Health Information Systems: Architectures and Strategies [2 ed.] 1849964408, 9781849964401, 1849964416, 9781849964418
Health Information Systems: Architectures and Strategies is a definitive introductory resource that tackles the pivotal
214 72 6MB
English Pages 340 [382] Year 2011
Front Matter....Pages i-xliii
Introduction....Pages 1-2
Health Institutions and Information Processing....Pages 3-23
Information System Basics....Pages 25-32
Health Information Systems....Pages 33-42
Modeling Health Information Systems....Pages 43-73
Architecture of Hospital Information Systems....Pages 75-183
Specific Aspects for Architectures of Transinstitutional Health Information Systems....Pages 185-199
Quality of Health Information Systems....Pages 201-236
Strategic Information Management in Hospitals....Pages 237-282
Strategic Information Management in Health Care Networks....Pages 283-289
Final Remarks....Pages 291-292
Back Matter....Pages 293-337
Recommend Papers
![IT architectures and middleware strategies for building large, integrated systems [2. ed., 4. print]
0321246942, 9780321246943](https://ebin.pub/img/200x200/it-architectures-and-middleware-strategies-for-building-large-integrated-systems-2-ed-4-print-0321246942-9780321246943.jpg)
![Health Care Information Systems - 5th Edition [5 ed.]
9781119853879](https://ebin.pub/img/200x200/health-care-information-systems-5th-edition-5nbsped-9781119853879.jpg)
![Health Information Systems: Architectures and Strategies [2 ed.]
1849964408, 9781849964401, 1849964416, 9781849964418](https://ebin.pub/img/200x200/health-information-systems-architectures-and-strategies-2nbsped-1849964408-9781849964401-1849964416-9781849964418.jpg)
- Author / Uploaded
- Alfred Winter
- Reinhold Haux
- Elske Ammenwerth
- Birgit Brigl
- Nils Hellrung
- Franziska Jahn (auth.)
- Similar Topics
- Art
- Design: Architecture
File loading please wait...
Citation preview
Health Information Systems Second Edition
AlfredWinter • ReinholdHaux ElskeAmmenwerth • BirgitBrigl NilsHellrung • FranziskaJahn KathrynJ.Hannah • MarionJ.Ball (SeriesEditors)
Health Information Systems Architectures and Strategies Second Edition WithaForewordbyReedM.Gardner
Authors Alfred Winter, PhD Professor of Medical Informatics Institute for Medical Informatics, Statistics and Epidemiology University of Leipzig, Germany [email protected] Reinhold Haux, PhD Professor of Medical Informatics Peter L. Reichertz Institute for Medical Informatics University of Braunschweig – Institute of Technology and Hannover Medical School, Germany [email protected] Elske Ammenwerth, PhD Professor of Medical Informatics Institute for Health Information Systems University for Health Sciences, Medical Informatics and Technology (UMIT), Austria [email protected]
Birgit Brigl, PhD Friedrichsdorf, Germany [email protected] Nils Hellrung, PhD Peter L. Reichertz Institute for Medical Informatics, University of Braunschweig – Institute of Technology and Hannover Medical School, Germany [email protected] Franziska Jahn, Dipl.-Inf. Institute for Medical Informatics, Statistics and Epidemiology University of Leipzig, Germany [email protected]
ISBN 978-1-84996-440-1 e-ISBN 978-1-84996-441-8 DOI 10.1007/978-1-84996-441-8 Springer London Dordrecht Heidelberg New York British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Control Number: 2010933608 © Springer-Verlag London Limited 2011 Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patents Act 1988, this publication may only be reproduced, stored or transmitted, in any form or by any means, with the prior permission in writing of the publishers, or in the case of reprographic reproduction in accordance with the terms of licenses issued by the Copyright Licensing Agency. Enquiries concerning reproduction outside those terms should be sent to the publishers. The use of registered names, trademarks, etc., in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore free for general use. Product liability: The publisher can give no guarantee for information about drug dosage and application thereof contained in this book. In every individual case the respective user must check its accuracy by consulting other pharmaceutical literature. Cover design: eStudioCalamar Figueres/Berlin Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)
DedicatedtoProf.Dr.ClausO.Köhler(1935-2008)
“Anytechnologysetsarelationshipbetweenhumanbeingsandtheirenvironment, bothphysicalandhuman.Notechnologycanbeseenasmerelyinstrumental.This is especially relevant when dealing with large automatic information systems, developedtocontributetothemanagementandintegrationoflargeorganizations, suchashospitals.” Jean-MarieFesslerandFrancoisGrémy(irstrecipientoftheIMIAAwardofExcellence).Ethical ProblemswithHealthInformationSystems.MethodsofInformationinMedicine2001;40:359-61.
“1.1WhyDoWeNeedBiomedicalandHealthInformatics[BMHI]Education?
Despitethedocumentedbeneits,therearestillbarrierstoHIT[healthinformation technology] in clinical settings, including a mismatch of return on investment betweenthosewhopayandthosewhobeneit,challengestoameliorateworklowin clinicalsettings,lackofstandardsandinteroperability,andconcernsaboutprivacy andconidentiality…. Anotherbarrier,lesserstudiedandquantiiedbutincreasinglyrecognized,isthelack of characterization of the workforce and its training needed to most effectively implementHITsystems…ThishasledtocallsforBMHItobecomeaprofessional discipline … and for it to acquire the attributes of a profession, such as a welldeinedsetofcompetencies… … Table2:Recommended…learningoutcomesintermsoflevelsofknowledgeand skillsforprofessionalsinhealthcare… … 1.6 Characteristics,functionalitiesandexamplesofinformationsystemsinhealthcare… 1.7 Architecturesofinformationsystemsinhealthcare… 1.8 Managementofinformationsystemsinhealthcare… …” RecommendationsoftheInternationalMedicalInformaticsAssociation(IMIA)onEducationin BiomedicalandHealthInformatics-1stRevision.MethodsofInformationinMedicine2010;49: 105-120.
Foreword from the 1st Edition in 2004
Healthcare management is a complex and ever-changing task. As medical knowledge increases, as clinical management strategies or administrative management strategies change,andaspatientsmovefromonecitytoanotherorfromonecountrytoanother,the challengesofmanaginghealthcarehavechangedandwillcontinuetochange.Centralto allofthesechangesisaneedtostoreandprocessadministrativeandclinicalrecordsfor thepatient.Forthereasonslistedabove,computerizationofrecordsystemsinhospitals andclinicshasbeenandcontinuestobeaslowandcomplexprocess.Developingastrategytoprovidethebesthealthcareserviceatthelowestpossiblecostisacommongoalof almosteveryhealthcaresystemintheworld.Caregiveninthehospitalistypicallythe mostadvanced,themostcomplex,andthemostexpensive.Asaconsequence,understandingandmanaginghealthcareinhospitalsiscrucialtoeveryhealthcaredeliverysystem. Thisbookprovidesawonderfuloverviewforstudentsandmedicalinformaticsprofessionals. It also provides the background that every medical informatics specialist needs to understandandmanagethecomplexitiesofhospitalinformationsystems. Thisbookdealsprimarilywiththeunderlyingadministrativesystemsthatareinplace inhospitalsthroughouttheworld.Thesesystemsarefundamentaltothedevelopmentand implementationoftheevenmorechallengingsystemsthatacquire,process,andmanage thepatient’sclinicalinformation.Hospitalinformationsystemsprovideamajorpartofthe informationneededbythosepayingforhealthcare,betheyhospitaladministrators,health insurancecompanies,publichealthauthorities,orlocalornationalpoliticalleaders.Asa consequenceanimportantandcomplexsetofstrategieshasbeenimplementedtodocumentmedicalproblemsandproceduresthathospitalsaredealingwith.ProblemsareusuallycodedwithInternationalClassiicationofDiseases(ICD-9orICD-10)codingsystems while medical procedures are designated using Current Procedural Terminology (CPT) codes.Typically,thesecodesareusedtogeneratebillstoaninsurancecompanyorgovernmentalunit.Asaconsequence,thesedatamustbegenerated,transmitted,andprocessed accurately and promptly. Computer technology enhances the ability of hospital clinical andadministrativestafftoprovidethesedata.
vii
viii
Foreword from the 1st Edition in 2004
Becauseofthecomplexitiesandchangingneedsofmedicalinformation,theieldof medicalinformaticsisinneedofagrowingnumberofprofessionalswhounderstandhow tousecomputersandarefamiliarwiththeadministrativerequirementsofthehealthcare ieldandclinicalmedicine.Havingapersonwhohasknowledgeinalloftheseieldsis unusual.However,Iamconvincedthattherateatwhichmedicineisabletobetteruse computertechnologyislimitedbythelackofasuficientnumberofwell-trainedprofessionalswhohaveanunderstandingofalloftheseields.Asaconsequence,Icongratulate eachofyouwhoisstudyinghospitalinformationsystemsandencourageyoutotakewhat youwilllearnfromthisbookandmovetheieldforward. Afteryouhaveanunderstandingofwhatispresentedinthistext,Iencourageyouto takeonthechallengeofclinicalinformatics.Studyandlearnhowcomputerscanbeused toadvantagebythoseprovidingclinicalcare–physicians,nurses,pharmacists,therapists, andothercaregivers.Inthefuturewemustallworktowarddevelopingcomputerandcommunicationssystemsthatwillenhancetheacquisitionofclinicaldatasothatthedatacan be used to provide better patient care and more eficient and better administrative documentation. Enjoythisbook.Itsclearlywrittenmaterialsandexercisesshouldgiveeveryreadera challengeandopportunitytolearn.IfoundAppendixA,thethesaurus,atreasureofimportantinformation.Thethesauruswillbeveryhandyforeveryoneforyearstocome.Icongratulate the authors for their knowledge, skillfulness, and dedication in writing and publishingthisbook. SaltLakeCity,Utah,USA
ReedGardner
Series Preface
Thisseriesisdirectedtohealthcareprofessionalsleadingthetransformationofhealthcare byusinginformationandknowledge.Forover20years,HealthInformaticshasoffereda broad range of titles: some address speciic professions such as nursing, medicine, and healthadministration;otherscoverspecialareasofpracticesuchastraumaandradiology; stillotherbooksintheseriesfocusoninterdisciplinaryissues,suchasthecomputer-based patientrecord,electronichealthrecords,andnetworkedhealthcaresystems.Editorsand authors,eminentexpertsintheirields,offertheiraccountsofinnovationsinhealthinformatics.Increasingly,theseaccountsgobeyondhardwareandsoftwaretoaddresstherole ofinformationininluencingthetransformationofhealthcaredeliverysystemsaroundthe world.Theseriesalsoincreasinglyfocusesontheusersoftheinformationandsystems: theorganizational,behavioral,andsocietalchangesthataccompanythediffusionofinformationtechnologyinhealthservicesenvironments. Developmentsinhealthcaredeliveryareconstant;inrecentyears,bioinformaticshas emergedasanewieldinhealthinformaticstosupportemergingandongoingdevelopments in molecular biology. At the same time, further evolution of the ield of health informatics is relected in the introduction of concepts at the macro or health systems deliverylevelwithmajornationalinitiativesrelatedtoelectronichealthrecords(EHR), datastandards,andpublichealthinformatics. These changes will continue to shape health services in the twenty-irst century. By makingfullandcreativeuseofthetechnologytotamedataandtotransforminformation, HealthInformaticswillfosterthedevelopmentanduseofnewknowledgeinhealthcare. KathrynJ.Hannah MarionJ.Ball
ix
Preface for the 2nd Edition
In2004,thetextbook“StrategicInformationManagementinHospitals–AnIntroduction toHospitalInformationSystems”appearedinthisHealthInformaticsSeriesofSpringer. Thebookwasreceivedwellandbelongs–accordingtothepublishinghouse–tothetop sellingbooksofthisseries. Fiveyearsafteritsappearance,bothSpringerandweasauthorsfelttheneedtoprepare a2ndedition.Inthis2ndeditionwewantedtoconsidertheprogressinourieldandalso the lessons learned from our students, when using the book in our lectures, e.g. in the international “Frank van Swieten Lectures on Strategic Information Management in Hospitals”(InternationalJournalofMedicalInformatics2004;73,97-100and807-15). Also,duetothechangedperceptionofinformationsystemsinhealthcare,whichareno longerlimitedtosingleinstitutionslikehospitals,butcanembraceseveralhealthcareinstitutionswithinhealthcarenetworks,thisrevisionhasbecomenecessary. With this book on “Health Information Systems – Architectures and Strategies” we havepreparedasubstantiallyrevisedandelaborated2ndedition. What are the major differences from the 1st edition, which appeared in 2004? We shifted the focus from hospital information systems and their strategic management to strategicinformationmanagementofhealthinformationsystems.However,information systemsinhospitalsstillplayamajorroleinthebook.Allcontentshavebeencarefully updated. The book has been restructured in order to improve the use as a textbook for lecturesonhealthinformationsystems. Inadditiontothefourauthorsofthe1stedition,twonewauthors(N.H.,F.J.)havealso contributedtothisedition.Forthe2ndeditionwealsoinvitedaninternationalboardof experts–CIOsandresearchersintheieldofhealthinformationsystems–togiveustheir adviceandcomments,includingexamplesandusecases.
xi
xii
Preface for the 2nd Edition
Allauthorshavebeeneitherdirectlyorindirectlyinluencedbythevisionaryviewson health information systems of Claus O. Köhler. Dr. Köhler was Professor of Medical Informatics at the German Cancer Research Center in Heidelberg, Germany, and longtermfacultymemberattheHeidelberg/HeilbronnMedicalInformaticsProgram.Hisbook onthe‘integratedhospitalinformationsystem’,whichappearedinGermanyin1973,signiicantlyinluencedthedevelopmentofhealthinformationsystemsatleastinGermany. Clauspassedawayin2008.Wewanttodedicatethisbooktohim.
AlfredWinter ReinholdHaux ElskeAmmenwerth BirgitBrigl NilsHellrung FranziskaJahn
Acknowledgements for the 2nd Edition
Alsoforthis2ndeditionwewanttoexpressourcordialthankstoallcolleagues,contributingtothisbook,inparticulartoBakheetAldosari,CarlDujat,ChristopherDuwenkamp, MarcoEichelberg,GertFunkat,MowafaHouseh,AlexanderHörbst,FlorianImmenroth, HagenKosock,MichaelMarschollek,BassimaSaddik,PaulSchmücker,ClaudiaSiemersMarschollek,andThomasWendt. Wearegratefultothemembersofthebook’sInternationalAdvisoryBoard–leadingCIOs andresearchersintheieldofhealthinformationsystems–forspendingtheirtimeincommentingonthisbookandgivingussigniicantadvice.Itisourhopethatwecouldwell includethesecomments. Themembersofthebook’sInternationalAdvisoryBoardareDominikAronsky(Nashville, USA),WillemJanterBurg(Amsterdam,TheNetherlands),YoungMoonChae(Seoul, Korea), Andrew Grant (Sherbrooke, Canada), Rada Hussein (Cairo, Egypt), Georg Lechleitner (Innsbruck, Austria), Christoph U. Lehmann (Baltimore, USA), Dirk May (Hannover,Germany),FernánGonzálezBernaldodeQuirós(BuenosAires,Argentina), ChristophSeidel(Braunschweig,Germany),AmnonShabo(Haifa,Israel),JanErikSlot (Amsterdam, The Netherlands), Katsuhiko Takabayashi (Chiba, Japan), and Majid Al Tuwaijri(Riyadh,SaudiArabia). Again,ourstudentshelpedusalotbyevaluatingourlecturesandbyprovidinguswith manyconstructiveideasandhelpfulcomments.Inaddition,wethankUlrikeWeberwho designedalotoftheiguresinthisbook.
xiii
Acknowledgements for the 1st Edition
Wewouldliketoexpressourthankstoallofourcolleagueswhocontributedtothisbook, especiallyReedGardner,whocommentedonitandwrotetheForeword.Thanksalsoto manyotherpeoplewhohelpedtoproducethisbook,especiallyFriedaKaiserandGudrun Hübner-Bloder. Wewouldalsoliketothankthefollowingcolleaguesforhelpingtoobtainiguresandscreen shots:MarcBatschkus,ThomasBürkle,AndrewGrant,TorstenHappek,MarianneKandert, Thomas Kauer, Georg Lechleitner, Otwin Linderkamp, André Michel, Gerhard Mönnich, OliverReinhard,ChristofSeggewies,PierreTetrault,RaimundVogl,andImmanuelWilhelmy. Inparticular,wearegratefultoUrsulaandMarkusBeutelspacherforallowingustotakea pictureoftheirHeidelbergquintupletsforthecover(quintupletpicturebyBerndKrug). Notleast,wewanttothankourstudents,whokeptaskingcriticalquestionsanddrew ourattentiontoincompleteandindistinctarguments.
xv
Annotation to the Figures
Allpersonsshowninthephotoshavegiventheirpermission.Withtheexceptionofthe Heidelbergquintuplets,norealpatientsareshown.Thepatientsintheiguresaremostly theauthors,theirfamilies,ormedicalinformaticsVIPs.Wehavepartlyusedscreenshots fromcommercialsoftwareproductsinthisbook.Thisusecannotberegardedasarecommendationforthoseproducts.Weonlywanttoillustratetypicalfunctionalityandtypical userinterfacesofsoftwareproductsthatsupportspeciichospitalfunctions.Therefore,we didnotmentiontheproductnames.
xvii
Contents
1 Introduction........................................................................................................
1
2 HealthInstitutionsandInformationProcessing............................................. 2.1 Introduction................................................................................................. 2.2 SigniicanceofInformationProcessinginHealthCare.............................. 2.2.1 InformationProcessingasQualityFactor....................................... 2.2.2 InformationProcessingasCostFactor........................................... 2.2.3 InformationasProductivityFactor................................................. 2.2.4 HolisticViewofthePatient............................................................ 2.2.5 HospitalInformationSystemasMemoryandNervousSystem..... 2.3 ProgressinInformationandCommunicationTechnology......................... 2.3.1 ImpactontheQualityofHealthCare............................................. 2.3.2 ImpactonEconomics...................................................................... 2.3.3 ChangingHealthCare..................................................................... 2.4 ImportanceofSystematicInformationManagement.................................. 2.4.1 AffectedPeopleandAreas.............................................................. 2.4.2 AmountofInformationProcessing................................................. 2.4.3 SharingtheSameData.................................................................... 2.4.4 IntegratedInformationProcessingtoSatisfyInformationNeeds...... 2.4.5 RaisingtheQualityofPatientCareandReducingCosts................ 2.4.6 BasisofSystematicInformationProcessing................................... 2.5 Examples..................................................................................................... 2.5.1 KnowledgeAccesstoImprovePatientCare................................... 2.5.2 NonsystematicInformationProcessinginClinicalRegisters......... 2.5.3 TheWHOeHealthResolution........................................................ 2.5.4 EstimatedImpactofeHealthtoImproveQuality andEficiencyofPatientCare........................................................ 2.6 Exercises..................................................................................................... 2.6.1 AmountofInformationProcessinginTypicalHospitals................ 2.6.2 InformationProcessinginDifferentAreas..................................... 2.6.3 GoodInformationProcessingPractice........................................... 2.7 Summary.....................................................................................................
3 3 3 3 4 6 6 7 8 8 10 11 12 12 13 14 15 16 16 17 17 18 19 21 22 22 22 23 23
xix
xx
Contents
3 InformationSystemBasics................................................................................ 3.1 Introduction................................................................................................. 3.2 Data,Information,andKnowledge............................................................. 3.3 InformationSystemsandTheirComponents.............................................. 3.3.1 SystemsandSubsystems................................................................. 3.3.2 InformationSystems....................................................................... 3.3.3 ComponentsofInformationSystems.............................................. 3.3.4 ArchitectureandInfrastructureofInformationSystems................ 3.4 InformationManagement............................................................................ 3.5 Exercises..................................................................................................... 3.5.1 OntheTermInformationSystem.................................................... 3.5.2 OnEnterpriseFunctions.................................................................. 3.5.3 OnApplicationComponents........................................................... 3.5.4 OnArchitecturesandInfrastructures.............................................. 3.5.5 OnInformationManagement.......................................................... 3.6 Summary.....................................................................................................
25 25 25 26 26 26 27 29 30 30 30 31 31 31 31 31
4 HealthInformationSystems.............................................................................. 4.1 Introduction................................................................................................. 4.2 HospitalInformationSystems..................................................................... 4.3 TransinstitutionalHealthInformationSystems........................................... 4.4 ElectronicHealthRecordsasaPartofHealthInformationSystems.......... 4.5 ChallengesforHealthInformationSystems............................................... 4.6 Example....................................................................................................... 4.6.1 ArchitectureofaHospitalInformationSystem.............................. 4.7 Exercises..................................................................................................... 4.7.1 HospitalInformationSystemasaSystem...................................... 4.7.2 BuyingaHospitalInformationSystem........................................... 4.7.3 TransinstitutionalHealthInformationSystems.............................. 4.8 Summary.....................................................................................................
33 33 33 36 38 38 40 40 41 41 41 41 42
5 ModelingHealthInformationSystems............................................................ 5.1 Introduction................................................................................................. 5.2 OnModelsandMetamodels....................................................................... 5.2.1 Deinitions....................................................................................... 5.2.2 TypesofModels.............................................................................. 5.3 AMetamodelforModelingHealthInformationSystemsonThreeLayers: 3LGM²......................................................................................................... 5.3.1 UMLClassDiagramsfortheDescriptionof3LGM²..................... 5.3.2 3LGM²-B......................................................................................... 5.3.3 3LGM²-M........................................................................................ 5.3.4 3LGM²-S......................................................................................... 5.4 OnReferenceModels.................................................................................. 5.5 AReferenceModelfortheDomainLayerofHospitalInformationSystems....
43 43 43 43 45 51 52 55 66 67 68 70
Contents
xxi
5.6 Exercises..................................................................................................... 5.6.1 TypicalImplementationofHospitalFunctions............................... 5.6.2 3LGM²asaMetamodel.................................................................. 5.6.3 Modelingwith3LGM².................................................................... 5.7 Summary.....................................................................................................
71 71 71 72 73
6 ArchitectureofHospitalInformationSystems................................................ 6.1 Introduction................................................................................................. 6.2 DomainLayer:DatatobeProcessedinHospitals...................................... 6.2.1 EntityTypesRelatedtoPatientCare.............................................. 6.2.2 EntityTypesAboutResources........................................................ 6.2.3 EntityTypesRelatedtoAdministration.......................................... 6.2.4 EntityTypesRelatedtoManagement............................................. 6.3 DomainLayer:HospitalFunctions............................................................. 6.3.1 PatientCare..................................................................................... 6.3.2 SupplyandDisposalManagement,Scheduling, andResourceAllocation................................................................. 6.3.3 HospitalAdministration.................................................................. 6.3.4 HospitalManagement..................................................................... 6.3.5 ResearchandEducation.................................................................. 6.3.6 ClinicalDocumentation:AHospitalFunction?.............................. 6.3.7 DomainLayer:Exercises................................................................ 6.3.8 DomainLayer:Summary................................................................ 6.4 LogicalToolLayer:ApplicationComponents............................................ 6.4.1 PatientAdministrationSystem........................................................ 6.4.2 MedicalDocumentationSystem..................................................... 6.4.3 NursingManagementandDocumentationSystem......................... 6.4.4 OutpatientManagementSystem..................................................... 6.4.5 ProviderorPhysicianOrderEntrySystem(POE).......................... 6.4.6 PatientDataManagementSystem(PDMS).................................... 6.4.7 OperationManagementSystem...................................................... 6.4.8 RadiologyInformationSystem....................................................... 6.4.9 PictureArchivingandCommunicationSystem(PACS)................ 6.4.10 LaboratoryInformationSystem...................................................... 6.4.11 EnterpriseResourcePlanningSystem............................................ 6.4.12 DataWarehouseSystem.................................................................. 6.4.13 DocumentArchivingSystem.......................................................... 6.4.14 OtherComputer-BasedApplicationComponents.......................... 6.4.15 ClinicalInformationSystemandElectronicPatientRecord SystemasComprehensiveApplicationComponents..................... 6.4.16 TypicalNon-Computer-BasedApplicationComponents............... 6.5 LogicalToolLayer:IntegrationofApplicationComponents..................... 6.5.1 TaxonomyofArchitecturesattheLogicalToolLayer................... 6.5.2 Integrity........................................................................................... 6.5.3 TypesofIntegration........................................................................
75 75 75 76 77 78 78 79 79 93 96 104 104 106 107 108 110 111 113 115 116 118 120 122 124 125 127 128 129 131 133 134 135 137 138 144 146
xxii
Contents
6.5.4 Standards......................................................................................... 6.5.5 IntegrationTechnologies................................................................. 6.5.6 LogicalToolLayer:Example......................................................... 6.5.7 LogicalToolLayer:Exercises........................................................ 6.5.8 LogicalToolLayer:Summary........................................................ 6.6 PhysicalToolLayer:PhysicalData-ProcessingSystems........................... 6.6.1 ServersandCommunicationNetworks........................................... 6.6.2 Clients............................................................................................. 6.6.3 Storage............................................................................................ 6.6.4 TypicalNon-Computer-BasedPhysicalData-ProcessingSystems 6.6.5 Infrastructure................................................................................... 6.7 PhysicalToolLayer:IntegrationofPhysicalData-ProcessingSystems.... 6.7.1 TaxonomyofArchitecturesatthePhysicalToolLayer.................. 6.7.2 PhysicalIntegration......................................................................... 6.7.3 ComputingCenters......................................................................... 6.7.4 PhysicalToolLayer:Example........................................................ 6.7.5 PhysicalToolLayer:Exercises....................................................... 6.7.6 PhysicalToolLayer:Summary....................................................... 6.8 SummarizingExample................................................................................ 6.8.1 HealthInformationSystemsSupportingClinical BusinessProcesses.......................................................................... 6.9 SummarizingExercises............................................................................... 6.9.1 HospitalFunctionsandProcesses................................................... 6.9.2 ApplicationComponentsandHospitalFunctions.......................... 6.9.3 MultiprofessionalTreatmentTeams................................................ 6.9.4 InformationNeedsofDifferentHealthCareProfessionals............ 6.9.5 HISArchitectures............................................................................ 6.9.6 CommunicationServer................................................................... 6.9.7 AnatomyandPhysiologyofInformationProcessing..................... 6.10 Summary..................................................................................................... 7 SpeciicAspectsforArchitecturesofTransinstitutional HealthInformationSystems.............................................................................. 7.1 Introduction................................................................................................. 7.2 DomainLayer.............................................................................................. 7.2.1 SpeciicAspectsforHospitalFunctions......................................... 7.2.2 AdditionalEnterpriseFunctions..................................................... 7.3 LogicalToolLayer...................................................................................... 7.3.1 IntegrationofApplicationComponents.......................................... 7.3.2 StrategiesforElectronicHealthRecordSystems........................... 7.4 PhysicalToolLayer..................................................................................... 7.5 Examples..................................................................................................... 7.5.1 “GesundheitsnetzTirol(GNT)”:TheTyrolean HealthCareNetwork...................................................................... 7.5.2 VeteransHealthInformationSystemsandTechnology Architecture(VISTA)......................................................................
149 155 161 163 167 168 169 169 170 170 171 172 172 174 175 176 177 177 178 178 181 181 181 182 182 182 182 182 183
185 185 186 186 188 188 188 190 192 193 193 196
Contents
xxiii
7.5.3 TheHypergenesBiomedicalInformationInfrastructure................ 7.5.4 TheNationalHealthInformationSysteminKorea........................ 7.6 Exercises..................................................................................................... 7.6.1 ChallengesofTransinstitutionalHealthInformationSystems....... 7.6.2 StrategiesforTransinstitutionalElectronicHealthRecords........... 7.6.3 TheTerm“ElectronicHealthRecord”............................................ 7.6.4 TransinstitutionalInformationSystemsinOtherSectors............... 7.7 Summary.....................................................................................................
196 197 198 198 198 199 199 199
8 QualityofHealthInformationSystems........................................................... 8.1 Introduction................................................................................................. 8.2 QualityofStructures................................................................................... 8.2.1 QualityofData................................................................................ 8.2.2 QualityofComputer-BasedApplicationComponents andTheirIntegration....................................................................... 8.2.3 QualityofPhysicalDataProcessingSystems................................. 8.2.4 QualityoftheOverallHISArchitecture......................................... 8.2.5 Exercises......................................................................................... 8.2.6 Summary......................................................................................... 8.3 QualityofProcesses.................................................................................... 8.3.1 SingleRecording,MultipleUsabilityofData................................ 8.3.2 NoTranscriptionofData................................................................ 8.3.3 LeannessofInformationProcessingTools..................................... 8.3.4 EficiencyofInformationLogistics................................................ 8.3.5 Patient-CenteredInformationProcessing....................................... 8.3.6 Exercises......................................................................................... 8.3.7 Summary......................................................................................... 8.4 QualityofOutcome..................................................................................... 8.4.1 FulillmentofHospital’sGoals....................................................... 8.4.2 FulillmentoftheExpectationsofDifferentStakeholders............. 8.4.3 FulillmentofInformationManagementLaws............................... 8.4.4 Exercises......................................................................................... 8.4.5 Summary......................................................................................... 8.5 BalanceasaChallengeforInformationManagement................................ 8.5.1 BalanceofHomogeneityandHeterogeneity.................................. 8.5.2 BalanceofComputer-BasedandNon-Computer-BasedTools....... 8.5.3 BalanceofDataSecurityandWorkingProcesses.......................... 8.5.4 BalanceofFunctionalLeannessandFunctionalRedundancy........ 8.5.5 BalanceofDocumentationQualityandDocumentationEfforts.... 8.5.6 Exercises......................................................................................... 8.5.7 Summary......................................................................................... 8.6 EvaluationofHealthInformationSystemsQuality.................................... 8.6.1 TypicalEvaluationPhases.............................................................. 8.6.2 TypicalEvaluationMethods........................................................... 8.6.3 Exercises......................................................................................... 8.6.4 Summary........................................................................................
201 201 202 202 203 205 206 206 207 208 208 208 208 210 210 211 212 212 213 213 215 216 216 216 217 217 218 219 219 220 220 221 221 224 227 228
xxiv
Contents
8.7 SummarizingExamples.............................................................................. 8.7.1 TheBaldrigeHealthCareInformationManagementCriteria........ 8.7.2 InformationManagementStandardsoftheJointCommission....... 8.7.3 TheBabyCareLinkStudy............................................................... 8.7.4 In-DepthApproach:TheFunctionalRedundancyRate.................. 8.8 SummarizingExercises............................................................................... 8.8.1 EvaluationCriteria.......................................................................... 8.8.2 JointCommissionInformationManagementStandards................. 8.9 Summary.....................................................................................................
228 228 229 230 230 235 235 235 235
9 StrategicInformationManagementinHospitals............................................ 9.1 Introduction................................................................................................. 9.2 Strategic,TacticalandOperationalInformationManagement................... 9.2.1 InformationManagement................................................................ 9.2.2 InformationManagementinHospitals........................................... 9.2.3 StrategicInformationManagement................................................ 9.2.4 TacticalInformationManagement.................................................. 9.2.5 OperationalInformationManagement............................................ 9.2.6 RelationshipBetweenITServiceManagement andInformationManagement......................................................... 9.2.7 Example.......................................................................................... 9.2.8 Exercises......................................................................................... 9.2.9 Summary......................................................................................... 9.3 OrganizationalStructuresofInformationManagement.............................. 9.3.1 ChiefInformationOficer............................................................... 9.3.2 InformationManagementDepartment............................................ 9.3.3 Example.......................................................................................... 9.3.4 Exercises......................................................................................... 9.3.5 Summary......................................................................................... 9.4 StrategicPlanning....................................................................................... 9.4.1 Tasks................................................................................................ 9.4.2 Methods........................................................................................... 9.4.3 TheStrategicInformationManagementPlan................................. 9.4.4 Example.......................................................................................... 9.4.5 Exercises......................................................................................... 9.4.6 Summary......................................................................................... 9.5 StrategicMonitoring............................................................................... 9.5.1 Tasks................................................................................................ 9.5.2 Methods........................................................................................... 9.5.3 Examples......................................................................................... 9.5.4 Exercises......................................................................................... 9.5.5 Summary......................................................................................... 9.6 StrategicDirecting...................................................................................... 9.6.1 Tasks................................................................................................ 9.6.2 Methods...........................................................................................
237 237 238 238 241 242 243 244 246 248 248 249 249 249 251 251 252 253 254 254 256 257 261 261 263
263 264 267 268 272 272 273 273 273
Contents
9.6.3 Example.......................................................................................... 9.6.4 Exercise........................................................................................... 9.6.5 Summary......................................................................................... 9.7 LastButNotLeast:Education!................................................................... 9.8 SummarizingExamples.............................................................................. 9.8.1 DeicienciesinInformationManagement....................................... 9.8.2 ComputerNetworkFailures............................................................ 9.8.3 InformationManagementResponsibilities..................................... 9.8.4 SafelyImplementingHealthInformation andConvergingTechnologies......................................................... 9.8.5 IncreasedMortalityAfterImplementationofaComputerized PhysicianOrderEntrySystem........................................................ 9.9 SummarizingExercises............................................................................... 9.9.1 ManagementofOtherInformationSystems................................... 9.9.2 BeginningandEndofInformationManagement........................... 9.9.3 CultivatingHospitalInformationSystems...................................... 9.9.4 HospitalInformationSystemFailure.............................................. 9.9.5 IncreasedMortality......................................................................... 9.9.6 RelevanceofExamples................................................................... 9.9.7 ProblemsofOperationalInformationManagement....................... 9.10 Summary..................................................................................................... 10 StrategicInformationManagementinHealthCareNetworks..................... 10.1 Introduction................................................................................................. 10.2 DescriptionofHealthCareNetworks......................................................... 10.3 OrganizationalStructuresofInformationManagement inHealthCareNetworks............................................................................. 10.3.1 CentralityofInformationManagementinHealthCareNetworks. 10.3.2 IntensityofInformationManagementinHealthCareNetworks... 10.4 TypesofHealthCareNetworks.................................................................. 10.5 Example....................................................................................................... 10.5.1 RegionalHealthInformationOrganizations................................... 10.6 Exercise....................................................................................................... 10.6.1 ThePlötzbergHealthCareNetwork............................................... 10.7 Summary.....................................................................................................
xxv
274 274 274 275 275 275 276 277 278 279 279 279 279 280 280 280 280 280 281 283 283 284 284 284 286 286 287 287 288 288 288
11 FinalRemarks.................................................................................................... 291 Thesaurus................................................................................................................... 293 RecommendedFurtherReadings............................................................................ 327 Index........................................................................................................................... 331
List of Figures
Fig.2.1 Radiologicalconferenceinaradiodiagnosticdepartment.........................
4
Fig.2.2 Theoficeofaseniorphysician.................................................................
6
Fig.2.3 A nexampleofthepatientsummary withinanelectronicpatientrecord.............................................................
7
Fig.2.4 Apaper-basedpatientrecordarchiveasoneinformation storingpartofthehospital’smemoryandnervoussystem........................
8
Fig.2.5 S napshotinaserverroomofahospitalshowing thecomputer-basednervecordsofthehospital’smemory andnervoussystem....................................................................................
9
Fig.2.6 Physicianusingapicturearchiving andcommunicationsystemfordiagnostics................................................
10
Fig.2.7 Amobilecomputeronawardtosupport medicaldocumentationandinformationaccess.........................................
12
Fig.2.8 Astudynurseinanoutpatientunitdealing withamultitudeofpaper-basedforms.......................................................
13
Fig.2.9 Regularclinicalroundbydifferent healthcareprofessionalsonaward............................................................
14
Fig.2.10 Duringawardround:Healthcareprofessionals jointlyusinginformationprocessingtools.................................................
15
Fig.2.11 Prof.L.,HeadoftheDepartmentofPediatrics, workingwithaliteratureserver..................................................................
17
Fig.2.12 TheHeidelbergquintuplets........................................................................
18
xxvii
xxviii
List of Figures
Fig.3.1 Anexampleforformsandfoldersfornursingdocumentation, representingaphysicaldataprocessingsystem.Therulesthat describewhomayusetheseforms,andhowtheyshouldbeused, makeuptheapplicationcomponent...........................................................
28
Fig.3.2 Typicalphysicaldataprocessingsystemsinanoutpatientunit (e.g.,printer,telephone,andnon-computer-basedpatientrecord).............
29
Fig.4.1 Ahealthcareprofessionalaccessingpatientinformation..........................
35
Fig.4.2 Ageneralpractitioneraccessingdocuments ofahospitalinformationsystem................................................................
37
Fig.5.1 AnextractofatechnicalHISmodelwithsomephysicaldata processingsystemsandtheirdatatransmissionlinksofthe hospitalinformationsystemofthePlötzbergMedicalCenter andMedicalSchool(PMC)........................................................................
47
Fig.5.2 AnextractofatechnicalHISmodelwithsomeapplicationcomponents andtheircommunicationlinksofthehospitalinformationsystem ofthePlötzbergMedicalCenterandMedicalSchool(PMC)....................
47
Fig.5.3 AnextractfromtheorganizationalmodelofPlötzbergMedical CenterandMedicalSchool(PMC)............................................................
48
Fig.5.4 Asimpliieddatamodel(UMLclassdiagram),describingthe relationshipsbetweentheentitytypespatient,case,andprocedure, asextractfromthedatamodeloftheHISofthePlötzbergMedical CenterandMedicalSchool(PMC)............................................................
49
Fig.5.5 Exampleofabusinessprocessmodel,basedonaUMLactivity diagram,describingapartoftheadmissionprocessinthe DepartmentofChildandJuvenilePsychiatryatPlötzberg MedicalCenterandMedicalSchool(PMC)..............................................
50
Fig.5.6 AbstractclassesandsubclassesinaUMLclassdiagram..........................
53
Fig.5.7 AssociationsbetweenclassesinaUMLclassdiagram.............................
53
Fig.5.8 AssociationclassesinaUMLclassdiagram.............................................
53
Fig.5.9 Decompositionasonetypeofassociationbetween oneclassanditselfinaUMLclassdiagram..............................................
54
List of Figures
xxix
Fig.5.10 Specializationasanothertypeofassociationbetweenone classanditselfinaUMLclassdiagram.....................................................
54
Fig.5.11 Conceptsofthe3LGM²DomainLayer......................................................
57
Fig.5.12 Exampleofa3LGM²DomainLayer.........................................................
58
Fig.5.13 Conceptsofthe3LGM²logicaltoollayer.Linesdenote interlayerrelationshipsbetweenlogicaltoollayeranddomainlayer........
59
Fig.5.14 Exampleofa3LGM²logicaltoollayer.....................................................
60
Fig.5.15 Conceptsofthe3LGM²physicaltoollayer.Dottedlines denoteinterlayerrelationshipsbetweenlogicaltoollayer andphysicaltoollayer................................................................................
61
Fig.5.16 Onthetop,theconceptofaclustervirtualizingseveralphysical dataprocessingsystemsisillustrated.Atthebottom,thereisone physicaldataprocessingsystemwhichisvirtualizedintoseveral virtualmachines.........................................................................................
62
Fig.5.17 Exampleofa3LGM²physicaltoollayer...................................................
63
Fig.5.18 CompleteUMLdiagramofthe3LGM²-Bmetamodelshowing allconcepts,intra-andinterlayerrelationships..........................................
64
Fig.5.19 Exampleofthree3LGM²layersandtheirinterlayerrelationships............
66
Fig.5.20 Conceptsof3LGM²-M.Thediagramonlycontainsthenew conceptsmessagetypeandcommunicationstandardandtheir relationshipswithotherconcepts...............................................................
67
Fig.5.21 Conceptsof3LGM²-S.Thediagramonlycontainsthenew conceptsservice,serviceclass,invokinginterfaceandproviding interfaceandtheirrelationshipswithotherconcepts.................................
68
Fig.5.22 HospitalfunctionsoftheReferenceModelfortheDomain LayerofHospitalInformationSystems(presenteduntil secondhierarchylevel)...............................................................................
70
Fig.6.1 Apatientbeingadmittedinapatientadmissiondepartment.....................
80
Fig.6.2 Extractofthedomainlayerofthe3LGM²-basedreference modeldescribingtheenterprisefunctionpatientadmission, itssubfunctions,andinterpretedandupdatedentitytypes.........................
80
xxx
List of Figures
Fig.6.3 Typicalorganizationalmedia:amagneticcardandstickers withpatientidentiicationdata...................................................................
82
Fig.6.4 Informingpatients’relativesonaward......................................................
84
Fig.6.5 Extractofthedomainlayerofthe3LGM²-basedreferencemodel describingtheenterprisefunctiondecisionmaking,planning andorganizationofpatienttreatment,itssubfunctions, andinterpretedandupdatedentitytypes....................................................
84
Fig.6.6 Paper-basednursingdocumentationonaward..........................................
85
Fig.6.7 Extractofthedomainlayerofthe3LGM²-basedreferencemodel describingtheenterprisefunctionorderentry,itssubfunctions, andinterpretedandupdatedentitytypes....................................................
86
Fig.6.8 Apaper-basedorderentryformforlaboratorytesting...............................
87
Fig.6.9 Clinicalexaminationconductedbyapediatrician.....................................
88
Fig.6.10 Extractofthedomainlayerofthe3LGM²-basedreference modeldescribingtheenterprisefunctionexecutionofdiagnostic, therapeuticandnursingprocedures,itssubfunctions,andinterpreted andupdatedentitytypes.............................................................................
89
Fig.6.11 Extractofthedomainlayerofthe3LGM²-basedreference modeldescribingtheenterprisefunctioncodingofdiagnosesand procedures,itssubfunctions,andinterpretedandupdated entitytypes.................................................................................................
90
Fig.6.12 Preparingforthedischargeofapatientfromaward................................
91
Fig.6.13 Extractofthedomainlayerofthe3LGM²-basedreference modeldescribingtheenterprisefunctionpatientdischargeandtransfer tootherinstitutions,itssubfunctions,andinterpretedandupdated entitytypes.................................................................................................
92
Fig.6.14 Extractofthedomainlayerofthe3LGM²-basedreferencemodel describingtheenterprisefunctionsupplyanddisposalmanagement, itssubfunctions,andinterpretedandupdatedentitytypes.........................
94
Fig.6.15 Thestockofdrugsonahospitalward........................................................
94
List of Figures
xxxi
Fig.6.16 Extractofthedomainlayerofthe3LGM²-basedreference modeldescribingtheenterprisefunctionschedulingand resourceallocation,itssubfunctions,andinterpretedand updatedentitytypes....................................................................................
95
Fig.6.17 Extractofthedomainlayerofthe3LGM²-basedreferencemodel describingtheenterprisefunctionhumanresourcesmanagement, itssubfunctions,andinterpretedandupdatedentitytypes.........................
96
Fig.6.18 Extractofthedomainlayerofthe3LGM²-basedreferencemodel describingtheenterprisefunctionarchivingofpatientinformation, itssubfunctions,andinterpretedandupdatedentitytypes.........................
97
Fig.6.19 Apaper-basedhospitalarchivewitharobotsystemforstoring andgatheringboxesilledwithpatientrecords..........................................
98
Fig.6.20 Extractofthedomainlayerofthe3LGM²-basedreferencemodel describingtheenterprisefunctionqualitymanagement, itssubfunctions,andinterpretedandupdatedentitytypes.........................
99
Fig.6.21 Extractofthedomainlayerofthe3LGM²-basedreferencemodel describingtheenterprisefunctioncostaccounting,itsinterpreted andupdatedentitytypes............................................................................. 100 Fig.6.22 Extractofthedomainlayerofthe3LGM²-basedreferencemodel describingtheenterprisefunctioncontrollinganditsinterpreted andupdatedentitytypes............................................................................. 101 Fig.6.23 ExcerptfromaDRGscorecard,whichisanimportantpart ofahospital’scontrollingreport(inGerman).Theleftcolumn listsasetofkeyperformanceindicators(KPIs)tobecontrolled (e.g.,“Verweildauer(Mittel)”:patients’averagelengthofstay). Thenextcolumnscontainforsometimeperiodsvaluesas recorded(e.g.,“IstMonat122009”:valuesrecordedin December2009)orasplanned(e.g.,“PlanMonat122009”: valuesplannedforDecember2009)........................................................... 102 Fig.6.24 Extractofthedomainlayerofthe3LGM²-basedreference modeldescribingtheenterprisefunctioninancialaccounting, itsinterpretedandupdatedentitytypes...................................................... 103 Fig.6.25 Extractofthedomainlayerofthe3LGM²-basedreferencemodel describingtheenterprisefunctionfacilitymanagement, itssubfunctions,andinterpretedandupdatedentitytypes......................... 103
xxxii
List of Figures
Fig.6.26 Extractofthedomainlayerofthe3LGM²-basedreferencemodel describingtheenterprisefunctioninformationmanagement, itssubfunctions,andinterpretedandupdatedentitytypes......................... 103 Fig.6.27 Extractofthedomainlayerofthe3LGM²-basedreferencemodel describingtheenterprisefunctionhospitalmanagement, itsinterpretedandupdatedentitytypes...................................................... 105 Fig.6.28 Extractofthedomainlayerofthe3LGM²-basedreferencemodel describingtheenterprisefunctionresearchandeducation, itssubfunctions,andinterpretedandupdatedentitytypes......................... 105 Fig.6.29 Referencemodelofthehospitalfunctionsonthedomainlayer................ 109 Fig.6.30 Screenshotofapatientadministrationsystemshowingthepatient listofadepartmentofneurosurgeryinthebackgroundanda windowforassigninganappointmenttoapatientinthefront.................. 111 Fig.6.31 Ascreenshotofamedicaldocumentationsystemshowingalistof patient-relateddocumentsontheleftandanopeneddischarge letterontheright........................................................................................ 113 Fig.6.32 Screenshotofanursingdocumentationsystem.Ontheleft,it showstheselectednursingpathwaysforthegivenpatient. Ontheright,itshowsthecorrespondingopentasksthatnow havetoperformed....................................................................................... 115 Fig.6.33 Screenshotofanapplicationcomponentforscheduling inanoutpatientunit.................................................................................... 117 Fig.6.34 Screenshotofalaborderentrysystem.Anordersetofelectrolyte hasbeenchosentoseedetails(lowerright).Ifavailable,results canalreadybereviewed(lowerleft).......................................................... 119 Fig.6.35 Screenshotofapatientdatamanagementsystemshowinga patient’svitalparametersandgivendrugsduringaday............................ 120 Fig.6.36 Screenshotofanoperationmanagementsystem,showingthe surgeonsaswellasthepatientsthatareassignedtothem.For eachpatient,theplannedprocedureandthestatusoftheoperation isdisplayed.Thestatuscanalsobedisplayedforeachoperation roomusingatimeline(notshown)............................................................ 122 Fig.6.37 Screenshotofaradiologyinformationsystem(RIS)showing allradiologicdocumentsofonepatient..................................................... 124
List of Figures
xxxiii
Fig.6.38 ScreenshotfromaPictureArchivingandCommunication System(PACS)applicationcomponent,presentingdifferent imagesofapatient..................................................................................... 126 Fig.6.39 Inalaboratoryunit.Thelaboratoryinformationsystem(LIS; runningonthePCinthefront)managestheanalysisof samplesbylaboratorydevices(inthebackground)................................... 127 Fig.6.40 Screenshotofadatawarehousesystem.Thediagramshowsa target/actualcomparisonforthebedoccupationofa universityhospital...................................................................................... 130 Fig.6.41 Screenshotofadocumentarchivingsystem.Besidestext documentslikeindingsfromthelaboratory,operationreports, letters,anamneses,etc.,italsocontainsradiologicalimages..................... 132 Fig.6.42 Thepatientchart......................................................................................... 136 Fig.6.43 Thepaper-basedpatientrecord.................................................................. 136 Fig.6.44 DBnstylewithmultiplecomputer-basedapplicationcomponents, eachwithitsowndatabasesystem,using3LGM2symbols. Thecloudinthecenterindicatesthatsomeasyetunknown meansisneededtolinkthecomponents.................................................... 140 Fig.6.45 (DB1,AC1)architectureusing3LGM²symbols.Thegrayrectangle denotesthecomputer-basedapplicationcomponentthatcontainsadatabase system(denotedbythecylinder)................................................................ 141 Fig.6.46 (DB1,ACn)architecturewithmultiplecomputer-based applicationcomponents,using3LGM²symbols.Onlyone computer-basedapplicationcomponent(inthecenter) containsadatabasesystem......................................................................... 141 Fig.6.47 (DBn,ACn,Vn,CPn)architecturewithmultiplecomputer-based applicationcomponents,using3LGM²symbols,withseveral bidirectionalcommunicationinterfaces.Thisrepresentation isalsocalleda“spaghetti”architecturalstyle............................................ 143 Fig.6.48 (DBn,ACn,Vn,CP1)architecturewithmultiplecomputer-basedapplication componentsconnectedbyaspeciicapplication component,using3LGM²symbols.Thisrepresentationisalso calleda“star”architecturalstyle................................................................ 143
xxxiv
List of Figures
Fig.6.49 Assignmentofindingstoordersandtocases,andofthosecases toaparticularpatient,inaUML-baseddatamodel................................... 145 Fig.6.50 Event-drivencommunicationwithHL7..................................................... 150 Fig.6.51 DBnarchitecturalstylewithmultiplecomputer-basedapplication components,eachwithitsowndatabasesystem,using3LGM symbols.Acommunicationserverlinksthecomponents.......................... 158 Fig.6.52 Anintegratedclinicalinformationsystem(CIS)asanexample fora(DB1,V1)architecture.TheCISoftencontainsthemedical documentationsystem,thenursingmanagementanddocumentation system,theoutpatientmanagementsystem,andthephysician orproviderorderentrysystem(POE)asmodules..................................... 162 Fig.6.53 A(DB1,V1)-styledsubinformationsystematChibaUniversity Hospital,Japan.Thecentralelectronicmedicalrecord(EMR) systemhasadatabasewhichcanalsobeaccessedbyother applicationcomponents.............................................................................. 162 Fig.6.54 AtypicalsituationinGermanUniversityHospitals:theenterpriseresource planning(ERP)systemincludesacomponentforpatientadministration. 163 Fig.6.55 Atanophthalmologyunit(1)..................................................................... 164 Fig.6.56 Atanophthalmologyunit(2)..................................................................... 164 Fig.6.57 Atanophthalmologyunit(3)..................................................................... 165 Fig.6.58 Atanophthalmologyunit(4)..................................................................... 165 Fig.6.59 Awardina“paperless”hospital................................................................ 166 Fig.6.60 Ahealthcareprofessionalaccessingpatientinformation.......................... 170 Fig.6.61 Typicalpaper-basedphysicaldataprocessingsystems.............................. 171 Fig.6.62 Amodernserverroominacomputingcenter ofauniversityhospital............................................................................... 176 Fig.7.1 Patientadmissionintheoficeofageneralpractitioner............................ 187 Fig.7.2 OverviewofsenseArchitecturebasedonseveralIHEIntegration Proiles.Actorsaredepictedasboxesandtransactionsaslines................ 194
List of Figures
xxxv
Fig.8.1 Exampleofatranscription(1).................................................................... 209 Fig.8.2 Exampleofatranscription(2).................................................................... 209 Fig.8.3 Extractfromthebusinessprocess“mealordering”................................... 212 Fig.8.4 Astylishclientatapatientadmissionunit................................................. 218 Fig.8.5 Themajorstepsofanevaluationstudy.Boxescompriseactivities; arrowsintoaboxfromtopareinput;arrowsoutfroma boxareoutput.Feedbackloopsindicatethatearliersteps mayhavetoberedoneorreined............................................................... 222 Fig.8.6 MatrixSUP:rectanglesdenoteenterprisefunctions,rounded rectanglesdenoteapplicationcomponents,andconnectinglines illustratea“1”intherespectivepositionofthematrix,thatis, thatacertainenterprisefunctionissupportedbyacertainapplication component.Forexample,enterprisefunctionEdecisionmaking canbesupportedbyapplicationcomponent5decisionsupport systemor7pathologyinformationsystemalternatively............................ 232 Fig.9.1 Relationshipbetweenplanning,directing,andmonitoringduring strategic,tactical,andoperationalinformationmanagement. Forexplanationseeparagraphbefore........................................................ 240 Fig.9.2 Three-dimensionalclassiicationofinformation managementactivities................................................................................ 240 Fig.9.3 Strategic,tactical,andoperationalinformationmanagement inhospitals,HISoperation,andtheirrelationships.................................... 241 Fig.9.4 Typicalphasesoftacticalinformationmanagementprojects.................... 243 Fig.9.5 Monitoringoftheserverofahospitalinformationsystem........................ 245 Fig.9.6 Animmediatesupportcenterforthird-levelsupportofavendor.............. 247 Fig.9.7 Aninformationmanagementboardmeetingatauniversity hospital.Participantsinthismeetingare(fromtheleft):the directorofprocurement,thechairofthestaffcouncil,the assistantoftheinformationmanagementdepartment’sdirector, thedirectoroftheinformationmanagementdepartment,asenior physicianaschairoftheboard,amedicalinformaticsprofessor asvice-chairoftheboard,adirectorofamedicalresearch departmentasrepresentativeofthemedicalfaculty andmedicalschool,thedirectorofnursing.Thedirectorof inanceandvice-directorofadministrationtookthephoto........................ 250
xxxvi
List of Figures
Fig.9.8 OrganizationofinformationmanagementatthePlötzberg MedicalCenterandMedicalSchool(PMC).............................................. 252 Fig.9.9 Strategicinformationmanagementplanningofhospitals.Astrategic informationmanagementplangivesdirectivesfortheconstruction anddevelopmentofahospitalinformationsystem.Itdescribes therecentandtheintendedhospitalinformationsystem’s architecture.(Detailsareexplainedinthefollowingsections).................. 258 Fig.10.1 Typesofhealthcarenetworks.................................................................... 287
Figure Credits
•
ASpecialistPracticeinBammental,Germany:Figure6.9,Figure7.1
•
ASpecialistPracticeinInnsbruck,Austria:Figure4.2
•
BraunschweigMedicalCenter,Braunschweig,Germany:Figure6.41
•
CernerCorporation:Figure2.3,Figure6.33,Figure6.34,Figure6.36
•
CernerImmediateSupportServices,KansasCity,Missouri,USA:Figure9.6
•
ChibaUniversityHospital,Chiba,Japan:Figure6.19
•
DiakonissehjemmetsSykehusHaraldsplass,Bergen,Norway:Figure8.2
•
GermanCancerResearchCenter,Heidelberg,Germany:Figure8.1
•
I THicoservetechnologyforhealthcareGmbH,Innsbruck,Austria: Figure2.6,Figure6.38
•
LDSHospital,SaltLakeCity,Utah,USA:Figure2.9
•
Oneoftheauthor’sformerhomeofice,Meckesheim,Germany:Figure6.61
•
elayangHospital,KualaLumpur,Malaysia:Figure2.7,Figure6.12, S Figure6.59,Figure9.5
•
UniversityMedicalCenterErlangen,Germany:Figure4.1,Figure6.6
•
niversityMedicalCenterHeidelberg,Germany:Figure2.2,Figure2.4, U Figure2.11,Figure2.12,Figure6.1,Figure6.3,Figure6.15,Figure6.55, Figure6.56
xxxvii
xxxviii
Figure Credits
•
niversityMedicalCenterInnsbruck,Austria:Figure2.1,Figure2.7,Figure2.10, U Figure3.1,Figure3.2,Figure6.32,Figure6.39,Figure6.43,Figure6.57, Figure6.58
•
niversityMedicalCenterLeipzig,Germany:Figure2.5,Figure2.8,Figure6.8, U Figure6.23,Figure6.30,Figure6.31,Figure6.35,Figure6.37,Figure6.40, Figure6.42,Figure6.62,Figure9.7
•
WeinbergCancerCenterJohnsHopkins,Baltimore,Maryland,USA:Figure6.4
List of Tables
Table2.1 ExampleofSimpson’sparadox–SuccessratesofNovumandVerum treatmentsforpatientswithdiagnosisD,treatedduringtheyearsdatthe PlötzbergMedicalCenterandMedicalSchool(PMC)........................... 19 Table5.1 AnextractfromthefunctionalHISmodel,describinghospital functionsrelevantforpatientcareandhospitaladministration atthePlötzbergMedicalCenterandMedicalSchool(PMC).................
46
Table6.1 Typicalsupportedhospitalfunctionsandrelated featuresofthepatientadministrationsystem.......................................... 112 Table6.2 Typicalsupportedhospitalfunctionsandrelated featuresofthemedicaldocumentationsystem........................................ 114 Table6.3 Typicalsupportedhospitalfunctionsandrelated featuresofthenursingmanagementanddocumentationsystem............ 116 Table6.4 Typicalsupportedhospitalfunctionsandrelated featuresoftheoutpatientmanagementsystem........................................ 118 Table6.5 Typicalsupportedhospitalfunctionsandrelated featuresoftheprovider/physicianorderentry(POE)system................. 119 Table6.6 Typicalsupportedhospitalfunctionsandrelated featuresofthepatientdatamanagementsystem..................................... 121 Table6.7 Typicalsupportedhospitalfunctionsandrelated featuresoftheoperationmanagementsystem......................................... 123 Table6.8 Typicalsupportedhospitalfunctionsandrelated featuresofthe“radiologyinformationsystem”(RIS)............................ 125 Table6.9 Typicalsupportedhospitalfunctionsandrelated featuresofthe“radiologyinformationsystem”(RIS)............................ 126 xxxix
xl
List of Tables
Table6.10 Typicalsupportedhospitalfunctionsandrelated featuresofthelaboratoryinformationsystem(LIS)............................... 128 Table6.11 Typicalsupportedhospitalfunctionsandrelated featuresoftheenterpriseresourceplanningsystem............................... 129 Table6.12 Typicalsupportedhospitalfunctionsandrelated featuresofthedatawarehousesystem.................................................... 131 Table6.13 Typicalsupportedhospitalfunctionsandrelated featuresofthedocumentarchivingsystem.............................................. 133 Table6.14 Furtherspeciicapplicationcomponentsinhospital informationsystems(HIS)...................................................................... 134 Table8.1 ThematrixSUPforEFandAC.Thematrix isillustratedinFig.8.6............................................................................ 231 Table8.2 isupp......................................................................................................... 232 min Table8.3 Vector ISUP ......................................................................................... 235 Table9.1 Dimensionstobeconsideredforoperationalinformation managementofthecomputer-basedpartof hospitalinformationsystems................................................................... 246 Table9.2 Structureofthestrategicinformationmanagementplan (2010–2015)ofthePlötzbergMedicalCenterand MedicalSchool(PMC)............................................................................ 262 Table9.3 ExtractfromthePMCHISbenchmarkingreport2010 (KPI=keyperformanceindicator).......................................................... 269 Table9.4 ExamplesofCCHITfunctionalcriteriaformedicationordering........... 271 Table10.1 Attributesforthedescriptionofhealthcarenetworks............................ 285
About the Authors
AlfredWinter AlfredWinterisProfessorforMedicalInformaticsattheInstituteforMedicalInformatics, Statistics,andEpidemiologyoftheUniversityofLeipzig,Germany. HestudiedinformaticsattheTechnicalUniversityinAachen,Germany,andreceived hisPh.D.andalicenseforlecturing(German“Habilitation”)formedicalinformaticsfrom theFacultyofTheoreticalMedicineattheUniversityofHeidelberg. His research focuses on methods and modeling tools for the management of health informationsystems.HeteachesinformationmanagementinhealthcareinamedicalinformaticsmasterscourseatLeipzigUniversityandinmasterprogramsforhealthcaremanagement and for health information management at the private Dresden International UniversityinDresden.Heworksasaconsultantandisresponsibleforcoordinatedstrategic information management at Leipzig University Hospital and Leipzig University MedicalFaculty.HeismemberoftheboardoftheGermanprofessionalassociationof medical informaticians (BVMI), member of the board of the joint medical informatics divisionoftheGermanAssociationofMedicalInformatics,BiometryandEpidemiology (GMDS)andtheGermanAssociationofInformatics(GI)andchairoftheirworkinggroup “Methodsandtoolsforthemanagementofhospitalinformationsystems.”
ReinholdHaux ReinholdHauxisProfessorforMedicalInformaticsandDirectorofthePeterL.Reichertz Institute for Medical Informatics of the University of Braunschweig – Institute of TechnologyandofHannoverMedicalSchool,Germany. HestudiedMedicalInformaticsattheUniversityofHeidelberg/UniversityofApplied SciencesHeilbronn,Germany,wherehegraduatedasM.Sc.(German“Diplom”)in1978. HereceivedaPh.D.fromtheFacultyforTheoreticalMedicine,UniversityofUlm,in1983 andaPostdoctoralLectureQualiication(German“Habilitation”)forMedicalInformatics andStatisticsfromtheMedicalFacultyofRWTHAachenUniversityin1987. Hiscurrentresearchieldsarehealthinformationsystemsandmanagementandhealthenablingtechnologies.Sinceitsstartin2001,theinternationalFrank-vanSwieten-Lectures on Strategic Information Management in Hospitals are part of his teaching activities. ReinholdHauxwasandischairpersonormemberofITstrategyboardsofvarioushospitalsinGermanyandAustria,amongothersoftheHeidelberg(1989–2001)andErlangen xli
xlii
About the Authors
University Medical Centers (1999–2001), of TILAK (2002–2004), of Braunschweig MedicalCenter(since2005),andofHannoverMedicalSchool(since2007).HeisEditor of Methods of Information in Medicine and, for the term 2007–2010, President of the InternationalMedicalInformaticsAssociation(IMIA).
ElskeAmmenwerth ElskeAmmenwerthisprofessorforhealthinformaticsandheadoftheInstituteforHealth Information Systems at the University for Health Sciences, Medical Informatics and Technology(UMIT)inHallinTyrol,Austria. ShestudiedmedicalinformaticsattheUniversityofHeidelberg/UniversityofApplied SciencesHeilbronn,Germany.Between1997and2001,sheworkedasaresearchassistant at the Institute for Medical Biometry and Informatics at the University of Heidelberg, Germany.InHeidelberg,herworkcomprisedtheevaluationofhospitalinformationsystems,nursinginformatics,andrequirementsanalysisforhospitalinformationsystems.In 2000,shereceivedherPh.D.fromtheMedicalFacultyoftheUniversityofHeidelbergfor herworkonrequirementsmodeling. At UMIT, she works on the electronic health record, electronic medication systems, evaluationmethods,nursinginformatics,andonprocessmodelingandoptimization.She regularlylecturesonhospitalinformationsystemsandtheirmanagement,onevaluationof informationsystems,andonprojectmanagement. ElskeAmmenwerthisAustrianrepresentativewithinEFMI,theEuropeanFederationfor MedicalInformatics,andwithinIMIA,theInternationalMedicalInformaticsAssociation. SheisalsoheadingtheEFMIworkinggroup“AssessmentofHealthInformationSystems.”
BirgitBrigl BirgitBrigliscurrentlyresponsiblefortheservicemanagementoftheITfullserviceprovider oftheGermanFederalMinistryofFinance.Earlier,sheworkedasaresearchscientistatthe InstituteofMedicalInformatics,Statistics,andEpidemiologyoftheUniversityofLeipzig, Germany.Herresearchinterestscoverthemanagementofhospitalinformationsystems,with aspecialfocusonhospitalinformationsystemsmodelingandITservicemanagement. ShestudiedmedicalinformaticsattheUniversityofHeidelberg/UniversityofApplied Sciences,Heilbronn,Germany.Between1992and1998sheworkedasaresearchassistant at the Institute for Medical Biometry and Informatics at the University of Heidelberg, focusingontheintegrationofdecisionsupportsystemsinhospitalinformationsystems.In 1997,shereceivedherPh.D.fromtheFacultyofTheoreticalMedicineattheUniversityof Heidelbergforherworkonknowledgeacquisition.
NilsHellrung NilsHellrungisseniorresearcheratthePeterL.ReichertzInstituteforMedicalInformatics of the University of Braunschweig – Institute of Technology and of Hannover Medical School,Germany.
About the Authors
xliii
He studied industrial engineering at the Berlin – Institute of Technology, Germany, wherehegraduatedasadegreedengineerin2003andMedicalInformaticsattheHealth andLifeScienceUniversityHall/Tyrol,Austria,wherehegraduatedasM.Sc.in2005.He receivedaPh.D.fromPeterL.ReichertzInstituteforMedicalInformaticsoftheUniversity ofBraunschweig–InstituteofTechnologyandofHannoverMedicalSchoolin2009. Hiscurrentresearchieldsarehealthinformationsystemsandmanagementwithafocus ontransinstitutionalinformationmanagementandhealthcarenetworks.Sinceitsstartin 2008heiscoordinatorforthepubliclyfundedprojectPAGE(PlatformforIntegrationof HealthEnablingTechnologiesinHealthCareNetworks).
FranziskaJahn FranziskaJahnisresearchassistantattheInstituteforMedicalInformatics,Statistics andEpidemiologyoftheUniversityofLeipzig,Germany. ShestudiedcomputersciencewithmedicalinformaticsasmainsubjectattheUniversity ofLeipzigandgraduatedin2008.Herresearchfocusesonhospitalinformationsystem architecturesandtheirmanagement.ShecurrentlyworksonherPh.D.thesisaboutprocess benchmarkingofhospitalinformationsystems.Sheteachesmedicalinformaticsstudents andcoordinatestheireducationattheUniversityofLeipzig. SheorganizesannualPh.D.students’symposiaoftheGermanAssociationofMedical Informatics,BiometryandEpidemiology(GMDS).
Fromlefttoright:F.J.,R.H.,B.B.,N.H.,E.A.,A.W.
Introduction
1
What is a health information system? The literature deines health information systems (HIS)inmanydifferentwaysandpresentsvariousviews.Somearticlesfocusontheorganizationalaspectsofinformationprocessing,whileothersfocusonthetechnologyused. Tobeginwith,weunderstandahealthinformationsystemastheinformationprocessing andinformationstoringsubsystemofahealthcareorganization,whichmaybeasingle institution,forexample,ahospital,oragroupofhealthcareinstitutionslikeahealthcare network. Thisbookdiscussesthesigniicanceofinformationprocessinginhealthcare,withan emphasisoninformationprocessinginhospitals,theprogressininformationandcommunicationtechnology,andtheimportanceofsystematicinformationmanagement.Nearly allpeopleworkinginhealthcareinstitutionshaveanenormousdemandforinformation, whichhastobefulilledinordertoachievehigh-qualityandeficientpatientcare.For example,physiciansandnursesneedinformationconcerningthehealthstatusofpatients fromdifferentdepartmentsoftheirhospital.Theyalsoneedcurrentmedicalknowledgeas abasisfortheirclinicaldecisions.Inaddition,themanagementofahospitalneedsup-todateinformationaboutthehospital’scostsandservices.Ofcourse,thequalityofinformation processing is important for the competitiveness of a hospital. Consequently, this systemofinformationprocessingcanberegardedasthememoryandnervoussystemof therespectivehealthcareinstitution. Thesubjectofinformationprocessingisquitecomplex.Nearlyallgroupsandallareas inahealthcareinstitutiondependonthequalityofinformationprocessing.Theamountof informationprocessingistremendous.Additionally,theinformationneedsofthedifferent groupsareoftenbasedonthesamedata.Therefore,integratedinformationprocessingis necessary. If health information systems are not systematically managed and operated, theytendtodevelopchaotically.This,inturn,leadstonegativeconsequencessuchaslow dataquality,resultinginlowqualityofpatientcareandincreasingcosts.SystematicinformationmanagementcanhelptopreventsuchHISfailuresandcontributetoahigh-quality andeficientpatientcare. Well-educated specialists in health informatics/medical informatics, with the knowledgeandskillstosystematicallymanageandoperatehealthinformationsystemsaretherefore needed to appropriately and responsibly apply information and communication technologytothecomplexinformationprocessingenvironmentofhealthcaresettings. Thisbookdiscussesthetypicalarchitecturesofhealthinformationsystemsandtheir systematicstrategicmanagement.Alotofexampleswillshowhowcertainmethodsand A.Winteretal.,HealthInformationSystems, DOI:10.1007/978-1-84996-441-8_1,©Springer-VerlagLondonLimited2011
1
2
1
Introduction
toolscanbeusedtodescribeandassessarchitecturesofhealthinformationsystemsandto supportthevariousinformationmanagementtasksinanintegratedfashion. This textbook addresses you as a health care and health/medical informatics professionalaswellasastudentinhealth/medicalinformaticsandhealthinformationmanagement. It should be regarded as an introduction to this complex subject. For a deeper understanding,youwillneedadditionalknowledgeand,foremost,practiceinthisield. Ifyouarenotfamiliarwithpatientcareandmedicalresearch,youcanindanintroductorychaptertohealthinstitutionsandtheirrespectiveinformationprocessingtasks.Ifyou arenotfamiliarwithinformationsystems,youcanindanintroductorychaptertoinformationsystembasics. Wewanttoprovideyouwithaterminologyabouthealthinformationsystemswhichis ascompleteandsoundaspossible.Tosupportthis,wecompiledacatalogofthemost importanttermsasathesaurusattheendofthebook.Thetermscatalogedareunderlined inthetext,wheretheyareexplained.Ifyouindtermswhichareprintedinitalics,these termswillreferto • actionstobeundertakeninhealthcareinordertoprocessdataandinformationandto therebycontributetothemissionofhealthcare(i.e.,hospitalfunctions). • computer-basedtoolstosupportprofessionalsinhealthcaretoundertaketheseactions (i.e.,applicationcomponents). YouwillindexplanationsoftherespectivetermsinSects.6.3,6.4. Afterreadingthisbook,youshouldbeabletoanswerthefollowingquestions: • Whyissystematicinformationprocessinginhealthcareinstitutionsimportant? • Whatareappropriatemodelsforhealthinformationsystems? • Howdohealthinformationsystemslooklikeandwhatarchitecturesareappropriate? • Howcanweassessthequalityofhealthinformationsystems? • Howcanwestrategicallymanagehealthinformationsystems? Intheend,weareconidentthatyouwillbeabletoanswerthequestion“Howcangood informationsystemsbedesignedandmaintained?” Ifyouarealecturer,wewouldliketosupportyoubysomesupplementarymaterials basedonthebookwhichcanbedownloadedfromhttp://www.3lgm2.de/en/Publications/ Materials/HealthInformationSystems.jsp.”
Health Institutions and Information Processing
2
2.1 Introduction Healthinformationsystemsstronglyinluencequalityandeficiencyofhealthcare,andtechnical progressoffersadvancedopportunitiestosupporthealthcare.Inthischapter,wewilldiscussthe interrelationbetweenhealthinformationsystemsononesideandhealthcareontheotherside. Afterreadingthischapter,youshouldbeabletoanswerthefollowingquestions: • Whatisthesignificanceofinformationsystemsforhealthcare? • Howdoestechnicalprogressaffecthealthcare? • Whyissystematicinformationmanagementimportant?
2.2 Significance of Information Processing in Health care 2.2.1 Information Processing as Quality Factor Decisionsofhealthcareprofessionalsarebasedonvastamountsofinformationaboutthe patient’shealthstate.Itisessentialforthequalityofpatientcareandforthequalityof hospitalmanagementtofulilltheseinformationneeds. Whenapatientisadmittedtoahospital,aphysicianornurseirstneedsinformation aboutthereasonforpatientadmissionandthepatienthistory.Later,sheorheneedsresults fromservicessuchaslaboratoryandradiology(Fig.2.1),whicharesomeofthemostfrequent diagnostic procedures. In general, clinical patient-related information should be availableontime,anditshouldbeup-to-dateandvalid.Forexample,therecentlaboratory reportshouldbeavailableonthewardwithin2haftertherequest.Ifthisisnotthecase,if itcomestoolate,orisoldorevenwrong,qualityofcareandpatientsafetyisatrisk.An incorrectlaboratoryreportmayleadtoerroneousandevenharmfultreatmentdecisions. Additionally,ifexaminationshavetoberepeatedorlostindingshavetobesearchedfor, the costs of health care may increase. Information should be documented adequately, enabling health care professionals to access the information needed and to make sound decisions. A.Winteretal.,HealthInformationSystems, DOI:10.1007/978-1-84996-441-8_2,©Springer-VerlagLondonLimited2011
3
4
2 Health Institutions and Information Processing
Fig. 2.1 Radiologicalconferenceinaradiodiagnosticdepartment
Peopleworkinginhospitaladministrationalsomustbewellinformedinordertocarry outtheirtasks.Theyshouldbeinformedtimelyandreceivecurrentinformation.Iftheinformationlowistooslow,billsarewrittendaysorevenweeksafterthepatient’sdischarge.If informationismissing,payableservicescannotbebilled,andthehospital’sincomewillbe reduced.Forexample,undercertaincircumstances,theamountpayablebythehealthinsuranceisreduced,iftheinvoicefromthehospitalarrivesafteracertaindeadline. Hospitalmanagementalsohasanenormousinformationneed.Up-to-dateinformation aboutcostsandproceedsarenecessaryasabasisforcontrollingtheenterprise.Information aboutthequalityofpatientcareisequallyimportant,forexample,abouttheformandseverityofpatients’illnesses,aboutnosocomialinfections,oraboutcomplicationratesoftherapeuticprocedures.Ifthisinformationisnotaccurate,notontime,orincomplete,thehospital’s workcannotbecontrolledadequately,increasingtherisksofmanagementerrors. Thus,informationprocessingisanimportantqualityfactorinhealthcareand,inparticular,inhospitals.
2.2.2 Information Processing as Cost Factor In 2007, member states of the Organization for Economic Cooperation and Development (OECD)spentbetween6and15%oftheirtotalgrossdomesticproduct(GDP)onhealthcare.1 OrganizationforEconomicCo-operationandDevelopment(OECD).OECDHealthData2008. StatisticsandIndicatorsfor30countries.http://www.oecd.org
1
2.2 Significance of Information Processing in Health care
5
In 2006, the costs for the approximately 2,100 German hospitals with their 510,000 beds amountedto€60billion;1.1millionpeopleworkedintheseorganizationsinGermany,and17 millioninpatientsweretreated.2IntheUSA,hospitalspendingwasnearly$600billion.The overallUSnationalhealthexpenditurereached$2.2trillionin2007,andaccountedfor16.2% oftheGrossDomesticProduct.3 Arelevantpercentageofthosecostsisspentoninformationprocessing.However,the totalpercentageofinformationprocessingcanonlybeestimated.Alreadyinthe1960s, studies observed that 25% of a hospital’s costs are due to (computer-based and non- computer-based)informationprocessing.4However,suchanestimatedependsonthedeinition of information processing. In general, the investment costs (including purchase, adaptation, introduction, and training) must be distinguished from the operating costs (includingcontinuedmaintenanceandsupportaswellasstaff),andthecostsforcomputerbasedfromthecostsfornon-computer-basedinformationprocessing(whichstillareoften muchhigherinhospitals). Lookingatcomputer-basedinformationprocessing,theannualbudgetthathealthcare institutionsspendoninformationandcommunicationtechnology(ICT)(includingcomputersystems,computernetworks,andcomputer-basedapplicationcomponents)wasin 2006between2.5%and3.3%ofthetotalhospitaloperatingexpense,dependingonthe numberofbeds.5Inmanyhospitals,theannualbudgetisevenlower.Mosthospitalchief informationoficers(CIOs)expectanincreasingbudget.6 Whenlookingatnon-computer-basedinformationprocessing(seeFig.2.2,forexample),thenumbersbecomeincreasinglyvague.However,wecanexpectthat,forexample, theannualoperatingcosts(includingpersonnelcosts)foranon-computer-basedarchive, storingabout300,000–400,000newpatientsrecordseachyear,mayeasilyamounttomore than€500,000.Atypical,standardized,machine-readableform,includingtwocarboncopies(aradiologyorder,forexample)costsapproximately€0.50.Atypicalinpatientrecord atauniversityhospitalconsistsofabout40documents. Basedonthesefigures,itbecomesapparentthatinformationprocessinginhealth careisanimportantcostfactorandconsiderablysignificantforanationaleconomy. Itisclearthat,ontheonehand,efficientinformationprocessingoffersvastpotential forcostreductions.Ontheotherhand,inefficientinformationprocessingleadstocost increases.
ederalStatisticalOfice.StatisticalYearbook2008fortheFederalRepublicofGermany.http:// F www.destatis.de 3 USDepartmentforHealthandHumanServices.NationalHealthExpenditureData2007.http:// www.cms.hhs.gov/NationalHealthExpendData 4 JydstrupR,GrossM.Costofinformationhandlinginhospitals.HealthServicesResearch1966; 1:235–71. 5 HealthcareInformationandManagementSystemsSociety(HIMSS):2007AnnualReportofthe USHospitalITMarket.http://marketplace.himss.org 6 Healthcare Information and Management Systems Society (HIMSS). The 19th Annual 2008 HIMSSLeadershipSurveyCIOResultsFinalReport.2000.http://www.himss.org 2
6
2 Health Institutions and Information Processing
Fig. 2.2 Theoficeofasenior physician
2.2.3 Information as Productivity Factor In the nineteenth century, many societies were characterized by rising industry and industrial production. By the second half of the twentieth century, the idea of communicating and processing data by means of computers and computer networks was already emerging. Today we speak of the twenty-irst century as the century of information technology, or of an “information society.” Informatics and information andcommunicationtechnology(ICT)areplayingakeyrole.Information,boundtoa mediumofmatterorenergy,butlargelyindependentofplaceandtime,shallbemade availabletopeopleatanytimeandinanyplaceimaginable.Informationshallindits waytopeople,notviceversa. Today, information belongs to the most important productivity factors of a hospital. Productivityisdeinedasaratioofoutputandinput.Allresourceslikepersonnel,medical devices,etc.arepartoftheinput.Therefore,fromaneconomicpointofview,productivity ofahospitalmightbedeinedastheratioofthenumberofcasesandfull-timeemployees. If,however,outputisconsideredasqualityofpatientcare,itismuchmoredificultto calculateproductivity.Therefore,alotofreliableclinicaldataareneeded.Forhigh-quality patientcareandeconomicmanagementofahospital,itisessentialthatthehospitalinformationsystemcanmakecorrectinformationfullyavailableontime.Thisisalsoincreasinglyimportantforthecompetitivenessofhospitals.
2.2.4 Holistic View of the Patient Informationprocessinginahospitalshouldofferacomprehensive,holisticviewof the patient and of the hospital. “Holistic” in this context means to have a complete pictureofthecareofapatientavailable,independentofthehealthcareinstitutions
2.2 Significance of Information Processing in Health care
7
Fig. 2.3 Anexampleofthepatientsummarywithinanelectronicpatientrecord
andhospitaldepartmentsinwhichthepatienthasbeenorwillbetreated.Thisholistic viewonthepatientcanreduceundesiredconsequencesofhighlyspecializedmedicine with various departments and health care professionals involved in patient care. Despitehighlydifferentiateddiagnosticsandtherapy,andthemultitudeofpeopleand areasinahospital,adequateinformationprocessing(andagoodhospitalinformation system)canhelptomakeinformationaboutapatientavailablecompletely(Fig.2.3). Asspecializationinmedicineandhealthcareincreases,sodoesthefragmentationof information,whichmakescombininginformationintosuchaholisticviewmoreand morenecessary.However,itmustbeclearlyensuredthatonlyauthorizedpersonnel canaccesspatientdataanddataaboutthehospitalasanenterprise.
2.2.5 Hospital Information System as Memory and Nervous System Figurativelyspeaking,ahospitalinformationsystemmightberegardedasthememory andthenervoussystemofahospital.Ahospitalinformationsystem,comprisingthe informationprocessingandstorageinahospital(Figs.2.4and2.5),toacertainextent canbecomparedtotheinformationprocessingofahumanbeing.Thehospitalinformation system also receives, transmits, processes, stores, and presents information. Thequalityofahospitalinformationsystemisessentialforahospital,againiguratively,inordertobeabletoadequatelyrecognizeandstorefacts,torememberthem, andtoactonthem.
8
2 Health Institutions and Information Processing
Fig. 2.4 Apaper-basedpatientrecord archiveasoneinformationstoring partofthehospital’smemoryand nervoussystem
2.3 Progress in Information and Communication Technology 2.3.1 Impact on the Quality of Health Care Progress in information and communication technology (ICT) changes societies and affectsthecostsandqualityofinformationprocessinginhealthcare.Itisthususefulto takealookattheworldofinformationandcommunicationtechnology. Tremendousimprovementsindiagnosticshavebeenmadeavailablebymoderntechnology,forexample,intheareaofmedicalsignalandimageprocessing.Magneticresonance imaging and computer tomography, for example, would not have been possible withoutimprovementsininformationprocessingandinformationmethodologyandwithoutmoderninformationandcommunicationtechnology(seeFig.2.6).Improveddiagnosticsthenleadtoanimprovementintherapy.Sometherapies,forexample,inneurosurgery orradiotherapy,arepossiblemainlyduetotheprogressinICT.Thesameistrueintheield of medical biotechnology: the development of new drug agents, research in molecular principlesofdiseases,andtheresultingnewpatient-speciictherapeuticoptionenablea bettertreatmentofpatients. Nowadays,clinicalresearchcantoanincreasingdegreebeconductedwithsuccess,and be internationally competitive, only if carried out on an interdisciplinary, often also
2.3 Progress in Information and Communication Technology
9
Fig. 2.5 Snapshotinaserver roomofahospitalshowing thecomputer-basednerve cordsofthehospital’s memoryandnervous system
inter-regionalorinternational,andcollaborativebasis.Thiscollaborationhasbeenandis fostered by integrated systems of information technology. The translation of medical researchoutcomesintonewtherapiesneedstightinformationexchangefrom“bench”to “bed”,thatis,fromresearchtopatientcare.Viceversainformationaboutexperienceswith therapiesiscollectedbycomputer-basedtoolswithinclinicaltrialsandisresenttoand carefullyanalyzedatresearchinstitutions. ImportantprogressduetoimprovementsinmodernICTcanalsobeobservedininformation systems of health care organizations. The role of computer-based information systems,togetherwithclinicaldocumentationandknowledge-baseddecisionsupportsystems,canhardlybeoverestimatedinrespecttothequalityofhealthcare,asthevolumeof dataavailabletodayismuchgreaterthanitwasafewyearsago.
10
2 Health Institutions and Information Processing
Fig. 2.6 Physicianusingapicturearchivingandcommunicationsystemfordiagnostics
2.3.2 Impact on Economics Formanycountries,thevisionofan“informationsociety”hasbecomeareality.Nearly everymoderneconomicbranchisshapedbyinformationprocessingandinformationand communicationtechnology. Theworldwideinformationandcommunicationtechnologymarketvolumeisestimated atnearly€2.5trillionin2009withagrowthrateofabout5%peryear.7Germany’sexpected total annual turnover on information and communication technology was approximately €146billionin2008.Generally,halfofthismoneyisspentoninformationtechnology(data processinganddatacommunicationequipment,software,andrelatedservices)andtheother halfoncommunicationtechnology(telecommunicationequipmentandrelatedservices).8 ICThasbecomeamajorfactorforqualityandeficiencyofhealthcareworldwide. ICTinhealthcarealsoemergedtoaleadingindustrybranch.Thepercentageofhealth careICTontheworldwideICTmarketisdificulttoestimate.Thefollowingnumbers mayindicatethesigniicanceofICTinhealthcare:IntheU.S.theestimatedtotalexpenditures of ICT equipment and software in health care were about $21 billion in 2007, whichis8.1%ofthetotalUSICTexpenditures.9ReportsfromtheEuropeanUnion(EU)
EuropeanInformationTechnologyObservatory(EITO).ICTMarketOverview.http://www.eito. com/reposi/FreeDataSheets/ICT-MarketOverview-world 8 Bitkom.http://www.bitkom.de 9 US Census Bureau. Information & Communication Technology Survey. http://www.census. gov/csd/ict 7
2.3 Progress in Information and Communication Technology
11
statethattheeHealthindustryintheEU(deinedascomprisingclinicalinformationsystems, telemedicine and homecare, and regional networks) was estimated “to be worth closeto€21billionin2006”andthattheglobaleHealthindustry“hasthepotentialtobe thethirdlargestindustryinthehealthsectorwithaglobalturnoverof€50–60billion.”10 ManycountriesestablishedprogramstoforceinformationandcommunicationtechnologyespeciallyintheeHealthsegmentwithinvestmentvolumesbetween$50millionand $11.5billion.11 Onemighthavedoubtsaboutthevalidityoftheseratherroughnumbers.However,they allexemplifythefollowing:Thereisasigniicantandincreasingeconomicrelevancenot onlyforinformationandcommunicationtechnologyingeneralbutalsoinhealthcare.
2.3.3 Changing Health Care Nowoncemore,whatchangesinhealthcaredoweexpectthroughinformationandcommunicationtechnology? Thedevelopmentsmentionedwillprobablycontinueintothenextdecadeatleastatthe samerateasgiventoday.Thedevelopmentofinformationandcommunicationtechnology willcontinuetohaveaconsiderableeffectonoursocietiesingeneralandonourhealth caresystemsinparticular.12 The use of computer-based tools in health care is dramatically increasing, and new technologiessuchasmobiledevicesandmultifunctionalbedsideterminalswillproliferate.Thosemobileinformationprocessingtoolsofferbothcommunicationandinformation processingfeatures.Wirelessnetworksarestandardinmanyhospitals.Computer-based training systems strongly support eficient learning for health care professionals. Documentationeffortsarecontinuouslyrisingandleadtomoresophisticatedcomputerbaseddocumentationtools(seeFig.2.7).Decisionsupporttools,forexample,inthecontext of drug prescription, support high-quality care. Communication is increasingly supportedbyelectronicmeans.Theglobalizationofprovidinghealthcareandthecooperationofhealthcareprofessionalsisincreasing,andpatientsandhealthcareprofessionals seekreliablehealthinformationontheInternet.Largehealthdatabasesareavailablefor everyone at his or her work place and global companies offer personal health records worldwideforeveryone.Providinghigh-qualityandeficienthealthcarewillcontinueto bestronglycorrelatedwithhigh-qualityinformationandcommunicationtechnologyanda
EuropeanCommunities.AcceleratingtheDevelopmentoftheeHealthMarketinEurope.2007. Luxembourg: Office for Official Publications of the European Communities; 2007. http://ec. europa.eu/information_society/activities/health/downloads/index_en.htm,lastaccessedMay20, 2009. 11 G.F.Anderson,B.K.Frogner,R.A.Johns,U.E.Reinhardt,HealthcareSpendingandUseof InformationTechnologyinOECDCountries,HealthAffairs,May/June200625(3):819–31. 12 President’s Information Technology Advisory Committee (PITAC). Transforming Healthcare Through Information Technology – PITAC report to the president. Arlington: Nation CoordinationOfficeforComputing;2001.
10
12
2 Health Institutions and Information Processing
Fig. 2.7 Amobilecomputeronaward tosupportmedicaldocumentation andinformationaccess
soundmethodologyforsystematicallyprocessinginformation.However,thenewestinformationandcommunicationtechnologiesdonotguaranteehigh-qualityinformationprocessing. Both information processing technologies and methodologies must adequately andresponsiblybeappliedand,aswillbepointedoutlateron,systematicallymanaged.
2.4 Importance of Systematic Information Management 2.4.1 Affected People and Areas Nearlyallpeopleandallareasofahospitalareaffectedbythequalityoftheinformation system,asmostofthemneedvarioustypesofinformation(e.g.,aboutthepatient)intheir dailywork.Thepatientcancertainlyproitmostfromhigh-qualityinformationprocessing sinceitcontributestothequalityofpatientcareandtoreducingcosts. The professional groups working in a hospital, especially physicians, nurses, and administrativepersonnel,andothersarealsodirectlyaffectedbythequalityoftheinformationsystem.Astheyspend25%orevenmoreoftheirtimeoninformationhandling, theydirectlyproitfromgoodandeficientinformationprocessing.However,theywill alsofeeltheconsequencesifinformationprocessingispoor.
2.4 Importance of Systematic Information Management
13
2.4.2 Amount of Information Processing Theamountofinformationprocessinginhospitals,especiallyinlargerones,shouldnotbe underestimated.LetuslookatatypicalGermanuniversitymedicalcenter.Itisanenterprise encompassing staff of approximately 4,500 people, an annual budget of approximately €250 million, and, as a maximum care facility, numerous tasks in research, education,andpatientcare.Itconsistsofupto60departmentsandupto100wardswith upto1,500bedsandabout100outpatientunits.Annually,approximately,50,000inpatientsand250,000outpatientsaretreated,and20,000operationreports,250,000discharge letters,20,000pathologyreports,100,000microbiologyreports,200,000radiologyreports, and800,000clinicalchemistryreportsarewritten. Each year, approximately 300,000–400,000 new patient records, summing up to approximatelyeightmillionpiecesofpaper,arecreated(Fig.2.8).Whenstoredinapaperbasedway,anannualrecordvolumeofapproximately1,500misgenerated.InGermany, forexample,theyshouldbearchivedoveraperiodof30years.Whenstoreddigitally,the annual data volume needed is expected to be around 10–15 terabytes, including digital imagesanddigitalsignals. Thecomputer-basedtoolsofauniversitymedicalcenterencompassmorethanhundred of the computer-based application components, thousands of workstations and other
Fig. 2.8 Astudynurseinanoutpatientunitdealingwithamultitudeofpaper-basedforms
14
2 Health Institutions and Information Processing
terminals,andmorethanhundredservers(largercomputersystemsthatofferservicesand featurestoothercomputersystems),andtherespectivenetwork. Thenumbersinthemajorityofhospitalsaremuchsmaller.Inlargeroneswewillind, forexample,abouttendepartmentswith600bedsandabout20,000inpatientseveryyear. Inindustrializedcountries1,500staffmemberswouldworkthere,andtheannualbudget ofthehospitalwouldbeabout€80million.Especiallyinruralareas,wecanalsoindhospitalswithonlyonedepartmentandfewerthan50beds.
2.4.3 Sharing the Same Data There are different reasons for pursuing holistic and integrated information processing. Themostimportantreasonisthatvariousgroupsofhealthcareprofessionalswithinand outsidehealthcareinstitutionsneedthesamedata(Fig.2.9). Forexample,asurgeoninahospitaldocumentsthediagnosesandtherapiesofanoperatedpatientinanoperationreport.Thisreportservesasbasisforthedischargeletter.The dischargeletterisalsoanimportantdocumenttocommunicatewiththeadmittinginstitution,normallyageneralpractitioner.Diagnosisandtherapyarealsoimportantforstatistics aboutpatientcareandforqualitymanagement.Equally,theycontainimportantinformationforthesystematicnursingcareofapatient.Diagnosticandtherapeuticdataarealso relevantforbilling. InGermany,forexample,somebasicadministrativedatamustbecommunicatedtothe respectivehealthinsurancecompanyonlinewithin3daysafterpatientadmissionandafter discharge.Inacodedform,theyarethebasisforaccounting.Additionally,managingand controlling a hospital is possible only if the cost (such as consumption of materials or drugs)ofthetreatmentcanbecomparedtothecharacteristicsandseverityoftheillness, characterizedbydiagnosisandtherapy.
Fig. 2.9 Regularclinical roundbydifferenthealth careprofessionalsonaward
2.4 Importance of Systematic Information Management
15
2.4.4 Integrated Information Processing to Satisfy Information Needs Informationprocessinghastointegratethepartlyoverlappinginformationneedsofthe differentgroupsandareasofahospital(seeFig.2.10). Systematic,integratedinformationprocessinginahospitalhasadvantagesnotonlyfor thepatient,butalsoforthehealthcareprofessionals,thehealthinsurancecompanies,and thehospitalowners.Ifinformationprocessingisnotconductedgloballyacrossinstitutions, but locally, for example, in professional groups (physicians, nurses, and administrative staff)orareas(clinicaldepartments,institutes,andadministration),thiscorrespondstotraditionalseparationpoliticsandleadstoisolatedinformationprocessinggroups,suchas“the administration”or“theclinic.”Inthiscase,thequalityofthehospitalinformationsystem clearlydecreaseswhilethecostsforinformationprocessingincreaseduetothenecessity formultipledatacollectionandanalysis.Finally,thishasdisadvantagesforthepatientand, whenseenfromanationaleconomicalpointofview,forthewholepopulation. However,integrationofinformationprocessingshouldconsidernotonlyinformation processinginonehealthcareorganization,butalsoinformationprocessingamongdifferent institutions (such as integrated health care delivery systems). The achievements of modernmedicine,particularlyintheieldofacutediseases,haveledtotheparadoxical result that chronic diseases and multimorbidity increasingly gain in relevance. Among otherreasons,thisisduetomorepeoplebeingabletolivetooldage.Moreover,inmany
Fig. 2.10 Duringawardround:Healthcareprofessionalsjointlyusinginformationprocessingtools
16
2 Health Institutions and Information Processing
countries,anincreasingwillingnesstoswitchdoctorsandahigherregionalmobilityexist among patients. The degree of highly specialized and distributed patient care creates a greatdemandforintegratedinformationprocessingamonghealthcareprofessionalsand among health care institutions such as hospitals, general practices, laboratories, etc. In turn,thisraisestheneedformorecomprehensivedocumentationandeficient,comprehensiveinformationsystems.
2.4.5 Raising the Quality of Patient Care and Reducing Costs Systematic information processing is the key factor for raising quality and reducing costs.Whatdoes“systematic”meaninthiscontext?“Unsystematic”can,inapositive sense,meancreative,spontaneous,orlexible.However,“unsystematic”canalsomean chaotic,purposeless,andineffective,andalsoentailhighcostscomparedtothebeneits gained. “Systematic” in this context means purposeful and effective, and with great beneit regardingthecosts.Bearingthisinmind,itisobviousthatinformationprocessingina healthcareinstitutionshouldbemanagedsystematically.Duetotheimportanceofinformationprocessingasaqualityandcostfactor,aninstitutionhastoinvestsystematically in its health information system. These investments deal with both staff and tools for information processing. They aim at increasing quality of patient care and at reducing costs. Unsystematicinformationprocessingnormallyleadstoalowqualityofhealthinformation systems,andtheinformationneedsofthestaffanddepartmentscannotbeadequatelysatisied.Whenhealthinformationsystemsarenotsystematicallymanaged,theytendtodevelop in a chaotic way. This has severe consequences: decreased data quality, and higher costs, especiallyfortoolsandinformationprocessingstaff,nottomentionaspectssuchasdataprotectionanddatasecurityviolation.Evenworse,insuficientlymanagedinformationsystems cancontributetobreakdownsinestablishedclinicalworklows,toareducedeficiencyof care,touserboycott,todecreasedqualityofcareand–intheend–evenendangerpatient safety. Toadequatelyprocessinformationandapplyinformationandcommunicationtechnology,knowledgeandskillsforthesetasksarerequired.
2.4.6 Basis of Systematic Information Processing Ifthehospitalmanagementdecidestoinvestinsystematicinformationprocessing(andnot in ighting the effects of chaotic information processing, which normally means much higherinvestments),itdecidestomanagethehospitalinformationsysteminasystematic way.Themanagementofahospitalinformationsystemformsandcontrolstheinformation system,anditensuresitseficientoperation.
2.5 Examples
17
2.5 Examples 2.5.1 Knowledge Access to Improve Patient Care Imaginethefollowingsituation:UrsulaB.waspregnantwithquintuplets.Shehadalready spentmorethan5monthsinaUniversityMedicalCenter.Shehadtospendmostofthis timelyinginbed.Duringthecourseofherpregnancy,herphysicalproblemsincreased. Fromthe28thweekon,shesufferedsevererespiratorydistress. ProfessorL.,thepediatrician,whowasalsoinvolvedinhertreatment,hadthefollowingquestion:Whatarethechancesoftheinfantsbeingbornhealthyatthisgestationalage? Hewenttoacomputer,whichisconnectedtothecomputernetworkoftheUniversity MedicalCenter.Thephysiciancalledupafunction“knowledgeaccess”and,asapplicationcomponent,aliteraturedatabase(MEDLINE13),whichcontainsthecurrentstateof theartofmedicalknowledgeworldwide(Fig.2.11). Thefollowinginformationresultedfromthisknowledgeaccess:Severalpublications statedthatonlyslimchancesexistforallinfantstosurviveingoodhealth.Iftheyareborn duringthe28thweekofpregnancy,thechanceforsurvivalisabout15%.Incaseofbirth duringthe30thweek,theirchanceswouldimprovetoabout75%.Also,accordingtothe literature,furtherdelayofthedeliverydoesnotimprovetheprognosisofthequintuplets.
Fig. 2.11 Prof.L.,Headof theDepartmentof Pediatrics,workingwith aliteratureserver
13
OfferedforfreebytheNationalLibraryofMedicine(NLM),Bethesda,USA,http://www.ncbi. nlm.nih.gov/entrez/query.fcgi
18
2 Health Institutions and Information Processing
Fig. 2.12 TheHeidelbergquintuplets
The physician discussed the results with the expectant mother. Despite her respiratory problems,shehadthestrengthtoendure2moreweeks. OnJanuary21stoftherespectiveyear,thequintupletswerebornwellandhealthyat theUniversityMedical Center (Fig.2.12). A team of 25 physicians, nurses, and midwives assisted during the delivery. Appropriate knowledge access was of crucial importance. Today,knowledgeaccessatawardisinmanyhospitalsanintegratedpartofahospital informationsystem. Youmaywonder,inwhichyearandinwhichmedicalcenterthequintupletswereborn. Itwasin1999attheHeidelbergUniversityMedicalCenter.Althoughthisrealexample datesbackmorethan10yearswefounditstillworthtoreportonbecauseitshowsclearly howimportantitistohaveinformationandknowledgeavailable,andinthisrespectis pioneeringforthefuturetrends. Bytheway,allquintuplesarewellandProf.L.,theattendingphysician,becametheir godfather.
2.5.2 Nonsystematic Information Processing in Clinical Registers Thefollowingexampleshowswhatcanhappenwheninformationprocessingisdoneina nonsystematic (or, better, chaotic?) manner from yet another point of view.14 Let us heexampleisbasedonasimilaronein:GreenSB,ByarDP.UsingObservationalDatafromRegistries T toCompareTreatments:TheFallacyofOmnimetrics.StatisticsinMedicine1984;3:361–70.
14
2.5 Examples
19
Table 2.1 Example of Simpson’s paradox – Success rates of Novum and Verum treatments for patientswithdiagnosisD,treatedduringtheyearsdatthePlötzbergMedicalCenterandMedical School(PMC) Success Success Yes No Rate S Allpatients Novum
333
1,143
1,476
(23%)
Verum
243
1,113
1,356
(18%)
S
576
2,256
2,832
Novum
24
264
288
Verum
147
906
1,053
S
171
1,170
1,341
Novum
309
879
1,188
(26%)
Verum
96
207
303
(32%)
405
1,086
1,491
Malepatients (8%) (14%)
Femalepatients
S
analyzea(ictitious)clinicalregisterfromthe(ictitious)PlötzbergMedicalCenterand MedicalSchool(PMC).PMCwillbeusedinexamplesandexercisesinthisbook. Table2.1showsstatisticswithpatientshavingdiagnosisD,forexample,rheumatism, andtreatedduringtheyearsd,forexample,1991–2001,atPMC.Thepatientshaveeither receivedstandardtherapy,Verum,oranewtherapy,Novum. Comparing the success rates of Novum and Verum, one might conclude that the newtherapyisbetterthanthestandardtherapy.Applyinganappropriatestatisticaltest wouldleadtoalowpvalueandasigniicantresult.Thesuccessratewasalsoanalyzed bysex.ThisresultedinVerumleadinginfemalepatientsaswellasinmalepatients. Isoneofourconclusionserroneous?Ormaybeboth?Whatwouldasystematicdesign andanalysisofsucharegisterbe?Afterlookingatthedata,onecanidentifyafairlysimple reasonforthisso-calledSimpson’sparadox.Themethodologyforprocessinginformation systematicallyoughttopreventsucherrors;however,itisfarmorecomplex.
2.5.3 The WHO eHealth Resolution15 Nowadays computer-based information processing in health institutions and moreover health networks is referred to as eHealth. In its eHealth resolution, the World Health 15
orldHealthOrganization.eHealthResolution.58thWorldHealthAssembly,Resolution28. W May25,2005.Geneva:WHO;2005.58thWorldHealthAssembly’shomepage:http://www. emro.who.int/HIS/ehealth/PDF/EB115_R20-en.pdf
20
2 Health Institutions and Information Processing
Organization (WHO) strongly recommends to systematically introduce, improve, and manageeHealthworldwide,whichshouldleadto“eHealthforallby2015.”16 “TheFifty-EighthWorldHealthAssembly… 1. URGESMemberStates: 1. toconsiderdrawingupalong-termstrategicplanfordevelopingandimplementing eHealthservicesthatincludesanappropriatelegalframeworkandinfrastructureand encouragespublicandprivatepartnerships; 2. todeveloptheinfrastructureforinformationandcommunicationtechnologiesfor health as deemed appropriate to promote equitable, affordable, and universal accesstotheirbeneits,andtocontinuetoworkwithinformationtelecommunication agencies and other partners to strive to reduce costs to make eHealth successful; 3. to build on closer collaboration with the private and not-for-proit sectors in information and communication technologies, to further public services for health; 4. toendeavortoreachcommunities,includingvulnerablegroups,witheHealthservicesappropriatetotheirneeds; 5. to mobilize multisectoral collaboration for determining evidence-based eHealth standardsandnorms,toevaluateeHealthactivities,andtosharetheknowledgeof costeffectivemodels,thusensuringquality,safetyandethicalstandards; 6. toestablishnationalcentersandnetworksofexcellenceforeHealthbestpractice, policycoordination,andtechnicalsupportforhealthcaredelivery,serviceimprovement,informationtocitizens,capacitybuilding,andsurveillance; 7. toconsiderestablishingandimplementingnationalpublic-healthinformationsystemsandtoimprove,bymeansofinformation,thecapacityforthesurveillanceof, andrapidresponseto,diseaseandpublichealthemergencies. 2. REQUESTStheDirector-General: 1. topromoteinternational,multisectoralcollaborationwithaviewtoimprovingcompatibilityofadministrativeandtechnicalsolutionsintheareaofeHealth; 2. to document and analyze developments and trends, inform policy and practice in countries,andreportregularlyonuseofeHealthworldwide; 3. to provide technical support to Member States in relation to eHealth products and services by disseminating widely experiences and best practices, in particularontelemedicinetechnology;devisingassessmentmethodologies;promoting research and development; and furthering standards through diffusion of guidelines; 4. tofacilitatetheintegrationofeHealthinhealthsystemsandservices,includingin thetrainingofhealth-careprofessionalsandincapacitybuilding,inordertoimprove accessto,andqualityandsafetyof,care;
ealy JC. The WHO eHealth Resolution – eHealth for all by 2015? Methods Inf Med 2007; H 46(1):2–4.
16
2.5 Examples
21
5. tocontinuetheexpansiontoMemberStatesofmechanismssuchastheHealthAcademy whichpromotehealthawarenessandhealthylifestylesthrougheLearning17; 6. toprovidesupporttoMemberStatestopromotethedevelopment,applicationand managementofnationalstandardsofhealthinformation;andtocollectandcollate availableinformationonstandardswithaviewtoestablishingnationalstandardized health information systems in order to facilitate easy and effective exchange of informationamongMemberStates; 7. tosupportregionalandinterregionalinitiativesintheareaofeHealthamonggroups ofcountriesthatspeakacommonlanguage.”
2.5.4 Estimated Impact of eHealth to Improve Quality and Efficiency of Patient Care ThisexampleistakenfromastudyreportpresentedbyGartneronbehalfoftheSwedish MinistryofHealthandSocialAffairs.18 ThisstudyanalyzedthepotentialbeneitsofanincreasedusageofeHealthinsixEU memberstates: “…ThereisasigniicanthealthcareimprovementpotentialusingeHealthasacatalyst.… Examplesofquantiiedpotentialsinclude: • Fivemillionyearlyoutpatientprescriptionerrorscouldbeavoidedthroughtheuseof ElectronicTransferofPrescriptions. • 100,000yearlyinpatientadversedrugeventscouldbeavoidedthroughComputerized Physician Order Entry and Clinical Decision Support. This would in turn free up 700,000bed-daysyearly,anopportunityforincreasingthroughputanddecreasingwaitingtimes,correspondingtoavalueofalmost€300million. • 49,000 cases of inpatient Hospital Acquired Infections could be avoided every year collectively through the use of Business Intelligence and Data Mining for real time detection of in-hospital infections. This could increase availability by over 270,000 bed-days,resultinginopportunitysavingsofover€131million. • 11,000 deaths caused by complications related to diabetes could collectively be reduced through Electronic Medical Records with Chronic Disease Management capabilities. • 5.6millionadmissionstohospitalsforchronicallyillpatientscouldbeavoidedcollectivelythroughtheuseofTelemedicineandHomeHealthMonitoring. • Nine million bed-days yearly could be freed up through the use of Computer-Based PatientRecords,anopportunityforeitherincreasingthroughputordecreasingwaiting times,correspondingtoavalueofnearly€3.7billion.
e Learningisunderstoodinthiscontexttomeanuseofanyelectronictechnologyandmediain supportoflearning. 18 http://www.sweden.gov.se/content/1/c6/12/98/15/5b63bacb.pdf 17
22
2 Health Institutions and Information Processing
• Patients can become more involved and accountable for the management of their
chronicconditionsthroughaccesstoknowledgebasedbestpracticesviaanEMRwith Chronic Disease Management capabilities and communication with their physicians throughaPatientPortal. • Patients can have more control on how and when to engage with their physicians throughtechnologiessuchasPatientPortalsandPersonalHealthRecordsthatenable alternativewaysofcommunicationandconsultationsuchase-mailande-visits.”
2.6 Exercises 2.6.1 Amount of Information Processing in Typical Hospitals Estimatethefollowingiguresforatypicaluniversitymedicalcenterandforatypicalrural hospital. To solve this exercise, look at the strategic information management plan for informationprocessingofahospital,orproceedwithyourownlocalinvestigations. • Numberof(inpatient)clinicaldepartmentsandinstitutes • Numberofwardsandoutpatientunits • Numberofemployees • Annualbudget • Numberofbeds,inpatients,andoutpatientsperyear • Numberofnewpatientrecordsperyear • Numberofdischargelettersperyear • Numberofcomputerservers,workstations,andterminals • Numberofoperationreports,clinicalchemistryreports,andradiologyreportsperyear
2.6.2 Information Processing in Different Areas Findthreeexamplesofinformationprocessingforeachofthefollowingareasinahospital,takingintoaccountthedifferenthealthcareprofessionalgroupsworkingthere.Which information is processed during which activities, and which tools are used? Take noncomputer-based and computer-based information processing into consideration in your examples. • Informationprocessingonaward • Informationprocessinginanoutpatientunit • Informationprocessinginanoperatingroom • Informationprocessinginaradiologydepartment • Informationprocessinginthehospitaladministration
2.7 Summary
23
2.6.3 Good Information Processing Practice Havealookatthefollowingtypicalareasofhospitals.Trytoindtwoexamplesof“good” informationprocessingpracticesintheseareas,andtwoexamplesof“poor”information processingpractices.Whichpositiveornegativeconsequencesforthepatientscouldthey have? • Patientadministration • Cardiologicward • Laboratory
2.7 Summary Informationprocessingisanimportantqualityfactor,butanenormouscostfactoraswell. Itisalsobecomingaproductivityfactor.Informationprocessingshouldofferaholistic viewofthepatientandofthehospital.Ahospitalinformationsystemcanberegardedas thememoryandnervoussystemofahospital. Informationandcommunicationtechnologyhasbecomeeconomicallyimportantand decisiveforthequalityofhealthcare.Itwillcontinuetochangehealthcare. Theintegratedprocessingofinformationisimportant,because • allgroupsofpeopleandallareasofahospitaldependonitsquality, • theamountofinformationprocessinginhospitalsisconsiderable,and • healthcareprofessionalsfrequentlyworkwiththesamedata. Thesystematicprocessingofinformation • contributestohigh-qualitypatientcare,and • reducescosts. Informationprocessinginhospitalsiscomplex.Therefore, • thesystematicmanagementandoperationofhospitalinformationsystems,and • medical informatics specialists responsible for the management, and operation of h ospitalinformationsystemsareneeded.
Information System Basics
3
3.1 Introduction Every domain usually has its own terminology, which often differs from the ordinary understandingofconceptsandterms.Thischapterpresentstheterminologyforinformationsystemsandtheirmanagement,asusedinthisbook.Itis,therefore,essentialtoread thischaptercarefully.AllrelevantconceptscanalsobefoundintheThesaurusattheend ofthebook. Afterreadingthischapter,youshouldbeabletoanswerthefollowingquestions: • Whatisthedifferencebetweendata,information,andknowledge? • Whatareinformationsystems,andwhataretheircomponents? • Whatisinformationmanagement?
3.2 Data, Information, and Knowledge Dataconstitutereinterpretablerepresentationsofinformation,orknowledge,inaformalized manner suitable for communication, interpretation, or processing by humans or machines.Formalizationmaytaketheformofdiscretecharactersorofcontinuoussignals (e.g.,soundsignals).Tobereinterpretable,therehastobeanagreementonhowdatarepresentinformation.Forexample,“PeterSmith”or“001001110”aredata.Asetofdatathat isputtogetherforthepurposeoftransmissionandthatisconsideredtobeoneentityfor thispurposeiscalledamessage. Thereisnouniquedeinitionofinformation.Dependingonthepointofview,thedeinitionmaydealwithasyntacticaspect(thestructure),asemanticaspect(themeaning),ora pragmaticaspect(theintentionorgoalofinformation).Wewillsimplydeineinformation asspeciicdeterminationaboutentitiessuchasfacts,events,things,persons,processes, ideas,orconcepts.Forexample,whenaphysiciandeterminesthediagnosis(facts)ofa patient(person),thenheorshehasinformation.
A.Winteretal.,HealthInformationSystems, DOI:10.1007/978-1-84996-441-8_3,©Springer-VerlagLondonLimited2011
25
26
3
Information System Basics
Knowledgeisgeneralinformationaboutconceptsinacertain(scientiicorprofessional) domain(e.g.,aboutdiseases,therapeuticmethods).Knowledgecontrastswithspeciicinformationaboutparticularindividualsofthedomain(e.g.,patients).Theknowledgeofanurse, forexample,compriseshowtotypicallydealwithpatientssufferingfromdecubitus. Forthesakeofsimplicity,wewilloftenusetheterminformationprocessingwhenwe meanprocessingofdatatogetherwithitsrelatedinformationandknowledge.
3.3 Information Systems and Their Components 3.3.1 Systems and Subsystems Beforetalkingaboutinformationsystems,letusirstdeinetheconceptsystem.Asdeined here,asystemisasetofpersons,things,events,andtheirrelationshipsthatformsanintegratedwhole.Wedistinguishbetweennaturalsystemsandartiicial(man-made)systems. Forexample,thenervoussystemisatypicalnaturalsystem,consistingofneuronsand their relationships. A man-made system is, for example, a hospital, consisting of staff, patients, and relatives, and their interactions. If a (man-made) system consists of both humanandtechnicalcomponents,itcanbecalledasocio-technicalsystem. Asystemcan,inprinciple,bedividedintosubsystemsthatcompriseasubsetofthe componentsandtherelationshipsbetweenthem.Forexample,apossiblesubsystemofthe nervoussystemisthesympatheticnervoussystem.Asubsystemofahospitalis,forexample,awardwithitsstaffandpatients. Subsystemsthemselvesareagainsystems.
3.3.2 Information Systems Aninformationsystemisthatpartofaninstitutionthatprocessesandstoresdata,information,andknowledge.Itcanbedeinedasthatsocio-technicalsubsystemofaninstitution, whichcomprisesallinformationprocessingaswellastheassociatedhumanortechnical actorsintheirrespectiveinformationprocessingroles.Thismeansthat,forexample,the computers,printers,telephones,aswellasthestaffusingthemtomanageinformationare partoftheinformationsystemofaninstitution. “Socio-” refers to the people involved in information processing (e.g., health care professionals,administrativestaff,andcomputerscientists),whereas“technical”refersto informationprocessingtools(e.g.,computers,telephones,andpatientrecords).Thepeople andmachinesinaninstitutionareconsideredonlyintheirroleasinformationprocessors, carryingoutspeciicactionsfollowingestablishedrules. An information system that comprises computer-based information processing and communication tools is called a computer-based information system. An information
3.3 Information Systems and Their Components
27
system can be divided into subsystems, which are called sub-information systems. For example,theinformationsystemofaninstitutioncanbesplitintotwosub-information systems:thepartwherecomputer-basedtoolsareusediscalledthecomputer-basedpart; therestiscalledthenon-computer-basedpartofaninformationsystem.
3.3.3 Components of Information Systems Whendescribinganinformationsystem,itcanhelptolookatthefollowingtypicalcomponentsofinformationsystems:enterprisefunctions,businessprocesses,applicationcomponents,andphysicaldataprocessingsystems.1 Anenterprisefunctiondescribeswhatactinghumanormachineshavetodoinacertain enterprisetocontributetoitsmissionandgoals.Forexample,patientadmission,medical andnursingcareplanning,orinancialaccountingdescribetypicalenterprisefunctions. Enterprisefunctionsareongoingandcontinuous.Theydescribewhatistobedone,nothow itisdone.Enterprisefunctionscanbestructuredintoahierarchyofenterprisefunctions, where an enterprise function can be described in more detail by reined sub-functions. Enterprisefunctionsareusuallydenotedbynounsorgerunds(i.e.,wordsendingwith-ing). Theactionssummarizedbyanenterprisefunctionareinmostcasessigniicantlydealing withinformationprocessing.Lateronwewillfocusmorestrictlyonthisaspectandthereforerestricttoinformationprocessingenterprisefunctions(seeSect.5.3.2.1). Forthesakeofsimplicity,wewillrefertoenterprisefunctionsashospitalfunctions,if therespectiveenterpriseisahospital. Anactivityisaninstantiationofanenterprisefunction.Forexample,“thephysician admitsthepatientSmith” is an activity of the enterprise function patient admission. In contrasttoenterprisefunctions,activitieshaveadeinitebeginningandend. Todescribehowanenterprisefunctionisperformed,notonlymayinformationabout itsreinedsub-functionsbeneeded,butinformationabouttheirchronologicalandlogical sequencemayalsobeneeded.Withbusinessprocesses,thesequenceof(sub-)functions togetherwiththeconditionsunderwhichtheyareperformedcanbedescribed.Business processesareusuallydenotedbyverbs,whichcanbefollowedbyanoun(e.g.,“admitting apatient,”“planningcare”or“writingadischargeletter”).Processinstancesarecomposed oftheindividualactivities;hencetheyalsohaveadeinitebeginningandend.Whileenterprisefunctionsconcentrateonthe“what,”businessprocessesfocusonthe“how”ofactivities.Enterprisefunctionscanbeconsideredasrepresentativesofbusinessprocesses. Whereasenterprisefunctionsdescribewhatisdone,wenowwanttoconsidertoolsfor processingdata,inparticularapplicationcomponentsandphysicaldataprocessingsystems.Bothareusuallyreferredtoasinformationprocessingtools.Theydescribethemeans usedforinformationprocessing. Application components support enterprise functions. We distinguish computer-based fromnon-computer-basedapplicationcomponents.Computer-basedapplicationcomponents arecontrolledbysoftware products. A software product is an acquired or self-developed 1
WewillgivelittlebitmoreformaldefinitionofthesetermslateroninSect.5.3.2.
28
3
Information System Basics
pieceofsoftwarethatcanbeinstalledonacomputersystem.Forexample,thecomputerbasedapplicationcomponentpatientadministrationsystemstandsfortheinstallationofa softwareproducttosupportenterprisefunctionssuchaspatientadmissionandadministrativedischargeandbilling. Non-computer-based application components are controlled by working plans that describehowpeopleusecertainphysicaldataprocessingsystems.Forexample,anoncomputer-based application component called nursing management and documentation systemiscontrolledbyrulesregardinghow,bywhom,andinwhichcontextgivenforms fornursingdocumentationhavetobeused.Inthisexample,thepaper-basedformsthatare usedrepresentphysicaldataprocessingsystems(seeFig.3.1). Communicationandcooperationamongapplicationcomponentsmustbeorganizedin suchawaythattheenterprisefunctionsareadequatelysupported. Physicaldataprocessingsystems,inally,describetheinformationprocessingtoolsthat areusedtoimplementcomputer-basedaswellasnon-computer-basedapplicationcomponents.Physicaldataprocessingsystemscanbehumanactors(suchasthepersondelivering mail), non-computer-based physical tools such as forms for nursing documentation, paper-basedpatientrecordsortelephones,orcomputersystems(suchasterminals,servers, and personal computers). Computer systems can be physically connected via data wires,leadingtophysicalnetworks.Figure3.2showssometypicalphysicaldataprocessing
Fig. 3.1 Anexampleforformsandfoldersfornursingdocumentation,representingaphysicaldata processingsystem.Therulesthatdescribewhomayusetheseforms,andhowtheyshouldbeused, makeuptheapplicationcomponent
3.3 Information Systems and Their Components
29
systems.Theprinter,forexample,couldcontributeintheimplementationoftheapplicationcomponentmedicaldocumentationsystembyprintingdocumentationforms. Detailsonthemostrelevantinformationprocessingtoolsinhospitalscanbefoundin Sects.6.4and6.6.
3.3.4 Architecture and Infrastructure of Information Systems Thearchitectureofaninformationsystemdescribesitsfundamentalorganization,representedbyitscomponents,theirrelationshipstoeachotherandtotheenvironment,andby theprinciplesguidingitsdesignandevolution.2Thearchitectureofaninformationsystem can be described by the enterprise functions, the business processes, the information processingtools,andtheirrelationships. There may be several architectural views of an information system, for example, a functionalviewlookingprimarilyattheenterprisefunctions,aprocessviewlookingprimarily at the business processes, etc. Architectures that are equivalent with regard to certaincharacteristicscanbesummarizedinacertainarchitecturalstyle.
Fig. 3.2 Typicalphysicaldataprocessingsystemsinanoutpatientunit(e.g.,printer,telephone,and non-computer-basedpatientrecord) I nstituteofElectricalandElectronicsEngineers(IEEE).Std1471-2000:RecommendedPracticefor ArchitecturalDescriptionofSoftware-IntensiveSystems.September2000.http://standards.ieee.org
2
30
3
Information System Basics
Whenthefocusisputontothetypes,number,andavailabilityofinformationprocessingtoolsusedinagivenenterprise,thisisalsocalledtheinfrastructureofaninformation system.
3.4 Information Management Ingeneral,managementcomprisesallleadershipactivitiesthatdeterminetheinstitution’s goals,structures,andbehaviors.Accordingly,informationmanagement(ormanagement of information systems) comprises those management activities that deal with the managementofinformationprocessinginaninstitution,forexample,ahospital.Thegoal ofinformationmanagementissystematicinformationprocessingthatcontributestothe institution’sstrategicgoals(suchaseficientpatientcareandhighsatisfactionofpatients andstaffinahospital).Informationmanagementthereforedirectlycontributestotheinstitution’ssuccessandabilitytocompete. Thegeneraltasksofinformationmanagementareplanning,directing,andmonitoring.In otherwords,thismeans • planningtheinformationsystemanditsarchitecture, • directingitsestablishmentanditsoperation,and • monitoringitsdevelopmentandoperationwithrespecttotheplannedobjectives. Information management encompasses the management of all components of an informationsystem–themanagementofenterprisefunctionsandbusinessprocesses,of applicationcomponents,andofphysicaldataprocessingsystems. Informationmanagementcanbedifferentiatedintostrategic,tactical,andoperational information management. Strategic information management deals with information processingasawhole.Tacticalinformationmanagementdealswithparticularenterprise functions or with application components that are introduced, removed, or changed. Operationalinformationmanagement,inally,isresponsibleforoperatingthecomponents of the information system. It cares for its smooth operation, for example, by planning necessary personal resources, by failure management, or by network monitoring. InformationmanagementinhospitalsisdiscussedindetailinSect.9.2.
3.5 Exercises 3.5.1 On the Term Information System Trytodescribeinyourownwords,whattheterminformationsystem,asintroducedin Sect.3.3,means.
3.6 Summary
31
3.5.2 On Enterprise Functions Choosetwodifferenttypesofenterprises,forexample,abankandatheatre.Trytolistive majorenterprisefunctionsfortheseenterprises.
3.5.3 On Application Components PleaselookatFig.3.1.Itshowsaphysicaldataprocessingsystem.Pleasetrytoformulate some rules as to how the different parts are to be used, and by whom, to implement a non-computer-basednursingdocumentationsystemasapplicationcomponent.Doyouneed anyotherphysicaldataprocessingsystemstoimplementthisapplicationcomponent?
3.5.4 On Architectures and Infrastructures Letusforthisexercisefocusonthearchitectureofhouses(notonthearchitectureofinformationsystems).Describetwodifferentarchitecturalstylesforhouses.Identifyiveitems, whichareimportanttodescribetheinfrastructureofacertainhouse.
3.5.5 On Information Management Whatdoesinformationmanagementmean?Describethreeinformationmanagementtasks inyoureverydaylife.
3.6 Summary Whenworkingoninformationsystems,wemustdistinguishbetweendata,information, andknowledge: • Datacanbedefinedasarepresentationofinformation,orknowledgeinaformalized manner,suitableforcommunicating,interpreting,orprocessing. • Information can be deined as speciic determination about entities, such as facts, events,things,persons,processes,ideas,orconcepts. • Knowledgecanbedefinedasgeneralinformationaboutconceptsinacertaindomain.
32
3
Information System Basics
Asystemisasetofpersons,things,events,andtheirrelationshipsthatformanintegrated whole.Systemscanbedividedintosubsystems. Aninformationsystemcanbedeinedasthesocio-technicalsubsystemofaninstitution,whichcomprisesallinformationprocessingaswellastheassociatedhumanortechnical actors in their respective information processing roles. Typical components of informationsystemsare: • theenterprisefunctionssupported; • thebusinessprocessesthattakeplace; • theapplicationcomponentsthatsupporttheenterprisefunctions; • thephysicaldataprocessingsystemstheapplicationcomponentsareexecutedon. Thesubsystemofaninformationsystemwherecomputer-basedtoolsareusediscalledthe computer-basedpartoftheinformationsystem.Thearchitectureofaninformationsystem describesitsfundamentalorganization,representedbyitscomponents,theirrelationships to each other and to the environment, and by the principles guiding its design and evolution. Informationmanagementcomprisesthosemanagementactivitiesinaninstitutionthat dealwiththemanagementofinformationprocessingandthereforewiththemanagement oftheinstitution’sinformationsystem.
Health Information Systems
4
4.1 Introduction An information system was previously deined as the socio-technical subsystem of an institution,whichcomprisesallinformationprocessingaswellastheassociatedhumanor technicalactorsintheirrespectiveinformationprocessingroles.Healthinformationsystems(HIS)aredealingwithprocessingdata,information,andknowledgeinhealthcare environments. Especially with regard to chronic diseases, it becomes more and more importanttoorganizehealthcareinapatient-centricway,suchthatallparticipatingin-or outpatientcareinstitutionscooperateveryclosely.Thisisalsodenotedasintegratedcare. Inintegratedcareitisnecessarytoproviderelevantinformationnotonlywithinasingle institution,butwhereverandwheneveritisneeded.Thisincludesmedicalpractices,rehabilitationcenters,nursingcenters,andeventhehomeofthepatient.Wethereforedifferentiateinstitutionalandtransinstitutionalhealthinformationsystems.Inthefollowing,we will introduce hospital information systems, which are the most complex instances of institutionalinformationsystems,andtransinstitutionalinformationsystems.Throughout thebookwewillusethetermhealthinformationsystemandtheabbreviationHISifwe discussaspectsconcerningbothhospitalinformationsystemsandtransinstitutionalinformationsystems.Ifwedealwithpropertiesbeinguniqueforoneofthese,wewillusethe termshospitalinformationsystemandtransinstitutionalinformationsystem,respectively. Afterreadingthischapter,youshouldbeabletoanswerthefollowingquestions: • Whatarehospitalinformationsystems? • Whataretransinstitutionalhealthinformationsystems? • Whatarethechallengesforhealthinformationsystems? • Whatareelectronichealthrecords?
4.2 Hospital Information Systems Withthedeinitionofinformationsystemsinmind,ahospitalinformationsystemcanbe easilydeined.Ahospitalinformationsystemisthesocio-technicalsubsystemofahospital,whichcomprisesallinformationprocessingaswellastheassociatedhumanortechnical A.Winteretal.,HealthInformationSystems, DOI:10.1007/978-1-84996-441-8_4,©Springer-VerlagLondonLimited2011
33
34
4
Health Information Systems
actors in their respective information processing roles. Typical components of hospital informationsystemsareenterprisefunctions,businessprocesses,applicationcomponents, andphysicaldataprocessingsystems(seeSect.3.3.3).Forthesakeofsimplicity,wewill refertoenterprisefunctionsofahospitalashospitalfunctions. Asaconsequenceofthisdeinition,ahospitalhasahospitalinformationsystemfrom thebeginningofitsexistence.Therefore,thequestionisnotwhetherahospitalshouldbe equipped with a hospital information system, but rather how its performance can be enhanced,forexample,byusingstate-of-the-artinformationprocessingtools,orbysystematicallymanagingit. Allgroupsofpeopleandallareasofahospitalmustbeconsideredwhenlookingat informationprocessing.Thesensibleintegrationofthedifferentinformationprocessing toolsinahospitalinformationsystemisimportant. Hospitalstaffcanbeseenaspartofthehospitalinformationsystem.Forexample,when workinginthedepartmentofpatientrecordsorasanoperatorinadepartmentforinformationandcommunicationtechnology,staffmembersdirectlycontributetoinformationprocessing. However, in their role as user of the hospital information system, they use information processing tools (e.g., a nurse may use a telephone or a computer). Each employeemaycontinuouslyswitchbetweenthesetworoles. Thegoalofahospitalinformationsystemistosuficientlyenabletheadequateexecution of hospital functions for patient care, including patient administration, taking into account economic hospital management as well as legal and other requirements. Legal requirements concern, for example, data protection or reimbursement aspects. Other requirementscanbe,forexample,thedecisionofahospitalexecutiveboardonhowto storepatientrecords. Tosupportpatientcareandtheassociatedadministration,thetasksofhospitalinformation systemsare: • Tomakeinformation,primarilyaboutpatients,available:currentinformationshouldbe providedontime,attherightlocation,toauthorizedstaff,inanappropriateandusable form.Forthispurpose,datamustbecorrectlycollected,stored,processed,andsystematicallydocumentedtoensurethatcorrect,pertinent,andup-to-datepatientinformation canbesupplied,forinstance,tothephysicianoranurse(Fig.4.1). • Tomakeknowledge,forexample,aboutdiseases,sideeffects,andinteractionsofmedicationsavailabletosupportdiagnosticsandtherapy. • Tomakeinformationaboutthequalityof patientcareandtheperformanceandcost situationwithinthehospitalavailable. Inadditiontopatientcare,universitymedicalcentersundertakeresearchandeducationto gainmedicalknowledgeandtoteachstudents. Whenhospitalinformationsystemsmakeavailable • therightinformationandknowledge • attherighttime • attherightplace • totherightpeople • intherightform,
4.2 Hospital Information Systems
35
Fig. 4.1 Ahealthcareprofessional accessingpatientinformation
sothatthesepeoplecanmaketherightdecisions,thisisalsodescribedasinformationand knowledgelogistics. Hospitalinformationsystemshavetoconsidervariousareasofahospital,suchas • wards, • outpatientunits, • serviceunits:diagnostic(e.g.,laboratorydepartment,radiologydepartment),therapeutic(e.g.,operationroom),andothers(e.g.,pharmacy,patientrecordsarchive,library, bloodbank), • hospital administration areas (e.g., patient administration department, patient record archive, department of quality management, inancial and controlling department, department of facility management, information management department, general administrationdepartment,humanresourcesdepartment), • officesandwritingservicesfor(clinical)reportwriting. Inaddition,therearethemanagementareas,suchashospitalmanagement,management of clinical and non-clinical departments, administration management, and nursing management. Theseareasarerelatedtopatientcare.Theycouldbebrokendownfurther.Foruniversitymedicalcenters,additionalareas,neededforresearchandeducation,mustbeaddedto theabovelist. Obviously,themostimportantpeopleinahospitalarethe • patientsand,incertainrespects,their • visitors.
36
4
Health Information Systems
Themostimportantgroupsofpeopleworkinginahospitalare • physicians, • nurses, • administrativestaff, • technicalstaffand • medicalinformaticiansandotherhealthinformationmanagementstaff. Within each group of people, different needs and demands on the hospital information systemmayexist,dependingontherole,tasks,andresponsibilities.Wardphysicians,for example,requireinformationthatisdifferentfromthatrequiredbyphysiciansworkingin serviceunitsorbyseniorphysicians.Patientssometimesneedsimilarinformationasphysiciansbutinadifferentform.
4.3 Transinstitutional Health Information Systems In many countries, the driving force for health care and for ICT in health care has recently been the trend toward a better coordination of care, combined with rising costpressure.Oneconsequenceistheshifttowardbetterintegratedandsharedcare. Thismeansthatthefocuschangesfromisolatedproceduresinonehealthcareinstitution(e.g.,onehospitaloronegeneralpractice)tothepatient-orientedcareprocess, encompassing diagnosis and therapy, spreading over institutional boundaries (Fig.4.2). Agroupoftwoormorelegallyseparatedhealthcareinstitutionsthathavetemporarily andvoluntarilyjoinedtogethertoachieveacommonpurposearedeinedasahealthcare network. The information system of a health care network is called a transinstitutional health information system. Typical examples are regional health information systems, comprisingthehealthcareenvironmentinacertainregion,including,forexample,hospitals,oficesofgeneralpractitioners,pharmacies,rehabilitationcenters,homecareorganizations,andevenhealthinsurancesandgovernmentalauthorities. In the United States, for example, health care institutions are merging into large integrated health care delivery systems. These are systems of health care institutions thatjointogethertoconsolidatetheirroles,resources,andoperationstodeliveracoordinatedrangeofservicesandtoenhanceeffectivenessandeficiencyofpatientcare. ThesituationinEuropeisalsochangingfromhospitalsascentersofcaredeliveryto decentralizednetworksofhealthcaredeliveryinstitutionsthatarecalledregionalnetworksorhealthcarenetworks.Enterpriseboundariesareblurring.Hospitalinformation systems will increasingly be linked with information systems of other health care institutions. Forexample,theHannoverMedicalSchool,alargehospitalwith75departmentsand morethan1,400patientbeds,providesawebportalforitspartners,wherepatientdata can be shared and mutually updated by all institutions participating in patient care. Among others, a rehabilitation clinic in Fallingbostel, 50 km north of Hannover, has
4.3 Transinstitutional Health Information Systems
37
Fig. 4.2 Ageneralpractitioneraccessingdocumentsofahospitalinformationsystem
accesstothedataoftransplantpatients,whoneedtimetorecoveraftertheiroperationhas takenplace,withouttheneedforexpensiveintensivecaretreatment.Byelectronically linking both facilities, the most cost-effective treatment process and optimal time of transfertotherehabilitationfacilitycanbechosenwithoutlossofqualityduetoinformationlags. Thearchitectureofhospitalinformationsystemsmusttakethesedevelopmentsinto account.Theymustbeabletoprovideaccessortoexchangepatient-relatedandgeneral data (e.g., about the services offered in the hospital) across its institutional boundaries. Alotoftechnicalandlegalissueshavetobesolvedbeforecomputer-basedtransinstitutionalhealthinformationsystemswilladequatelysupporttransinstitutionalpatientcare. For example, a general willingness to cooperate with other health care providers must exist;optimalcareprocessesmustbedeined,andrecentbusinessprocessesberedesigned; accounting and inancing issues must be regulated; questions of data security and data conidentialitymustbeanswered,togetherwithquestionsondataownership(patientor institution)andonresponsibilitiesfordistributedpatientcare;issuesonlong-termpatient records (centralized or decentralized) must be discussed; and technical means for integrated,transinstitutionalinformationprocessingmustbeoffered(telemedicine,eHealth), includinggeneralcommunicationstandards.
38
4
Health Information Systems
4.4 Electronic Health Records as a Part of Health Information Systems Themostimportantenterprisefunctionsinhealthcarearerelatedtodiagnosticsandtherapy.Obviously,datathatarerelevanttomedicaldecisionmakingneedtobecollectedand presentedinapatientrecord. Apatientrecordingeneraliscomposedofalldataanddocumentsgeneratedorreceived during the care of a patient at a health care institution. Nowadays, many documents in the paper-basedpatientrecordarecomputerprintouts,suchaslaboratoryresults,ordischargesummariestypedintoatextprocessingsystem.Theportionofdocumentscreatedincomputerbasedformwillfurtherincrease.Thus,itseemsnaturaltostriveforapatientrecordthatispartly orcompletelystoredonelectronicdocumentcarriers:theelectronichealthrecord(EHR). Theelectronichealthrecord(EHR)isthecollectionofmedicaldatarelatingtoonesubject ofcare,i.e.,thepatient,thatisstoredinthecomputer-basedpartofahealthinformationsystem.“TheEHRforasubjectofcaremightbescatteredphysicallyacrossmultiple(discreteor interconnected)clinicalsystemsandrepositories,eachofwhichwillholdandmanageapartialEHRforeachofitsdatasubjects,scopedaccordingtotheserviceorcommunitysettings, clinicaldomainsandtimeperiodsofuseofthatsysteminthelifeofeachperson.”1 Primarily, EHRs are used to support patient care by providing relevant information about a patient whenever and wherever it is needed. Furthermore, they are needed for administrativefunctions,suchasbillingandqualitymanagement. Inthepast,EHRusedtobeprovider-centric,i.e.,theyonlycontainedpatientinformationthatwasrecordedinoneinstitution,forexample,inahospitalorinaphysician’sofice. ThoseEHRareusuallycalledelectronicpatientrecords(EPR).Hence,potentiallyrelevant informationaboutthemedicalhistoryofapatientthatwasrecordedinotherinstitutions wasmissingorhadtoberecordedagain.Thisledtoqualityandeficiencyproblems. Althoughthissituationcanstillbefoundinmanyinstitutions,effortsarebeingmade today to organize EHRs patient-centric, i.e. independent of institutional boundaries. To achievethevisionofacompleteandlifetime-spanningEHRwhichsupportshealthcareon theonehand,butrespectslegalandethicalissuesontheotherhand,differentstrategies canbefound.ThesearedescribedinSect.7.3.2.
4.5 Challenges for Health Information Systems Inspiteofthepositivegeneraldevelopmentofhealthinformationsystems,challengescan beidentiiedthathavetoberesolved.Evolutionarygrowninformationsystems(remember thathospitalshaveaninformationsystemfromthebeginningoftheirexistence)consistof avarietyofcomponentsandtendtobeveryheterogeneous.Asaconsequence,majorchallengesexist.Inanticipationofconceptsthatareintroducedlaterinthisbook,wewantto sensibilizeyouforthefollowingchallenges: StandardISO/DraftInternationalStandard18308
1
4.5 Challenges for Health Information Systems
39
• Thechallengeofuseracceptance.Healthcareprofessionalsorhospitalmanagers
asusersofheterogeneoushealthinformationsystemsmayhavetheneedtooverview a broad variety of data. They will have dificulties or will at least have the problemofhavingtouseasetofapplicationcomponents,oftenwithdifferentuser interfaces,overlappingfeatures,andseparateuseridentiicationprocedures.Thisis time-consuming and potentially dangerous for the patient, as important data may not be available when needed, leading to wrong diagnostic or therapeutic decisions. • Thechallengeofdataredundancy.Asdifferenthealthcareprofessionalsoftenneed thesamedata(seeSect.2.3),heterogeneousinformationsystemstypicallyleadtodata duplication:Relevantdatamaybedocumentedseveraltimesatdifferentsitesandorby differentproviders.Thisdouble-documentationistime-consumingforstaffandpatients anderror-prone,asdocumenteddatamaybeinconsistentorincomplete.Inaddition, uncontrolledredundancycausesconsiderableadditionalmaintenancecostsforupdatingduplicateddatain(redundant)databases. • Thechallengeoftranscription.Inheterogeneousarchitecturesthereisaconsiderable amount of transcription, i.e., the transfer of data from one storage device to another (e.g., the transfer of a patient’s diagnoses from the patient record to an order entry form).Mediacracks,i.e.thechangeofthestoragemediaduringthetranscriptionof data,areoftenthecauseoferrors.Bothmaydecreasethequalityofdataandbothdo increasecostsaswellasthehealthcareprofessional’stimeneededforrecordingand accessingpatientdata. • Thechallengeofmaintainingreferentialintegrity.Forredundantdata,eitherasreplicatesorevenasduplicates,itisdificulttoobtainandlatermaintainreferentialintegrity,i.e.,thecorrectassignmentofentities,forexampletheassignmentofdatatoa certainpatient(seeSect.6.5.2.2). • Thechallengeofcosts.Toohigh,inparticularuncontrolledredundancycausesconsiderable additional maintenance costs for updating replicate data in (redundant) databases. • Thechallengeofprivacyandsecurity.Patients’healthdatabelongtothemostsensitivedataabouthumans.Forthisreason,individualpatientdatamustonlybeaccessible tothosepersonspreviouslyauthorizedbythepatient.Ahealthinformationsystemmust guaranteethisclaimwhichiscalledprivacyofpatientdata.Likewise,theinformation system has to ensure data security which comprises availability, conidentiality and integrityofpatientdata. Intransinstitutionalhealthinformationsystems,theproblemofheterogeneitycanbeestimatedevenhigher.Sincethesesystemsinvolvemanyoriginallyautonomousinformation systems,someadditionalchallengescanbeidentiied: • Thechallengeofterminology.Havingdatastoredindifferentdatabases(andwithout auniqueandcomprehensivedatamodelordatadictionary)atdifferentsites,thereisno immediateneedforaunifiedterminologyandsemantics.Whenstartingtocommunicatebetweentheseapplicationssystems,thiscan,however,causesevereproblems,and anagreementonterminologyforcommunicatinginformationondiagnosesandproceduresisneeded.
40
4
Health Information Systems
• Thechallengeof stability. Down times of one or more computer-based application
systemswiththeconsequenceofsigniicantlyrestrictingcareprocessesmaybeseldom, but they exist. The same holds for problems in accessing the non-computer-based record.Butinatransinstitutionalorevenregionalcontext,riskofinstabilityiscomparativelyhigh. • Thechallengeoftransinstitutionalinformationmanagement.Nowadays,inmany hospitals,theoperational,tactical,andstrategicinformationmanagementisorganized professionally(seeSect.9.3).Dedicated,specializedstaffistakingcareoftheinformationsystem.Forinformationprocessingandstoringwithinhealthcarenetworks,there isoftennospecificpersonorgroupbeingresponsibleandhavingauthoritytodecide andact.
4.6 Example 4.6.1 Architecture of a Hospital Information System Hereisanextractofthedescriptionofthearchitectureofthehospitalinformationsystem ofthePlötzbergMedicalCenterandMedicalSchool(PMC).Asmentioned,PMCisaictitiousinstitution,whichwillbeusedinexamplesandexercisesinthisbook. ThehospitalinformationsystemofPMCenablesthehospitalfunctionsofpatientcare with patient admission, decision making and patient information, medical and nursing careplanningorderentry,executionofdiagnostic,therapeutic,andnursingprocedures, codingofdiagnosesandprocedures,patientdischargeandtransfertootherinstitutions. Inaddition,supplyanddisposalmanagement,schedulingandresourceallocation,hospitaladministration,andhospitalmanagementaresupported. Thosehospitalfunctionsaresupportedbysomebiggerandoverahundredsmaller applicationcomponents(partlycomputerbased,partlynon-computerbased).Themost important application component is the patient administration system, the computerbasedapplicationcomponentthatsupportspatientadmissionandpatientdischargeand transfertootherinstitutions.Inaddition,severalcomputer-baseddepartmentalapplicationcomponentsareusedforworkorganizationandresourceplanning(e.g.,intheradiology department, in the laboratory department, and in outpatient units). Nearly all computer-based application components are interconnected, using a communication server. Some computer-based application components are isolated systems without interfaces. Non-computer-based application components are used for special documentation purposes(e.g.,documentationinoperationrooms),andfororderentry.
4.7 Exercises
41
Theapplicationcomponentsareinstalledonphysicaldataprocessingsystems.Ascomputer-basedphysicaldataprocessingsystems,approximately40applicationanddatabase serversareoperated,andover4,000personalcomputersareused.Over1,000printersof differenttypesareinstalled.Mostcomputer-basedphysicaldataprocessingsystemsare interconnectedbyahigh-speedcommunicationnetwork. Asnon-computer-basedphysicaldataprocessingsystems,over2,000telephones and800pagersareused.About1,500differenttypesofpaper-basedformsareused to support different tasks. More than 400,000 patient records are created and used eachyear,andadozenlocalarchivesareresponsibleforpatientrecordarchiving.A paper-basedmailingsystemallowsfornon-computer-basedcommunicationbetween departments.
4.7 Exercises 4.7.1 Hospital Information System as a System Asintroduced,asystemcanbedeinedasasetofpeople,things,and/oreventstogether withtheirrelationshipsthatformsanintegratedwhole.Whichpeople,things,oreventscan youindwhenlookingatahospitalinformationsystem?Inwhatrelationshipdotheystand tooneanother?Tosolvethisexercise,takeintoaccountthecomponentsofhospitalinformationsystemsasdeinedinSect.3.3.3.
4.7.2 Buying a Hospital Information System LookatthedeinitionofhospitalinformationsystemsinSect.4.2.Basedonthisdeinition, isitpossibletobuyahospitalinformationsystem?Explainyouranswer.Whatdovendors ofhospitalinformationsystemsthusreallysell?
4.7.3 Transinstitutional Health Information Systems Patient-centered (not just institution-centered) care is obviously playing a major role. Mention the challenges, which transinstitutional health information systems are facing. Trytoarguefromyourpointofview,whythesechallengesaregreaterthanthesameones withinonehospital.
42
4
Health Information Systems
4.8 Summary Informationsystemsthataredealingwithprocessingdata,information,andknowledgein healthcareenvironmentsarecalledhealthinformationsystems. Healthinformationsystemcanbedifferentiatedasinstitutionalhealthinformationsystem,forexample,hospitalinformationsystems,andtransinstitutionalhealthinformation systemsthatspanthebordersoftwoormorelegallyseparatedinstitutions.Transinstitutional healthinformationsystemsplayavitalroleinsupportingintegratedcare. Importantchallengesofhealthinformationsystemsare: • thechallengeofuseracceptance, • thechallengeofdataredundancy, • thechallengeoftranscription, • thechallengeofmaintainingreferentialintegrity, • thechallengeofcosts, • thechallengeofprivacyandsecurity. In transinstitutional health information systems, some additional challenges can be identiied: • thechallengeofterminology, • thechallengeofstability, • thechallengeoftransinstitutionalinformationmanagement. Theelectronichealthrecord(EHR)isthecollectionofmedicaldatarelatingtoonepatient thatisstoredinthecomputer-basedpartofahealthinformationsystem.EHRsareneeded tosupportfunctionsofpatientcareaswellasforadministrativefunctions.
Modeling Health Information Systems
5
5.1 Introduction Modeling HIS is an important precondition for their management: What we cannot describe,weusuallycannotmanageadequately. After deining the concepts necessary for dealing with this chapter, we will present sometypesofinformationsystemmodelsdescribingdifferentaspectsofHIS.Wewillthen focus on the so-called “three-layer graph-based metamodel” (3LGM²), which has been developed for describing, evaluating and planning health information systems. 3LGM² models and their application for information management will play an important role throughout this book. Finally, we introduce reference models and a speciic reference modelthatfacilitatesmodelinginformationsystemsinhospitals. Afterreadingthissection,youshouldbeabletoanswerthefollowingquestions: • Whataremodels,metamodelsandreferencemodels? • WhataretypicalmetamodelsformodelingvariousaspectsofHIS? • Whatis3LGM²? • WhataretypicalreferencemodelsforHIS?
5.2 On Models and Metamodels When dealing with systems in general and with HIS in particular, we need models of systems.
5.2.1 Definitions Deinition5-1:Model Amodelisadescriptionofwhatthemodelerthinkstoberelevantofasystem.
A.Winteretal.,HealthInformationSystems, DOI:10.1007/978-1-84996-441-8_5,©Springer-VerlagLondonLimited2011
43
44
5 Modeling Health Information Systems
Inthesciences,modelscommonlyrepresentsimpliieddepictionsofrealityorexcerpts ofit.Modelsareadaptedtoanswercertainquestionsortosolvecertaintasks.Models shouldbeappropriatefortherespectivequestionsortasks.Thismeansthatamodelis only“good”whenitisabletoanswersuchaquestionorsolvesuchatask.Forexample, amodelthatonlycomprisesthepatients(andnotthenurses)ofawardcannotbeusedfor nurse stafing and shift planning. Since we are dealing with information management, thismeansthatmodelsshouldpresentasimpliiedbutappropriateviewofaHISinorder tosupportinformationmanagement.Examplesofrespectivequestionstobeanswered couldbe: • WhichhospitalfunctionsaresupportedbyaHIS? • Whichinformationprocessingtoolsareused? • Whatinformationisneeded,ifapatientshallbeadmittedtothehospital?Whatinformationcanbeprovidedafterwards? • Whatwillhappenifaspeciicserverbreaksdown? • Howcanthequalityofinformationprocessingbejudged? Thebetteramodelassistsitsusers,thebetterthemodelis.Thus,theselectionofamodel dependsontheproblemsorquestionstobeansweredorsolved. A large number of different classes of models exists. Each class of models is determinedbyacertainmetamodel.Ametamodelcanbeunderstoodasalanguageforconstructingmodelsofacertainclassandaguidelineforusingthelanguage. Deinition5-2:Metamodel Ametamodelisamodelingframework,whichconsistsof • modelingsyntaxandsemantics(theavailablemodelingconceptstogetherwiththeir meaning),i.e.themodelinglanguage, • therepresentationoftheconcepts(howtheconceptsarerepresentedinaconcrete model,e.g.,inagraphicalway), • and(sometimes)themodelingrules(e.g.,themodelingsteps),i.e.theguidelinefor applyingthelanguage.
JustasdifferentviewsonHISexist,therealsoexistvariousmetamodels.Typicaltypesof metamodelsforHISare: • functionalmetamodels,focusingonhospitalfunctionsthataresupportedbytheinformationsystem; • technicalmetamodels,whichareusedtobuildmodelsdescribingtheinformationprocessingtoolsused; • organizationalmetamodels,whichareusedtocreatemodelsoftheorganizationalstructureofHIS; • datametamodels,whichareusedforbuildingmodelsofthestructureofdataprocessed andstoredinsideaHIS;
5.2 On Models and Metamodels
45
• businessprocessmetamodels,whichfocusonthedescriptionofwhatisdoneinwhich chronologicalandlogicalorder;
• information system metamodels, which combine different metamodels into an integrated,enterprise-wideviewoninformationprocessinginaninstitution.
TheartofHISmodelingisbasedontherightselectionofametamodel.Thus,forHIS modeling,youshouldconsiderthefollowingsteps: 1. DeinethequestionsortaskstobesolvedbytheHISmodel. 2. Selectanadequatemetamodel. 3. Gathertheinformationneededformodeling. 4. Createthemodel. 5. Analyzeandinterpretthemodel(answeryourquestions). 6. Evaluateiftherightmetamodelwaschosen,thatis,ifthemodelwasadequatetoanswer thequestions.Ifnot,returntostep2.
5.2.2 Types of Models Dependingonthetypeofmetamodelused,modelscanbearrangedaccordingtodifferent types.
5.2.2.1 Functional Models Functionalmodelsrepresenttheenterprisefunctionsofaninstitution(whatistobedone). Inahospital,theirelementsarethehospitalfunctionsthataresupportedbytheapplication componentsofthehospitalinformationsystem.Therelationshipsofthehospitalfunctions can,forexample,representtheinformationexchangedbetweenthem.Inaddition,enterprisefunctionsareoftendescribedinahierarchicalway,comprisingmoreglobalenterprise functions (such as patient administration) and more speciic (reined) enterprise functions(suchasadministrativedischargeandbilling). Typicalquestionstobeansweredwithfunctionalmodelsare: • Whichenterprisefunctionsarerelevantinagiveninstitution? • Whichspeciicenterprisefunctionsarepartofwhichglobalenterprisefunctions? • Whichenterprisefunctionssharethesamedata? Typicalrepresentationsoffunctionalmodelsare(hierarchical)listsofenterprisefunctions, aswellasgraphicalpresentations.Table5.1presentsanextractfromathree-levelhierarchyofhospitalfunctions:
46
5 Modeling Health Information Systems
Table 5.1 An extract from the functional HIS model, describing hospital functions relevant for patient care and hospital administration at the Plötzberg Medical Center and Medical School (PMC) Patientcare
Orderentry
Preparationofanorder Appointmentscheduling
Patientdischarge andtransfertoother institutions
Administrativedischargeandbilling
Patient administration
Patientidentiicationandcheckingfor recurrent
Medicaldischargeandmedicalreportwriting Nursingdischargeandnursingreportwriting
Hospital administration
Administrativeadmission Visitorandinformationservice Information Management
Strategicinformationmanagement Tacticalinformationmanagement Operationalinformationmanagement
5.2.2.2 Technical Models Technical models describe the information processing tools used. As concepts, they typicallyprovidephysicaldataprocessingsystems(e.g.,computersystems,telephones, forms, pagers, records) and application components. As relationships, they provide the datatransmissionbetweenphysicaldataprocessingsystems(e.g.,networkdiagrams),or thecommunicationbetweenapplicationcomponents. Typicalquestionsthatcanbeansweredwithtechnicalmodelsare: • Whichinformationprocessingtoolsareused? • Whichapplicationcomponentscommunicatewitheachother? • What are the data transmission connections between the physical data processing systems? • Whatdoesthenetworktechnologylooklike? • Whattechnicalsolutionsareusedtoguaranteethesecurityandreliabilityofinformationprocessingtools? Technicalmodelsaretypicallypresentedaslists(e.g.,listsofinformationprocessingtools used)orasgraphs(e.g.,graphofthenetworkarchitectureofcomputersystems).Examples ofgraphicalmodelsarepresentedinFigs.5.1and5.2.
5.2.2.3 Organizational Models Organizational models describe the organization of a unit or area. For example, they may be used to describe the organizational structure of a hospital (e.g., consisting of
5.2 On Models and Metamodels
Clinical Appliction Server Cluster
47
IN
CIS printer
LIS inbox
CWP outbox
LIS form reader
CIS PC1
Application server 1
Application server 2
LIS PC2 Switch A4
ComServ Hardware
ComServ 1
LIS Application Server LIS PC3
ComServ 2 Switch A3
Switch X34 PC4
ComServ RAID
RIS Application Server
Switch X35
PACS Application Server
PC5
PC6
RIS-Gateway-Server
PC7 PACS DB Cluster PACS DB Server1
PACS DB Server2
RIS-DB-Cluster RIS DB Server1
PC8
RIS-DB-Server2
Fig. 5.1 AnextractofatechnicalHISmodelwithsomephysicaldataprocessingsystemsandtheir datatransmissionlinksofthehospitalinformationsystemofthePlötzbergMedicalCenterand MedicalSchool(PMC)
Modalities
Snail Mail System
Clinical Documentation and Management System (Clinix)
Radiology Information System (Radix)
Communication Server (Comix)
PACS (Pix)
Laboratory Information System (Lix)
Fig. 5.2 AnextractofatechnicalHISmodelwithsomeapplicationcomponentsandtheircommunication links of the hospital information system of the Plötzberg Medical Center and Medical School(PMC)
48
5 Modeling Health Information Systems
Hospital Executive Commitee
Dept. of Internal Medicine
Paediatrics surgery
Dept. of Surgery
Dept. of Information Management
General surgery
Inpatient unit A
Dept. of Pathology
Hospital Administration
Patient management
Financial accounting
Facility management
Outpatient unit B
Fig. 5.3 AnextractfromtheorganizationalmodelofPlötzbergMedicalCenterandMedicalSchool (PMC)
departmentswithinpatientandoutpatientunits).InthecontextofHIS,theyareoften usedtodescribetheorganizationofinformationmanagement,thatis,howitisorganized inordertosupportthegoalsofthehospital. Theconceptsofthosemodelsareusuallyunitsorrolesthatstandinacertainorganizational relationship to each other. Typical questions to be answered with organizational modelsare: • Whichorganizationalunitsexistinahospital? • Whichinstitutionsareresponsibleforinformationmanagement? • Whoisresponsibleforinformationmanagementofagivenareaorunit? Organizationalmodelsaretypicallyrepresentedasalistoforganizationalunits(e.g.,list ofthedepartmentsandsectionsinahospital),orasagraph(e.g.,graphicaldescriptionof theorganizationalrelationships).AnexampleispresentedinFig.5.3.
5.2.2.4 Data Models Datamodelsdescribethedataprocessedandstoredinaninformationsystem.Theirconcepts are typically entity types (compare sect. 5.3.2.1) and their relationships. Typical questionstobeansweredwithdatamodelsare: • Whatdataareprocessedandstoredintheinformationsystem? • Howaredataelementsrelated? A typical metamodel for data modeling is offered by the class diagrams in the Uniied ModelingLanguage(UML).1AnexampleispresentedinFig.5.4.
1
ObjectManagementGroup(OMG):UnifiedModelingLanguage–UML.http://www.uml.org
5.2 On Models and Metamodels
49
Case
Patient -identification number -name -birthday -address
Procedure
-identification number -insurance 1
0..
Inpatient -admission date -date -ward identification
1
0..
-identification number -type -date -provider
Outpatient -treatment date -clinic identification
Fig. 5.4 Asimpliieddatamodel(UMLclassdiagram),describingtherelationshipsbetweenthe entitytypespatient,case,andprocedure,asextractfromthedatamodeloftheHISofthePlötzberg MedicalCenterandMedicalSchool(PMC)
5.2.2.5 Business Process Models Businessprocessmodelsfocusonadynamicviewofinformationprocessing.Theconceptsusedareactivitiesandtheirchronologicalandlogicalorder.Often,otherconcepts areadded,suchastheroleorunitthatperformsanactivity,ortheinformationprocessing toolsthatareused.Thefollowingperspectivesusuallycanbedistinguished: • Functionalperspective:Whatactivitiesareperformed,andwhichdataflowsareneeded tolinktheseactivities? • Behavioralperspective:Whenareactivitiesperformed,howaretheyperformed,and wherearetheyperformed?Dotheyusemechanismssuchasloopsandtriggers?What mechanismstriggerthestartoftheoverallprocess? • Organizationalperspective:Whereandbywhomareactivitiesbeingperformed?Which differentrolesparticipateintheactivities? • Informationalperspective:Whichentitytypesorentities(documents,data,products) arebeingproducedormanipulated?Whichtoolsareusedforthis? Typicalquestionstobeansweredwithbusinessprocessmodelsare: • Whichactivitiesareexecutedwithregardtoagivenenterprisefunction? • Whoisresponsibleandwhichtoolsareusedinagivenprocess? • Whichactivityisthepre-orpostconditionforagivenactivity? • Whataretheweakpointsofthegivenprocessandhowcantheybeimproved? Duetothenumberofdifferentperspectives,variousbusinessprocessmetamodelsexist. Examplesaremetamodelsforsimpleprocesschains,event-drivenprocesschains,activity diagrams,andPetrinets. Simpleprocesschainsdescribethe(linear)sequenceofprocesssteps.Theydescribethe speciicactivitiesthatformaprocess,inadditiontotheresponsiblerole(e.g.,aphysician).
50
5 Modeling Health Information Systems
Event-drivenprocesschainsadddynamicpropertiesofprocesssteps:eventsandlogicaloperators(and,or,xor)areaddedtotheenterprisefunctions,allowingthemorecomplexmodelingofbranchingandalternatives.Inaddition,someinstancesofevent-driven processchainsallowtheadditionofentitytypes(e.g.,achart).2 Activitydiagrams(aspartofthemodelingtechniqueoftheUniiedModelingLanguage, UML)alsodescribethesequenceofprocesssteps,usingactivities,branching,conditions, andentitytypes(Fig.5.5).Inaddition,themethodallowsforsplittingandsynchronization ofparallelsubprocesses.3
physician
secretary
administrative staff [patient calls to get an appointment]
answer call
check need for admission
check need for admission
foward call to physician
[necessity unclear]
[not necessary] finish call
[necessary] arrange date
[not necessary] check if it is patient’s first admission
inform administration finish call
note [no]
find old patient record
[yes]
prepare new patient record
[patient has appointment]
Fig. 5.5 Exampleofabusinessprocessmodel,basedonaUMLactivitydiagram,describingapart oftheadmissionprocessintheDepartmentofChildandJuvenilePsychiatryatPlötzbergMedical CenterandMedicalSchool(PMC)
ScheerAW.ARIS–BusinessProcessModeling.Berlin:Springer;2000. ObjectManagementGroup(OMG):UnifiedModelingLanguage–UML.http://www.uml.org
2 3
5.3 A Metamodel for Modeling Health Information Systems on Three Layers: 3LGM²
51
Finally,Petrinetsdescribethedynamicpropertiesofprocessesinamoreformalway thantheothermethods.4
5.2.2.6 Information System Models Informationsystemmodelscompriseallmodelingaspectsdiscussedsofar,thatis,functionalmodeling,technicalmodeling,organizationalmodeling,datamodeling,andprocess modeling.Butbeyondthis,informationsystemmodelsconsiderthedependenciesofthese modelsand,therefore,offeramoreholisticview. Typicalquestionstobeansweredwithenterprisemodelingare: • Whichhospitalfunctionsaresupportedbywhichinformationprocessingtools? • Aretheinformationprocessingtoolssuficienttosupportthehospitalfunctions? • Isthecommunicationamongtheapplicationcomponentssuficienttofulilltheinformationneeds? • Whichaimsoftheenterprisewillbeaffectedbyacertainapplicationcomponent? • Inwhichareaoftheinstitutionarespeciicdataonspeciicobjectsused?
5.3 A Metamodel for Modeling Health Information Systems on Three Layers: 3LGM² Letusnowintroduceaspeciicinformationsystemmetamodelformodelinghealthinformationsystems.Thismetamodeliscalledthethree-layergraph-basedmetamodel(3LGM²). ItaimstosupportthesystematicmanagementofHISandespeciallythestructuralquality assessmentofinformationprocessinginhealthcareinstitutions.Wewillusethismetamodelfurtheroninthisbook. Typicalquestionstobeansweredwithmodelsderivedfromthe3LGM²metamodelare: • Whichhospitalfunctionsaresupported? • Whichinformationisneededorupdatedwhenperformingahospitalfunction? • Whichapplicationcomponentsareused,andhowdotheycommunicate? • Whichphysicaldataprocessingsystemsareused? • Whichhospitalfunctionsaresupportedbywhichapplicationcomponent? • Which application components are installed on which physical data processing systems? • Whatistheoverallarchitectureofthehospitalinformationsystem? 3LGM²combinesfunctional,technical,organizational,andsomeaspectsofbusinessprocessmetamodels.ItisrepresentedinUMLnotation.Asthenameindicates,the3LGM² distinguishesthreelayersofinformationsystems: • domainlayer; • logicaltoollayer; • physicaltoollayer. 4
MortensenKH,ChristensenS,editors.PetriNetsWorld.http://www.daimi.au.dk/PetriNets
52
5 Modeling Health Information Systems
Therearethreevariantsofthe3LGM²:3LGM²-B,3LGM²-Mand3LGM²-S.Themain differencebetweenthevariantsistheprovisionofdifferentconceptsfordescribingthe communication between application components in an information system. 3LGM²-B, whichisthebasisfor3LGM²-Mand3LGM²-S,consistsofconceptsdescribingbasicelementsofaHISarchitecture.3LGM²-Mextends3LGM²-Bbyconceptsformodelingmessage-based communication. For architectures in which computer-based application componentsprovideservicestobeusedbyothercomputer-basedapplicationcomponents (so-called service-oriented architectures), the concepts of 3LGM²-S are useful for modeling. Thefollowingsectionsprovidethetheoreticalbackgroundforthefurtheruseof3LGM² inthisbook.Section5.3.1introducesthoseUMLconceptsneededforunderstandingthe descriptionof3LGM²inUMLnotation.Section5.3.2explains3LGM²-Bwithitsbasic conceptsofthethreelayersof3LGM².Sections5.3.3and5.3.4introducetheadditional conceptsof3LGM²-Mand3LGM²-S.
5.3.1 UML Class Diagrams for the Description of 3LGM² The metamodel 3LGM² is formalized with the help of class diagrams according to the Uniied Modeling Language (UML). For a more thorough understanding of UML, you mayrefertoarespectivetextbook,butthenecessarybasicsareexplainedinthissection. 3LGM² class diagrams consist of classes, abstract classes and associations between classes. Every 3LGM² concept is described by a class. Classes contain the attributes and methodsofasetofsimilarentities.Everyclassisgivenanunambiguousclassname. For example, computer-based application component is a class of the 3LGM² metamodel.Inaconcrete3LGM²model,weuseinstancesofclassesformodeling.Apatient administration system could be an instance of the class computer-based application component. AsintroducedinSect.3.3.3,bothcomputer-basedapplicationcomponentsandnoncomputer-based application components are application components. In 3LGM², we thereforerefertoapplicationcomponentasanabstractclass,becausewecanbuildno instances of application component in a concrete model. Instead, we have to use its 3LGM²subclassescomputer-basedapplicationcomponentornon-computer-basedapplicationcomponentsforbuildinginstances.Thenamesofabstractclassesarewrittenin italics in UML class diagrams; arrows relate the subclasses to the abstract class (seeFig.5.6). Associationslinkclasses.Forexample,applicationcomponentislinkedwithcomponent interface by the association “has” in 3LGM² (see Sect. 5.3.2.2). Associations are representedbylinesbetweenclassesandareoftennamed.Atthetwoendsofanassociationline,themultiplicityspeciieshowmanyinstancesofaclasscanbelinkedwithhow manyinstancesoftheoppositeclass(seeFig.5.7). If it is necessary to specify an association by attributes or by associations to other classes, an association class is used (see Fig. 5.8). In 3LGM², we use for example the
5.3 A Metamodel for Modeling Health Information Systems on Three Layers: 3LGM²
53
associationclass‘support’tofurtherspecifytherelationshipsbetweenenterprisefunctions andapplicationcomponents(seeSect.5.3.2.4). Inthe3LGM²metamodel,weoftenuseassociationsbetweenoneclassanditselfin ordertobeabletosub-orsuperordinatecertaininstancesofthisclass.Forexample,both
Class -attributes
Sub class 1
Sub class 2
-attributes
-attributes
Fig. 5.6 AbstractclassesandsubclassesinaUMLclassdiagram
Class 1
Class 2
-attributes
-attributes 1
1..*
Fig. 5.7 AssociationsbetweenclassesinaUMLclassdiagram
Association class -attributes
Class 1
1
-attributes
Fig. 5.8 AssociationclassesinaUMLclassdiagram
1..*
Class 2 -attributes
54
5 Modeling Health Information Systems
theoperationtheatrewithroomnumberOP1andthebuildingB123areinstancesofthe sameclassoflocationsinacertainhospital.ButsinceroomOP1islocatedinB123itcan besubordinatedtothatbuilding.Inasimilarbutnotidenticalwaytheenterprisefunction ‘executionofradiologicalprocedures’canbesubordinatedtoexecutionofdiagnosticand therapeuticprocedures. Astheseexamplesshow,itishelpfultodistinguishbetweentwodifferentmeaningsof subordinationin3LGM²models:decompositionandspecialization. Decompositionsdescribepart-ofrelationshipsaswithOP1andB123.Therespective superordinationiscalledacomposition.SeeFig.5.9foritsdescriptioninaUMLclass diagram. Specializations describe reinement relationships as with ‘execution of radiological procedures’ in comparison to execution of diagnostic and therapeutic procedures. The respectivesuperordinationiscalledgeneralization.SeeFig.5.10tounderstandhowthisis describedinaUMLclassdiagram.Inspecializations,attributevaluesofthesuperordinated instance are inherited to the subordinated instance. For example, if execution of diagnosticandtherapeuticproceduresissupportedbyacertaincomputer-basedapplication component X, then ‘execution of radiological procedures’ is also considered to be supported by X since it is actually just one special way of execution of diagnostic and therapeuticprocedures.
Class
1
Fig. 5.9 Decompositionasonetypeofassociation betweenoneclassanditselfinaUMLclassdiagram
1..*
Class
1
Fig. 5.10 Specializationasanothertypeofassociation betweenoneclassanditselfinaUMLclassdiagram
1..*
5.3 A Metamodel for Modeling Health Information Systems on Three Layers: 3LGM²
55
5.3.2 3LGM²-B 5.3.2.1 Domain Layer Thedomainlayerof3LGM²describeswhatkindsofactivitiesinahealthcareinstitution areenabledbyitsinformation system and what kind of data should be stored and processed.Thislayerisindependentoftheimplementedinformationprocessingtools. Therearemyriadsofphysicalorvirtualobjectsinahealthcareinstitution(e.g.,patient John Doe, or epicrisis of patient Jane Smith dated from 2010-06-10). Data represent informationaboutthoseobjectsandarethevaluesofattributesoftheobjects.Objects beingsimilarwithrespecttotheirattributecategoriesaresummarizedasobjectclasses (e.g.,classpatientassetofallobjectswhicharehumanbeingsandreceivecareinahealth careinstitution). Deinition5-3:Entitytype Anentitytypeisarepresentationof 1. anobjectclassandof 2. thedatarepresentinginformationconcerningtheobjectsofthisobjectclass,ifthese dataarestoredorcouldorshouldbestoredintheinformationsystem. Weusethelabelofanobjectclassastermfortherespectiveentitytype.Forthesakeof simplicitywesay“dataaboutanentitytype”ifwethinkaboutthedatabeingrepresented bytheentitytype. Forexampleanentitytypepatientrepresents • theclassofallpatientsinacertainhospitaland • thedatadescribingthesepatientsliketheirname,birthday,homeaddress,socialsecuritynumber,height,weightetc.asitisstoredinacertaininformationsystem. Note,thatwemustnotconfusetheentitytypewiththerespectiveobjectclassandthedata describingitsobjects.Theentitytypeisonlytheirrepresentationorsurrogateinaninformationsystem’smodel. Informationprocessingactivitiesatacertaintimeandplaceinaninformationsystem usecertaindatainordertoupdateordeleteotherdata.Usingtheconceptentitytypewe cansay,thatanactivityinterpretsdataaboutcertainentitytypesandupdatesdataofcertain entitytypes.Againsimplifyingwesaythattheactivityinterpretsentitytypesandupdates entitytypes. IfwelookatenterprisefunctionsasintroducedinSect.3.3.3andfocusontheirinformationprocessingaspects,wecanthendeinethemformallyasfollows: Deinition5-4:Enterprisefunction Theclassofallactivitiesinterpretingthesamesetofentitytypesandupdatingthesame setofentitytypesiscalledaninformationprocessingenterprisefunction(short:enterprise function).Anenterprisefunctionisadirectiveinaninstitutiononhowtointerpretdata
56
5 Modeling Health Information Systems
about entity types and then update data about entity types as a consequence of this interpretation.Thegoalofdatainterpretationandupdatesispartoforcontributesto(sub) goalsoftheinstitution.Afunctionhasnodeinitivebeginningorend. Similartoanactivity,anenterprisefunctionissaidtointerpretentitytypesandupdate entitytypes. Forexample,ifDr.DoeplansclinicalcareforherpatientSmiththismorning(i.e.an activity),thiswillresultininterpretingdataaboutSmith’sdiagnosesandthehospital’s clinicalpathwaysandthenupdatingdataaboutconcreteplansformedicalornursingprocedures.Theclassofallsimilaractivitiesofalldoctorsandnursesmaybecalledmedical andnursingcareplanningandisanenterprisefunctionofahospital.Theenterprisefunction contributes signiicantly to the hospital’s goal of treating patients and updates the entitytypesmedicalprocedureandnursingproceduredependingontheinterpretedentity typesdiagnosisandclinicalpathway. Enterprisefunctionsandentitytypescanbestructuredhierarchicallybyspecializationanddecomposition.Whenanenterprisefunctionoranentitytypeisspecialized, allitssubelementsareafurtherreinementoftheenterprisefunctionortheentitytype andindependentoftherespectivesuperelement.Foranenterprisefunctionitmeans thattheactivitiesregardingthisenterprisefunctionareexecuteddifferentlyindifferentcontexts.Theenterprisefunctionexecutionofdiagnosticprocedures,forexample, has different specializations in different diagnostic departments. Similarly, an entity typecanhavedifferentformsforslightlydifferentpurposes:Aindingfromaradiologydepartmentisdifferentfromaindingfromalaboratory;butbotharespecializations of indings, which is the generalized term. By contrast, when an enterprise functionoranentitytypeisdecomposed,allitssubelementsformapropersubsetof theenterprisefunctionortheentitytype.Anactivityregardinganenterprisefunction isonlycompletedifallactivitiesregardingallitsdecomposedsubfunctionsarecompleted.Forexample,theactivitiesregardingpatientadmissionareonlycompletedif among others administrative admission, medical admission and nursing admission havebeenperformed.Adecomposedentitytypeisonlycompletewhenallitssubordinateentitytypesareavailable.Apatienthistory,forexample,mustcontainallclinical data about a patient, including medical and nursing anamnesis, indings, discharge summaryetc. Interpretingandupdatingrelationshipsbetweenenterprisefunctionsandentitytypes are inherited to their subelements, no matter whether the enterprise functions or entity typesweredecomposedorspecialized. Enterprisefunctionsareperformedincertainorganizationalunitsofhealthcareinstitutions. Organizational units like a radiology department can be decomposed, but not specialized. Deinition5-5:Organizationalunit An organizational unit is a part of an institution which can be deined by responsibilities.
5.3 A Metamodel for Modeling Health Information Systems on Three Layers: 3LGM²
57
Notethatenterprisefunctions,entitytypes,andorganizationalunitsarejustpartofa static view of a hospital. However, so-called information processes can be modeled by bringingenterprisefunctions,whichinterpretandupdateentitytypes,intoalogicaland chronologicalorder. Deinition5-6:Informationprocess Aninformationprocessisalogicalandchronologicalsequenceofenterprisefunctionswhichinterpretorupdatedataaboutentities. Incontrasttomostbusinessprocessmodels(seeSect.5.2.2.5),informationprocesses donotcontainabehavioralperspective. TheconceptsintroducedhereandtheirmutualassociationsareillustratedinFig.5.11 byusingUMLasexplainedinSect.5.3.1. A graphical representation of enterprise functions and entity types on the domain layerisshowninFig.5.12whereenterprisefunctionsarerepresentedasrectanglesand entitytypesarerepresentedasovals.Anarrowfromanentitytypetoanenterprisefunctionrepresentsaninterpretingaccess,fromanenterprisefunctiontoanentitytypean updatingaccess.Dashedarrowspointfromdecomposedelementstowardstheircomposition.Thismayalsobeillustratedbypositioningdecomposedelementsonorintotheir composition. Arrows with empty heads point from specialized elements to their generalization. Whichentitytypesandwhichenterprisefunctionsofaninformationsystemaremodeled,dependsonthehealthcareinstitutionandonthemodelingpurpose.Referencemodelsmayofferrecommendationsaboutimportantentitytypesandenterprisefunctionsfor certainkindsofhospitals(seeSects.5.4and5.5).
-has part 1
*
Entity type
-specializes
0..1 -is specialized by
* -is part of
* -interprets
-updates *
-is carried out through
Use
* -is interpreted by *
-is updated by * Information process
* -is executed in
Function * -executes -has part
*
* -carries out * -is specialized by
* 1
-is part of * -specializes
* -is used by Organizational unit
-has part * 0..1
Fig. 5.11 Conceptsofthe3LGM²DomainLayer
-uses
-is part of
58
5 Modeling Health Information Systems
Patient admission
Administrative admission
Patient
Medical admission
Medical anamnesis Diagnosis
Nursing admission
Medical and nursing care planning
Execution of diagnostic, therapeutic and nursing procedures
Nursing anamnesis
Clinical pathway Nursing procedure
Patient history
Medical procedure Order
Finding Execution of radiologic procedures
Execution of laboratory procedures
Fig. 5.12 Exampleofa3LGM²DomainLayer
5.3.2.2 Logical Tool Layer At the logical tool layer (UML metamodel, Fig. 5.13), application components (see Sect.3.3.3.)arethecenterofinterest. Deinition5-7:Applicationcomponent Anapplicationcomponentisasetofactuallyusablerules,whichcontroldataprocessingofcertainphysicaldataprocessingsystems.Rulesareconsideredtobeactually usable,iftheyareimplementedsuchthattheyarereadytosupportcertainenterprise functions in a certain enterprise or support communication between application components. Iftherulesareimplementedasexecutablesoftware,theapplicationcomponentiscalled computer-basedapplicationcomponent.Otherwiseitiscallednon-computer-based. Applicationcomponents,eithercomputer-basedornon-computer-based,supportenterprisefunctions.Asoftwareproductisasetofrules,representedbysoftwarebeingstored atacertainmediumbutnotimplementedandactuallyexecutableyet.Thisiswhatwecan buy.Acomputer-basedapplicationcomponentcannotbeboughtinashopanywaybuthas
5.3 A Metamodel for Modeling Health Information Systems on Three Layers: 3LGM²
59
Communication link
{Organizational system must not be part of Computer-based application component}
-is part of
Sending in terface
* -receives from
* -sends to
Receiving inter face
1
Application componen t * -has part
* -belongs to
Computer-based application componen t
*
* -has
Communication inte rface
Organizational system
-is basis for
* -bases on Software produc t
Fig. 5.13 Conceptsofthe3LGM²logicaltoollayer.Linesdenoteinterlayerrelationshipsbetween logicaltoollayeranddomainlayer
tobeconstructedbycustomizingabuyablesoftwareproductonsite.Non-computer-based applicationcomponents(synonym:organizationalsystems)arecontrolledbyconventional rules, which can be summarized as working plans describing how people use non- computer-baseddataprocessingsystems(seeSect.3.3.3.). Applicationcomponentsareresponsibleforthestorageandforthecommunicationof dataaboutcertainentitytypes.Applicationcomponentsusecommunicationinterfacesfor the communication among each other. A communication interface can either send or receivedataaboutentitytypes.Forcommunicationamongapplicationcomponents,communicationlinkscanbedeined.Acommunicationlinkconnectsacommunicationinterfaceofoneapplicationcomponentwithacommunicationinterfaceofanotherapplication componentandcommunicatesdataaboutacertainentitytype.Applicationcomponents havetocommunicatebyrespectiveuseoftheirinterfacestoensurethatenterprisefunctionscaninterpretandupdateentitytypesasdescribedatthedomainlayer.Application componentsmaybedecomposed. Applicationcomponentsofaninformationsystemareobjects,whichactuallycanbe experienced by staff members in an institution. But nevertheless, they are not tangible. Therefore,werefertoapplicationcomponentsalsoaslogicaltools.Consequently,wecall thelayerdescribingtheapplicationcomponentsthelogicaltoollayer.Thisisincontrastto thetangibletools,whichwerefertoasphysical(seeSect.5.3.2.3).
60
5 Modeling Health Information Systems
Patient administration system (PatAd)
Medical documentation system (Medix)
Nursing management and documentation system (NursDoc)
Communication server (Comix)
Organizational system for order entry
Radiology information system (Radix)
Laboratory information system (LabSys)
Fig. 5.14 Exampleofa3LGM²logicaltoollayer
Figure5.14showsanexampleofalogicaltoollayer.Inthisexample,theapplication componentsaredepictedaslargeroundedrectangles,andtherelationshipsbetweenthem viacommunicationinterfaces(smallcircles)aredepictedasarrows.Thedirectionofthe arrowsrepresentsthedirectionofthecommunication.Namesofsoftwareproductsbeing usedforacomputer-basedapplicationcomponentareputinparentheses.
5.3.2.3 Physical Tool Layer Thephysicaltoollayer(UMLmetamodel,Fig.5.15)isasetofphysicaldataprocessing systems. Deinition5-8:Physicaldataprocessingsystem A physical data processing system is a physically touchable object or a simulated physicallytouchableobjectbeingabletoreceive,store,forward,orpurposefullymanipulatedata.Wedenotereceiving,storing,forwardingandpurposefulmanipulationofdata asdataprocessing.Thisdataprocessingiscontrolledbyrules.
5.3 A Metamodel for Modeling Health Information Systems on Three Layers: 3LGM²
61
Data transmission connection Location
-is located at 0..1
-specifies
-submits data to * -houses * Physical data processing system
* -is of type
0..1 Tool class
1 -has part
* * -is virtualized by -is part of
-is submitted via *
* -submits * Transmitting medium -receives data from -signal based : bool * -virtualizes
* -is specialized by 0..1 -specializes
Fig. 5.15 Conceptsofthe3LGM²physicaltoollayer.Dottedlinesdenoteinterlayerrelationships betweenlogicaltoollayerandphysicaltoollayer
Therulesmentionedhereareidenticaltothosementionedinthedeinitionofanapplicationcomponent. Physicaldataprocessingsystemscanevenbehumanactors(suchaspersonsdelivering mail),non-computer-basedphysicaltools(suchasprintedforms,telephones,books,paperbased patient records, administrative stickers), or computer systems (such as terminals, servers,personalcomputers,switches,routers). Physicaldataprocessingsystemslikeaspeciicserveroraspeciicpersonalcomputer canbeassignedtoatoolclass(e.g.,server,personalcomputer)andalocation.Physical dataprocessingsystemsarephysicallyconnectedviaso-calleddatatransmissionconnections(e.g.,communicationnetwork,courierservice)whichcanusedifferenttransmitting media.Atransmittingmediumiseithersignal-based(e.g.,coppercable,opticaliber)or non-signal-based(e.g.,sheetofpaper,CD-ROM,memorystick). Physicaldataprocessingsystemscanbereinedbydecomposition(seeSect.5.3.1).A physicaldataprocessingsystemcanbepartofexactlyonephysicaldataprocessingsystem. Additionally,physicaldataprocessingsystemscanbevirtualized.Wespeakofvirtualizationwhen,either,oneormorephysicaldataprocessingsystemssimulateonephysical dataprocessingsystem,oronephysicaldataprocessingsimulatesoneormorephysical dataprocessingsystems.Intheirstcase,wecallthesimulatedphysicaldataprocessing systemacluster.Inthelattercase,wecallthesimulatedphysicaldataprocessingsystems virtual machines. In a cluster, for example, different servers could, depending on their capacity,alternativelyrunacertaincomputer-basedapplicationcomponent.However,the clustercanbeadministratedasone(virtual)server.Bycontrast,withthehelpofvirtual machines,differentoperatingsystemsordifferentinstancesofoneoperatingsystemcan berunononephysicalserver,forexample.Bothvirtualmachinesandclustersarecalled virtualphysicaldataprocessingsystems,sincetheycanbeusedasphysicaldataprocessingsystemsforexampletoimplementapplicationcomponents,butarenotphysicalinthe senseofbeingtouchablephysically.SeeFig.5.16fortherepresentationofvirtualization in3LGM²models.
62
5 Modeling Health Information Systems
Fig. 5.16 Onthetop,theconceptofaclustervirtualizingseveralphysicaldataprocessingsystems isillustrated.Atthebottom,thereisonephysicaldataprocessingsystemwhichisvirtualizedinto severalvirtualmachines
Figure5.17showsanexampleofaphysicaltoollayer.Datatransmissionconnections between physical data processing systems are depicted as lines. The igure also shows decompositions,virtualmachinesandclusters.Informationaboutlocationsandtoolclasses isnotrepresentedgraphically.
5.3.2.4 Interlayer Relationships Avarietyofdependencies,calledinterlayerrelationships,existamongcomponentsofdifferentlayers.Relationshipsexistbetweenconceptsofthedomainlayerandthelogicaltoollayer andbetweenconceptsofthelogicaltoollayerandthephysicaltoollayer(seeFig.5.18). Consideringthedomainlayerandthelogicaltoollayer,themostimportantrelationshipsarebetweenenterprisefunctionsandapplicationcomponents.Theserelationships arehandledbytheclassuseandtheassociationclasssupportin3LGM².Ifacomputerbasedapplicationcomponentisusedforacertainenterprisefunctiontherearetwopossibilities. Either the computer-based application component is immediately used for supporting the activities regarding an enterprise function (class use), or the computerbasedapplicationcomponentonlymediatestheuseofanothercomputer-basedapplication component which supports the enterprise function (class support). The answers to the
5.3 A Metamodel for Modeling Health Information Systems on Three Layers: 3LGM²
Virtual machine B.1
63
Virtual machine B.2
Printer Ward PC1 Server MN PAS server
out
Ward outbox
Ward PC2 Network CS server Mrs Doe RIS PC1
IN
Radiology inbox
Delivery system
RIS Cluster LIS PC2 LIS server
RIS database server
IN
Laboratory inbox
RIS application server Mrs Roe
Fig. 5.17 Exampleofa3LGM²physicaltoollayer
followingtwoquestionshelptounderstandhowtoassignenterprisefunctionstoapplicationcomponents. • Which application components are necessary to support an enterprise function completely? Ina3LGM²model,wespecializeordecomposeenterprisefunctionstothatlevelof detailneededtodescribethesupportoftheenterprisefunctionsbysingleapplication components. That means, if we think of the hierarchy of enterprise functions in a 3LGM²modelasatreeingraphtheory,theneachofthetree’s“leaffunctions”must completelybesupportedbyoneapplicationcomponentoftheinformationsystem.We onlyassignapplicationcomponentstothe“leaffunctions”ofthetree.Forexample,if weindthattheenterprisefunctionmedicalandnursingcareplanningneedsjointsupportoftwoapplicationcomponentsXandY,wehavetospecializeordecomposethe enterprisefunctioninaway,thattheresultingsubfunctionsaresupportedbyXandY respectively.IfXisusedbycliniciansandYisusedbynursesitcouldbeasolutionto decomposetheenterprisefunctioninto‘medicalcareplanning’and‘nursingcareplanning’. If X is used in department D1 and Y in department D2, a specialization into ‘medicalandnursingcareplanninginD1’and‘medicalandnursingcareplanningin D2’couldbeasolution.Thenextparagraphwillhelptoidentifytheexample’slatter caseasafunctionalredundancywhichmightnotbeexpressedbyspecializingclinical andnursingcareplanningbutbyassigningD1andD2asalternative/redundantapplicationcomponentssupportingtheenterprisefunction.
1
*
* -is received by * -sends * -is sent by
-can support
Software product
*
-bases on
*
-has part
* -specializes
-specializes
* -uses
*
-is carried out through * Use -supports
0..1
*
Support
-is supported by
-is saved by
-is master : boolean
Storage
*
0..1 -is part of
-is basis for
*
-is part of
Communication interface
* -sends to
* -can receive Receiving interface
{Organizational system must not be part of Computer-based application component}
* -has
Organizational system
1 * -has part
-is needed for -is needed functionality * for saving Application component * -belongs to
*
* -is sent by
-signal based : boolean
Communication link -can send * Sending interface * -sends from *
* -is submitted via
-submits * Transmitting medium
Data transmission connection
-submits -receives * -houses data to * * data from -specifies * Physical data processing system * -virtualizes -virtual : boolean -is part of * -has part 1 -needs -needs for * * * -is virtualized by functionality saving
* -is used by -is used by * Organizational unit -uses * Computer-based application component
* -carries out * -is specialized by
-is interpreted by
-saves -interprets
-can be * supported by
*
Function
*
*
-specializes * Entity type
-is of type 0..1
-is located at 0..1
Fig. 5.18 CompleteUMLdiagramofthe3LGM²-Bmetamodelshowingallconcepts,intra-andinterlayerrelationships
{Function m (m elem N) interprets Entity type, which is updated by Function n:n elem N and n0beasetofenterprisefunctions,and AC := { AC1 ,..., ACn }, N>0asetofapplicationcomponents. Thetwo-dimensionalmatrixSUPdescribingtherelationshipbetweenenterprisefunctions andapplicationcomponentsisdeinedas SUP := ( sup p,n ) p =1...P , n =1... N with 1 if function ef p is supported by application component acn sup p , n = 0 else
{
Let us look at the following example: Suppose a set of enterprise functions EF:={A,B,C,D,E,F,G},wherethelettersrepresententerprisefunctionsasfollows:Afor medical admission, B for administrative admission (inpatients), C for administrative admission(outpatients),Dforexecutionofradiologydiagnostics,Efordecisionmaking, F for visitor and information service, and G for medical and nursing care planning. Additionally,supposeasetofapplicationcomponentsAC:={1,2,3,4,5,6,7,8,9},wherethe numbersrepresentapplicationcomponentsasfollows:(1)formedicaldocumentationsystem,(2)forpatientadministrationsystem,(3)forcardiovascularinformationsystem,(4) for RIS, (5) for decision support system, (6) for pharmacy information system, (7) for pathologyinformationsystem,(8)fordialysisinformationsystem,and(9)forteleradiologysystem. Table8.1containsthematrixSUPdescribingtheinterlayerrelationshipsbetweenthese functionsandapplicationcomponents,whereasFig.8.6illustratestheserelationships. Table 8.1 ThematrixSUPforEFandAC.ThematrixisillustratedinFig.8.6 Applicationcomponentsn=1,…,9 1 2 3 4 5 6 7 Enterprise functionsp=1,…,n
8
9
A
1
0
0
0
0
0
0
0
0
B
0
1
0
0
0
0
0
0
0
C
1
1
1
1
0
0
0
0
0
D
0
0
0
1
0
0
0
0
0
E
0
0
0
0
1
0
1
0
0
F
0
0
0
0
0
1
0
0
0
G
0
0
0
0
1
0
1
1
1
232
8
A medical admission
B administrative admission (inpatients)
1 medical documentation system
C administrative admission (outpatients)
3 cardiovascular information system
2 patient administration system
D execution of radiology diagnostics
5 decision support system
Quality of Health Information Systems
E decision making
F visitor and information service
7 pathology information system 6 pharmacy information system
4 RIS
G medical and nursing care planning
9 teleradiology system 8 dialysis information system
Fig. 8.6 MatrixSUP:rectanglesdenoteenterprisefunctions,roundedrectanglesdenoteapplication components,andconnectinglinesillustratea“1”intherespectivepositionofthematrix,thatis, thatacertainenterprisefunctionissupportedbyacertainapplicationcomponent.Forexample, enterprisefunctionEdecisionmakingcanbesupportedbyapplicationcomponent5decisionsupportsystemor7pathologyinformationsystemalternatively
InordertoreducecomplexityoftheHISitisinterestingtoknowwhichapplication componentscouldbeomittedwithoutlossoffunctionality,thatis,withouthinderingthe execution of any enterprise function. Before we deine a measure for functional redundancywewanttoexplain,howwecandetectredundantsupportofenterprisefunctionsby applicationcomponents. N
Foreveryenterprisefunction ef p ∈ EF wecaneasilycalculate isupp := ∑ supp,n ⋅ n =1
Everyisuppdenotesthenumberofapplicationcomponentsactuallysupportingtheindividual enterprise function efp; we call it its individual degree of support by application components.Everyisupp>1maybeanindicatorthatsomeapplicationcomponentsare dispensable, with isupp-1 indicating the number of possibly superluous components. However,thisnumberneedsacarefulinvestigationbecausesomeoftheapparentlysuperluousapplicationcomponentsmaybenecessaryforotherenterprisefunctions.Obviously, measuringfunctionalredundancyinaway,whichissupportiveforinformationmanagement,needsameasure,whichtakestheseinterrelationshipsintoaccount. Continuingpart1ofourexamplewecaneasilycalculateisuppasshowninTable8.2.
Table 8.2 isupp
p
efp
isupp
1
A
1
2
B
1
3
C
4
4
D
1
5
E
2
6
F
1
7
G
4
8.7 Summarizing Examples
233
The value of isup3=4 indicates that perhaps there are three superluous application componentssupportingC(administrativeadmission(outpatients)).Butdetailedanalysis shows that the application components 1 (medical documentation system), 2 (patient administrationsystem),and4(RIS)cannotbeomitted,becausetheyareneededforthe functionsA(medicaladmission),B(administrativeadmission(inpatients)),andD(execution of radiology diagnostics). However, application component 3 (cardiovascular informationsystem)seemstobeagoodcandidateforbeingremovedfromthisinformation system because function C as the only function it supports is also supported by applicationcomponents1,2,and4.Ameasureofredundancyshould,therefore,correctly indicate that considering the enterprise function C (administrative admission (outpatients))onlyoneapplicationcomponentcouldbeomitted(namely,cardiovascularinformationsystem). Howcananalgorithmautomaticallycheck,whetherparticularapplicationcomponents canbeomittedornot,givenEF,ACandSUP? ThechallengeistocalculateaminimalsubsetACmin ⊆ ACofapplicationcomponents, whichguaranteesthatallfunctionsaresupportedandthattherearenosuperluousapplicationcomponentsinuse.EachsetACminwecalla“minimalfunctionallynonredundantset ofapplicationcomponents.”Ingeneral,thereismorethanonesuchsetACminforagiven information system, that is, there is more than one way to cut down functional redundancy. LetusdescribeanysubsetAC’⊆ACofapplicationcomponentsbeingactuallyinuse USE by a vector , indicating whether application components are member of the subset AC’ornot: USE := (usen ) n=1... N with
1 if acn ∈ AC ′ usen = 0 else
min min Hence,ACmincanbedescribedby USE := (usen ) n =1... N and usenmin ∈{0,1}. Givenwhat applicationcomponentsareinuse,thatis,giventherespectivevectorUSE,wecancalculatetheindividualdegreeofsupportforallenterprisefunctionsaswellas: T ISUP = SUP ∗ USE with ISUP = isup p p =1... P . Asstatedabove,wewantthat,despiteofsomeapplicationcomponentsbeingnotinuse, everyfunctionissupportedbyatleastoneapplicationcomponent.Weintroduceavector e oflengthPcontainingonly“1”: e = e p p =1...P with e p := 1, p = 1...P. Nowwecanstatetheirstpostulation(P1):Foreveryvector USE ,whichisasacandidateforbeingconsideredasapossiblereducedsetofapplicationcomponents,thefollow ingconstraintholds: SUP ∗ USE ≥ e T . Second,wewanttohaveasfewapplicationcomponentsinuseaspossible.Weintro duceavector c oflengthNcontainingonly“-1”: c = (cn )n =1... N with cn := −1, n = 1...N . Thisleadstothesecondpostulation(P1): c T ∗ USE → max.
(
)
( )
234
8
Quality of Health Information Systems
SinceSUPisamatrixofzeroesandones,wehaveapure0-1linearprogrammingproblem.Thisparticularproblemhereiswellknowninliteratureasthe“setcoveringproblem.” Correspondingtoourstatementthattherewillbemorethanone“minimalfunctionally nonredundantsetofapplicationcomponents”therearealsodifferentsolutionsfortheset coveringproblem. The simplest algorithm for computing these solutions, known as “brute-force,” checksallcombinationsofapplicationcomponentsforpostulations(P1)and(P2).Of course this would need too much computing resources for realistic information systems with several tenths of application components. Moreover, set covering is an NP-complete problem generally, which, roughly, means that the complexity of any algorithmwillbeintheorderofanexponentialfunctionofN.Fortunately,forusual modelsofaHIScertainassumptionscanbemade,whichcanbeusedtodeineaheuristicalgorithm.The3LGM²tool(seefootnote8inChap.5)providessuchanalgorithm,whichactuallycomputessolutionsforthisparticularsetcoveringproblemeven inlargeHISmodelsveryfast. min min Thesolutionisaset USE ofallvectors USE k := usekmin , n n =1... N ,forwhich(P1)and min min (P2)hold.Thesetisdeinedas USE min := USE1 ,...,USE K .
(
{
{
)
}
}
This corresponds with the set AC := AC1 ,..., AC K of minimal functionally nonredundantsetsofapplicationcomponents AC min Inthesenseofthesetcoveringprobk . min min lemwecouldsayevery AC k coversEF.Becauseof(P2),allthosesets AC k areofthe min samecardinality M := AC k . N −M WecannowdeineFunctionalRedundancyRate(FRR)as FRR := . N FRRcanbeinterpretedasthepercentageofapplicationcomponents,whichcouldbe removedfromtheinformationsystemwithoutlossoffunctionality. Sincethegiveninformationsysteminourexampleisverysmall,wecanimmediately identifytwominimalfunctionallynonredundantconigurationswithoutusingcomplicated algorithms,butbymerelylookingaccuratelyatFig.8.6: min
min
min
AC = {1, 2, 4,5,6} and AC 2 = {1, 2, 4,6,7} , which correspond to the vectors min1 min USE1 = (1,1,0,1,1,1,0,0,0) and USE 2 = (1,1,0,1,0,1,1,0,0). min For USE1 as one of the two minimal solutions in our example holds: min min SUP ∗ USE1 = ISUP1 T = (1,1,3,1,1,1,1) (seeTable8.3). min
min
(
)
min min Thus,(P1)holdsfor ISUP1 . Inthesameway(P1)canbeshowntoholdfor ISUP 2 as well. FRRisonlydependentonNandonM,beingthenumberofapplicationcomponents andthecardinalityofallminimalfunctionallynonredundantsetsofapplicationcompo9−5 nents,respectively.WithN=9andM=5,weget: FRR = = 0, 44. 9 Hence,44%oftheapplicationcomponentsinourexamplecouldberemoved.
8.9 Summary
235
min
Table 8.3 Vector ISUP
p
efp
isupp
1
A
1
2
B
1
3
C
3
4
D
1
5
E
1
6
F
1
7
G
1
8.8 Summarizing Exercises 8.8.1 Evaluation Criteria Lookatthequalitycriteriaforstructures,forprocesses,andforoutcomesasdescribedin this chapter. Please select three criteria that you personally ind most important. Please discussyourchoice.
8.8.2 Joint Commission Information Management Standards PleaselookatthetenmajorJointCommissionInformationManagementStandardspresentedinSect.8.7.2.Analyzehowtheycorrespondtothequalitycriteriaforstructures,for processes,andforoutcomesasdescribedinthischapter.Towhichqualitycriteriacanthey bematched?
8.9 Summary Threemajorapproachestoqualityassessmentaretypicallydistinguished:qualityofstructures, quality of processes, and quality of outcome. Quality characteristics may help to describe“HISdiseases,”ind“HISdiagnoses,”andderiveadequate“HIStherapies.” Inthecontextofhospitalinformationsystems,qualityofstructuresreferstotheavailabilityofresourcesneededforinformationprocessing.Itcomprisesqualityofdata,quality
236
8
Quality of Health Information Systems
ofcomputer-basedapplicationcomponents,qualityofphysicaldataprocessingsystems, andqualityofHISarchitecture. Quality of processes deals with the quality of the information processes, which are necessarytomeettheuser’sneeds.Itcomprisessinglerecordingandmultipleusabilityof data,notranscriptionofdata,leannessofinformationprocessingtools,eficiencyofinformationlogistics,andpatient-centeredinformationprocessing. Qualityofoutcomedescribeswhetherthegoalsofinformationmanagementhavebeen reached,or,inabroadersense,towhatextentthehospitalinformationsystemcontributes tothegoalsofthehospitalandtotheexpectationsofdifferentstakeholders,andthefulillmentofinformationmanagementlaws. Itisaparticularchallengeforinformationmanagementtotakeintoaccountthosecriteriawhereonlyanadequatebalanceisadeterminantofthequalityofahospitalinformation system. This comprises the balance of homogeneity and heterogeneity, the balance of computer-basedandpaper-basedtools,thebalanceofdatasecurityandworkingprocesses, thebalanceoffunctionalleannessandfunctionalredundancy,andthebalanceofdocumentationqualityanddocumentationefforts. Evaluation studies should follow a well-deined, structured approach, comprising: study exploration; irst study design; operationalization of methods and detailed study plan;executionofstudy;andreportandpublicationofstudy. Evaluationmethodsthatcanbeusedcomprisequantitativemethodssuchastimemeasurement,eventcounting,andusersurveys,andqualitativemethodssuchasqualitative interviews,qualitativeobservations,andqualitativedataanalysis.
Strategic Information Management in Hospitals
9
9.1 Introduction Untilnowwehavediscussedhowhealthinformationsystemslooklikeandhowtheirqualitycanbedescribedandmeasured.Wewillnowexaminehowhighqualityhealthinformation systems can be achieved and how high quality can be maintained, especially in hospitals. HighqualityHIScanonlybyachievedandHISfailurescanonlybepreventedifthe HISaresystematicallyplanned,monitoredanddirected.Wesummarizethistriadbythe term‘informationmanagement’. Inthischapter,wewillirstdifferentiateinformationmanagementwithregardtodifferentscopes.Henceweintroducedeinitionsforstrategic,tacticalandoperationalinformationmanagement.Butthischapter,liketheentirebook,willfocusonstrategicinformation management. Systematic information management as a whole requires clear organizational structures. Implementing these is one of the irst tasks of strategic information management. After discussing appropriate organizational structures for information managementinhospitalswewillexplainwhatplanning,monitoringanddirectingmeans especiallyinastrategiccontext. Afterreadingthissection,youshouldbeabletoanswerthefollowingquestions: • Whatdoesinformationmanagementmeanandhowcanstrategic,tacticalandoperationalinformationmanagementbedifferentiated? • Whatorganizationalstructuresareappropriateforinformationmanagementinhospitals? • WhatarethetasksandmethodsforstrategicHISplanning? • WhatarethetasksandmethodsforstrategicHISmonitoring? • WhatarethetasksandmethodsforstrategicHISdirecting? • Howcanexpertsforinformationmanagementinhospitalsbegained?
A.Winteretal.,HealthInformationSystems, DOI:10.1007/978-1-84996-441-8_9,©Springer-VerlagLondonLimited2011
237
238
9
Strategic Information Management in Hospitals
9.2 Strategic, Tactical and Operational Information Management Inthissection,wepresentinmoredetailthetasksofinformationmanagementinhospitals. Afterreadingthissection,youshouldbeabletoanswerthefollowingquestions: • Whatdoesinformationmanagementingeneralandinhospitalsencompass? • Whatarethethreemainscopesofinformationmanagement? • Whatarethetasksof strategic,tactical,andoperationalinformationmanagementin hospitals? • What is meant by IT service management and how is it related to information management?
9.2.1 Information Management The concept ‘management’ can stand for an institution or for an enterprise function (seeSect.6.3.3.8andfootnote3inChap.6).Asaninstitution,managementcomprises allorganizationalunitsofanenterprisethatmakedecisionsaboutplanning,monitoring,anddirectingallactivitiesofsubordinateunits.Asanenterprisefunction,managementcomprisesallleadershipactivitiesthatdeterminetheenterprise’sgoals,structures, andactivities. Wecandistinguishbetween(general)managementdealingwiththeenterpriseasawhole andmanagementdealingwithdistinguishableunitsoftheenterprise.Themanagementof the business unit information processing is called information management. In general, informationmanagementshouldcontributetofulillstrategicenterprisegoals.Information managementinanenterprisemanagesitsinformationsystemandthusdealswiththefollowingobjects: • enterprisefunctionsandentitytypes, • applicationcomponents,and • physicaldataprocessingsystems. Informationmanagement • planstheinformationsystemofanenterpriseanditsarchitecture, • directsitsestablishmentanditsoperation,and • monitorsitsdevelopmentandoperationwithrespecttotheplannedobjectives. Differentmanagementscopeshavedifferentperceptionsandinterests.Hence,itishelpfulto divideinformationmanagementwithregardtoitsscopeintostrategic,tactical,andoperationalmanagement,whichallcompriseasmaintasksplanning,directing,andmonitoring. • Strategicinformationmanagementdealswiththeenterprise’sinformationprocessing asawholeandestablishesstrategiesandprinciplesfortheevolutionoftheinformation system.Animportantresultofstrategicmanagementactivitiesisastrategicinformationmanagementplan.
9.2 Strategic, Tactical and Operational Information Management
239
• Tactical information management deals with particular enterprise functions or application components that are introduced, removed, or changed. Usually these activitiesaredoneintheformofprojects.Suchtacticalinformationmanagement projectsareinitiatedbystrategicinformationmanagement.Thus,strategicinformationmanagementisavitalnecessityfortacticalinformationmanagement.The result of tactical information management projects is the enterprise information system. • Operationalinformationmanagementisresponsibleforoperatingthecomponentsof theinformationsystem.Itcaresforitssmoothoperationinaccordancewiththestrategicinformationmanagementplan.Additionally,operationalinformationmanagement plans, directs and monitors permanent services for the users of the information system.
Thisseparationisessentialbecauseeachoftheseinformationmanagementscopeshasdifferentperspectives,andthereforeusesdifferentmethodsandtools.Forexample,strategic information management focuses on strategic information management plans (compare Sect.9.4.3).Tacticalmanagementneeds,forexample,methodsforprojectmanagement, userrequirementsanalysis,andsoftwaredevelopmentorcustomizing.Operationalmanagementrequiresmethodsandtoolsfortopicsthatrangefromintra-enterprisemarketing ofservicestoservicedeskandnetworkmanagement. Figure 9.1 presents the relationships among strategic, tactical, and operational informationmanagement.Withinstrategicinformationmanagementastrategicinformation management plan has to be created as result of planning activities. This dependsclearlyonstrategicgoalsoftheenterprise,whicharegivenbythestrategic enterprisemanagement.Sincethestrategicinformationmanagementplancontainsa listofprojectstobeperformedinthecomingyears(compareSect.9.4.3),strategic directingmeanstoinitiatetheseprojectsastasksoftacticalinformationmanagement. Strategicmonitoringcollectsdifferentinformationregardingthestateoftheinformation system, its operation, users’ opinions and directives of the strategic enterprise management.Withineachprojectoftacticalinformationmanagementthecourseof theprojecthastobeplanned(projectplan)andtheprojectwillbedirectedaccording tothisplan.Theresultsarecomponents(applicationcomponentsand/orphysicaldata processingsystems)oftheinformationsystem.Again,monitoringisneededincollectinginformationoftheinformationsystem’sstateandderivingconsequencesfor therespectiveprojectsbutalsoforstrategicdecisions.Whenaprojectends,theresults have to be operated and thus we enter the scope of operational information management. Managementcomprisesonlythosetasksthatarenonexecutive.Therefore,operational tasks(suchasoperatingacomputerserver)arenotpartofmanagement’stasks.However, thoseoperationaltaskshavetobeplanned,directed,andmonitored.Thisiscarriedoutby operationalinformationmanagement. Figure 9.2 presents a three-dimensional classiication of information management activities.Itshowsthethreeobjectsofinformationmanagement(functions,application components,andphysicaldataprocessingsystems),thethreetasks(planning,directing, monitoring),andthethreescopes(strategic,tactical,operational).
240
9
Strategic Information Management
Strategic Enterprise Management
Planning
Strategic Information Management in Hospitals
Tactical Information Management
Operational Information Management
Planning Operation
Strat. IM Plan, Project Portfolio
Project Plan
Directing
Directing
Monitoring
Monitoring
User HIS component
nent
nent
Fig. 9.1 Relationshipbetweenplanning,directing,andmonitoringduringstrategic,tactical,and operationalinformationmanagement.Forexplanationseeparagraphbefore
Fig. 9.2 Three-dimensionalclassiicationofinformationmanagementactivities
9.2 Strategic, Tactical and Operational Information Management
241
9.2.2 Information Management in Hospitals We can now apply the deined management concepts to the enterprise “hospital”. Informationmanagementinhospitalsisthemanagementofhospitalinformationsystems; hencewecanuse“HISmanagement”asasynonym.Thetasksofinformationmanagement inhospitalsare: • planningthehospitalinformationsystemanditsarchitecture; • directingitsestablishmentanditsoperation; • monitoringitsdevelopmentandoperationwithrespecttotheplannedobjectives. Informationmanagementinhospitalsisperformedinanenvironmentfullofinluencing factors.Forexample,decisionsmadebythehospital’smanagementdirectlyinluence informationmanagement(e.g.,adecisiontocooperateinahealthcarenetwork).New legalregulationshaveaneffectoninformationmanagement(e.g.,alawenforcingthe introductionofanewbillingsystembasedonpatientgrouping).Patientsandusersof thehospitalinformationsystemwiththeirvalues,attitudes,comments,demands,and fearsalsoinluenceinformationmanagement.Ontheotherside,informationmanagementitselfmayaffect,forexample,themanagementoftheenterprise(e.g.,information management may propose to introduce a hospital-wide, multiprofessional electronic patient record system; this must in turn lead to strategic activities such as process reorganization). Figure9.3presentsthisrelationshipbetweenHISmanagementandHISoperation,and theinluencingfactors. Wenowlookattheactivitiesofstrategic,tactical,andoperationalinformationmanagementinhospitals.
Fig. 9.3 Strategic,tactical,andoperationalinformationmanagementinhospitals,HISoperation, andtheirrelationships
242
9
Strategic Information Management in Hospitals
9.2.3 Strategic Information Management Strategicinformationmanagementdealswiththehospital’sinformationprocessingasa whole.Itdependsstrictlyonthehospital’sbusinessstrategyandstrategicgoalsandhas to translate these into an appropriate information strategy. The planning activities of strategic information management result in a speciic strategic information managementplan.Thisplanincludesthedirectionandstrategyofinformationmanagementand givesdirectivesfortheconstructionanddevelopmentofthehospitalinformationsystembydescribingitsintendedarchitecture.Aproposalforthestructureandcontentof strategicinformationmanagementplansispresentedinSect.9.4.3.Thestrategicinformationmanagementplanisthebasisforstrategicprojectportfolios.Theycontainconcrete projects, which implement the objectives of the strategy, and shall be revised regularly.Forexample,thestrategicinformationmanagementplanmightcontainthe introduction of a clinical information system on all wards within the next 2 years to providehealthcareprofessionalswiththerightinformation,intherightplace,atthe right time. The strategic project portfolios could then contain individual projects, for example, on decision making, planning and organization of patient treatment, order entry,andmedicaladmission. Directingaspartofstrategicinformationmanagementmeanstotransformthestrategic information management plan into action, that is, to systematically manipulate the hospitalinformationsystemtomakeitconformtothestrategicplan.Thesystem’smanipulationisusuallydonebytheinitiationofprojectsofthestrategicprojectportfolio.The projectsdealwiththeconstructionorfurtherdevelopmentandmaintenanceofcomponentsofthehospitalinformationsystem.Planning,directing,andmonitoringtheseprojectsarethetasksoftacticalinformationmanagement.Operationalmanagementwillthen beresponsiblefortheproperoperationofthecomponents.Anexampleofstrategicdirectingwouldbetoinitiateaprojectfortheintroductionofaprovider/physicianorderentry system. Monitoringaspartofstrategicinformationmanagementmeanscontinuouslyauditing HISqualityasdeinedbymeansofitsstrategicinformationmanagementplan’sdirectives andgoals.Auditingshoulddeterminewhetherthehospitalinformationsystemisableto fulillitstaskseficiently,thatis,whetheritcancontributesigniicantlytothehospital’s goals(Sect.8.4.1),meetthestakeholders’expectations(Sect.8.4.2)andfulillthelaws (Sect.8.4.3). Themanagement’staskistoinstall“sensors”toaudittheinformationsystem’squality (compareSect.8).Managementhastoreceiveinformationfromthecurrentprojects,from operationalmanagement,fromusers,andfromthevariousstakeholders.Additionalinformationcanbegainedthroughevaluationprojects(seeSect.8.6). Monitoringresultsareusedasinputforthedirectingtasksofinformationmanagement, whichcouldforexampleinitiatefurtherprojects.Monitoringresultswillalsogivefeedbacktoupdatethestrategicinformationmanagementplan,whichcouldforexamplelead tofurtheractivitiesofstrategicmanagement. Strategicinformationmanagementanditsstrategicinformationmanagementplanare thevitalrequirementsfortacticalandoperationalinformationmanagementinahospital.
9.2 Strategic, Tactical and Operational Information Management
243
9.2.4 Tactical Information Management Tacticalinformationmanagementdealswithparticularenterprisefunctionsorapplication components.Itaimstointroduce,remove,change,ormaintaincomponentsofthehospital informationsystem.Suchacomponentcouldbeaprovider/physicianorderentrysystem. Relatedactivitiesareusuallyperformedwithinprojects.Theseprojectshavetobeinitiated aspartofaninformationstrategy,whichisformulatedintheprojectportfolioofastrategic informationmanagementplanasdrawnupbythestrategicinformationmanagement.The resultofalltacticalinformationmanagementprojectsistheHISitself. Theorganizationoftheoperationandmaintenanceofinformationprocessingtoolsis partofoperationalinformationmanagement.However,ifproblemsoccurduringtheoperationofHIScomponents(e.g.,frequentusercomplaintsaboutamedicaldocumentation system),appropriateprojectsmaybeexecutedbytacticalinformationmanagement(e.g., introducingabetterversionofthemedicaldocumentationsystem).Typically,thosetactical information management projects comprise a planning phase, a running phase (which couldcontain,forexample,systemanalysis,evaluation,selection,speciication,orintroduction),andainishingphase(seeFig.9.4). Planning in tactical information management means planning projects and all the resourcesneededforthem.Eventhoughtacticalinformationmanagementprojectsare basedonthestrategicplan,theyneedaspeciictacticalprojectplan.Thisplanhasto describetheproject’ssubjectandmotivation,theproblemstobesolved,thegoalstobe achieved, the tasks to be performed, and the activities to be undertaken to reach the goals.
Fig. 9.4 Typicalphasesoftacticalinformationmanagementprojects
244
9
Strategic Information Management in Hospitals
Directing in tactical management means the execution of such tactical information managementprojectsinhospitals,basedonaprojectplan.Therefore,itincludestypical tasksofprojectmanagementsuchasresourceallocationandcoordination,motivationand trainingofthestaff,etc. Monitoringmeanscontinuallycheckingwhethertheinitiatedprojectsarerunningas plannedandwhethertheywillproducetheexpectedresults.Monitoringresultsinluence projectplanning,asaproject’splanmaybeupdatedorchangedaccordingtotheresultsof theproject’smonitoringinagivensituation.
9.2.5 Operational Information Management Operationalinformationmanagementisresponsibleforoperatingthecomponentsofthe hospitalinformationsystem.Ithastocareforitsoperationinaccordancewiththestrategic informationmanagementplan. Planning in operational information management means planning organizational structures,procedures,andallresourcessuchasinances,staff,rooms,orbuildingsthat arenecessarytoensurethefaultlessoperationofallcomponentsofthehospitalinformation system. For example, operational information management may require the installationofaservicedeskandaservicesupportsystemthatenablesthequicktransmissionofusers’errornotestotheresponsibleservices.Suchsystemsbutalsorespective staff resources need to be available for a longer period of time. Therefore, they should be allocated as part of a strategic information management plan. Moreover, planninginthiscontextconcernstheallocationofpersonnelresourcesonaday-to-day basis (e.g., planning of shifts for staff responsible for user support or network management). Directing means the sum of all management activities that are necessary to ensure properresponsestooperatingproblemsofcomponentsofthehospitalinformationsystem,thatis,toprovidebackupfacilities,tooperateaservicedesk,tomaintainservers, and to keep task forces available for repairing network components, servers, personal computers, printers, etc. Directing in this context deals with engaging the resources plannedbythestrategicinformationmanagementplaninsuchawaythatfaultlessoperationofthehospitalinformationsystemisensured.Operationalinformationmanagement does not mean to exchange a server, but to organize the necessary services and resources. Monitoringdealswithcaringfortheveriicationoftheproperworkingandeffectivenessofcomponentsofthehospitalinformationsystem.Forexample,anetworkmonitoring system may regularly be used to monitor the availability and correct working of networkcomponents(compareFig.9.5). ToguaranteethecontinuousoperationofthemostimportantcomponentsofaHIS,itis helpful to draw up a concept for operational information management. Such a concept shouldclarifythefollowing:
9.2 Strategic, Tactical and Operational Information Management
245
Fig. 9.5 Monitoringof theserverofa hospitalinformation system
• Whichcomponentshavetobesupported? • Whattaskscompriseoperationalsupport? • Whoisresponsiblefortheoperationalsupport? • Whatshouldbetheintensityofoperationalsupport? Typically,threelevelsofoperationalsupportcanbedistinguished.First-levelsupportisthe irstaddressforallusergroupswithanykindofproblem.Itmayconsist,forexample,ofa central24-h-hotline(servicedesk)thatisresponsibleforthemanagementofuseraccounts andirsttroubleshooting,orofdecentralizedinformationprocessingstaff.Whentheirstlevel support cannot solve the problems, it hands them over to the second-level support, speciallytrainedinformaticsstaffinthecentralinformationmanagementdepartmentwhoare usuallyresponsiblefortheoperationofthespeciicapplicationcomponents.Thethird-level support,inally,addressesthemostsevereproblemsthatcannotbesolvedbythesecond-level support.Itcanconsist,forexample,ofspecialistsfromthesoftwarevendor(Fig.9.6). Table 9.1 presents objects, responsibilities, tasks, and the intensity that should be deinedaspartoftheoperationalmanagementconceptforthecomputer-basedpartofa HIS.Asanexample,aconceptforoperationalmanagementinahospitalcouldclarify
246
9
Strategic Information Management in Hospitals
Table 9.1 Dimensionstobeconsideredforoperationalinformationmanagementofthecomputerbasedpartofhospitalinformationsystems Dimension Facets Objects
Decentralizedapplicationcomponents(e.g.,in departments) Centralapplicationcomponents(e.g.,patientadministrationsystem) Workstations Decentralizedservers Centralservers Networks Backbone
Responsibility
Local(indepartments) Central(indepartmentsforinformationprocessing) Vendors
Task
First-levelsupport(incidenttaking,incidentanalysis, problemsolvingifnecessary,usertraining) Second-levelsupport(trainingcourses,regularoperation, dataprotection) Third-levelsupport(softwaredevelopment,problem solving,contactwithvendors)
Intensity
Availability(e.g.,24h/day,7days/week) Presence(e.g.,locally,bypager,byhotline) Timeliness(e.g.,answeringtime