups (0-4 years of age, 5-17 years of age, 18-64 years of age, 65 years and older)
newborns)
Discharges per 1,000 population for each service area or the county as a whole for the top five dischargers
ETHNICITY
■
Percentage ofWhites
■
Percentage of African Americans
■ ■
Percentage of Hispanics Percentage of Asians, etc.
Potentially avoidable hospitalizations
ED use by type of visits—nonemergent, emergent/treatable primary care, emergent/ED care needed/ preventable avoidable, emergent/ ED care needed
SOCIOECONOMIC DATA
Percentage of people below federal poverty guidelines Total number of households
Estimated per capita income Estimated average household income Percentage of households with incomes
PUBLIC HEALTH DATA
■
Rates of communicable diseases
■
Rates of sexually transmitted diseases
■
Rates of vector-borne diseases
SURVEY DATA:
4. Mobilize partnerships 5. Develop policies/plans
21% ■t
88% ●I
50% I
'Jiy
■
●j
H
49%
●Si
6. Enforce laws
70%
t—i
7. Link to health services
29% I
1-
8. Ensure workforce 1-
9. Evaluate services
:
H
39% t-
10. ResearcWinnovations
70%
57%
Overall
H
52%
0%
10%
20%
30%
40%
50%
60%
70%
T
T
80%
90%
Priority Rating
Essential Service
100%
Performance Score
(Level of Activity)
Quadrant I (high priority/low performance) - these Important activities may need increased attention. g
49 (Moderate)
4. Mobilize community partnerships to identify and solve health problems
10
21 (Minimal)
5. Develop policies and plans that support individual and community
7
49 (Moderate)
1. Monitor health status to identify community health problems
health efforts
7. Link people to needed personal health services and ensure the 9 29 (Moderate) provision of healthcare when otherwise unavailable 9. Evaluate effectiveness, accessibility, and quality of personal and 10 39 (Moderate) popuation-based health services Quadrant II (high priority/high performance) ● these activities are being done well, and it is important to maintain efforts. 7 70 (Significant) 6. Enforce laws and regulations that protect health and ensure safety Quadrant III (low priority/high performance) - these activities are being done well, but the system can shift or reduce some resources or attention to focus on higher-priority activities. 5 88 (Qptimat) 2. Diagnose and Investigate health problems and health hazards 8. Ensure a competent public
personal healthcare workforce
10. Research for new insights and innovative solutions to health problems
4
70 (Significant)
4
57 (Significant)
Quadrant IV (low priority/low performance) ● these activities could be improved, but are of low priority. They
may need little or no attention at this time. 3. Inform, educate, and empower people ^ut health issues
2
50 (Significant)
Summary of the EPHS Performance Scores and Overall Scores: Score
EPHS 1,
Monitor health status to Identify community health problems
49
2.
Diagnose and investigate health problems and health hazards
88
3.
Inform, educate, and empower people about health issues
50
4.
Mobilize community partnerships to identify and solve health problems
21
5.
Develop policies and plans that support individual and community health efforts
49
6.
Enforce laws and regulations that protect health and ensure safety
70
7,
Link people to needed personal health services and ensure the provision of healthcare when othenwise unavailable
8.
Ensure a competent public and personal healthcare workforce
9. 10,
Evaluate effectiveness, accessibility, and quality of personal and population-based health services
Research for new insights and innovative solutions to health problems
Overall performance score
29
70 39 57 52
329
330
II: POINT-OF-CARE TECHNOLOGY
Group Evaluation;
Community Health Assessment and Group Evaluation COMMUNITY-AT-LARGE
additional information about the community can be included in the comment box COMMUNITY'S NAME:
Southern County Module score sutmnarfes
Policy (%)
Environment (%)
Module
6970
52-31
Physical activity
56.45
61.19
Nutrition
52,73
54,55
Tobacco use
64.44
75,56
Chronic disease management
52.73
43,64
Leadership
Needs and Assets: ASSETS
NEEDS Low
0-20%
High 21-40%
61-80%
41-60%
81-100%
Community at targe (CAL) Physical activity
CALE
CALP
Nutrition
CALP
CALE
Tobacco
CALP. CALE
Chronic disease mgt.
CALP. CALE
Leadership
CALP CALE Assessment Data on Health Outcomes and Behavioral Risk Factors;
Health Outcome
County
Diabetes
11.0%
9%
Prevalence for those >20 yrs
Cardiovascular disease
7.8%
6.6%
Percentage of heart disease >18 yrs 2005-2008
Adult obesity
30%
29%
Percentage of adults reporting
Preventable hospital stays
90
68
State
Measure
5MI i3Q
Hospitalization rate per ambulatory care-sensitive condition per 1,000 Medicare enroilees
Environmental Conditions
Limited access to healthy
14.0%
9%
foods
Percentage of low-income individuals who do not live close
to a grocery store Access to parks
14%
33%
Access to exercise
68%
74%
opportunities
Percentage of those living within 0.5 miles of a park
Percentage of population with adequate access to locations for
physical activibes Behavioral Conditions Tobacco use
25%
17%
Percentage of adults reporting smoking 00 cigarettes and currently smoking
Physical inactivity
24%
24%
Percentage of adults >20 yrs reporting no leisure time or
physical activity
13s PUBLIC HEALTH DATA TO SUPPORT HEALTHY COMMUNITIES
a community benchmark used to measure the progress of community health improvement plan goals and objectives over time. This information was used to identify gaps in and needs of the community. For example, we obtained and determined that areas receiving scores of 60% or less would be gaps and considered for inclusion in the community health improvement plan. In contrast, areas receiving scores of 61% to 100% would be classified as assets and would, therefore, not be a
priority to be addressed in the community health improvement plan. Additional county data were gathered to support findings from the CHANGE tool and to assist stakeholders with the development of the community health improvement plan. One of the principal challenges to collecting these data was finding data at the county level because of data protection constraints in public domain data. Many health indicators, such as social determinants of health and rates of chronic disease, were either difficult to locate or simply not available. One issue relates to the sample size of the state Behavioral Risk Factor Surveillance Survey.The absence of county benchmark data makes it difficult to accurately reflect the success of the community health improvement plan.To supplement these data insufficiencies at the county level, data from the County Health Rankings and the state as well as regional data for cardiovascular disease and social determinants of health were used. The report card for the community assessment in the case study is reflected in Figure 13.11.The goals established for the community based on the assessment are noted In Figure 13.12.
FIGURE 13.12 Goals and Objectives for the Community. Decrease the number of residents without access to healthy food from
Goall.'
14% to g% by December 31,2025.
Objective
; Expand & develop existing farmers’ market program to include locations
1.1
! accessible to all community members by December 31, 2025.
Objective
; Establish and promote a county-wide community garden program, ' increasing number of sites accessible to outlying areas from 0 to 4
1.2
by December 31.2025.
Objective
Initiate the Expanded Food and Nutrition Education Program by
1,3
December 31, 2025.
Reduce tobacco use in the county from 25% to 17.5% by December
Goal 2.1
31,2025.
Objective
Develop a county-wide tobacco-cessation program by May 31,2023.
2.1
Objective
Develop a county Smoke-Free Order, with community stakeholders, for
2,2
adoption by December 31,2025.
Objective 2.3
I
Implement a tobacco use awareness campaign with community
! partners to reach alt sectors of the county by December 31,2024.
331
332
II: POINT-OF-CARE TECHNOLOGY
After a review of the case study, one should reflect on the following questions: What tools were deployed in this case study?
1.
How effective was the assessment and goal planning? Defend your position
2.
using elements of the chapter to reinforce your position. 3.
Why do you think the public health professional selected the tools and the model for assessment used?
4.
What data challenges were evident in this case?
5.
How might an HIE in the region have helped this community?
CASE STUDY 2
Rapid Community Response to COVID-19 Through Advanced Uses of Data—A PCCI Case Study Introduction
Because of our core experience in the practical applications of advanced data science and social determinants of health (SDOH), PCCI became involved with Dallas County Health and Human Services (DCHHS), Parkland Health & Hospital System (Parkland) and other civic, hospital, and public health leaders at the outset of the COVID-19 pandemic. Immediate Challenges
COVID-19 presented several immediate challenges to Dallas County and Parkland and its care providers.Table 13.6 lists those in detail. TABLE 13.6 COVID-RelatedChallengesto Public Health THE CHALLENGE
DETAILS OF THE CHALLENGE
The Public Health
DCHHS needed to launch public health initiatives for approximately 2.5 million county residents to reduce the spread of COVID-19. Given the highly contagious nature of the virus and finite resources, public health officials needed access to real-time, hyper-localized data to track infection rates, monitor effectiveness of local efforts, identify emerging "hot spots" and high-risk areas for proactive education and testing, and forecast community needs.
Challenge:
The Care Delivery Capacity Challenge:
Parkland is a focal point for the Dallas COVID-19 response. Current evidence shows that individuals with underlying medical conditions and/or those who are immunosuppressed are more susceptible to presenting with more serious forms of the disease. Given the large and diverse patient population that Parkland serves as the area's largest safety-net hospital, it sees a disproportionate number of complex patients, which can quickly exhaust critical care bed capacity and intensive care unit resources.
The Frontline Care
Management Challenge:
Initial criteria used at the pandemic's outset, such as travel history, were not specific enough since the infection rates (and disease progression) became highly localized. Frontline staff needed additional real-time information about patient exposure to ensure effective testing, triage, prioritization, and follow-up.
1 3: PUBLIC HEALTH DATA TO SUPPORT HEALTHY COMMUNITIES
Our Beliefs
The Information needed to address the public health, care delivery capacity, and frontline care management challenges is vital for coordinating efforts between local healthcare providers, health, and human services. Local government is needed to develop and implement rapid, targeted, and impactful responses to the ongoing waves of the pandemic.Table 13.7 illustrates important requirements for managing a pandemic. PCCI's Data-Driven Approach to Pandemic Management
To effectively equip public health, civic, and hospital leaders with the best information to address the challenges and preserve health across our communities, we identified the following needs (see Figure 13.13): 1. A PHI/Health Insurance Portability and Accountability Act (HIPAA) secure data hosting environment that is flexible, configurable, and scalable and can accommodate data derived from disparate structured and unstructured sources. This is essential to execute our solution.
2. Tools to predict at the community/individuallevelto hyper-localize interventions and inform public health guidelines. 3. Tools to predict at the care delivery level to help predict disease surge, project capacity, and guide development of alternative clinical care practices. 4. A communication portal with multiple dynamic dashboards makes data outputs easy to interpret and act upon by clinical, public health, and civic leaders.
TABLE 13.7 Needs for Proactive Pandemic Management WHAT IS NEEDED FOR PROACTIVE PANDEMIC MANAGEMENT?
Region-wide data repository and timely, dynamic dashboards for
positive COVID-19 cases.
Robust risk
Localized
measurement
community data showing greater
Mobility pattern modeling within
models capture and analyze real-time data for a person's proximity to infected individuals
heightened risk
about movement
as a risk indicator forCOVID-19
factors and
and its impact on disease spread.
exposure.
and across
concentrations
communities to
of individuals
create correlations
with COVID-19
and predictions
mapped access points for medical and non-medical
support. Routine access to
Customized
Nimble methods
Better information
reliable, localized
forecasting models, with scenario mapping capabilities, to
for altering clinical delivery workflows
on the changing
to ensure proper
model demand for
hospital beds, ICU,
appropriate level of safety for patients
ventilators, and
and staff.
hospitalized patients to impact individual patient care and overall capacity
projections of pandemic onset, peaks, and plateaus.
PPE.
care and the
condition of
management
333
334
II: POINT-OF-CARE TECHNOLOGY
FIGURE 13.13 Data Driven Approach to the Pandemic.
Source: Courtesy Steve Miff, PhD, PCCInnovation.org
Response/Impact
1. The Public Health Challenge -The access to information (e.g., real-time, hyper-localized data) is vital for public officials to coordinate efforts between local healthcare providers, health and human services, and local government to develop and implement rapid, targeted, and impactful responses to the pandemic.
■ Leverage Our Connected Communities of Care Network to Mitigate Mortality and MorbidityTied to COVID-19
13: PUBLIC HEALTH DATA TO SUPPORT HEALTHY COMMUNITIES
The COVID-19 outbreak has uncovered the extreme exposure to harm faced by those most at-risk in our communities. Absent a pandemic, the community's underserved residents face extreme difficulties addressing their social and health needs. With the pandemic, these troubles have multiplied. Like Dallas, communities that have created Connected Communities of Care (CCC) linking together healthcare providers and community-based organizations (CBOs), via a network of configured electronic information exchanges, can provide cross-sector care coordination while also helping to quickly stabilize community areas devastated by major crises, such as COVID-19. PCCIhas been able to use the Da//asCCCto quickly assemble data to help identify hotspot neighborhoods locations where the virus is having a disproportional impact on residents, many of whom are poor and underserved, and then turn that information into targeted communications and tactical containment efforts through community-wide awareness and education messaging. Communications delivered to residents through familiarfood pantries, homeless shelters, and places of worship are often much more effective than community-wide public information campaigns. By connecting local CBOs and faith-based organizations with public health workers and clinicians, we have been able to facilitate effective contact
tracing, provide critical community access points for services, and implement care plans for high-risk individuals in a more efficient and scalable manner. ■ Create a Vulnerability Index to Identify Sub-Populations With Heightened Vulnerabilities for Proactive Interventions (see Figure 13.14) Factors such as healthcare disparities, socioeconomic status, and race/ ethnicity have increasingly become key factors for developing ongoing COVID-19 management strategies. As a result, PCCI created a Vulnerability Index (VI) tool by geo-mapping data from disparate sources to identify sub-populations who are at high risk for complications and mortality from COVID-19 and/or who harbor other SDOH factors that place them at greater risk of infection. PCCI's VI considers a variety of risk factors—social, clinical, and demographic—all to identify communities especially at-risk for exponential spread. The VI not only captures each risk factor's individual effect but also synergies across risk factors.
Our Vi's significant finding shows that social deprivation is a leading factor in determining the risk for COVID-19 infection and the primary reason for racial/ ethnic disparities in COVID-19 risk, more so than age, race, or comorbidity rates. This information Is helping public health officials and other city and county leaders to coordinate efforts for more targeted testing, outreach, assistance to mitigate disease surge, and save lives until vaccines are available. Through increased testing and disease identification sites, rent or mortgage moratoriums, food distribution, utility forgiveness, and other basic needs assistance, community leaders can decouple a community's COVID-19 risk from economic distress (see Figure 13.15). 2. The Care Delivery Capacity Challenge-Hos'p\ta\administrative leaders needed real-time data to prioritize, prepare, and forecast utilization to proactively plan for this uncertain and unprecedented potential rise in healthcare delivery need, both during the initial disease wave and subsequent surges.
335
336
II: POINT-OF-CARE TECHNOLOGY
FIGURE 13.14 Geo-Map ofVulnerability.
Source: Courtesy Steve Mif,f PhD, PCCInnovation.org.
FIGURE 13:15 Vulnerability Assessment Framework Details.
Vulnerability Assessment Framework
m
n
PCCi Vulnerability Index: Biology, £nv. Demographics
Health comorbidity
Social Determinants
●
Multi-dimensional
●
Data-Driven
■
Equitable
*
Adjustable w/ changing circumstances
Lockdown adherence
Components stable in short interval
(Skhange in mobility)
1. 2-
Comorbidily Age
3.
Social Determinants
Emerging Risk COVID-19 (CLI symptomatic or positive test cases)
Proprietary & Confidential O PCCI
Components changing in real time 4.
Mobility
S.
Incidence of COVIO-L9-like-illness(CLI)
V
PCCI
Source: Courtesy Steve Miff, PhD, PCCInnovation.org.
■ Develop a Capacity Forecasting ModelThat Can Evolve With Disease Surges
In collaboration with Parkland, we developed a forecasting model that modified the CHIME (COVID-19 Hospital Impact Model for Epidemics) and NorthShore models currently available to a new model that we believed was a more realistic predictor of the COVID-19 trajectory based on Parkland and Dallas' unique local variables/factors, such as social distancing and shelter-in-place timing. We believed that parametric SIR (Susceptible, Infected and Recovered
13: PUBLIC
HEALTH DATA TO SUPPORT HEALTHY COMMUNITIES
(or Removed) modeling was unlikely to capture the full picture of the outbreak emerging in Asia and Europe, given that for COVID-19, available information had unmeasurable differences in local adherence, there were unaligned definitions for non-pharmaceutical interventions (NPI).There were differences in data collection, each of which impacts SIR parametric assumptions.
Our approach maximized the available information to provide clarity for public health leadership by using deployed frameworks for geocoded information and forecasts for tracking the course of the disease usable by hospitals and communities. Our approach used empirical data from other geographical areas and communities (such as Hubei, China, and King County, Seattle) and customized their epidemic curve to Dallas County's and Parkland's unique characteristics. This was a successful approach and we were the only organization in Dallas to predict an initial late April 2020 peak and plateau of COVID-19 numbers (see Figure 13.16). 3. The Frontline Management Challenge Frontline clinical staff needed real-time information about patient proximity to infected individuals to use in addition to clinical manifestation to assess the risk of COViD-19 exposure for effective testing, triage, prioritization, and follow-up. ■ Create an Index to Manage Proximity Risk through Early Identification of High-Risk Individuais To effectively manage future risk, first responders, clinical teams, and public health leaders needed- current information about patient proximity to infected individuals to inform appropriate intervention strategies. PCCI developed a novel Proximity Index (see Figure 13.17) to be used for earlier identification of higher risk individuals by integrating the following inputs:
FIGURE 13.16 COVID-19 Projections. Performance of Nonparametric Model Scaling Dallas's Epidemic Curve of Deaths against Seattle, King County, WA. Mean Square Integrated Area 87
Source: Courtesy Steve Miff, PhD, PCCInnovatiori.org.
337
338
Ih POINT-OF-CARE TECHNOLOGY
FIGURE 13:17 Proximity Index Factors.
PCCI Source: Courtesy Steve Miff, PhD, PCCInnovation.org.
■ Geocoded, confirmed County
COVID-19 cases at the block level across
Dallas
■ Density details of the population living in that proximity ■ Mobility information of Individuals in that block (shelter-in-place and block level mobility data as available)
We incorporated Proximity index scores in existing Parkland workflows to (a) better manage scheduled outpatient (OP) visits and (b) inform care management for unscheduled ED visits. We integrated this information onto Parkland COVID-19 dashboards and made it available to the Parkland
outpatient task force.
PCCI's Proximity Index has been able to identify patients with a potential risk of contracting COVID-19, during the social distancing phase, with six times the likelihood over the general population. With limited test-kit availability, this model's insights have helped spur proactive testing of high-risk Individuals before they enter clinical buildings and have helped to prevent exposure, especially in asymptomatic COVID patients. In the first 30 days, the model proactively screened 63,230 patients scheduled for a non-COVID-related clinic visit at a health system. Of these, 1,986 were identified as high risk for COVID exposure and were phone triaged and offered testing at convenient locations 24 to 48 hours before their visit date. Until results were available, their health needs were met by telemedicine and virtual care. Only 5.7% of COVID-19 positive patients were missed by this predictive model. On chart review, most of these cases had missing or incorrect data, such as addresses.
13: PUBLIC HEALTH DATA TO SUPPORT HEALTHY COMMUNITIES
Conclusion
PCCI's data-driven approach has helped Dallas leaders make pandemicmanagement activities more transparent and precise through an enhanced approach to disease prevention and mitigation.This approach has enabled a more coordinated, community-wide response, standardization of clinical practice across multiple locations and different frontline care teams, identification of hotspots and taiiored/targeted approaches to those most vulnerable and at-risk, improved public awareness, and enhanced capacity for increased testing through mobile units and collaboration with local stores (for increased testing centers).The results and lessons learned in Dallas have also laid the groundwork for expansion of the work into other communities.
SUMMARY
In today's environment of cost containment, accountable care organizations (ACOs), and healthcare reform, the financial viability of healthcare providers depends on creating and maintaining healthy communities and the ability to rapidly respond to immediate health threats. The community assessment methodology described within this chapter
is provided to outline a systematic approach to use data and information available within the community to pinpoint community service locations and public health outreach activities, as well as to measure the health status outcomes of the residents of
those communities. They also provide the data infrastructure to help address vulnerable special needs communities as well as external threats to the health of our communities. Community assessments are valuable methods used by healthcare institutions nowadays for strategic planning to support healthy communities and outreach programs; however, to sufficiently inform the process, methods such as those described within the chapter are required. This chapter has provided models, tools, metrics, and data-analysis approaches to inform community health assessment and planning. The chapter has also provided a case study to demonstrate how to effectively apply these methods to a community for health assessment purposes. Finally, the chapter discusses how new and improved HIT infrastructure can be used for public health hazards to respond promptly and effectively to minimize risk for populations.
END-OF-CHAPTER
RESOURCES
EXERCISES AND QUESTIONS FOR CONSIDERATION
You are approached by your local community to provide leadership in a community assessment and planning process. Consider the content covered concerning community assessment and planning, models and tools presented, and primary and secondary data sources, and reflect on the following questions; 1. What is the first step in this process in aligning stakeholders for planning? Why is it important to consider this as the first step in the process?
339
340
II: POINT-OF-CARE TECHNOLOGY
2. What model(s) will you use to structure the process, and why are models such as these important to the process?
3. What data will you need to collect as primary data? What data are available in the public domain that you may be able to use to inform the assessment and planning? 4. Why are data important to the community health assessment? 5. If you are assessing a rural community, what might be the constraints concerning data availability that you need to factor into the analysis? How will you address these constraints?
6. What special populations might exist in your community that requires separate analysis, and what are the data constraints in doing so? ADDITIONAL RESOURCES
SPtllKfl mUSKIk'
CONNECT
A robust set of instructor resources designed to supplement this text is located at http://connectspringerpub.com/content/book/978-0-8261 -8526-6. Qualifying instructors may request access by emailing textbook@springerpub. com.
REFERENCES
Adeleye, O. A., & Ofili, A. N. (2010). Strengthening intersectoral collaboration for primary health care in developing countries: Can the health sector play a broader role? Journal of Efwironmental and Public HeaWi 2010,1-6. https://doi.org/10.1155/2010/272896 Agency for Healthcare Research and Quality. (2014a). Inpatient quality indicators overvieio. http: / / WWW
.qualityindicators.ahrq.gov/modules/iqi_resources.aspx Agency for Healthcare Research and Quality. (2014b). Prevention quality indicators overvieio. http;// www.qualityindicators.ahrq.gov/modules/pqi_resources.aspx
Artiga, S., Orgera, K., Pham, O., & Corallo, B. (2020). Growing data underscore that communities of color are being harder hit by COVID-19. Kaiser Family Foundation, 2020(Apr), 1-1. https: / / WWW
.kff.org/policy-watch/growing-data-underscore-communities-color- harder-hit-covid-19/ Asset-Based Community Development Institute. (n.d,-a). About. Retrieved October 4, 2021, from https://community-wealth.org/content/asset-based-community-devel opment-institute-abcd
-northwestern-university Asset-Based Community Development Institute, (n.d.-b). Research: Deborah Puntenney: Neighborhood health status improvement. Retrieved October 4, 2021, from https://resources.depaul.edu/abcd -institute/initiatives/Pages/research.aspx Association of State and Territorial Health Officials. (2014). ASTHO profile of state public health (Vol. 3). Author, http://www.astho.org/Profile/Volume-Three Association of State and Territorial Health Officials. (2015). Cojnmunity health needs assessments, http;/ /
www.astho.org/Programs/Access/Community-Health-Needs-Assessments
Bauer, U. E., Briss, .P A., Goodman, R. A., & Bowman, B. A. (2014). Prevention of chronic disease in the
21st century: Elimination of the leading preventable causes of premature death and disability in the USA. The Lancet, 354(9937), 45-52. https:/ /doi.org/10.1016/S014 0-6736(14)60648-6 Beck, A.Boulton, M. L., Lemmings,& Clayton, J. L. (2012). Challenges to recruitment and retention of the State Health Department epidemiology workforce. American Journal of Preventive Medicine, 42(1), 76-80. https://doi.org/10,1016/j.amepre.2011.08.021 Brown, L. D., Feinberg M. E., & Greenberg M. T. (2012). Measuring coalition functioning: Refining constructs through factor analysis. Health Education & Behavior, 39(4), 486-497. https://doi .org /10.1177/1090198111419655 California Endowment. (2015). Zipcode or genetic code: Which is a better predictor of health? http:// www.calendow.org/zipcode-or-genetic-code-which-is-a-better-predi ctor-of-health Centers for Disease Control and Prevention. (1999). Prameioorkfor program ei'aluation in public health: Morbidity and Mortality Weekly Report, 45(RR-11), 1-40. http:/ /www.cdc.gov/mmwr/PDF/RR/RR4811.pdf Centers for Disease Control and Prevention. (2010a). National piiblic health performance standards, http: / / www.cdc.gov/nphpsp/theinstruments.html
13: PUBLIC HEALTH DATA TO SUPPORT HEALTHY COMMUNITIES
Centers for Disease Control and Prevention. (2010b). The public health system and the ten essential
public health
services:
Overvieiv. https://www.cdc.gov/stltpubli chealth/publichealthservices/
essentialhealthservices.html
Centers for Disease Control and Prevention. (2013). Community health assessment and group ez’aluation
(CHANGE); Building a foundation of knowledge to prioritize community needs, http://www.cdc.gov/
nccdphp / dch / programs / healthycommunitiesprogram / tools / change.htm
Centers for Disease Control and Prevention. (2014, December). STLT public health professionals gateway:
National voluntary accreditation for public health departments, http://www.cdc.gov/stltpublichealth/ accreditation
Centers for Medicare & Medicaid Services. (2014). Waivers, http: / / www.medicaid.gov/medicaid-chip
-program-information / by-topics / waivers / waivers.html Dallas County Health and Human Services. (2018). West Nile watches 2018. https: / /www.dallascounty .org/department/hhs/documents/WestNileWatchMapping6-14-2018.pdf Davis, R. A., Cohen, L., & Rodriguez, S. (2010). Toward health equity: A prevention framework for reducing health and safety disparities. In B. J. Healey & R. S. Zimmerman (Eds.), The neto world of health promotion: New program development, implementation, and evaluation (pp. 153-194). Jones & Bartlett. Edwards, ]., Pickens, S., Schultz, L., Erickson, N,, & Dykstra, D. (2012). Horizons: The Dallas County
community health needsassessment. Dallas County Health and Human Services and Parkland Health and Hospital System, http://www.dalIascounty.org/department/hhs/docu ments/FINALCHNA_000 .pdf
European Centre for Health Policy. (1999). Health impact assessment: Main concepts and suggested approach. World Health Organization Regional Office for Europe. http://www.healthedpartners,org/ceu/ hia/hiaOl /01_02_gothenburg_paper_on_hia_l999.pdf Foster-Fishman, P. G., Berkowitz, S. L., Lounsbury, D. W., Jacobson, S., & Allen, N. A. (2001). Building collaborative capacity in community coalitions: A review and integrative framework. American journal of Community Psychology, 29(2), 241-261. https: / /doi.org/10.1023/A:1010378613583 Ghandhi, S., & Sabik, L. (2014). Emergency department visit classification using the NYU algorithm. The Atnerican Journal of Managed Care, 20(4), 315-320. Glaser, J. (2006). Health information exchange (HIE) business models: The path to sustainable financial success. Deloitte Center for Health Solutions, http://www.providersedge.c om/ehdocs/ehr_articles/ Health_Info_Exchange_Business_Models.pdf Haughton, B., & Stang, J. (2012). Population risk factors and trends in health care and public policy. Journal of the Academy of Nutrition and Dietetics, I12(Suppl. 3), S35-S46. https://doi.org/10.1016/j .jand.2011.12.011 Health and Human Services. (2013). National Archives and Records Administration. Community
health needs assessments for charitable hospitals. Federal Register, 20523-20544. https://www
. federalregister.gov/docLiments/2013/04/05/2013-07959/community -health-needs-assessments
-for-charitable-hospitals Hilltop Institute, (n.d.). Community benefit state laxo profiles: A 50-state sun>ey of state community benefit laws through the lens of the ACA. Retrieved October 4, 2021, from http://www.hilltopinstitute.org/ hcbp_cbl.cfm Hoelscher, D., & McBride, S. (2020). Usability and the rapid deployable infectious disease decision
support system. Computers, Inprmatics, Nursing, 38(10), 490-499. https://doi.org/10.1097/CIN .0000000000000654
Internal Revenue Service. (2017). Neiv requirements for 501(c)(3) hospitals under the Ajfbrdable Care Act. https://www.irs.gov/charities-non-profits/charitable-organizations/new-requirements-for-501c3 -hospitals-under-the-affordable-care-act Johns Hopldns Medicine. (2014). Health Library: Diabetes, http://www.hopkinsmedicine.org/
healthlibrary/conditions/diabetes/diabetes_statistics/diabetes/d iabetes_home_85,P00343
Kama, J., & Kramer, M. (2011). Stanford social innovation reinew: Collectwe impact (Review No. LSU-2011).
LeLand Stanford Jr. University. http://c.ymcdn.com/sites/www.lan o.org/resource/dynamic/ blogs/20131007_093137_25993.pdf
Kindig, D., & Stoddart, G. (2003). What is population health? American Journal of Public Health, 93(3),
380-383. https://doi.org/10.2105/ajph.93.3.380 Kretzmann, J. P, & McKnight, J. L. (1993). Building communities from the inside out: A path toxoard finding and mobilizing a community's assets. ACTA Publications. McBride, S. (2005). Geo-map trending of the MRSA infections in hospital discharge data. In Combining the AHRQ indicator sets to assess the health of communities: Poxverful information for planning purposes (1st ed., pp. 20-28). Dallas-Fort Worth Hospital Council Data Initiative, http:/ /www.qualityindicators
.ahrq.gov / Downloads /Resources/Presentations/ 2005 / 2005AHRQQI_McBride_Combining_Sets
_to_Assess_Health_of_Communities.ppt
341
342
II: POINT-OF-CARE TECHNOLOGY
McBride, S. (2006). Epidemiologic evaluation of rising MRSA in Texas. Paper presented at the Poster Session of the CDC Assessment Conference, Atlanta, GA.
McBride, S, (2007, November 13). AHRQ corijerence: A regional commitment to improve the healthcare
workforce, patient safety, quality & the health of populations sensed, presented. Agency for Healthcare Research and Quality. National Association of County and City Health Officials. (2009). Integrating performance improvement processes: MAPP, National Public Health Performance Standards, and Accreditation, http; / / www.naccho
.org/topics/infrastructure/mapp/framework/mapppubs.cfm Nicholas, J. A., & Hall, W. J. (2011). Screening and preventive services for older adults. Mount Sinai Journal of Medicine, 78{4), 498-508. https: / /doi.org/10.1002/m sj.20275 Office of Adolescent Health. (2014). Trends in teen pregnancy and childbearing, http; / / www.hhs.gov/ash/ oah/adolescent-health-topics/reproductive-health/teen-pregnancy/ trends.html Parkland Health & Hospital System/Dallas County Health and Human Services (2019) Dallas county community health assessment 2019. https://www.parklandhos pital.com/Uploads/public/ documents/PDFs/Health-Dashboard/CHNA%202019.pdf Parkland Hospital System. (2008). Community health checkup 2008: Trends, highlights and new features. Parkland Hospital System. Parkland Hospital System. (2015). Community health assessment: Cedar Hill sendee area. Author, https: / / www.parklandhospital.com/Uploads/Public/Documents/PDFs/Health-Da shboard/Cedar-Hill -SA-2015.pdf Pennel, C., McLeroy, K., Burdine, J., & Matarrrita-Cascante. (2015). Nonprofit hospitals' approach to community needs assessment. American Journal of Public Health, 205(3), el03-ell3. https://doi .org /10.2105 / AJPH.2014.302286 Public Health Accreditation Board. (2013). Public Health Accreditation Board: Standards & Measures Version
1.0. http: / / www.phaboard.org/accreditation-overview/getting-s tarted
Public Health Community Platform. (2015). Condition referral from EHR to public health use case, http: / / www.thephcp.org/about-PHCP/background Public Health Informatics Institute, (n.d.). Business process matrix template, https://www.digitalbridge .us/
Robert Wood Johnson Foundation. .counlyhealthrankings.org
(2014, January).
County health
rankings.
http://www
Roussos, S. T., & Fawcett, S. B. (2000). A review of collaborative partnerships as a strategy for improving
community health. Annual Review of Public Health, 21(1), 369-402. https://doi.org/10.1146/annurev
.publhealtfi.21.1.369
Somerville, M. H., Nelson, G. D., & Mueller, C. H. (2013). Hospital community benefits after the ACA: The state law landscape (Issue Brief No. 6). The Hilltop Institute at UMBC. Retrieved from http://www.hilltopinstitute.org/publications/HospitalCommunityBe nefitsAfterTheACA
-StateLawLandscapeIssueBrief6-March2013.pdf Starfield, B., So, L., & Macinko, J. (2005). Contribution of primary care to health systems and health. MilBank Quarterly, 93(3), 1-76. http://www.commonwealthfund.org/ usr_doc/Starfield_Milbank .pdf Stevens, C. D., Schriger, D. L., Raffetto, B., Davis, A. C., Zingmond, D., & Roby, D. H. (2014). Geographic clustering of diabetic lower-extremity amputations in low-income regions of California. Health
Affairs, 33(8), 1383-1390. https:/ /doi.org/10.1377/hlthaff.2014 .0148
University of Michigan. (2013). Enumeration and characterization of the public health nursing workforce: Findings of the 2012 public health nursing sunmjs (No. RWJ-062013). University of Michigan Center for Excellence in Public Health Workforce Studies.
https://sph.umich.edu/cephw/pdf/Nurse%20
Workforce-RWJ%20Report.pdf University of Wisconsin Population Health Institute. (2014). County Health Rankings model, http:/ / WWW
.countyhealthrankings.org/county-health-rankings-model U.S. Census Bureau. (2013). State and county quickfaefs. https://www.census.gov/quickfacts/fact/ table/US/PST045217
U.S. Department of Health & Human Services, (n.d.). Healthy People 2020. http://healthypeopIe.gov Zakocs, R. C., & Edwards, E. M. (2006). What explains community coalition effectiveness? A review of the literature. American Journal of Preventive Medicine, 30(4), 351-361. https://doi.0rg/lO.lOl6/j .amepre.2005.12.004
Privacy and SeCm^jtyln aU.bi quite us Health InformationTechnology WoiSd SUSAN MCBRIDE. HELEN CATON-PETERS, KRISTINTESMER. AND JONATHAN ISHEE
OBJECTIVES ●
Discuss the need, history, and principles of the Health Insurance Portability and Accountability Act (HIPAA), including transactions, privacy, and security components.
●
Discuss the increased requirements of HIPAA outlined in the Health Information Technology for Economic and Clinical Health (HITECH) Act and the purpose of increased protections.
●
Discuss the importance of clinicians fully understanding and being the trusted agents who protect patients' healthcare information.
●
Describe common issues seen in the clinical setting that constitute privacy and security violations and how to mitigate these issues.
●
Describe cybersecurity threats and the need to enhance security in the health informatics (HI) environment as a result of these threats,
●
Evaluate a case study magnifying the importance of protected health information (PHI)
protections and violations to the trust of consumers. ●
Develop a strategy and plan to address privacy and security with an exercise outlined for the reader,
CONTENTS INTRODUCTION
344
Regulatory Environment FederalTrade Commission
345 345
Food and Drug Administration
346
Substance Abuse and Mental Health Services Administration
State Regulatory Requirements International Law
346
347
Health Insurance Portability and Accountability Act ElectronicTransactions and Code Sets Requirements Privacy Rule
346
347 349
350
Guidance Regarding Methods for De-Identification of Protected Health Information Security
352
351
344
I: POiNT-OF-CARE TECHNOLOGY
Enforcement
353
Business Associate Agreements
354
HITECH ACT INCREASED PROTECTIONS 2013 Modifications
355
356
Proposed 2020 Modifications
357
ROLE OFTHE CLINICIAN INTHE PROTECTION OF PROTECTED HEALTH INFORMATION Clinicians' Responsibilities
358
358
Promoting Interoperability Program
358
The Basics of a Security Risk Assessment
359
Risk Assessments—An Important Role for the Nursing Informaticist and Nursing Leadership 359 THE NURSE'S ROLE IS IMPORTANT IN ESTABLISHING PUBLICTRUST POPULATION HEALTH AND RESEARCH DATA SUMMARY CASE STUDY
360
361 361
EXERCISES AND QUESTIONS FOR CONSIDERATION REFERENCES
359
362
363
INTRODUCTION
In recent years, the significance of safeguarding the privacy and security of health information lias become a primary and critical issue for healthcare providers and informatics professionals. Though not a new issue, the relevance to practice is more pressing than ever as new ways in which health data are collected, used, and shared promise innovation and increase the potential for misuse and breach. The adoption of electronic health records (EHRs) and other forms of health technology has grown at such a rapid rate that existing privacy policy protections are not sufficient and gaps in safeguards are evident as data flow across and through this ever-changing healthcare ecosystem. Studies have shown that patients generally have a large amount of trust in providers, but that trust can very quickly erode if a breach of any magnitude occurs with their health information (Hall et al., 2002). Consequences stemming from system disruption and data theft and loss can lead to significant patient harm and result in organizational, reputational, and financial damage. Patients and providers must be able to trust the technology they use to make the most gains in healthcare. It is up to all nursing professionals to establish and maintain this trust so that the benefits of a fully interoperable and learning health system can be realized (Eden et al., 2008). Bedside nurses and APRNs have an obligation as essential clinical providers and professionals to maintain the privacy and security of the EHR. Understanding the responsibilities of PHI security and the organization's compliance plan is a critical role for all nurses (Dolan & Farmer, 2016). Nursing informatics specialists have an increasing obligation to safeguard health data and information systems on behalf of the clinical team and educate end-users on how to maintain privacy and security according to current regulations. The threat to healthcare data increases significantly with the digital age and new cyber threats to health information (Nahm et al., 2019). This chapter covers the critical information needed by professional nurses related to the Health Insurance Portability and Accountability Act (HIPAA) and other regulations to safeguard protected health information (PHI).
14: PRIVACY AND SECURITY IN A UBIQUITOUS HIT WORLD
Regulatory Environment To begin to understand how to safeguard health information appropriately, one must understand the regulatory environment as it applies to the type of data and the intended use. Perhaps the most recognizable health privacy regulation is the Health Insurance Portability and Accountability Act (HIPAA). Enacted in 1996, this regulation was primarily enacted to streamline and digitize claims submission through the adoption of electronic transaction standards. Included within the Administrative Simplification section of HIPAA was a requirement that certain healthcare providers, clearinghouses, and plans keep health information private and secure. HIPAA is considered a seminal event in health privacy regulation since it was the first time that the federal government set wide-ranging privacy and security requirements for patient medical information. HIPAA sets a floor for privacy and security practices for defined "Covered Entities" (CEs) with states being able to enact more stringent requirements.^ In addition, providers and organizations participating in government health information technology (HIT)-related payment programs are required by the Centers for Medicare & Medicaid Services (CMS) to attest that they meet certain privacy and security measures in place under the prior Meaningful Use Program that is now the Promoting Interoperabilit y Program. However, a healthcare professional must consider numerous other policy-related considerations when ensuring that patient rights to confidentiality are respected and upheld, including understanding the entire regulatory picture and maintaining awareness of the capabilities and weaknesses of the technological environment in which they practice. Federal protections for the privacy and security of health data beyond HIPAA apply depending on circumstances, and APRNs need to be aware of the implications for practice. Federal Trade Commission
The traditional approaches to privacy and security may require a broader understanding of how health data may be regulated. The Federal Trade Commission (FTC) enforces privacy and security practices and describes its mission as follows: "The Federal Trade Commission (FTC or Commission) is
an independent U.S. law enforcement
ageJKi/ charged zoith protecting consumers and enhancing competition across broad sectors of the economy. The FTC's primary legal authority comes from Section 5 of the Federal Trade Commission Act, which prohibits unfair or deceptive practices in the marketplace. The FTC also has authority to enforce a variety of sector specific lazvs, including the Truth in Lending Act, the CAN-SPAM Act, the Children's Online Privacy Protection Act, the Equal Credit Opportunity Act, the Fair Credit Reporting Act, the Fair Debt Collection Practices Act, and the Telemarketing and Consumer Fraud and Abuse Prevention Act. This broad authority allows the Commission to address a zoide array of practices affecting consumers, including those that emerge zuith the development ofnezv technologies and business jnodels." (FTC, 2014. p. 1) The recent case of LabMD v. FTC demonstrates how the FTC uses its enforcement
authority to hold companies accountable for the protection of personal medical information. In this case, the FTC alleges that LabMD failed to reasonably protect the security of consumers' personal data, including medical information. The complaint alleges that LabMD collectively exposed the personal information of approximately 10,000 consumers in two separate incidents. The complaint alleges that LabMD billing information for more than 9,000 consumers was found on a peer-to-peer file-sharing ' Covered entities (CE) are defined as "health care providers who conduct covered health care transactions electronically, health plans, and health care clearinghouses" (Federal Register/Vol. 78, No. 17/Friday, January 25, 2013/Rules and Regulations, p. 5567).
345
346
I: POINT-OF-CARE TECHNOLOGY
network. Then, in 2012, LabMD documents containing sensitive personal information of at least 300 consumers were found in the hands of identity thieves. This case is ongoing, but it highlights an important lesson for healthcare professionals to heed. Health data deserve special attention and protection regardless of location or form, and we are challenged to continually address these concerns in light of all possible regulatory requirements. Food and Drug Administration
Another federal regulatory agency with a role in the privacy and security of healthcare data is the U.S. Food and Drug Administration (FDA). The FDA oversees the safety of medical devices, which includes addressing the management of cyberseairity risks and hospital network security. Recent guidelines (FDA, 2013) recommend that medical device manufacturers and healthcare facilities take steps to ensure that appropriate safeguards are in place to reduce the risk of failure cau.sed by a cyber attack. The introduction of malware can initiate cyber attacks into the medical equipment or unauthorized access to configuration settings in medical devices and hospital networks. The consequences of not adequately addressing these risks could be dire. As medical devices are increasingly integrated within healthcare environments, there will be a need for vigilance toward cybersecurity practices to ensure all systems are adequately protected and patients remain safe from harm. Nursing informatics (NI) are frequently called on to evaluate the safety and effectiveness of new devices and software. Considerations of cybersecurity must be included in any evaluation process. Substance Abuse and Mental Health Services Administration
Certain types of health data considered especially sensitive have special protection under the law. In the case of substance abuse treatment data, heightened confidentiality protections are afforded by 42 CFR Part 2 enforced by the Substance Abuse and Mental Health Services Administration (SAMHSA, 2014). These regulations exist to protect
patients receiving substance abuse treatment in federally funded facilities against possible discrimination. Patients must give their consent to share data collected from these facilities with other healthcare providers. Those protections persist in that a patient must give express consent for any future disclosures. It should be noted that the increased
privacy protections provided in 42 CFR Part 2 only apply to healthcare providers (a) receiving federal substance abuse treatment funding and (b) who hold themselves out as providing such treatment or referral.^
State Regulatory Requirements A discussion of privacy regulations would not be complete without mentioning that healthcare professionals also need to be aware of specific state and international laws that impact data use. State privacy laws differ widely, sometimes pre-empting federal privacy regulations by establishing greater protections (Health Information Law [HIL],
n.d.). This situation adds a layer of complexity that must be navigated carefully to ensure adherence to federal requirements while at the same time respecting the timely and
secure sharing of patient data across state lines. In many cases, special protection exists for data such as HIV and sexually transmitted disease diagnose.s, mental health records, information related to minors, and the prohibition of transmitting or storing PHI outside the United States. Informatics professionals must be familiar with what state law requires - For 42 CFR Part 2 to apply. A provider must be federal assisted and hold itself out as providing and provides alcohol or drug abuse diagnosis treatment or referral for treatment. See 42 CFR § 2.11.
14: PRIVACY AND SECURITY IN A UBIQUITOUS HIT WORLD
and understand potential consequences of sharing data in new models of care such as accountable care organizations (ACOs) and health information exchanges (HIEs). International Law
International law will play an increasing role in privacy protections for patients as data are shared across U.S. borders (NitroSeairity & FairWarning/ n.d.). Data can be stored offshore and fall under international law, which may affect rights and use. Health technology
developers are based across the globe and are challenged with incorporating privacy and security practices that must meet a complex web of regulatory requirements. Frequently, those developers are innovative nurses and providers who are addressing patient care and building mechanisms for clinical decision support (CDS) by designing creative and simple-to-use applications and tools. Building privacy and security best practices, such as strong authentication procedures and encryption, into these tools as a part of the software development cycle will prevent many potential vulnerabilities. Using the regulatory framework in place will guide developers toward key practices and robust risk assessment procedures that will circumvent a vast array of possible negative consequences. As of May 25, 2018, the General Data Protection Regulation has become effective. With this new regulation, the European Commission intends to strengthen and unify data protection for individuals within the European Union (EU). The regulation also addresses the export of personal data outside the EU, which has implications for U.S. businesses and patients overseas (http: / /ec.europa.eu/justice/ data-protection/reform/ index_en.htm).
Health Insurance Portability and Accountability Act The fundamentals of HIPAA, including its history, requirements, additions to the regulatory requirements under the Act, implications for clinicians, and how EHRs, Promoting Interoperability Program measures, and emerging innovative technologies are likely to push the constraints of the regulations, are described in this section. HIPAA was passed into law in 1996 and was subsequently amended and expanded to address the increasing privacy of Protected Health Information (PHI). We review the HIPAA regulations passed in 1996 and discuss new privacy and security regulations under the Health Information Technology for Economic and Clinical Health (HITECH) Act, including an emphasis on enforcement penalties. With this emphasis on enforcement, the HITECH Act is often described as having "put teeth in HIPAA" (Clearwater Compliance, 2012). A number of resources are available that support organizations in adhering to HIPAA regulations, including guidance developed by the Office for Civil Rights (OCR) and the Office of the National Coordinator for Health Information Technology (ONC). We provide information on accessing and utilizing these expansive resources to help understand and adhere to HIPAA regulations. Finally, we discuss the significance of added security related to cyber threats and the importance of all healthcare professionals in maintaining a heightened awareness related to this relatively new threat. HIPAA Background
HIPAA is also known as the Kennedy-Kassebaum Act after former senators Nancy Kassebaum (R-Kansas) and Edward M. Kennedy (D-Massachusetts). Before the passage of the Affordable Care Act (ACA), HIPAA was considered the most significant federal healthcare reform since the enactment of Medicare and Medicaid in 1965 (Atchinson &
Fox, 1997). This expansive Act covered health insurance reform in five titles. Titles I and
347
348
II: POINT-OF-CARE TECHNOLOGY
II of the Act focus on health insurance reform around the portability of health insurance between jobs and limitations to pre-existing conditions and, through administrative simplification, to contain fraud, waste, and abuse within healthcare. For the purpose of this chapter, title II is the emphasis of discussion because it deals primarily with the
protection of clinical data. Title II of HIPAA required the Department of Health and Human Services (DHHS) to establish national standards for electronic healthcare transactions and national
identifiers for providers, health plans, and employers (CMS, 2013). With the increased use of electronic claims and billing transactions data came the understanding of the need to protect the privacy and security of the health data captured within those transactions and, more broadly, by the entities responsible for collecting and using those data. Therefore, the DHHS was also required to develop rules for privacy and security that would apply when electronic transactions and code sets were used. Finally, it was important to establish enforcement procedures that the DHHS would follow to investigate reports of noncompliance and fines that organizations would be subject to. These rules are outlined in 45 CFR Parts 160, 162, and 164. The transaction and code set, employer identifier, and National Provider Identifier (NPl) mles are administered and enforced by the CMS, whereas the privacy and security mles are administered and enforced by the OCR (https://www.hhs.gov/ocr/index.html). The DHHS OCR is responsible for enforcing the Privacy Rule and Security Rule. HIPAA CEs were required to comply with the Security Rule beginning on April 20, 2005. On July 27, 2009, the OCR became responsible for enforcing the Security Rule (OCR, n.d.-e). The OCR enforces the Privacy Rule and Security Rule by investigating complaints, conducting compliance
reviews, and issuing guidance on how to comply with the regulations and mles. The OCR refers to possible criminal violations of HIPAA to the Department of Justice (DOJ; OCR, n.d.-a).
HIPAA regulations apply to organizations defined as CEs, which include health
plans, healthcare clearinghouses, and certain healthcare providers. To be considered a CE and therefore subject to HIPAA regulations, a healthcare provider, clearinghouse, or health plan must be using an electronic standard adopted by the Secretary of the DHHS. Key dates for HIPAA enactment are noted in Table 14.1. The initial stages of HIPAA involved a complicated regulatory requirement with extensive comment periods in the TABLE 14.1 Key HIPAA Dates and Deadlines DATE
DEADLINE FOR NOTED ACTION
August 21, 1996
HIPAA Public Law 104-191 signed
November 3, 1999
HIPAA Privacy Rule proposed
December 28, 2000
HIPAA Final Privacy Rule initially posted but underwent revisions
October 16, 2002
Electronic healthcare transactions and code sets—all CEs except those that filed for an extension and small health plans
April 14, 2003
Privacy Rule in effect for all CEs except small health plans
April 16, 2003
Electronic healthcare transactions and code sets—all CEs must have
started software and system testing April 14, 2004
Privacy Rule in effect for small health plans
April 20, 2005
Security Rule comes into effect
March 16, 2006
Enforcement compliance in effect
CEs, covered entities; HIPAA, Health insurance Portability and Accountability Act, Source: Centers for Medicare & Medicaid Services. (2013). HIPAA: General information, https://www.cms
.gov/Regulations-and-Guidance/HIPAA-Administrative-Simplificatio n/HIPAAGenInfo/index.html.
14: PRIVACY AND SECURITY IN A UBIQUITOUS HIT WORLD
FIGURE 14.1 Health Insurance Portability and Accountability Act (HIPAA) Sections and Important Dates.
Health Insurance
Portability and Accountability Act
(HIPAA) 1996
Administrative
Insurance Reform
Simplification Accountability
E-Transactions, Code Sets & Identifiers
October 16,2003
Portability
Privacy Compliance April 14,2003 The Administrative Simplification sections of HIPAA are more applicable to healthcare providers, while the Insurance Reform sections are more relevant to payers.
Security Compliance April 20,2005
Enforcement Compliance March 16,2006
V
/
rule-making process, modifications to the final rules, delays, and final effective dates that were often accompanied by considerable compliance efforts by CEs within the industry.
Figure 14.1 provides an overview of HIPAA, which notes that the Administrative Simplification sections of the Act are more applicable to healthcare providers, whereas
to payers. We focus more here on covering the sections relevant to healthcare providers and minimally on the insurance the Insurance Reform sections are the most relevant
reform components.
ElectronicTransactions and Code Sets Requirements According to Healthcare Information and Management Systems Society (HIMSS, 2015), transactions "are electronic exchanges involving the transfer of information between two parties for specific purposes." The transactions component of HIPAA requires that the DHHS adopt national standards for electronic healthcare transactions, which constitute a large part of the administrative simplification component. The transaction formats and standards are specified in the regulation, and they are aimed at creating administrative simplification so that payers, providers, and claims adjudication third parties are aligned on standard formats for processing claims and payments as well as for the maintenance and transmission of electronic healthcare information and data. These standards include
the following specifications:
■ Eligibility for Health Plan Inquiry and Response {270 / 271) ■ Healthcare Claim (837)
349
350
il: POINT-OF-CARE TECHNOLOGY
■ Healthcare Claim Status Request and Notification (276/277) ■ Referral Certification and Authorization (278)
■ Healthcare Claims Payment and Remittance Advice (835; Kushniruk & Borycki, 2013, p. 108) HIPAA transaction standards require that any provider or payer transmitting specified by the most current regulations dictating the code sets within the format for the type of electronic transaction. For example, the standardized billing format for a hospital is the 837 institutional (837i)
information as noted earlier must do so in the format
format. The American National Standards Institute (ANSI) ASC X12 837 is the claim
or encounter format. Specified code sets fall within that standard, and the data within the form must conform to the standards, including rigorous edits that ensure the data
within the form meet compliance. CEs are required to use the format if they submit the transaction electronically, and the claims payer is required to receive the format. Claims that are not considered complete or that contain errors must be corrected before they can be processed, or receive denial, rejection, or remittance advice, all of which have their electronic ANSI transaction formats to automate the process. In the event the claim submitted by the hospital or provider does not meet the definition of a "clean claim" or lacks complete or correct information, the claim will be rejected and sent back to the hospital througli this process. The International Classification of Diseases, 10th revision, clinical modification (ICD10-CM) codes is an example of a code set requiring the hospital to use the 837i electronic format. These codes were updated to ICD-IO-CM on October 1, 2015, with hospitals and providers expected to use the new code set starting in that year (CMS, 2015). X12 formats are messaging standards developed to transmit data between two entities referred to as "trading partners" in the HIPAA legislation. These file formats are periodically updated; for example, the ASC X12 837 has revisions that include a 5010 version providing a mechanism for allowing the use of ICD-IO-CM and other improvements. In the example given, trading partners are the hospital, the clearinghouse transmitting the claim to the payer, and the payer entity. All CEs must be able to utilize up-to-date HIPAA standards
under the electronic transactions and code sets requirements. In addition to the institutional formats for hospitals, there is a professional format for provider billing (837 Professional, or 837P), as well as dental (837 Dental, or 837D) and retail, pharmaceutical transactions (National Council for Prescription Drug Programs; CMS, 2017). Figure 14.2 depicts the typical claims processing and electronic billing provider flow. The claims payer can be Medicare, Medicaid, or private payers; this process
follows federal billing transaction requirements regardless of the payer. Clearinghouse intermediary service providers often manage the electronic data transmission (EDT) .services for a managed care organization (MCO).
Privacy Rule
Although Congress passed HIPAA in 1996, the Privacy Rule was not promulgated until 2003. The Privacy Rule is intended to protect the rights of individuals concerning the confidentiality of PHI while simultaneously allowing legitimate use of these data by governing the disclosure of PHI. According to the DHHS, the purpose of the Privacy Rule was as follows:
"The HIPAA Privacy Rule establishes national standards to protect individuals' medical records and other personal health mformation and applies to health plans, health care clearinghouses, and those health care providers that conduct certain health care transactions
14: PRIVACY AND SECURITY IN A UBIQUITOUS HIT WORLD
FIGURE 14.2 Provider ElectronicTransaction Process.
837 Medical
837 Payment request
V >
■o
■> o
&
paper remit Electronic or check
payment
All EDI transactions should be acknowledged by the ASCX12 997 (unctional acknowledgment A clearinghouse is often utilized for the provider and MCO transaction processing.
EDI, electronic data interface; MCO, managed care organization. Source: Ohio Bureau of Workers' Compensation, (n.d.). EDI implementation guide. Provider billing flow. Retrieved October 4, 2021, from https://www,bwc.ohio.gov/p rovider/services/techlevel.asp,
electronically. The Rule requires appropriate safeguards to protect the privacy of personal health information and sets limits and conditions on the uses and disclosures that may be made of such information without patient authorization. The Rule also gives patients rights over their health information, including the right to examine and obtain a copy of their health records and request corrections.” (OCR, n.d.-d, para. 1)
An important distinction to be made between the Privacy Rule and the Security Rule is that the provisions of the Privacy Rule apply to all PHI regardless of form, whereas the Security Rule governs electronic PHI. The Privacy Rule has undergone modifications after the final ailes posted in 2003, the most recent of which are a part of omnibus legislation in the HITECH Act provisions discussed later in this chapter. However, the goals of the Privacy Rule remain true to initial intent by mandating federal protections for individually identifiable health information, establishing rights to access and control health information, and preserving essential uses of PHI, with examples such as research for improving quality of care. The critical elements of the Privacy Rule include comprehensive specifications about CEs and business associates (BAs), permitted uses and disclosures, research, individual rights, administrative requirements, and compliance and enforcement.
Guidance Regarding Methods for De-Identification of Protected Health Information
There are two different methods recommended for de-identifying PHI in compliance with HIPAA. PHI is defined as individually identifiable health information that is transmitted or maintained by a CE or its BEs in any form or medium (45 CFR 160.103).
351
352
II: POINT-OF-CARE TECHNOLOGY
The definition exempts a small number of categories of individually identifiable health information, such as that found in employment records held by a covered entity in its role as an employer.
The two methods recommended are (a) expert determination through applying statistical or scientific principles (algorithms) to de-identify the PHI and (b) safe harbor methods, including the removal of 18 types of identifiers with no ability for residual information to identify the person (45 CFR 164.514; OCR, n.d.-b). The 18 elements that must be removed are as follows: 1. Names 2.
All geographical subdivisions smaller than a state, including street address, city, county, precinct, ZIP Code, and their equivalent geocodes, except for the initial three digits of a ZIP Code, if according to the current publicly available data from the Bureau of the Census: (a) the geographic unit formed by combining all ZIP Codes with the same three initial digits contains more than 20,000 people; (b) the
initial three digits of a ZIP Code for all such geographic units containing 20,000 or fewer people are changed to 000 3.
All elements of dates (except year) for dates directly related to an individual, including birth date, admission date, discharge date, and death date; all ages older than 89 and all elements of dates (including year) indicative of such age, except that such ages and elements may be aggregated into a single category of age 90 or older
4. Phone numbers 5. Fax numbers
6. Electronic mail addresses 7.
Social Security numbers
8. Medical record numbers 9.
Health plan beneficiary numbers
10. Account numbers 11.
Certificate/license numbers
12.
Vehicle identifiers and serial numbers, including license plate numbers
13. Device identifiers and serial numbers
14. Web universal resource locators (URLs) 15.
Internet protocol (IP) address numbers
16.
Biometric identifiers, including finger and voice prints
17.
Full-face photographic images and any comparable images
18.
Any other unique identifying number, characteristic, or code (note this does not mean the unique code assigned by the investigator to code the data; OCR, n.d.-b)
When de-identi fying PHI most CEs use the safe harbor described above since reliance on an expert determination may expose the CE to liability if OCR determines that the methodology used by the expert was faulty resulting in the ability to re-identify the data. Security
The DHHS issued the Security Rule in 2003 with a compliance date of 2005. The Security Rule applies to electronic health information created, received, used, and maintained by CEs and contains scalable and flexible measures that CEs must address
14: PRIVACY AND SECURITY IN A UBIQUITOUS HIT WORLD
as a part of their electronic PHI security-related activities. Security safeguards fall into three areas of compliance: (a) administrative, (b) physical, and (c) technical. The administrative component requires policies and procedures to be in place within CEs to demonstrate how the entity complies with the Act. The physical requirements relate to controlling physical access or inappropriate access to protected information. Finally, the technical component requires that CEs protect PHI when the data are transmitted or exchanged. Enforcement
In 2006, the DHHS promulgated the initial set of regulations surrounding enforcement of HIPAA by setting civil penalties for violations, procedures for investigations, and hearings. The rules and regulations were in place until this point, but no clear enforcement requirements were spelled out. There are a number of recent examples of enforcement cases with issues related to unpatched and unsupported software resulting in a $150,000 fine, a medical records dumping case resulting in an $800,000 fine, and larger fines ranging from $1.7 to $3.3 million for settlements on potential violations (OCR, n.d.-c). It should also be noted that there are separate criminal provisions for HIPAA violation, which the DOJ enforces.^ While criminal prosecution has been historically low, there has been a recent uptick in DO] referrals and proseaition. As of July 31, 2021, more than 29,000 cases have been investigated and resolved by OCR that required changes in privacy practices and corrective actions by providing technical assistance to HIPAA-covered entities and their business associates (DHHS,
2021). Of these cases, OCR imposed Civil Money Penalties (CMPs) in over 100 cases resulting in a total dollar amount of $135,328,482 (Table 14.2). TABLE 14.2 Status of All Privacy Rule Complaints—June 2021 Complaints Remaining Open
3,880
1%
Complaints Resolved
266,362
99%
Total Complaints Received
270,242
Referrals to DOJ for Criminal Enforcement
1,167
Most investigations were opened after the offending provider notified the OCR of a breach. The OCR begins very few investigations upon independent or internally
generated evidence or complaint. When the OCR imposes CMPs, a CE is entitled to a hearing before an administrative law judge (OCR, n.d.-f). Several instructive themes to the industry have emerged from resolution agreements. In most instances, the following inadequacies in HIPAA compliance were found in the OCR's investigations:
1. Incomplete or nonexistent risk analyses of HIPAA privacy and security policies and procedures;
2. Failure to have adequate business associate agreements (BAAs) in place between organizations sharing patient data; 3. Delayed reporting of breach notification to the OCR; and/or 4. Weak auditing proce.sses designed to detect, prevent, and mitigate HIPAA violations. H2U.S. Code§1320d-6
353
354
II: POINT-OF-CARE TECHNOLOGY
Risk analyses, business associate agreements (BAAs), effective auditing practices, and timely reporting of violations to the OCR are all required by HIPAA and HITECH. Effective compliance programs must be in place to achieve these requirements. The amount of CMPs imposed for failure to follow these required regulatory practices considers both the egregiousness of the substantive deficiencies and the nature and size of the entity found in violation of statutory requirements. Examples of the up-to-date OCR resolution agreements can be found at www. hhs.gov/hipaa/for-professionals/ compliance-enforcement/agreemen ts/index .html. This site is regularly updated. Case examples are an excellent means to learn about operational issues that result in HIPAA violations and fines. In January 2018, Aetna signed a $1.15 million settlement with New York for breaches related to the disclosure of the HIV status of individuals.
This settlement involved mailing
of information to New York residents in 2017 that contained
HIV drug information visible
through the envelope windows. This mistake inadvertently revealed 2,460 New York
Aetna members' HIV status. This incident resulted in an additional class action lawsuit
filed against the company for a reported $17.2 million settlement (Gordon, 2018). The most recent cases demonstrate high risk related to several situations that bear mentioning for the practitioner. First, most cases involve the physical loss of mobile devices that contain PHI. These include laptops, mobile phones, USBs, and back-up tapes. Storage of any PHI on a mobile device is a high-risk operational decision. Allowing PHI to reside only on network servers or cloud-based storage is a better practice in all instances to reduce risk. If your work necessitates the storage of PHI in mobile computing environments or access to PHI through these devices, proper security precautions must be followed. This includes encryption, the ability to "kill" information in the device from a
central platform, double authentication access, and physical restrictions to areas in which the mobile devices are kept. More than half of the 2016 to 2017 resolution agreements involve organizations that did not follow this advice. The DHHS has issued new security guidance for the use of mobile devices that can be found at the following websites:
■ www.healthit.gov/ providers-professionals/your-mobile-device-a nd-health -information-privacy-and-security
■ www.hhs.gov/sites/default/files/ocr/privacy/hipaa/administrati ve/securitynile/ remoteuse.pdf?language = es
Second, OCR imposed CMPs on multiple types of organizations during the 2016 to 2017 cycle. These organizations include physician offices, hospitals, insurers, ancillary service providers, technology firms, universities, and others. Small nonprofit organizations and large for-profit organizations paid CMPs for violations. No organization large or small is
immune to the requirements of HIPAA and HITECH regulations. Informaticists must be ever cognizant of their responsibilities to protect PHI, know the policies and procedures of their employer or contracting partners, and follow them without exception. If an organization's policies or practices seem deficient, contact the organization's compliance officer and recommend changes or improvements to assist in the protection of PHI.
Business Associate Agreements
The HIPAA privacy and security rules apply to CEs that are health plans, healthcare clearinghouses, or healthcare providers; however, many healthcare providers require the services of organizations under contract for certain functions that may requi re the handling of PHI. The Privacy Rule allows for these contractual relationships with
a business associate (BA; OCR, 2003). Under HIPAA, a BA is considered an extension
14: PRIVACY AND SECURITY IN A UBIQUITOUS HIT WORLD
of a covered identity, meaning that the requirements to safeguard PHI are considered to extend to the BA. A BA conducts activities on behalf of the CE. Activities that are
considered BA activities are things such as claims processing or administration, data analysis, processing or administration, utilization review, quality assurance, billing, benefit management, practice management, and repricing. The use of a contract is required to ensure the relationship between the CE and BA is clearly defined. For HIPAA, this contract is also known as a BAA. This contract between the CE and the BA must comply with certain Privacy Rule and Security Rule requirements and the BA is directly liable for violations (OCR, 2003). Examples of some BAAsthat are becoming more prevalent are cloud or service providers for EHR or data repository hosting and developers outside
of organizations. Organizations, such as Health Information Exchanges, e-Prescribing Gateways, or those that manage the transmission of PHI or require access to PHI on a routine basis, are considered a BA under the definitions noted in 45 Code of Federal
Regulations (CFR) 160, subpart A in §160.103 (eCFR.gov, 2014). A discussion of the HIPAA mles would be incomplete without remarking on the identifier standards and their significance to the transactions conducted in healthcare. As mentioned earlier, the DHHS has developed an employer identifier number (EIN) and an NPI for use in healthcare transactions. Use of such identifiers provides a means for standardized identification of employers, providers, and others, both within and across HIT systems. This simplifies the process of electronic data sharing, leading to greater efficiencies and improved care delivery. An early plan for identifier standards under HIPAA was the development of a patient identifier that was also intended for use with electronic transactions. This rule has never been developed, and organizations are challenged with finding alternative solutions for dealing with the very important task of correct patient identification, not only within an electronic transaction used under HIPAA, but also across all electronic health technology and care delivery systems where patients seek care. The consequences of mismatched and misrouted health information represent a real threat to patient safety that camiot be ignored in the absence of federal regulation. A master patient index (MPI) is one strategy that care delivery organizations employ to connect patient identities across their systems and maintain the integrity of data they collect, use, and share. Advanced practice nurses should develop and adopt safe and secure practices for patient identity matching to improve the coordination of care while upholding patients' rights. Consumer privacy advocates have voiced serious concerns over the implications of an NPI. As uses of electronic health information expand, so do the potential abuses. All efforts aimed at solving this complex problem should seek to engender patient and provider trust and balance this with safety and security risks associated with increased HIE.
HITECH ACT INCREASED PROTECTIONS
With the passage of the HITECH Act of 2009 as a part of the American Recovery and Reinvestment Act (see Chapters 1 and 4), the movement from paper-based records to EHRs and the expansion of HIE anticipated under the HITECH Act constituted unprecedented amounts of PHI data being exchanged in electronic format, and, as such, new vulnerabilities of exposure were anticipated. As a result, along with the HITECH Act came additional provisions for HIPAA that went above and beyond the original
protections established in 1996. Also known as the HIPAA Omnibus rule, these changes significantly expand individual rights and provide increased protection and control over health information. The HITECH Act requires the DHHS to perform audits, increases penalties for noncompliance based on the level of negligence, and outlines breach notification requirements. These substantial modifications have been described as the
355
356
li: POINT-OF-CARE TECHNOLOGY
DHHS "putting teeth in HIPAA/' meaning that these provisions or their violations are likely to be much more painful economically and with penalty subject to jail time for organizations or individuals who violate these protections intentionally. 2013 Modifications
Modifications to the HIPAA Privacy, Security, Enforcement, and Breach Notification Rules Under the HITECH Act and the Genetic Information Nondiscrimination Act
(GINA) along with other modifications to the HIPAA regulations were filed in the Federal Register in a final ruling on January 25, 2013, with entities required to comply with the final rules by September 23, 2013 (DHHS, 2013). So what changed with this new regulation, and why did additional protections need to be put into place? In 2013, DHHS Secretary Kathleen Sebelius stated the following regarding these final rule changes: "Much has changed in health care since HIPAA was enacted over fifteen years ago. The new mle will help protect patient privacy and safeguard patients' health information in an ever-expanding digital age" (OCR, 2013). The final rules filed in 2013 Rirther reinforced the HITECH Act changes, including modifications to;
■ Privacy, Security, and Enforcement rules, to strengthen privacy and seairity protections for health information and to improve enforcement initially provided for by the HITECH Act in 2009
■ Breach Notification Rule, which replaces the interim final mle published initially with the HITECH Act in 2009
■ Privacy protections under the Privacy Rule for genetic information as required by the GINA of 2008
■ Rules that are intended to increase workability and flexibility by decreasing the burden and better harmonizing the requirements with those under other DHHS departmental regulations (DHHS, 2013)
One of the most important things to note with the latest 2013 changes is that they indicate that regulations continue to evolve and change within the landscape of health information and are responsive to healthcare consumers' needs in the digital age. The changes related to genomic data are reflective of this ever-expanding landscape. The final mle is ba.sed on statutory changes under the HITECH Act, enacted as part of the American Recovery and Reinvestment Act of 2009, and the GINA of 2008, which clarifies that genetic information is protected under the HIPAA Privacy Rule. The most important aspect of these changes is that this mle prohibits most health plans from using or disclosing genetic information for underwriting purposes. Additional requirements that may be particularly relevant to APRNs and others providing healthcare services and billing for those services include the following: ■ Restrictions on disclosure of PHI that are relevant to genomic information about the individual.
■ Information about services paid for out of pocket must be withheld from the payer on the patient's request. ■ Clarification and guidance regarding patient access rights to protected health information.
■ Treatment, payment, and healthcare operations disclosures must be tracked and records should be maintained for 3 years. ■ CEs with EHRs must provide or transmit PHI in electronic format as directed by the patient.
14: PRIVACY AND SECURITY IN A UBIQUITOUS HIT WORLD
■ Limits for uses and disclosures related to marketing and fund-raising are prohibited. ■ Accountability to BAs and subcontractors is extended to those entities for protection of information and regulatory penalty.
The privacy and security changes in the final rulemaking provide the public with increased protection and control of PHI, and individual rights are expanded in important
ways. Patients can ask for a copy of their record in an electronic form. When individuals pay in cash, they can instruct the provider not to share information about their treatment with their health plan. The final omnibus rule sets new limits on how information is used and disclosed for marketing and fund-raising purposes and prohibits the sale of an individual's health information without their permission. BAs and subcontractors must
now comply with the requirements and are directly liable for violations. An additional emphasis of the HITECH Act is patient engagement. The patient engagement movement will continue to drive forward new and innovative ways to involve patients in care. Patients wishing to incorporate applications and technological tools for health and fitness into their healthcare regimen .should be able to feel .secure in the knowledge that these tools have built-in safeguards and that providers have evaluated the safe and secure use of such tools. This will require the involvement of nurses in all stages of development, including representation at the federal policy level. These new advances will also require enhanced cyberseairity because the majority of these applications are internet based. Proposed 2020 Modifications On December 10, 2020 OCR published the latest round of proposed changes to HIPAA via a Notice of Propo.sed Rule Making (NPRM). Included in the proposed changes are provisions to increase permissible disclosures of PHI and improve care coordination by: a.
adding definitions for the terms "electronic health records" and "personal health application";
b. strengthening a patients right to inspect their PHI in person, shortening a CE's c.
d.
e.
response time from 30 days to 15 days; requiring CEs to inform patients that they retain their right to obtain or direct copies of PHI to a third party when a summary of PHI is offered in lieu of a copy; reducing identification verification burdens on patients exercising their access rights; requiring healthcare providers and health plans to submit a patient's access request to another healthcare provider and to receive back the requested electronic copies of the patient's PHI in an EHR;
specifying when electronic PHI must be provided to patient's at no charge; g- specifying other Items related to the permitted charge CEs can charge patients; h. amending the definition of healthcare operations to clarify permitted uses and f.
disclosures for individual-level care coordination
and case management; and
i. replacing the privacy standard that permits CEs to make certain uses and disclosures of PHI based on their "professional judgment" with a standard permitting such uses or disclosures based on a CE's good faith belief that the u.se or disclosure is in the best interests of the individual. Public comment for
the NPRM closed on May 6, 2021
and any finalization of the proposed mle will likely occur in 2022.^ ' 2020 Fed Reg 27157 (Jan 20,2021).
357
358
II: POINT-OF-CARE TECHNOLOGY
ROLE OFTHE CLINICIAN INTHE PROTECTION OF PROTECTED HEALTH INFORMATION
Surveys indicate that the majority of privacy and security breaches often result from human error or negligence. In fact, the fourth annual benchmark study on privacy and data security released by the Ponemon Institute in March 2014 indicates that 75% of organizations surveyed say that employee negligence is their biggest concern (Ponemon Institute, 2014). Organizations recognize that there are gaps in policy, technology, and education that can lead to negligence. This is an area that nurses must be cognizant of and look for ways to mitigate.
Clinicians' Responsibilities
What can clinicians do to help protect healthcare consumers' PHI? All clinicians should take steps to comply with HIPAA, include awareness of the different components of HIPAA, stress professional commitments to advocacy for healthcare consumers and patients, and have heightened awareness of where PHI might be exposed. Examples of issues arising that constitute HIPAA violations are:
■ Text messages that constitute unsecure messaging (routine mobile phone texts); ■ Photos taken with mobile devices in the workplace with the potential for exposing computer screens with subsequent posting to Facebook and other social media pages; and
■ Student nurses taking photos of the first injection given with patient labels on syringes may constitute HIPAA violations.
These actions, though seemingly imiocent, violate patients' rights under HIPAA. The profession of nursing is responsible to patients for those protections. In 2014, the American Nurses As.sociation (ANA) issued a privacy and confidentiality statement for members:
"Ongomg advances in technology, including computerized medical databases, telehealth, social media and other Internet-based technologies, have increased the likelihood of potential and unintentional breaches of private/confidential health information. The purpose of this position statement is to speak on the role of nurses in protecting privacy and confidentiality and in providing recommendations to avoid a breach." (ANA, 2015, "Purpose")
In addition, nurses should stay up to date with all changes to HIPAA, including new provisions under the HITECH Act with an understanding of how Meaningful Use (MU) measures relate to privacy and security regulations. The CMS defines the EHR incentive program, and the ONC defines the EHR certification criteria. The use of certified EHR technology ensures that the technology can support the requirements of MU.
Promoting Interoperability Program The CMS Promoting Interoperability Program (PIP; previously Meaningful Use Program)
requires that electronic health information created or maintained in a certified EHR be
protected with appropriate technical capabilities. Required by these provisions are basic security risk assessments and mitigation of any areas that present risk.
14: PRIVACY AND SECURITY IN A UBIQUITOUS HIT WORLD
The Basics of a Security Risk Assessment
A security risk assessment (SRA) is an important step a hospital or clinic can take to identify risks and vulnerabilities to PHI (e.g., breaches of HIPAA requirements). Such incidents might include roles within the organization that are not properly assigned, allowing individuals to see PHI inappropriately, and in misuse of portable devices that store PHI. The basic steps for an assessment recommended by the ONC include 1. Reviewing the existing seairity of PHI; 2. 3. 4. 5.
Identifying threats and vulnerabilities; Assessing risks for likelihood and impact; Mitigating security risk; and Monitoring results (ONC, n.d.).
Risk Assessments—An Important Role for the Nursing Informaticist and Nursing Leadership The role of the APRN involves several important responsibilities such as knowing and implementing the requirements of the PIP including HIPAA requirements. One of those requirements is an SRA. The ONC and the OCR, recognizing the challenging task of SRAs, have provided an online tool that can be downloaded in Windows or mobile device
applications. Tlie tool walks the end user through the process of assessing the organization. It is a self-contained tool that is question based, guiding the organization through a series of 156 questions. In addition to the software application, the ONC has provided a comprehensive user's guide (ONC, n.d.). This tool was designed for small providers to use and helps comply with HIPAA and the PIP requirements. The authors recommend that readers review the SRA tool along with this chapter (www.healthit.gov/providersprofessionals/security-risk-assessment-tool). The clinical informaticists and HIT professionals require additional expertise as to
how a security audit might take place and the awareness of what constitutes a full audit. Generally, in larger institutions, security professionals are responsible for conducting and analyzing audits and the nurse informaticist may be involved with gathering data for the actual audit. However, nurses may be responsible for conducting the audit in small clinics. Critical Access Hospitals (CAHs), and small rural community hospitals. With this type of assessment tool, the scrutiny of how the organization adheres to policy is critical. An important step in assessment is the observation of practices in place, not simply a policy stating that the staff are adhering to policy and procedures. The staff must be aware of all policy and procedures to protect PHI, maintain the security of PHI, and follow the policies and procedures. As noted earlier, the human factor is often the largest challenge in organizations' violations of HIPAA law. Training and education of all staff may also be the responsibility of nursing leadership, particularly in smaller facilities with limited resources.
THE NURSE'S ROLE IS IMPORTANT IN ESTABLISHING PUBLICTRUST
Beyond state and federal implications, nursing boards are also taking actions regarding issues related to disclosures and on events related to social media breaches, and they are ensuring nurses are accountable for their actions related to patient health information (ANA, 2014). Nurses should familiarize themselves with the changes under way and demonstrate behaviors that exemplify these standards.
359
360
II: POINT-OF-CARE TECHNOLOGY
Nurses play an important role in establishing and maintaining patient trust. To preserve that special relationship, we must take steps to understand the mles that exist, implement ways to protect and secure information, and educate staff and patients. To maintain and protect the confidentiality of patient information, it is critical that nurses understand the intersection of HIT and how it contributes
relationship.
to the clinician-patient trust
Nurses should educate themselves on federal and state regulations and policies
that impact their patient populations and, where possible, become involved in policy making activities to ensure the voice of nursing is represented. Formulating institutional
and organizational policies that represent the rights of patients guaranteed by the law and implementing them accurately can be highly effective for mitigating the damaging effects that result without rules being articulated clearly for healthcare providers and patients to understand.
Nurses should take time to understand the security potential and limitations related to technology that are a part of their work environment. This professional obligation requires that nurses think through the workflows and how they impact the security of patient information and work with vendors to build solutions that work and that providers and patients can trust. Other roles that nursing leaders may play include national standards work on security and transport of data, maintaining institutional policies, particularly in areas related to encryption and use of email, texting, and mobile devices. Nursing leadership is also responsible for the education of the nursing staff at all levels, acting as a role model and mentor by demonstrating effective actions and steps to protect the PHI of the public. These responsibilities are significant in terms of maintaining the public trust. POPULATION HEALTH AND RESEARCH DATA
Although provisions for research are clearly stated in the Privacy Rule, access to data has, in some respects, become more restrictive as public and private entities, including research repositories, have tightened the access to data in the name of maintaining secure PHI. Other avenues have opened up public domain data in remarkable ways under federal initiatives for open access to data; this can improve the nation's health, safety, and strength. Under the Open Government Initiative established by the Obama administration, the president states:
"My administration is committed to creating an unprecedented level of openness in Government. We loill work together to ensure the public trust and establish a system of transparency, public participation, and collaboration. Openness will strengthen our democracy a?id promote efficiency and effectiveness in Government." (Obama, 2015)
We encourage the reader to access www.data.gov and explore sources available in the healthcare domain. The website is home to the U.S. government's open data with data, tools, and other resources for researchers and developers. However, audits coupled with
the fear of increasing penalties under the HITECH Act could result in organizations being less likely to take any risks associated with managing the access to data that may identify individuals within the research repository or data set in the private sector. Although advances in technology have vastly improved the amount and quality of data collected and could facilitate probing analyses not done three decades ago, researchers are rarely given access even when detailed protocols are provided (Wartenberg & Thompson, 2010). Examples of public domain data are the National Center of Health Statistics (NCHS) compilation of de-identified birth, death, and fetal death data for the entire country.
14: PRIVACY AND SECURITY IN A UBIQUITOUS HIT WORLD
Many attest to significant value in these types of de-identified sources as recently reflected in public testimony provided in December 2014 to the ONC Health IT Policy Committee Privacy and Security Work Group (2015). Others express concern that the de identification provides limited use for epidemiologic studies, particularly with respect to people with chronic illness (McGraw, 2009; Wartenberg & Thompson, 2010). An example of expressed concern is reflected in Wartenberg and Thompson's claim (2010) that the Department of Veterans Affairs had instructed its hospitals to protect patient privacy; as a result, protections are no longer provided for cancer surveillance data at federal and state levels. These researchers indicate that the lack in submission creates a gap in
the overall interpretation of data and is a disservice to our veterans in preventing and
treating cancer (Wartenberg & Thompson, 2010). Researchers have voiced their concerns regarding the negative impact that the HIPAA Privacy Rule has on access to data that are necessary to achieve credible and reliable study results; however, measurable proof in the protection of data is still lacking (Nass et al., 2009). As patient advocates, healthcare providers, and researchers, we must strike a balance between patient privacy protection and the research process (Bova et al., 2012). We should gain the trust of patients by possessing a thorough understanding of the protection of personal information (Rho et al., 2013). Work is under way to define the permitted uses under HIPAA for research in a big data world with initiatives such as Patient-Centered Outcomes Research. The goal is to build a trust framework to support such data uses (Patient Centered Outcomes Research Institute [PCORI], 2015). CASE STUDY
You are the newly appointed advanced nurse practitioner to a large children's hospital specialty clinic and oversee all nurse practitioners in the clinic. Your Chief Executive orients you to the various clinics and states, 'We need you to be aware we have a real issue with the way our clinicians are using the EHR, waiting until the end of the shift to document the clinicai care. We think they might be carrying paper around in their pockets, maybe even taking it home with them to document in the evening. They could even be violating HIPAA. Can you make this a priority for your first few weeks in the clinics to investigate?" As a result of this conversation during orientation, you realize one of your first tasks is to fully review the federally posted Security Risk Assessment (SRA) on HealthlT.gov and encourage your clinic leadership to join you in implementing a full risk assessment. You determine that there have been breaches in several
clinics with PHI data. Consider the following:
What is your first course of action? Are you required to report the breaches?
How will you handle communication and improvement strategies within clinic?
SUMMARY
We have reviewed background information on the protection of healthcare data under the HIPAA regulatory requirements established in 1996 and updates to that regulation enacted with the HITECH Act to increase the protection of data. Each component of HIPAA was discussed, including transactions, privacy, and seairity rules. We have discussed the
361
362
II: POINT-OF-CARE TECHNOLOGY
enforcement component and how enforcement and audits have increased with sizable penalties for disclosures by organizations across the United States. In addition, we have examined other regulatory requirements, including state laws that can override federal law when considered more stringent with respect to protections. We have also discussed tlie importance of public trust and nursing roles that are important for establishing strong policies and procedures in protecting the PHI and identity of individuals we care for daily. Most issues with security breaches are related to vulnerabilities created by human errors; therefore, we have also related common incidences of HIPAA violations and reviewed why they occurred and how to mitigate these types of incidents while emphasizing the importance of SRAs in that process. Population health data and the "push and pull" between protecting privacy and disclosing adequate information to address epidemiologic and other research questions were discussed. Also examined was whether we as a nation have the rightbalance between open government and public disclosure of healthcare data for the common good versus potential risks of disclosing public domain data that might be used to identify individuals. New threats regarding cybersecurity were reviewed with special attention as to why these new threats are occurring, what they are, and what we can do as an industry to guard against these exposures. Technical terms related to cybersecurity were noted and used within the context of threats that all healthcare professionals should be aware of to help guard against disclosures. Finally, an exercise is presented to consider the SRA tool established by the ONC and the Office for Civil Rights (OCR) to assess organizations for adequate protections. END-OF-CHAPTER
RESOURCES
EXERCISES AND QUESTIONS FOR CONSIDERATION
Considering information related in this chapter, identify a clinical environment in which PHI or personal identifiers might be vulnerable to exposure. Download the SRA tool available on the HealthIT.gov website: www.healthit.gov/provider s-professionals/ security-risk-as.sessment-tool.
The SRA takes you through each HIPAA requirement by presenting questions about your organization's activities. Your "yes" or "no" answer will show you whether you need to take corrective action for that particular item. There are a total of 156 questions. Use the SRA to assess the healthcare organization you have chosen and write a report for the organization as to areas it needs to consider. Consider the following questions: 1. Which areas do you consider as high risk and what actions should be taken with these vulnerabilities?
2. Whose responsibility is it to address vulnerabilities and risks to PHI?
3. In the event an audit occurs at this point, what do you believe the organization would do regarding adherence to HIPAA? What recommendations would you make to the organization to prepare for such an audit? 4. How important are policies and procedures for adherence to the HIPAA protections of PHI? Is having a policy in place adequate evidence of meeting requirements under HIPAA of protecting PHI?
I4s PRIVACY AND SECURITY IN A UBIQUITOUS HIT WORLD
ADDITIONAL RESOURCES
SnuUfUNtllSr.’.'
CONNECT
A robust set of instructor resources designed to supplement this text is located at http://connect.sprlngerpub.com/content/book/978-0-8261 -8526-6. Qualifying instructors may request access by emailing textbook@springerpub. com.
REFERENCES
American Nurses Association. (2014). 6 tips for nurses using social media, https://www.nursingworld
.org/~4af5ec/globalassets/docs/ana/ethics/6_tips_for_nurses_using_social_-media_card_web.pdf
American Nurses Association. (2015). Privacy and confidentiality , https://www.nursingworld.org/
practice-policy/nursing-excellence/official-position-statements/ id/privacy-and-confidentiality
Atchinson, B. K., & Fox, D. M. (1997), The politics of the Health Insurance Portability and Accountability
Act, Health Affairs, 16(3), 146-150, https://doi.org/10,1377/hlt haff.l6.3.146; http://www,library .armstrong.edu / eres / docs / eres/ MHSA8635-l_CROSBY / 8635_week2_HIPAA_politics.pdf Bova, C., Drexler, D., & Sullivan-Bolyai, S, (2012), Reframing the influence of the Health Insurance Portability and Accountability Act on research. Chest, 141(3), 782-786. https://doi.org/10.1378/ chest.11-2182
Centers for Medicare & Medicaid Services, (2013). HIPAA: General information, https;/ /www.hhs.gov/
hipaa / for-professionals / privacy / la ws-regulations / index.html
Centers for Medicare & Medicaid Services, (2015). lCD-10. https;//www.cms.gov/Medicare/Coding/ ICDIO/ index.html?redirect = / icdlO Centers for Medicare & Medicaid Services. (2017). Transactions ovenneio. https://www.cms.gov/
ReguIations-and-Guidance/Administrative-Simplification/Transacti ons/TransactionsOverview .html
Clearwater Compliance. (2012). Preparingfor the HIPAA security rule again; now, unth teeth from the HITECH Act! (White Paper for Health Care Professionals), https:/ /clearwatercompliance.com/wp-content/ uploads/HIPAA-HlTECH-Security-Rule-FAQ_WhitePaper_v6.pdf Dolan, C. M.,& Farmer, L.J. (2016). "Let the record speak..The power of the medical record. The journal
for Nurse Practitioners, 12(2), 88-94. https; / / doi.org/https: / /doi.org/10.1016/j.nurpra.2015.09.018 eCFR.gov, (2014, December). Electronic code of federal regulations, https://www.ecfr.gov/cgi-bin/text
-idx?SID = 6476a9ffd68705614e3599e553a393fe&node = se45.1.160_1103&rgn = div8 Eden,)., Wheatley, B., McNeil, B., & Sox, H. (2008). Knoxoing xohat ivorks in health care: A roadmap for the nation (Brief). National Academies Press, https: / /books.nap.edu /openbook.php?record_id = 12038 Federal Trade Commission. (2014). 2014 privacy and data security update, https; / / www.ftc.gov/system/ files / documents / reports / privacy-data-security-update-2014 / -privacydatasecurityupdate_2014 .pdf Gordon, E. (2018). Aetna agrees to pay SI7 million in HIV prixmcy breach, https://www.npr.org/sections/ health-shots/2018/01/17/572312972/aetna-agrees-to-pay-17-million -in-hiv-privacy-breach Hall, M., Dugan, E., Camacho, F., Kidd, K., Aneil, M., & Balkrishnan, R. (2002). Measuring patients' trust in their primary care providers. Medical Care Research and Reviexo, 59, 293-318. https://doi .org /10.1177 /1077558702059003004 Health and Human Services. (2021). Health Information Prii’acy Enforcement highlights, https; / / www.hhs .gov / hipaa / for-professionals / compliance-enforcement / data / enforcement-highlights / index.html Healthcare Information and Management Systems Society. (2015). Standards for health informatics and HIT systems, https://www.himss.org/library/interoperability-standard s/standard-health-informatics -hit-systems Health Information Law. (n.d.). Health information and the laio: States. Retrieved October 4, 2021, from http:/ /www .healthinfolaw.org/state Health IT Policy Committee Privacy and Security Workgroup. (2015). Health big data recommendations. https://www.healthit.gov/sites/default/files/facas/HITPC_Health_Big_Data_Report_FINAL.pdf Kushniruk, A., & Borycki, E. (2013), Human factors in healthcare IT. In KM McCormick & B Gugerty (Eds.), Healthcare information technology exam guide. McGraw-Hill. McGraw, D. (2009). Privacy and health information technology. Journal ofLaxo Medicine & Ethics, 37(3), 121-149. https: / /doi.org/10.1111 /j.l748-720X.2009.00424.x Nahm, E.-S., Poe, S., Lacey, D., Lardner, M., Van De Castle, B., & Powell, K. (2019). Cybersecurity essentials for nursing informaticists. CIN: Computers, Informatics, Nursing, 37(8). https://journals .lww.com/cinjournal/Fulltext/2019/08000/Cybersecurity_Essentials_for_Nursing.l.aspx
363
364
II: POINT-OF-CARE TECHNOLOGY
Nass, S.
Levit, L. A., Gostin, L. O., & Institute of
Medicine. (2009), Beyond the HIPAA prwacy rule:
Enhancing privacy, improving health through research. National Academies Press, http://www.ncbi .nlm.nih.gov/books/NBK9576/ NitroSecurity & FairWarning. (n.d.). Security and prwacy of electronic medical records (White Paper). Healthcare Information and Management Systems Society. Retrieved October 4, 2021, from https: / /
www.himss.org/nitro-white-paper-security-and-privacy-electronic- medical-records Obama, B, (2015). State of the union, https://www.cnn.eom/2015/0 1/20/politics/state-of-the-union -2015-transcript-full-text/index.html Office for Civil Rights, (n.d.-a). Enforcement process. Retrieved October 4, 2021, from https: / /www.hhs .gov/ hipaa/ for-professionals/ compliance-enforcement / enforcement-process / index.html Office for Civil Rights. (n.d.-b). Guidance regarding methods for de-identification ofprotected health information in accordance with the Health Insurance Portability and Accountability Act (HIPAA) privacy rule. Retrieved October 4, 2021, from https:/ / www.hhs.gov/ocr/privacy/hipaa/un derstanding/coveredentities/ De-identification/guidance.html#_edn2 Office for Civil Rights, (n.d.-c). Health information privacy: Stolen laptops lead to important HIPAA settlements. Retrieved October 4, 2021, from https://www.hhs.gov/ocr/privacy/ hipaa/enforcement/examples/ stolenlaptops-agreements.html
Office for Civil Rights, (n.d.-d). Health information privacy: The HIPAA privacy rule. Retrieved October 4, 2021, from https: / /www.hhs.gov/ocr/privacy/hipaa/administrativ e/privacyrule/index.html Office for Civil Rights, (n.d.-e). HIPAA enforcement. Retrieved October 4, 2021, from https: / / www.hhs .gov/ hipaa / for-professionals / compliance-enforcement / index.html Office for Civil Rights, (n.d.-f). How OCR enforces the HIPAA Privacy & Security Rules. Retrieved October 4, 2021, from https://www.hhs.gov/hipaa/for-professionals/compli ance-enforcement/examples/ -how-OCR-enforces-the-HIPAA-privacy-and-security-rules/index.htm l Office for Civil Rights. (2003). OCR HIPAA privacy: Business associates, https://www.hhs.gov/sites/ default / files / ocr / privacy / hipaa / understanding / coveredentities / -businessassociates.pdf Office for Civil Rights. (2013). New rule protects patient privacy, secures health information, https: / / www .businesswire.com/ news/home/20130117006506/ en/New-rule-protect s-patient-privacy-secures -health
Office of the National Coordinator for Health Information
Technology, (n.d.). Security risk assessment tool. Retrieved October 4, 2021, from https://www.healthit.gov/p roviders-professionals/security
-risk-assessment-tool
Ohio Bureau of Workers' Compensation, (n.d,). EDI implementation guide: Provider billing flow. Retrieved October 4, 2021, from https: / / www.bwc.ohio.gov/provider/servi ces/techlevel.asp Patient Centered Outcomes Research Institute. (2015). About us. http:/ /www.pcori .org/about-us Ponemon Institute. (2014). Learn to manage your prwacy & security risks, https://www.ponemon.org/ local/upload/file/ID%20ExpertsPatient%20Privacy%20%26%20Data%20S ecurity%20Report%20 FINALl-l.pdf
Rho, M. J., Jang, K. S., Chung, K. Y, & Choi, I, Y, (2013). Comparison of knowledge, attitudes, and trust for the use of personal health information in clinical research. Multimedia Tools and Applications, 67(3), 1^. https: / / doi.org /10.1007 / sll042-013-l 772-6 SubstanceAbuse and Mental HealthServicesAdministration.(2014). Laws and regulations, http: / / www .samhsa.gov/about-us/who-we-are/laws-regulations U.S. Department of Health & Human Services. (2013). Modifications to the HIPAA privacy, security, enforcement, and breach notification rules under the Health Information Technology for Economic and Clinical Health Act and the Genetic Information Nondiscrimination Act; other modifications to the HIPAA rules: Final rule. https://www.gpo.gOv/fdsys/pkg/FR-2013-01-25/pdf/2013-01073 .pdf U.S. Food & Drug Administration. (2013). Cybersecurity for medical devices and hospital networks: FDA safety communication, https://www,fda.gov/medicaldevices/safety/ -alertsandnotices/ucm356423 .htm
Wartenberg, D., & Thompson, D. W. (2010). Privacy versus public health: The impact of current confidentiality rules. American journal of Public Health, 200(3), 407-412. https://doi.org/10.2105/ AJPH.2009.166249
Personal Health
cords aod
Patient Portals MARI TIETZE AND STEPHANIE H. HOELSCHER
OBJECTIVES ●
Appraise the relationship among the personal health record (PHR)/portal. patient engagement/ activation, and patient safety and quality.
●
Identify factors associated with increased patient PHR/portal use.
●
Examine advantages and disadvantages of patient-generated health information.
●
Identify components of the ideal patient portal.
●
Predict achievable levels of patient engagement/activation in one's practice.
●
Model the PHR/portal implementation using the interprofessiona l approach to increase patient use. CONTENTS
INTRODUCTION
366
REGULATORY REQUIREMENTS FORTHE PATIENT PORTAL
Meaningful Use
368
368
MU Stage 3 and MIPS ACI Objectives and Measures DIGITAL DIVIDE AND HEALTH LITERACY
369
371
PHR/PORTAL IMPACT ON PATIENT SAFETY, UNINTENDED CONSEQUENCES CONTRIBUTION OF PERSONAL DATA-TELLINGTHE PATIENT'S STORY A Patient's Questions Are the Answer and New Question Builder App Patient-Generated Health Information
375
Personal Health Data for Public Good
376
IndustryTrends and Community Involvement
373
373
373
376
INTERPROFESSIONAL INFORMATICS TO INCREASE PERSONAL HEALTH RECORD/PORTAL
USE
377
Methods of Portal Adoption Characteristics for Success
377 378
The contributions of Cristina Winters to this cha}}tcr in the preihoiis edition of this book arc acknowledged here.
366
II: POINT-OF-CARE TECHNOLOGY
CASE STUDY: MYCHARTCARE SUMMARY
380
382
EXERCISES AND QUESTIONS FOR CONSIDERATION REFERENCES
382
382
INTRODUCTION
The Federal Health information technology (IT) Strategic Plan 2020-2025 final report had a clear message for patients and providers that access to health data by the patient is key (Office of the National Coordinator for Health Information Technology [ONC], 2020). This strategy is an evolution of the 2015 efforts to "collect data, share data and use data" (ONC, 2014, p. 5). The strategies set forth by the latest plan are as follows (ONC, 2020, pp. 21-22): ■ Enable individuals to access their health information by to view and interact with their data via secure mobile other tools.
ensuring that they are able apps, patient portals, and
■ Promote greater portability of health information through application program interfaces (APIs) and other interoperable health IT permits individuals to readily send and receive their data across various platforms.
■ Improve access to smartphones and other technologies needed to attain and use health information, especially for at-risk, minority, mral, disabled, and tribal populations. ■ Build the evidence base on the use of health information, including the types of information that will benefit individuals most and
the best way to present information
to patients and caregivers. ■ Provide resources on how to access and use health
information so that patients and caregivers understand how to use their data safely, securely, and effectively.
The use of EHRs has risen dramatically in recent years in private practice. According to a National Electronic Health Records survey in 2017 with the National Center for Health Statistics, the amount of office-base providers using a certified electronic health record (EHR) technology (CEHRT) system in the United States was almost 86%, and using a certified EHR was almost 80% (Myrick etal., 2019). Important to this growth is the PHR, a component associated with the EHR that provides specific access via an electronic portal for the patient's view of their information. A PHR is defined by the Social Security Act (42 use 1320d[6]) as including "individually identifiable health information that includes, with respect to an individual, information: (A) that is provided by or on behalf of the individual; and (B) that identifies the individual, or with respect to which there is a reasonable basis to believe that the information can be used to identify the individual" (see Congressional Record—House, February 12, 2009, p. H1348, which is available online at www.ssa.gov/OP_Home/ssact/ titlell / 1171.htm).
The National Learning Consortium indicated that a patient portal, on which a PHR exists, is a "secure online website that gives patients convenient 24-hour access to personal health information from anywhere with an Internet connection" (HealthIT.gov National
IS: PERSONAL HEALTH RECORDS AND PATIENT PORTALS
Learning Consortium, 2014, p. 1). Using a secure username and password, patients can view health information, such as the following: ■ Summarized health profile (e.g., most recent vital signs or weight) ■ Recent provider visits and notes
■ Discharge summaries ■ Medications ■ Immunizations
■ Allergies
■ ■ ■ ■
Lab results, pathology reports Radiology results Procedure history Trackers (e.g., patient can self-report weight, steps, blood pressure, blood glucose)
Some patient portals also allow patients to do the following: ■ Exchange secure emails with their healthcare teams ■ Request prescription refills ■ Schedule non-urgent appointments ■ Check benefits and coverage ■ Update contact information ■ Make payments ■ Download and complete forms ■ Pre-register for visits
■ Send images such as photos of rashes and wounds ■ Input patient-generated health data (PGHD) such as histories, allergies, medications ■ View educational materials (ONC, 2017, p. 1) With patient portal implementation, an organization can enhance patient-provider communication, empower patients, support care between visits, and, most important, improve patient outcomes (HealthIT.gov National Learning Consortium, 2014, p. 1). As illustrated in Figure 15.1, the components of the PHR/patient portal typically include (a) patient records/history (from the provider's main EHR); (b) educational/training documents; (c) collaboration methods, such as email 24 hours a day, to communicate with healthcare professionals; and (d) quality metrics, such as outcome measures, that demonstrate progress over time (Cognator.com, 2014). Regarding efficiencies, the portal provides a means for the provider to send messages
to the patient and ease workflow by reducing phone messages and unscheduled visits by the patient (Clarke et al., 2013). Patients benefit from portal use, as they are allowed to use the services provided rather than waiting for long periods for phone calls to be returned by the clinic. Portal u.se may also be cost-effective in that it may decrease the need for repeated tests and procedures by specialists or emergency care providers, as the patient has the ability to access diagnostic procedures and interventions from a laptop, computer, tablet, or smartphone (Di Maio, 2010).
367
368
I; POINT-OF-CARE TECHNOLOGY
FIGURE 15.1 Patient Pathway (Portal) and Associated Sources of Health Improvement Information and Professionals.
^^Patient
I
|V^
HQuality
Patient
1
i,
"ij
!> Educational/ -
* training
documents
f
\-o/ Collaboration
mu r Source; Cognator.com. (2014). Empowering the patient, transforming healthcare. http://www .cognitor.com/patient-pathway.
REGULATORY REQUIREMENTS FOR THE PATIENT PORTAL
Meaningful Use According to the 111th U.S. Congress, H.R.l - American Recovery and Reinvestment Act of 2009 which became public law on February 17, 2009, meaningful use (MU) was originally defined by the use of certified EHR technology in a meaningful manner (e.g., electronic prescribing); ensuring that the certified EHR technology is connected in a manner that provides for the electronic exchange of health information to improve the quality of care; and that in using certified EHR technology the provider must submit to the Secretary of DHHS information on quality of care and other measures (U.S. Congress, 2009, p. 123 Stat. 470). During stage 2 of MU, healthcare providers had to implement a patient portal as a part of the EHR in 2014. The requirements were that the eligible provider must provide patients with the ability to view online, download, and transmit their health information. An eligible provider is any physician or advanced practice provider who participates in the Medicare or Medicaid EHR incentive program (McGraw, 2012). The eligible provider must provide "50% of all unique patients seen within a reporting period access within 4 business days of when the information is available to them" (Netti, 2013, p. 1). The requirements for stage 2 of MU are that 5% of patients must download and/or transmit information via the portal (Netti, 2013). In 2017, stage 3 was implemented and moved the focus to improve patient health outcomes through the use of CEHRT (Cherry & Jacob, 2019).
15: PERSONAL HEALTH RECORDS AND PATIENT PORTALS
As stage 2 of MU requirements existed, the patient portal had to be implemented and utilized by patients and providers for the providers to continue to receive financial reimbursements for services. These services are based on submitted allowable Medicare
charges or a set payment for the Medicaid incentive pathway. A penalty was imposed on eligible providers who did not successfully report electronic health record MU (EHRMU) criteria by 2015 (Harrison et al, 2007). With 2016 came the introduction of the new Medicare reimbursement model, the Medicare Access and CHIP Reauthorization Act (MACRA) of 2015, in which variance
of MU objectives and measures were outlined. Merit-Based Incentive Payment System (MIPS) is one of MACRAs Quality Payment Program (QPP) measurement routes. MIPS is divided into four categories: quality, cost and resource use, clinical practice improvement activities, and advancing care information (ACI). The MIPS ACI closely echoes stage 3 MU measures and objectives. Between the two, the new Riles for eligible providers (EPs), eligible hospitals (EHs), and critical access hospitals (CAHs) include the following (CMS, 2016). Another critical, overarching requirement is that of privacy and security of the
patient data. Most PHRs are offered by healthcare providers and health plans covered by the Health Insurance Portability and Accountability Act of 1996 (HIPAA) Privacy Rule. Some PHRs are covered under other applicable laws and it is important to be aware of this distinction (Office for Civil Rights, 2020).
MU Stage 3 and MIPS ACI Objectives and Measures Patient Electronic Access to Health Information
■ Measure 1: Patient access—view, download, transmit (VDT) via patient portal and
application programming interface (API) ■ Measure 2: Patient-specific education Coordination of CareThrough Patient Engagement
■ Measure 1: VDT via patient portal, API, or both ■ Measure 2: Secure messaging
■ Measure 3: Patient-generated health data (PGHD) It is easy to see how large a role the patient portal can play in patient engagement. Table 15.1 provides an overview of the main objectives and associated measures for MIPS. The italicized text directly addresses the expected functionality for patient engagement.
Therefore, it is financially important for providers to educate patients on the use and benefits of the patient portal. Subsequently, providers hope that these patients will enroll in and use this EHR feature that has been made available to them. From a clinical perspective and the approach used in this chapter, the underlying rationale for PHRs and associated patient portals is to facilitate the patient's engagement/activa tion in the care delivery process. Specifically, as noted, patient activation is defined as outlined in Table 15.1. Other considerations are:
■ Understanding that one must take charge of one's health and that actions determine health outcomes
■ A process of gaining skills, knowledge, and behaviors to manage health ■ Confidence to make needed changes (Hibbard et al., 2004, p. 1010).
369
370
II: POINT-OF-CARE TECHNOLOGY
TABLE 15.1 MIPS Advancing Care Information Objectives and Measures The following is the full list of objectives and measures for MIPS indicators. Italics indicates the expected functionality for patient engagement. OBJECTIVE
MEASURE
Protect Patient Health Information
Security Risk Analysis'
Electronic Prescribing
ePrescribing*’
Patient Electronic Access
Patient Access
Patient-Specific Education
Coordination of CareThrough Patient Engagement
View, Download, andJransmit (VDT)'=
Secure Messaging Patient-Generated Health Data
Health Information Exchange
Exchange Information With Other Physicians or Clinicians Exchange Information With Patients Clinical Information Reconciliation®
Public Health and Clinical Data
Registry Reporting
Immunization Registry Reporting
(Optional) Syndromic Surveillance Reporting (Optional) Electronic Case Reporting
(Optional) Public Health Registry Reporting (Optional) Clinical Data Registry Reporting '■Required measures.
Required to fill out the measure (either yes/no or numerator/denominator). ' Required to fill out the measure, may be selected as a part of the performance score. Source: Centers for Medicare & Medicaid Services, (2016). Merit-Based Incentive Payment System: Advancing care information (p. 6|. https://www.cms.gov/Medicare/Quality-Payment-Program/Resource -Library/Advancing-Care-Information-Performance-Category-Fact-Sh eet.pdf (p. 7).
As such, patient portal, patient activation, and subsequent patient safety and quality of care delivery are intertwined. One contributes to the other and vice versa (HealthIT.gov National Learning Consortium, 2013). Promoting Interoperability Program
In 2018, the CMS made a move to rename MUs. The Promoting Interoperability Program (PIP) was rebranded out of necessity. The CMS wanted to demonstrate its continued commitment to enhancing the patientexperience with improved access to their health data, improved interoperability functionality, as well as improved use of health information exchanges (HIEs). This change not only rebranded MU but also moved MIPS to the "Promoting Interoperability performance category" to make sure the alignment between programs was maintained (Bresnick, 2018, para. 3). In aligning PIP with MU, historically, there were three stages (CMS, 2020):
1. Stage 1 set the foundation for PIP; set requirements for EHI data, including providing patients with their EHI
2. Stage 2 focused on advancing clinical processes and ensuring that the MU of EHRs supported the National Quality Strategy. Encouraged use of CEHRT for continuous quality improvement at the point of care
3. Stage 3, from 2017 to now, focuses on CEHRT to improve health outcomes. (This rule modified Stage 2 to ea.se reporting requirements and align with other CMS programs.)
15: PERSONAL HEALTH RECORDS AND PATIENT PORTALS
DIGITAL DIVIDE AND HEALTH LITERACY
As noted, patient portals are online applications that allow the patient access to their medical information and enable communication
with their healthcare provider
(HealthIT.gov National Learning Consortium, 2012). According to Emont (2011), the use of health information technologies (HITs) and online resources has great potential to boost care quality by improving care access, efficiency, chronic disease management, and patient and family involvement. Barriers to the use of the portal exist, despite advantages of the service. Many patients do not know about the service, as implementation of a patient portal is a relatively new concept. One aspect of patient portal use that has been debated is the notion of access to the internet. In one study by Kruse et al. (2012), 638 family practice clinic patients completed questionnaires about their internet use. Of these, 499 (78%) were internet users and 139 (22%) were nonusers. Lack of computer access and not knowing how to use email or the internet were the most common barriers toilnternet use. Younger age, higher education and income, better health, and absence of a chronic illness were associated with internet use. The major factor associated with internet use among patients with chronic conditions was age. As such, the authors suggested that if older adults with chronic illness are to reap the benefits of HIT, their internet access will need to be improved. Institutions planning to offer consumer HIT .should be aware of groups with lower internet access (Kruse et al., 2012).
Despite access, the patient portal and associated PHR must be presented at a low reading level, and understandability should be matched to the target population. A 2010 report by the U.S. Office of Disease Prevention and Health Promotion defined health literacy as "the degree to which individuals have the capacity to obtain, process, and understand basic health information and services
needed to make appropriate
health decisions" (Office of Disease Prevention and Health Promotion, 2010, p. 6). The
report claims that limited health literacy affects people of all ages, races, incomes, and education levels. Still, the impact of limited health literacy disproportionately affects lower socioeconomic and minority groups. It affects people's ability to search for and use health information, adopt healthy behaviors, and act on important public health alerts. Limited health literacy is also associated with worse health outcomes and higher costs (Berkman et al., 2004).
The national action plan to improve health literacy contains seven goals to improve health literacy and suggests strategies for achieving them. They are as follows: 1. Develop and disseminate health and safety information that is accurate, accessible, and actionable.
2. Promote changes in the healthcare system that improve health information, communication, informed decision-making, and access to health .services. 3. Incorporate accurate, standard.s-ba.sed, and developmentaliy appropriate health and science information and curricula in child care and
education at the university level.
4. Support and expand local efforts to provide adult education, English- language
instruction, and culturally and linguistically appropriate health information services in the community.
5. Build partnerships, develop guidance, and change policies. 6. Increase basic research and the development, implementation, and evaluation of
practices and interventions to improve health literacy. 7. Increase the dissemination and use of evidence-based health literacy practices and
interventions (Office of Disease Prevention and Health Promotion, 2010, p. 7).
37!
372
II: POINT-OF-CARE TECHNOLOGY
An accompanying report by the Agency for Healthcare Research and Quality (AHRQ) published extensive guidelines for developers of materials and websites for the low-literacy population (Eichner & Dullabh, 2007). The report's appendix provides a comprehensive "yes/no" checklist for developers to guide them in written materials, PHRs, and patient portal creation. A few representative items for consideration are as follows:
■ Words are short, simple, and familiar (one to two syllables, no jargon, acronyms, abbreviations);
■ Unavoidable technical terms are explained; ■ Sentences are short;
■ Written in "active" voice rather than "passive" voice (use "Mary visited the clinic" rather than "The clinic was visited by Mary"); ■ Consistent use of words throughout; and ■ Reading level is not above sixth grade (Eichner & Dullabh, 2007, p. A-1).
The National Learning Consortium has supported providers with guidance and materials to optimize patient engagement in their care. One fact sheet specifies how to optimize patient portals for patient engagement and meet MU requirements (HealthIT. gov National Learning Consortium, 2013). The materials list the following actions: 1. 2. 3. 4. 5.
Make sure the portal engages patients. Have providers learn the benefits of patient portals. Understand the relationship between the patient portal and the MU. Implement portal features that support engagement. Implement the portal systematically via provider control, team focus, and efficiency. 6. Promote and facilitate patient use (HealthIT.gov National Learning Consortium, 2013).
Others have emphasized the relationship between healthcare literacy and patient safety and quality (Berkman et al., 2004; Sarkar et al., 2010; Tomsik & Briggs, 2013). Common findings are that patients with low-level literacy are at high risk for unintended healthcare events and unsuccessful care delivery. For example, compared with those who did not report any health literacy limitation, even among those with internet access, patients with diabetes reporting limited health literacy were less likely to access and navigate an internetbased patient portal than tho.se with adequate health literacy (Sarkar et al, 2010). Thu.s, the internet has the potential to greatly expand the capacity and reach of healthcare sy.stems. Current use patterns suggest that, in the absence of participatory design efforts involving those with limited health literacy, those most at risk for poor diabetes health outcomes will
fall further behind if health systems increasingly rely on internet-based services alone.
Compounding this issue, a review of the literature has suggested that effective interventions integrate strategies that motivate, empower, and encourage patients to make informed decisions and assume responsibility for self-care. However, gaps in current evidence lack information on improving adherence and .self-care for patients at an increased risk of poor adherence, including those with cognitive and functional impairments and low health literacy (Evangelista Shinnick, 2008). Not every patient will enroll in and use the portal. Although the portal is no longer a novel concept, many patients are still unfamiliar with it and the benefits that it can provide. The increase in mobile devices and health applications will be interoperable with a patient's EHR, thus providing real-time access to the patient's most recent data.
15: PERSONAL HEALTH RECORDS AND PATIENT PORTALS
On the other hand, providers are often reluctant to engage in the use of the portal for various reasons, with many fearing that patient portal access will increase their workload
or worry that their notes will be questioned (Walker et ah, 2011). The most recent information from OpenNotes (www.opennotes.org) shows that 96% of patients "all or nearly all of their clinician notes," which disputes the common concerns that patient data access could confuse the patient (Heath, 2020, para. 1). PHR/PORTAL IMPACT ON PATIENT SAFETY, UNINTENDED CONSEQUENCES
As noted in Chapter 5, "Consumer Engagement/Activation Enhanced by Technology," the book by James details events in which a failure of integrated care ended in the death of his 19-year-old .son. Among many other observatioas, James claimed that in a truly patient-centered medical system, "laws must be written to require providers to offer medical records to their patients after every office visit and hospital stay" (James, 2007, p. 122). The PHR and associated patient portals are platforms where patient-generated data, portal management, and bidirectional interaction exemplify patient-centered medical systems. However, in addition to putting the patient in the center of the medical record system, the PHR/patient portal tool, when actively used, improves the safety and quality (effectiveness and efficacy, if you will) of care delivery (James, 2013). Other studies have indicated that the level of communication between patient/family members and providers is inversely related to unintended consequences of healthcare delivery (Sittig & Ash, 2007). Historically, in 1999, with the first Institute of Medicine (lOM) report on errors in healthcare delivery, the need for increased communication by patients to healthcare providers and by healthcare providers to healthcare providers was a key recommendation (lOM, 2000), reiterated in further lOM reports (2001, 2012). The increase in technology has brought an increase in patients using that technology to access their healthcare team. With this comes the need to address the patient's level of
health literacy and technology literacy. "Despite having increased access to their health data, patients do not always understand this information or its implications, and digital health data can be difficult to navigate when displayed in a small-format, complex interface" (Baldwin et al., 2016, p. 1). It is paramount that as this patient-facing tool continues to expand in usability, the design must keep the patient-level needs in mind. For example, providing a graph with no frame of reference within a portal could cause confusion and encourage the patient to less extended use of the tool. Infantine stated, "Despite the increasing use of mHealth apps, up to 80% of apps are abandoned after only 2 weeks, suggesting more research is needed to understand what features engender longevity" (as cited in Baldwin et al., 2016, p. 2). CONTRIBUTION OF PERSONAL DATA-TELLING THE PATIENT'S STORY
A Patient's Questions Are the Answer and New Question Builder App
The AHRQ campaign to encourage communication between patients and providers focuses on patient-generated questions and associated follow-ups (AHRQ, 2014). Figure
15.2 represents the campaign's home web page where the campaign slogan displays, "Questions Are the Answers." The web page highlights the main features, such as "The 10 Questions You Should Know," before attending your next appointment with your provider. Patients are offered an option to customize those questions according to their specific needs
373
374
II: POINT-OF-CARE TECHNOLOGY
through the "Build Your Own Question List" feature. The website also guides patients with details about what to do during and after the appointment with their providers. In 2019, AHRQ, based on the success of the "Patient Questions Are the Answers campaign, advanced the program to include a consumer-facing mobile app (Agency for Healthcare Research and Quality, [AHRQ], 2021). Figure 15.2 depicts the general view of the app while the app can be obtained for free at the Apple store as the "AHRQ Question Builder." The app can be used to: //
■ Prepare and organize questions by type of medical encounter. ■ Take photos of insurance cards, pill bottles, or even a skin rash. ■ Access consumer education materials and videos.
Given the federal government's emphasis on the importance of communication (see Federal HIT Strategic Plan 2020-2025, Figure 15.3), it is suggested that, at minimum,
this web page should be prominently displayed on any PHR/patient portal for patient FIGURE 15.2 AHRQ Question Builder App.The app helps patients and caregivers prepare for medical appointments and to maximize visit time.
c CD > o
o
architects and engineers
Hi
●a
0) c& o
«
ortunities,
education and training, establishes communication
develop road maps and partnerships
plan, program road map
Q}
a. ●a
CD O
i &} 3 O CD
CD u
Stewardship e
u
Cross-functional teams ensuring standardized data definitions, tenninology, data consistency, quality controls and user acceptance testing across functional areas and geographic locations. Includes producer data stewards, consumer data stewards, technical data stewards, and business data stewards such as process owners and subject matter experts (SMEs)
●a o
< CD
B CD 3
Distributed Analytics Teams Enterprise Research
Bioinformatics
Clinical
Enterprise data and analytics assets
Rnance
Operaticms & business
17: STRATEGIC THINKING
and data exploration skills, data mining, knowledge of statistical analysis and techniques, machine learning, artificial intelligence, master data management, data quality, data modeling, user interface design, and Bl and visualization software platforms and tools. The core analytics team gathers and defines the voice of customer user needs, best practices for analytics solutions, roles and functional skill levels needed for analytics talent, titles, progression paths and salary ranges for analytics talent, and provides education, training and mentoring to the functional area analysts. Architecture/Data Management
Thearchitectureanddatamanagementfunctionsincludethetechnical infrastructure, development of the master data management plan, Bl tools administration, data model design, data quality scorecards, and ETL. Data architects and data engineers work together to design, build, and manage the enterprise architecture and data management framework. Quality
The quality of the data is measured by how effectively the data support the driven organization. Quality threads through each of the people, process, and technology aspects and throughout the lifecycle of the data. The data quality lifecycle involves three phases: define, audit, and improve. The define phase consists of defining the scope of the data quality, identifying the data quality criteria, and developing the preventive data quality processes. The audit phase involves developing a data quality measurement plan and auditing data quality by subject or functional area. The improve phase is where recommendations are made for strategic data quality programs and existing data quality issues are remediated, Data quality must be a priority for the enterprise, led by the business and ongoing (Giordano, 2015). transactions and decisions needed to create a value
Innovation and Strategy Innovation and strategy activities Include analyzing the market, seeking opportunities for product development, and developing roadmaps for short, medium, and long-term success. Responsibilities include identifying problems that can be solved with advanced analytics and data science solutions and cultivating partnerships with internal and external partners to develop and leverage innovative products. Program Management
Program management owns and manages the analytics program road map in partnership with other Center for Advanced Analytics teams under the direction of the information governance council. Program management professionals deploy and manage the user centric process for the launch of new initiatives and analytic content development; document and prioritize requirements; lead and manage deployment of Bl and analytic content; assist with education and training; and establish communication plans.
437
438
III: DATA MANAGEMENT
Stewardship Data stewardship is embedded in the business wherever data are produced,
curated, or consumed. Data stewards have a deep understanding of the data; they understand the workflows that create the data, and the business rules around
the data. Data stewards manage user access, data retention policies and user acceptance testing. Data stewards manage and monitor data quality standards and issues with the data and communicate those to the business and technical
stakeholders (Giordano, 2015). Data stewardship roles include: ■ Producer Data Stewards - produce data and lead data quality efforts in the source systems.
■ Consumer Data Stewards - consume data and focus on the interaction of data with the end users.
■ Technical Data Stewards - support the data integration processes; establish common definitions and calculations of the data; work with developers and end users in defining measures, calculations, and aggregations; and ensure the correctness and security of the data,
■ Business Data Stewards - may be departmentally or functionally focused and generally align closer to the executive or domain for that organizational area. Business data stewards may be process owners or subject matter experts for their functional area:
● Process Owners - manage the enterprise processes that can impact data creation and quality in the source systems. ● Subject Matter Experts-specialize in the subject or functional area where the data are produced, curated or consumed (Giordano, 2015). Distributed Analytics Teams
Analysts distributed across the enterprise execute most of the functionally focused analytics for their respective areas. In collaboration with the core analytics and data management teams, distributed analysts work closely with process owners, subject matter experts, and leaders to create and manage Bl and visualization solutions for their department or functional area.
SUMMARY
This chapter has provided a practical framework for HDOs as they launch and enhance their EDW and BI strategy to become analytic innovators and competitive market leaders. The framework is modular, scalable, and adaptable to meet the needs of HDOs, regardless of their size, budget, organizational complexity, and level of maturity in developing stand-alone data marts, and more complex EDW and BI technology platforms. Healthcare leaders should consider how best to leverage components of this framework and techniques to succeed under new payment reform and pay-for-performance mandates to meet their respective enterprise strategic plan goals. An organized approach to enterprise use of data includes addressing multiple dimensions of an organization including people, process, and technologies. Data integrity and governance to ensure
17; STRATEGIC THINKING
high-quality data are critical for organizations to consider with clinicians important
to all aspects of HDO's enterprise use of data. The authors conclude this chapter by presenting a robust case study with application of the E-DRAP to a healthcare system to
inform strategic development of an enterprise strategy for advanced analytics. Finally, the APRN and interprofessional teams should identify, understand, and engage those
responsible for reporting and analytics within tlieir HDO to ensure their end-user needs are being successfully met and their voices are well represented within the development of enterprise data systems. END-OF-CHAPTER
RESOURCES
EXERCISES AND QUESTIONS FOR CONSIDERATION
Consider content covered concerning data management, reporting, and analytics
development, and respond to the following questions: 1. Why should people and processes be structured and operationalized to meet your reporting and analytic business requirements? 2. Consider the assessment tool and use the tool to
assess your clinical organization. Perform the assessment, consider the challenges identified in the chapter, and reconsider your assessment, as you interview key individuals in your organization, including the chief executive officer, chief financial officer, chief nursing informatics officer, chief information officer, and chief medical informatics officer. Do their answers differ from yours? If so, why do you believe that is the case? What does that tell you about the readiness of your organization for an EDW/BI program?
3. Reflect on VOC described in this chapter for EDW/BI development and deployment. Determine which, if any, of these techniques are employed to meet your respective
reporting and analytics requests. Evaluate the effectiveness of using these techniques within your organization. 4. Consider the options for selecting vendors to assist with your E-DRAP project and evaluate which options may be best for your organization and why. ADDITIONAL RESOURCES A robust set of instructor resources designed to supplement this text is located srtiiiu>>viiiv-
CONNECT
,
at http://connect.springerpub.com/content/book/978-0-8261-8526 -6. Qualifying instructors may request access by emailing textbook@springerpub. com.
REFERENCES
AgileManifesto. (2001), The Agile manifesto, http; / / agilemanifesto.org
Ambler, S., & Lines, M. (2012). Disciplined Agile delivery: A practitioner’s guide to Agile software delwery in the enterprise.IBM Press. Ballard, D. J. (2003), Indicators to improve clinical quality across an integrated health care system.
International JournalJvr Quality in Health Care, I5(Suppl. 1), il3-i23. https:/ /doi.org/10.1093/intqhc/ mzgOSO
439
440
III: DATA MANAGEMENT
Bhatia, .P (2019). Data mining and data warehousing: Principles and practical techniques. Cambridge University Press.
Cagan, M. (2010). Inspired: Hoio to create products customers love. Silicon Valley Product Group. The Data Warehousing Institute. (2009). TDWI's Business Intelligence Maturity Model poster, https:// knowledgeworks. wordpress.com /2009 /02/171 the-tdwi-bi-maturity -model Di Maio, A. (2010). Gartner launches open government maturity model (No. G12010). http://blogs.gartner .com/andrea_dimaio/2010/06/28/gartner-launches-open-government-m aturity-model Evelson, B. (2010). Want to know what Forrester's lead data analysts are thinking about BI and data domain? https; / /go.forrester.com/blogs/want-to-know-what-forresters-le ad-data-analyst.s-are-thinking -about-bi-and-the-data-domain
Forrester Research. (2014). Business intelligence. https://www.forrester.com/Business-Intelligence George, M. L., & Vincent, .P (2002). Lean six sigma: Combining six sigma with Lean speed (1st ed.). McGrawHill.
Giordano, A. D. (2015). Perjbrming information governance: A step-by-step guide to making information governance work. IBM Press.
Informatica. (2017). Holistic data governance: A framework for competitive adi’antage. https://www .informatica.com/Ip/holistic-data-governance-framework_2297.html #fbid=Mt9HJgN6uNT Kimball, R., & Caserla, J. (2004). The data warehouse ETL toolkit: Practical techniques for extracting, cleaning, conforming, and delivering data. John Wiley & Sons, http://users,itk.ppke.hu/~szoer/DW/ Kimball%20&%20Caserta%20-The%20Data%20Warehouse%20ETL%20Toolkit% 20%5BWiley%20 2004%5D.pdf
Kiron, D,, Ferguson, R. B., & Prentice, P. K. (2013). From value to I’ision: Reimaging the possible with data analytics: What makes coinpanies that are great at analytics different from everyone else. (Research Report). MIT Sloan Management Review, http://www.sas.com/content /dam/SAS/en_us/doc/ whitepaper2/reimagining-possible-data-analytics-106272.pdf National Institutes of Health, U.S. National Library of Medicine. (2013). Supporting interoperability— Terminology, subsets and other resources from IheNLM. https: / / www.nlm.nih.gov/hit_interoperability .html
Orr, K. (1998), Data quality and systems theory. Communications of the ACM, 42(2), 66-71. https;//doi .org/10.1145/269012.269023 Sanders, D., Burton, D, A., & Protti, D. (2013). The healthcare analytics adoption model: A framework and roadmap (White Paper). HealthCatalyst. https://healthsystemcio.com/whitepapers/HC_analytics_ adoption.pdf Shaffer, V, & Beyer, M, (2014). Top actions for healthcare delivery organizations CIOs, 2U14: Awid 25 years of mistakes in enterprise data warehousing. Gartner Report, https://www.gartner.com/en/ documents /2664433
Soulsby, D. J. (2009). Creating the golden record: Better data through chemistry, https://dama-ny.com/ images/meeting/101509/ damanyc_mdmprint.pdf Stodder, D. (2013). Achieinng greater agility loith business intelligence: Improving speed and fexibility for BI, analytics, and data warehousing. TDWI Research, Best Practices Report, https://tdwi.org/~/media/ tdwi/tdwi/research/bpr/2013/tdwi_bpreport_qll3_achievingbiagilit y_web.ashx
Data Managem^;]^ The Foundations for Improvement ' SUSAN MCBRIDE, MARI TIETZE, AND SALLY BARLOW
OBJECTIVES ●
Discuss the basics of database management and related items that healthcare organizations need to consider.
●
Discuss metrics development and the complexity of designing and well documenting measures for patient safety, quality, and population health improvement.
●
Examine levels of measurement and the importance of correctly analyzing healthcare data.
●
Describe the challenges of utilizing data from the clinical setting and outline specific steps to address those challenges.
●
Describe analytic software in the healthcare setting, including business intelligence (Bl) tools and suites of products available to layer onto databases and data warehouses.
●
Discuss statistical analysis and common tests that are run for examining quality and patient safety issues.
●
Define and discuss data mapping, what it is, and how it is utilized in healthcare. CONTENTS
INTRODUCTION
442
DATA MANAGEMENT: FOUNDATIONS FOR ANALYSIS
442
Retrospective Data Warehouse, Operational, and Clinical Data Store Defined BASICS OF DATA ANALYSIS
Measurement Theory
444
Operationalizing a Measure
Exploring a Data Set
445
447
449
Graphically Examining the Data Set Data Transformation
450
451
Text Data to Numeric Values
Data Mapping
444
444
Levels of Measurement
Quality of the Data
443
451
452
Tfu’ contributions of Deb McCullough to this chapter in previous editions of this book are acknowledged.
442
III: DATA MANAGEMENT
Statistical Analysis
452
Parameter Estimates
453
Normal Distribution
453
Selecting the Right Statistical Test Control Chart
454
456
ANALYTICS AND BUSINESS INTELLIGENCE TOOLS
Spreadsheet Applications Statistical Packages
CASE STUDY SUMMARY
458
458
Business Intelligence Tools Data Visualization
460
460
467 468
EXERCISES AND QUESTIONS FOR CONSIDERATION REFERENCES
457
468
469
INTRODUCTIOIM
Data management, measures, and analytics are the foundations of improvement. In healthcare, tremendous volumes of data are available; however, very little of those data are matured into information and knowledge that generate the wisdom and critical
thinking needed in the industry to fully capitalize on the electronic data being amassed. Value-based purchasing models are driving the industry to use data in significantly new and innovative ways to compete in the healthcare industry. As with other industries, the healthcare industry is going to compete on analytics. Davenport (2005) indicates that many companies have built their businesses on "the ability to collect, analyze and act on data" (p. 2). Healthcare is no different, and this is particularly relevant with value-based purchasing models that require that we compete on quality and efficiency. So, how does the healthcare industry establish a strong base to prepare organizations to be strategic in managing data effectively and analyzing it for success? Chapter 17 discusses data management and establishing a strategy on designing and deploying an enterprise data, reporting, and analytics-driven organization. This chapter covers the basics of data management needed to mature a data set and to analyze it for improvement purposes. We discuss how to approach a data analysis project and how to evaluate a data set for data integrity and provide examples of common issues with data integrity. Levels of measurement and how those levels of measures are relevant to analytic approaches are discus.sed. Various analytic software applications are available for use with large data sets. Some of these common tools are examined, including spreadsheet applications, statistical packages, and business intelligence (BI) tool sets. We compare and contrast these tools and suggest applications for their use in common situations that are often encountered in the healthcare industry. Finally, we examine a case study using common statistical analysis and Microsoft Excel in an exercise to emphasize lessons learned. DATA MANAGEMENT: FOUNDATIONS FOR ANALYSIS
Master data management is the coordination of people, practices, and automation, which was largely covered in Chapter 17. We cover data management within this chapter as
18: DATA MANAGEMENT AND ANALYTICS: THE FOUNDATIONS FOR IMPROVEMENT
an approach to fully understanding the data in preparation for analysis. The first step is to consider the data source and the integrity of the information. Areas to understand include data creation, data structure, integration, metadata, data storage, data modeling and data usage. We define these terms for the reader and discuss the importance of the terms related to analysis. This section covers the operational data stores versus the data warehouse for analytics, the differences between them, and when one utilizes the former versus the latter.
Retrospective Data Warehouse, Operational, and Clinical Data Store Defined
A data warehouse is a retrospective store of data set up to report trends, offer comparisons, and provide strategic analysis. It can include clinical, operational, and financial data (Englebardt & Nelson, 2002). It is typically considered a nonvolatile store of data that does not change with time. In contrast, a clinical or operational data store accumulates clinical and operational data from many systems to assist clinicians in managing patient care at the point of care (Englebardt & Nelson, 2002). These types of data stores are expected to shift and change with time given the ever-changing nature of the patient. Clinical data repositories supporting the infrasti*ucture of the electronic health records (EHRs) are examples of operational and clinical data stores. There are differentiating characteristics between a data warehouse and an operational data store that are important considerations for data management and analytic methods. To summarize the differences, data from operational systems are extracted, transformed, modeled, and stored in the data warehouse. A data warehouse
refers typically to retrospective data, and it maintains both aggregate and detail-level data. Aggregate data are summarized data, whereas detailed data maintain patient level data. Aggregate data are not limited to one patient. Aggregate data may track across time, organizations, across populations, or across other types of variables. (Ryan & Thompson, 2002) The data warehouse centralizes data collection for the intended purpose, provides a common view of data reflecting the enterprise, supports analytic tools, is expandable,
and provides data marts within the infrastmcture (Imhoff et al., 2003). A data mart is a a specific purpose. For example, a data mart might be a subset of data for financial purposes that is isolated from clinical subsection of the data warehouse that stores data for
data that the end user does not need to have access to within the data warehouse. In
contrast, a clinical or operational data store is subject oriented; in the case of the clinical data store, the subject would be the patient. In a financial data store, it might be an account number associated with "the subject." Data are fully integrated across time and events. Data are current and also volatile, meaning they change based on events occurring with the patient at that point in time, and are typically detailed data and not aggregate information (Imhoff et al., 2003). There are some important considerations to remember when designing a data warehouse, and we cover some very basic design components that are important for data analysis and discovery. Databases are defined as a collection of related data organized for efficient storage and modification, and rapid search and retrieval (Kuo, 2017). The design strategies are relevant to "the rapid search and retrieval" requirements for analysis. It is not an ideal use of workforce resources to have an analyst wait for hours or even minutes to return a query or report. When amassing large volumes of data, the way the stmcture of the database or data warehouse is constructed is very relevant. Analysts can be involved in design strategies if they are knowledgeable about
443
444
Hi: DATA MANAGEMENT
what they want from the data and how they intend to use it. The authors advise a build that strategizes a "left to right" build strategy. In other words, the analyst needs to fully understand and operationalize (ability to measure) the outcomes and processes that they want to retrieve from the data store before laying out the design. Frequently, organizations will design a data warehouse by depositing multiple data sources, including admissions, discharge, and transfer (ADT) data, financial data, clinical data, and supply chain data, without thinking about the purpose and how these data will be used. The informed analyst can help with the design strategy by being clear on what they need from the data warehouse. In Chapter 17, we discussed strategies for convening the right people to design the enterprise data. This chapter addresses how an analyst determines what they want from the data and how to think in terms of appropriate data-analysis methods. BASICS OF DATA ANALYSIS
In designing a data-analysis strategy, the analyst wants to determine what it is that is needed for analysis. This strategy might sound like a very basic notion but designing accurate metrics and a data strategy that will populate the measure is often much more difficult than we initially think. Measurement Theory Measurement theory is a fundamental science u.sed to understand data. Krebs (1987), in a classic article on measurement theory, notes: "Measurement theory is the conceptual foundation of all scientific decisions. If the measurements are erroneous, no amount of
statistical or verbal sophistry can right them" (Krebs, 1987, p. 1834). In addition. Waltz et al. (2010) indicate that conceptual frameworks are critical to systematically guiding the measurement process by increasing the likelihood that concepts and variables universally salient to nursing and healthcare practice will be identified and explicated (Waltz et al., 2010).
Measurement theory can be considered the basis for evidence-based clinical practice and is an important consideration before any analysis of healthcare data. It is important to think from left to right—meaning we need to think about what it is we intend to measure or the effect we want to examine collection methods based on that outcome of interest. We
and build systems and dataidentify what those elements
or variables are by using conceptual and theoretical frameworks that many of us as clinicians know intuitively, but we are relying on theory we have learned over time and from the scientific literature to reinforce how we approach designing data-collection and analysis methods. Measurement can be defined as "the assignment of numbers to objects or events according to Riles" (Stevens, 1959, p. 25). The goal of measurement is to accurately evaluate a phenomenon of interest and reduce concepts to operational definitions with numeric values. These numeric values can take on different levels of measurement.
Levels of Measurement
The levels of measurement can be classified into scale, nominal, ordinal, interval, and ratio data. Once classified, the level of measurement specifies which statistical operations can be properly used. These statistical analysis decision points are reflected in Figure 18.1. These levels are defined as follows:
18; DATA MANAGEMENT AND ANALYTICS: THE FOUNDATIONS FOR IMPROVEMENT
Property Level of
measurement
Categories
Ranks
Equal intervals
True zero point
Nominal
- Categorical Ordinal
0
0 0
Interval
0 - Continuous
Ratio
0
0
0
0
FIGURE 18.1
Four Levels
of Measurement.
■ Nominal: Numbers assigned represent an object's membership in one of a set of mutually exclusive, exhaustive, and unorderable categories. ■ Ordinal: Numbers assigned represent an object's membership in one of a set of mutually exclusive and exhaustive categories that can be ordered according to the amount of the attribute possessed. ■ Interval: Numbers assigned represent an object's membership in one of a set of
mutually exclusive and exhaustive categories that can be ordered and are equally spaced regarding the magnitude of the attribute under consideration. ■ Ratio: Same as the interval, but also the distance from an absolute zero point is known.
Nominal and ordinal data can be considered categorical data, whereas interval or scale. Figure 18.1 reflects
and ratio data can be considered a continuous variable
this more simplistic approach toward data analysis. The figure reflects the level of measurement and properties that can help an analyst determine how to approach analyzing variables. Operationalizing a Measure An operational definition outlines precisely how a measure will be constructed. To specify dependent measures or outcome measures and independent variables that might impact the outcome of interest is an important component to improving science, as well as to the fundamentals of research. To fully understand operational definitions, we need to define several terms, including (a) variable, (b) dependent variable, (c) independent variable, (d) confounder factors, (e) outcome measure, and (f) process measure. We start
with defining the term "variable." A variable is defined as a quantity that may assume any of a set of values, such as the gender variable, which is either male or female; we can assign a value of 1 = male and 2 = female (Database, n.d.). A dependent variable in an
analysis or a study is the outcome of interest, such as mortality, total costs of a procedure, or 30-day readmissions. In many healthcare organizations today, "yes" or "no" are examples of variables of interest that can be considered dependent variables. Think in terms of these variables as dependent upon independent variables. An independent variable is a variable that is related to the dependent variable of interest, or it may be an intervention that is manipulated in a research study or improvement process. The intervention would be considered a process measure that impacts the outcome measure. Figure 18.2 depicts a model frequently used in health outcome research and
improvement science that was originally developed by Donabedian (1966). This model proposes relationships among components that are two dimensional, with interventions
445
446
III:
DATA MANAGEMENT
FIGURE 18.2
Examining Odds of Primary Cesarean Delivery Using a Quality Health Outcomes Model. A/ofe;The dashed
arrows represent interaction or effect
modifiers, whereas the
solid arrows depict the confounding effects or effect mediating factors.
Source: McBride, S. (2005). The effect of induction of labor on the odds of cesarean
delivery [Unpublished doctoral dissertation]. Texas Woman's
University, Denton, Texas.
acting through characteristics of the system and of the client, and vice versa. The effect of an intervention in this study (e.g., induction of labor) is either mediated or modified by client and system characteristics. An example of these relationships related to primary cesarean delivery is that the effect of labor induction varies across parity and gestational age {McBride, 2005). The Donabedian (1966) health outcomes model (discussed in Chapter 3) is excellent for framing improvements and healthcare outcomes studies and for thinking through relationships and processes that may influence some outcomes of interest. For example, this model is relevant to examining the effect of inductions on primary cesarean delivery and what factors we believe will influence the relationship of the intervention (induction) with the outcome of interest (primary cesarean delivery). Independent
variables, such as parity, gestation, race/ethnicity, maternal age, medical indication for induction, dystocia, fetal distress, and baby weight, were identified as significant factors related to the induction of labor's influence on whether or not a successful primary
cesarean or a successful vaginal delivery occurs. Confounding factors are situations
or
factors that influence the outcome of interest that a researcher or analyst must control for when examining theimpact of an intervention on some outcome of interest (Hosmer
& Lemeshow, 2000). In the case of examining the impact on primary cesarean rates in a hospital, we would want to control for the independent variables noted as independent
variables. However, how will we define and measure the odds of a primary cesarean and the influence of induction of labor as "yes" or as "no"? By clearly defining
what data will be used to measure these three factors, we are "operationalizing" the definitions. This discussion was not intended to be a full discussion on research and
improvement science. Therefore, we refer the reader to a research text for full guidelines on
research design and strategy. However, we are emphasizing that defining measures
18: DATA MANAGEMENT AND ANALYTICS: THE FOUNDATIONS FOR IMPROVEMENT
and operationalizing precisely how you will measure variables in any analysis are fundamental competencies.
Conceptual models are visual diagrams, such as the health outcomes model noted in Figure 18.2, that are particularly helpful in clarifying how and what one will be analyzing. It is important to involve all stakeholders from interprofession al teams who fully understand the clinical domain one is examining. When approaching and operationalizing a measure, a relatively new technique is to convene the interprofessional team and map out the thought processes related to how the measure will be constructed using a mind map. Figure 18.3 presents an example of a mind map. A mind map is a visual depiction of some phenomenon of interest. In this case, it is a map of factors influencing an outcome of interest. Figure 18.3 reflects a map conceptualizing a mea.sure for a catheter-associated urinary tract infection. In addition to conceptually mapping the process, more detail is needed to extract the data from the electronic environment for analyzing this outcome of interest. Figure 18.4 reflects the more detailed workflow map of the interprofessional departments that influence the outcomes of interest and have some impact on the actual data as they flow through the electronic environment. FIGURE 18.3 Mind Map for Operational Definition of a Measure for Catheter-Associated Urinary Tract Infection. catheter days
HAC, hospital acquired condition. Source; McBride, S., Fenton, S„ Valdes, M„ & Gilder, R. (2013). Transforming digital data into useful information. TNA-TONE Health IT Committee Educational Series. http://www.texa snurses.org/?page=HITWebinars2013,
Microsoft programs Excel and Word, along with several off-the-shelf applications, do a nice job of providing point-and-click tools with SmartArt features that support the construction of a mind map. Figure 18.3 presents an example of a measure mapped using a nice feature functionality that helps create a mind map. Microsoft Excel, Word, and PowerPoint can be used to create conceptual models or "mind maps" of data elements and relationships to measures using SmartArt or other chart features. Figure 18.4 uses Microsoft Visio to note workflow and roles related
to the mea.sure. The "traffic jam"
notation in the figure presents the convergence of data into information to operationally define the numerator and denominator for the measure.
Quality of the Data
Before using data, it is important to determine the quality of the data being used. Quality metrics for data are often a consideration for organizations, and these metrics might
447
CO
FIGURE 18.4 Transforming data Into Useful Information and Knowledge.
Q >
Transforming data (process noise) into useful information (stgnai) - CAUTi exampie of chalienges to signal propagation
> >
-»Network of temporal event data (process noise) sequence & flow of information (slgnat)-» 1 Catheter is placed - CAUTI protocol Is initiated
r
j >●
is in CAUn
denominator
FALSE
TRUE
Monitor for
change in CAl/Tl
^
H
CAUTI metric repordng,
CAUTI numerator and denominator
archive, codification
distribution, and messaging
metric data
L denominator status J
> o
V
TRAFFIC JAM ! Culture & sensrtMty positive
\
lessaging to physicians, nurses,
——_WUU megiQ_—
Lab repot
CPOE forandmicrobi^
Order cuttore &
^.Mnsitivity
4
Assess lab results, order antimicrobial, order cath
change or remove, monitor CAUTI messaging
Orders ot pharmacy
-.....^and nursing ^
Monitor forun
un suspected: nob^ physician, send lab, give ^ antibiotic, change or remove cath
4
^ Administer antimicrobial, change or remove
L catheter, continue to monitor patient and
CPOE messaging from CAUTI metric and EHR
7 J
A/ofe; Traffic jam notation presents the convergence of information into the operational definition of the numerator and denominator. CAUTI, catheter-associated urinary tract infection; CPOE, computerized provider order entry; EHR, electronic health record; UTl, urinary tract infection. Source: McBride, S., Fenton, S., Valdes, M., & Gilder, R. (2013). Transforming digital data into useful information. TNA-TONE Health IT Committee Educational Series. http://www.texasnurses.org/?page=HITWebinars 2013.
18: DATA MANAGEMENT AND ANALYTICS: THE FOUNDATIONS FOR IMPROVEMENT
include factors such as the accuracy of the data meaning the data are as free from error as possible; the consistency of the information across the institution; the timeliness of the data indicating that the data are as current as possible; completeness of the data with all of the values captured; conformity to standards; and the overall integrity of the information. Let us take one of these types of data-integrity issues and walk through a process for identifying the issue. A duplicate records issue can create challenges involved in delivering patient care and in examining the data for outcomes analysis. For example, a cardiac patient presents to the emergency department with a myocardial infarction. When the admitting clerk searches the clinical data store within the EHR using a common lookup feature to search for patient records, the clerk notes two records that appear to be of virtually the same individual but with different birth dates. Proper patient identification is a common problem within the EHR and is often an expensive issue to rectify within the clinical data store. The duplicate record must be merged into one master record. When performing data analysis, duplicate records are a common source of error; when different data sources or multiple extracts from the same data source are used, duplication of the cases can be introduced into an analysis file. Let us take the same example and consider that data have been extracted, transformed, and loaded into a data warehouse for the analysis of cardiac outcomes. Aclinical analyst is analyzing retrospective outcomes data for patients with myocardial infarction, tracking and trending outcomes for the facility. If there are duplicate medical records for the same individual within the clinical data, the data will inflate the overall denominator of total cardiac patients within the analytic files. Therefore, one of the first things an analyst must do before analyzing data is to inspect the data for data integrity on all factors noted earlier.
Exploring a Data Set In addition to examining data for the integrity of the information, an analyst should examine the data and explore the information before analysis. Open the data file and visually inspect it. ■ What rows and columns do the data set reflect?
■ How are the data structured? Are visibly missing data apparent in the file?
■ Do the data appear to be sorted in some order? ■ What variables in the data set represent dependent, independent, and grouping variables?
If the software you are using has a feature to generate a data dictionary of file information, start with running the report that will provide you variable information. Include in the report: position in the file, data labels on variables, measurement level (nominal, ordinal, scale, or string), column width, and variable labels, also referred to as "value sets." Examine all of these features in the data set. Figure 18.5 includes an example of an IBM SPSS software application and a file information display feature that is generated in the software. It was obtained by selecting File/Display Data File Information/Working File. Exploratory Data Analysis
Exploration of data can help determine whether data are accurate and complete, distributions of the data, and what statistical techniques mightbe appropriate for analyzing the data and examining initial relationships that might exist among variables. A data exploration analysis provides a variety of visual and numerical summaries of data and can be performed by all cases, a subset of cases, or separately for groups of cases. Grouping
449
450
I: DATA MANAGEMENT
FIGURE 18.5 Display File Information Feature in IBM SPSS. fs
o
'OulpuH [0oojm«ml| - ISM SPSS SWIstKS Vicw»f
FM
Qat
TlifKiwtn
yistrl
Fom«
ir-
DkMgtnMng
firaplH
UHn
U»2u
nyidsw
Htu
-Ti
ilB — B fi Ouvul Qdifio
4 Pile Infonnation
B S) Filfl lnfocPTUU»n td«C4S«tl] Cs\U9«rsVsti4Cse\DscpiKx\K«»t
tw
M
eRar(A
133710 HedcaiePsrtA
i59,
IS.
Dala Analysis lonU Tools lei Imancial and soenMic data
anaiyiK.
Q FUNCfttS.XLAM
33711 NueCioss/bVrSNeld
VS7.B9
OIHPEHA8lllTAnC«
33712 MeckaroRaitA
^.9
CVStmSUMSPEC
,3^13 Me4c«ePailA
^37.0
FOODAPMIT Pt«U:-K>Nltl5
33714 Mediate Pait A
«7.69
OTH REHABILITATION
$50,192
33715 HedctfePaitA
^.0
SIREPTOCOCCAl SEPTICEMIA
$11,696
VS7.89
OTHPEHABILITATICN
^.89
HAIJIC DEPRESSIVE OTH
OBSTRUCT CHRONIC 8PC«OSITlS W^XAC
33719 Ke4c«eP«tA
'S9I.31 ^.0
33720 KeOc«eP«tA
♦9I.21
OBSTRUCT CHRONIC BftOIXHlIIS W/EXAC
■33721 Elue CioosSBIue SKdd
VS7.89
OTHREHA8ILITATION
33722 MeOewS
*^.30
RECLRR MAJOR DEPRESS!^ UNSPEC
33716 Meeicare Part A ^33717 EXe Cioss/Kue STaeld ●33718 Ekse Cross/esue SSeld
33723
$I6.2» $9,S02
$9,569
$32,945 $7,549 $10.W7 $9,454
UfilN TRACT INFECTION UNSfeC
$7,773 $30,9SI $5,827
^.04
OTKR SKlNt.$lj&3I
$54,669
V57,89
OTH REHASILITATIC'N
'31Z.9
CONDUCT DISTLSBANCE UTSPEC
^.6
ORGANIC BRAIN SYND OTH
$13,662
33726 hie«cveP«tA
VS7.69
OTH REMABIIITATION
$63,419
33727 Me4c«e Part A
^.0
FOCXi/VOMir M4CWONIT1S
UC7 AO
miH BFUARII ITATinU
MedcarePailA
33724 l-SeOeare Pait A 33725 MedcarePailA
W9FT 14
4
r
Ready
H
n« ru » Oi»IH
lUDSOCt: CPU,STY
ti _
PiessFlTor add in help.
15,918
$4,607
$12,592 tS4 tM
0< Avtrage:$AS00
Count 2
]>
Sum; $9.000 liSaSQ 1011X0-
m
460
111: DATA MANAGEMENT
Business Intelligence Tools Business Intelligence (BI) tools are a set of software applications that typically help inform one's business as to profitability, quality, risk, and other factors that indicate how well one is doing within any given industry. Often, these tools allow multidimensional
analysis that is aimed at hitting "the sweet spot" in one's data. A multidimensional business analyst thinks in terms of when, who, what, where, and result. Figure 18.9 reflects this dimensional thinking and provides examples of what might constitute a BI multidimensional strategy for healthcare. In this example, the "when" is an ability to report the year, quarter, or month. The "who" is an ability to stratify or filter one's report on dimensions such as hospital system, hospital, ora specific provider. The "what" would be a similar ftmctionality allowing an organization to filter or stratify using different specialty services. The "where" feature allows one to look at data and reports by patient's county, city, or state. Finally, the "what" is tlie outcome measures the organization seeks to determine that truly get to the "sweet spot" in the data. In this example, authors note frequently reported measures of mortality, length of stay, and total charges. The sweet spot in these types of analytic tools provides the organization with information that will help inform quality and cost of care. These types of tools are excellent for driving improvement and mining data to determine where an organization is doing well and where it might need to improve. The tools are often easy to track, trend, filter, drag, and drop by point and click of a mouse to quickly mine the data for the information needed to inform the organization. FIGURE 18.9 Setting Up Multidimensional Analyses With BI Tools.
●Year ● Quarter ● Month
● Hospital system ● Hospital ● Provider
● Cardiac
● Mortality
● Obstetrics
● County ● City
* Length
● Ortho
● State
● Total
ofs^y charges
BI, business intelligence.
Data Visualization
Many software analytic applications provide means of visualizing data such that the data "tell a story." Dating back to Florence Nightingale, data visualization in healthcare is recognized as a powerful manner in which data can inform improvement strategies. Richardson et al. (2021) profiled the leaders for data visualization, defined by Marr (2017) as follows: "Data visualization is about how to pre.sent your data, to the right people, at the right time, in order to enable them to gain insights most effectively." The 2021 Magic Quaciran tfor Analytics and Business Intelligence Platforms done by Gartner notes the following vendors as top contenders: Tableau, Qlik, and Micro.soft. We will provide screen shots as examples of how these types of tools can be u.sed in healthcare. First, Tableau is reported as a top contender because it offers a powerful suite of data preparation, management, and visualization tools. Tableau Prep allows the analyst to clean, combine, transform, and prepare large data sets to make them easier to analyze (Allchin, 2020). Tableau Desktop can handle Big Data operations and integrates well with advanced data solutions such as Hadoop, My SQL, and Teradata—often seen in the healthcare industry (Marr, 2017). Apache Hadoop is an open-source software that manages and handles large data sets across clusters of computers (Apache Hadoop, n.d.). Open-source software is a computer software that is distributed with its source code and ohen managed by volunteer efforts
18: DATA MANAGEMENT AND ANALYTICS; THE FOUNDATIONS FOR IMPROVEMENT
of programmers sharing code to fix bugs, improve functions, or customize the software for the intended use (Data Dictionary, n.d.). Big Data and advanced analytics will be covered further in Chapter 27. Tableau Public offers a version of Tableau Desktop for free where users can connect to publicly available data sets and create visualizations and save
them to the web. Figure 18.10 represents a dashboard visualizing data within the Tableau
FIGURE 18.10 Illustration of a Typical Tableau Dashboard.
® STATES iBI I I
fORlOROHAVIRUS
1£■>
»sofS'»7taiMifi
■
166.2M people have received the vaccine
365.8M
62.1%
51.1%
40.6%
doses have seen
of the po^ulat on ages 16*
distributed across the US
received 1 or mere dosei
of the popiilat>cr) ages 18* are f jlly vaccinated
ofallpcoplc ntnoUSaic ftHy vaccinated
TOP58TAT8SPCREACHMETRIC I SElECTeuTTOfiTOViSUAlIZs
distributed per
100k people
)
Dntr
● More
2 HowTo
FiftefDala
Explore Market Data ■
y CureoCSeleclNns
Lines Selected
Current Selections
2013
20UQI
2012Q4 2013Q1
:2012Q2
●2013Q4
Case Level Data-
AHRQ Quarterly
AHRQ Yearly
^
I
^ - ■ NOT 2011Q1. 2011Q2, 2013Q-I HespitiShorlNeme^ ' ■ 18 of 1093 DsdrpQir
448
2011Q2 2011Q3 '2011Q-I 2012Qi
Report Builder
>
Market Capture DashPoard o
> o
Click lor help >»
FOUNDATION
2012
PivolTaoie*
Mirti Charts
DFWHCl^
2011
DFWHC vs State ●
RroducOjneDesc
^
PtType
.^*●1 ^ ■ -127.5 - CARDIAC ARREST
[>e9CodeOesc
● Cardologv
COUNTY
Payer Groups 186
TARRANT
Medcare
2.05% ●IS.7S
Unrated
11.6%
71
DENTON
38
COLUN
28
GRAYSON
26 LOS Group 12
(Blank)
3-SOavs
9
6-3 Days
,
I
1-2 Days
12
PWRKER
2013Q3
37.7H
t^edcakl
DALLAS
JOHNSON
2113Q2
O
Ir^ued
I
9-HDa>s
APR Risk Of MortaBty
.IrijCut
Avg LOS DF.-.VtC Ho^
'E
ED Case
Srxpery Case
O
ExWrr*
Total Chrg
IE
TotaK
IE
Trend
Trcitd Stacked
E
Major
a
Moderate
Cases for Inpatient for NOT 201 IQ 1, 2011Q2, 2013Q4 Quarters of Data
' iHM Health Arlirrgton Menonal Ho$p
^Rlf Health Hams Methodst HospiU
70
Tn^ Health Hams Methodst Ho»lU Tbtas Health Harris Methodst HosprU
line Health Hams Methodst Hosprb
ss Health Presbvtr»ian Ho9«ital-0« QMUHealth Presbyterian Hospitai-Ki
Health Heart & Vasoiar Hospitj
Ikxas Heahh Presbyterian Hosprtal-Pl
HMVHealth Presbytenari HoiiiibhRi Ibcas Health Presbyterian Hospit^ - '●
fhcas Health Presbyterian Hospital Al
In/Out
Abfcne Reoonal Medcal Center An^ton Danbury Medcal Center
Sapbst Hospitals of Southeast Texas Bac&st Medcal Center
62.5%
Major
29.5%
Mnor
1.8% 6.3%
Moderate
Ungroipable 60
/
AgeGroup
\
/
50
●18-+1 Years
\
■45-64 Years ■65-74 Yews
40
■75+Years
30
20-24 Years 25-29 Years
20
30-34 Years
IMlBHealth Presbyterian Hospital Dr
niw Health Presbviertan Ho^rtal Fk
Un^oupable
Extreme
HlKHHeaJth Hams Methodst Ho^b
21.7% 6.5%
APR Severity
Heahh Hams Methodist HEB 80
69.-IH
2.5%
Trend %> Share
BC'x
Trend
I
Hospital name
Pc Graph
AvgTotchrg
35-39 Yews
to
40-14 Yews 45-49 Years
0
O'
■V^
s>
sy
.'nP
'p,o
'p
.O'
,o
ft ■T
o
I
50-54 Years
'f'
55-59 Years
DschrgQIr
' Indicales OrugiHFV
61, business intelligence. Data Source: Texas Hospital Inpatient Discharge Public Use Data File, [Q1 2011 to Q42013]. Texas Department of State Health Services, Center for Health Statistics, Austin, Texas. [2015]
Source; The Qlikview Dashboard provided by DFWHC Foundation, used with permission.
H
18: DATA MANAGEMENT AND ANALYTICS: THE FOUNDATIONS FOR IMPROVEMENT
Another common feature of Power BI is how it is used
by the data analytic industry.
For example, consulting firms such as SNP Technologies, Inc. (2017) and Withum Digital (Tate, 2017) become "Microsoft Partners" and use products such as Microsoft Power BI
to build data infrastructures for companies without the expertise or the desire to do it themselves. At the same time, the product is user friendly enough to be used by most data analytics professionals. In one case, a health informatics student who had just learned how to use Power BI created a Power BI interactive dashboard for the providers in a psychiatric clinic.^ This Power BI interactive dashboard helped them to understand where they stood regarding outcomes of interest for their patients. These outcomes were known as industry standard
depression scores, anxiety scores, medication compliance, collaboration with initial referring physician/practitioner, and other physical parameters such as blood pressure and body mass index (see Figure 18.12 for a partial list). This interactive dashboard allowed the provider to "drill down" and explore more detail that comprised a given metric. It also allowed the chief psychiatrist to identify providers that might benefit from additional guidance and support to improve the patient outcomes. FIGURE 18.12 Partial List of Metrics (and Sample Values) That Were Set Up to Automatically Import Into Power BI. MEASURE FROM EHR
VALUE EXAMPLE
Blood pressure Diastolic blood pressure
90/52 mmHg 52 mmHg
GAD-7 score
15
Heart rate
70/min
PHQ-9 score
20
Respiratory rate
16 breaths per min
Sheehan
14
Systolic blood pressure
88 mmHg 155 lb
BI, business intelligence; EHR, electronic health record.
Once the list is identified, it is exported from the EHR and then imported into Power BI (see Figure 18.13). From there the provider sees the overall dashboard and can drill down to further
explore potential areas for improvement (see Figure 18.14).
Key features and benefits of Microsoft Power BI include a feature that improves the productivity of the analytics team. Ideally, BI tools that can integrate with data architecture exist in the cloud and provide real-time interactive visualizations/ dashboard. Specifically, regarding Power BI, four of the most useful advantages for Power BI have been reported (Tate, 2017). Advantage #1 —More Accessible Data Than Ever Before
Power BI was built to integrate with Microsoft technology, such as Azure, a cloud computing service created by Microsoft where organizations pay for access to a massive pool of computing resources, SharePoint documents, or SQL Server databases. However, it enhances non-Microsoft solutions, too. In fact. Power BI currently connects to hundreds 'The irnrh of Micluiel Etzel, Texas Woman's University 6S in Health Informatics, Clinical Application Minor, is acknowledged for these examples of Power BI.
465
466
IM: DATA MANAGEMENT
FIGURE 18.13 Raw Data Were Exported From the EHR Into Power B! in Preparation for Dashboard Creation. ID PATIENT
ID
COUNTY
PROVIDER
PT
ZIP PT
MONTHS
ASSESS
ASSESS
IN
1
2
PRACTICE
SCORE
SCORE
ASSESS 3 SCORE
1001
101
Dallas
75063
3
7
13
25
1002
101
Dallas
75063
4
5
10
23
1003
101
Dallas
75063
5
8
12
26
1004
101
Dallas
75063
2
8
12
27
1005
101
Dallas
75063
4
9
13
27
1006
101
Dallas
75063
2
8
13
27
1013
102
Dallas
75063
4
7
13
25
1015
103
Dallas
75063
8
7
13
27
1016
103
Dallas
75063
7
6
15
28
1017
103
Dallas
75063
7
6
16
29
1018
103
Dallas
75063
8
5
16
30
1019
103
Dallas
75063
7
8
15
30
1020
104
Dallas
75063
8
8
14
31
Bl, business intelligence; EHR, electronic health record.
Median of assess 3 score and
measure by provider Provider 101
FIGURE 18.14 Drill-Down of Provider 101 for Assessment
3 Score (Depression).
patient assessment
metrics
The median score is "29"
for depression, and this practitioner"101" has a median score that is on the better side of that median.
57
of common software solutions and pulls data into a centralized, easy-to-digest dashboard. Here are just a few: ■ Spark ■ Hadoop ■ Google Analytics ■ SAP
■ Salesforce
■ MailChimp (Tate, 2017) Advantage #2 —Ease of Implementation
Very little engineering or IT resources are needed to implement Power BI. In fact, some instances do not require any engineering. Managers simply need to create an application interface key and plug it into the software (Tate, 2017).
18: DATA MANAGEMENT AND ANALYTICS: THE FOUNDATIONS FOR IMPROVEMENT
FIGURE 18.15 A Complete List of Variables for Total Charges. TOTAL CHARGES Mean
27,424.85594
Standard error
569.5811108
Median
19,500,92
Mode
1,531.6
Standard deviation
28,261.78593
Sample variance
798,728,544.1
Kurtosis
27,59007693
Skewness
4,193621611
Range
380,457.76
Minimum
0
Maximum
380,457.76
Sum
67,519,995.32
Count
2,462
Source: Microsoft Corp. (2018). Load the analysis ToolPak in Excel, https://support.office.com/en-us/article/ Ioad-the-analysis-toolpak-ln-excel-6a63e598-cd6d-42e3-9317-6b40b a1a66b4; Tableau, (n.d.). l/Vebinar series
achieving operational excellence in healthcare, https://www.tabl eau.com/learn/series/achieving-operational -excellence-healthcare.
Advantage #3 —Robust Access Control and Security
Power BI sets up access control through Active Directory (AD), the same control panel your organization uses for other Microsoft solutions. What makes Power BI different is row-level security, which allows users to grant and rescind access on a very controlled level (Tate, 2017). Advantage #4—A Simple Learning Curve
The fourth Power BI advantage is its simple learning curve. Everyone uses Micro.soft
products, so the ribbons and other user interface elements will be instantly familiar. This means basic users can explore simple Power BI services right away, and advanced users can jump right into exploring advanced data modeling. The interface also allows users to easily export data to other .systems (like Excel), which gives users the flexibility to work with their data in other environments if they choose (Tate, 2017). CASE STUDY
Describe How This Situation May or May Not Be Present in Your
Organization and Why
A hospital has just been publicly reported as using its state regulatory hospital discharge data.The data have been risk adjusted to control for comorbidities with which the patient may have entered the hospital to effectively "level the playing field" for comparative reporting of hospitals. When your hospital's report comes, you note that you risk adjusted poorly for stroke mortality. Your hospital has just received stroke center designation and you feel confident that best practice protocols with stroke management are being followed. The analyst tracks and
467
468
III:
DATA MANAGEMENT
trends your mortality data on stroke and you note a spike in mortalities for a given quarter, which appears to be why your state public report is unsatisfactory. Drilling into the data to examine the numerators (deaths) due to stroke (denominator and total patient population), your analyst determines that the risk of mortality has a score of 1 to 4, with 1 being a minor risk of mortality and 4 being severe. All the deaths that occurred as a result of stroke had a minor risk of mortality. Clearly, this is why the hospital did not risk adjust well, as these patients did not appear to have any risks associated with severe illness that would reasonably explain the mortalities. Next, the analyst drills further into the detailed subject-oriented data of actual patient records, followed by a full chart review to
obtain the "sweet spot" that would inform improvement. Consider the following questions:
1. Would the analyst be examining aggregate data, detailed data, or both to investigate this quality issue?
2. Would the analyst likely use a retrospective data warehouse, clinical data store, or both to investigate the mortality rate?
3. What type of tools or analytic approaches do you believe this analyst might use?
SUMMARY
Data analysis, an important component for understanding today's complex healthcare delivery system, affects provider, vendor/supplier, and payer/insurers. Given that the provider's data analysis directly affects patient care delivery, we have focused on the critical aspects for consideration. As noted in the previously described Nursing Education for Health Care Informatics (NEHI) model (McBride, Tietze, & Fenton,
2013), data management and analytics, linked with patient safety, quality, and pointof-care technology, create the culminating process through which optimum healthcare improvement may occur. This chapter provides the specifics for supporting analytic skills development.
END-OF-CHAPTER
RESOURCES
EXERCISES AND QUESTIONS FOR CONSIDERATION
Figure 18.15 (see text) comes from an existing data set of patients with type 2 diabetes who were hospitalized in 2012. The StatPak feature was used to create the associated descriptives. Using the chart, please respond to the following questions: 1. What is the total count of case records?
2. What is the standard deviation of the total charge for hospitalization? 3. What is the median total charge? 4. What is the mode for the total charge?
18: DATA MANAGEMENT AND ANALYTICS: THE FOUNDATIONS FOR IMPROVEMENT
5. What is the overall sum for total charges? 6. What is the maximum total charge? ADDITIONAL RESOURCES
SMillCt* FUIUSHIN-.
CONNECT
_
A robust set of instructor resources designed to supplement this text is located at http://connect.springerpub.com/content/book/978-0-8261-8526 -6. Qualifying instructors may request access by emailing textbook@springerpub. com.
REFERENCES
Adams, J,, & Carets, D. (2014). The healthcare analytics evolution: Moving from descriptive to predictive
to prescriptive. In R. A. Gensinger (Ed.), Analytics in healthcare: An introduction (pp. 13-20). Health
Information Management Systems Society. Allchin, C. (2020). Tableau prep: Up & running. O'Reilly Media.
Apache Hadoop. (n.d.). What is Apache Hadoop? http; / /hadoop.apache.org/index.html Carey, R. G. (2003). Improi’ing healthcare with control charts: Basic and advanced SPC methods and case studies. ASQ Quality Press. Centers for Disease Control and Prevention. (2015). International Classification of Diseases, tenth reinsion,
clinical modification (ICD-IO-CM). http;//www.cdc.gov/nchs/icd/i cdlOcm.htm Classification of Diseases (ICD-IO-CM/ PCS) transition: Frequently asked questions, http;//www.cdc.gov/ nchs/icd/icdlOcm_pcs_faq.htm Database, (n.d.). Merriam-Webster's online dictionary. Retrieved August 30, 2021, from http://vvww .merriam-webster.com/dictionary/database Data Dictionary, (n.d.). Open source. Random House, Inc. Retrieved August 30, 2021, from http: / / www .dictionary.com/browse / open-source Davenport, T. H. (2005). Competing on analytics (No. R0601H). Harvard Business School. Donabedian, A. (1966). Evaluating the quality of medical care. Milbank Memorial Fund Quarterly, 44, 166-206. https://doi.oig/10.2307/3348969 Englebardt, S. P, & Nelson, R. (2002). Healthcare informatics: An interdisciplinary approach. Gale, Cengage. Evergreen, S. D. (2017). Presenting data effectively: Communicating your findings for maximum impact. Sage Centers for Medicare & Medicaid Services. (2014). International
Publications.
Paul, F., Erdfelder, L., Lang, A. G. & Buchner, A. (2015). G*Power: Statistical power analyses for windows and mac. http;/ /www.gpower.hhu.de/en.html Hosmer, D., & Lemeshow, S. (2000). Applied logistic regression (2nd ed.). John Wiley & Sons. Imhoff, C., Galemmo, N., & Geiger, J. (2003). Mastering data warehouse design: Relational and dimensional techniques. John Wiley & Sons. Krebs, D. E. (1987). Measurement theory. Physical Therapy, 67(12), 1834-1839. https; / /doi.org/10.1093/ ptj/67.12.1834 Kuo, A. M. H. (2017). Chapter 7; Database, data warehousing, clouds for storage & data mining for healthcare. In K. A. McCormick, B. Gugerty, & J. Mattison (Eds.), Healthcare information technology exam guide for CHTS and CAHIMS certifications (2nd ed., pp. 133-167). McGraw-Hill Education. Langley, G. J., Nolan, K. M., Nolan, T. W., Norman, L., & Provost, L. P. (2009). The improvement guide: A practical approach to enhancing organizational performance (2nd ed.). Jossey-Bass. Marr, B. (2017). The 7 best data visualization tools in 2017. Forbes, https://www.forbes.com/sites/ bernardmarr / 20171071201 the-7-best-data-visual ization-tools-i n-2017/ # 2c016b7c6c30
McBride, S. (2005). The effect of induction of labor on the odds of cesarean delivery [Unpublished doctoral dissertation]. Texas Woman's University, Denton, TX. McBride, S., Fenton, S., Valdes, M., & Gilder, R. (2013). Transforming digital data into useful information. TNA-TONE Health IT Committee Educational Series, https://www.tex asnurses.org/page/ HIT#tab7
McBride, S. G., Gilder, R-, Davis, R., & Fenton, S. (2006). Data mapping./ournu/o/f/ie Afwmcfln Hefl/t/i In/onMflfion
Management Association, 77(2), 44-48. http://library.ahima.org/d oc?oid=65895#.WzPdUPVrzcs
McBride, S. G., Tietze, M., & Fenton, M. .V (2013). Developing an applied informatics course for
a doctor of nursing practice program. Nurse Educator, 38(1), 37-42. https://doi.org/10.1097/ NNE.0b013e318276df5d
Microsoft Corp. (2018). Load the analysis ToolPak in Excel, https://support.office.com/en-us/article/ Ioad-the-analysis-toolpak-in-excel-6a63e598-cd6d-42e3-9317-6b40bala66b4
469
470
III: DATA MANAGEMENT
Munro, B. H. (2005). Statistical methods for health care research (5th ed.). Lippincott Williams & Wilkins. Pallant J- (2010). SPSS suridval manual: A step by step guide to data analysis using SPSS (4th ed.). Open University Press/McGraw-Hill. R Foundation. (2014). The R project for statistical computing: Getting started, https: / / www.r-project.org Richardson, }■, Schlegel, K., Sallam, R., Kronz, A, & Sun, J. (2021) Magic quadrant for analytics and business intelligence platforms, https: / /www.gartner.com/docum ent/3996944?toggIe=l&refval= 289152310&ref=solrResearch
Ryan, S. A., &: Thompson, C. B. (2002). The use of aggregate data for measuring practice improvement. Seminars for Nurse Management, 20(2), 90-94. SNP Technologies. (2017). 20 advantages of Microsoft power Bl. https://www.snp.com/blog/10 -advantages-microsoft-power-bi Stevens, S. S. (1959). Measurement, psychophysics and utility. In C. W. Churchman & .P Ratoosh (Eds.), Measurement: Definitions and theories, (pp. 18-63). John Wiley & Sons. Tableau, (n.d.). Webinar series achieving operational excellence in healthcare, https://www.tableau.com/ learn/series/achieving-operational-excellence-healthcare
Tate, J. (2017). Top 4 adz'antages of power BI. http://digital.w ithum.com/blog/top-4-advantages-of -power-bi Troyansky, O., Gibson, X, & Leichtweis, C. (2015). QlikView your business: An expert guide to business discovery zoith QlikViexo and Qlik Sense. John Wiley & Sons. Tufte, E. R, (2001). The znsual display of quantitative information (2nd ed.). Graphics Press.
Variable, (n.d.). Merriam-Webster's online dictionary, http://www.merriam-webster.com/dictionary/ variable
Waltz, C. F., Strickland, O. L., & Lenz, E. R. (2010). Measurement in nursing and health research (4th ed.).
Springer Publishing.
Clinical Decisio JONI S. PADDEN, DWAYNE HOELSCHER, SUSAN MCBRIDE, AND MARI TIETZE
'■■Vv
OBJECTIVES ●
●
Define clinical decision support (CDS) programs, the Five Rights of CDS as they relate to supporting clinical practice, and how CDS programs utilize an evolving array of tools to improve the user experienceand encourageevidence-basedcare by frontline clinicians. Define predictive analytics and prescriptive analytics and discuss how they fit into advanced CDS methodology.
●
Discuss the potential for CDS tools to either solve or exacerbate problems.
●
Explain how a structured CDS methodology for either a single organization or any shared initiative, such as promoting interoperability programs (PIP) among diverse organizations, can improve outcomes.
●
Discuss the importance of collaborative stakeholderteams to improve outcomes, compliance, and sustainability of CDS interventions.
●
Describe and apply a structured methodology for using CDS interventions to improve outcomes examining the American Heart Association Million Hearts® campaign for improving cardiovascular (CV) health outcomes.
●
Examine case studies that provide strategies for bringing specific CDS performance improvement into organizations with a systematic and structured approach. CONTENTS
INTRODUCTION
472
THE BASICS OF CLINICAL DECISION SUPPORT
473
Clinical Decision Support: Definitions, Goals, and Objectives
475
Clinical Decision Support; Importance of Human Factors Design Principles
477
STRATEGIES FOR IMPLEMENTING A SUCCESSFUL CLINICAL DECISION SUPPORT PROGRAM
478
Characteristics and Elements of a Successful Clinical Frameworks for Success
Decision Support Team
478
480
TOOLS ANDTYPES OF CLINICAL DECISION SUPPORT
481
USE OF CLINICAL DECISION SUPPORTTO ALIGN IMPROVEMENT INITIATIVES
482
The contributions of Maxine Ketcham, Tanna Nelson, and Mike Ecklwrd to this chapter in the previous editions of this book
are acknowledged.
472
III: DATA MANAGEMENT
Clinical Decision Support Oversight Committee
483
Deployment of Clinical Decision Support Interventions Measuring the Success of the Program
483
484
MEANINGFUL USE AND CLINICAL DECISION SUPPORT PROGRAMS
485
Improving Population Health Using Healthcare InformationTechnoio gy and Clinical Decision Support Strategies 485 Clinical Decision Support Intervention Workflow Redesign
488
488
Education andTraining to "Hardwire" Improvements Evaluation and Monitoring for Success
489
490
CHALLENGES AND ISSUES WITH CLINICAL DECISION SUPPORT
Challenges to Implementing Clinical Decision Support
490
Features of Safe Healthcare InformationTechnoiogy to Address Challenges Legal Implications
490
491
491
CASE STUDY 1: OBSTETRICAL SCREENING CASE STUDY 2:THINK SEPSIS
492
494
CASE STUDY 3: NORMAL NEWBORN
495
CASE STUDY 4: COVID-19 AND INFECTIOUS DISEASE SCREENING/INTERVEN TION SUPPORTED BY CLINICAL DECISION SUPPORT SUMMARY
497
499
EXERCISES AND QUESTIONS FOR CONSIDERATION REFERENCES
499
499
INTRODUCTION The Office of the National Coordinator for Health Information
Technology (ONC) defines clinical decision support (CDS) as a component that "provides clinicians, staff,
patients, or other individuals with knowledge and person-specifi c information, intelligently filtered or presented at appropriate times, to enhance health and healthcare (ONC, 2018, p. 1). CDS encompasses a variety of tools to enhance decision-making
in the clinical workflow. These tools include computerized alerts and reminders to care providers and patients; clinical guidelines; condition-spec ific order sets; focused patient data reports and summaries; documentation templates; diagnostic support, and contextually relevant reference information, among other tools" (n.d., para. 1). In addition, CDS is a set of tools within the electronic health record (EHR) to encourage the healthcare team to do the right thing at the right time with correct interventions within the clinical workflow. This seemingly simple concept is often a complicated setup in the EHR to support the overwhelmed clinician with an efficient workflow. To do that, one must set up a process of CDS to strategically design a program using the EHR as a tool to enhance patient safety, quality, and population health. For these reasons, CDS is a key strategy within the federal Health Information Technology for Economic and Clinical Health (HITECH) Act, which is used for attaining a promoting interoperability program (PIP) of EHRs.
CDS is an effective tool at improving outcomes in many areas, including improving adherence to clinical initiatives such as deep vein thrombosis (DVT) prophylaxis, cardiac mortality prevention strategies outlined in the Million Hearts campaign, and areas
19: CLINICAL DECISION SUPPORT SYSTEMS
associated with the regulatory reporting requirements of quality measures. We examine some of these use cases in light of the methods described and discuss the use of CDS
within the PIP guidelines and the reasons for this emphasis within the federal regulations. Also, we discuss challenges and issues that arise with the inappropriate use of CDS in organizations. CDS is a powerful tool; however, without design strategies, CDS can result in misuse, creating potential patient safety and legal implications for organizations. We
discuss design strategies to address these challenges through a strategic approach to CDS deployment within organizations and to adhere to best practices to improve interventions and patient outcomes that will help mitigate these issues. Finally, with the help of case studies, we demonstrate the use of these methods in clinical examples, which include aligning CDS with patient-centered outcomes research. THE BASICS OF CLINICAL DECISION SUPPORT
CDS tools existed before the development of EHRs under the national program to certify EHRs under the PIP. Historical examples include practice guidelines carried in clinicians' pockets, patient cards used by providers to track a patient's treatments, and tables of important medical knowledge (Clinfowiki.org, 2015). The Oregon Health & Science University (OHSU) houses the Clinical Informatics Wiki (a.k.a. ClinfoWiki), a website devoted to topics in biomedical informatics. Many of these CDS tools continue to be relevant to the electronic age of healthcare, but they do so by integrating CDS within the EHRs, presenting an opportunity for the various types of decision support to be immediately available within the workflow at the correct time in the clinical decision making process. CDS can be more relevant and accurate, can facilitate, and can be integrated into the clinical workflow when designed well and deployed effectively. It is this innovative use of technology that increases the magnitude of CDS impact on patient care with respect to patient safety and quality. We examine the basics of a CDS program that helps organizations achieve success with CDS tools. The first fundamental step in any CDS project is to understand the five rights of CDS. These five elements should be identified for any project (Osheroff et al., 2012). Adjustments should be made to close any gap in any of the five components. The five rights may be used as a checklist before deploying a solution into the care environment (see Table 19.1 TABLE 19.1 Five-Rights Framework for Success of Clinical Decision Support (CDS) CATEGORY
DEFINITION
Right information
Evidence-based and actionable information constitutes the "what" of
the CDS program. Right person
Clinicians and the patient constitute the correct individuals impacted by the CDS program, identifying the "who."
Right CDS
The tools that include documents/forms, data display, answers, order sets, algorithms, and alerts define the "how."
intervention format
Right channel
The vehicle for delivering the CDS program, such as within the EHR, or supporting technology, such as smartphones or dashboards, reflect the "where."
Right point in the workflow
The process within which the clinical care is delivered that will be impacted by the CDS program comprising the workflow for redesign using CDS outlining and diagramming constitutes the "when."
EHR, electronic health record.
Source: Osheroff, J. A.,Teich, J. M., Levick, D., Saldana, L., Velasco, .FT., Sittig, D. F, & Jenders, R. A, (2012).
Improving outcomes with CDS: An implementer's guide (2nd ed,). Healthcare Information and Management Systems Society. http://www,himss.org/ResourceLibrary/ResourceDe tail.aspx?ltemNumber=11590.
473
474
II: DATA MANAGEMENT
as a checklist reference). These elements are the foundation for a successful CDS project.
The five rights address the who, what, where, when, and how of a CDS program while emphasizing the importance of clear articulation of goals and objectives that identify all five components (Osheroff et al, 2012). Each of the five rights presents unique challenges for CDS design. There are often several evidence-based sources for information on a topic that contradict each other or have slight differences in standards. A large part of defining the right information on which to base a project is a passionate debate among stakeholders regarding which standard to adopt. The decision-making process frequently takes longer than the build and implementation process. In the digital world, there is no room for gray nuance, which is why consensus on what single standard to apply to any given population is difficult. The system needs to know to trigger off a specific value. A simple example is a hypertension alert. The system must know exactly what the parameter for hypertension is, say a systolic blood pressure of 150 and greater. Even in sophisticated tiered response alerts, each tier must have the trigger parameter defined for the system to work properly. A vague term such as “unstable vital signs" cannot be supported in most CDS models. Defining the right person or persons in a CDS model presents challenges due to the many roles clinicians have in any given organization. Saying the information needs to be available to the nurse must be more definitive. Is it available to all nurses, to nurses
only in the emergency department (ED), to all adult nurses but not to pediatric nurses, to behavioral health nurse managers but not to med-surgical nurse managers? These detailed questions must be answered for the CDS tools to work as intended. Moreover,
with each role added to a process, the matrix becomes more complex. This level of detail is daunting but manageable once system requirements are understood by all stakeholders.
The format is the most recognizable component of the five rights as it is what the end-user sees. Although "hard-stop" best practice alerts (BPAs) are the most familiar CDS tool, this kind of disruptive device should be employed only for life-threatening critical decision points where the only safe choice is the one the BPA allows. The overuse of BPAs leads to all BPAs being ignored as the alert has become meaningless to busy staff. However, other types of CDS tools may be more appropriate to the majority of clinical workflows (eHealthUniversity, 2014). These types of alerts may include passive alerts, information sections, links to source documents, single select paneled choices, and smart order sets. These types of alerts are just a few of the many formats available as CDS tools. A best practice for designing an alert is that the design must be informed by the discipline it will fire to; physicians should never decide what fires to the nurse, just as the nurse should not decide what fires to the physician. The channel used to deploy a CDS tool may not be within the EHR itself. Infinite technology options exist within the clinical environment. Information display boards, wearable communication devices, smartphones/mobile devices, smart intravenous (IV) pumps, and implanted or wearable medical devices are only part of the ever-growing list of devices available to deploy CDS tools. How those tools interface with the EHR, what is the benefit of one kind of device over another, and whether several devices need
to be used are all part of determining the right channel for the CDS interventions. CDS interventions should not be limited to being only inside the EHR, as most clinicians spend as little time as possible logged into a screen. The correct placement in the workflow is one of the hardest of the five rights to determine. Thinking about and designing a process is always different than performing the process. Usability testing by end-u.sers is the gold standard for making sure CDS tools are in the right spots to be most effective at supporting the desired clinical outcomes. Even wlien usability testing has been extensive, every stakeholder engaged, and every
19: CLINICAL DECISION SUPPORT SYSTEMS
nuance accounted for, there will always be a need to adjust once the tools are in the live environment. It is best practice to pay attention when users voice difficulties with
using CDS tools and provide end-users with multiple avenues of feedback such as email address, phone numbers, committee meetings, and so on. The difficulty of a change needs to be weighed along with the change not being what is needed. If a process is too difficult or not seen as beneficial, the clinicians will find a way to work around what they
perceive as a barrier.
Clinical Decision Support: Definitions, Goals, and Objectives Business decision support systems focus on financial metrics and models, whereas CDS focuses on healthcare outcomes by encouraging clinicians to follow best practices
and evidence-based guidelines. The word "support" in the term "CDS" points to the fundamental goals of CDS. CDS is an informatics term that involves technology to aid decision-making, guiding the end-user through complex systems to achieve a targeted outcome (Health Information and Management Systems Society [HIMSS], 2011). When used effectively, the specific build of the technology can make using a system easier and more clinically relevant for the end user. CDS is not meant to make decisions for the or clearer by offering evidence-based choices determined by practice standards, regulatory compliance elements, current literature, and other determinants. Look beyond the myopic view that CDS consists only of evidence-based order sets or hard-stop BPAs. This view of CDS is shortsighted and does not allow for the full scope of what CDS can do to help clinicians and healthcare systems achieve higher quality standards, cost-efficient care, improved patient safety, and better compliance with regulatory reporting. CDS can help achieve these goals for an organization if the tools employed are well designed and user friendly. The primary goal of a CDS program is to leverage data and the scientific evidence clinician but rather to make clinical decisions easier
to help guide appropriate decision-making. When looking at ways in which CDS tools can be leveraged in a clinical process, the CDS team needs to approach the project in a data-driven manner supported by the evidence. This requires an in-depth analysis of the scientific evidence coupled with data-analysis methods to identify gaps in practice within the organization. It is equally important to identify where there are gaps in the organization's ability to report how they are doing concerning the quality of patient care and whether recommended practice guidelines are being followed. The absence of data captured to track adherence to guidelines or quality measures may indicate where an organization needs to focus on CDS strategies. It is the role of the CDS team to identify all of the elements of a process and use data to identify areas where processes might be enhanced with the use of CDS tools to provide users with the best evidence and to support appropriate decision-making and treatment decisions (HIMSS, n.d.-b).
Two tools available to CDS teams are predictive analytics and prescriptive analytics. Predictive analytics takes the available data and applies logic or algorithms to calculate the likelihood of an event. A fall risk tool is a simple form of a predictive analytic tool. The tool provides a calculated score. The score is used to predict a low, moderate, or high risk of the patient falling. Predictive analytic tools are evolving into sophisticated data-mining calculations and away from the antiquated manual checkbox methods.
The cognitive burden of complicated scoring is what computers are meant to do. It is up to the clinician to determine how to act on the information instead of wasting time finding the information. More than three or four questions feels like a burden to the busy clinician while the computer can access hundreds of data points with minimal burden on
475
476
III: DATA MANAGEMENT
the system. Prescriptive analytics takes the concept of predictive analytics a step further. Prescriptive analytics are intended to lead the clinician on a defined pathway to address the identified issue. An example of a predictive analytic tool is the system looking at hundreds of data points to give a weighted sepsis risk score. If the score following certain parameters of risk triggers a sepsis guideline order set, then the tool falls within the category of predictive analytic CDS tools.
Caution must be used when designing and implementing predictive and prescriptive analytic tools. While most algorithms work for most patients, healthcare deals with human beings, not widgets. Clinical judgment and special circumstances must always be allowed for any CDS process and should be seamless within the clinical workflow. For example, CDS can reinforce nationally established guidelines to address patient safety, quality, and population health; however, care must be exercised to ensure best practices are maintained. Content creation and maintenance are labor intensive and keeping up with best practices requires vigilance. Vaccination adherence is one example that is useful to consider, particularly given that vaccinations have become controversial as a public health issue, with many families electing not to vaccinate because of personal beliefs about vaccination safety (Lieu et al, 2015). However, there are national quality measures that healthcare providers and hospitals are expected to report and to perform well on concerning adhering to vaccination algorithms (National Quality Fomm, 2008). In the event one's organization resides in a community with large numbers of individuals who reject vaccinations, one's institution will appear to perform poorly related to federal guidelines on vaccination unless data are captured that indicate "patient refuses vaccination." In this scenario, if an institution's quality goal is to achieve 100% compliance with influenza screening, vaccination, and required reporting, the CDS tools would be designed to support the entire process to reinforce quality and efficiency, not just to provide data capture for regulatory and reporting requirements. In this example, not only would the workflow of the clinicians be supported with efficiency, but also the screening tool would be designed to lead the clinician to the correct orders for the patients and to trigger the best decision on behalf of the patient. Data can be structured in such a way that compliance with the measure is accurate and easily reportable regardless of indication, including the patient's refusal to be vaccinated. Also, when data are captured in a structured format, the reporting tools generating data from the EHR can also alert leadership when measures are not being met so that improvement strategies can be launched to address poor performance. It is not the job of the EHR or CDS to enforce compliance but instead to make compliance with evidence-based protocols easier and more accurately reportable. The goals of a CDS program to address adherence to an influenza algorithm would be to: 1.
Use the relevant data and information about the patient to determine whether the patient meets the clinical requirements for vaccination.
2.
Note any contraindications for vaccination.
3.
Document reasons for not administering the vaccination for a patient meeting clinical criteria within the algorithm, such as refusal of the vaccination. Capture the data in a structured format to guide the clinician through the algorithm.
4.
5.
6.
Structure data and information to document compliance with the regulatory requirements and to support quality improvement (QI). Align workflow of the clinicians during assessment and treatment with the efficient administration of vaccinations as necessary (Texas Health Resource.s, 2014).
19: CLINICAL DECISION SUPPORT SYSTEMS
Because the CDS tools often have unintended consequences, such as leading is suggested by the CDS tool, it is the responsibility of the CDS team to keep the process transparent so these kinds of pitfalls can be foreseen and avoided. If it is the intention of the tool to eliminate alternatives to a process, that too must be vetted by the clinicians during the design process. By making it easy to do the right thing at the right time, CDS tools support
clinicians to think there is no alternative but what
safe practice. However, this does not negate the need for clinicians to know what is safe or unsafe but instead helps make the safe choices clearly evident to the user within the documentation (Institute of Medicine [lOM], 2011). Additionally, one way to
address unintended consequences is to consider human factors design principles when developing CDS strategies.
Clinical Decision Support: Importance of Human Factors Design Principles
Human factors design has developed from human factors design engineering and takes into account basic principles. These principles, according to Phansalkar et al. (2010),
apply the knowledge about human capabilities and limitations to the design of products, processes, systems, and work environments. This research team applied these principles to the design and implementation of CDS alerts with an extensive review of the scientific literature. This team's actionable recommendations
for the design and implementation of
future clinical information systems are as follows: 1. Alert philosophy: Develop an approach to "categories of problems" that should be included in the alerting system and consider "how many priorities there should be for each category of risk." These authors make a distinction between medication alerts and alerts to mitigate errors.
2. Prioritization of alerts: Alerts should be developed with levels for low, medium, and
high, with coding schemas such as "word, color, shape, position on the screen, and other indicators known to influence urgency."
3. Low-priority alerts: Avoid the use of low-priority alerts and instead provide information to the end user.
4. Linked information: If information is linked in forming a holistic judgment, then the information should reside together perceptually; however, information that is dissimilar and should be considered separately should be easy to differentiate. 5. Time sensitivity: Contemporaneous information should be presented at the same time in the system. 6. Tailored interfaces to the end-user: This is likely to reduce false alarms, be more satisfying to the end-user, and reduce error. 7. Hard stops: Alerts that require acknowledgment before the user moves on
minimally or not at all constitutes a "hard stop." Use these types of alerts minimally and cautiously.
8. Auditory alerts: Combine auditory alarms with visual alerts. 9. Mental models of the end user: The presentation of alert information should attempt to match the mental models of the end user.
10. Format of alerts: The format should be designed to avoid habituation (Phansalkar etal., 2010).
477
478
III:
DATA MANAGEMENT
STRATEGIES FOR IMPLEMENTING A SUCCESSFUL CLINICAL DECISION SUPPORT PROGRAM
First and foremost, the CDS program must be strategically aligned with the mission, vision, and values of the organization (Kendall & Kendall, 2014). A successful CDS implementation requires a balance aniong people, processes, and technology. The people aspect of this balance is not only the most important but also the most challenging. It is CRicial to have engagement and buy-in at the top levels of the executive team and to permeate the entire organization with the commitment to change aligned with quality strategies. Because CDS programs involve changes in process and workflow, the CDS team must involve the stakeholders who are the most impacted by the process redesign and strategically design the technology component using the EHR functionality appropriately. Q1 professionals are important partners in this strategic alignment of the organization and optimize the technology using CDS as a tool in the toolkit for improvement. Key strategies for success with CDS share common themes with other success strategies and align with recommendations outlined in Chapter 8 dealing with the systems development life cycle. In addition to addressing the human factors design principles noted previou.sly, these strategies include the following:
1. Ensure the right stakeholders participate in the process. 2. Understand the full process before beginning to design CDS solutions and tools. 3. Recognize that documentation cannot solve problems, but it can make solutions easier; conversely, it can also further exacerbate issues.
4. CDS leaders must be strong enough to do what is right instead of what is easy. Often, addressing an issue means an enormous amount of work in the background where the user sees only a slight change in the EHR.
5. Resources should be considered and should justify the need for the CDS. Sometimes it takes a complete rebuild to help address an issue.
6. Stakeholders need to be engaged throughout the process, not just at the beginning. 7. Ensure design, vetting the build, testing, and evaluation/fol low-up after installation is done with the frontline users.
8. Ensure leadership and the groups using the data for reporting and outcome tracking are engaged in the process to verify that the strategic goals of the organization are met, as well as the needs of patients and clinicians. 9. Recognize the CDS team needs to understand the strategic goals before designing a process with the frontline clinicians. All too often, what is done in day-to-day practice is not what is spelled out in the policy or called for by regulation (Kendall & Kendall, 2014).
Characteristics and Elements of a Successful Clinical Decision
Support Team A successful CDS program requires builders who understand the clinical relevance of the care being addressed by the CDS and relevant workflows and processes (Osheroff et al.,
2012). The ideal situation is to have builders who are also clinicians and who understand the workflow. The CDS team members should mirror the roles in the organization they are
designing support tools for, meaning if the team is going to support physicians, nurses.
19:
CLINICAL DECISION SUPPORT SYSTEMS
and other allied health professionals, the CDS team should have those same clinicians represented on the team. Problems quickly emerge when physicians try to design processes for nurses or vice versa. Crucial disconnects occur when the builder and the
user do not speak the same language, in this case, healthcare-specific terminology. Even among healthcare providers, the terms used by a neonatal nurse may be very different from those used by a geriatric oncology nurse. Based on the authors' experience, these kinds of disconnects must be identified and eliminated.
Having a skilled informatidst
as part of the design team can save time and effort in being able to have a usable end product.
Translation between technical IT people and clinicians is more of an art than a science. From the experience of the authors working with interprofessional teams on CDS, a program requires leaders who can clearly translate between "IT speak" and clinical terminology. Figure 19.1 notes the importance of a team approach to the success of a program. Simple terms, such as "close," can cause huge confusion if not defined in
the group using the term. For example, the IT builder thinks "close" means to collapse or not see all of something. The surgeon thinks "close" means finishing the task. The nurse thinks "close" means to go to the next task. There needs to be facilitation by the CDS team to ensure every stakeholder fully understands the terms and functionality of the tools being developed and deployed. As users become savvier with functionality and IT terminology, this process will improve, but the need to clarify so that all team members understand will always be there. This facilitation often includes educating users as to what the systems can and cannot do. Many clinicians think the computer can do whatever they want it to do and are surprised to learn that there are limitations within any software. From the experiences of the authors, a key role of nursing informatics (NI) is to be able to articulate to end users what CDS can and cannot do effectively. For example, the group of stakeholders wants there to be a hard stop in an order entry. The CDS builder needs to be able to articulate that putting the hard stop in place will cause FIGURE 19.1 InterprofessionalTeams and Clinical Decision Support (CDS).
Note: EHR, electronic health record; HIE, health information exchange; PHR, personal health record.
479
480
ill: DATA MANAGEMENT
the order to function differently than an order without a hard stop. From the authors' experience, clinicians often think they want something until they learn the downside of their request. Once again, this is an important role that the nursing informaticist plays: educating stakeholders on the capabilities of the system before asking them to make design decisions. Often, clinicians will ask for what is familiar to them instead of what the system is capable of doing for them. The CDS team needs to understand the goals of a project so they can recommend the best ways for the system to support the users in meeting their objectives. The discussion should focus on the problem the user is trying to solve instead of what tools should be used. Once the desired outcomes are defined, the NI
and CDS team can recommend the most appropriate option or options available within the system. Mismanagement of tools, such as BPAs, will lead to fatigue and, ultimately, cau.se more to be missed than caught because of users ignoring alerts. The CDS team is responsible for addressing this kind of poor decision support to do a better job of making the right thing easy instead of relying solely on reminders or alerts. As sociotechnical theory would indicate, a truly successful CDS process is experienced by the end user as a seamless and unobtrusive process while still guiding the end user to the safest, best choices (lOM, 2011, p. 77). To work effectively, these features need to work in tandem and be well designed by effective multidisciplinary teams. Frameworks for Success
Bates and colleagues have outlined a framework for effective CDS that they refer to as the "10 commandments" for success. The framework is noted in Table 19.2, and it includes
recommendations on timeliness, end-user needs, addressing resistance, simplicity, monitoring impact, and managing the system based on the evidence. These factors reinforce end-user acceptance as well as best practices for seeking the best evidence to inform patient care through the use of safe, efficient, and effective CDS strategies (Bates et al., 2003). Failing to follow these factors can cause frustration for the end-user, decreasing compliance with the original CDS planned.
TABLE 19.2 Success Factors for Clinical Decision Support (CDS) 1.
Speed is everything.
2.
Anticipate needs and deliver in real-time.
3.
Fit into the user's workflow.
4.
Little things can make a big difference.
5.
Recognize that physicians will strongly resist stopping.
6.
Changing directions is easier than stopping.
7.
Simple interventions work best.
8.
Ask for additional information only when you need it.
9.
Monitor impact, get feedback, and respond.
10.
Manage and maintain your knowledge-based systems.
Source;Adapted from Bates, D.W., Kuperman, G. J„ Wang, S., Gandhi,!, Kittler, A., Volk, L., Spurr, C„ Khorasani,Tanasijevic, M., & Middleton, B. (2003).Ten commandments for effective clinical decision support: Making the practice of evidence-based medicine a reality. Journal of the American Medical
Informatics Association. 10{6), 523-530. https://doi.org/10.1197 /Jamia.M1370.
19: CLINICAL DECISION SUPPORT SYSTEMS
TOOLS AND TYPES OF CLINICAL DECISION SUPPORT
CDS encompasses a wide variety of tools. These tools include, but are not limited to, computerized alerts and reminders for providers and patients, drug-drug interaction alerts, underdose or overdose alerts (based on renal or liver function or age or drug
levels), actionable clinical guidelines, condition-specific order sets, focused patient data
reports and summaries, documentation templates, diagnostic support, and contextually relevant reference information (HIMSS, n.d.-a). Table 19.3 includes a description of some
of the most commonly used tools within CDS programs. According to Osheroff et al. (2012), another consideration for CDS is when and how the data needed to support decision-making are presented. An example noted is patientspecific data of relevant labs, such as the results of a patient's renal and liver function during the computerized provider order entry (CPOE) of medications that might be contraindicated based on certain lab results. Population-specifi c data are also used; for
example, micro-biograms, which are tables of local bacterial flora and their sensitivity and susceptibility to various antibiotics, can be used for CDS. TABLE 19.3 Types of Clinical Decision Support (CDS)Tools With Descriptions TYPES OF CDS TOOLS
DESCRIPTION
Smart documentation forms
Forms that are tailored based on patient data to emphasize data elements pertinent to the patient's conditions and healthcare needs
Order sets, care plans, and protocols
Structured approaches to encourage correct and efficient ordering, promote evidence-based best practices, and provide different management recommendations for different patient situations
Parameter guidance
Algorithms to promote correct entry of orders and documentation
Critiques and "immediate" warnings
Alerts that are presented just after a user has entered an order, a prescription, or a documentation item, to show a potential hazard or a recommendation for further information
Relevant data
summaries
A single-patient view that summarizes, organizes, and fitters a patient's information to highlight important management issues
Multi-patient monitors
A display of activity among all patients on a care unit, which helps providers prioritize tasks and ensures that important activities are not omitted while providers are multitasking among patients
Predictive and
Analytic methods that combine multiple factors using statistical and artificial intelligence techniques to provide risk predictions, stratify patients, and measure progress on broad initiatives
retrospective analytics Info" buttons
Expert workup and management advisors
Filtered reference information and knowledge resources within fields or "buttons" where information is provided to the end user in the context of the current data display, also referred to as metadata, or "data about data"
Diagnostic and expert systems that track and advise a patient workup and management of the patient based on evidence-based protocols
Event-triggered alerts
Warnings triggered within the system based on data that alert the clinical user to a new event occurring asyndnronously, such as an abnormal lab result
Reminders
Time-triggered events within the system reminding the clinical user of a task needed to be based on predetermined time within the system
Source: Health Information and Management Systems Society, (n.d.-a). CDS 101: Fundamental issues. Retrieved August 30, 2021, from http://www.himss.org/lib rarY/clinical-decision-support/ issues?navltemNumber=13240.
481
482
III: DATA MANAGEMENT
CDS functionalities may be deployed on a variety of platforms (e.g., mobile, cloudbased, locally installed). CDS is not intended to replace clinician or patient judgment but is deployed as a tool to assist care team members in making timely, informed, higher quality decisions. CDS is frequently not only an integrated part of the provider's EHR but may also be present in a variety of other technologies such as pharmacy systems, patients' personal health records (PHRs), or patient portals. Some providers use CDS as a "service" by seairely sending patient information to a registry, implementing cloud-based CDS interventions, or using forecaster programs that can respond about what treatments or diagnostic testing might be appropriate for the patient (Osheroff et al., 2012). As technology advances, new processes for creating and using CDS must be conceptualized, investigated, and implemented. One approach is the exercise around maintaining evidence-based CDS. As previously mentioned, content creation and authoring CDS is time-consuming. The contemporary thought is developing content
that would then be deposited into repositories. The repository would be maintained by federal agencies such as ONC, AHRQ, and so forth.
The content creation would be written in a language such as clinical quality language (CQL), which is easy to read and understand. These content items are referred to as "artifacts." Artifacts rely on Boolean logic to advance through the process. An example would be Abnormal Blood Glucose and Type 2 Diabetes Mellitus: Part One, Screening (AHRQ, 2019). This scenario could be used in different ways, such as order sets, reminders, or alerts. The inclusion criteria would include items such as age and BMI with personal history (including gestational diabetes) or family history of diabetes, and so on. Exclusion criteria would include items such as active pregnancy, impaired glucose tolerance testing, recent hemoglobin AlC, and so forth. Each of the inclusion and exclusion criteria would be selected when setting up the artifact using national value sets and vocabularies such as the Value Set Authority Center (VSAC) housed in a repository maintained by the National Library of Medicine (NLM, 2021). This library includes clinical vocabularies such as SNOMed, RxNorm, Loinc, and so on. The concept is to develop the artifact in the repository website and then submit it for review and approval. When creating the artifact, one would select the concept terminology instead of the codes or identifier numbers. The requirement within the EHR would be to ensure procedures, tests (laboratory, radiology), and documentation are mapped or aliased with the value set identifier. Once this task has been completed, the mapped item can be used anywhere within the EHR from different areas or documented by different end users. Once the values are set up in the EHR, an application programming interface (API) could be used to leverage the information and either pull the information from the NLM on demand, or it could be set up to reside internally and download on a routine basis. The concept of on-demand CDS is showing great promise. The ONC and the MITRE corporation are two organization that have taken the lead on moving content from narrative to staictured. As mentioned, content creation and maintenance are laborious.
Using the methods mentioned herein will hopefully keep CDS evidence based. There will always be local practice patterns, and with any CDS, one must take into account the culture of the facility to prevent alert fatigue and documentation burden as well as enduser acceptance.
USE OF CLINICAL DECISION SUPPORTTO ALIGN IMPROVEMENT INITIATIVES
CDS can be governed within the quality framework of the organization and used strategically to reinforce quality and patient safety initiatives within organizations. The
19: CLINICAL DECISION SUPPORT SYSTEMS
core to a successful implementation of any QI is strong leadership. This is particularly
true of CDS programs because they require commitments throughout the organization because of impacts with workflow and ongoing investment of capital and personnel. As with any organization committed to QI, teams led by strong leaders must be brought together to develop a shared vision of quality and patient safety. This includes physician
champions, chief nursing officer, chief financial officer, chief information officer, and their staff. Accountability can then be established for the desired outcomes. Clinical champions are required to gain buy-in for the CDS effort. Champions serve as change agents, represent their groups in reviewing design and prioritizing projects, and communicate effectively to and from the clinician groups impacted by the changes (Osheroff et al., 2012). CDS programs also require stakeholders who are most impacted by the changes to clinical workflow to help in designing strategies and implementing plans. Successful CDS programs are implemented with the stakeholders rather than bringing forced change to the stakeholders (HIMSS, n.d.-b).
Clinical Decision Support Oversight Committee Osheroff et al. (2012) recommend that a CDS oversight committee needs to be established with the support of senior administrators to own and manage decision support workflows and functions. Members should represent a cross-pollination from the pharmacy and therapeutics committee, EHR committee, patient quality and safety, nursing unit directors, senior leadership, as well as members responsible for the CDS build. Their first actions would be to develop a charter as well as processes and procedures, and then identify committees that they need to interact with to improve care processes and workflows. For example, if the CDS program does not fall under the QI department, a key partner in the process is to engage the QI leadership and staff in the process. Conversely, if the CDS program falls under the QI department, the key to success is a strong relationship with the IT department. Another critical partner in the process is the NI content expert. It is the experience of the authors that nursing informaticists frequently lead the CDS initiative with support from physician colleagues, and the authors recommend if this is not the case, the NI content experts are important stakeholders to engage in strategizing the use of CDS.
Strategy sessions should be held to determine the CDS program scope that will best support the organization's goals and programs (Osheroff et al., 2012). For example, should there just be a few tools, to begin with, and should one build on them as needed or start with many tools and then systematically turn off those not needed? A clear understanding of the organization's prioritized opportunities for improvement, as well as CDS functionality and review capabilities, are needed to ultimately determine which users will benefit the most from the various types of decision support tools selected for use (Osheroff et al., 2012).
Deployment of Clinical Decision Support Interventions Once the executive oversight committee and the team responsible for the intervention have aligned on the CDS program strategies, the design of the intervention takes place (Osheroff et al., 2012). The design phase should be validated with all stakeholders who will be impacted by the process; the intervention should be developed, followed by full testing before taking the intervention into the full production environment. Once the testing is complete, the intervention is ready for deployment. Evaluating and measuring the impact following the cycle of improvement is a critical step.
483
484
I: DATA MANAGEMENT
Measuring the Success of the Program Evaluation strategies to measure the impact of the CDS program are an important consideration for the team. CDS program evaluation can include both quantitative and qualitative methods. It is important to strategize before implementation; the team will decide whether the intervention is working as expected and improving patient care (Osheroff et al., 2012).
One qualitative evaluation strategy is to hold focus groups or survey the stakeholders most impacted by clinical workflow changes. Suggested questions to ask clinicians based on recommendations of the authors are: (a) How is the process working for the nurses and physicians? (b) Does the CDS program interfere with patient care or create unintended patient safety consequences? Feedback is important to creating a continuous learning environment to inform improvement. Quantitative measures to monitor improvement are equally as important. Methods described in Chapter 21 recommend control charts and various tools for QI that should also be considered in designing quantitative evaluation strategies. Quantitative outcomes and process measures are also important to monitor for reporting to leadership on the effectiveness of the program and to share as best practices when successful programs have had a significant impact (Osheroff et al., 2012). Another consideration for organizations that design CDS strategies is to align the program with measurement to improve pay-for-performance programs and accreditation requirements. The strategies outlined earlier follow a typical QI initiative life cycle. These strategies can be depicted in the life-cycle process reflected in Figure 19.2 (Osheroff et al., 2012).
FIGURE 19.2 Clinical Decision Support (CDS) Life Cycle Intervention.
f'
Identify area for improvement using CDS CDS Interventions;
Secure management buy-in and executive leadership support
Design^Validate-^Oevelop->Test-»Deploy
¥
CDS Interventions;
Implementation/project management with input from management, end users, and related staff
Evaluate Ure intervention effect; MEASURE IMPACT
Implementation/project management, using framework developed by management/oversight and with attention to effects on end users and related staff, and on the overall strategies, priorities, and clinical standards set by executive leadership
Source; Adapted from Osheroff, J. A.,Teich, J. M„ Levick, D., Saldana, L., Velasco, FT, Sittig, D. R, & Jenders, R. A. (2012). Improving outcomes with CDS: An implementer's guide (2nd ed., p. 45). Healthcare Information and Management Systems Society. http://www.himss.org /ResourceLibrary/ResourceDetail .aspx?ltemNumber=11590.
19: CLINICAL DECISION SUPPORT SYSTEMS
MEANINGFUL USE AND CLINICAL DECISION SUPPORT PROGRAMS
PIP was developed as an incentive and compliance program to advance the adoption and optimization of the EHR. The achievement of promoting interoperability programs (PlP)-clinical decision support (CDS) is a core measure to both stages 1 and 2 of PIP. In stage 1, providers and hospitals are required to implement one CDS rule. However, in stage 2 of PIP, there are more extensive requirements, including a connection in the strategy aligned with core measures of quality within the organization. The requirement under stage 2 of PIP is stated as follows: To meet the decision support requirements for stage 2 of PIP, there must be five CDS interventions related to four or more clinical quality measures at a relevant point in the patient care for the entire reporting period. Effective January 1, 2017, a reduction in reporting requirements for modified stage 2 and stage 3 resulted in the elimination of the CDS requirement for Medicare as well as dualeligible Medicare/Medicaid hospitals. However, the CDS requirement remained intact for Medicaid hospitals (Department of Health and Human Services [DHHS] Centers for Medicare & Medicaid Services [CMS], 2016).
Improving Population Health Using Healthcare Information Technology and Clinical Decision Support Strategies A broader application of CDS can be seen in nationwide efforts to deploy these types of strategies to impact populations. An example of a program that constitutes a national strategy focused on improving population health outcomes that can effectively apply CDS strategies is the Million Hearts campaign (CDC, 2011). The Million Hearts campaign aims at improving cardiovascular (CV) health in the United States; it was launched by the DHHS to prevent 1 million heart attacks and strokes in 5 years. Partners span from across the public and private health sectors, including the Centers for Disease Control and Prevention (CDC) and the CMS; healthcare professionals; private insurers; businesses; health advocacy groups such as the American Heart Association (AHA) and the American Stroke Association; and community
organizations. These partners support the Million Hearts campaign through a wide range of activities. The purposes are to coordinate efforts to reduce the number of people who need treatment, optimize treatment for those who need it, and realize the full value of prevention in CV health. The Million Hearts campaign is based on four tenets (ABCS) aligning with recommended evidence-based practice guidelines. These four areas of focus are: A = aspirin use for secondary prevention (occurs in 47% of patients who could benefit)
B = blood pressure (BP) control (only 46% of people with high BP have it controlled)
C = cholesterol control (only 33% of people with high cholesterol have it controlled) S = smoking cessation (only 23% of people who try to quit get help with combined nicotine replacement and behavioral therapy; CDC, 2011) The strategies that correspond to the ABCS of the Million Hearts campaign are to prevent heart disease and stroke in participants and their families by understanding the risks and what can be done to lower or reduce them (CDC, 2011). Knowing the ABCS
485
486
III: DATA MANAGEMENT
profile and committing to a plan that would lead to reduced irsk is a key strategy for the organization. Table 19.4 provides a sample of what the overall goal and strategy for improvement might look like for a clinic that is focused on participating in a program to improve CV care using CDS (HIMSS, n.d.-a). A vision for the program, goals, objectives, and measurement criteria is important to establish in tlie planning and assessment phase of any CDS program. The stars in Figure 19.3 indicate compliance with indicators for meeting a PIP measure. Table 19.4 outlines the measures, goals, and objectives of what a CDS program might look like related to CV care and the ABCS of the Million Hearts campaign.
TABLE 19.4 Measures of Success for a Cardiovascular
Clinical Decision Support (CDS)
Improvement Program DATA ELEMENTS FACTOR
PERCENTAGE CHANGE
ENDPOINT
METRIC
Aspirin prophylaxes
65% compliance or 38% improvement from
Daily use of 81 mg and percentage increase among those
1. Population
current levels
in the cohort seen by participating clinicians BP*
65% compliance or a 41% improvement from current levels
Cholesterol
65% compliance or a 97% improvement from current levels
Smoking
Daily medication compliance and percentage increase among those in the cohort seen by participating clinicians
Daily medication compliance and percentage increase among those in the cohort seen by participating clinicians
17% compliance with stage of change shift
Stage of change shift at
or an 11% reduction in
quitting and prevalence
prevalence from current
percentage
least one level toward
levels
Weight control
Reduction of weight in 65% of population
Increase in fitness in 30%
of population
measure
1. Rx refills
2. Population surveillance
3. Identify patients failing measure 1. Rx refills
2. Population surveillance
3. Identify patients failing measure 1. Rx nicotine
replacement
2. Identify patient reason for failing target
1, Move 20% of obese to
overweight status 2. Move 20% of
Fitness
surveillance
2. Identify patients failing target
1. Identify patients failing target measure
overweight to normal weight 3. Sentinel changes in the program plan
2. Population
1. Reduce resting heart
1. Identify patients failing target
rate
2. Lower BP in 50%
3. Increase vital capacity by 50%
surveillance
3. Self-report dashboard
measure
2, Population surveillance
3. Self-report dashboard
*MUP configuration template/EHR analytics. BR biood pressure: EHR, eiectronic heaith record; MUR a meaningful use provider; Rx, prescribe.
19: CLINICAL DECISION SUPPORT SYSTEMS
Medicai assistants Medical assistants update patient's vital signs In structured data fields and review or update the medical summary Information Record blood pressure
Record height, weight, calculate BMI Patient Intake
Plot and display growth chart (age appropriate)
Record or review smoking status Verify, update allergy list, or NKDA Verify, update current medications, or annotate “none'
^ H vital signs are ml clinically relevant or appropriate
♦ Provider conducts patient consult or procedure Provider documents consult or procedure r
V.
Provider visit
★
Provider determines diagnosis code
Update patient problem list, or document “none' The use of templates can increase speed, efficienr,y and accuracy
but is mt required for MU. The use of dictation, voice recognidon, or free text is possible. Out you may lose We ability to use E&M codem.
♦ Provider determines patient’s care plan Review alerts, reminders, quality indicators Provider visit
Use diagnosis-based order sets or clinical decision tools
Use EHR to order and transmit lab request ^ A lab interface is not required for stage I but facilitates the ability to comply with COM results management and patient engagement.
Provider selects and prescribes medication as needed Review drug-to-drug and drug-to-allergy interactions
FIGURE 19.3
^ Provider visit
AWorkflow Redesign
Strategy to Support
measure.
Note: BMI, body mass index; COM, clinical quality measures; E&M, evaluation and management; NKDA, no known drug allergies; PIP, promoting interoperability programs.
Use EHR to generate prescription and transmit to pharmacy financial impact on the patient based upon the medications selected by provider
Patient receives information before leaving the practice
EHRs. Stars indicate
compliance with indica tors for meeting the PIP
★
^ Fomulary checking Is not required for stage 1 but may have direct
Electronic Health Record
(EHR): Optimization for cardiac care improve ment using certified
Review patient's Insurance formulary
Patient provided with educational irformation
Check¬ out
★
Patient provided with clinical summary Patient provided with CD of medical information If requested Clinical information and results are sent to Patient Portal
● Generating educational material through the EHR is a menu item but makes it easier to keep up-to-date information ● Patient n>rtal is ml required for stage 1 but facililates patient engagement and communication
★
487
488
M: DATA MANAGEMENT
Clinical Decision Support Intervention
Through the use of standardized clinical documentation forms and CDS alerts based on guidelines, omissions in medications and better integration of multimodal approaches, such as ABCS, can be tailored to individual patient-centered needs in clinical and lifestyle change efforts to reduce CV risks. For example, if a particular patient is a smoker, the EHR would capture smoking "yes" or "no" and, subsequently, trigger the clinician to counsel the patient accordingly and provide support in smoking-cessation suggested services. When CDS tools are designed according to protocol and use-certified robust EHRs, they can provide a collection of data that healthcare organizations can use to track and trend provider performance based on protocol adherence (HIMSS, n.d.-a). Certified EHRs have functionality within the products that can support the creation of clinical documentation forms to standardize the collection of important data elements in a structured data field such that CDS ailes can trigger the clinician to collect the right data and intervene according to protocols such as guidelines outlined in Million Hearts (CDC, 2011). However, EHRs directly from the vendor do
not automatically have the capacity to capture and trigger these types of adherence to evidence-based practice protocols; hence, the CDS implementation team must design and deploy the forms and CDS triggers to intervene effectively. These applications are considered more advanced implementation efforts to effectively optimize the use of these components of the EHRs and to structure the CDS and the quality reporting to work in tandem. In addition to being available to trigger algorithms for the clinical alerts to providers, the structured data fields provide a better ability to track and trend important data elements for quality indicators, including both processes- and outcome-based quality measures. When organizations have reached the level of EHR optimization to effectively trigger alerts based on these types of guidelines, a Texas statewide study of nurses' experiences with their EHRs indicates that clinicians are more likely to be satisfied with the EHR. However, drug-drug and drug-allergy alerts present in the EHR are associated with the dissatisfaction of the EHR. This might reflect the ovemtilization of CDS within the medication-ordering process and that using CDS for guideline adherence is a more mature u.se of CDS within an organization. Researchers interpreted these findings as reflecting a more advanced mature u.se of the EHR but note that further research is needed to clarify these findings (McBride et al., 2017). It is likely that the satisfaction of clinicians is reflective of CDS being used "meaningfully" within the clinical workflow to support decision-making. Examples of this type of use of CDS alerts to support clinicians are the presentation of laboratory values to use in counseling the patient and simplifying the periodic monitoring of key clinical indicators that chart the progress on such things as CV risk-reduction plans. Interventions focused on integrated CDS tools within the clinical workflow can improve the appropriateness of lab and pharmaceutical interventions. Also, these strategies provide a powerful motivation to patients as well as trigger action in providers. Another example of how CDS tools and QI methods can support CV disease interventions is to enable EHRs to more easily collect data to evaluate the impact of QI efforts in rapid cycles of improvement on process and outcome measures.
Workflow Redesign CDS protocols can effectively be "hardwired" into clinical workflows to maximize the opportunity for provider compliance with protocols driven by consistent documentation of structured data, CDS alerts for education, and training and reporting to monitor compliance. Within a CDS program focused on population health outcomes, such as CV
19: CLINICAL DECISION SUPPORT SYSTEMS
disease adherence protocols, it is important to address the current versus future state standardization of workflows.
In the paper-record world, frequently, clinicians have a different workflow for the management of patients and respective patient outcomes. In the process of implementing an electronic environment to help manage patient populations, an equal number of options are Lised to manage patients in the electronic environment as there are in the paper environment, but many more opportunities exist to standardize the delivery of care. Using best practices in clinical workflow analysis, change management, and EHR implementation to help maximize the use of EHRs and respective CDS tools, many of these barriers can be addressed. Refer to Figure 19.3 which reflects the strategies of how elements essential to the Million Hearts campaign can be built into certified EHR products (Tushan, 2012). These figures reflect the work of the AHA working with the ONC to establish recommendations on how certified products can be used to reinforce clinical workflows and documentation strategies under the campaign. These workflows can be used to strategize how and when the data capture needs to occur related to the protocol and how CDS can trigger the clinicians to adhere to documentation and workflow, as outlined in Figure 19.3 (Tushan, 2012).
Education and Training to "Hardwire" Improvements Educational intervention is also an important strategy that is used to support clinicians in maximizing CDS and EHR Rmctions. While training providers and clinical staff on the use of the EHR to standardize clinical documentation, CDS rule sets and custom reports
are pivotal to success in a population health-based CDS program. An education campaign is recommended by the authors before implementation to help reinforce functionalities of the EHR system and the CDS program implemented. Standard educational campaigns have proven effective in services historically provided to regional extension center (REC) member providers and hospitals, and similar methods will be deployed to educate providers on the utilization of clinical documentation forms, CDS, and reporting. Based on the work of the authors with the RECs and preparing clinics to address the Million Hearts campaign, an educational program for a clinic might include the following elements: ■ Review the Million Hearts campaign and best practices related to CV disease prevention, including the ABCS protocol. ■ Review baseline reports on the performance of ABCS measures (if available). The education program ideally should incorporate reports as to current provider adherence to the metrics in the campaign, if possible, although prior to instating the CDS and structured data, often these types of reports can be collected with the manual abstraction of records. Baseline measures prior to the implementation and incorporation of the baselines into the education for providers help reinforce why the CDS program is needed. ■ Train providers and clinical staff on the use of CDS and custom reports. ■ Provide an overview of EHR functionality and opportunities to improve the quality of care and outcomes.
■ Present a vendor-specific overview on how to maximize the use of the EHR in the delivery of care. Vendor-specific education on CDS alerts may be available to utilize in combination with specific training materials that include customized forms and reports to collect structured data that are important to the CV protocol, CDS luiles, and the process and outcome measures. ■ Suggest techniques to adopt and implement new CDS aile sets clinic-wide while ensuring the highest level of compliance by all providers (CDC, 2011).
489
490
I: DATA MANAGEMENT
Evaluation and Monitoring for Success Utilizing a specific report functionality or building reporting capability within an EHR for monitoring progress is an important development strategy (Osheroff et al., 2012). In addition, regularly recurring observational studies, focus groups with providers, and surveys can be used as qualitative methods for the CDS team to evaluate the effectiveness and to improve the program in the long term. For example, if hard stops in an EHR of clinical decision support (EHR-CDS) strategy are enabled, they can be overridden with documentation in the record of the reason(s). These reasons can help document the compliance with the CV algorithm, and if there are apparent clinical reasons for noncompliance, justification is captured in a text field for further evaluation. Along with provider debriefing, these reasons can help inform upgrades and changes in updates to the CDS tools. The CDS team brings together all stakeholders as core team participants in the organization, with the primary goal of the program to fight heart disease and stroke partnered with AHA and other participating organizations across the country. CHALLENGES AND ISSUES WITH CLINICAL DECISION SUPPORT
Although the ONC endorsed certification bodies to test and certify EHRs that require CDS functionality, these products do not come "out of the box" ready to achieve results such as those described by the Million Hearts campaign example noted earlier. They require a significant strategy and infrastructure that is evident from the strategies described. In addition, there can be unintended consequences, including patient safety issues and legal liability concerns with CDS, that are critical to consider in all organizations using CDS within EHRs (lOM, 2011).
Challenges to Implementing Clinical Decision Support Implementing CDS effectively and without provider resistance presents challenges. These challenges are noted in the report by the Agency for Healthcare Research and Quality (AHRQ), Clinical Decision Support Systems: State of the Art, to be primarily related to misalignment in the CDS intent and what the end users intended to do prior to receiving the alert, timing, and autonomy (Berner, 2009). Timing is noted as an issue in the report, indicating that providers may agree they need alerts on preventive services but disagree on the timing of when to receive the alerts within their workflow. Additional issues are speed and ease of access to alerts. The third and likely the most significant issue, according to the report, is the autonomy desired by clinicians related to how much control end users have over their response to the CDS. This area relates to whether the CDS alert is a "hard stop," preventing the clinicians from moving forward in the EHR until the alert is addressed, and whether it takes significant effort to override the alert (Berner, 2009, p. 8). Timeliness of documentation is also an important consideration when presenting a clinician with a critical alert. If the alert is intended to warn of a critical issue with a
patient in an early warning system such as the Modified Early Warning Score (MEWS), the documentation must be timely. If, for example, a CDS alert is set up to alert the nursing staff related to patient deterioration based on the MEWS score and the vital signs were taken on the entire ward prior to entry into the EHR, then once the vital signs are entered, it could be too late to intervene due to the age of the vital signs. As discussed by Capan et al. (2018), early recognition of physiological deterioration requires data-
19: CLINICAL DECISION SUPPORT SYSTEMS
driven approaches to allow frontline clinicians to detect, communicate, and treat early deterioration.
Features of Safe Healthcare Information Technology to Address Challenges The lOM (2011) report Health IT and Patient Safety: Building Safer Systems for Better Care outlines several recommendations. The report focuses on end-users and recommendations as to what constitutes safety from an end-user standpoint. These recommendations are relevant to CDS programs and how the tools are deployed, which aligns with these recommendations. As noted earlier, some of the major challenges related to CDS are that the CDS does not align with end-user expectations. As a result, aligning CDS development with the lOM recommendations helps the industry design systems that are safer and more effective. The recommendations are as follows:
1. Retrieval simplicity that is accurate and timely with both native and imported data
2. A system that the end-user desires to interact with 3. Simple and intuitive displays of data 4. Easy navigation
5. Evidence at the point of care to aid decision-making 6. Enhancements to workflow with automation of mundane
tasks, streamlining tasks
rather than increasing physical or cognitive workload 7. Easy transfer of information to and from provider organizations 8. No unanticipated downtime (lOM, 2011) The recommendations also indicate that the cognitive workflow and the decision making process in healthcare are complex, with the need for clinicians to rapidly simulate massive amounts of data within the decision-making process in complex, rapidly changing environments. The report further emphasizes the need for timely information and not a cumbersome, time-consuming, or too rigid pathway that the clinician must navigate. Further, there is an emphasis on the fact that the most vulnerable period of time for patient safety issues occurs at the initiation of new technology. This period of vulnerability further reinforces the importance of training, particularly with changes in workflow that are typical of CDS implementation. Usability guidelines are also an important consideration. To address these issues, the lOM report recommends several guidelines established by the National Institute for Standards and Technology that address design strategies in incorporating best practices in usability design, methods for evaluation and improvement, usability engineering, recommendations on organization commitments to usability, and testing guidelines related to usability and patient safety (lOM, 2011).
Legal Implications Although there are some noted liability risks to providers who use CDS, there could also be exposure to not acting on a CDS alert. Greenberg and Ridgely (2011) evaluated the malpractice risk associated with CDS use and concluded that the most important issue with regard to liability is whether CDS tools are well designed and well implemented. They determined that a well-designed CDS process should provide only those alerts that
491
492
III: DATA MANAGEMENT
are clinically relevant, reduce the likelihood of alert fatigue, and allow clinicians to detect adverse events. This led them to conclude that adopting a well-designed CDS program would reduce the overall malpractice risk. An article by Kesselheim et al. (2011) reached similar conclusions regarding the impact of reducing alert fatigue and emphasized the importance of using clinical judgment when interpreting the output of a CDS system. To further reduce liability, they also recommend a stronger government regulation of CDS and the development of international practice guidelines (Kesselheim et al., 2011). The issue of sharing CDS content among institutions has been discussed in the context of using a Web 2.0 architecture to encourage interoperability (Wright et al, 2009). Citing other references, Wright and colleagues concluded that the patient's healthcare provider is responsible for making the final decision on the clinical relevance of any shared CDS content. This conclusion basically indicates that existing "hard copy" references that aid in clinical decision-making are no different than electronic support documentation. However, they caution that there is little case law as precedent to address these questions (Wright et al., 2009). With this position, it could be further concluded that clinicians would be held to the same standard of care regardless of whether a CDS system is used, and as long as clinicians make the final decision, the use of CDS should not increase liability risk. In other words, clinicians are held to the same level of accountability with the EHR as with the paper-based record. Information-based liability is an important issue for all clinicians. The rapid expansion of medical information databases, EHRs, and associated medical expert systems and CDS has the potential for impacting medical malpractice (Greenberg & Ridgely, 2011). This is particularly tnie given that the electronic world has the ability to track actions and potentially follow the "train of thought" the clinician had given the footprint in the electronic record, which tracks clicks, routes taken through the documentation in the EHR, and whether or not the clinician responded appropriately to the algorithms and CDS built into the system. Providers could potentially be held liable for failing to access a computerized medical database, failing to use available software providing decision support, or using this technology in an improper, inexpert, or inappropriate fashion with the electronic record admissible in a court of law. Likewise, nurses, hospitals, and healthcare systems may also be liable for staff failing to adhere to alerts. However, an area that is unclear is what actually constitutes the permanent legal record, and whether or not the CDS alerts, data that trigger the alerts, the algorithms behind the alerts, and the actions that result based on the alerts are a part of that permanent record (Kesselheim et al., 2011). A well-designed CDS system should provide alerts that are clinically relevant, reduce the likelihood of alert fatigue, and allow clinicians to detect adverse events (HIMSS, n.d.-a). CASE STUDY 1
Obstetrical Screening
Prob/em;When a potentially laboring mother presented for an obstetric screening exam to determine whether she was truly in labor, there was no standard workflow for the placement of orders necessary for the exam and no standard order set for the exam.The divergent practices In placing orders and workflows for conducting the exams caused several problems for the multiple hospital system. Problems included issues with appropriate billing and reimbursement for the exam, putting the nurse in jeopardy of practicing outside the scope of licensure, potentially
19: CLINICAL DECISION SUPPORT SYSTEMS
missing elements of the exam needed for best patient care, and the risk of violating Emergency Medical Treatment and Active Labor Act laws. To understand all the elements needed to address the problem, the CDS team partnered with the Nl team to learn the entire process and identify the areas in need of impact by CDS tools. Stakeholders:The stakeholders identified for engagement in the process were obstetrics (OB) triage nurses, obstetricians, quality/risk department representatives, billing/finance representatives, compliance department representatives, individual member hospital nursing leadership, the CDS team, and Nl. Individual and group meetings with the various stakeholder groups were conducted to get a clear understanding of all the components needed for a successful process. Design:The CDS team did an analysis of orders used for obstetrical triage across the system to find commonalities, identify differences, and see which hospitals had gaps in available orders. Comparisons were made while looking at denials for reimbursement and successful reimbursement to identify any potential missed revenue opportunities. The CDS team also did an analysis with Nl and nursing leadership to identify the appropriate scope-of-practice issues with the obstetrical triage exam screening orders and workflows.The two teams (CDS and Nl) worked with obstetricians to ensure that appropriate care standards for the potentially laboring mother were addressed. Once the clinical workflow and necessary orders were identified, CDS and Nl took the information to other stakeholders to ensure
the process met with reimbursement, quality, and legal and regulatory guidelines. A standardized system order set was built that included all the necessary orders for a potentially laboring mother with the verbiage necessary in the orders to qualify for maximum reimbursement.This system order set replaced all the other hospital-specific sets that had been in use. Within the approved workflow for obstetrical screening, a properly certified RN can conduct portions of the exam without contacting the physician. The order set included prechecked orders to cover these items so the nurse can easily use the order set with the fewest number of clicks and know the orders are appropriate to be placed without physician input. Depending on the results of the exam, the nurse must contact the physician to receive further orders for the laboring mother. The order set can be cleared and then reused by the nurse to place the orders being received from the physician. This prevents the nurse from having to go to multiple order sets and allows the nurse to sign new orders appropriately to protect their licensure.The design of the order set allows for easy maintenance by the CDS team, which is very important for the sustainability of the use of the set; reduces the number of order sets the clinician has to search for, which is a huge satisfierfor staff; and contains evidencebased orders with best practice recommendations to provide the best care for the patient.
lmplementation:Jhe implementation plan for the new obstetrical triage exam order set included training for all staff who use the set, notification of the billing and coding departments of the new orders being used, and training and support staff for the EHR. Users were very satisfied with the reduced number of clicks to get what they needed and with the ease of use of the set. Users also indicated that "knowing they were within their scope of practice" made them more confident in using the new set. Adoption of the set was quick. Within 60 days of implementation, reimbursements increased, and denials markedly decreased. Physician feedback
493
494
III: DATA MANAGEMENT
was very positive, as the implementation led to fewer calls to the physician, and the calls that were necessary were shorter and easier to manage. This allows nurses to place initial orders with reduced clicks and to be able to clear the set and use it while speaking to the physician to obtain further orders if necessary. OutcomeiThe use of CDS tools to better design and build the obstetrical screening order set has resulted in increased revenue capture for all system hospitals, ease of use for all clinicians, promotion of nurses practicing within their scope of licensure, and increased satisfaction of nurses and physicians with the EHR.
CASE STUDY 2
Think Sepsis
Problem:The organization's original process for early identification was based on a paper tool built into the EHR where 100% of patients entering the ED were screened. Nurses were hurried, missing key signs of infection, and did not consider the questions to be applicable for certain patients. Adherence to the processes was poor, and nearly half of all sepsis patients presenting in the ED were not identified through the screening process. Stakeholders: The stakeholders in the process are the ED physicians, ED nurses, coding/compliance department, CDS, Nl, hospital leadership, quality/risk department, quality abstractors, and sepsis coordinators. Des/gn; Since there is no definitive test or single finding to diagnosis sepsis, the design team instead worked on the premise that a patient must have an infection to have sepsis,Therefore, the infection confirmation is the trigger event within the clinical workflow relevant to CDS design. A process to provide early recognition of potential sepsis patients by leveraging EHR capabilities to identify patients at high risk of infection was developed. All of the available information pertinent to the risk of infection is presented to support the optimal clinical judgment. Once presented with an identified patient that might have an infection, staff are reminded to Think Sepsis as part of developing their clinical plan for the patient. Patient flow through the ED was segmented into phases.The use of historical data available in the EHR for any returning patients serves as a foundation then, layering new information as it is captured creates a dynamic and comprehensive clinical picture. Figure 19.4 reflects a visual representation ofthe flow of information. The CDC tools used to support the new workflow were well received by stakeholders, and they were actively involved in the design of determining where the tools should be seen and used.
■ The advisory immediately appears as soon as sepsis criteria were met.
■ Display all information as to why the advisory triggered. ■ Provide quick access to specific patient information and documentation to address the advisory.
Outcome: After the deployment of the new CDS tools, several lessons were learned. Unintended consequences required an agile approach to optimization.
19: CLINICAL DECISION
SUPPORT SYSTEMS
FIGURE 19.4 Recent historical Information Is Merged With the Current Clinical Picture to identify Infection Risks and Early Signs of Body Functional Abnormalities.
Patient arrival
Extended Contact VVIttl immuno
compromist Cancer HWAIOS
Asplenia
Healthcare
Registration
Tralge
● Arrival
● ctiie)
oincleM
comol^ ● Vital signs
Institution
● Homemeds
● Recert admission
● Patient
● Livesina
● Uvlng
nursing home
history arrangements
or insthution
Though the ultimate aim was early identification and treatment of sepsis, the criteria were designed to recognize a broader population, those with infection risk accompanied by abnormal nursing assessment findings.The concise delineation between sepsis and infection risk was not consistently maintained throughout the deployment.This resulted in a misunderstanding of the criteria to be more like a diagnostic tool instead of the intended information-gathering tool. Many times, clinicians encountered the alert immediately and were required to make a clinical decision even before having assessed the patient.The redesign allowed clinicians to review the chart, assess the patient, and determine a course of care prior to the alert firing to the clinician.The advisory showed only if exclusion criteria had not been met and fired at the time of placing orders. This change greatly improved physician and nurse satisfaction. Clinicians felt the advisory contained too much information, took too long to read, and didn't offer the ability to document or place orders directly from the advisory. An agile redesign eliminated wordy explanations and simply provided quick documentation and access to orders that greatly improved usage and compliance of the expected workflows.
CASE STUDY 3
Normal Newborn
Prob/em; Because of changes in both state law and federal regulations surrounding how physician orders must be authorized and standards related to the standing delegated order (SDO) set, the leadership of the enterprise directed the Nl and the CDS teams to lookat ways to make the order sets compliant with the new regulations. The orders and processes surrounding normal newborns were chosen as the test
495
496
III: DATA MANAGEMENT
case for a new model for order sets and workflow improvement. Normal newborn orders must include orders for time-sensitive medications given to the infant in the first hour of life. These time-sensitive orders posed the greatest difficulty in developing an order set that would be compliant with the new regulations and that helped the nurses' practice within their scope of licensure while still being able to appropriately care for the infant Although there are caveats in both state and federal regulations that relate to normal newborn care, the authentication process still had to be addressed for the orders to be valid. Stakeholders: The stakeholders in the process are the labor/delivery nurses, newborn nursery nurses, pediatricians, obstetricians, coding/compliance department, CDS, Nl, hospital leadership, quality/risk department, accreditation department, and legal department, Design:The Nl collaborated with the CDS team to develop a model for new order sets. Reasons that related to dissatisfaction with order sets were that nurses could
not clearly tell what orders to place were within their scope, and the physicians disliked having to deal with purely nursing orders. Mixing nursing orders with orders that need to be initiated or signed by the physician makes most order sets too long and too cumbersome for easy utilization by users. Given both regulatory needs and the input from users, a decision was made to break the normal newborn orders into three sets (time-sensitive set, nursing scope set, and physician set) that work together instead of one massive set. The CDS team did a statistical analysis of the different normal newborn sets across the system to identify commonalities, delineate differences, and determine the usage of various orders. An analysis was done of the established evidence and best practices related to normal newborn care to determine whether any gaps existed in the current orders that could be addressed by the new sets. Because the time-sensitive orders need to be placed and acted on potentially before a pediatrician has examined the infant, they had to be designed as an SDO. State and federal rules indicate that SDOs must be reviewed by medical, nursing, and pharmacy leadership at least annually and be tied to a policy to be valid.The development of the time-sensitive set corresponded to the creation of a system-wide policy, and the Nl team ensured the policy and orders aligned, as well as that the policy and order set were approved by all necessary committees. Both teams (CDS and Nl) vetted the content and workflow for all three sets with the end-users and system leadership, a process that took many months as each revision had to go back to all three groups for validation. The physicians were ultimately responsible for the content of the orders that required their validation and signature.The nurses and nursing leadership ensured the orders on the nursing action set were within the scope of a nurse's practice without needing authorization from a physician. The physician set contains only orders needing physician initiation or authorization that were not already ordered on the time-sensitive set. This set can be used directly by the physicians or can be done as a telephone order by the nurse with the physician if necessary. The design of the sets allowed for the correct orders to be placed at the correct time by the correct discipline. The design also supported already existing workflows surrounding the care of the normal newborn infant and is flexible enough to be used easily by both large and small hospitals within the system.The sets utilize new functionality available for conditional orders, which allows the clinician to choose
19: CLINICAL DECISION SUPPORT SYSTEMS
the correct order that is dependent on other assessment or history Information,
giving very clear direction to the staff on why the order is indicated for the patient. Outcome:Jhe outcome from the launch of the normal newborn order sets was
overwhelmingly positive. Nurses stated that they felt much more confident that they were ordering appropriately to stay within their scope of practice. Physicians were very satisfied with the decrease in calls for orders. A minor glitch related to common practice versus recommended practice was identified surrounding orders for holding cord blood for additional testing. Many of the physicians were used to this being done automatically without orders and were unhappy when the process stopped because no order had been placed by the nurse.This concern was corrected by including an order to hold cord blood in the nurses' order set. If the physician wished to order any additional tests, the cord blood would be available.
CASE STUDY 4
COVID-19 and Infectious Disease Screening/Intervention Supported
by Clinical Decision Support Problem: The COVID-19 virus is no longer contained to China.The evolution of the virus to a global pandemic called for new Infection prevention measures to be developed and instituted across the world.This new infectious disease potential in the United States necessitated the CDC and the Departments of Health in each of the states to create new screening tools and subsequent infection prevention measures. Completely new isolation and exposure precautions were mandated for any potential exposure to or risk of COVID-19. All healthcare providers must follow the new screening guidelines to protect patients and address public health concerns. The new screening criteria, isolation type, and required subsequent actions need to be integrated into the EHR.The focus is on protecting the patient, the staff, and public health. Identification and containment are paramount. Hospital leadership and infection prevention staff want to leverage the EHR to help meet the new federal and state criteria related to the COVID-19 virus.
Stakeholders: Stakeholders in the process are the infection prevention staff, all clinical staff in all areas (inpatient, emergency, ambulatory, outpatient clinics), CDS team, EHR builders, Nl team, hospital and system leadership, health information management (HIM), legal, quality/risk department, regulatory reporting, and compliance department. Design: The Nl team leads the initiative to build and implement a new screening tool, orders, and necessary documentation for the staff to safely care for the patient, themselves, and public health. The N! coordinated with infection prevention and HIM to ensure the EHR builds to match the paper screening tool being used for areas that do not use the EHR.The screening tool was developed to screen for any newly emerging disease as well as known contagious diseases such as viral hemorrhagic fever. Middle Eastern respiratory syndrome, corona virus (MERS-CoV), tuberculosis, and others. Development also included the ability to rapidly change the screening questions and alerts depending on which most prominent threat to public health develops.
497
498
II: DATA MANAGEMENT
The design was intended to be broader than looking for just one disease and being flexible in response. Because of the critical need to identify any potential exposure or risk, the screening questions were made hard stops so that the staff had to complete the entire screen. CDS tools were employed to reduce the number of questions asked that are dependent on the answers to the first three questions. If the response to any of the required questions was positive, additional hard-stop questions would appear for the clinician to answer. CDS tools for alerts were utilized to give the clinicians reinforcement when the screen was successfully completed so they knew they were done and to alert the staff that further actions were needed. If further actions were needed and actionable, BPA was built that would allow the clinician to do what was needed with minimal
clicks and eliminated any guesswork. New orders for the new kind of isolation had to be built, as did the new alerts. Links to the CDC and state websites for
infection prevention are available for staff to use. For the infection prevention staff
to follow and manage any potential emerging disease patient, a new tool was built for them as well as for new reports for monitoring and compliance. Because communication among caregivers is key to safety, if the patient screened positive, alerts were fired throughout the record to ensure any clinician participating in care would be aware of the patient's infectious disease status. Great care was taken in developing tools for publicly viewable status boards, so that patient privacy and confidentiality were protected. CDS safety-net features were employed to prevent any patient from being dismissed without being screened by not allowing discharge instructions to be printed. The screening tool and associated actions were tested for usability by frontline staff, both nurses and physicians. Once the workflow and documentation elements were approved by infection prevention and hospital leadership, the measures were deployed. Reporting on any aspect of care related to COVID-19 has been very challenging as the different agencies such as the CDC, Food and Drug Administration (FDA), state, and local health departments all require different information reported to them. Many of the required reporting elements have been included in the design for the new screening tool, as well as In the COVID-19 testing orders and other documentation elements related to COVID-19, such as COVID convalescent plasma therapy and COVID monoclonal antibody therapy documentation. Outcome: The deployment of the new tool made the system compliant with the new CDC and state recommendations. Staff feedback was very positive, including comments about how easy it was to use the tool, and the staff felt confident they knew the steps to take if a positive screening waereto occur. Because the screening was developed as more than just a COVID-19 screening, it was easy for staff to adopt into their workflow. As the COVID-19 virus information is updated, such as what qualifies as exposure or time frames for needing isolation, the screening tool is updated to be current with CDC and other regulatory guidelines. A coordinated effort from all teams contributing to the COVID-19 response helped to make the required reporting to state and federal agencies much more reliable.The build efforts went beyond the screening tool and impacted all areas of the EHR, including inpatient, ambulatory, surgical, and emergency department documentation workflows. The leadership gave very positive feedback on the coordination and standardization of the tools and actions that were to span the EHR and paper workflows.
19: CLINICAL DECISION SUPPORT SYSTEMS
SUMMARY
This chapter has covered the basics of CDS definitions, types of CDS tools, and the essential elements that organizations should have in place for ensuring a successful CDS program. These elements include leadership and exeaitive support, as well as interprofessional teams representing the stakeholders most impacted by changes to workflow. These teams are fundamental to helping design the CDS strategies. The alignment of the CDS program with fundamental strategies for QI has also been discussed, along with the importance of involving departments that can help with optimizing the overall impact of the CDS intervention on patient safety and quality. The life cycle intervention was reviewed with important steps outlined, and concerns with liability for clinicians and hospitals have been discussed. Finally, four case studies were presented. END-OF-CHAPTER
RESOURCES
EXERCISES AND QUESTIONS FOR CONSIDERATION
Considering the four case studies presented, let us outline the life cycle intervention for each of the programs and consider the following questions: 1. Identify the area that will be impacted by the intervention. Be specific in terms of major stakeholders involved and address management and executive leadership buy-in. Why is it important to consider executives and management for CDS interventions? 2. Examine the design and validate, develop, test, and deploy processes for each of the four case studies. How was the intervention deployed within each of the case studies? Which of the tools discussed in the chapter were used within each of the case studies?
3. In examining the four case studies, how effective were the interventions? Support your position with evaluation measures noted in each of the case studies. Are they qualitative or quantitative approaches? Could the team have improved its evaluation strategies? If so, how? If not, why not? 4. Utilize the Five Rights Checklist in Table 19.1 and evaluate one or more of the four cases to determine if these organizations addressed all five rights. ADDITIONAL RESOURCES
CONNECT'
A robust set of instructor resources designed to supplement this text is located at http://connect.springerpub.com/content/book/978-0-8261-8526-6. Qualifying instructors may request access by emailing textbook@springerpub. com.
REFERENCES
Agency for Healthcare Research and Quality. (2019). Abnormal blood glucose and type 2 diabetes mellitus: Part one, screening. CDS.AHRQ. https://cds.ahrq.gov/cdsconnect/a rtifact/abnormal-blood -glucose-and-type-2-diabetes-mellitus-part-one-screening Bates, D. W., Kuperman, G. ]., Wang, S., Gandhi, T, Kittler, A., Volk, L., Spurr, C., Khorasani, Tanasijevic, M., & Middleton, B. (2003). Ten commandments for effective clinical decision support; Making the practice of evidence-based medicine a reality, journal of the American Medical Informatics Association, 20(6), 523-530, https://doi.org/10,1197/jamia,M1370
499
500
M: DATA MANAGEMENT
Berner, E. S. (2009), Clinical decision support systems: State of the art (No. 09-0069-EF). Agency for Healthcare Research and Quality, https://www.healthit.ahrq.gov/sites/defaul t/files/docs/page/09-0069 -EF_l.pdf
Capan, M., Hoover, S., Miller, K. E., Pal, C., Glasgow, J. M., Jackson, E. V., & Arnold, R. C. (2018), Datadriven approach to early warning score-based alert management. BMf Open Quality, 7(3), e000088 https://doi.Org/10.1136/bmjoq-2017-000088 Centers for Disease Control and Prevention. (2011). Million Hearts: Strategies to reduce the prevalence of leading cardiovascular disease risk factors—United States, 2011. Morbidity and Mortality Weekly
Report, 60(36), 1248-1251. https://www.cdc.gov/mmwr/preview/mmwr html/mm6036a4.htm Clinfowiki.org. (2015). Clinical decision support—History, http; / / www.clinf0wiki.0r2/wiki/index.php/
CDS#History Department of Health and Human Services Centers for Medicare and Medicaid Services. (2016). 42 CFR parts 414,416, 419,482,486,488, and 495 [CMS-1656-FC and IFC] RIN 0938-AS82. Federal Register, 56,
1159,1194. https://s3.ama2onaws.com/public-inspection.federalreg ister.gov/2016-26515.pdf eHealthUniversity. (2014, September). Clinical decision making: More than just 'alerts' tipsheet. https: / /
www.cms.gov/regulations-and-guidance/legislation/EHRincentivepro grams/downloads/
clinicaldecisionsupport_tipsheet-.pdf Greenberg, M., & Ridgely, M. S. (2011). Clinical decision support and malpractice risk. Journal of the American Medical Association, 306(1), 90-91. https://doi.0rg/lO. lOOl /jama.2011.929 Health Information and Management Systems Society. (2011). So you wan f to do CDS: A C-level in troduction to clinical decision support (Webinar Slides No. HIMSScds201lX Author. Health Information and Management Systems Society, (n.d.-a). CDS 101: Fundamental issues. Retrieved August 30, 2021, from http://www.himss.org/library/cli nical-decision-support/issues? navItemNumber=13240
Health Information and Management Systems Society, (n.d.-b). Wljat is clinical decision support? Retrieved August 30, 2021, from http://www.himss.org/library/clinical-deci sion-support/what-is?navItem Number=13238#promi.seperil Institute of Medicine. (2011). Health IT and patient safety: Building safer systems for better care (Consensus Report). National Academies Press. Kendall, K., & Kendall, J. (2014). Systems analysis and design (9th ed.). Pearson. Kesselheim,A. S., Cresswell,K., Phansalkar,S.,Bates, D. W., & Sheikh, A. (2011), Clinical decision support systems could be modified to reduce 'alert fatigue' while still minimizing the risk of litigation. Health Affairs, 30(12), 2310-2317. https:/ /doi.org/10.1377/hlthaff.201 0.1111 Lieu, G., Ray, T., Klein, N., Chung, C., & Kulldorff, M. (2015). Geographic clusters in underimmunization and vaccine refusal. Pediatrics, 135(2), 280-289. https://doi.or g/10.1542/peds.2014-2715 McBride, S., Tietze, M., Hanley, M., & Thomas, L. (2017). Statewide study to assess nurses' experiences with meaningful use-based electronic health records. Computers, Informatics, Nursing, 35(1), 18-28. https://doi.Org/10.1097/ClN.0000000000000290 National Quality Forum. (2008). National z’oluntary consensus standards for influenza and pneumococcal hnmunizations. http://www.qualityforum.org/Publications/2008/12/ National_Voluntary_Consensus _Standards_for_lnfluenza_and_Pneumococcal_lmmunizations.aspx NLM.gov. (2021). NLM imiue set authority center (VSAC); Welcome. NLM.gov. https://vsac.nlm.nih.gov/ Office of the National Coordinator for Health Information Technology. (2018). Clinical decision support.
https: / /www.healthit.gov/topic/safety/clinical-decision-suppor t Osheroff, J. A., Teich, J. M., Levick, D., Saldana, L„ Velasco, F. T, Sittig, D. F., & Jenders, R. A. (2012).
Improving outcomes with CDS: An implementer's guide (2nd ed.). Healthcare Information and Management Systems Society. Phansalkar, S., Edworthy, J., Hellier, E., Seger, D. L., Schedlbauer, A., Avery, A. j., & Bates, D. W. (2010). A review of human factors principles for the design and implementation of medication safety alerts in clinical information systems. Journal of the American Medical Informatics Association, 27(5), 493-501. https: / / doi.org/10.1136/jamia.2010.00526 Texas Health Resources. (2014). Vaccination protocol (No. THRvaccine2014). Author. Tushan, M. (2012). Weekly webinar series for oi^ercoming meaningful use barriers: Solutions from the field for Million Hearts introduction, REC experiences, and implementation into workflow (Presentation No. ONC2012). Office of the National Coordinator for Health IT.
Wright, A., Bates, D. W., Middleton, B., Hongsermeier, T, Kashyap, V, Thomas, S. M., & Sittig, D. F. (2009). Creating and sharing clinical decision support content with Web 2.0: Issues and example. Journal of Biomedical Informatics, 42(2), 334-346, https://doi.Org/10.1016/j.jbi.2008.09.003
PATIENT SAFETY/QUALITY AND POPULATION HEALTH
Health lnfornnati®nTechnolagX, and Implicatioris^r Patletit Safety MARI TIETZE AND SUSAN MCBRIDE
OBJECTIVES ●
Discuss the implementation of the Patient Safety Act and the relationship of that Act to the national agenda to implement electronic health records {EHRs) with rapid deployment methods.
● ●
Discuss patient safety organizations (PSOs) and what they are designed to accomplish. Describe the roles of the Agency for Healthcare Research and Quality (AHRQ) as the overseer for Health InformationTechnology (ONC).
of PSOs and the Office of the National Coordinator
●
Discuss the history of the lOM report and its impact today.
●
Outline how the Future of Nursing (FoN) report 2020-2030 addresses patient safety associated competencies.
●
Explain use of SAFER guidelines development, maintenance, and use.
●
Explore the safety needs and associated competencies in the delivery of teleheaith services.
●
Discuss the importance of having an interoperability approach for safety.
●
Describe the relationship between the EHR and patient safety.
●
Explore how to optimize EHR downtime preparedness to maintain patient safety.
●
Define actions that can be taken by healthcare professionals to report patient safety issues appropriately. CONTENTS
INTRODUCTION
504
SAFETY HISTORY RESULTING INTHE PATIENT SAFETY ACT
505
PATIENT SAFETY ISSUES AND HEALTH INFORMATIONTECHNOLOGY
Models of Unintended Consequences
508
Operational Model forThree Domains of Expertise
509
Harrison's Interactive Sociotechnical Analysis Model
Guide to Unintended Consequence Management
509
510
Hazard Manager by the Agency for Healthcare Research and Quality Emergency Care Research Institute Annual Report of Errors
511
510
506
504
IV: PATIENT SAFETY/QUALITY AND POPULATION HEALTH
Clinical Impact of Health InformationTechnology Error/Unintended Consequences Downtime and Disaster Preparedness Procedures and Protocols Telehealth Care and Associated Patient Safety Needs
513
515
516
ACTIONS FOR WHEN HEALTH INFORMATIONTECHNOLOGY PATIENT SAFETY ISSUES ARISE 516
Data Collection and Reporting
Levels of Reporting NationalTrend
516
516
518
STRATEGIESTO MITIGATE HEALTH INFORMATIONTECHNOLOGY PATIENT SAFETY ISSUES
518
Multifaceted Options to Optimize Safety Healthcare Delivery Systems TeamSTEPPS Model
520
520
521
The Leapfrog Group, an Employer-Based Safety Initiative
522
Technology Informatics Guiding Education Reform Leadership to Transform Education
522
Remediating Unintended Consequences of Health InformationTechnol ogy Pinpoint the Cause
523
523
Prioritize, Plan, and Execute the Remediation for a Health Information
Technology Patient Safety Issue
524
Statewide Nursing Approach for Error Mitigation Ethics of Decision Making CASE STUDY SUMMARY
525
526
528 528
EXERCISES AND QUESTIONS FOR CONSIDERATION REFERENCES
529
530
APPENDIX 20,1 HIMSS PATIENT ENGAGEMENT NETWORK SCHEMATIC
534
INTRODUCTION
Rapid deployment of electronic health records (EHRs), health information exchanges, and their support technology have created fertile ground for patient safety issues to arise. This chapter discusses national issues concerning patient safety and quality related to the rapid deployment of health information technology (HIT). Also, national strategies to address patient safety under the Patient Safety and Quality Improvement Act of 2005 (Public law 109-41 ,2005) are reviewed, and the healthcare professional's role is emphasized as to when and how to act on patient safety events related to technology. What are these unintended consequences? EHRs can offer many benefits to healthcare providers and their patients, including better quality of medical care, greater efficiencies, and improved patient safety. Even if these benefits are achieved, one will almost certainly face some unanticipated and undesirable consequences from implementing an EHR. These consequences are often referred to as unintended consequences (Campbell, Sittig, Ash, Guappone, & Dykstra, 2006). They can undermine provider acceptance, increase costs, sometimes lead to failed implementation, and result in harm to patients. However, learning to anticipate and identify unintended consequences promotes effective decisions, clarifies trade-offs, and addresses problems as they arise (Campbell et ah, 2006).
20: HEALTH INFORMATION TECHNOLOGY AND IMPLICATIONS FOR PATIENT SAFETY
One example of work conducted on the topic of unintended consequences of IT is reflected in the study by Sittig and Ash (2007). Their study collected unintended adverse consequences of EHRs from five hospitals representing 2,346 beds having implemented an EHR system, specifically computerized provider order entry (CPOE). Practitioners provided their experiences associated with the implementation, and themes emerged that helped answer the question as to what were some examples of unintended consequences (Sittig & Ash, 2007). Nine common examples follow later in this chapter. SAFETY HISTORY RESULTING IN THE PATIENT SAFETY ACT
The safety of patients in American hospitals was called into question well over a decade ago after the release of To Err Is Human: Building a Safer Health System, a report by the Institute of Medicine (lOM) detailing the number of patients harmed in hospitals (Kohn etal., 1999). According to the report, more people die in a given year as a result of medical errors than from motor vehicle accidents, breast cancer, or AIDS (Kohn et al., 1999). This
report, among others, .such as Crossing the Quality Chasm: A Nezo Health System for the 21st Century (Committee on Quality of Healthcare in America, 2001), created a national focus on safety in healthcare organizations. APRNs are in a unique position to prevent errors by raising awareness about and understanding of how and when they occur and leading teams to implement improvements to address quality concerns. According to the classic book titled Keeping Patients Safe: Transforming the Work Environment ofhlurses (Page, 2004), nurses' proximity to and continual ob.servation of the patient places the profession in a position to prevent errors before the patient is impacted. An lOM report. The Future of Nursing: Leading Change, Advancing Health, further emphasizes the importance of nursing's role in designing safe systems that work well for patients (Cipriano, 2011). The report emphasizes nurses' proximity to the patient, as well as the magnitude of nurses' impact as "the largest segment of the health care workforce with some of the closest, most sustained interactions with patients" (Cipriano,
2011, p. 143). In 2021, the follow-up report, Future of Nursing 2020-2030: Charting a Path to Achieving Health Equity, was created (National Academies of Sciences, Engineering, and Medicine, 2021). This report, written by a team of subject matter experts, emphasized the issues of health inequity, inclusiveness, and environmental natural resources preservation. Sponsored by Robert Wood Johnson Foundation, this work builds on the foundation set by The Future of Nursing: Leading Chajige, Advancmg Health (lOM, 2011). Nurses have a critical role to play in achieving the goal of health equity, but they need robust education, supportive work environments, and autonomy. The ultimate goal is the achievement of health equity in the United States built on strengthened nursing capacity and experti.se. By leveraging these attributes, nursing will help to create and contribute comprehensively to equitable public health and healthcare systems that are designed to work for everyone. The identification of this equitable public health and healthcare system includes being able to manage, analyze and interpret data, such as the social determinants of health (National Academies of Sciences, Engineering, and Medicine [NASEM], 2021, p. 2). Addressing the need to focus efforts on improving patient safety in the United States, Congress passed the Patient Safety Act and Quality Improvement Act (2005), which launched a national push for patient safety organizations (PSOs). PSOs were created for healthcare organizations to share information on patient safety events without the fear that the information might be used against them in a lawsuit. By providing confidential mechanisms in a secure environment, information on patient safety events can be collected, aggregated, and analyzed to help develop approaches to systematic
505
506
IV: PATIENT SAFETY/QUALITY AND POPULATION HEALTH
errors that aid the health delivery system (AHRQ, n.d.-b). Given this role, PSOs are an important avenue for nurses to consider when an event occurs that either injured or might have potentially injured a patient (near miss). Many healthcare organizations are associated with PSOs or work closely with a PSO to report patient safety events. Nurses should be aware of reporting mechanisms in place for their institutions, particularly as we move rapidly to deploy EHRs that impact so many critical elements of the patient care process.
PATIENT SAFETY ISSUES AND HEALTH INFORMATION TECHNOLOGY
In 2011, the lOM Committee on Patient Safety and Health Information Technology released a report titled Health IT and Patient Safety: Building Safer Systems for Better Care (lOM, 2012). The report outlined observations of potential HIT errons, concluding that several factors must be taken into consideration with implementation to prevent these errors from occurring. The factors included:
■ Implementation strategies (fast or slow progression onto an EHR) ■ The influence of end users regarding the configuration of the EHRs and training of clinicians
■ Workflow using the paper record versus the new electronic system ■ Availability of data for the analysis of quality (lOM, 2012) Recognizing that evidence was mounting with health IT safety and unintended consequences (Singh, Wilson, et al., 2009; Myers et al., 2011), in 2014 the Office of the National Coordinator for Health Information Technology (ONC) commissioned RTI International to convene a HIT safety task force comprised of subject matter experts to advi.se national direction to address the challenges with safe and effective use of HIT (RTI International, 2014). The task force resulted in recommendations for a HIT
safety center and developed a road map entitled Health IT Safety Center Roadmap (RTI International, 2015). The full road map can be downloaded from the following website: www.healthitsafety.org. The authors encourage a full review of the recommendations and strategy to establish a PSO. The intent of the center is to capitalize on existing infrastructure nationally with the PSOs and convene safety researcher.s, vendors, and end users within the center to share resources and best practices for solutions with the ultimate goal of making HIT safer. The road map consists of moving from the current environment to development of the Health IT Safety Center to the ultimate Health ITenabled Learning Health system. From this work, as well as landmark research, several models for safe HIT practices have evolved, as well as a SAFER (Safety Assurance Factors for EHR Resilience) Guides for assessing organizations' patient safety practices (ONC, 2018). We will review a sample of these models beginning with an examination of the SAFER Guides and the intended purpose. This review is not intended to be comprehensive but instead a brief discussion of the types of models that can be helpful in addressing HIT safety and unintended consequences.
The SAFER Guides provide self-assessment tools for healthcare facilities to utilize for heightening their EHR safety (ONC, 2018). They were designed with a "multifaceted system-based approach" (Sittig & Singh, 2015, p. XVIII). The self-assessments can be utilized by a variety of audiences (internal and external stakeholders) to identify gaps and
20: HEALTH INFORMATION TECHNOLOGY AND IMPLICATIONS FOR PATIENT SAFETY
promote quality improvement initiatives for safety of the EHR. The SAFER Guides are noted in Table 20.1 and have three categories: foundational guides, infrastructure guides, and clinical process guides.
Since the period when To Err Is Human (Kohn et al, 1999) was published, much has been learned about the epidemiology of safety. Despite progress, for example, in
hospital-acquired infections and medication safety, there remains substantial chances for improvement (Bates & Singh, 2018). Another example where improvement is warranted is the "Never Events," such as wrong patient and wrong-site surgery that still occur with unacceptable frequencies. For the future, the focus is on tools and strategies that allow organizations to measure and reduce harm consistently and continuously. As this is achieved, the policies must encourage and, as needed, "require" use of such tools (Bates & Singh, 2018). Notably, in the 2022 Fiscal Year Centers for Medicare & Medicaid Services (CMS) guidelines, hospitals must "attest to having completed an annual assessment of all nine guides in the SAFER Guides measure, under the Protect Patient Health information objective" (Centers for Medicare & Medicaid Services, 2021, p. 1). In 2018, Sittig et al. conducted a stmctured eight-organization risk assessment which demonstrated that many of the SAFER Guides recommendations were lacking in implementation (Sittig et al., 2018). Their study highlighted the need for initiatives through national policy to further promote implementation of the guides. Sittig etal. (2020) acknowledge challenges in HIT and patient safety due to the complexities of healthcare systems of care and HIT systems. The EHR has created unintended consequences and safety challenges; incorporating the SAFER Guides at the appropriate stages of the system lifecycle can improve safety (Sittig et al., 2020). A recent study supported by the American Nursing Informatics Association (ANIA) study on the SAFER Guides to assess the utilization of the SAFER Guides by nursing strategies to improve their utilization and application found poor utilization and awareness rates in the nursing informatics leaders across the United States and inform
informatics community. The study did find that the higher the level of education the more likely the nursing informatics professional was aware of the guides. This study has resulted in ANIA focusing on improving awareness and use of the guides by the nursing informatics community of professionals. ANIA will be working on a toolkit including top areas of interest for nursing-sensitive SAFER Guides beginning with a focus nursing and EHR downtime (McBride et al., 2020 August 20). TABLE 20.1 The SAFER Guides Categories SAFER GUIDES Foundational Guides
High Priority Practices Organizational Responsibilities
Infrastructure Guides
Contingency Planning System Configuration System Interfaces
Clinical Process Guides
Patient Identification
Computerized Provider Order Entry With Decision Support Test Results Reporting and Follow Up Clinician Communication
Note:The SAFER Guides consist of nine guides organized into three groups.The SAFER Guides can be accessed at https;//www.healthit.gov/topic/safety/safer-guides.
507
508
IV: PATIENT SAFETY/QUALITY AND POPULATION HEALTH
Models of Unintended Consequences Unintended Adverse Consequences
Much of the early research on HIT safety involved studying healthcare professionals, such as physicians and nurses, during and after the implementation of HIT systems in their organization. Key observations and recommendations, based on these studies, are reflected
in the book titled Clinical Infonnation Systems: Overcoming Adverse Consequences (Sittig & Ash, 2007). The study conducted by Campbell et al. (2006) using qualitative methods and an expert panel to gather and analyze examples of five successful CPOE sites also provided direction. Nine unintended adverse consequences of the CPOE implementations evolved from this aimulative research. The unintended consequences apply not only to CPOE implementation but also to other HIT systems such as EHRs. The nine unintended consequences are briefly summarized here: 1.
More zvork for clinicians. Example: After the introduction of an EHR, physicians often have to spend more time on documentation because they are required to (and facilitated to) provide more and more detailed information than with a paper chart. Although this information may be helpful, the process of entering the information may be time consuming, especially at first.
2.
Unfavorable workfloiv changes. Example: The CPOE automates the medication- and test-ordering process by reducing the number of clinicians and clerical staff involved, but by doing so it also eliminates checks and counterchecks in the manual ordering process. That is, with the older system, nurses or clerks may have noticed errors, whereas now the order goes directly from the physician to the pharmacy or lab. Never-ending demands for system changes. Example: As EHRs evolve, users rely more heavily on the software and demand more sophisticated functionality and new features (e.g., custom order sets). The addition of new functionalities necessitates
3.
that more resources be devoted to EHR implementation and maintenance. 4.
5.
Conflicts between electronic and paper-based systems. Example: Physicians who prefer paper records annotate printouts and place these inpatient charts as formal documentation, thus creating two distinct and sometimes conflicting medical records. Unfavorable changes in communication patterns and practices. Example: EHRs create an "illusion of communication" (i.e., a belief that simply entering an order ensures that others will see it and act oii it). For example, a physician fails to speak with a nurse about administering medication, assuming that the nurse will see the note in the EHR and act on it.
6.
Negative user emotions. Example: Physicians become frustrated with hard-to-use software.
7.
Generation ofneiv kinds of errors. Example: Busy physicians enter data in a miscellaneous section, rather than in the intended location. Improper placement
can cause confusion, duplication, and even medical 8.
error.
Unexpected and unintended changes in institutional power structure. Example: IT, quality-assurance departments, and the administration gain power by requiring physicians to comply with EHR-based directives (e.g., clinical decision support [CDS] alerts).
9.
Overdependence on technology. Example: Physicians dependent on CDS may have trouble remembering standard dosages, formulary recommendations, and medication contraindications during system downtimes (Campbell et al., 2006; Sittig & Ash, 2007).
20: HEALTH INFORMATION TECHNOLOGY AND IMPLICATIONS FOR PATIENT SAFETY
Operational Model for Three Domains of Expertise Interprofessional collaboration has been said to improve patient safety and quality outcomes and this also may be applied in the HIT environment (Interprofessional Education Collaborative, 2016). We have observed that there are three key players in the effort to manage HIT safety. They are as follows: 1. Patient safety and risk-management specialists 2. Quality-improvement specialists 3. Nursing informaticists
All three domains of expertise have made unique contributions to the end goal of improved care through EHRs and interoperability. In many organizations, these operational models work on separate departments that are not always tightly aligned. Practical implications for department independence are an operational model that is vital to successful HIT implementations. Interdepartmental expert contributions that support interoperability are as follows: ■ Quality departments maintain quality assurance and peer review protection. ■ Patient safety information has historically been maintained in risk-management departments as a result of the litigious nature of the information relating to a patient safety event resulting in injury or death.
■ HIT deployment is managed by the IT departments of most institutions. Quality departments maintain committee structure and operational procedures within healthcare organizations to address peer review protections with differing laws depending on the state policy. Patient safety information is particularly sensitive information to organizations, and at the same time, we have rapid deployment of EHRs resulting in the potential for unintended consequences. For good reason, these three departments—patient safety, quality improvement, and nursing informatics—often operate independently of one another and have different specialists with expertise in these three areas worldng from their particular domain of expertise. Harrison's Interactive Sociotechnical Analysis Model Many unintended and undesired consequences of HIT flow from interactions between the HIT and the healthcare organization's sociotechnical system. Factors such as workflows, ailture, social interactions, and technologies are affected. Harrison and colleagues present a conceptual model of these proces.ses they call interactive sociotechnical analysis (ISTA). ISTA is said to capture common types of interaction with special emphasis on recursive processes, such as feedback loops, that alter the newly introduced HIT and promote second-level changes in the social system. ISTA draws on prior studies of unintended consequences, along with research in sociotechnical systems, ergonomics, social informatics, technology-in practice, and social constmction of technology. The ISTA model provides a guide for further research on emergent and recursive processes in HIT implementation and their unintended consequences. Familiarity witli the model can also foster practitioners' awareness of unanticipated consequencesthatbecomeevidentonly during HIT implementation (Harrison et al., 2007). Specifically, five interactive components of the ISTA model are as follows: 1. New HIT changes the existing social system 2. Technical and physical infrastaicture mediated HIT use—the interaction of new HIT with existing technical and physical conditions affects HIT use
509
510
IV: PATIENT SAFETY/QUALITY AND POPULATION HEALTH
3. Social system mediates HIT use—the interaction of new HIT with social system affects HIT use
4. HIT use changes the social system—the interaction of new HIT with social system affects HIT use, which then further changes the social system 5. HIT-social system interactions engender HIT redesign—the interaction of new HIT with the social system affects HIT use, which then leads to changes in HIT properties (Harrison et al., 2007)
As noted in Chapter 3 (see Figure 3.1), the arrows in the ISTA schematic show the impact of one sociotechnical subcomponent on another and correspond to the five interaction types (Harrison et al, 2007).
Guide to Unintended Consequence Management On the HealthIT.gov website, the ONC has created a number of resources, one of which is the Guide to Reducing Unintended Consequences of Electronic Health Records (RAND Corporation, 2011). The guide addresses all care settings and notes a number of common patient safety issues related to the unintended consequence of HIT. This online guide for healthcare providers, IT specialists, and system administrators helps in planning and avoiding possible problems when implementing and using an EHR. The guide was developed to provide practical knowledge and resources for all types of healthcare organizations. '"The guide is based on the research literature, other practice-oriented guides for EHR implementation and use, research by its authors, and interviews with organizations that have recently implemented EHR" (RAND Corporation, 2011, para 3). Bates et al. provided the guide and it is organized into four modules;
1. Introduction to unintended consequences 2. How to avoid unintended consequences 3. Understand and identify unintended consequences
4. Remediate unintended consequences (covered later in this chapter)
Hazard Manager by the Agency for Healthcare Research and Quality The Hazard Manager is a federally funded project focused on developing and testing a software tool to capture and manage information about prospectively identified HIT hazards before they have the potential to cause harm. Rather than looking retrospectively at accidents or near misses, this tool is designed to collect structured information about potential hazards associated with specific HIT products. Figure 20.1 depicts the HIT Hazard Manager Database form that is available online. There are four main categories of hazard attributes: 1. Discovery 2. Causation
3. Impact
4. Mitigation/corrective action
20: HEALTH INFORMATION TECHNOLOGY AND IMPLICATIONS FOR PATIENT SAFETY
FIGURE 20.1 HIT Hazard Manager Database. 20j
HIT Hazard Manager
t
AdffHn ●
^
Flrp« "
MyAccovnt* NftI «■
m«y
M^cn Ofilei A{ « KdXAt tl. I
i.Dtsv^t«n
^.SjVtmilnvohtd O I49«»
IIk
iZhHk
CeHabwatIveCn
■ Osiropractk
● Dentiny
-
interepenbk RKOrds
Celaborative Care
H
C
admirustralive data
tortg-term
c
“O
Inlegiated with claims and
Acute
-
Big data, hadoop, rise of GPUs’ for AI, efficient
Storage, faster computing, loT miniaturization,
networks everywhere, data science skills, public cloud emerges
Perfect storm
Open source algorithms and frameworks, high performing cost-effective computing,
2020
History of Artificial
advanced data science skills,
more data than ever before,
FIGURE 27.1
,,
AI solves complex problems
Intelligence Transformation.
As a result of technological advances, we have storage and computing speed; it is reasonable that we can ask and answer nearly any question we might conceive at negligible costs given our advanced analytics and capability with technology. However, to do so means we must learn how to discover new knowledge by the way we ask questions. As John Naisbitt indicated in his 1982 book Megatrends, "We are drowning in data and starving for knowledge" (Naisbitt, 1982, p. 17). This is certainly the case in the healthcare industry today and is expected to get much more profound as we contemplate full use of genomic and epigenetic data. We have in some respects, "the perfect storm." Reflecting on the history we see that in the 1950's math, statistics and algorithms were in use. By the 1980s expansive computing and early AI was emerging supported by expert systems and lules-based engines. By 2000, central processing units (CPU) and storage of enterprise data had gotten much cheaper, and the internet was exploding with all kinds of devices. A new term emerged from this connectivity. The Internet of Things (loTs). By 2010, big data and new data management tools such as Hadoop were on the scene as well as graphic processing units (GPUs) with speed of processing for AI and better and more efficient storage of data. The loT was getting smaller with more and more mobile and digital devices, and data science skills were developing. By 2020, we have open-source algorithms and frameworks, highperforming and lower cost computer with advanced analytic skills rapidly developing across industries. AI is poised to solve complex problems and help healthcare innovate new and better solutions (Ray, 2018). BIG DATA AND AI USE CASE CATEGORIES
Healthcare is one of the industries pushing the constraints of using traditional methods to
manage and analyze data. The masses of unstructured textually rich data within the EHR
687
688
V: NEW AND EMERGING TECHNOLOGIES
are among the prime examples of data that constitute big data. Mayer-Schonberger and Cukier (2013) explain the phenomenon of big data as "things one can do at a large scale that cannot be done at a smaller one to extract new insights or create new forms of value" (p. 6). Genomics and epigenetics are examples of big data creating substantial value. Additionally, the (loTs is exploding with data and information, including such things as mobile health devices, wearables tracking health and wellness data, nanotechnology, and other mechanisms connecting, sensing, and collecting massive amounts of data (Clancy, 2017). Data sources in healthcare that collect massive amounts of data have several features
that have been worthy of examination. These features include massive parallelism (common record types such as an EHR entry for a patient, or admission transaction discharge data) where the fields are synonymous from case to case. Each, to a greater or lesser degree, requires enormous data volume storage and distribution for the purpose of manipulation through analytics or for use in high-performance computing and data mining such as pattern or thread recognition and, more recently, for ML algorithms. Because these use cases and these common techniques have been around historically, why are these use cases so prominent now and what is new about these techniques? This is a result of the fact that we are collecting and storing more data because it has become affordable and easier to use primarily because of an increase in processing speed. Also, unlike a decade ago, open-source coding and new sources of information with widespread access to the internet are much more common and have led to much easier, more efficient orders of magnitude and more elegant solutions in using and managing big data. Finally, commodity hardware and software have made utility-level usage available to all. We examine techniques for using big data in healthcare and consider
some of these use cases in large healthcare systems for operational and clinical analysis to improve healthcare outcomes using advanced analytics software in the areas of cognitive computing and data mining. New trends in AI in healthcare have been in the works
for years and are now being put into operation. In the near future we can expect that AI will be able to help patients and healthcare professionals from the dispatch of first responders, to transportation to the medical facility, during emergency triage and interventions, through surgery and post-surgical recovery and intensive care, to physical tlierapy and home care. At every step of the patient's journey, AI is helping aid medical professionals to correctly diagnose and more effectively treat medical issues such as sepsis that can substantially benefit from the enhancements in accuracy, decision-making, and speed offered by AI. AI cannot replace human training, skills, and cognitive abilities. It can, however, simplify a host of complicated processes, routines, and tasks by rapidly synthesizing massive quantities of data into a form that can be easily processed by healthcare professionals for improved decision-making. BIG DATA AND AI IN HEALTHCARE At the forefront of healthcare in the United States is the unquestionable need to contain
costs while simultaneously improving quality. Healthcare costs represent approximately 18% of the U.S. GDP and have increased at a rate greater than the U.S. economy for 31 of the past 40 years (Executive Office of the President Council of Economic Advisers,
2009). Coupled with these rising costs is the unprecedented complexity of healthcare information, particularly in oncology where significant advances are being made in genomic sequencing, immunotherapy, and targeted therapies (Malin, 2013). Of significant concern is the expenditure of approximately $95 billion annually on medical research in
27; “BIG DATA" AND ADVANCED ANALYTICS
the United States, with only 6% of all clinical trials completed on time (American Cancer
Society, 2014). In 2012, the Institute of Medicine (lOM) issued a document outlining three charges that would help address these challenges. This initiative called for the following from healthcare providers and institutions: (a) using tools such as computing power in the form of big data and analytics; (b) improving connectivity; and (c) improving organizational capabilities and ensuring collaboration between teams of clinicians and with patients (lOM, 2012). In essence, the lOM (2012) emphasized the need for healthcare systems that provided rapid, real-time data for use in routine clinical care, comparative effectiveness research, quality improvement, safety, and generation of new hypothesis for investigation. These uses of data and information support the call by lOM for creating learning healthcare organizations (lOM, 2012, pp. 55-57). These recommendations set the stage for cognitive computing power to be introduced as a potential solution. Cognitive computing is defined by Malhotra (2018) as a collective intelligence of people combined with computers using intelligent systems in the context of anticipatory analytics, further defined by Jones (2017) as “applying knowledge from cognitive science to build systems that simulate human thought processes." Clancy differentiates ML as “the science and technology of systems that learn from data" (2017). These systems present challenges with adoption and implementation of technology that require nursing expertise and leadership given the impact on clinical and operational workflows often managed by nursing.
Improvement in healthcare delivery through technology has received support and reinforcement by legislation, such as the Health Information Technology for Economic and Clinical Health (HITECH) Act, passed in 2009 (U.S. Congress, 2009). In June 2014, as the HITECH Act moved into its fifth year of implementation, Dr. Karen DeSalvo, the national coordinator for health information technology (HIT) at the time, stated: "There's ^reat promise in what we can do with information, whether that's to improve systems around quality and safety or whether it's to advance science andjor cure and treatment for individuals zoith genomics at the bedside all the way to population-level advancements. There's azz array of opportunities for the use of the data. At the end of the day, it's the patient’s data ... and zoe have to get ahead of the privacy and security challenges that arc going to arise as Big Data gets more common." (Fluckinger, 2014, para. 10) Subsequently, in 2012, the Obama administration instituted the “Big Data Research and Development Initiative," which received $200 million of federal support (Jee & Kim, 2013, p. 81). The purpose of this national initiative was to maximize the use of big data (massive bodies of digital data) into information to inform science and discovery in biomedical research (Jee & Kim, 2013). Prior to the HITECH Act, most of our healthcare data in the United States had been
kept in silos within institutions,
clinics, insurance companies, and government agencies. Although we continue to face integration challenges, the HITECH Act and the accompanying financial EHR Incentives Program of the Centers for Medicare & Medicaid Services (CMS), which is aimed at providers adopting and implementing EHRs, have created the infrastructure needed to capture and store electronic data at unprecedented levels of detail. This is further coupled with data from genomics and epigenetics, thus propelling healthcare into the era of big data.
The Triple Aim framework supported by the Institute for Healthcare Improvement (IHI) calls for transformation of the healthcare industry and creates a need to embrace the use of big data, analytics, and innovative tools such as cognitive computing systems. The
goal of the Triple Aim is to “improve the patient experience of care, improve the health
689
690
V: NEW AND EMERGING TECHNOLOGIES
of the population, and to reduce the per capita cost of health care" (IHI, 2015). By using big data, healthcare has the opportunity to affect all three dimensions that are needed to improve the healthcare system.
With this emerging wealtli of information, and potentially superimposed, useless or misinformation, healthcare is beginning to identify how and where big data and advanced analytics can improve healthcare costs and help shape the reform of clinical patient care. In addition, big data analytics can serve to provide solutions in public health, such as the ability to predict future healthcare needs of a population or infectious disease outbreaks (Webster, 2014). Big data analytics may also be used to improve disease and anomaly detection by providing comprehensive information, trends, and comparisons that result in increased accuracy for initial, and subsequent, diagnoses. This could result in reduced medical errors such as those resulting from mis-/missed diagnosis or omissions of care. These types of utilities for big data and advanced analytics could ultimately translate to improved patient outcomes and reduced costs. Big data and cognitive computing analyses of data can deliver information in "real time" to clinicians. CONSIDERATIONS FOR USE OF A! IN THE CLINICAL DECISION SUPPORT
There are important considerations that clinicians should be aware of when considering how AI is being used within EHRs and other technologies supporting clinical care, in,lehealth, mobile applications and other digital point-of-care devices. The AI algorithms are often embedded into the technology and support clinical decision-making. Consider this scenario: An AI model is used in the EHR to alert you of a clinical course of action that does not seem to make sense to you. However, because the model cannot explain itself, you have no insight into the reasoning behind the recommendation . Your only options are to trust it or not, but without any context. The following are questions clinicians should consider when supporting the u.se of AI within technology: 1. How does AI software fit into the clinical workflow? 2. How does the AI software work?
3. What types of AI does this software use? 4. How much should I trust this recommendation made
by AI?
5. How was this AI software developed?
6. Who was involved with the development of the AI and how was it tested? BIG DATA AND NURSING KNOWLEDGE
In 2013, recognizing the importance of big data and knowledge generation to nursing practice, the University of Minnesota convened an invitational conference of nursing leaders, informatics subject matter experts, and researchers for the first Nursing Knowledge: Big Data Science Conference. Sub.sequently, the American Academy of Nursing supported a "call to action" for executives and nursing leadership to continue to convene and generate recommendations for important components of the action plan. This initiative has resulted in subgroups working on important components of the action and has matured over the last .several years with active working groups throughout the year. Components of the action plan are noted in Table 27.2 (Clancy etal, 2014).
27: “BIG DATA" AND ADVANCED ANALYTICS
TABLE 27.2 Key Components of the Action Plan for Nursing Knowledge: Big Data Science Conference
BIG DATA AND ANALYTICS NATIONAL NURSING INITIATIVE Action Plan
Components of the Action Plan
1
Develop a strategy/campaign for educating frontline nurses, students, and faculty on informatics competencies and the value of standardized nursing data.
2
Advocate for the adoption of Systematized Nomenclature of Medicine— ClinicalTerminology and Logical Observation Identifiers Names and Codes as national standards for clinical data, and link them with nursing terminologies through mappings.
3
Convene a consensus conference with leaders of the major nursing
organizations and interprofessional stakeholders to educate them, hear their views, and ultimately speak in one voice. 4
Refresh and activate the American Nurses Association's Nursing Information & Data Set Evaluation Center criteria to advance systems that
represent and value nursing data. 5
Continue bold participation in standards and EHR standards development to ensure a nursing voice.
EHR, electronic health record.
Source: Clancy,T, Bowles, K., Gelinas, L., Androwich, I., Delaney, C., Matney, S., Sensmeier, J., Warren, J., & Westra, B. (2014). A call to action; Engage in big data science. Nursing Outlook, 62, 64-65. https;//doi. org/10.1016/j.outlook.2013.12.006.
DATA MINING
Knowledge can be discovered by mining large data sets. In the era of gold mining there were three distinct phases of the process—prospecting for the vein, following the lode, and smelting (removing impurities) of the ore into refined gold—so too there are three similar phases in data mining. Often, in the Gold Rush era, prospectors would find nuggets of coarse gold, usually when a stream eroded a point of the vein and washed the nuggets downstream. Almost all the initial finds were by happenstance, and that is what led to major discoveries. In that era, thousands of tons of gold-containing material were leached to extract gold ore deposits that could then go to the smelter for refining and pouring into bullion bars.
In data mining, prospecting for the vein focuses on findings in the data—anomalies, correlations, patterns, or trends. Once these data nuggets are discovered, the focus is to mine the lode by determining trajectories that produce dimensionality that gives the data their characteristic features. This ultimately forms a picture of the data lode, which is a modeling function that has predictive or at least systematic valuation. Finally, the ore-rich material is smelted or processed by removing variation and sorting causation to eliminate the noise in the data system. Here, dependencies of the data that are useful to clarifying and shaping the picture are used, and graphic representations are often ways in which the data are portrayed.
Data Mining Defined Historically, data mining has been defined as mining data and information from large databases and is often associated with ML or advanced analytics techniques (Chen et al., 1996). Data mining is also defined as a method in computer science that is used to discover patterns and trends within large data sets. Data mining techniques contain many
691
692
V: NEW AND EMERGING TECHNOLOGIES
Specialized classifications and subclassifications involving various methods that intersect with AI, MU statistics, and database systems (Chakrabarti et al., 2006; Clifton, 2014). In a classical work on data mining, Chen et al. (1996) classify the types of data mining in terms of the type of databases being mined, knowledge to be mined, and techniques to be utilized in mining.
Data Mining, Al, and MLTechniques
The CRoss-Industry Standard Process for Data Mining (CRISP-DM) is a standardized approach in use across industries when data mining that provides structure to help you plan, organize, and implement data science methods. CRISP-DM addresses the following important strategies of generating value from big data: (a) construct a clear question or problem to solve appropriate to the data; (b) develop the team with the right competencies; (c) identify the data source, understand the data, and prepare the data; (d) explore the data using data mining techniques, advanced analytics, statistical methods, and identify the right question. CRISP-DM suggests six distinct phases. Table
27.3 identifies these phases with a description of the process within each phase (Process Data Science Alliance, 2021). TABLE 273 CRISP-DM Six Phases SIX PHASES
DESCRIPTION
1. Business
Determine business objectives, assess the situation, determine data mining goals, and produce the project plan.
Understanding 2. Data
Understanding 3. Data Preparation
Collect the initial data, describe the data, explore the data, verify the data quality. Select the data, clean the data, construct the data, integrate the data, and format the data.
4. Modeling
Select the modeling techniques, generate test design, build the model, and assess the model.
5. Evaluation
Evaluate results, review the process, and determine next steps.
6. Deployment
Plan deployment, plan monitoring and maintenance, produce final report/product and review the project.
Source: Derived from Process Data Science Alliance (2021). What is CRISP DM? https://www.datasciencepm.com/crisp-dm-2/#crisp-dm-phases.
Additional techniques in data mining include anomaly detection, association rule learning, cluster, network and relationship analysis, classification, regression modeling, and summarization. In addition, AI is receiving increasing emphasis in complex patient environments such as the intensive care unit for early detection of complications due to sepsis and in order to improve outcome (Lovejoy et al., 2019). When linked to AI, ML capabilities of electronic devices has exponentially improved the processing speed and memory capacity, particularly in complex clinical problem-solvin gand conceptualization. AI has four major components: ML, natural language processing (NLP), deep learning, and robotics. (Robotics was covered in Chapter 26.) These advanced analytic techniques will be cautiously applied to healthcare as increasingly complex problems involving unstmctured data require time-critical clinical decision-making. Hence, the skill base of the APRN necessitates an applied understanding of these analytic techniques, their advantages, and limitations. A convergence of AI and analytics may be a high priority for healthcare organizations; however, as the complexity of the algorithms
27: “BIG DATA" AND ADVANCED ANALYTICS
increases, the end-user's confidence and trust typically diminish (Kolyshkina & Simoff, 2021). However, there are many promising applications of AI (Lovejoy et al, 2019).
Often, ML is used interchangeably with deep learning; however, there are nuances in differentiating these levels of AI. Figure 27.2 reflects the relationships of these methods. Deep learning is associated with neural networks with three layers, including input layer, neural network, and output layer. The main differentiation is how the algorithm "learns" with increasing levels of complexity. The exact detail of how these layers of AI work are beyond the scope of this text.
Artificial
Intelligence
Machine
learning
Deep learning
FIGURE 27.2 Deep Learning Versus Machine Learning.
All these techniques are associated with knowledge discovery within databases. Table 27.4 reflects these techniques, definitions, and uses of these types of techniques in healthcare. Anomaly/change detection is used in public health to detect early, subtle patterns and trends in disease outbreaks, environmental and population health (Noah et al., 1998). This technique is used in some of the software used by the Centers for Disease Control and Prevention (CDC) for syndromic surveillance systems using emergency department data to detect disease outbreak from aberrations in the data (Henning, 2004). In addition, associations and relationships in healthcare data are often used to examine outcomes by examining relationships in variables, or they may be used in hierarchical data models. This type of data mining may also be a preliminary step to building predictive models to examine variables that are predictive of some outcome of interest. The regression models fall into this type of category of data mining, but a process of association rules learning may be a first step in the process. Cluster analysis discovers groups or structures in the data, such as clusters of patients who tend to go to one hospital in a given Zip Code or county An example of this type of work related to COVID-19 clusters can be seen at the Center for Applied Research and Engagement Center (https: / /extension2.missouri.edu/programs/cares). Summarization is used in many of our business intelligence (BI) tools that aggregate cubic views of data or report certain outcomes. These data-mining summarization tools allow an end user to drag and drop and quickly identify patterns and trends in the data based on the summarization of tables. These tools often have data visualization capability to see graphic relationships in the data as well. Figure 27.3 demonstrates this capability with the IBM Cognos BI toolset.
693
694
V: NEW AND EMERGING TECHNOLOGIES
TABLE 27.4 Types of Data-MiningTechniques TECHNIQUE
DEFINITION
HEALTHCARE APPLICATION
Anomaly detection
Pattern detection in identifying
Rule-based anomaly detection
data errors or unusual deviation from the norm
for detection of disease
Association rule
Identifying association between
identifies relationships in
learning
variables
variables associated with an
outbreaks
outcome of interest; can be
preliminary work to predictive modeling Cluster analysis
Discovering groups or structures
Clusters of market segments on patient preferences by
in the data
healthcare market Classification
Generalizing known structure to
Classifying patient safety errors related to HIT can support taxonomy development
new data or information
Regression modeling
Modeling data for prediction or explaining some phenomenon with the least amount of error as
possible Summarization
Data aggregation or compacting information, including visualizations and report generation
Regression models are often used for predictive analytics such as predicting factors that are associated with mortality or 30-day readmissions Often used in Bl tools to
aggregate data in cubic views of data by category and by some outcome measure
Bl, business intelligence; HIT, health information technology. Source: Fayyad, U., Piatetsky-Shapiro, G., & Smyth, R (1996). From data mining to knowledge discovery in
databases. Ame^icar^ Association for Artificial Intelligence, 77(3), 37-54. https://doi.org/10.1609/aimag.v17i3.1230.
FIGURE 27.3 Data Summarization Business Intelligence (BI)Tool. fllr
tdV
ymar
Rua
HHp
xrtoiQ.
ft-SB-
Gp
● S-fieiber ● t) .■■Oia%aigeO*li d (^P*MnlSak/c«Gcogttf>y «
Fitct ¥t lpcil4. Sm EV I’
P«M»i SATce Geeflrjohv
t) ●●■eviu.coiWTrjjtiti
ptn efr mere detiis.
eooo
Gender
7000
■ Mels Female
jj .●.0yMSCR»9cn li ■9.ttU€Cfrotrlf’‘
60CO
Jj .T, 8y ftCC Pr«gram
K .*.H(MPUIo'D«»Mre»(SO*} ti .* PtBcriioe s .P.SenSB'
I
^ 5000 a
5
4000
tt 3000
ti
".P^rcrSoMce
±p.>».W>C»6 tt
PrneM
2000
1000
Mnltrg 11
ami scMT^v
H
Smm 0^ Adnason
0
Case Count
£ .P. Unnviikdir £ .*.ScMUityUni
* Cc'^MflCMe
£ ●P.CcfSAciWnSuiut
m
I I I I
I I I I I lull Majer
Extreme
Hsrtallty RJak
£ ●* Typei^Adno*en
e
i
Moderate
Minor
tl -".PiteniStiU
m
fewuk
'subtotal (Bduded) Gender
0
Moderiie
Kajv
Exvew
Subtotal (oduded)
HortabtyKtsk
m
»n
i>4
7J»
144«S
14S7S
to*
l*t*
MB
MSO
1M7S
1477*
IS3*
4S«1
eT*]
1427*
2*1*1
2*3S4
1S72
4739
92S2
1537*
3093*
31132
II
Source: McBride, S. (2012, September 26), Use of state-level data by universities for improving patient safety, quality and population health. Paper presented at theTIHCQE Conference, Lubbock,Texas.
27: "BIG
DATA" AND ADVANCED ANALYTICS
Advanced Data Mining Tools
New tools are entering the market that are available in open sources that combine all these methods into one product. The IBM modeler is an example of this type of tool. According to IBM, this tool offers "an extensive predictive analytics platform that is designed to bring predictive intelligence to decisions made by individuals, groups, systems and the enterprise" (IBM, 2015). This tool combines many of the techniques noted earlier, in addition to text analyzers, data optimization tools, Bayesian and neural network compilers, chi-square automatic interaction detection (CHAID), and advanced 3D data visualization capabilities. Animation coupled with 360 x 360 x 360 free rotation of visualizations greatly enhance the ability to detect subtle patterns and anomalies of data mining tools are capable of visualizing the higher order dimensions in complex data by displaying them in the lower order dimensions of 3D space. The centroids and vector sum products of n-dimensional interest. The IBM modeler and other similar advanced
distributions are u.sually invisible in terms of graphic visualization techniques that are available in common statistical suite software such as SAS, SPSS, Stata, and others. In
contrast, the highly advanced graphics visualization engines of the IBM modeler and other similarly dedicated and highly specialized data mining applications not only display those hyper-dimensions as a default but also allow them to be displayed and
manipulated in multiple types of graphics. These graphics include such things as 3D-animated scatterplots, multidimensional Trellis arrays, multidimensional heat maps, and multidimensional probability density gradient topological maps. Visualization and
manipulation of ordination in higher dimensions of data is a hallmark of data mining and a core skill of the data scientist.
These advanced exploration capabilities are what place data mining in its own class, distinct and separate from standard and archaic hypothesis testing mathematical methods such as the parametric general linear model of regression. Data mining will not ever replace the standard models, but it will serve to enhance, support, and expand their application to the newly discovered phenomena that data mining techniques bring to first light. Dedicated data mining suites, such as the IBM modeler, make it possible to automate the discovery process. This is accomplished by mnning the same dependent (target) variable against all available independent (predictors of the target) variables, by sample testing multiple serial predictive models of the target with the predictors, with every appropriate mathematical model available, depending on the declared level of measure (categorical, ordinal, or scale) and role (target, predictor, and both) of the entire list of variables in the data set. For a complex data set with multiple millions of rows of data observations and multiple thousands of columns of data variables, the process can take days to complete. The resultant key output of this process is a short list of the top 10 to 20 predictive modeling methods that achieved a viable model, arranged in best to worst scoring predictor importance, given the target being predicted, as defined when the automation is initialized. Each of the top three or four models will echo the findings of each other, but each may show a completely different synoptic view of the data. The top five models may be factor analysis, cluster analysis, binary logistic regression, CHAID analysis, and a neural network of the predicted target variable, using a similar subset of independent predictor variables. Across all five models, the list of independent variables in the models will almost always agree (with some minor variations) in terms of predictor importance. Corroboration of the same predictors across multiple different modeling methods is also a key feature that sets data mining apart from more traditional analytical approaches that focus on one method as superior to all, with subsequent rejection of any consideration of alternate methods as being applicable to the development of valid and reliable predictive models
695
FIGURE 27.4 KNIME Workbench and Data Workflow.
Data Access
Transformation
Analysis and Data Mining
Visualization
Deployment
Database Reader
PMML Writer Decision
Tree Learner
Ido*]
in
'1
Access
Merge
Oracle DB
Auto-Binner
Data
XLS Reader
Enrich database with external data
Twitter API Connector
[dQ®
ddji
Scorer
■t ^ [dd*
LQO*
(do* Determine
Text Analytics
Preprocessing
i
>
mm
|dd#l
>
roo*
ld~d»l
Call out to Rio view
Include tbis Image
Contingency Table
iQpji
Extract
Filtering
important terms
and stemming
IOq#1 Save model tor later re-use
ft*
in a report Image to Report
XLS WrHer
loio* Export to Excel
Q lOO*
JavaScript Scatter Plot
Include diis Image in a report
JS
Interactive Table
Loading and Preprocessing on Hadoop Database Table Selector
Database
Statistics
Group By Database
mm
dd®
Database Joiner
Create
Connection Table Reader
Database
Table Selector
Database Row Filter
Color Joiner
L** t> r*
aggregate
mm
Image to Report
R View (Table)
Tag Cloud
Big Data
Hive Connector
> >
model accuracy Concatenate
Search
t
Tree Predictor
Text
Twitter
iiT
Decision
Partitioning
-t> ™ >
/ > (dd*
PUML
(OO®
idd®
[Od®
Manager
[dd®
mm
(dCLtl
Database
Highlight
Writer
Scatter Plot
[dd®
Ido®
taJ-
Bar Chart
lP_Qj»J
(J Free Chart)
Write results
to database
fop® Interactive
^ Id ■ Crosstab
[QO#l
data plot
ldj5®
Data to
Report
t> B
l> Ido®
dd® Include this table
in a report
API. application program interface; DB, database; KNIME, Konstanz Information Miner; PMML, predictive model markup language. Source: https://www.knime.com/downloads/fulI-license.
27: "BIG DATA” AND ADVANCED ANALYTICS
that demonstrate high degrees of sensitivity, specificity, positive predictive value, and negative predictive value. The KNIME (Konstanz Information Miner) tool is an example of an open-source
product that is based on the R statistical package. This analytics platform is a leading open-source platform that is downloadable from the KNIME website: www.knime. org/knime. According to KNIME Innovation (n.d.), "KNIME, pronounced [naim], is a modern data analytics platform that allows an analyst to perform sophisticated statistics and data mining on data sets to analyze trends and predict potential results." It also has a visual workbench that combines data access, data transformation, initial investigation,
powerful predictive analytics, and visualization. Figure 27.4 reflects how this workbench appears to the end user with an ability to drag and drop for managing the workflow of the data, transforming the data, and ultimately mining the data set. SAS Enterprise Miner is yet another very sophisticated data-mining tool available in the commercial market. This tool, similar to the IBM tool, has the ability to provide descriptive and predictive analytics to find patterns and trends hidden within healthcare data sets (SAS Institute, n.d.). All three of these tools noted earlier have the ability to
manage data within workflows similar to the one noted in Figure 27.5.
FIGURE 27.5
Baylor Scott & White data scientist and
contributing author Richard Gilder in a
data mining laboratory demonstrating some of the work done in
the laboratory.
THE ROLE OFTHE DATA SCIENTIST
A new and emerging role in the healthcare industry is the advanced analytics professional who is capable of managing and analyzing massive amounts of data and using the techniques and tools noted earlier. The average healthcare analyst is not capable of
fully understanding and managing these types of tools and techniques without special training and a solid foundation in statistics. The Harvard Business Review used the term "data scientist" to describe the competencies required as "part hacker, part analyst, part communicator" (Davenport & Patil, 2012, p. 16). The job of the data scientist is focused on using analytics to solve problems, and the data scientist has the competencies to understand how to "fish out answers to important business questions from today's
697
698
V: NEW AND EMERGING TECHNOLOGIES
tsunami of unstructured information" (Davenport & Patil, 2012, p. 73). In this review, the case of the analysts who developed the "People You May Know" feature on Facebook is used as an example of this new skill required to support the management and analysis
of big data. It is indicated that these new roles are required to exploit vast new flows of information for transforming industries (Davenport & Patil, 2012). It is further indicated that the rush to capitalize on these types of data is likely to face human capital constraints because of a significant lack of individuals adequately trained to work on big data and these data-mining techniques and tools.
So what does a data scientist look like and what type of work do they actually do within the scope of work? The Baylor Scott & White healthcare system, in their Dallas, Texas, facility has implemented a data mining laboratory (DML) and utilizes the IBM modeler to determine patterns and trends in the healthcare enterprise data. The data scientist who has oversight of this laboratory and helps train other analysts and researchers to utilize the tool is a master's-prepared nursing informaticist with extensive training in statistics, computer science, and biomedical and clinical informatics and with a solid base as a mathematician. Richard Gilder is noted in Figure 27.5 at the Baylor Scott & White healthcare system DML. Mr. Gilder, who is a data scientist and coauthor of this
chapter, is shown in front of projections of data mining graphics; he notes patterns and trends and uses data mining tools to visualize these patterns. A typical week in the life of a data scientist is similar to driving in heavy rush-hour traffic in a major city: unpredictable. The various healthcare projects that the data scientist works on could be compared to traffic lanes, where many projects mn simultaneously
in parallel. Each project has its lane, and each vehicle in that project convoy carries the next milestone in the project. It is only when the entire convoy rests safely in the parking lot after the last milestone has been passed that the objectives have been achieved or surpassed and tlie project is considered a success. But as we all know, everything can be rolling along smoothly in both directions and all lanes on the expressway, and suddenly with no warning, a vehicle has a blowout and has to pull off to the side of the road. This event has a ripple effect on the overall traffic pattern in both directions, is disruptive, and always causes problems. In data mining, something similar happens as a direct result of the data-mining process. Data mining can be disruptive as in the previously noted example where clinicians and administrators consider day-to-day activity to be satisfactory. It can be very disruptive to the status quo or normal workflow in the healthcare environment where the process is occurring; ultimately, it is a positive dismption with measurable outcomes. These outcomes may include measures such as the number of lives saved, number of preventable adverse events avoided, number of patients returning to normal function sooner rather than later, and amount of money spent. The chief scientific scope of work that the healthcare data scientist works under falls into one of the categories of translational or applied research, One example of this use in the industry is Baylor Scott & White's DML. The DML supports the aims of translational research, supporting
biomedical informatics science with state-of-the-science
analytical capabilities, in an interdisciplinary collaborative environment. Translation of vast and complex data into
actionable information resulting in evidence-based practices with outcomes of improved safety and quality of life is the primary mission. The DML serves as a resource to the community of healthcare sciences research. Translational research aspires to
re-use
data" (Schaffer, 2008).
Data mining requires removal of barriers around isolated data known as data silos. Data governance models within organizations can streamline and organize the removal of restrictions and barriers around data silos through hierarchical administrative
oversight that trumps the subordinate individual departmental and service-line
27: "BIG DATA" AND ADVANCED ANALYTICS
restrictions around data sources and data objects. Data mining requires data governance.
Capturing and meaningfully using the data artifact stream that is constantly generated by the process of healthcare delivery in the environment-.specif ic context of care is what the science of translational research aspires to do in healthcare data mining. The chief aim of healthcare data mining is identification and delivery of actionable insights to the frontline healthcare provider, sooner rather than later. Successful application of the data mining process to the vast, rich, complex, and chaotic streaming mixture of informationsignal and artifact-noise characteristics of healthcare data is self-sustaining as a result of dramatically increased demand for data mining services. Clinically significant MU of the information-signal characteristic to modulate the healthcare delivery process in a timely and efficient manner toward improvement, safety, and optimization for all involved, and especially the patient and the patient's family, is the heart and soul of data-mining science in healthcare. The cost savings that result from successful data mining can be significant. Increased patient and family satisfaction resulting from improved outcomes with the reduction of preventable adverse events, and identification of other factors that impact patient satisfaction and comfort, such as environmental noise, can be detected and measured through healthcare data mining. The data mining process that the data scientist experiences on an everyday basis revolves around the linchpin of MU through practical applications of translational research. The data scientists' primary operational objectives are consolidating, validating, vetting, and communicating findings in a way that can be understood by those with the accountability and authority to act upon them, in a safe, timely, effective, efficient, equitable, and patient-centered manner. Production and delivery of actionable insight from valuable information that would otherwise be lost as noise, confusion, and missed
opportunities for improvement is the scope of work for data mining endeavors in general. In a healthcare data mining operation, the results of delivering actionable insight to frontline practitioners of healthcare delivery are life-saving and have a positive impact on quality of life and return to function, which are, after all, the overarching objectives of all healthcare delivery. The following illustration in Figures 27.6 and 27.7 are provided as an example of the actual output that resulted from a typical DML project involving the public domain
data provided by the CDC consisting of national birth certificate data from more than out in the year 2010 in the United States. The project described next requires the data scientist to perform the functions of the data mining process, and the example serves as a stepwise walkthrough essential to the cross
4 million birth certificate forms that were filled
functional elements of the data mining process. The Problem
Every project is designed to solve a problem, and the problem for this project was that the desired analytical data set was only available for download, as an unparsed, flat even for a modern high-end gaming machine, 4 million records (rows) containing 234 variables (columns) pose a severe and text file of 4 million records. Let us consider that
computationally intense challenge to even successfully open such a file in a text-editing application, such as Microsoft Word, without crashing the computer in the process. The processing time to just open and view a file such as this could take 30 to 90 minutes, and editing could take multiple hours per "find and replace" operation. Extraction of the data from the birth certificates
was the first step in the data mining
process. This step was performed by the CDC. Everyone would agree that the process of filling out a birth certificate form when more than 4 million babies were born in the United
699
FIGURE 27.6 Chi-Squared Automatic Interaction Detector (CHAID)Tree Diagram. S(r*MitoAKAA7i
CMQcry
%
mA ZStUT4 3.6
■ns
■NO
■ns
TottI
95>;57
loao 2.613.431
Q Esttrabd 9estt1lDe SSmeka or lea
Ad|. pfatoess inoCU'S^Bi s 6IISXA tffsl
( YES
no
1 Model
CotoQory
Node 2
%
●HD ■ns
Csiodory
as.1
210641
116
34CC$
M
244.676
low
■M
% 9JA 2»fjsa ZA
ei,)22
njb
?«’Ca75S
■ TES Tow
E
u laia P»i 24Bg
Prlmarv CsacUon Ad], p values = .OOO Chl-Sqeare s 27308SI,
M). pvalun B JXIO CU-SquM > 6764J26, disl
dlsl
£
I YB
(
1
MO
1
YB
i
Ni3W4
Cettgory ■NO
■ YB ToW
%
C4Nv>«y
B07
80062
13
$571
3.4
86.663
81M
m,6W
112
26.464
BO
15601.1
0
■HO ToW
Calegofy ●ns Total
Category
%
9u
am
u
XbiA
23
ft.SM
■HO ■ YES Tew
%
Cdegory
IT
●1.2
tOJBO
II
1.915
OJ
21.704
●m ■TB
Tew
1.928.582
913
■ YB Toial
4»6i
3.7
1131B
lia
440,173
0
tftsl
dfBl
d(s1
Buttneelght efa thee TBSSg
1
J
1
■0
YES
I
NO
ns
1
1
NO
1
1
YB
1
Nodes
Node 10
%
Caiegofy
Calegoiy
73J92
■no
7M
SAIS7
■YB
1M
14.421
■YB
20J
14JM3
14
67313
23
61200
TeW
Node 12
Hodell
%
aU
0
Cabgory
718
%
■NO
A^ pvMiae < OOD Cki-Sdsve B S3713S,
●m TeW
44.804
SinitwelgM leatNa 2B3SS A^ pvafeies B .000 Ohl-Spun s S14J41
r
YB tedeB
13
0
1
■80
07.7 f JBOTn
■YB
Prtirary CsecOon
1
NO
Cotegorr
A4. p vWiB s mo CU'^ve c 447.191.
df*l
Node?
Nodes
%
□
PHnwY CeectiM p vaM B mo CW-eqiive s 31IBSO,
f
CatoQMT
■YB
TobI
1
Nodes
%
nm
●m ■YES
%
Cattgory
97J imsn
TeW
13
41.999
703
1 6S136&
0
Mode 13
%
Cabgory
■no
914
74212
■YES
U
2305
2J
77.017
TdW
■■0 ■YB Tew
Node 14
%
Catogoiy
915
3B7.SM
33
t4i66
116
412377
0
■K ■TB TeW
0
% 933
96344
12
175?
1.1
280»
0
Fathan age 23 or lea Ad^ prabes s .OOOCM-n99.00; Um pnrdf il&boo 9U
Oly-
FACOBDCk
t
RtAilid PiJR.ACYCUA«Ai;TrSD* -
« ● _
frifc*i€R RESOLD'’' - * *
« Ufrk FULLZ C DCCUUD JT GCVtC « Oecuri*^^ 04 M ●ma t
irc3u«ennsint9innAQ;ri «ulr●ab40t^sR.l b'TroiO'^ w*9c« o«a Titatvi mursn:*
Source: IBM Corporation. (2017a). IBM X-Force threat intelligence index 201.7 IBM Security. X-Force. IBM.
HS/
752
V: NEW AND EMERGING TECHNOLOGIES
over the past few years and is now a widely acknowledged potential that can covertly operate within any organization and one where immediate steps must be taken by the risk and security departments to mitigate. Less acknowledged and equally insidious is
the threat from trusted third-party providers whose lapse in security diligence could easily become the conduit that a malefactor leverages to apply compromised seairity credentials and network access rights exfiltrate partner data and intellectual property (IBM Corporation, 2017b).
Within the current paradigm, most companies struggle with the complexities of change and consequently the second- and third-order effects these changes have on their security posture. Securing the company's data is a strain for most security departments
with today's attack techniques, so imagine the difficulty most companies would have if the criminal community was given a technological gift which shifted the balance of power to their advantage. In 2013, Edward Snowden exfiltrated top-secret data from the U.S. government, in addition to information on secret programs. Snowden captured
information on highly sophisticated programs designed to compromise computers for national security purposes (The Courage Foundation, 2018). These complex computer programs and doaimentation, designed by the significant resources of a national government, were released to the public through various sources including wikileaks. org. When this type of information is added to stolen as.sets from other breaches, the data can be placed in the hands of a resourceful criminal network and small, poorly funded bands of criminals with all the tools needed to launch highly sophisticated and automated attacks. This type of activity may be why companies reported in 2016 the number of cyberattacks went down, but their losses increased. Fewer attacks are needed when the attacks are more targeted at what the criminal wants and when they succeed on the initial attempt (IBM Corporation, 2017b). Data breaches have real costs; according to a study published by the Ponemon Institute in 2016, healthcare had the highest cost of data breach per lost or stolen record at $380 compared to financial services' average cost of $245 and the global average cost at $141. Additionally, the United States has higher breach notification costs, ex-post response costs, and lost business costs than other nations in the study (Ponemon, 2017). Figure 29.3 depicts a few of the more notable 2016 leaks of unstructured data containing personally identifying information, financial data, personal secrets to remain hidden and, as detailed in the IBM X-Force Threat Intelligence Index for 2017, some of the leaked documents directly influenced global politics. For example, from the Panamanian law firm "Mossack Fonseca," the leak exposed offshore accounts of prominent people worldwide; other leaks targeted the financials of current and former heads of state, their family and friends, and business people, and celebrities. In April 2016, the Prime Minister of Iceland stepped down in the aftermath of a leak (IBM Corporation, 2017b). A hacker can obtain access to critical administrator credentials in 3 days to compromise the organization's data after initially gaining access to the environment. The global median time from compromise to discovery is 99 days. Given that it takes a very short time to exfiltrate the data once the compromise is in place, detection and reaction are still 96 days too late (Mandiant, 2017). Despite the best efforts of dedicated security professionals, it has become clear that humans cannot perceive (sense) the threats in the environment and respond with defensive measures quickly enough to mitigate the current threats and are woefully inadequate to respond to the new threats being prepared. Malefactors improve their technology and techniques rapidly and adapt to defenses quickly. Thus, a new approach to security is needed that can sense the environment and adapt and respond to the threat at the speed with which the environment changes and the attack executes. This type of activity constitutes the realm of a cognitive system.
29: ENHANCING CYBERSECURITY
FIGURE 29.3 Notable 2016 Global Leaks of Unstructured Data.
Turkey
Canaita
17 GB archive offilee from
5 GB data stolen from a
casino chain^ including national ID numbers, photo ID copies and other personally identifiable information (PH)
a Turkish police server^
Germany/Europe 1.9 TB of insider information
Philippines
atxxjt European football
300 GBofRIipino voter
players,-their salaries
Poland
and contracts
14 GB fromaRilish internet
of half the country's voters
service provider (ISP)'*
and Including fingerprints
US
data^ (Comelec) consisting
and passport scans
150 GB from an Ohio
urology group' including protected health information (PHI) data
France 3 GB leak of data from the 400 GB of Habitat for
Humanity® volunteer data including background checks
French masonic lodge,"’ providing an insider look Into the highly secretive Freemason organization
h
us
500 GB from Gorilla
Glue^ including intellectual property
Kenya 1 TB of data from
Kenyan Ministry of Foreign
Affairs®^ including trade secrete and classified Information
Qatar 1.4 GB of data from a Qatar
bank®^ including intelligence reports on people of interest
India
100 GB from a Kerala, India,
government^ server to Facebook including names, addresses, and Income
Source: IBM Corporation, (2017a). IBM X-Force threat intelligence index 201.7 IBM Security, X-Force. IBM.
Cognitive Security Systems Sharing cyberattack and breach information, not company-sensitiv e information (perhaps through generic case studies), between machines at the speed of electrons holds the promise to halt the spread of viral infections such as the WannaCry ransomware attack on May 12, 2017. This ransomware attack reportedly affected 230,000 computers in over 150 countries (Perlroth & Sanger, 2017). A cognitive system must first learn;
ergo, a cognitive security system must be taught about security. Think about this as sending the computer to school to earn a PhD in security. This knowledge and learning is stored in a special data store called its corpus or memory. The first step is the computer is given a vocabulary consisting of all the common words in the language, and then the vocabulary is expanded to include the words and meaning unique to the computer science and security domains. Second, the computer is taught to read each type of literature that must be consumed and how to interpret its idioms. A conference paper is stylistically different from a blog and a journal is different from a tweet. This is accomplished by selecting a sample set of documents for each type of literature to train the computer with; the computer will read and understand by carefully decomposing the document's sentences by mapping the relationships and concepts of the words within them. In this way, the computer begins to learn, through example, how the material in the literature and its concepts relate to each other. After the initial training.
753
754
V: NEW AND EMERGING TECHNOLOGIES
new documents, situations, and problems are presented to the computer to evaluate if the right answer is derived or the correct action is executed. If not, the computer is corrected through adding additional information and/or examples and the test is repeated. Through this iterative process, the computer builds new knowledge in its corpus, learns by example, and accuracy increases. Special programs are developed to
accelerate the learning process by enabling the computer to read unstructured material, such as textbooks, at a rate of millions of words per minute, and the human to more quickly relate words, concepts, and categories together and introduce new examples and lessons. This type of machine learning is termed supervised machine learning (SML) because a human must prepare, build, monitor, and adapt (supervise) the training until the computer exhibits sufficient proficiency and accuracy. At the heart of all cognitive systems is the ability to be presented with a problem that can be shaped into the form of questions. By using the available information and experiences to date, the cognitive system then answers the question(s) and optionally takes a desired action. The purpose of the cognitive system determines the content of the questions that can be asked, and actions performed. For example, IBM's Watson for cybersecurity is a cognitive security system that can be used as an archetype to illustrate how the question/answer process works. As seen in Figure 29.4, first the question itself is decomposed, marking it for grammatical elements such as subject, verb, object, and modifiers and detecting the "focus" that is being solved for (e.g., the focus of "What malware leaves the file qwerty.doc on systems" is "What malware"). Next, Watson generates hundreds or thousands of candidate answers or "hypotheses" by performing numerous natural-language searches within its corpus. Watson casts a very wide net when generating hypotheses to ensure all possibilities are included for consideration. Next, Watson applies dozens of different scoring algorithms that use many different ML and natural language processing (NLP) variations. Each of these scorers votes on each of the hundreds or thousands of hypotheses, resulting in a large array of possible answers FIGURE 29.4 Deep QA IBM Watson. Deep QA
C2:
Learned Models
Help Combine and Weigh the Evidence
Models
Answer Question
Sources
Primary Search
Question
and Topic Analysis
i Candidate
Answer
Evidence
Scoring
Retrieval
E)eep
Models
^odete^ yodels
Evidence
Scoring
Models
Answer
Models
Generation
Question
Hypothesis
Hypothesis and
Decomposition
Generation
Evidence Scoring
Hypothesis Generation
Hypothesis and Evidence Scoring
Final Confidence
Synthesis
Merging and
Ranking
n
Answer and Confidence
Source: IBM. (n.d.). IBM—DeepQA project. Retrieved October 6, 2021, from httpsi//www.research.ibm.com/ deepqa/deepqa.shtml.
29: ENHANCING CYBERSECURITY
with many scores applied to each. For each possible answer, the original information that supports it is tracked as supporting evidence. Once all the scorers have voted on all the hypotheses with their supporting passages from the corpus, the array of results is then weighed according to the ML weights (specifically. Logistic Regression) that were calculated during the most recent training. The highest confidence answers and their supporting information are presented to the user through an interface or, if desired and the confidence level is high enough, the system can take autonomic actions. This new approach to security is anticipatory (as is a living organism), and expectations are high regarding security breaches and uncertainty. A security breach is defined as one that impacts services, data integrity, and access to information and may be covert or overt.
Preparing to Perform a Vulnerability or Risk Assessment in the Healthcare Environment
We can surmise that cognitive security is interdependent on data sources (i.e., devices) in the healthcare environment. Hence, the health loT is characterized by ubiquitous sensing and knowledge understanding and is often referred to as sense making. This step of data ingestion and sourcing is essential to threat characterization. In other words, this new approach to security is anticipatory and defined by its essential elements of information (as is a living organism with its DNA and cellular structure), and expectations are defined by historical events, learning, and assimilation. The result is user-defined levels of uncertainty. As a result, security breaches may impact services, data integrity, and access to information and define future risk tolerance.
Approach to Risk Assessments
for specific systems. Although generalizable, risk assessments must be precise to be meaningful and actionable. To define appropriate criteria, the following approach is recommended:
Contextual information defines risk and vulnerabilities
■ Understand baseline operational conditions (human, software, hardware) from individual asset to system perspectives.
■ Map system redundancies in information presentation, and junctures for decision making requirements to include critical nodes of information flow. ■ Understand tolerance for dismption of services or data integrity. ■ Determine minimally acceptable workflow efficiency and effectiveness and healthcare risk tolerance. (W. Rhodes, personal communication, 2018) BLOCKCHAIN, HEALTH CLOUD, AND HEALTH SECURITY
On October 31, 2008, an anonymous person using the pseudonym Satoshi Nakamoto published a research paper designing a cashless currency called bitcoin that transacted payment in a peer-to-peer manner, bypassed banks as intermediaries, lowered transaction fees, and averted accounts being frozen by law enforcement authorities (Nakamoto, 2008). Privacy and security were a key concern for the author, so to store the transactions in a tamper-free, transparent, and anonymous manner, the author devised a new type of data storage methodology called "blockchain." For further clarity, bitcoin is a cryptocurrency or a digital-only form of currency. Blockchain is a new data storage method that acts as a distributed ledger for securely .storing the transaction data and its
755
756
V: NEW AND EMERGING TECHNOLOGIES
history in chronological order in a tamper-proof, immutable, and transparent manner where the concerned parties can verify transaction authenticity. Bitcoin and Blockchain
Bitcoin and blockchain technology are continuously referred to as integral to the cybersecurity environment. Initially adopted by criminals as a way of avoiding money laundering detection by law enforcement authorities, the value of the blockchain storage method was soon realized for its commercial value and was enhanced by several companies working together in an open fomm to adapt to the more rigorous requirements of business and healthcare.
Blockchain derives its name from its chronological structure. Thestmcture is enforced through a chain of forward and backward pointers that record the previous and the next record in the chain. Once a transaction is written to the blockchain data store, it becomes a link in the chain and is never erased. If a mistake occurs, another transaction is posted
that corrects the problem, leaving the mistake and the correction permanently recorded for full transparency. Blockchain for business and healthcare differs from the original blockchain technology in four key areas: 1. Identity over anonymity—The goal of bitcoin is to (always) keep identities secret (permissionless system). In contrast, the goal for business and healthcare is to ensure each participant has a unique identity (permissioned systems) that is validated, and only authorized caregivers are given access to the information they need while also providing the transparency required by auditors and regulators. This ability to identify individuals and apply policies to con.strain their access allows compliance with data protection regulations such as those stipulated in HIPAA. 2. Selective endorsement over proof of work—In keeping with existing laws, transactions are endorsed by relevant participants to prove acceptance of terms and receipt of goods and services. The proof of work method is computationally expensive and not needed if the identity is known. 3. Assets over cryptocurrency—Assets of all types are recorded in the blockchain ledger while with bitcoin or any other cryptocurrency the only asset recorded is the cryptocurrency.
4. Speed and scale—The original design of blockchain could handle only a fraction of the concurrent transactions load required by business and healthcare, so the design was enhanced to improve speed, scalability, and resilience. Blockchain for business is airrently used to track the mining and distribution of diamonds, financial trades, payments, clearing and settlements, insurance contracts, and inventory, and is being adopted by the largest banks to streamline bureaucracy and reduce costs. From a 2016 study by the IBM Institute for Business Value, blockchain implementations by early adopters (trailblazers) have experienced that the greatest positive impact considering time, cost, and risk was achieved through applying blockchain technologies to clinical trial records, regulatory compliance, and nredical health records (IBM Corporation, 2016). Health Cloud
McGovern et al. (2014) note that a person's health information comprises enormous amounts of data about the individual representing millions of gigabytes of data. To a
29: ENHANCING CYBERSECURITY
computer system a person is represented as a collection of data stored in files. Intimacy is garnered through collecting additional data from different events of anyone's life such as medical tests, doctor's reports, medicines you have taken in the past and those you are taking currently, allergies, blood type, where you have traveled recently such as an overseas trip, what life changes you are facing such as a new baby; anything that has ever been determined about you medically. Valuable insight can be gleaned from understanding the health history of your parents, your grandparents, aunts, uncles, brothers, and sisters. Each additional piece of information begins to paint a portrait of who you currently are and gives insight into your probable future health conditions or perhaps conditions that may be passed to an unborn child. All these data have value when unlocked through powerful analytic programs and may be the key to diagnosing an illness much more quickly, preventing a probable future ailment, or informing emergency care professionals with critical information needed to provide you with urgent care. When integrated with an existing electronic health record (EHR) system and accessed through a standard application programming interface (API), blockchain enables new options to share the data securely and maintain compliance with HIPAA regulations (Ekblaw et al., 2016). A long-standing objective for medical professionals has been to significantly improve the quality of care provided to patients at the time and place of need. One obstacle, above all others, that has been the most difficult to overcome is that throughout a person's life, their medical records are stored in various computer systems used by different hospitals,
doctors, and specialists at various locations and perhaps with caregivers in different states or a different country. Most of the systems housing your medical history cannot easily share information with each other, and when urgent care is needed, there are few methods available to inform the responding emergency persoiinel of your medical history adequately. One approach to solve this challenge is a mechanism to easily validate your medical information from any source it is generated from, at the time it is created, and safely store it in a manner where any authorized caregiver can access it to provide you the best care possible.
CASE STUDY
Proposed Solution;The Health Cloud Option Recent advances in new comm unication technologies might enable new solutions to challenges of improving cybersecurity noted inthischapter.These new technologies include communications such as the wide availability of cloud computing, more ubiquitous access to high-speed wireless internet, the general availability of the Blockchain for Business software, and advances in the standardization of APIs to exchange data and enable systems to interoperate. A pictorial overview is noted in
Figure 29.5, which the authors will proffer as a proposed solution, the Health Cloud. The Health Cloud solution proposed to address the issues noted in this chapter would have six fundamental attributes to address security threats: Cloud infrastructure consists of a series of highly secure Cloud Data Centers that provide all the application services with sufficient
■ Robustness—The Health
performance, availability, redundancy, scalability, resilience, and disaster recovery. These qualities ensure that the information is available at any time it is needed. Although the health data would be stored in a physically decentralized manner, to
757
758
V: NEW AND EMERGING TECHNOLOGIES
FIGURE 29.5 The Health Cloud.
Source: Morrow, W. (2017, December 26). How is cloud computing revolutionizing healthcare? https://www.huffingtonpost.com/William-morrow/how-cloud-computin g-is-re_b_11675810.html.
the patient it would appear as a centralized repository of all their health records and medical artifacts no matter where the medical service was administered.
■ Secure—The health data are protected by state-of-the-art cognitive security capabilities and the data are always encrypted no matter if being transmitted or at rest in its data stores. Cognitive security technology can sense, respond, and adapt to a highly dynamic threat environment, protecting the data from being breached. Given that even the most secure technology can be breached, the data will be protected with two additional technologies unique to sophisticated Cloud Data Centers, which ensure the data are not used outside the geographic locations they are authorized for. Should the data be stolen, special encryption and storage distribution techniques render the data completely and physically unreadable.
■ Blockchain Transparency—The health data will be kept in a robust btockchain data store so that the patient's complete and up-to-date health records are available upon request to authorized healthcare givers. Additionally, the transparent nature of the blockchain technology will ensure compliance with government regulations such as HIPAA.
■ Internet ofThings—The Health Cloud is engineered to ingest and store all the various sensor inputs available from home and hospital devices extending the data that can be considered in the patient's health evaluation as well as new services the sensors enable.
■ Improved Convenience/Reliabllity/Interoperability—Having all the data available to new healthcare providers means that the patient does not have to repeat
29: ENHANCING CYBERSECURITY
information already provided. Additionally, new services can be evaluated and authorized quickly, insurance issues resolved swiftly, and new insights from powerful analytics can be made available.
■ Improved Safety—Today, in emergency situations, the medical responders must work quickly and often administer powerful drugs without an adequate knowledge of your medical history because it is simply not available, and action must be taken immediately. When your records are on the Health Cloud, emergency medical responders can be authorized to Immediately see the vital information they need to consider to best respond to your condition. In the scenario depicted in Figure 29.5, the patient's data reside in the Health Cloud, and these data are available for the insurance company to preauthorize
treatment and settle expenses efficiently. New doctors and specialists can quickly review a complete history with their findings being made immediately available. During an emergency, the data will be made available to the responder, but more than that, the data will be enriched and filtered to serve the needs of the specific type of responder. If a helicopter, plane, or ambulance is providing transportation, its equipment capabilities, its communication capabilities, and unique information needs can be taken into consideration. During this entire process, the data and applications can be reformatted to fit the limitations of the equipment and transmitted securely using highly sophisticated security technologies kept constantly updated. During each step of the emergency process, the patient's treatment can be updated to the Medical Cloud for others to see and monitor. New cognitive technologies such as "Active Listening" will enable the Health Cloud to listen to and understand the emergency responder and then communicate the vital information to the hospital instead of burdening the responder with this added responsibility as is done today. Additionally, with the entire medical history available, should advanced/alternative medical treatment be necessary, such as that found in a clinical trial, the online medical community would be able to quickly evaluate and match patients with the appropriate clinical trials available.
SUMMARY
In summary, the speed at which technology and innovation is disrupting business models and the unprecedented speed at which the threat matrix evolves and adapts is outpacing the mitigation capabilities and budgets of even the most generously funded healthcare organizations. This has supercharged the race to develop next-generation cloud cognitive security systems that can more quickly sense their environment, decide on the proper course of action, and respond more quickly than their adversaries. The Health Cloud is one embodiment of some of the thinking that securely collects, stores, and disseminates the critical health-related data and insight to the authorized caregiver at the time and place it is needed.
In this chapter, we have reviewed the basics in terms of cybersecurity and new developments related to cybersecurity. We have discussed the necessity for enhanced security as a result of new and emerging global threats and attacks. Additionally, nextgeneration health information environments and challenges are described, including implications for security and further challenges we can expect in the future. Further, the movement of personalized medicine and genomics with massive amounts of personal
759
760
V: NEW AND EMERGING TECHNOLOGIES
health information potentially vulnerable to disclosure via breaches are discussed. The pros and cons of bitcoin and blockchain with the malicious and positive uses of both were tangentially referred to in this chapter. Finally, a proposed case study is presented addressing how we might create solutions to address these global threats to safely and effectively exchange massive amounts of personal health information on behalf of the populations served. We conclude with reflections for the reader to consider given both the threats and the opportunities presented within the chapter in the Exercises and Questions for Consideration.
END
1
CHAPTER
RESOURCES
EXERCISES AND QUESTIONS FOR CONSIDERATION
Considering the information presented in this chapter as well as the case study presented on the Health Cloud, reflect on the following questions: 1. What are the environmental factors globally and new developments in healthcare information technology that are precipitating new and emerging cybersecurity threats? 2.
Given increased amounts of information available on personal health information, including genomic information noted in Chapter 25, how can healthcare organizations assess for risk and potentially design programs to guard against new and emerging cybersecurity threats?
3.
What are bitcoin and blockchain and why are they used both maliciously as well as to protect health information?
4.
In the case study presented, how is this scenario different from what many states and regions have in place for sharing and exchanging health information on behalf of the populations served within their communities via health information exchanges (a review of Chapter 11 might be helpful for this reflection)? ADDITIONAL RESOURCES A robust set of instructor resources designed to supplement this text is located CONNECT ,
at http://connectsprlngerpub.com/content/book/978-0-8261-8526- 6. Qualifying
instructors may request access by emailing textbook@springerpub. com.
REFERENCES
Briggs, B. (Ed.). (2017). Tech Trends 2017: The kinetic enterprise. Deloitte University Press. Cisco. (2013). Cisco CIO summit 2013: Summary, https://www.cisco .com/web/offer/ciosummit2013
-london/CIOSummit2013_Summary.pdf The Courage Foundation. (2018). Courage Snowden, https:/ Zedwardsnowden.com
Ekblaw, A., Azaria, A., Halamka, J. D., & Lippman, A. (2016). A case study afr blockchain in healthcare: "Medrec" prototype afr electronic health records and medical research data, https; / / www.media.mit.edu / publications/ medrec-whitepaper G—, T. (2014). Renting a zombie farm: Botnets and the hacker economy, http://www.symantec.com/ connect / tr/ blogs / renting-zombie-farm-bo tnets-and-hacker-ec onomy
HITRUST. (2021). HITRUST CSF Our Framework, https: / /hitrustalliance.net/product-tool/hitrust-csf/
29: ENHANCING CYBERSECURITY
IBM. (n.d.). IBM—DeepQA project. Retrieved October 6, 2021, from https://www.research.ibm.com/ deepqa / d eepqa. sh tml IBM Corporation. (2016). IBM X-Force threat intelligence report 2016. IBM Security, X-Force. Author. IBM Corporation. (2017a). IBM X-Force threat intelligence index 2017. IBM Security, X-Force. Author. IBM Corporation. (2017b). Security trends in the healthcare industry. IBM Security, X Force. Author.
Identity Theft Resource Center. (2015). Identity Theft Resource Center breach report hits record high in 2014.
http: / / www.idtheftcenter.org/Press-Releases/2014breachstatistics.htm
INTERPOL. (2020). Preventing crime and protecting policeL: INTERPOL'S COVID-19 global threat assessment. https:/ / www.interpol.int/en/News-and-Events/News/2020/Preventi ng-crime-and-protecting
-police-INTERPOL-s-COVID-19-global-threat-assessment
Mandiant. (2017). M-Trends 2011A r>iVn»from thefron f lines, https: / / www2.fireeye.com / rs / 848-DID-242 / images / RPT-M-Trends-2017.pdf McDowell, M. (2013). Security tip (ST04-015): Understanding denial-of-service attacks, http:/ /www.us-cert .gov/ncas/tips/ST04-015 McGovern,L.,MilIer,G.,&Hughes-Cromwick,P. (2014). The relativecontributionofmultipledeterminants
to health. Health Affairs, https: / /www.healthaffairs.org/do/10 .1377/hpb20140821.404487/full
Morrow, W. (2017, December 26). How cloud computing is revolutionizing healthcare? https://www .huffingtonpost.com/william-morrow/how-cloud-computing-is-re_b_1 1675810.html Muthuppalaniappan, M., & Stevenson, K. (2021, February 20). Healtlicare cyber-attacks and the COVID-19 pandemic: an urgent threat to global health. International Society for Quality in Health Care, 33(1). https: / / doi.org/10.1093/intqhc/mzaall7 Nakamoto, S. (2008). Bitcoin: A peer-to-peer electronic cash system, https:/ /bitcoin.org/bitcoin.pdf National Institute of Standards and Technology (NIST), Computer Security Resource Center. (2021). Glossary, https: / /csrc.nist.gov /glossary /term / cybersecurity Perlroth, N\, & Sanger, D. (2017). Hackers hit dozens of countries exploiting stolen N.S.A. tool. Neio York Times, https: / / www.nytimes.com/2017/05/12/world/europe/u k-national-health-service -cyberattack.html Ponemon Institute. (2017). Cost of data breach study: Global overview. Ponemon Institute, https: / /www-01 .ibm.com / common/ssi / cgi-bin / ssiaIias?htmlfid=SEL03130WWEN
761
pplicaftQn of H Interprofession InformationTechnology in Educatibn MARI TIETZE, CHRIS McCLANAHAN, AND DAVID GIBBS
OBJECTIVES ●
Outline interprofessional education (IPE) components and their relationship to patient outcomes.
●
Discuss/explore the relationship between IPE and health information technology (HIT) for the delivery of healthcareservices now and in the future,
Describe the processes and professional standards (best practices or competencies) needed for the delivery of interprofessional education/collaboration (IPE/C) -based healthcare and quality patient outcomes. ● Describe exemplars in which IPE are used to deliver healthcare services. ● Utilize a toolkit to define steps in implementing an IPE program. ●
CONTENTS INTRODUCTION
764
History of Interprofessional Education
764
Defining Interprofessional Education and the Significant Role of Ethics COMIVIUNICATION
765
766
RELATIONSHIP BETWEEN INTERPROFESSIONAL EDUCATION AND HEALTH INFORMATIONTECHNOLOGY 768 Stakeholders
769
Team Practice/Simulation
770
CONSIDERATIONS FORTHE FUTURE
Organizations to Support the Process
771
771
Interprofessional Informatics Program Efforts CASE STUDY SUMMARY
773
776 780
EXERCISES AND QUESTIONS FOR CONSIDERATION REFERENCES
780
780
The contributions of Stacey Brown to this chapter in previous editions of this book are acknowledged.
764
V; NEW AND EMERGING TECHNOLOGIES
INTRODUCTION
The purpose of this chapter is to introduce interprofessional education (IPE) and describe IPE's relationship with health information management (HIM) and patient outcomes. The chapter also explores what is on the horizon for the IPE and HIM relationship, introduces professional standards, best practices, and competencies necessary for quality patient outcomes, and provides practical examples of the relationship. Note that the National Center for Interprofessional Practice and Education has indicated that we should adopt Interprofessional Practice and Education (IPE) as the "new" IPE (Regents of the University of Minnesota, 2021). This would replace those terms that previously contained the Collaborative.
A century after Flexner, Goldmark, and Welsh-Rose revolutionized postsecondary education for health professionals, two significant reports from the Lancet and the Institute
of Medicine (lOM) sought to similarly redesign the education of health professionals for the 21st century. The independent Lancet Commis.sion led by Julio Frenk and Lincoln Chen released Health Professionals for a New Century: Transforming Education to Strengthen Health Systems in an Interdependent World (Frenk et al., 2010). The lOM [now called the National
Academy of Medicine] produced The Future of Nursing: Leading Change, Advancing Health (lOM, 2010). The National Academy of Medicine (NAM) recently updated the 2010 version of The Future of Nursing: Leading Change, Advancing Health 2020-2030 (National Academies of Sciences, Engineering, and Medicine [NASEM], 2021a). All of these reports provide highlevel visions for the health professions but rely on educators to identify the best relevant practices and mechanisms for expanding proven, improved approaches to integrated health professional education. Considering the importance of IPE to the safe and effective delivery of healthcare, the NAM created an ongoing, evidence-based forum for multidisciplinary exchange on innovative health professional education initiatives. Known as the Global Forum on Innovation in Health Professional Education (NASEM, 2021b), this forum
not only convenes stakeholders to highlight contemporary issues in health professional education but also supports an ongoing, innovative mechanism to grow and evaluate new ideas. This is a mechanism that is multifocal, multidi.sciplinar y, and global (Cuff, 2013). The work of the Global Forum has been reflected throughout the previous chapters in this book. For example, disaission about roles in Chapter 2 included content about the expert panel on IPE and related the use of information technology (IT) in those roles. Similarly, the discussion about patient safety and quality in Chapter 20 included content about how the integration of IPE-skilled health professional teams tend to yield greater efficiency and more positive outcomes than those who are not IPE skilled. The association of IPE with consumer/patient engagement and activation is such that IPEskilled health professionals are additive to the engagement/acti vation model in that these professionals can enhance patient engagement/activation efforts compared with the traditional involvement of health professionals. Finally, data analytics and clinical
decision support systems (CDSS) are strengthened when IPE-skilled team members do the work to build and use these tools. These component relationships along with the Nursing Education Health Informatics (NEHI) model (McBride et al., 2013) provide the context for managing the IT of the future.
History of Interprofessional Education
The IPE movement became active in the mid-1990s. Many of the pioneers were foundations, such as the John A. Hartford Foundation, Robert Wood Johnson Foundation (RWJF), and Josiah Macy, Jr. Foundation. Each foundation identified the need for professional collaboration. In addition, the lOM presented alarming rates of multiple problems facing the nation related
to quality care. The lOM laid out visions of how systems must change in practice. To Err Is
30: INTERPROFESSIONAL APPLICATION OF HIT
Human: Building a Safer Health System (lOM, 2000); Crossing the Quality Chasm: A Nezv Health System for the 21st Century (Committee on Quality of Healthcare in America, 2001); and, in education, Health Professions Education: A Bridge to Quality (lOM, 2003). By 2005, professional organizations solidified the lOM vision of focusing on interprofessional collaborative practice as the primary means to address international quality problems. Significant among the practices were the Canadian Interprofessional Health Collaborative (CIHC) and the American Interprofessional Health Collaborative (AIHC). Both organizations teamed together to form the Collaborating Across Borders (CAB) initiative to accelerate the already rising IPE movement. Between 2005 and 2012, accrediting agencies and professional organizations redefined competencies of individual healthcare professional education curricula.
Professional organizations, in particular, the World Health Organization (WHO) and the Interprofessional Education Collaborative Expert Panel (IPEC), created sentinel reports defining IPE and identifying core competencies for interprofessional collaborative practice. Today, the reports serve as foundational documents for all health professions chartering a course of IPE. IPEC continues to implement goals and strategies around: ■ serving as a thought leader for the advancement of IPE, ■ advancing academic efforts around IPEC, and
■ informing policy (IPEC, 2021). The United States has begun building resources to support IPE. In 2012, the Health Resources and Services Administration (HRSA) of the U.S. Department of Health
and Human Services (DHHS) awarded the University of Minnesota $4 million over 5 years to establish a national coordinating center for IPE and collaborative practice, the National Center for Interprofessional Practice and Education. In addition, the Josiah Macy, Jr. Foundation, the RWJF, the Gordon and Betty Moore Foundation, and the John A. Hartford Foundation have collectively committed a maximum of $8.6 million in grants over 5 years to support and guide the National Center for Interprofessional Practice and Education (Health Professions Accreditors Collaborative, 2019).
Defining Interprofessional Education and the Significant Role of Ethics Patient rights and the delivery of safe care are expected to be facilitated by information technologies such as the electronic health record (EHR). This chapter also focuses on the key elements needed for that to happen and the associated role of patient engagement/ activation.
Noting issues with communication and using the EHR, the Texas Nurses Association and Texas Organization of Nurse Executives partnered to evaluate the changing health technology environment in Texas. In particular, the statewide study of nur.ses' satisfaction with their use of clinical information systems assesses the experience of nurses with their EHRs (McBride et al., 2015). A descriptive exploratory study using the Clinical Information System Evaluation Scale (CISIES) and a newly developed Demographic Survey and Meaningful Use Maturity-Sensitive Index (MUMSI), with a narrative component, was conducted in 2014 and 2015. Nurses across Texas received an electronic invitation to
participate in the survey, resulting in 1,177 respondents. Further immersion and analysis of the categories with the two lead researchers with informatics expertise was done and resulted in a synthesis of comments within the categories that revealed several primary themes noted in Figure 30.1. A conceptual model was developed by the research subgroup that reflects the overall common concepts detected in the thematic analysis, also noted in Figure 30.1. Note that "communication" is a major theme with a minor theme "reduced consultation among clinicians."
765
766
V: NEW AND EMERGING TECHNOLOGIES
FIGURE 30.1 Narrative Themes From Qualitative Data: Collected as Part of theTexas
Nurses Association/Texas Organization of Nurse Executives Survey. NARRATIVE THEMES REGARDING NURSE EXPERIENCES WITH CIS/EHR Major and minor themes
System design/usability
Intormation system Ign
Patient safety and quality
c/ Organ cu urs/
Tune tonal
sociaNustIce
Support Workflow
Distress
Communication
interoperability Documentation/legaitty
Nurse-patient lime reduced/work inefficiency Competency of IT, administrative/leadership Medication admin., other work-arounds
Aggravation, voice not heard
Reduced consultation among clinicians
CIS, clinical information system; EHR, electronic health records.
COMMUNICATION
The following shows examples of how EHR-related issues with interprofessional communication are evident in some of the free-text their EHRs.
comments provided by nurses about
"The concept of EHR is a good one. The practicality is not, especially as the documentation requirements for core measures, etc., is increased. All areas need to be on board for all aspects of the system to be effective. Members of the healthcare team 7W longer communicate and rely on a EHR to do the communication for them. It affects a7\d delays the care of the pyatient. While I agree uyith the concept and all the safety features available in a EHR, I do feel the delivery of care is hindered as well as the communication amongst the healthcare team." "Multiple systetns, different modules don't co7?wmnicate." "Our outpatie7it record is different and the tzvo do 77ot commimicate. This is a total disaster." "Our facility needs an updated syste7n. We docut77ent in txuo different databases and information is nof shared with pyliysicmi's offices." "The syste77i we use (XYZ vendor) has iwt C077ipletely worked as it should. Currefitly, we have 7nany billuig problems as the 'f7iodules' within the syste7U do not C077i7mu7icate
effectively. The nursing assess7nents are 7Wt visible for our doctors to k770W the status of their patients. This system is very 'frag777e7ited' a7id 07ie has to go to 777a7iy different 'zoi77dozLys' to get a good picture of zohat is going on zvith the patient."
IPE occurs "when students from two or more professions learn about, from and
with each other to enable effective collaboration and improve health outcomes" (WHO,
2010, p. 7). The other report that supports IPE effort is a 2011 report of an expert panel (Interprofessional Collaborative Expert Panel [ICEP], 2011). A more recent 2016 version of this 2011 report, also of an expert panel, updates some of the initial content (Interprofessional Education Collaborative, 2016). Both of these reports, for example, illustrated that a community- and population-oriented approach is central to the IPE model (see Figure 30.2). The reports emphasize that teamwork, communication value/ ethics, and roles are the actions of the IPE-skilled health professionals from the trajectory of prelicensure through practice.
30: INTERPROFESSIONAL APPLICATION OF HIT
FIGURE 30.2 interprofessional Collaborative Practice Domains.
Source: Interprofessional Collaborative Expert Panel. (2011). Core competencies for interprofessional collaborative practice: Report of an expert panel (No. ICEP-2011). Interprofessional Education Collaborative. https://www.aacom.org/docs/default-source/insideome/ccrpt05-10-1 1.pdf?sfvrsn=77937f97_2,
It has been suggested that the challenges of health systems are fundamentally ethical. These ethical principles consider health and healthcare a right. These principles support balance in the distribution of resources for health to both individuals and populations.
Thus, cooperation is seen as the central tenet in achieving this principle (ICEP, 2011). Figure 30.3 illustrates the four competency domains of IPE, which in the 2016 report were expanded upon by the addition of sub-competencies, appearing in bold in the figure
(Interprofessional Education Collaborative, 2016, p. 13); Competency Domain 1: Values/ ethics for interprofessional practice clearly addresses this issue. The background and rationale of related ethics are an important, new part of crafting a professional identity, one that is both professional and interprofessional in nature. As noted, these values and
ethics are patient centered with a community/population orientation, grounded in a sense of shared purpose to support the common good in healthcare, and reflect a shared commitment to creating safer, more efficient, and more effective systems of care. All four of the IPI competencies influence the use of information technology by
balancing individual and team competencies (Troseth, 2017). For example, the addition of data driven out of machine learning/artificial intelligence (ML/AI), may be heavily influenced by interprofessional teams and teamwork, communication, values/ethics?.To foster a culture of interprofessional collaboration for the complexity of such work, teams must be brought together to dialogue, build understanding, and create new ways of being and practicing together (Troseth, 2017, p. 16). Studies have indicated that increasing trustful dialogue and interaction is important to ensuring safe technology development and quality care (Barr et al., 2017; Verville et al., 2017). The relationship among the four main competencies of the IPE model and the work of the IPE team is illustrated in Figure 30.4. Providing patient-centered care is the core of the competencies supported by the other three competencies: utilize informatics, employ evidence-based practice, and apply quality improvement.
767
768
V: NEW AND EMERGING TECHNOLOGIES
FIGURE 30.3 Interprofessional Collaborative Practice Competency Domains:The four competency domains of interprofessional education, with subcompetencies appearing in bold. Competency 1 Work with Individuals of other professions to maintain a climate of mutual respect and shared values. (Values/Ethics for Interprofessional Practice) Competency 2
Use the knowledge of one's own role and those of other professions to appropriately assess and address the health care needs of patients and to promote and advance the health of populations. (Roles/Resprxisib ilities) Competency 3
Communicate with patients, families, communities, and professionals in health and other fields in a responsive and responsible manner that supports a team approach to the promotion and maintenance of health and the prevention and treatment of disease. (Interprofessional Communication) Competency 4
Apply relationship-building values and the principles of team dynamics to perform effectively in different team rotes to plan, deliver, and evaluate patient/popuiationcentered care and population health programs and policies that are safe, timely, efficient, effective, and equitable. (Teahis and Teamwork)
Source: Core competencies for interprofessional collaborative practice: 2016 Update (No. ICEP-2016). Interprofessional Education Collaborative. https;//hsc.unm.edu/i pe/resources/ipec-2016-core-competencies.pdf.
FIGURE 30.4 InterprofessionalTeamwork and Institute of Medicine Core Competencies.
Source: Interprofessional Collaborative Expert Panel. (2011). Core competencies for interprofessional collaborative practice: Report of an expert pane/(No. ICEP-2011). Interprofessional Education Collaborative. https://www.aacom.org/docs/default-source/insideome/ccrpt05-10-1 1.pdf?sfvrsn=77937f97_2.
RELATIONSHIP BETWEEN INTERPROFESSIONAL EDUCATION AND HEALTH INFORMATION TECHNOLOGY
As noted, informatics is one of the concepts of the interprofessional model suggested
by the expert panel (ICEP, 2011). In addition, ubiquitous in today's healthcare delivery environment, informatics is a key facilitator, for example, in support of both
30: INTERPROFESSIONAL APPLICATION OF HIT
communication and clinical simulation. Aside from the
faculty and students, patients
and their families benefit from interprofessional IT. Stakeholders Patients/Families
Engaging patients and families in quality improvement efforts is becoming more commonplace as studies indicate that engaged patients/families yield better outcomes more efficiently and at less cost (Hibbard & Greene, 2013). Technology can facilitate this process, especially when used in an interprofessional team. This multifaceted approach to using the
technology among IPE professionals as well as with patients in support of their medical management is ideal. One example of such a program is the Partnership for Patients program created by the Centers for Medicare & Medicaid Services (CMS, 2014) where providers partner with patients in shared decision-making models supported by technologies. Providers
Providers such as nurses, physicians, and healthcare organizations can also benefit from interprofessional collaborative practice supported by technology. Some providers use the interprofessional approach of psychiatric care delivery along with IT (Akroyd et al., 2014). Others have similarly experienced such success through rapid response teams in the emergency department (ED; Allen, Jackson, & Elliott, 2015). In other situations, faculty have benefited from the use of technology in interprofessional teaching (King et al., 2012; Paquette-Warren et al., 2014; Pfaff et ah, 2014; Pulman et ah, 2009). Another such example is seen in post-acute care services such as home health and remote patient monitoring for patients in their homes. CareCycle Solutions provides case managers from nursing, physical therapy (PT), and respiratory therapy backgrounds who work together collaboratively in a virtual environment supported by a data warehouse and a decision support system. From day 1, the training of these professionals is interprofessional and involves technology. In addition, CareCycle Solutions utilizes information from its data warehouse to support business decisions involving accountable care organizations (ACOs) and other healthcare insurance entities (Noble & Casalino, 2013; Torres & Loehrer, 2014). Suppliers/Vendors
The role of technology vendors in interprofessional education/collaboration (IPE/C) is to understand the needs of these initiatives. An example of technology applied to interprofessional practice is to have an application available locally to provide the
interprofessional information needed (Youm & Wiechmann, 2015). Vendors can support communication efforts among providers engaged in interprofession al collaborative practice by using a smartphone (Djukic et al., 2012; King et al., 2014; Peluchette et al., 2012; Smith, 2014; Youm & Wiechmann, 2015).
Since 2015, vendors and practitioners have been partnering in an interprofessional way to focus on the needs of the older population. The proliferation of wearable technology such as smart watches, fitness trackers, or wearable microprocessors that integrate with smartphones have given patients and healthcare professionals unprecedented access to real time health data. Smart technology can capture and transmit electrocardiogram wave forms for those individuals with cardiac issues or provide instant, pain free monitoring of blood sugar levels for diabetics. Providers can accurately track, monitor, and evaluate patient conditions in real time, without using traditional telemedicine practices. Vendors
769
770
V: NEW AND EMERGING TECHNOLOGIES
and practitioners have been partnering in an interprofessional way to foais on the needs
of the older population. Reports indicate that those people aged 65 years and older are expanding at a faster rate than there are younger people to care for them (Ortman et al., 2014). Studies also show that these older populations have adopted technology and use remote care technology to take better care of themselves (Intel-GE Care Innovations, 2015). These trends are a driving force for recent collaborative approach between innovative technology and interprofessional care delivery. One example of a blend between older populations, innovative remote technology, and better healthcare delivery is the Kinesis QTUG product (Kinesis Health Technologies, 2018). QTUG represents the standard "Timed Up and Go" test that employees perform on patients. The employees use these body-worn sensors, applied to the patient, and associated assessment by ML algorithms. Kinesis QTUG is used for the assessment of falls risk, frailty, and mobility. Health professionals or social care workers can utilize QTUG for assessing mobility and falls risk of older adults via body-worn inertial sensors along with questionnaire data. A detailed 69 parameter gain and mobility assessment estimates level of fall risk by comparing to a large reference data set (Kinesis Health Technologies, 2018). This much information optimally benefits the patient when an interprofessional collaborative approach is used; for example, among social work, nursing, FT, and occupational therapy (OT). APRNs should know that the interprofessional collaborative approach along with QTUG services reduce falls (Greene & Kenny, 2012), prevent readmissions (Care Innovations, 2018), and improve quality of life for the frail elderly (Wuest et al., 2016). Unfortunately, the rate of development of innovative technologies coupled with continued reports of information data breeches, hacking, and information ransoming, are creating an insecurity and lack of trust in many new technologies. A lack of understanding by individuals creates a mistrust in technology. Although wearable technology has changed acce.ss to immediate information for the better, distrust in the
technologies ability to protect their personal health information has limited the universal acceptance of these systems (Ramsetty, & Adams, 2020). Team Practice/Simulation
Progress in stabilization and accessibility of information and communication technology
have allowed for more widespread use by the organizations, as well as by the general
public. One of those advancements has been in the area of simulation methodologies in education. Sometimes called e-learning technologies, they can prove beneficial to both faculty and students (Carbonaro et al, 2008; King et al, 2012: Palaganas et al 2020). In these scenarios, interprofessional teams process skills development, pedagogical integrity, and instructor/faculty balance between face-to-face and online interaction and student perspectives can be assessed. In addition to this approach for simulation
practice, students also benefit from interprofessional practice with skills development using a polarity thinking framework (Adams, et al, 2019, p. 1). With polarity thinking as a framework, the IPE roles are strongly supported because each independent value of team members is neutral versus negative or positive. Usability testing of IT is another aspect of simulation .supporting interprofessional practice. An example is the medical simulation center for an EHR laboratory (Landman
et al, 2014). This approach is becoming an important component of safety testing of IT; however, it is also a viable approach used to engage multiple di.sciplines and department staff in testing. Once such a laboratory is set up, it can be used by all stakeholders, thereby favorably addressing issues such as return on investment, cost, and benefits.
The use of in-situ simulation methods can test HIT systems in real-time and witliin
the real environment. For example, healthcare systems may use in-situ simulations to
30; INTERPROFESSIONAL APPLICATION OF HIT
assess both the operational readiness of HIT systems and the employee's ability to safely, efficiently and accurately interface, (access, input, export, and transmit data) with the HIT
systems within a new working environment. This same process also provides stakeholders the ability to assess how newly formed provider team's and departments integrate, communicate, form small time limited work teams, and collaborate to provide safe patient
care within the new hospital environment. Using simulation methods such as this, gives .stakeholders information on:
■ Ease of use and accessibility limitations of HIT systems, ■ Environmental issues with the arrangement of equipment and necessary tools for patient care,
■ Errors in system programming or identifying limitations in capabilities of systems, ■ Process improvement opportunities with both HIT and patient care systems (Palaganas, et al, 2020).
Regarding team practice/simulation, one program has evolved where students can participate in an established "interprofessional didactic and experiential project" for course credit (Texas Tech University Health Sciences Center [TTUHSC], 2017). Patient safety and quality is the focus of the project, and the domain content is clinical workflow improvement. Students from biomedical science, medicine, nursing, pharmacy, and other health professions are in the course together, experiencing the interprofessional approach to communication, project management, and ultimately problem solving. Interprofessional teams are created from the students in the course and the teamwork exercises are conducted throughout the year. Students are assigned preceptors in a hospital setting as a way to practice skills development (TTUHSC, 2017). For further information. Dr. Susan McBride can be contacted at [email protected].
CONSIDERATIONS FORTHE FUTURE
As discussed, many organizations exist with the focus on advancing IPE/Cs, and those should continue to be followed as sources to advance IPE initiatives. In addition, at
least three organizations represent efforts to infuse IT into interprofessional-based education.
Organizations to Support the Process TheTechnology Informatics Guiding Education Reform Initiative
Technology Informatics Guiding Education Reform (TIGER) is an interprofessional community focused on providing educational support for the advancement of technology used for optimal healthcare delivery. TIGER began in 2006 within the nursing community and in 2014 transitioned to be part of the Health Information and Management Systems Society (HIMSS) with an interprofessional approach spanning clinical and administrative disciplines (Shaw et al., 2017). As noted in the "core" section of the 10-year vision (Figure 30.5), TIGER was envisioned as interdisciplinary (Shaw et al., 2017). TIGER also provides the virtual learning environment (VLE; Kuo et al., 2018). The TIGER VLE, powered by HIMSS, is a dynamic and unique one-stop online portal for academic professionals, students, adult learners, and clinical educators. The VLE contains competencies to take one from A to Z in HIT. This personalized learning experience is designed to expand skill sets in a self-paced format. On the VLE home page, one may integrate readily available HIT
vetted resources that are reflective of core international
modules and resources into the current curriculum.
771
772
V: NEW AND EMERGING TECHNOLOGIES
FIGURE 30.5 Ten-Year Vision of theTechnology Informatics Guiding Education Reform (TIGER) Initiative Includes Interdisciplinary Education and Collaboration. (To YEAB VISION.!
«F tMsa.iKftefeMu
f^Su«6>ice vtfni*s n*r ISwer etWATxW t fucna
Of
^ JiHP ?ArgTl}
6DUCAcTI^?M
●A
,yKfc«K^ Awvi
wflwancing health (No. FoN20H). National Academies Press. http://www.nationalacademies.org/hmd/R eports/2010/The-Future
-of-Nursing-Leading-Change-Advancing-Health.aspxx Intel-GE Care Innovations. (2015). Older populations have adopted technology for health: People over 65 zoill use remote care technology to take better care of themselves. (1 No. 1). Intel-GE Care Innovations, LLC. Interprofessional Collaborative Expert Panel. (2011). Core competencies for interprofessional collaborative practice: Report of an expert panel (No. ICEP-2011). Interprofessional Education Collaborative, https: / /
www.aacom.org/docs/default-source/insideome/ccrpt05-10-ll.pdf?sf vrsn=77937f97_2 Interprofessional Education Collaborative. (2016). Core competencies for interprofessional collaborative practice: 2016 update. (Report No. IPE2016). Interprofessional Education Collaborative, https://hsc .unm.edu/ipe/resources/ipec-2016-core-competencies.pdf Interprofessional Education Collaborative Expert Panel. (2021). Strategic Plan, Interprofessional Education Collaboration, www.ipecollaborative.org /strategic-plan. Kinesis Health Technologies. (2018). Kinesis QTUG, mobility and falls risk assessment technology: Evidence base (1 No. 1). Kinesis Health Technologies. King, S., Carbonaro, M., Greidanus, E., Ansell, D., Foisy-Doll, C., & Magus, S. (2014). Dynamic and routine interprofessional simulations: Expanding the use of simulation to enhance interprofessional competencies. Journal of Allied Health, 43(3), 169-175. King, S., Chodos, D., Stroulia, E., Carbonaro, M., MacKenzie, M., Reid, A., & Greidanus, E. (2012). Developing interprofessional health competencies in a virtual world. Medical Education Online, 17, 1-11. https://doi.org/10.3402/meo.vl7i0.11213
Kuo, M., Ball, M., Skiba, D.Marin, H., Shaw, T, & Chan^ P. (2018). Technology informatics guiding
education reform (TIGER), the future of interprofessional education is here. Studies in Health Technology and Informatics, 250, 65-66. https: / /www.ncbi.nlm.n ih.gov/pubmed/29857374 Landman, A. B., Redden, L., Neri, R, Poole, S., Horsky, J., Raja, A. S., & Poon, E. G. (2014). Using a medical simulation center as an electronic health record usability laboratory. Journal of the American Medical Informatics Association, 21(3), 558-563. https: / / doi.org/10.1136/amiajnl-2013-002233 Lee, J. K., McCutcheon, L. R. M., Fazel, M. T„ Cooley, J. H., & Slack, M. K. (2021). Assessment of interprofessional collaborative practices and outcomes in adults with diabetes and hypertension in primary care: A systematic review and meta-analysis. JAMA Netxoork Open, 4(2), e2036725. https: / / doi.org/10.1001 /jamanetworkopen.2020.36725 McBride, S. G., Tietze, M., & Fenton, M., V. (2013). Developing an applied informatics course for a doctor of nursingpracticeprogram.N«rse Educator, 38(1 ),37-42.https://doi.org/10.1097/NNE.0b013e318276df5d McBride, S, G., Tietze, M., Hanley, M. A., & Thomas, L. (2015). Stateioide study assessing the experience of nurses xoith their electronic health records. Texas Nurses Association, https://c.ymcdn.com/sites/ www.texasnurses.org / resource / resmgr /HIT_Files / HIT_Survey_Report-Final.pdf MedBiquitous Consortium. (2015). MedBiquitous standards make it easier to track professional achievements, access learningand measure improz’ements. https://education2prac tice.org/interprofessional-toolkit National Academies of Sciences Engineering and Medicine. (2021a). The future of mirsing 2020-2030: Charting a path to achiez’e health equity. (Report No. 2). The National Academies Press, https;//doi .org/10.17226/25982 National Academies of Sciences Engineering Medicine. (2021b). Global forum on innoz’ation in health profossional education. NationalAcacemies.org. https://www.natio nalacademies.org/our-work/global -forum-on-innovation-in-health-professional-education Noble, D. & Casalino, L. P (2013). Can accountable care organizations improve population health?: Should they try? Journal of the American Medical Association, 309(11), 1119-1120. https;//doi
.org / lO.lOOi / jama.2013.592
Ortman, J. M., Velkoff, V. A., & Hogan, H. (2014). An aging nation: The older population in the United States. https: / /www.census.gov/prod/2014pubsZp25-1140.pdf Palaganas, J., Ulrich, B., Mancini, M, (2020). Foundations of simulation. Mastering simulation, 2nd Ed.: A handbook for success. Sigma Theta Tau International, Indianapolis,IN. Paquette-Warren, J., Roberts, S. E., Fournie, M., Tyler, M., Brown, j., & Harris, S. (2014). Improving chronic care through continuing education of interprofessional primary healthcare teams: A process evaluation. Journal of Interprofessional Care, 28(3), 232-238. https://doi.org/10.3109/13561820.2013 .874981
Peluchette, J., Karl, K., Coustasse, A., & Emmett, D. (2012). Professionalism and social networking: Can
patients, physicians, nurses, and supervisors all be "friends?" Health Care Manager, 31(4), 285-294. https://doi.org/10.1097/HCM.0b013e31826fe252
30: INTERPROFESSIONAL APPLICATION OF HIT
Pfaff, K., Baxter, P., Jack, S,, & Ploeg, J, (2014). An integrative review of the factors influencing new graduate nurse engagement in interprofessional collaboration. Journal of Advanced Nursing, 70(1), 4-20. https://doi.org/10.1111/jan.l2195 Pulman, A., Scammell, J., & Martin, M. (2009). Enabling interprofessional education: The role of technology to enhance learning. Nurse Education Today, 29(2), 232-239. https://doi.org/10.1016/] .nedt.2008.08.012
Ramsetty, A., & Adams, C. (2020). Impact of the digital divide in the age of COVID-19. Journal of the American Medical Infonnatics Association: JAMIA, 27(7), 1147-1148. https://doi.org/10.1093/jamia/ ocaa078
Regents of the University of MinnCvSota. (2021). About interprofessional practice and education: The "new" IPE. NexusIPE.org. https://nexusipe.org/informing/about-ipe
Shaw, X, Sensmeier, J., & Anderson, C. (2017). The evolution of the TIGER initiative. Computers, Informatics, Nursing, 35(6), 278-280. https://doi.org/10.1097/CI N.0000000000000369 Smith, K. A. (2014). Healthcare interprofessional education: Encouraging technology, teamwork, and team performance. Journal of Continuing Education in Nursing. 45(4), 181-187. https://doi.org/ 10.3928/00220124-20140327-01
Texas Tech University Health Sciences Center. (2017). Redesigning clinical workflow to optimize patient safety: An interprofessional didactic and experiential project. (1 No. 1). Author. Torres, X, & Loehrer, S. (2014). ACOs: A step in the right direction. Accountable care may achieve better care at lower cost. Healthcare Execulwe, 29(4), 62-65.
Troseth, M. (2017), Interprofessional collaboration through technology. Nursing Management, 48(8), 15-17. https;//doi.org/10,1097/01.NUMA.0000521583.55623.c0
Verville, ]., Palanisamy, R., & Taskin, N. (2017). Impact of trust and technology on interprofessional collaboration in healthcare settings; An empirical analysis. International Journal of E-Collaboration, 13(2), 10-44. https://doi.Org/10.4018/IJeC.2017040102 Wilbur, L. (2014), The University of Arkansas' five-pillar plan for an institutional Triple Aim culture. https; / / nexusipe.org / news / university-arkansas%E2%80%99-fi ve-pillar-plan-institutional-triple -aim-culture
World Health Organization. (2010). Framework for action on interprofessional education & collaborative practice (No. WHO/HRH/HPN/10.3). Author. Wuest, S., Masse, F., Aminian, K., Gonzenbach, R., & de Bruin, E, D, (2016). Reliability and validity of the inertial sensor-based timed "up and go" te.st in individuals affected by stroke. Journal of Rehabilitation Research and Development, 53(5), 599-610. https; / / doi.org/10.1682/JRRD.2015.04.0065 Youm, J,, & Wiechmann, W, (2015), The med Appjam: A model for an interprofessional student-centered mHealth app competition. Journal of Medical Systems, 39(3), 34, https; / Zdoi.org/10.1007/sl0916-015 -0216-4
783
J ABBREVIATIONS
ATA
American Telemedicine Association
BA
business associate
BAA
ABC
American Academy of Anrbulatory Care Nursing American Academy of Colleges of Nursing altemate billing codes
BAM
business associate agreement binary alignment/map
ACA
Patient Protection and Affordable Care Act;
AAACN AACN
BC
birth certificate
also know as the Affordable Care Act
BCMA
bar codes medication administration
or Obamacare
BCP
BPA
AD
advancing care information accountable care organization active directory
ADA
Americans with Disabilities Act
CAB
ADT
CAM
AF4Q
admit-discharge-transfer Aligning Forces for Quality
business continuity plan Blu-ray disc business intelligence biomedical health informatics professionals body mass index best practice alert Behavioral Risk Factor Surveillance System bring your own device collaborating across borders critical access hospital
CAHIIM
Commission on Accreditation for Health
AHA
American Heart Association
AHIMA
American Health Information Management
ACB
authorized certification bodies
BD
ACE
angiotensin-converting enzyme
Bl
ACGME
Accreditation Council for Graduate Medical
BMHI
Education ACI ACO
Association AHRQ
BMI
BRFSS BYOD
Informatics and Information CARES
Management Act Coronavirus Aid, Relief, and Economic
CASE
Competency Assessment in Simulation of
CBC
Colorado Beacon Consortium
CBO CBOs
Congressional Budget Office community-based organizations
Security Act
Agency for Healthcare Research and Quality
Electronic Health Records
AIHC
Agency for Healthcare Research and Quality-Prevention Quality Indicators artificial intelligence American Interprofessional Health
CCC
Clinical Care Classification
AMA
American Medical Association
CCC
Connected Communities of Care
AMI
acute myocardial infarction
CCD
continuity of care document
AMIA
American Medical Informatics Association
CCHP
ANA
American Nurses Association
CCM
Center for Connected Health Policy chronic care management
ANCe
American Nurses Credentialing Center American Nursing Informatics Association
CCR
continuity of care record
ANIA
CDA
clinical document architecture
ANSI
American National Standards Institute
CDC
Centers for Disease Control and Prevention
ANT
CDISC
Clinical Data Interchange Standards
CDM
clinical decision-making;
ARRA
Actor-Network Theory application program interface alternative payer models Alternative Payment Models advanced practice registered nurse American Recovery and Reinvestment Act
ASPE
Association of Standardized Patient
CDS
AHRQ-PQIs
Al
Collaborative
API
APMs APMs
Consortium
CDMH
common data model harmonization
CDOs
ASQ
American Society for Quality
CE
ASTHO
Association of State and Territorial Health
CEHRT
care delivery organizations clinical data repository clinical decision support clinical decision support system covered entity certified EHR technology
CEN
European Committee for Standardization
CEO
chief executive officer
APRN
Educators
Officials ASTM
American Society for Testing Materials
CDR
CDSS
785
786
ABBREVIATIONS
CFR CGM CHA
CHAID CHANGE
Code of Federal Regulations continuous glucose monitor coinmunity health assessment chi-square automatic interaction detection Community Health Assessment and Group Evaluation
CHCS CHIN CHIP CHIP
CHMIS
DSHS
Department of State Health Services
DSRIP
delivery system reform incentive payment
DTC
direct-to-consumer
DURSA
Data Use and Reciprocal Support Agreement dependent variable evaluation and management evidence-based practice eligible clinician
DV
E&M
composite healthcare system community health information network
EBP
Children's Health Insurance Program community health improvement plan community health management information
EClC
Evidence Communication Innovation
eCQM
electronic clinical quality measures Emergency Care Research Institute emergency department
system
EC
Collaborative ECRI
CHNA
community health needs assessment
Cl
confidence intervals
EDI
electronic data interface
CIHC
Canadian Interprofessional Health
EDRAP
Enterprise Data Management, Reporting, and Analytics Program
EDT
electronic data transmission
EDW
enterprise data warehouse enterprise environment factors eligible hospital eHealth Exchange
ED
Collaborative CIMIT CIS CISIES
Center for Integration of Medicine and Innovative Technology clinical information system Clinical Information System Implementation Evaluation Scale
EEF EH eHEX
CME
continuing medical education
EHI
CMMI
Center for Medicare and Medicaid Innovation
EHR
CMPs
civil money penalties
EHR-CDS
CMS
Centers for Medicare & Medicaid Services
EHR-MU
CNIO
chief nursing informatics officer
EIN
CNM
certified nurse-midwife
ELR
CNO
chief nursing officer certified nurse practitioner clinical nurse specialist community of practices commercial off-the-shelf products
CNP CNS COPS COTS
ELT
extract, load and transform electronic medication administration record
EMPI
enterprise master patient index
EMR
electronic medical record
EMRAM
electronic medical record adoption model engineered nanomaterial eligible provider essential public health services electronic submission request for medical
computers on wheels
ENM
CPHIMS
Certified Professional in Healthcare
EP
CPOE CPT CPU CQL COM CRiSP-DM
CRNA
CSF CVl CY
EHR of clmical decision support electronic health record meaningful use employer identifier niunber electronic laboratory reporting
eMAR
COWS
Information and Management Systems computerized provider order entry current procedural terminology central processing unit clinical quality language clinical quality measures Cross-Industry Standard Process for Data Mining certified registered nurse anesthetist common security framework content validity index calendar year
electronic health information electronic health records
EPHS esMD
document EU FACA FCM FDA
European Union Federal Advisory Committee Act Four Component Model U.S, Food and Drug Administration
FFS
fee for service
FHIMSS
Fellow of the Healthcare lirformation and
FHIR FMEA
Management Systems Society fast healthcare interoperability resources failure mode effect analysis family systems theory
DB
database
FST
DCHHS
Dallas County Health and Human Services data flow diagrams U.S. Department of Health and Human
FTC
Federal Trade Commission
GAO
Government Accountability Office
GINA
Genetic Information Nondiscrimination Act
GIS
geographical information system graphic processing units global language of busmess Genetic Testing Registry graphic user interface genome-wide sequencing hospital acquired condition
DFD DHHS
Services DICOM
Digital Imaging and Communication in
DIKW
Data Information Knowledge to Wisdom distance learning and telemedicine data-mining laboratory doctor of nursing practice
GTR
Department of Defense Department of Justice doctor's office quality-information technology doctor of physical therapy diagnosis-related groups
HAI
healthcare-associated infection
HCCN
health center controlled network
HCI
human-computer interaction Healthcare Common Procedure Coding
Medicine DLT DML DNP DoD
DOJ DOQ-IT
DPT
DRGs
GPUs GS1 GUI GWS HAC
HCPCS
System HCPFC
Health Care and Promotion Fund Committee
ABBREVIATIONS
787
HCUP
Healthcare Cost and Utilization Project
ISO
International Organization for
HDD HDO
hard disk drive; hospital discharge data healthcare delivery organization health IT health information technology
ISTA
HEDIS
Healthcare Effectiveness Data and
IT-Hl
interactive sociotechnical analysis information technology Interprofessional Team for Health Informatics
Standardization IT
Model
Information Set HI
health informatics
IT-Hl
HIE
health information exchange
JEDEC
HIL
health information law
JMJF
HIM
health information management Health Information and Management Systems Society
KPis
HIMSS
KSAs
Interprofessional Team for Health Informatics Joint Electron Device Engineering Council Josiah Macy, Jr. Foundation key performance indicators knowledge representation knowledge, skills, and abilities
HINs
health information networks
LAN
local area network
HIPAA
LCL
lower control limit
LGBTQ
LOINC
lesbian, gay, bisexual, transgender, questioning Lab-on-a-chip Logical Observation Identifiers Names and
HITECH
Health Insurance Portability and Accountability Act health information service provider health information technology Health Information Techonology Competencies Act Health Information Technology for Economic
MA
Medicare Advantage
MACRA
Medicare Access and CHIP
HL7
Health Level 7
HMO
health maintenance organization
MAP
HPAC
Health Professions Accreditors Collaborative
MATH
Measure Applications Partnership Modeling and Analysis Toolsuite for
HQMF
health quality measure format
HRI
Health Research Institute
MC
HRSA
Health Resources and Services
MCO
HISP HIT
HITCOMP
and Clinical Health Act
Codes
Reauthorization Act
Healthcare
MERS-CoV
International Council of Nurses
MEWS
Modified Early Warning Score
International Classification of Nursing
mHealth
mobile health
MHIM
master of health information management master in health systems management Multi-Interprofessional Center for Health
International Classification of Diseases, 9th edition. Clinical Modification
ICD-10-CM
LOG
managed care managed care organization PnP Medical Device Plug-and-Play major diagnostic categories master data management Medical Expenditure Panel Survey Middle Eastern Respiratory Syndrome,
Administration ICD-9-CM
KR
International Classification of Diseases, 10th edition. Clinical Modification
ICEP
Interprofessional Collaborative Expert Panel
ICN ICNP’^
Practice
MD MDGs MDM MEPS
Corona Virus
MHSM
ICT
Interprofessional Collaborative Practice information and communication technology
lEC
International Electrotechnical Commission
IEEE
MIME
IMDRF
Institute of Electrical and Electronics Engineers Institute for Healthcare Improvement International Health Terminology Standards Development Organization International Medical Device Regulators Forum
IMIA
International Medical Informatics Association
MMS
INACSL
mPERS
INS
International Nursing Association for Clinical Simulation and Learning informatics nurse specialist
Massachusetts Institute of Technology machine learning mobile medical application Measures Management System mobile personal emergency response
MPi
master patient index
lOM
Institute of Medicine
MRN
medical record number
loMT
Internet of Medical Things internet protocol interprofessional practice interprofessional collaboration interprofessional education Interprofessional Education Collaborative Expert Panel Institute for Patient- and Family-Centered
MRSA MRTCs Draft 2
methicillin-resistant Staphylococcus aureus Minimiun Required Terms and Conditions
MU
Meaningful Use
ICP
IHI IHTSDO
IP IP IPG IPE
IPEC IPFCC
Care IRB
Institutional Review Board
IRS
Internal Revenue Service
ISA
Interoperability Standards Advisory Information Systems Evaluation Tool Intelligent Sensor Information Processing
ISET ISIP
MICHi
Informatics MIPS MIT ML MMA
Multipurpose Internet Mail Extensions Merit-Based Incentive Payment System
services
Draft 2 MUMSI
Meaningful Use Maturity-Sensitive Index
MUP
meaningful use provider National Association of County and City
NACCHO
Health Officials
NACCHO-MAPP
National Association of County and City Health Officials Mobilizing for Action
NAM
National Academy of Medicine
NBHWC
National Board for Health & Wellness
Through Planning and Partnerships Coaching
788
ABBREVIATIONS
NBS
newborn screening
PATCH
NCBC
National Center for Biomedical Computing National Center for Cognitive Informatics & Decision Making in Healthcare
PC
NCHS
National Center for Health Statistics
NCPDP
National Council for Prescription Drug Programs National Committee for Quality Assurance
NCCD
NCQA
Planned Approach to Community Health personal computing
PCCI
Parkland Center for Clinical Innovation
PCMH POOR
patient-centered medical homes patient centered outcomes research
PCORI
Patient Centered Outcomes Research
PCORNet
Patient-Centered Outcomes Research
Plan-Do-Study-Act Physician Fee Schedule patient-generated health data
Institute
NCRR
National Center for Research Resources
NDF-RT
National Drug File-Reference Terminology
PDSA
NEC
not elsewhere classified
PFS
NEDS
Nationwide Emergency Department Sample
PGHD
NeHC
National eHealth Collaborative
PGHI
patient-generated health information
NEHI
PGS
NHANES
Nursing Education for Healthcare Informatics Nation^ Health and Nutrition Examination Sur\'ey
Personal Genome Service Public Health Accreditation Board
NHIN
National Health Information Network
PHD
NHIS
PHI
NIC
National Health Interview Survey National Health Safety Network nursing informatics Nursing Interventions Classification
NICU
neonatal ICU
NIH
National Institutes of Health
PIP
NIS
National Inpatient Sample
PMBOK
NIST
National Institute of Standards and
PMI
Technology Nurse Licensure Compact
PMML
NHSN Nl
NLC NLM
National Library of Medicine
NLN
National League for Nursing natiual language processing natural language understanding nursing minimum data set nursing management minimum data set Nursing Outcomes Classification National Organization of Nurse Practitioner
NLP NLU NMDS NMMDS
NOC NONFP
Faculties NPHPS
National Public Health Performance Standards
NPI
National Provider Identifier
NPRM
Notice of Proposed Rule Making National Quality Forum National Quality Strategy National Vital Statistics System
NQF NQS NVSS NwHIN
Network
PHAB PHC
primary healthcare personal health data protected health information
PHIN
Public Health Information Network
PHM
population health management
PHR
personal health record
PHRI
Public Health Reporting Initiative Promoting Interoperability Program Project Management Body of Knowledge Project Management Institute predictive model markup language project management professional
PMP
or
PMP
PNDS PoC PPCPs PPPY PQI PROS PSI PSOs PT PULSE
QCQI QDM QDs
Promoting Interoperability Program Perioperative Nursing Data Set point of care priority primary care providers per patient per year Prevention Quality Indicator patient-reported outcomes patient safety patient safety organizations physical therapy/therapists Patient Unified Lookup System for Emergencies Quantum Computing Quantum Information Quality Data Model quantum dots
Nationwide Health Information Network Outcome and Assessment Information Set
QHIN
OCR
optical character recognition
Ql
OHDSI
Observational Health Data Sciences and
QIOs
OHSU
QPP
OIG
Oregon Health & Science University Office of Inspector General
OMIM
Online Mendclian Inheritance in Man
QSEN
OMOP
Observ'ational Medical Outcomes Partnership
QTF
Qualified Technical Framework Draft 1
ONC
Office of the National Coord inator for Health
QTUG
Information Technology ONC Issue Tracking System
RAD
Quantitative Assessment for Timed-Up and Go rapid application development
outpatient Organizational Process Assets operating room objective structured clinical examinations
RCA
OASIS
Informatics
ONCITS OP OPA
OR OSCE
QHP
QIRB QRDA
RAM
RCE REC
RFI
OT
occupational therapy
RFP
OTC
over-the-counter
RFQ
OTs
occupational therapists physician assistants
RHIOs
PAs
RIBA
Qualified Health Information Network Qualified Health Plan
quality unprovement quality-improvement organizations Quality Improvement Review Boards Quality Payment Program quality-reporting data architecture Quality and Safety Education for Nurses
random access memory root cause analysis Recognized Coordinating Entity regional extension center request for information request for proposal request for a quote Regional Health Information Organizations Robot for Interactive Body Assistance
ABBREVIATIONS
ROl
return on investment
R-PACS
radiology-picture archiving and communication systems remote patient monitoring/management Risk Priority Number Real-Time Location System
RPM RPN RTLS
SAFER
Robert Wood Johnson Foundation Standards and Interoperability Safety Assurance Factors for EHR Resilience
SAMHSA
Substance Abuse and Mental Health Services
SANICS
Self-Asssessment of Nursing Informatics Competency Scale statistical analysis software State Ambulatory Surgery and Services
RWJF S&l
Administration
SAS SASD
Database
789
STEEEP
Safe, Timely, Effective, Efficient, Equitable,
STS
TEPs
Science and Technology Studies System Usability Scale strengths, weaknesses, opportunities, threats Technology Acceptance Model Transforming Clinical Practice Initiative Trusted Exchange Framework Trusted Exchange Framework and Common Agreement technical expert panels
THClC
Texas Health Care Information Council
THS TIGER
Texas Health Steps Technology Informatics Guiding Education
TJC
The Joint Comnrission
Patient-centered sus SWOT TAM
TCPI TEF
TEFCA
Reform
SCIP
Surgical Care Improvement Project
TNA
Texas Nurses Association
SD
standard deviation
TONE
SDLC
TONL
Texas Organization of Nurse Executives Texas Organization for Nurse Leaders
SDO
system development life cycle standing delegated order sot
TRCs
telehealth resource centers
SDOH
social determinants of health
TTUHSC
Texas Tech University Health Sciences
SDOs SEDD
standards development organizations State Emergency Department Databases
TURF
SGR
Sustainable Growth Rato
UCL
SHARP
Strategic Health IT Advanced Research
UDP
Projects Strategic Health IT advanced research projects on Security Strategic Health Information Exchange
UlUC UMLS
Task, User, Representation and Function upper control limit unified data protection University of Ulinois at Urbana—Chanrpaign Unified Medical Language System
URI
Uniform Resource Identifiers
URL
uniform resource locator
USCDI
United States Core Data for Interoperability
USPHS
U.S. Public Health Serv’ice
SIG
State Inpatient Databases special interest group
UT
SIM
State Innovation Models
UTA
SIPOC
stands for suppliers, inputs, process, outputs,
UTAUT
University of Texas University of Texas at Arlington Unified Theory of Acceptance and Use of Technology
SHARPS SHIEC
Collaborative SID
Center
customer
SIR SMART SME SML SMS
SMTP SNOMED
Susceptible, Infected and Recovered Substitutable Medical Applications, Reusable Technologies subject matter expertise supervised machine learning short message service Simple Mail Transfer Protocol Systematized Nomenclature of Medicine Systematized Nomenclature of Medicine-
VA
Veterans Affairs
VBP
value-based payment
VCF
Variant Call Format
VDT
view, download, transmit
VI VLE
Vulnerability Index virtual learning environment
VOC
voice of customer
VSAC
Value Set Authority Center
WANs
wide area networks
W8S
Work Breakdown Structure
WGS WHO
whole-genome sequencing World Health Organization
WIA
wisdom-in-action
SSA
School of Nursing Structured Query Language seairity risk assessment Social Security Administration
WOWs
Workstations on Wheels
SSD
solid-state drive
WTO
SSL
secure socket layer
XML
World Trade Organization extensible markup language
SNOMED-CT
Clinical Terms SON SQL SRA
NDEX
Note: Page numbers in italics denote figures and tables. 5C Model, 15-16
acute care workflows, 220
7 Cups of Tea app, 110 21“ Century Cures Act. See Cures Act 23andMe company, 645, 646,657,
ADA. See Americans with Disabilities Act
659
admission, discharge and transfer (ADT), 89, 264,266, 273, 275
25 by 5 initiative, 170
ADT. See admission, discharge and
1000 Genomes Project, 643 1135 waiver authority, 400
advanced directives, 296
transfer
advancing care information (ACI), 369
AAACN. See American Academy of Ambulatory Care Nursing AACN. See American Academy of Colleges of Nursing; American Association of
Colleges of Nursing ABCS, of Million Hearts campaign, 485^86
Accelerated Model for Improvement (Ami^M),551-552 Accenture 2014 Patient
Engagement Survey, 98 Accenture survey (2012), 271 acceptance testing, 195 accountable care organizations (ACOs), 79,89,101,266,347,572 Accreditation Council for Graduate Medical Education
(ACGME), 620 ACGME. See Accreditation Council for Graduate Medical Education
ACI. See advancing care information ACOs. See accountable care
organizations Active Listening, 759 Activity Theory, 52 Actor-Network Theory (ANT), 51, 52,55-56
significance of, 58 users, viewing reality, 56-57 actualization utility, 245
AF4Q. See Aligning Forces for Quality Agency for Healthcare Research and Quality (AHRQ), 5, 57, 101, 210,275,293, 372-374, 490, 574, 582,583
Hazard Manager by, 510-511 Inpatient Quality Indicators, 319 and Prevention Quality Indicators (AHRQ-PQIs), 317, 322
PSOs and, 516
quality indicators, 320 Question Builder app, 374 agile development, 169-170, 394, 195,201,432
AHA. See American Hospital Association AHIMA. See American Health
Information Management Association
AHRQ. See Agency for Healthcare Research and Quality AHRQ.gov website, 210 AI. See artificial intelligence (AI) AIHC. See American
Interprofessional Health Collaborative AIM-AHEAD initiative, 643
Aimi charter, 552,557
actionable improvement efforts, 556, 558
Change Packages, 559 parent, 552 AimiHub, 559 research tabs, 553 website, 552
alert fatigue, 514 alerts, 266
Aligning Forces for Quality (AF4Q), 99,101
All of Us Research Program (AOU), 108,653, 655 all-payer claims data, 574-575 alternative payment models (APMs), 5-6,89,591 AMA. See American Medical Association
Ambulatory Care Nurses Certification Exam, 397
ambulatory care procedures, 292 ambulatory workflows, 219-220 American Academy of Ambulatory Care Nursing (AAACN),397 Board of Directors, 397
American Academy of Colleges of Nursing (AACN), 773 American Academy of Nursing, 690
American Association of Colleges of Nursing (AACN), xxix, 29, 618
American Association of Nurses Practice and Advocacy Resources for Ge}ietics and Personalized Medicine, 660 American Health Information
Management Association (AHIMA),657 American Hospital Association (AHA), 106, 605, 606,733, 734 American Institutes for Research, 57, 522 791
792
INDEX
American Interprofessional Health Collaborative (AIHC), 765
American Lung Association, 108 American Mathematical Monthh/
(journal), 128 American Medical Association
(AMA),112,193,292, 515 Code of Ethics, 644
Physician Innovation Network, 396
policy, on telemedicine, 396 Telehealth Quick Guide, 396 American Medical Informatics
Association (AMIA), 10,246,618 American National Standards
Institute (ANSI), 135,350 American Nurses Association
(ANA), 28,29,46,91, 298-299, 358,394, 397, 658,724
Nursing Informatics Scope and Standards of Practice, 201
recognized data sets and terminologies, 299 American Nurses Credentialing Center (ANCC),201 American Nurses Foundation, 565
American Nursing Informatics Association (ANIA), 10, 507
American Recovery and Reinvestment Act (ARRA) (2009), 7, 77,154, 283,356, 617
American Society for Quality (ASQ), 542
American Society for Testing Materials (ASTM), 289 Americans with Disabilities Act
(ADA), 192 design consideration resources. 192
American Telemedicine
Association (ATA), 29, 394 AMIA. See American Medical Informatics Association Ami'". See Accelerated Model for
Improvement ANA. See American Nurses Association
analytics and reporting, 266 Analytics in Healthcare (Adams and Carets),458 ANCC. See American Nurses
Credentialing Center ANIA. See American Nursing Informatics Association
ANT. See Actor-Network Tlieory
asynchronous telemedicine, 394
AOU. See All of Us Research
ATA. See American Telemedicine Association AUDREY. See Assistant for
Program Apache Hadoop, 460 APIs. See application programming interfaces APMs. See Alternative Payment Models
application programming interfaces (APIs), 8,90,155, 156,265,267,289
Applied Clinical informatics (journal), 270 AQL Injection, 750 architectural and data-exchange models, for HIEs, 260-265 architecture and data
management, 437
ARRA. See American Recovery and Reinvestment Act
artificial intelligence (AI), 50, 136-137,669,703
application in healthcare, 670, 688
augmenting pre-hospital health care, 710
Healthcare, 711
automated test scripts, 196 Azure, 465
BAAs. See business associate
agreements
Baby's First Test, 660 Baptist Health South Florida, 243 barcode medication administration
(BCMA), 45-46,157,159,167, 171
Baylor Health Care System. See Baylor Scott & White Health System Baylor Scott & White Health System, 422,698 BCMA. Set'barcode medication administration
in clinical decision support, 690 clinical settings application of,
Beacon program, 78,82-83, 580
709-714
clinician, 713
implications, to nursing informatics and
interprofessional teams, 676-679
and machine learning, definition and history of, 686-687 transformation, history of, 687 ASPE. See Association of Standardized Patient Educators
ASQ. See American Society for Quality Asset-Based Community Development Institute, 315 Assistant for Understanding Data through Reasoning, Extraction, and Synthesis (AUDREY), 710
Association for Community Health Improvement Toolkit, 311
Association of American Medical
Colleges, 620 Association of Standardized
anomaly detection and data mining, 694
association rule learning and data
ANSI. See American National Standards Institute
Reasoning, Extraction, and Synthesis Augmented Personalized
case categories and, 687-688
Animated Genome, The, 659
Ansatz conflict, 126,127
Understanding Data through
Patient Educators (ASPE), 622 mining, 694
ASTM. See American Society for Testing Materials
BCP. See business continuity plan Behavioral Risk Factor
Surveillance System (BRFSS), 331, 574
Berg Insight, 399 best practice alerts (BPAs), 474 Betty Neuman Model of health, 106
BI. See business intelligence bidirectional maps, 452 big-bang approach, to system implementation, 197 big data, 135-137,419,461 advanced nursing practices and, 708-709
case studies, 709-714
characteristics and descriptions of, 685
data mining and, 691-697 data scientist and, 697-703 definition of, 685-686
nursing knowledge and, 690, 691
as prolific data, 686 software for advanced data-
mining analytics and, 703-708 See also artificial intelligence Big Data Research and Development Initiative, 689 binary digits. See bit binary logistic regression model. 701
INDEX
biome, 749 biomedical health informatics
professionals (BMHI), 628 biorepositories, 651-653 bit, 127-128 bitcoin, 750 Blackboard software, 775 black box, 135 blockchain, 289 framework, 290
HIE challenges and, 291 blogs, 726 blood glucose monitoring, 158 BLS. See Bureau of Labor Statistics
BMC Medical Iiifonnatics and Decision Making, 271 BMHI, See biomedical health
informatics professionals boot loader/ bootstrap, 137 both-and approach, 36-37 BPAs. See best practice alerts BRFSS. See Behavioral Risk Factor
Surveillance System Bridge2AI initiative, 643
Building Communities from the Inside Out (McKnight and Kretzmann), 314 Bureau of Labor Statistics (BLS), 25-28
bus configurations, 140,141,142 advantages and disadvantages of, 142
business associate agreements (BAAs), 353,354-355
business continuity plan (BCP), 164
business data stewards, 438
business intelligence (BI), 418 definition of, 457 KPIs in, 422
library, 418,420,422-423 tools of, 460-467
business-led program management, 425^26 business software, 138
byte, 128 CAB. See Collaborating Across Borders CAHIIM. See Commission on Accreditation for Health Informatics and Information
Management Education California Department of Energy, 669
California Endowment, 318 California Institute for Telecommunication and
Information Technology, 376
Canadian Interprofessional Health Collaborative (CIHC), 765
cancer treatment, nanotechnology drug delivery in, 675 Capgemini, 414 CareConnect EHR system, 200 CareCycIe Solutions, 769 Carequality initiative, 77,259 CARES Act. See Coronavirus Aid,
Relief, and Economic Security Acts
CASE. See Competency Assessment Simulation of Electronic Health Records
cause-and-effect diagram, 523, 540-541 CBC. See Colorado Beacon Consortium
CBOs. See community-based organizations CCC. See Connected Communities of Care
CCD. See Continuity of Care Document
C-chart, 545 CDC. See Centers for Disease Control and Prevention CDISC. See Clinical Data
Interchange Standards Consortium
CDS. See clinical decision support CEHRT. See certified EHR
technology CEN. See European Committee for Standardization (CEN)
Center for Advanced Analytics and E-DRAP framework, 435^38,436 Center for Health Statistics, 574
Center for Integration of Medicine and Innovative Technology (CIMIT), 85 Center for Medicare and Medicaid
Innovation (CMMl), 82,88 Centers for Disease Control and
Prevention (CDC), 78,233, 286, 397,527, 642, 693
Healthy Communities Program, 571
Centers for Medicare & Medicaid
Services (CMS), 6, 77-78, 79, 88,89,154-155,188,189,345, 348, 370,389, 398,572,591, 593, 594,598, 600, 607-608,769
chronic care management (CCM), 400^01
793
measure development principles, 595 measures management system, 582 patient engagement and, 99 Rule, 267
centralized repository model, 260 Cerner Open Development Experience, 289 certified EHR technology (CEHRT), 288,366,370,488 certified nurse midwives (CNMs), 25
certified nurse practitioners (CNPs),25 certified registered nurse anesthetist(CRNA), 25 CHAID. See chi-squared automatic interaction detector
CHANGE. See Community Health Assessment and Group Evaluation
change management, 199-200 sustainability and, 161-162 charts, significance of, 450-451 CHCS. See composite healthcare system check sheets, 540
Chief Nursing Informatics Officer (CNIO), 171 Children's Health Insurance
Program (CHIP), 88, 267,572 chronic care management telehealth toolkit for Medicaid and, 400-401 Children's Medical Center Dallas, 243
Children's Mercy Hospital, 237 CHIME. SeeCOVID-19 Hospital Impact Model for Epidemics CHINs. See Community Health Information Networks CHIP. See Children's Health
Insurance Program; community health improvement plan chi-squared automatic interaction detector (CHAID), 702 tree diagram, 700 CHMIS. See Community Health Management Information Systems chronic care management services, 399-401
background of, 399-400
telehealth toolkit for Medicaid
EHRs and, 232-233
and CHIP, 400^01 chronic disease hierarchical classification, 274
interoperability, and patient
CHSI. See community health status
EHR Incentive Program, 288,689 access, 90, 267
indicators
794
INDEX
Cl. See confidence interval CIHC. St’c Canadian
Interprofessional Health Collaborative
CIMIT. See Center for Integration of Medicine and Innovative
Technology CISIES. See Clinical Information
System Implementation Evaluation Scale
citizen science and community engagement, 108-109
tools and types of, 481-482,481 workflow redesign, 220,488-489 Clinical Decision Support Systems
command interpreter, for 6GL, 136 commercial off-the-shelf (COTS),
report (AHRQ), 490 clinical documentstandards,289 clinical/genomics research,
Commission on Accreditation
common data models for, 654-655 Clinical Informatics Wiki
(ClinfoWiki), 473
Citizen Scientist Curriculum
clinical information server, sample specifications for, 134 Clinical Information System
(University of Florida), 109 Civil Money Penalties (CMPs),
Clinical Information System
353,354
classification and data mining, 694 client-server architecture, 143
clinical and operational data store, 443
clinical champions, 483 Clinical Data Interchange Standards Consortium
(CDISC), 288
clinical decision support (CDS), 9,126,127,144,147,158, 159,160,167, 243,295, 347, 471-473, 526-527,764 basics of, 473-475 case studies, 492-498
challenges to implement, 490-492 characteristics and elements of successful, 478-480
clinical quality measures workflow and, 603-604 definition of, 472 educational inter\'ention and, 489-190
for electronic clinical quality measures, 604
evaluation strategies to measure success of, 484
five-rights framework of, 473-474,473 frameworks for success of, 480
goals and objectives of, 475-477
human factor design principles and, 477
for improvement initiatives, 482-484
interprofessional teams and, 479
interventions, deployment of, 483,488
legal implications in using, 491^92
life cycle intervention, 484 meaningful use and, 485—490 oversight committee, 483 population health improvement using, 485^87
Evaluation Scale, 765
Implementation Evaluation Scale (CISIES), 30, 237,248 trifold paper-based, 238
Clinical Information Systems (Sittig and Ash), 508
clinical messaging and notifications, 266
clinical nurse specialists (CNS), 25 Clinical Quality Language (CQL), 482, 598
example of, 599
cloud computing, 142-143 vulnerabilities of, 746-747
cluster analysis and data mining, 694
CMMI. See Center for Medicare and Medicaid Innovation
CMPs. See Civil Money Penalties CMS. See Centers for Medicare & Medicaid Services
CNIO. See Chief Nursing Informatics Officer CNM. See certified nurse midwives
CNS. See clinical nurse specialists cognitive computing, 458,689, 703-704
programs, for oncology research, 709-710
cognitive security definition of, 749-750
global problem with world changing leaks and, 750,752
risk assessment approaches, 755 risk assessment preparation and, 755
systems, 753-755 threats and vulnerabilities, 750
Collaborating Across Borders (CAB) initiative, 765 collaborative, notion of, 33
collective community impact, 12 collective impact, 310 Colorado Beacon Consortium
(CBC), 580
177,185,186 for Health Informatics and
Information Management Education (CAHIIM), 773 common-cause variation, 456 common data model, 294-295
Common Well Health Alliance, 76-77,259
communication protocols, 142 communications/messaging software, 138
community assessment surveys, 571 community asset mapping, 314-315,326
community-based organizations (CBOs), 335
Community-Based Participatory Research, 108
Community College Consortia Program, 86 community comparative analysis and analytic score card and confidenceinterval, 324 Community Health Assessment and Group Evaluation (CHANGE), 310, 327,328,331
community health improvement plan (CHIP), 307, 313,328, 331 intervention and evaluation of, 325-327
Community Health Information Networks (CHINs), 255-256, 283
Community Health Management Information Systems (CHMIS), 255, 283 community health needs assessment, 305-307
background of, 307 birth and birth-related
information for, 318 case studies, 327-331
CDC voluntary accreditation program and, 308-309 checklist for variables in, 326 community asset mapping and geomapping methods for, 314-315
community health improvement plan and, 325-327 data and, 317,322-324 data analytic and statistical approaches and, 315-317 economics of, 318
historical and current policydriven requirements for, 307-308
INDEX
models, 320
Connecting Health and Care for
mortality and morbidity and,
the Nation, 7 consensus standards, 287 consumer data stewards, 438
318-319
PCCI case study and, 332-339 population variables and, 318 primary care access for, 319-322 process of, 309-314 purpose of, 307 rural and small communities
consumer engagement and personalized medicine era, 644
DTC testing, 645-647 genetic testing and, 644-645
and special considerations
Consumers Union, 112
and, 324-325
content management, 726 Continuity of Care Document
community health status indicators (CHSI), 310
Community Impact Forum, 109 community involvement and industry trends, 376-377 competency, definition of, 616 Competency Assessment Simulation of Electronic
Health Records (CASE), 627-628
Complexity Tlieory, 52 composite healthcare system (CHCS), 153
Comprehensive Primary Care Initiative, 580
Computer-Based Patient Record, 153
computer-based provider order entry (CPOE), 26,57, 58, 77, 154,167,481,505,508,513, 522,537
barcoding and, 549-550 case study of medication administration with closed-
loop, 59-63
program evaluation and, 233 workflow of, 550
computers, in healthcare, 121 bits and bytes and, 127-130
(CCD), 289 control chart, 456-457,538,540, 541, 545-546 decision tree for, 546
of monthly stock-outs in supply room, 548 controlled medical vocabularies, desired characteristics for, 297-298,298
Cook Children's Hospital Community Assessment, 322 coordination of care, through patient engagement, 369 Copyright Act of 1976, 731
Core Competencies for Interprofessional Education Collaborative, 15,34
Core Entrustable Professional Activities for Entering Residency
human factors and system ergonomics and, 131-132, 144-146
software and, 134-139
Computer Vision, 711,713,714 confidence inter\’al (Cl), 453
confounding factors, 446 Connected Communities of Care
(CCC), 335
Connecticut Hospital Association (CHA), 322
Creative Commons, 731
CRISP-DM. See CRoss-Industry Standard Process for Data
Mining CRNA. See Certified registered nurse anesthetist
crossdisciplinary, notion of, 33 cross-functional diagrams, See swimlane diagrams/ flowcharts
Cross-Industry Standard Process for Data Mining (CRISP-DM), 692
phases of, 692 Crossing the Qualit]/ Chasm report (NAM), 154, 505,765
crowdsourcing, 108 Cures Act (2016), 8, 76,77,89-90, 200,204,255, 265,566
health information exchanges (HIEs) and, 267-268 Current Procedural Terminology (CPT), 292 cybersecurity, 741-743
Coronavirus Aid, Relief, and
case study, 757-759 cognitive security and global
Economic Security Acts (CARES Act), 81, 389 Coronavirus Preparedness and
threats and, 749-756 enhanced, 745-747 Health Cloud and, 756-757
Response Supplemental Appropriations Act, 399
COTS. See commercial off-the-shelf COVlD-19 Citizen Science research initiative, 108
HIT and, 143
Terminology CQL. See Clinical Quality Language
bitcoin and blockchain and, 756
Courage Foundation, 752
122-123
CPT. See Current Procedural
Medical Colleges), 620
communication fundamentals of, 123-125 hardware and, 130-134,139-143 healthcare informatics and, 125-127
Coyle and Battles model, 57-58 CPOE. See computer-based provider order entry
(Association American
case studies, 144-147
historical perspective of,
795
COVID-19 Hospital Impact Model for Epidemics (CHIME), 336 COVID pandemic, 9,10,16, 81,89, 112, 220,276, 286, 515, 626, 729
case study of rapid community response to, 332-339 chronic care management services, 399—400,401
data-driven approach to, 334 infectious disease screening/ intervention and, 497-498
needs for proactive management of, 333
-related population health, 565-566
-related public health challenges, 332
implications, for health informatics professional, 747 mitigation, merging with cognition, 748-749 new and emerging attack threats, due to pandemic vulnerabilities, 743-745
Dallas County assessment, 324, 325
Dallas County Health and Human Services (DCHHS),332 dark web, 750, 751
data, information, knowledge, and wisdom (DIKW), 45—46 data analysis control chart and, 456-457
data mapping and, 452 data quality and, 447-449,448 data set and, 449-451 data transformation and, 451^52
796
INDEX
data analysis exploratory, 449-450 measurement theory and, 444-447 normal distribution and, 453-454
parameter estimates and, 453 priority based on, 323-324 statistical analysis, 452-453 statistical test selection and, 454^56
Data Analysis ToolPak (Excel software), 458, 459 data architecture foundation, 417
data architecture strategies, 417^20
data availability and settings, 576 data blocking, 267 data breaches, 752
data complexity and genomics, 648-651
data dictionary, 422,449 data exchange/management, Set’
health information exchanges data feed and interfacing, 266 data-first mentality, 709 Data flow diagrams (DFD), 211 data-flow maps, 183 data integrity, 449 data lake, 418,420
data management, 5,13,22,90, 211,441-443, 566, 584,619, 677,687
analytic and business intelligence tools and, 457-467 analytics and, 13, 22-23 architecture and, 437
case study, 467-468 data analysis and, 444-457 decentralized, 260
as foundations for analysis, 442-444
master, 419 software, 138
See also clinical decision support (CDS); Enterprise Data Management, Reporting, and Analytics Program (E-DRAP) data mapping, 296,452 data marts, 418,443
data mining, 702 advanced analytics, software for, 703-708
advanced tool of, 695-697
artificial intelligence and machine learning techniques and, 692-694 definition of, 691-692
significance of, 691 techniques, types of, 694 data-mining laboratory (DML), 698,699
data model, 418
data quality, 437,447-449,448 capture, assessment building for, 602-603
integrity and, 434 life cycle, 437 rules, of Orr, 434
data reliability and validity, 159-160 data research models and harmonization need and
genomics, 653 data scientist, 697-699
problem tackling by, 699-703 data set, 449-451
exploration of, 449—150 graphical examination of, 450-451
diagnosis-related groups (DRGs), 574
Diagnostic and Statistical Manual of Mental Disorders {Fifth Edition), 28
DICOM. See Digital Imaging and Communication in Medicine Dietetics Workforce Demand
Study (2011), 107 digital divide, 731 digital health and informatics, comparison of, 11
digital health literacy, 731 Digital Imaging and Communication in Medicine
(DICOM), 292
Digital Millennium Copyright Act, 731
data sources, for future, 573
DIKW. See data, information,
data stewardship, 438
knowledge, and wisdom direct-messaging model, 262-264 Direct Project
data summarization business
intelligence tool, 694 data transformation, 451-452
data triangulation, 322-323, 323, 329-330
Data Use and Reciprocal Support Agreement (DURSA), 271
abstract model, 285 website, 285
direct-to-consumer (DTC), 644
risks and benefits of testing of, 646-647
data visualization, 460-467 data warehouse, 443
DireetTrust, 262-263
DCHHS. See Dallas County Health
disparate systems, integration of,
and Human Services
De-centralized Hospital Computer Program, 153 decimal to binary prefix terms, comparison of, 129 decision-making, ethics of, 526-527 decision-support systems, 45 deep learning, 693 de facto standards, 287
Delivery System Reform Incentive Payment (DSRIP) program, 88 Delphi study, 107 demographic and clinical data exchange, 266 demographic survey, 765 denial-of-service attacks, 745
Department of Homeland Security Science and Technology Directorate, 710
Department of Veterans Affairs, 153,361,524
dependent variable, 445
design, document, and build cycle, 431
design strategies, 443-444 device drivers, J37 Dexcom 6,112
DFD. See Data flow diagrams DHHS. See U.S, Department of Health and Human Services
testing, 645-646 156-157
distorted signal, 123 distributed analytics teams, 438 DLT. See U.S. Department of Agriculture's Distance Learning and Telemedicine
DML. See data-mining laboratory DNA analysis, 103 DNP. See Doctor of Nursing practice
doctor's office quality-information technology (DOQ-IT) program, 188,190
Doctor of Nursing practice (DNP), 377,550-551
DocuCare (Lippincott product), 624 Donabedian model, 51, 445—446 modified, 57
social media and public health and,64-65 structure-process-outcome framework of, 66-68, 576
DOQ-IT program. See doctor's office quality-information technology program downtime and disaster
preparedness procedures and protocols, 515
downtime contingency planning, for EHR, 164
INDEX
draft charter, 556,558
measurement and, 590-591
DRAFT The Essentials (AACN), 618
new and emerging trends in,
DRGs. See diagnosis-related groups
drug delivery systems and nanotechnology, 673 DSHS. See Texas Department of State Health Services
DSRIP program. See Delivery System Reform Incentive Payment program DTC. See direct-to-consumer
duplicate records, issue of, 449 DURSA. See Data Use and
Reciprocal Support Agreement eatrightpro.org website, 107 Ebola virus, 9
608-609
new telehealth requirements
and implications for, 608 principles of, 594,595 quality reporting initiatives and, 591-593
scribes to address regulatory burden of, 609
stakeholders in development
process of, 600-601, 600 standards and specifications, 597-598,597
strategy, blueprint for, 602 team, responsibilities of highfunctioning, 607 unintended consequences of, 607-608
ECIC. See Evidence Communication Innovation Collaborative
validation of, 605 electronic data transmission (EDT),
e-CQMs. See electronic clinical
electronic genetic data resources,
quality measures ECRI. See Emergency Care Research Institute
E-DRAP. See Enterprise Data
Management, Reporting, and Analytics Program EDT. See electronic data transmission
EDW. See enterprise data warehouse
EEF. See Enterprise Environment Factors
eHealth Exchange (eHEX), 80,254, 256-258
Direct Project for, 263 privacy, maintenance of, 271
EHR. See electronic health records
Eight Dimensional Model, 51, 53-55, 58
dimensions and descriptions of, 54,55
EIN. See employer identifier number
elbow-to-elbow training, 198 electronic clinical quality measures (e-CQMs), 220, 287, 292, 389, 589-590
case study, 609 CDS for, 604
development, process of, 594-601 efforts to improve, 593-594 EHR-based quality measurement and, 593 header, 598
implementation of, 601-605 iterative, development lifecycle. 597
350
for clinicians, 659-660 electronic health records (EHR), 4-5, 7,9,16, 26,76-80,122, 124,151-153,210,232,287, 358,482, 504,514-515, 575
and aspirin compared, 124 benefits, to improve safety and quality, 166-168 care coordination supported by, 168
case study, 632 certified, 78,156,167,488 communication skills, 620
competencies, 616-620 data standards and, 293-296 definition of, 143
documentation and providers of, 30-32 effective selection and
deployment of, 146-147 functional specifications of, 288 genomic data integration into, 653, 656
health information exchanges (HIES) and, 259, 269-271 in hospitals, history of, 153-154 human factors science, usability and,169
Incentive Program, 78, 87 -induced medical records, 514-515
informatics competency evaluation and, 627-628 International Medical
Informatics (IMIA) and, 628-629
interprofessional communication and, 766
797
Meaningful Use MaturitySensitiveIndex (MUMSI),30 nationwide adoption, drivers behind, 154-155
nursing informatics role and impact and, 165-166 optimization of, 515 optimization, for interoperability in obstetrics, 204-206
point of care device integration and interoperability and, 155-164
predictive analytics and, 168 project plan samples and implementation, 188,189-190 purpose of, 123 research and initiatives to address clinician
dissatisfaction and, 169-170 robotics and, 669
Safety Assurance Factors for the EHR Resilience (SAFER)
guide, 164-165,247-248 safety model dimensions, 247 satisfaction with, 31
SHARP grant and, 84-85 in simulation centers, challenges with, 624-626 simulations and, 621-624 software selection for, 138-139
and stethoscope compared, 125 superusers and, 166 telehealth and telemedicine in simulation and, 626-627
TNA/TONE and, 526
training sandbox, 626 Triple Aim and, 168-169 usability issues with, 168 workflow redesign and, 220, 221 See also point-of-care; program evaluation
electronic medical record (EMR), 98,143,271
publishing and transfer, 266 Emergency Care Research Institute (ECRI), 511-513 Health Devices Group, 512
Top Ten List of Healthy IT Safety Hazards, 512
emergency room, triage and interventions in, 712-713
EMPI. See enterprise master
patient index
employer identifier number (EIN), 355
EMR. See electronic medical record
engineered nanomaterials (ENMs), 679
798
INDEX
ENMs. See engineered nanomaterials
Enterprise Data Management, Reporting, and Analytics Program (E-DRAP), 413
best practice methodologies, 432-434
business intelligence (Bl) library, 420,422-423 business need and, 414-415
business-led program management and, 425-426 case study, 435-438 continuous and iterative process cycles and, 428-432 corporate culture and, 424 data architecture foundation
and, 417
data architecture strategies and, 417^20
ETL/ELT. See extract, transform, and load/extract, load, and transform
European Commission, 347 European Committee for Standardization (CEN), 288
EU/US partnership, for informatics competencies definition, 11,32
event management, 726
evidence-based practice, 36, 50, 104, 327,485,488, 511,617, 630, 631, 698,767, 111 Evidence Communication Innovation Collaborative
(ECIC), 99
Explorehealthcareers.org, 25 extract, transform, and load/extract,
load, and transform (ETL/ELT), 419
framework of, 415, 416 Information Governance
Council, and, 424-425
organizational components for, 425
road map for, 426,427 technology, 417,418-419, 423^24
enterprise data warehouse (EDW), 414,417, 418
products and solutions, for HDDs, 423
Enterprise Environment Factors (EEF), 180 enterprise master patient index (EMPl), 263
FACA. See Federal Advisory Committee Act
Facebook, 723
Visio workflow diagram of, 223 ergonomics and human factors, 131-132
in barcode scanning, 144-146 error mitigation, statewide nursing approach for, 525-526 Essential Genetic and Genomic
Competencies for Nurses With Graduate Degrees, 658, 660 essential public health services (EPHS),327,328 core functions and, 309 ethernet connections, 140
fishbone diagram. See cause-andeffect diagram Flex Medicare Beneficiary Quality Improvement Program, 100 flowcharts, 540 FMEA, See Failure Mode Effect
Analysis form utility, 245 Forrester Consulting, 684 fourth-generation machine language (4GL), 136 Freestyle Libre, 112 FST. See Family systems theory FTC. See Federal Trade Commission
FTP. See file transfer protocol Future of Nursing report. The
Family Health Model, 105-106
future state, mapping of, 216
Family systems theory (FST), 106 Fast Health Interoperability Resources (FHIR), 153, 262,
267, 286, 289
Developer Toolbox, 289 genes on, 650-651
feasibility studies, 181 Federal Advisory Committee Act
222-223
710-711
(FMEA), 216,540, 541,542-543 fair use doctrine, 731
EPHS. See essential public health Epic Care Everywhere, 259 Epic Share Everywhere, 259 epistemology, 44 e-prescribing workflow analysis,
(IGL), 134
first responder, readiness and dispatch to patient location,
Future of Nursing 2020-2030 report,
FDA. See Lf.S. Food and Drug
Epic, 259,289
first-generation machine language
as tool for public health, 63-67 Failure Mode Effect Analysis
enumerative studies, 454 ePatient, 727-728 services
filter bubble, 727 financial data store, 443 firm ware, 131
Administration
(FACA), 79
Federal Health l.T. Strategic report, 8 Federal Health information
technology (IT) Strategic Plan 2020-2025 report, 366,374
35, 505, 764
aOM), 388,505,522, 764
gaming/video software, 138 Gantt.com website, 187 Gantt charts, 187-188,229 GAO. See Government
Accountability Office gap analysis, 184 genebank. See biorepositories General Data Protection
Regulation, 347
General Hospitals, 89 Gene Reviews, 659 Genes in Life, 660 Genetic Alliance, 660
Federal Office of Rural Health
Genetic Information Nondiscrimination Act
Policy (2014), 100 Federal Trade Commission (FTC),
(GINA) (2013), 356,645 Genetics Home Reference, 659
25-26,42,345-346,644
federated/decentralized model. 263
fee for service (FFS), 577 FFS. See fee for service FHIR. See Fast Health
Interoperability Resources
fifth-generation machine language (5GL), 136
file transfer protocol (FTP), 140, 142
genetic testing, 644-645
Genetic Testing Registry, 659 genome-wide sequencing (GWS), 645
genomics, 641-642
benefits and challenges of resources of, 658
case study, 661
clinical decision making,
workflow, and HIT and,
647-656
INDEX
consumer engagement and personalized medicine era
consideration of, 130 infection-control issues and,
and, 644-647 definition of, 642 ethical considerations, 657-658
selection and specifications of,
history of, 642-644 nanotechnology and, 674-675 privacy and security of, 656-657 Genopedia, 659 Genos company, 646 geographical information system
132-133 130-131
specifications of, 133-134 Harrison Interactive Sociotechnical
Analysis Model, 509-510 HCPFC. See Health Care and Promotion Fund Committee HCUP. See Healthcare Cost and
of Coronavirus active cases, 317
Utilization Project HDO. See healthcare delivery organization
of vulnerability, 336
Health and Human Services
(GIS), 644
geomapping, 315 GINA. See Genetic Information Nondiscrimination Act
GIS. See geographical information system Global Forum on Innovation in Health Professional Education, 764 Global Health Informatics
Competency Recommended Framework, 35
Global Language of Business (GSl), 288
goal utility, 245 go-live. See system implementation Google Scholar, 723 Gordon and Betty Moore Foundation, 765
Government Accountability Office (GAO), 593
government-mandated standards, 287
grand theory, 44
graphical user interface (GUI), 237 graphics software, 138 graphs, significance of, 450-451 GSl. See Global Language of Business
GUI. See graphical user interface Guide to Reducing Uninteuded Consequences of Electronic Health Records (ONC), 510, 523
GWS. See genome-wide sequencing
Hadoop, 687 Haplotype Map (HapMap), 643 HapMap. See Haplotype Map hard-stop alert, 160,474,490 hardware
basic configurations of, 130 computer internal components and, 131
configurations, and networking and connectivity, 139-143
(HHS) measure inventory, 582 health behavior change,
technology and health coaching for, 109 patient-generated health information (PGHI) and, 110-111
wellness coaching and mobile devices and, 111-112 healthcare administration
professional, 28 Health Care and Promotion Fund
Conunittee (HCPFC), 518 Healthcare Cost and Utilization
Project (HCUP), 574-575 healthcare delivery organization (HDO), 414-415,419,423-424, 426,428
EDW/BI products and solutions for, 423
healthcare delivery systems, 520-521
healthcare ecosystem alignment, driving changes in healthcare outcome, 93 Healthcare Effectiveness Data and
Information Set (HEDIS), 591 Healthcare Information and
Management Systems Society (HIMSS), 10, 28,90,102-103, 169,349, 518, 771
Center for Patient FamilyCentered Care, 102 Certified Professional in Healthcare Information
and Management Systems (CPHIMS)!^’, 201 Davies award for, 242-244 Electronic Medical Record
Adoption Model (EMRAMSM), 241-242
interoperability and, 156 patient engagement framework with meaningful use categories, 118
799
patient engagement network, 534 program evaluation and, 241 healthcare providers and informatics, challenges for, 748 Health Cloud, 756-757
Health Department of Chicago, 729
HEALTHeLlNK, 275
health impact assessment (HIA), 312
health information exchanges (HIEs), 76,80-81, 87,154, 204, 286, 347
analytic methods and, 317 architectural and data-exchange models for, 260-265 business models, 265-266 case studies, 272-277 Cures Act and, 267-268
data exchange approach sample in, 263
eHEX privacy maintenance and, 271
EHRs as, 259
framework for success of, 272
history, compared with current HIEs, 255-256
master patient index (MPI)and, 268
models, 260-261
record linkage and, 268-269 standards and interoperability framework and, 265 state, 78
Trusted Exchange Framework and Common Agreement (TEFCA) and, 258-259 value, coupled with EHRs, 269-271
valued services associated with, 266
Health Information Law (HIL), 346 health information management (HIM), 764,777
physical therapy and, 778-779 radiation therapy and, 777-778 Health information networks
(HINs), 90
Health Information Portability and Accountability Act (HIPAA), 80,164,197,262,344,345,355, 730, 748
background of, 347-349 compliance, for telehealth, 405 dates and deadlines, 348
Privacy Rule, 350-351, 360,361 sections and important dates. 349
significance of, 347 See also privacy and security
800
INDEX
health information service
providers (HISPs), 263 health information technology (HIT), 77
alternative payer models and healthcare reform and, 5-6
changing landscape of, 4-5 Cures Act and, 8
demand for expanded clinical competencies in health and, 10
error/unintended consequences clinical impact, 513-515 Hazard Manager Database, 510-511,531 health and consumer
engagement and, 6 informatics to address
healthcare digital age and, 10-11
goals of, 254 health information exchanges (HIEs), 80-81 initiatives of, 81
Omnibus rule, 355,357
HEDIS. See Healtlicare Effectiveness Data and Information Set
proposed modifications (2020),
Hewner, Sharon, 273
modifications (2013), 356-357 357
protections and, 355-357 REC program and, 81-82 SHARP grants and, 83-85
workforce deployment program of, 86, 87
See also health information
technology (HIT) Health Information Technology Competencies (HITCOMP), 618, 627, 629
HealthIT.gov website, 90,165,293, 359, 380,510,518, 566
interoperability program
Health IT and Patient Safety report
interprofessional team for health
Health IT Certification Criteria
promotion and, 7^
informatics model (IT-HI model), 15-16
key terms of, 143 knowledge and wisdom and, 45-46
NEHI model and, 11-15
patient safety issues mitigation
(lOM), 491,506
(2015), 288
Health IT Playbook, 188,393 Health IT Safety Center Roadmap (RTl International), 506 Health Level Seven International
(HL7), 153, 246,262,267, 600, 625
strategies, 518-527,519 professional, 28
Clinical Document Architecture
reporting form and process for patient safety, 537 Sociotechnical Systems Theory
EHR-System Functional Model,
and QIand,57-68
technology optimization
balancing clinician well-being and, 5 theoretical foundations to examine, 47-57 TNA/TONE committee framework, 526
unintended consequences, remediating, 523,525 unprecedented pandemic and infectious disease challenges, 9 See also Health Information
Technology for Economic and CUnical Health (HITECH) Act (2009)
Health Information Technology for
(CDA), 289 288
messaging and, 288 Reference InformationModel, 288
health literacy, definition of, 371 health maintenance organizations (HMOs), 591
Employer Data and Information Data Set, 591
HealthMap, 729 health outcomesmodel, 64,446,447
Health Professionals for a Nezu Century report, 764 Health Professions Accreditors
Collaborative (HPAC), 773, 777
Health Professions Education report (lOM), 106,765
(HITECH) Act (2009), 4, 5,7,
Health Quality Measures Format (HQMF), 292, 598 Health Research Institute (HRI),
8, 28,76-79,99,154-155,283, 293,347,360,591,621, 689
Health Resources and Services
Economic and Clinical Health
Beacon program and, 82-83 EHR incentive program and, 78 federally funded programs of. 78
Healthy Communities Institute, 332 Health}/ People 2020 goals, 323 Health}/ People initiative, 571
109
Administration (HRSA), 626, 765,773
healthsystemCIO.com website, 517 Health Tech and You website, 103
HIA. See health impact assessment HIEs. See health information
exchanges HIM. See health information
management HIMSS. See Healthcare
Information and Management Systems Society HINs. See Health information networks HIPAA. See Health Information
Portability and Accountability Act HISPs. See health information
service providers histograms, 540 HIT. St-’e health information
technology HITcomp.org, 10,11 HITCOMP. See Health Information
Technology Competencies HITECH Act. See Health
Information Technology for Economic and Clinical Health Act
HITRUST, 746 HL7. See Health Level Seven International
HL7 v2.x, 288 HL7 V3,288 HMOs. See health maintenance
organizations
Hospital Compare website, 607 Hospital Inpatient Quality Reporting program, 591 HPAC. See Health Professions Accreditors Collaborative
HQMF. See Health Quality Measures Format
HRI. See Health Research Institute HRSA. See Health Resources and Services Administration
HTML, See hypertext markup language HTTP. See hypertext transfer protocol Humana and the Wisconsin HIE, 270
human-computer interaction (HCI), 178
human factors and ergonomics, 131-132
in barcode scanning, 144-146
INDEX
Human Genome Resources, 659
hybrid model, for HIEs, 261 hypertext markup language (HTML), 140 hypertext transfer protocol (HTTP), 140
i2b2. See Informatics for Integrating Biology and the Bedside IBM Cognos BI toolset, 693 IBM Corporate, 703-704, 710, 750,
INACSL. See International Nursing Association for Clinical
interactive sociotechnical analysis (ISTA), 48-49, 48, 509-510,
informatics
and digital health compared, 11 nanotechnology implications
ethical considerations for, 705-706
for oncology research, 709-710
nursing resistance to new tools and, 707-708
procedures, 706 workflow and data analysis, 706-707
IBM X-Force Exchange, 748 IBM X-Force Threat Intelligence
627-628, 628 definition of, 10-11, 618
for, 677-678
significance of, 768-769 See also individual entries
Informatics for Integrating Biology and the Bedside (i2b2), 107, 295, 654 tranSMART Foundation, 653
informatics nurse specialists (INSs), 28,165 information-based liability, 492 information blocking, 90-91 Information Governance Council, 424-425
Index, 752 ICD. See International Classification of Diseases
information pathway, 275 Information Systems Evaluation
ICEP. See Interprofessional Collaborative Expert Panel
inpatient data, 577
I-chart, 545 ICN. See International Council of
INSs. See Informatics nurse
Nurses ICNP®. See International
Tool (ISET), 237 insider threat, 750-751
specialists Instagram, 723 Institute for Healthcare
Classification of Nursing
Improvement (IHI), 5,101,
Practice
537, 773
Identify Theft Resource Center, 742,744
ID-reentry tool, 514 ID-verify tool, 514 lEC. See International Electrotechnical Commission
IHl.org website, 541,542 IHI. See Institute for Healthcare
Improvement IHTSDO. See International Health
Terminology Standards Development Organization Image File Attacks, 750
IMDRF. See International Medical
Device Regulators Forum IMIA. See International Medical Informatics
Driven Health Care, 99
infection-control interventions, 333 inferential statistics, 452-453 Informatica, 434
display file information in, 450 IBM Watson, for cybersecurity, 754 deep QA, 754 IBM Watson cognitive computing, 705
Roundtable on Value & Science-
integration testing, 195 Intelligent Sensor Information Processing (ISIP), 668
competencies, EU/U.S. partnership to define, 11 competencies, evaluating,
clinical use case to utilize, 704,
IPE, 104,105
Simulation and Learning incremental approach. See phased approach, to system implementation independent variable, 445 operational definitions of, 457
752,756 IBM Institute for Business Value, 756 IBM modeler, 695,698,702 IBM SPSS software, 449,458
703-704
801
Institute for Healthcare
Improvement Open School, 541
Institute for Healthcare Quality Improvement, 537 Institute for Patient- and FamilyCentered Care (IPFCC), 106 Institute of Medicine (lOM), 35,42, 98,112-113, 373,377,388,477, 619,689,764-765
Committee on Patient Safety and Health Information
Technology, 506 competency and, 616 core competencies, and interprofessional teamwork, 768
See also National Academy of Medicine (NAM)
intensive care unit and AI, 713
523,529
interdisciplinary, notion of, 33 International Classification of
Diseases (ICD), 292,297 1CD-9-CM, 292,452, 574 ICD-IO-CM, 350, 452, 574 ICD-IO-PCS, 292 ICD-11,292 International Classification of
Nursing Practice (ICNP®), 299,300 International Council of Nurses
(ICN), 299,300 International Criminal Police
Organization (INTERPOL), 743 International Electrotechnical
Commission (lEC), 128 International Health Terminology Standards Development Organization (IHTSDO), 288, 300
International Medical Device
Regulators Forum (IMDRF), 295
International Medical Informatics
(IMIA), 618,628-629 HEALTH model, to realize HIT benefits, 629
International Nursing Association for Clinical Simulation and
Learning (INACSL), 622 International Organization for Standardization (ISO), 288, 296
International Recommendation Framework of Core
Competencies in Health Informatics 2.0, 617-618 International Union Conservation
of Nature (lUCN), 187 Internet, 140
significance of, 720 Internet of Medical Tilings (loMT), 710-711, 714
Internet of Things (loT), 103,402, 687,688,742 Health Cloud and, 758
personalized medicine and, 748-749
802
INDEX
interoperability, xix, 5-6,27, 35, 49,138,143,170, 212, 372, 415, 419,492, 509, 566, 749 definition of, 267
disparate systems integration and, 156-157 EHRs and, 153-164,177,204206,246,257,259,267-268
genomics and, 650-651, 656 levels, descriptions of, 262 meaning and significance of, 8, 282-283
national healthcare transformation
and, 76,78-80,84,85,87,90 national standards for HIT and, 282-283, 287, 293,296, 300-301
promotion of, 7-8,80,89-90,93, 155, 204, 653
security for, 289 standards, 90,156,265,283-286
workflow redesign and, 212 See also Fast Health
Interoperability Resources (FHIR); Promoting Interoperability Program INTERPOL. See International
Criminal Police Organization interprofessional, notion of, 33 interprofessional collaboration (IPC), 33, 774
Interprofessional Collaborative Expert Panel (ICEP), 631,766, 777
interprofessional collaborative practice competency domains, 768 domains, 767
interprofessional competencies, patient-centered, 106-107 interprofessional education, 15, 32, 763-764
case studies, 776-779 communication and, 766-768 definition and ethics, 765-766 future considerations, 771-776
history of, 764-765 and HIT, relationship between, 768-770
patient engagement and, 104 pillars of, 34 practice organizations and initiatives and, 34-35 and traditional education
compared, 632 Interprofessional Education Collaborative (IPEC), 33,767 Expert Panel (IPEC), 15,32, 765 Interprofessional Education for Healthcare Informatics Framework, xx
interprofessional practice (IP), 32 competencies in support of, 35-38
Interprofessional Practice and Education, 32,33
interprofessional team for health informatics model (IT-HI model), 15,22,23 application to current healthcare environment, 16
education and practice organizations and initiatives, 15
high performing teams and challenges and, 15-16 informatics competencies for, 29-32
meaning and significance of, 24, 34-35
types of, 24-29 interprofessional team/teamwork, xvi, xix, xxviii, 4,11,13,104, 130,160,181,187,414, 439, 447,479, 536,537, 708
competencies in nursing and, 619,621,623, 627,630,631
electronic clinical quality measures and, 603-605, 607, 610
HIT and, 522-524,527, 767, 769, 770
lOM core competencies and, 105, 768
technologies and, 676-679 interval data, 445 interval estimate, 453 Invisible Disabilities Association website, 103 lOM. See Institute of Medicine loMT. See Internet of Medical
Things loT. See Internet of Things IP. See interprofessional practice IPC. See interprofessional collaboration
IPE. See interprofessional education
IPEC. See interprofessional education collaborative IPFCC. See Institute for Patient-
and Family-Centered Care ISET. See Information Systems Evaluation Tool
ISIP. See Intelligent Sensor Information Processing ISO. See International Organization for Standardization ISTA. See interactive sociotechnical
analysis iStat, 158
iterative development, 193 IT-HI model. See interprofessional team for health informatics model lUCN. See International Union Conservation of Nature
IV smart pumps, 159 JEDEC. Sec Joint Electron Device
Engineering Council John A. Hartford Foundation, 764, 765
Joint Electron Device Engineering Council QEDEC), 128-129 Josiah Macy, Jr. Foundation, 764, 765
Journal of American Medical Association, 521
journal of General Internal Medicine, 271
journal of the American Medical Informatics Association, 270
Kaiser Family Foundation, 88,318 Kaiser Permanente Community Health Assessment Toolkit, 322
Keeping Patients Safe, 505 Kennedy-Kassebaum Act. See Health Information Portability and Accountability Act keyloggers, 745 key performance indicators (KPIs), 418
business intelligence library and, 422
data definition, 418 Kinesis QTUG, 770 KNIME. See Konstanz Information Miner
knowledge and wisdom, 45^6 knowledge representation (KR), 136 Konstanz Information Miner
(KNIME), 697 workbench and data workflow, 696
KPIs. See key performance indicators
KR. See knowledge representation LabMD, 345-346
lab-on-a-chip (LOC), 671-673 components of functional, 672 size of, 672
Lakeland Health System, 243 LAN. See local area network
law making process, representation of, 92
INDEX
layered security, 744 LDH. See local health department Leadership and Workforce Survey (2017), 28
LEAN management, 212,216,432, 544-545, 704
project charter template, 223 Leapfrog Group, 247, 522 legacy systems, 156-157 Lifeinthefastlane.com blog, 726 light-controlled pain relief, 672 Linkedin, 723 live feeds, 726
live streaming, 726 load and volume testing, 196 LOC. See lab-on-a-chip local area network (LAN), 140,159 local health department (LDH), 146-147
local public health departments (LPHDs),312
local public health systems (LPHSs), 312, 324,327-328
Logical Observation Identifiers Names and Codes (LOINC), 262,292, 293, 299,604
panel, for nursing physiologic assessment panel, 300 logic model as framework, 572-573 structure, to program evaluation, 236
LOINC. See Logical Observation
bAagk Quadrant for Analytics and Business Intelligence Platforms, 460
measurement
(MCO), 350 manual association of devices, to
patient, 161 MAP. See measures application partnership Massachusetts General Hospital, 85,153 Massachusetts Institute of
Technology Center for Digital Business, 414
709-714
and deep learning compared, 693
definition of, 689
machine-organizable data, 262 machine-transportable data, 262 MACRA. See Medicare Access and CHIP Reauthorization Act
importance of, 590-591 operational, 445—416 measurement theory, 444 measures application partnership (MAP), 601
MedBiquitous, 772 Medicaid Waiver program, 88,309 Medical Device Plug-and-Play (MD PnP), 85
Medical Expenditure Panel Survey (MEPS), 574
complex project charter, 552,554 master data management, 419 master patient index (MPI), 268,
Medical University of South
355
materials management department and QI, 547-549
MATH. See Modeling and Analysis Toolsuite for Healthcare
Mayo Clinic College of Medicine of Rochester (Minnesota), 84
MC programs. See managed care programs
McAfee, Andrew, 686
organization 710
MDCs. See major diagnostic categories MD PnP, See Medical Device
Plug-and-Play MD SHARP, 85
Meaningful Measures Framework, 594
clinical settings application of,
definition of, 444 levels of, 444-445
medical liability coverage, for
MD Anderson Cancer Center, 704,
686-687
594-^01,596
mass vaccine administration
Nanotechnology, 671 LPHDs. See local public health departments LPHSs. See local public health
MA. See Medicare Advantage machine code programming language, 134-135 machine-interpretable data, 262 machine learning (ML), 692-694 artificial intelligence and,
248, 765
(MDCs), 574 malware, 745
MCO. See managed care
basic application of, 228
Sensitive Index (MUMSI), 30,
measure development process,
Identifiers Names and Codes London Center for
systems LucidChart, 218
Meaningful Use Maturity-
maintenance requests, 162 major diagnostic categories
managed care (MC) programs, 577 managed care organization
meaningful use (MU), 31,76,80, 82,99,155,162, 204,212, 591-592
criteria, 87 definition of, 381
HIMSS patient engagement framework and, 118 historical standards foundation
through, 294 Incentive Program, 288 patient engagement as success factor and, 234
patient portal and, 368-369 stage 3,369 See also Promoting
Interoperability Program (PIP)
803
telehealth, 404-405 Carolina, 270 Medicare Access and CHIP Reauthorization Act
(MACRA) (2015), 6,88-89, 255,369
Medicare Advantage (MA), 267 Medicare funding, 88 Medicare Physician Fee Schedule, 399
Megatrends (Naisbitt), 687 Memorial Sloan Kettering, 703-704 Mendeley, 723 MEPS, See Medical Expenditure Panel Survey Merit-Based Incentive Payment System (MIPS), 6,89,369 advancing care information objectives and measures, 370 methicillin-resistant Staphylococcus aureus (MRSA), 132 MEWS. See Modified Early Warning Score Microsoft, 460 Microsoft Cloud for Healthcare, 711-712
Microsoft Excel, 458 Microsoft Visio Professional, 217 functions of, 226-227
workflow diagram of e-prescribing process, 223 workflow diagram steps in, 217-218
middle ware, 237
midlevel practitioners (MLPs), 706-708
midrange theory, 44-45
nursing informatics specialty and, 45^7
804
INDEX
Million Hearts campaign, 485,489 ABCS tenets of, 485^86
MIME. See Multipurpose Internet Mail Extensions
mind map, 447 MIPS. See Merit-based Incentive
Nancy Staggers model of information technology, 107 nanoinformatics, 677 nanomedicine, 670
safety considerations of, 679 nanoparticles, 671
National Center for
Interprofessional Practice and Education, 32, 764, 765
National Center for Patient Safety, 524 National Center for Research
Payment System missed diagnosis, blog about, 518 MITRE Corporation, 482 ML. Sfe machine learning MLPs. See midlevel practitioners
nanorobotics, 675
nano tattoos, for diabetes, 673 mechanism of, 674
National Center of Health
MMAs. See mobile medical
nanotechnology, 669-670 application in healthcare,
National Committee for Quality
applications mobile access challenge, with POC device integration, 159 mobile computing, 142 mobile health (mHealth), 389 app category share percentages, 403
definition of, 402 devices of, 112
health promotion, 404 history, current use, and future and, 401-403
implementation, financing, and sustainability, 404 mobile medical applications (MMAs), 295
mobile personal emergency response services (mPERS), 404
Modeling and Analysis Toolsuite for Healthcare (MATH), 221 Modified Early Warning Score (MEWS), 490
MONAHRQ system, 583
Montefiore Medical Center survey, 513
Moore's Law, 123,686, 749
mPERS. See mobile personal emergency response services MPI. See master patient index MRSA. See methicillin-resistant
Staph\flococcus aureus
MU. See meaningful use multidisciplinary, notion of, 33 Multipurpose Internet Mail Extensions(MIME), 285 MUMSI. See Meaningful Use Maturity-Sensitive Index MyChartCare, 380
application in healthcare, 675-676
670-675
case study of, 679 implications, to nursing informatics and
interprofessional teams, 676-679
NIH projects profiled, 671 nanotek compounds, 676 nanotoxicology, 679 NASA Jet Propulsion Laboratory, 710
NASEM. See National Academies
of Sciences, Engineering, and Medicine
National Academies of Sciences,
Engineering, and Medicine (NASEM), 35, 764 National Academy of Medicine (NAM), 153,154,168, 254, 537 See also Institute of Medicine
(lOM)
National Association of County and City Health Officials Mobilizing for Action Through Planning and Partnerships (NACCHOMAPP),310 National Association of County and City Health Officials (NACCHO) website, 571 National Board for Health &
Wellness Coaching (NBHWC), m-112
Association of County and City Health Officials Mobilizing for Action Through Planning and Partnerships NAM. See National Academy of Medicine
Technology, 110 Statistics (NCHS), 360 Assurance (NCQA), 591
National Council for Prescription Drug Programs, 350 National eHealth Collaborative
(NeHC), 375,518 National Electronic Health
Records survey, 366 national healthcare transformation
and information technology, 75-77
federal law making and expert advocacy and, 91-93 Patient Protection and
Affordable Care Act (PPACA/ ACA) and, 88-91 See also Health Information
Technology for Economic and Clinical Health (HITECH) Act (2009) National Health Information
Network (NHIN), 282 National Health Information
Technology Policy Committee, 650
National Health Quality Roadmap, 608
National Health Safety Network (NHSN), 575, 582
National Inpatient Sample (MS), 574
National Institute for Standards
and Technology, 491 National Institute of Health (NIH), 85,101,643,669, 670
National Center for Biomedical
National Institute of Standards
Computing (NCBC), 107 National Center for Biotechnology
and Technology (NIST), 742 National League for Nursing
Information, 270
chromosome 21 with NACCHO-MAPP. See National
Resources (NCRR), 153 National Center for Telehealth and
Alzheimer's gene marker depiction by, 649 National Center for Health
StatisHcs (NCHS), 271,286,366
National Center for Human Factors in Healthcare, 16 National Center for
Interprofessional Education and Practice (2012), 15,34
(NLN), 618
National Learning Consortium, 366-367,372
National Library of Medicine (NLM), 170, 287,297,300,482, 600
National Nanotechnology Coordination Office, 669
National Organization of Nurse Practitioner Faculties
(NONFP), 627
INDEX
National Partnership for Women and Families study, 377 National Prevention Council
(2014), 100,103,395,568, 570
National Prevention Strategy, 100,
New York University (NYU) Center for Health and Public
Service Research, 319
Next Generation First Responder Apex program, 710
103, 568-571 framework, 570
NHIN. See National Health Information Network
key indicators and goals for.
NHSN. See National Health Safety
569
National Program of Excellence in Biomedical Computing, 107 national provider identifier (NPI), 25, 348, 355 National Public Health Performance Standards
(NPHPS), 311, 327,328
National Quality Forum (NQF), 267,476, 522, 582, 592,595, 600,601
National Quality Strategy (NQS), 4,6,11,79,100,415,590,592, 594
aims and priorities of, 592 National Strategy for Quality Improvement in Health Care (2016), 7 Nationwide Health Information
Network (NwHIN). See
eHealth Exchange natural language processing (NLP), 703 NBHWC. See National Board for
Health & Wellness Coaching NCBC. See National Center for
Biomedical Computing NCHS. See National Center of Health Statistics
NCQA. See National Committee
for Quality Assurance NCRR. See National Center for Research Resources NeHC. See National eHealth Collaborative
NEHI. See Nursing Education for Healthcare Informatics
Neighborhood Health Status Improvement, 315
Network
Nursing Home Quality Initiative, 591
nursing informaticists, 160,161, 163,165,166,192,196,199, 344,359,480,483,604,698
nursing informatics (NI), J3,28, 44,346,479, 507,509, 584,617, 627,676-679, 703, 721,743
competencies, levels of, 628
NI. See nursing informatics
definition of, 10-11
NICU. See neonatal ICU NIH. See National Institute of Health
impact of, 165-166
NIS. See National Inpatient Sample
scope and standards of, 29
NIST. See National Institute of
social media and, 734
Standards and Technology NLC. See Nurse Licensure
Compact ●NLM. See National Library of Medicine
NLN. See National League for Nursing NLP. See natural language processing noise and distortion, 123 noise variation. See common-cause variation
nominal data, 445,450 nonelectronic world, 262
NONFP. See National Organization of Nurse Practitioner Faculties
normal curve, 454 normal distribution, 453-454 normal newborn order sets, case
study of, 495-497 Northern Virginia Commimity College, 86 Notice of Proposed Rule Making (NPRM), 91-92,357 NPHPS. See National Public Health Performance Standards
NPI. See national provider identifier
NPRM. See Notice of Proposed Rule Making NQF. See National Quality Forum NQS. See National Quality Strategy
805
role of, 165
specialty, and traditional midrange theory, 45^7 See also individual entries
Nursing Knowledge: Big Data Science Conference, 690, 708
action plan components for, 691 nursing leadership community health information exchange strategy and, 272 risk assessment and, 359
nursing practice essentials, 29-30, 566-567
nursing robotics, 668 OASIS-C. See Outcome and Assessment Information Set
objective structured clinical examinations (OSCE), 627 Observational Health Data Sciences and Informatics
(OHDSl), 295,653 Observational Medical Outcomes
Partnership (OMOP), 295,653, 654
observed outcomes to expected outcomes (O/E) ratio, 578, 581
obstetrical screening, case study of, 492-494
occupational therapists (OTs), 26 OCR. See Office for Civil Rights O/E ratio. See observed outcomes
to expected outcomes
Neonatal eHandbook, 660
Nursebot, 668-669
Office for Advancement of
neonatal ICU (NICU), 171
Nurse Licensure Compact (NLC), 399 nursing data standards international nursing standards
Office for Civil Rights (OCR), 347,
networking communications and, 140-142
connectivity and hardware configurations and, 139-140 network typologies, 140-141,141 advantages and disadvantages of, 142
Neuman Systems Model, 106 New York Genome Center, 704 New York State Medicaid Data Warehouse, 273
and, 300 national data standards for
nursing, 298-300 Nursing Education for Healthcare Informatics (NEHI) Framework, xx, 23,468,584 for informatics content, 12-15,14
knowledge domains of, 11 significance of, 22-23
Telehealth, 396
348, 351,353,354 Office of Burden Reduction and Health Informatics, 608
Office of Inspector General (OIG), 400
Office of the National Coordinator
(ONC), 5, 7,8,99, 188, 392, 210,212, 235,347, 366, 375, 482, 518, 600
adoption level scale, 285
806
INDEX
Office of the National Coordinator
(co/it.) Common Data Model
Harmonization (CDMH), 294
direct-messaging project and. 263
EHRs and, 153,154,165,254,258, 271
Final Rule, 267-268
Health IT Policy Committee Privacy and Security Work Group (2015), 361 HITECH programs under, 82-85 interoperability and, 265,267, 293
Interoperability Standards Advisory (ISA), 283-284, 287 Issue Tracking System (ONCITS), 605 national healthcare
transformation and, 77-80,86, 89,90 Office of Standards and
Interoperability (S&I) Framework, 283,286 on SAFER Guides, 520 software selection and, 139 Tech Lab, 265
tips, for workflow design, 213 vision, for eHealth Exchange (eHEX), 257 website, 287, 293, 524 OHDSI. St’^ Observational Health Data Sciences and Informatics
OHSU. See Oregon Health & Science University OIG. See Office of Inspector General OMIM. See Online Mendelian Inheritance in Man
Omnibus rule, 355, 357 OMOP. Sfc Observational Medical
Outcomes Partnership ONC. See Office of the National Coordinator
Online Mendelian Inheritance in
Man (OMIM), 659
OPA. See Organizational Process Assets
Open Government Initiative, 360 OpenNotes website, 373 open-source software, 137,460-461 operating system, 137 ordinal data, 445,450
OSCE. See objective structured clinical examinations
OS Command Injection, 750 OTC market. See over-the-counter market Outcome and Assessment
Information Set (OASIS-C), 397 outcome measures, 214,445
outpatient data, 577 over-the-counter (OTC) market, 644
family needs and, 103-104 framework, with meaningful use categories, 118 healthcare costs and technology and, 112-114
interprofessionalism for, 104-105 using mobile applications, 111 national initiatives, 99-101 network schematic, 534
organizing frameworks of, 101-103
P4P. See Partnership 4 Patients parallel systems implementation. 197
parameter estimates, 453
precision medicine for consumers and, 107
provider-based models for, 105-106
Parent Charter, 554 mass vaccination, 555 Pareto charts, 540
provider competencies and,
Parkland Center for Clinical
technology and health coaching for health behavior change
Innovation (PCCI), 332-339
Parkland Health and Hospital System, 322,332 Partners Healthcare System (Boston, Massachusetts), 85,107
Partnership 4 Patients (P4P), 99, 769
PAs. See physician assistants PATCH. See Planned Approach to Community Health patient activation. See patient engagement
patient and family engagement framework, 102
patient-centered care, xvi, 36, 88, 104,205,577,617,630-631, 630,767
See also patient engagement; patient safety patient-centered medical homes (PCMH), 273,275,276 patient-centered outcomes research (PCOR), 653 Patient Centered Outcomes
Research Institute (PCORI), 361 Patient-Centered Outcomes Research Network
(PCORNet), 295,653, 655 Patient-Centered Outcomes
Research Trust Fund, 294
patient electronic access, to health information, 369
patient engagement, 6, 84,97,357, 370,372,375,381,389,518, 631,651,764,765
Oregon Health & Science University (OHSU), 473 Organizational Process Assets (OPA), 180 Orphanet, 660
citizen science and community
ORYX initiative, 591
by definition and, 98-99
engagement and, 108-109
collective community impact, 109 coordination of care through, 369
106-107
as success factor, 234-235
and, 109-112
Patient Engagement Committees, 235
Patient Engagement in Health and Health Care Framework, 101
patient-generated health data (PGHD),375 patient-generated health information (PGHI), 110-111, 375
patient identification errors, 513-514
patient portal. See personal health records
Patient Protection and Affordable
Care Act (PPACA/ACA), 7, 76, 88-91,100,101, 306-308, 566, 568, 592
patient response data, 577-578
patient safety, 4,11,13,16,23,110, 204,317,355,503-505,609
case study, 528 changes, 162-163 computers and, 124,126,132 CPOE and barcoding and, 549
data collection and reporting and, 516
data management and, 453,472, 473,475,476,482,483,490,491
device integration for, 171 downtime and disaster
preparedness procedures and protocols and, 515 EHRs and, 157,158,161,164, 233, 240,245-246, 248,255, 262, 615-633
electronic clinical quality measures and, 589-610
Emergency Care Research Institute (ECRI) annual report of errors and, 511-513
INDEX
Harrison Interactive
Sociotechnical Analysis Model and, 509-510
Hazard Manager by AHRQ and, 510-511
history, and Patient Safety Act, 505-506
HIT error/unintended
consequences clinical impact and,513-515
HIT patient safety issues mitigation strategies and, 518-527
PCOR. See patient centered outcomes research PCORl. See Patient Centered Outcomes Research Institute PCORNet. See Patient-Centered Outcomes Research Network
PDSA model. See Plan-Do-StudyAct model
Pearl robot, 668
persona! health data (PHD), 376 personal health information, 730 personal health records (PHRs), 108,143,365-367
interprofessional education and,
case study, 380-381
764, 771, 777-779 issues, and HIT, 506-510
digital divide and health literacy
national prevention strategy and, 563-585 national trend, 518 needs, and telehealth care, 516
operational model for expertise domains and, 509
PHR/portal impact on, 370,372, 373,375, 376
and quality, certified EHRs for, 167
quality improvement strategies and, 535-560
reporting levels, 516-517 scientific and theoretical foundations and, 44,51, 57, 60,61 telehealth and mobile health
and, 396-398,406 tools, 541-543 unintended adverse
consequences and, 508 unintended consequences
management guide and, 510 workflow redesign and, 210,212 Patient Safety and Quality Improvement Act (2005), 504, 505,516
patient safety organizations (PSOs), 505-506,516 patient satisfaction, with portal use, 378
patients/families and interprofessional education, 769
Patients Like Me website, 103
patient unified lookup system for emergencies (PULSE), 264-265 payer perspective and ACOs, 572 payloads, 745 PCCI. See Parkland Center for Clinical Innovation
P-chart, 545
PCMH. See patient-centered medical homes
definition of, 366 and, 371-373
impact on patient safety and unintended consequences, 373 interprofessional informatics to increase, 377-379
patient pathway, 368 patient portal adoption methods, 377-378
patient portal regulatory requirements and, 368-370 personal data contribution to, 373-377
success characteristics of, 378-379
personalized medicine, 644 consumer engagement and, 644-647
Internet of Things and, 748-749 personal/professional social media platforms, 723 pervasive computing. See ubiquitous computing Pew Research Center, 720, 721
PGHD. See patient-generated health data
PGHI. See patient-generated health information
pharmacist, 27 phased approach, to system implementation, 197 PHD. See personal health data PHI. See protected health information; public health
physician assistants (PAs), 26,79,221 Physician Quality Reporting Initiative, 591
physiomarkers, 713 Pinterest, 723
PIP. See Promoting Interoperability Program
place utility, 245 Plan-Do-Study-Act (PDSA) model, 51, 60-63, 537-538,538,556 Aimi Charter and, 554
improvement, 61 workflow redesign and, 537-538 Planned Approach to Community Health (PATCH), 310 PMBOK® Guide, 187, 201-204
PMC. See primary healthcare POC. See point-of-care point estimate, 453 point-of-care (POC), xxi, 11, 23,22, 23, 77,197,443,536,584,669, 690, 701,746
computers in healthcare and, 121-147
EHRs and, 151-171, 231-247, 253-277
genomics and, 648,652, 658
national standards for HIT and, 281-301
persona! liealth records and, 365-381 privacy and security and, 343-362 public health data and, 305-339 systems development life cycle and project management and, 175-206 telehealth and mobile health
and, 387-406
workflow redesign and, 209-223 point-of-care (POC) devices, 396-397 barcode medication
administration (BCMA) and, 159
challenges with, 159-164 communication technologies and,159
infrastructure PHIN. Sa'Public Health Information Network
IV smart pumps and, 159 for testing, 157-158 Polarity Map model, 523 polarity thinking, 36 Polarity Thinking”^” Model, 36
PHM. See population health
Ponemon Institute, 358,752
management
PHR. See personal health record PHRI. See Public Health Reporting Initiative
PHRs. See personal health records physical therapists (PTs), 26 physical therapy and rehabilitation, and AI, 714
807
population health, 360-361, 485^87,565-566 definition of, 567
issues, identification of, 568-575
quality metrics application in, 567-568
See also patient safety; quality metrics, for population health
808
INDEX
population health management (PHM), 568
possession utility, 245 postgenomics and nanotechnology, 674-675
Power BI (Microsoft), 463,465 advantages of, 465-467 interactive dashboard, 465 PPACA/ ACA. See Patient Protection and Affordable Care Act
PPCPs. See primary care providers PQIs. See Prevention Quality Indicators
Precision Medicine initiative, 107, 267
predictive analytics, 168,398, 458, 475
prescriptive analytics, 458,476 presentation software, 138 Prevention Quality Indicators (PQIs), 317, 322
PriceWaterhouseCoopers, 109 primary care providers (PPCPs), 82 primary healthcare (PMC), 236 logic model for, 236 privacy and confidentiality, of social media, 730
privacy and security, 343-344 BAAs, 354-355 case studies, 361-362
clinician role in PHI protection and, 358-359 electronic transactions and code
processes, meaning and significance of, 576 process-flow map, 548 process mapping, 211, 215 process measures, 214,445 process owners, 438 producer data stewards, 438 professional filters, applying, 732 recommended, 733
professional social media platforms, 723
Profound Knowledge Partners, 552 program evaluation, 233 clinician engagement as success factor and, 235
credible evidence gatliering and, 237
end users' satisfaction and, 237-239
evaluation design and, 235-236 framework for, 234 HIMSS Davies award for, 242-244
HIMSS EMRAM model and, 241-242
patient engagement and, 234-235 project description and, 235 project justification in, 240-241 return on investment and, 239-240
safe and effective use of
technology and, 247-248 stakeholder engagement and, 234
sets requirements and, 349-350 enforcement and, 353-354
success measures and, 244-246 time motion studies and, 239
Federal Trade Commission
use ensuring in, 241
(FTC) and, 345-346
Food and Drug Administration (FDA) and, 346 HIPAA and, 347-349 HITECH Act and, 355-357 international law and, 347
nurse's role and public trust and, 359-360 PHI de-identification methods
and guidance and, 351-352 population health and research data and, 360-361
Privacy Rule and, 350-351,360, 361
regulatory environment and, 345 Security Rule and, 352-353 state regulatory requirements and,346-347 substance abuse and mental health services administration
(SAMHSA) and, 346 Privacy Rule, 350-351, 354,360,361 complaints and status, 353
vendor evaluation of usability and adoption and, 239 program management, 422,423, 431,432,437 business-led, 425-426
programming languages, 134-135 project champion, 214 project charter, 181,182,236,540,543 designing, 212-214 quality improvement and, 539-540 template, LEAN, 213
Project Management Body of Knowledge. See PMBOK® Guide project management tools, in SLDC design phase, 187 Promoting Interoperability Program (PIP), 7,80,155,156, 177, 200, 294,473,485
clinician role in PHI protection and, 358
patient portal and, 370 See also interoperability; meaningful use (MU)
protected health information (PHI), 262, 271,344,347,354, 357
clinician role in protection of, 358 definition of, 351-352
prototypes and system-design goals, 185 provider-based models, for patient engagement, 105-106
provider competencies, patientengaging, 106-107 provider electronic transaction process, 351 providers and interprofessional education, 769
proximity index factors, 335 PSNet, 609
PSOs. See patient safety organizations
PTs. See physical therapists publication and subscribe exchange models, 264 Public Health Information
Network (PHIN), 282,286
Cascading Alert Checklist, 286 Preparedness Functional Requirement, 286 public health infrastructure (PHI), 306
Public Health Ontario, 732
Public Health Reporting Initiative (PHRI), 286
Puntenney, Deborah, 315
QCQI. See Quantum Computing Quantum Information
QDM. See Quality Data Model QHIN. See Qualified Health Information Networks
QHP, See Qualified Health Plan
QI. See quality improvement Qlik, 460,463 dashboard, illustration of, 464 Qlik Sense, 463 QlikView, 463
QPP. See Quality Payment Program
QRDA. See Quality Reporting Document Architecture
QSEN. See Quality and Safety Education for Nurses Qualified Health Information
Networks (QHIN), 258 Qualified Health Plan (QHP), 267 Quality and Safety Education for Nurses (QSEN), 36, 618,629, 631
competencies, levels of, 629-630, 629
INDEX
Quality and Safety Education in
random samples, 453
Nursing, xxviii Quality Data Model (QDM), 598
random variation. See noise
class attribute for, 608
example of, 599 quality improvement (QI), 44,169, 170,483, 535-536 aim statement of, 543 case studies, 547-550,552, 554-559
change implementation with, 544
control charts and, 545-546
defining and operationalizing, 50
Lean and, 544-545
nursing process and model for, 538-539,538
patient safety tools and, 541-543 PDSA cycle and, 537-538 review boards (QIRB), 550-551
significance of, 50 Six Sigma and, 544 Sociotechnical Systems Theory and,57-68 software as service for, 551-552
strategic plans focused on, 537 success measures of, 543-544 toolkit, in clinical case, 547 tools for, 539-541
workflow redesign and, 546-547 quality metrics, for population health
variation
rapid application development (RAD), 193,194 ratio data, 445
ratio of signal to noise, 123 RCA. See Root Cause Analysis RCE. See Recognized Coordinating Entity Readiness Assessment Checklist, 428, 429-433
readiness assessment cycle, 428-431
REC. See Regional Extension Center
Recognized Coordinating Entity (RCE), 258
record linkage methods for clinical quality management, 269
in HIEs, 268-269
referral management, 266 Regional Data Initiatives, 317 Regional Extension Center (REC),
273, 275, 283 Rochester, 270 valued services associated with. 266
mining, 694 regulatory changes, 162 release and training cycle, 431-432
sources of, 582-583
technology to support, 567-568 See also population health Quality Payment Program (QPP), 6, 287, 293,369, 593
of, 142
risk, meaning and significance of, 580-581
risk adjustment importance of, 581 interpretation of, 581 meaning of, 581 Risk Priority Number (RPN), 542 Robert Wood Johnson Foundation, 36,101,112, 237,308,388,505, 629, 764, 765
Robot for Interactive Body Assistance(RIBA), 668 robotics, 668-669
application in healthcare, 670, 673
implications, to nursing interprofessional teams,
case studies, 583-584
operationalization of, 567 providers and, 579-580 significance of, 579
Body Assistance ring networks, 141,142 advantages and disadvantages
informatics and
regression modeling and data
and, 576-583
RIBA. See Robot for Interactive
489
affected population and, 578 data to identify needs and generate, 573 development of, 578 measurable quality definition
RFPs. See requests for proposal RFQ. See request for a quote RHIOs. See Regional Health Information Organizations
78,80, 81-82,87,139,215, 222,
Regional Health Information Organizations (RHIOs), 256,
remote management
health promotion and, 397-398 history, current use, and future of, 396-397 workflow and volume
management predictive analysis and, 398 remote patient management (RPM), 391, 392,398
809
676-679
malfunctions, causes of, 678 Robot Laura®, 714
Robust Health Data Infrastructure, A report, 293 Rochester(NY) RHIO, 270 RockHealth, 10
Root Cause Analysis (RCA), 216, 540, 541-542 recommended actions list of, 542 rootkits, 745
RPM. See remote patient management
RPN. See Risk Priority Number "R" program, 458 RTI International, 506
rules and alerts, management of, 160
run charts, 540, 541
remote patient monitoring, 396, 516
Quality Reporting Document Architecture(QRDA),289,292 Quantified Self movement,108 Quantum Computing Quantum Information(QCQI), 686 query-based health information exchange, 262, 264
reporting systems, 575 request for a quote (RFQ), 184 request for information (RFIs), 181,
RACI diagram, 188 RAD. See rapid application development (RAD)
RESPECTS method, of communication skills, 619
184, 205,423
requests for proposal (RFPs), 181, 183,184 ResearchGate, 723
research question, 454 resilience coaching, 112
RFIs. See request for information
safe, timely, effective, efficient, equitable, and patientcentered (STEEEP), 214
descriptions of, 543 SAFER guide, See Safety Assurance Factors for the
EHR Resilience guide
Safety Assurance Factors for the EHR Resilience (SAFER)
guide, 164-165,247-248, 506-507,514-515,520
categories of, 507
810
INDEX
SAMHSA. See Substance Abuse and Mental Health Services Administration SANICS. See Self-Assessment
simulation, definition of, 618 simulation software, 138
of Nursing Informatics Competency Scale
SIPOC. See suppliers, inputs,
SAS® software, 583
single number estimate. See point estimate
process, outputs, customers
scatter diagram, 539,540 scatterplots, 541
situational analysis, in SLDC design phase, 187 Six Sigma, 544 sixth-generation programming language (6GPL), 135-136
schema, 417
SMART. See Substitutable Medical
SAS Enterprise Miner, 697 SAS software, 458
School of Nursing (SON), 273 Science and Technology Studies (STS), 53 scientific and theoretical
foundations, 43
building blocks of, 44-45 health information technology and,47-57
nursing informatics specialty and traditional midrange theory and, 45-47 Sociotechnical Systems Theory and QI and, 57-68
SCIP. Sc'c Surgical Care Improvement Project scope creep, 182 Scottsdale Institute, 515 scrum master, 195
SDLC. See Systems Dev'elopment Life Cycle SDOs. See standing delegated order sets
second-generation machine language (2GL), 134-135 secure socket layer (SSL), 140 security risk assessment (SRA), 359
Applications, Reusable Technologies smart ambulances, 712
SMART App Gallery, 289 Smart Contracts (Block Chain), 714 SMART platform, 84 time and trajectory (2009-2020), 85
SML. Sa? supervised machine learning SMTP. Set’ Simple Mail Transfer Protocol
Snapehat, 723 SNOMED-CT. See Systematized
social context, 577-578 social determinants of health
(SDOH), 108,109,332, 566 social media, 719-721
Sequoia Project, 258-259,265,271
data collection, analysis, and
Rate law
SHARP. See Strategic Health IT Advanced Research Projects SHARPS. See Strategic Health IT advanced research projects on Security SHIEC. See Strategic Health Information Exchange Collaborative SIM. See State Innovation Models
simple linear workflow diagrams, 211
Simple Mail Transfer Protocol (SMTP),285
sociotechnical-compatible conceptual perspectives, 51, 52-53
sociotechnical interaction, 132
sociotechnical systems, 47,53 in conjunction with qualityimprovement activities, 57-68 software
advanced programming languages and big data
selection, for EHR, 138-139
graphical representation of, 421 SNP Technologies, Inc., 465 social accountability partnership program, 395
case study, 735
language (7GPL), 136-137
social media traffic, 727 social robots, 668
snowflake schema, 417
(SANICS), 623 Sentinel Initiative, of FDA, 295
SGR law. See Sustainable Growth
722-723
social media analytics, 727 social media and public health, case study of, 63-68
Nomenclature of MedicineClinical Terms Snowden, Edward, 752
building connections and social
seventh-generation programming
in, 724-725
strategy, developing, 733-734 tools, platforms, and applications of, 723 user profile component of,
computing, 135-137 application, 138 internal component types of, 137 programming language classificationsand, 134-135
Self-Assessment of Nursing Informatics Competency Scale
service robots, 668,678
professional presence and strategies for use of, 731-734 sharing and content generation
networks in, 724
content management in, 726 feedback in, 726-727 definition of, 721-722
engaged, empowered, and evolved patients and, 727-728 ethical considerations for use of, 730-731
event management in, 726 impacting traditional informatics, 734 individual, societal, and
professional impact of, 728-734 interactive component of, 724 interprofessional implications of, 735 live feeds in, 726
live streaming in, 726 professional implications of, 730-731
types of, 137-138 SON. See School of Nursing source system data, 419 spaghetti diagram, 211 special-cause variation, 456 spreadsheet applications, 458 spreadsheet software, 138 Sprint process, 169,188 SQL. See Structured Query Language SRA. See security risk assessment SSL. See secure socket layer standards and interoperability framework, 265
standards development clinical document standards and, 289
clinical quality measures and, 289,292
code sets, vocabularies, and values and, 292-293 common data model and, 294-295
common data standards and, 287
EHR functional specifications and,288
interoperability and, 289 meaningful use program and. 294
medical devices and, 295
INDEX
medical imaging and communication and, 292
superusers, role of, 166, 198, 199
messaging and, 288 potential gaps in, 295-296 standing delegated order sets (SDOs), 495,496
supervised machine learning
star networks, 242, 142
suppliers/vendors and interprofessional education,
advantages and disadvantages of, 242 star schema, 417
graphical representahon of, 420 Stata software, 458
state agency connectivity, 266 State Innovation Models (SIM), 88
statistical analysis, 452-453 statistical analysis plan, 455 statistical packages, 458,459 statistical test, selection of, 454-456
Stats Plus (Mac version), 458
STEEEP. See safe, timely, effective, efficient, equitable, and patient-centered sticky-note method, 215 stock-out, 547-549
strategic development and EHR, 164
Strategic Health Information Exchange Collaborative (SHIEC),259 Strategic Health IT Advanced Research Projects (SHARP), 79 -C (SHARP-C) physician cognition, 83-84,246 grants, 83-85 research and technology development grants of, 83 -S (SHARP-S) privacy and security, 83 Strategic Health IT advanced research projects on Security (SHARPS), 405
structure, meaning and significance of, 576 Structured Query Language (SQL), 702
structure-process-outcome framework, 66-68,576
STS. See Science and Technology Studies
subject matter experts, 438
substance abuse and mental health services administration
(SAMHSA), 346 Substitutable Medical
Applications, Reusable Technologies (SMART), 650, 651
on FHIR genomics, 650, 652 summarization and data mining. 694
(SML), 754
suppliers, inputs, process, outputs, customers (SIPOC), 211
769-770
surgery and post-surgical recovery, and AI, 713
Surgical Care Improvement Project (SCIP), 162
survey data, 573-574 SUS. See Systems Usability Scale sustainability and change management, 161-162 SustainableGrowth Rate (SGR) law, 6
swimlane diagrams/flowcharts, 211,218
symmetrical and asymmetrical user connections, 725
synchronous telemedicine, 394 syndromic surveillance, 286 system analysis, 179, 202 goals, 183,184 outputs, 184-185,190-191 tools, 183-184
Systematized Nomenclature of Medicine-Clinical Terms
(SNOMED-CT), 288, 292, 293, 297,299
system configuration, in SLDC design phase, 186-187 system design, 179, 202 goals, 185-186,186 roles and skills, 191 tools, 186
system evaluation maintenance and support and, 279,202 metrics for, 200
support phase and implementation and, 196 system implementation communication plan for, 198 education and training considerations for, 198-200
evaluation and support phase and, 196-200
evaluation, support tools, and strategies and, 197 expectations and issues management with, 197 with new workflows, 216
phases of, 197 support goals and, 196-197 testing and, 179,202 system maintenance, 200 System of Systems solution, 711
811
system planning, 179, 202 goals, 179-180 outputs, 182 tools, 180-181
system prototyping, 193
Systems Development Life Cycle (SDLC), 165,175-177
agile development, 194,195 analysis phase of, 182-185 Design and Development phases of, 203-204 design phase of, 185-193 Disposition phase, 204 EHR and, 204-206 evaluation in context of, 233 informatics' roles in, 181-182, 185,191-192,196,199
phases of, 178 planning goals in, 180 planning phase of, 178-182,182 and professional project management framework compared, 200-204 rapid application development (RAD), 193,194
Requirements Analysis, 203 significance of, 177-178 structures to establish, 183
system implementation and, 196-200
systems testing and, 195-196 testing phase of, 195 waterfall development and, 193, 194
systems testing goals and considerations, 195 informatics' roles and skills in, 196
outputs, 196 tools, 196
Systems Usability Scale (SUS), 237, 239
system testing, 195 T2 Mood Tracker app, 110 Tableau, 460,461 dashboard, illustration of, 462
learning webinar series, 462-463 Public, 461^62
TAM. See Technology Acceptance Model
Task, User, Representation and Function (TURF), 84,246 TCP/IP. See transmission control
protocol/Internet protocol TCPI. See Transforming Clinical Practice Initiative
team practice/simulation and interprofessional education, 770-771
812
INDEX
TeamSTEPPS model, 521-522 technical data stewards, 438
Texas Board of Nursing, 394
technical exchange, of data, 262
Texas Department of State Health Services (DSHS), 147
technology, overreliance on,
Texas Health Care Information
160-161
Technology Acceptance Model (TAM), 47 Technology Informatics Guiding Education Reform (TIGER), 10, 35,617, 771
competencies by category, 37-38 competencies for interprofessional education, 37 Initiative Foundation, 523
International Competency Synthesis Project, 35, 617 patient safety and, 522-523 ten-year vision of, 772 technology standards, for healthcare, 294
TEFCA. Sc£,’Trusted Exchange Framework and Common
Agreement telehealth, 6, 387-390
case study, 405^06 characteristics by, 390 CMS chronic care management revenues and, 399-401
implementation, financing, and sustainability and, 394-396 implementation and licensure of, 398-399
meaning and significance of, 390 mobile health and, 401-404
pandemic impact on, 388-391 policies, regulations, and security, 404-405 problems to solve and, 391-393 regional resource centers of, 396 remote management and, 396-398 in simulation, 626-627
telemedicine and, 393-394,396
telehealth bridge course, example of, 776
telehealth care and patient safety needs, 516 telehealth nurse, 29
Telehealth Nursing Special Interest Group, 29 telehealth resource centers (TRCs), 396
telemedicine, 710
A MA's pol icy on, 396 history and current use, 393 nursing and, 394 in simulation, 626-627
types and trends of, 393-394 telephone triage, 397, 398
terminology standards supporting interoperability, 419
Council (THOC), 583 Texas Health Resources, 200,243
Texas Health Steps (THS), 147 Texas Nurses Association (TNA), 30,525,526, 765 evaluation of, 248
Texas Organization of Nurse Executives(TONE),30,525,765 Texas Organization for Nursing Leaders (TONE), 248 Texas State University College of
tree networks, 141,142
advantages and disadvantages of, 142
Tri-Council for Nursing, 565-566 Triple Aim framework, 689-690, 708, 773, 775
for clinician well-being, 168-169, 273
trojans, 745 Trump, Donald, 81
Trusted Exchange Framework and Common Agreement (TEFCA), 90,258 national initiatives, 259 Qualified Health Information
Health Professions, 777
Networks (QHIN) and, 258
Texas Tech University Health Sciences Center (TTUHSC),
Recognized Coordinating Entity
771
(RCE) and, 258 TTUHSC. See Texas Tech
Texas Woman's University interprofessional informatics program, 775, 776
TURF. See Task, User,
THCIC. See Texas Health Care Information Council
Representation and Function turnable QDs, for imaging cells,
The Joint Commission (TJC), 591 theoretical framework, 44
University Health Sciences Center
671
Twitter, 723
Think Sepsis, case study of, 494-495
third-generation machine language (3GL), 135 THS. See Texas Health Steps Tietze Telehealth Framework, 392, 516
TIGER. See Technology Informatics Guiding Education Reform TikTok, 723
time utility, 245 TJC. Set’ The Joint Commission TNA. See Texas Nurses Association
To Err Is Human report (NAM), 154, 505, 507, 764-765
TONE. See Texas Organization of Nurse Executives
TONL. See Texas Organization for Nursing Leaders Top 10 Health Technologi/ Hazards report, 512
Top 10 Patient Safet\/ Concerns for Healthcare Organizations report, 512
Top 10 Strategic Technology Trends for 2014, The report, 143 Transforming Clinical Practice Initiative (TCPI), 82
transmission control protocol/ Internet protocol (TCP/IP), 140 transparent transfer, of patient generated data, 204 TRCs. See telehealth resource centers
ubiquitous computing, 743-744, 746
U-chart, 545
UDP. Set.’ unified data protection UIUC. See University of Illinois at Urbana—Champaign UMC. See University Medical Center
Unfolding Case Studies for Genetics & Genomics
Healthcare Education, 660
unidirectional maps, 452 unified data protection (UDP), 746 Unified Theory of Acceptance and Use of Technology (UTAUT), 47
Uniform Resource Identifiers
(URI), 292 uniform resource locator (URL), 140
United States Core Data for
Interoperability (USCDI), 265, 267
unit testing, 195 University Medical Center (UMC), 380
University of Arkansas five-pillar IPE model, 773, 774
University of California, Los Angeles (UCLA)-HIA toolkit, 322
University of Florida, 109
INDEX I 8(3 University of IHinois at Urbana— Champaign (UIUC), 83 University of Michigan, 270 University of Minnesota, 15,34,690 University of Notre Dame, 271 University of Texas at Houston
UTAUT. See Unified Theory of Acceptance and Use of Technology utility softw'are, 337 utility/usefulness, categories of. 245
(UT Houston), 83
University of Utah, 153 University of Washington, 776 University of Wisconsin County Health Rankings model, 311-312,323,333 unstructured data and notable
global leaks, 753 URL See Uniform Resource Identifiers URL. See uniform resource locator
usability
value-based payment (VBP), 266 Value Set Authority Center (VSAC), 287,482
value stream map, 211 variable, definition of, 445
VBP. See value-based payment Veritas company, 646
virtual learning environment (VLE), 771 viruses, 745
definition of, 245
Vision for the Changing FacnHy Role,
improvement strategy, 544 testing, 474
VLE. See virtual learning
USCDI. See United States Core
Data for Interoperability U.S. Crowd Sourcing Citizen Science Act (2017), 108
U.S. Department of Agriculture's Distance Learning and Telemedicine (DLT), 405
U.S. Department of Defense, 110, 153
U.S. Department of Health and Human Services (DHHS), 4, 5,6, 8,78,81,91,100,109,154, 267,348, 368, 395,765
Use ofBlockchain in Health IT and Health-Related Research, The, 289
user requests, 163 U.S. Food and Drug Administration (FDA), 10, 158,185, 295, 346,644, 645-646 Sentinel Initiative of, 295 U.S. Office of Disease Prevention and Health Promotion, 371
U.S. Pharmacopeia, 644 U.S. Surgeon's General Initiative family history and, 647-648 My Family Health Portrait tool, 648
A (NLN document), 618 environment VOC. See voice of customer
voice of customer (VOC), 425,431, 432
VSAC. See Value Set Authority Center
vulnerability assessment framework, 336
Vulnerability Index tool, 335
wikis, 726
WinQl software, 583 wireframe, 432
example of, 433 Witlium Digital, 465 word processors, 338 Work Breakdown Structure (WBS), 188,201,203
workflow diagram steps, in Microsoft Visio Professional
2019,217-218
workflow redesign, 209-211, 549 best practices for, 215-216 case studies, 221-223
clinical decision support and, 488^89
for EHR support, 487 interoperability promotion and technology optimization and, 212
new advances in, 220-221
PDSA cycle and, 537-538 in quality-improvement modality, 212-214 software usage to support, 217-218
templates, 218-220,219 workflow software, 138 Workstations on Wheels (WOWs), 144
World Health Organization WAN. See wide area network
WannaCry ransomware attack (2017), 753
waterfall development approach, 193,394,201 WBS. See Work Breakdown Structure
Web 2.0 technologies, 720 WellPoint, Inc., 703
WGS. See whole-genome sequencing
(WHO), 292, 393,395, 513, 642, 765
World Trade Organization (WTO), 128
worms, 745 WOWs. See Workstations on Wheels WTO. See World Trade
Organization X-MR chart. See I-chart
WHO. See World Health
Organization
whole-genome sequencing (WGS), 645
Why Women Ar.e Stripey, 660 wide area network (WAN), 140
YouTube, 723