Collaborative Design of Process-Aware Information Systems ...

Rüdiger Breitschwerdt Aileen Collier Rick Iedema Oliver Thomas

Collaborative Design of

Process-Aware Information Systems on Handheld Devices for Mobile Health Workers

Living Lab Business Process Management Research Report, Nr. 7, Oktober 2013

RE

SE

AR

CH

www.living-lab-bpm.de

Living Lab Business Process Management Research Report

Herausgegeben von

Prof. Dr. Oliver Thomas Universität Osnabrück Fachgebiet Informationsmanagement und Wirtschaftsinformatik Katharinenstraße 3, 49074 Osnabrück Telefon: 0541/969–4810, Fax: –4840 E-Mail: [email protected] Internet: http://www.imwi.uos.de/ Bibliografische Information der Deutschen Nationalbibliothek

Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen National-bibliografie; detaillierte bibliografische Daten sind im Internet über http://dnb.d-nb.de abrufbar.

ISSN 2193–777X

Zitationshinweis

Breitschwerdt, R.; Collier, A.; Iedema, R.; Thomas, O. (2013): Collaborative Design of Pro-cess-Aware Information Systems on Handheld Devices for Mobile Health Workers. In: Thomas, O. (Hrsg.): Living Lab Business Process Management Research Report, Nr. 7, Osna-brück, Living Lab BPM e.V. Das Werk einschließlich aller seiner Teile ist urheberrechtlich geschützt. Jede Verwertung ist ohne Zustimmung des Living Lab Business Process Management e.V. unzulässig. Das gilt insbesondere für Vervielfältigungen, Übersetzungen, Mikroverfilmungen und die Ein-speicherung und Verarbeitung in elektronischen Systemen.

Copyright © 2013 Living Lab Business Process Management e.V. Living Lab Business Process Management e.V. Universität Osnabrück Katharinenstraße 3 49074 Osnabrück www.living-lab-bpm.de

Collaborative Design of Process‐Aware Information Systems on Handheld Devices for Mobile Health Workers

RüdigerBreitschwerdt1,AileenCollier2,3,RickIedema3,OliverThomas1

1ChairinInformationManagementandInformationSystems,UniversityofOsnabrück

Katharinenstraße3,49074Osnabrück,Germany{ruediger.breitschwerdt|oliver.thomas}@uni‐osnabrueck.de

http://www.imwi.uos.de/

2FlindersUniversity,PalliativeandSupportiveServices,RepatriationGeneralHospitalGPOBox2100,Adelaide5001

[email protected]://www.flinders.edu.au/

3UniversityofTechnology,SydneyJonesSt.,Building10,Level5Broadway,NSW2007,Australia

[email protected]://www.centreforhealthcom.org/

Services increasingly gain importance but the usage of corresponding guide‐lines in field scenarios hasnot beenparticularly supportedby IT.This paperwill suggest an approach to process‐aware information systems on portabledevicesintegratingtheusersthusaddressingthischallenge.Scenariosforjointmodelinganddevelopmentconsidering therequirementsof fieldservicepro‐viderswillbe studied tomeet thedemandedspecifics forworkflow tool sup‐portorrepresentation.Findings includeaweb‐basedmodeling toolselection,are evaluated with a use case in Australian palliative community nursing topresentexperiencesgenerallyworthwhileforsuchacollaborativeinitiative.

Keywords: Process modeling, Process‐aware information system, Workflow,Fieldservice,Palliativecare.

1 Introduction

Thedemandformobileservicesrisessteadilyandaffiliatedperformancequalitydependsagreatdealontheavailabilityofinformation(SherryandRatzan2012).Thatiswhytheuseofmoderninformationtechnology(IT)haspotentials:notonlytoavoid/correctmis‐takesbutalsotoimproverelationstotheclientortosupportdecisionsthusboostingout‐comes;forinstancewithawidelyavailableinformationbasisandhenceresultinginanac‐celeratedorientationforserviceproviders(Varshney2007;Akessonetal.2007;BatesandBitton2010).Therefore,manyinitiativesattempttoraiseefficiencyandqualityusingIT–alsoinfieldscenarios(Jhaetal.2008).Withadequatemodelingandelectronicprovisionofworkflowinformation,processqualitycouldbesupportedintheseenvironments(BergandToussaint2003).

2 RüdigerBreitschwerdt,AileenCollier,RickIedema,OliverThomas

Living Lab BPM Research Report 7 | 2013

Ammenwerthetal.(2010)requestcontemporarydataaccessmeansasanecessityforusableITinfrastructure.PositiveresultsforfieldserviceworkersusingmobileITapplica‐tionsaredescribedbyLegneretal.(2011),Birkhoferetal.(2007).AlthoughmobileITso‐lutionshavebeenexisting foryears (Tachakraetal.2003), therestill isdemand in fieldscenarios (Pryss et al. 2011;Wälivaara et al. 2009). Todate,most of portable solutionshavebeenanalyzedorevaluatedfor‘immobile’scenarios,e.g.inhospitalsordoctors’prac‐tices(Chatterjeeetal.2009;Breitschwerdtetal.2011).

FieldenvironmentsasinhomecareorambulancesstilllackIT‐supportedworkflows.There,correspondingstandardsandguidelinesdoexistandprovideinformationonindi‐cation,symptomsandexecutionofcareactivitiesormedication(SchoenbaumandGottlieb1990).Aportabletoolintuitivelyrepresentingthemmissesatthepointofcare:Ambulantproviderswouldthusbesupportedtoefficientlydeliverhighquality.Clientsshouldalsoprofit from the improved availability of – e.g. evidence‐based – information(Breitschwerdtetal.2011;Sadeghietal.2011).

It is explored subsequently how process‐oriented support can be realized tomeetmobile serviceproviders’need,here in thehealth arena.Consideringprospectiveusers’perspectiveshouldherebygenerateacceptance(Legneretal.2011).Therefore,welookatrelatedwork(chap.2)beforedefining(3),executing(3.1‐3.3),evaluating(4)anddiscuss‐ing(5)ourapproach.

2 Related Work

Process‐orientation represents an important instrument for ensuring quality (Lenz andReichert 2007). Breitschwerdt et al. suppose information deficits of mobile healthcareproviders regarding mandatory workflows. Modeling them could help with generatingadded value like achieved inmobile service industries also focusing on complex caringprocessesanddocumentation(Walter2009).Technicalfieldserviceprovidersareseeninanalogy(Breitschwerdtetal.2011;Rügge2007;Rügge2003;PicotandSchmid2007)be‐causeofsimilar(Winston1980)work(esp.complex‘maintenance’).

Here,theauthorsfocusontreatmentprocesses:prevention,diagnosis,therapy,nurs‐ing/ care and other client‐centered workflows. Compared to administrative processesthey are of higher complexity and therefore offermorepotential for support (Lenz andReichert2007).Standardslikeexpert‐consentedorevidence‐basedguidelinesarisemoreandmoreamongst them.Thesestandardizationeffortsare challengingbutcanbe facili‐tatedbyIT(Pedersenetal.2011).Inturn,theyrepresentabasisforIT‐supportedexecu‐tion(Kaiseretal.2011). IT, though,needstobe integrated further intoservicedelivery:workflowrepresentationshelpbetteriftheyarenotsimplypublishedonline,butembed‐dedusingprocess‐oriented systems. Thosehelpwith defining and setting standards re‐ducingcostswithoutalossofquality(BergandToussaint2003).Also,theyorganizetheinformationprovidedalongtheworkflowandaretobeflexiblyadjusted.Thatfunctionali‐ty is attributed to process modeling respectively process management tools (Lenz andReichert2007;Reichert2011;BenDhiebandBarkaoui2012).Amongst them,toolsrun‐ningonmobiledevicescurrentlygainsignificance(Pryssetal.2011;Houyetal.2011).

Process‐aware information systems (PAIS) supportworkflows based on underlyingprocessmodelsthuspreservinguserrequirementsandconfidenceinreliable,correctsys‐temoperationsbefore itsdeployment (Mansetal.2010).PAISprovideallphasesof theworkflowlifecycle,i.e.itsspecificationanddesign,developmentandconfiguration,valida‐tion test and implementation (Reichert2011;Mansetal.2010).Onmobiledevices they

CollaborativeDesignofProcess‐AwareInformationSystems 3


can be key for integrating such support into field service delivery like ambulant care(Pryssetal.2010;Pryssetal.2011;Breitschwerdtetal.2011).

3 Approach to Collaborative Realization in a Field Scenario

Awell‐definedprocedureisnecessarytoeventuallyachieveanimprovementforthisfield:Integratingdesign‐orientedmethodologiesforbothinformationsystems,ase.g.inHevneretal.(2004),Hevner(2007),Österleetal.(2010)andhealthcare(Rouse2009)ispursued,here:weconductiterativerevisionsthusapproachingtofindinnovativeproblemresolu‐tionsby

1. definingmethodsto

2. identify, describe problems in application domain as in the context of field serviceprovidershere,beforewe

3. createandevaluateourartifactsolutionsbasedontheprevioussteps.

Figure1.Guidelines,algorithmsorlegalaspectsneedtoberepresentedinprocessmodelsforex‐pert/userfeedbacksothattransferofvalidprocessestouserinterfacesofaPAISsucceeds



Asmethodsserveanalogy,focusgroups,theintegrationofworkshop‐basedandvirtu‐alenduser feedback fordevelopmentofartifacts likerequirements,processmodelsandsystems.Applicationscenarios (Frank2010;BergandToussaint2003)and– forevalua‐tion–ausecaseanalysis(Mouttham2012)inchapter4wereusedtochecktheviabilityofour solutions. The challenges described throughout the previous chapters – here e.g.guidelinestoberepresentedinmodelsforsolutiondevelopment(seeFig.1)–andthere‐quirementsascollected inchapter3.1aretheproblemstobeaddressed.Sinceeffective‐nessofITisacomplexcollaborativeachievement(LenzandReichert2007;Moutthametal.2012;Anetal.2009),weconsideredjointdesign,developmentand(test)usage(com‐bined:creation)concerningsoftwareanddeviceswiththeusers(e.g.byfeedback‐enabledworkflowmodeling).During thewholeapproachwe involve them forcollaborativesolu‐tionconception(Anetal.2009;vandeKaranddenHengst2009;Lee2007;Pedersenetal.2011;Ammenwerthetal.2010).

3.1 Approach to Collaborative Realization in a Field Scenario

The profile of stationarywork scenarios as per Pryss et al. (2011) has certain require‐ments also valid for field service delivery, e.g. synchronization, “physical problems likebrokenconnectionsormal‐functioningdevices(…)tobehandledbythesupportinginfra‐structure,butwithoutburdeningusers”.Buttheylacksomespecificonesdifferingmostlyregarding additionalmobility and information needs demanded (Chatterjee et al. 2009;Breitschwerdtetal.2011).AnalogtoPryssetal.(2011),weelicitedrequirementsR1‐R11(seeTable1)byscrutinizingdifferentfieldapplicationscenarios,e.g.of

communitynurses

paramedics,

rescuestaffactingindisastermanagementor

specifichomecareproviders(e.g.physicaltherapistsvisitingpatientsathome).

Theywereconfirmedduringconsecutive twohour focusgroupsessions(Tongetal.2007;vandeKaranddenHengst2009):oneinlate2011withtwelve(physicians,nurses,therapists,managers,ITadministrators)thenoneinearly2012withfive(nurses,manag‐erswithnursingbackground,ITadministrator)staffofAustraliancommunityhealthcareproviders(Tongetal.(2007)setaminimumoffour,vandeKaranddenHengst(2009)ofsix participants) from the Sydneymetropolitan area. Afterwards, we refined themwithfourITdevelopersholdingaB.Sc.degreeinInformationSystemsandaparamedicwithaMaster degree of the same subject. Finally,we categorized themas perMoutthamet al.(2012)foreasierallocationduringfurtherdevelopmentefforts.Therequirementsalsore‐flect(cp.R9)thatambulantserviceprovidersasusersareoftenontheflyandvirtuallyin‐teract with coworkers via mobile devices (Mouttham et al. 2012; Breitschwerdt et al.2011).



Table1.RequirementsidentifiedforPAISinfieldscenario

No. CategoriesaccordingtoMoutthametal.(2012)

Enduserrequirements Sampleconsequencesfordevelo‐pment

R1 Processes Intuitivelyunderstandablerepresentationofprocess‐esandinstancesistobesupported.

Deliberateselectionofprocessmodelinglanguage.

R2 Processes Processesinsystemmustbecomplianttomedicalguidelines,pathways,etc.

Useofstandardized/recom‐mendedprocesses.

R3 Processes/Data

Attachingdatatoexecutedprocessinstancesforpost‐processing,esp.viafree‐textentryfields,needstobepossible.

Enablingdatatransferwithoth‐ersystemsoftheprovider.

R4 Data Dataisexchangedcompat‐iblywithstandardinterfac‐es.

Datacreated/processedshouldbee.g.DICOM,HL7‐compliantforintegrationintoenterprisearchitecture.

R5 Data Systemconceptionfollowsdatasecurityandprivacyrequiredbypatient,care‐giverandlaw.

Thesystemconceptionneedstoconsiderdifferentnationallawsinaninternationalsetting.

R6 Data/Technology

interoperability

Thesystemenablesarti‐fact‐based(back‐)officeprocessing.

Offeringe.g.‘print‐to‐PDF’func‐tionality.

R7 Data/Technology

interoperability

Easyinvite‐functionalityforsharingpatient‐relateddatawithauthorizedper‐sonsmustbeprovided.

User/roleconceptanddatase‐curityaccordingtolegalre‐strictionsapplying.

R8 Technologyinteroperability

Theinstallationandupdateofthesoftwareiseasyandfast.

Softwareversiondeploymentprocedureneededfortesting.


Thesystemrunsbestpos‐siblyplatform‐independentonstationaryandmobiletouch‐screendevices.

Certainadequateoperatingsys‐temsneedtobeconsidered.


Thesolutionneedstobelowcostsincefundingcrit‐icalinthedomain,especial‐lyitsambulantscenarios.

Useofexistingsystemstobepursued(individualsolutionscostly):Freeoropensourcesoftwaremightbepreferred.


Asupportandmaintenanceconceptisrequired.

Administratorrolesandeventlogginghavetobeconsidered,apotentialhandovertothepro‐vider’sITserviceprepared.

3.2 Process Representation for Conceptual Modeling

Modeling as a necessary step during requirements engineering means abstracting andcapturing theessence for facilitating the interactionwith theusers (BergandToussaint



2003).Anadequateprocessmodeling languageasabasicprerequisitealsohelpswithasuccessfulimplementatione.g.of(clinical)pathwaysorguidelines(Reichert2011;Junetal.2012).Forintegratinghealthcareprocessknowledgeintoinformationsystems,ithastobe computer‐interpretably or even formallymodeled for executable usage (Kaiser et al.2011;Mansetal.2010).That“translation” intosuchaconceptualmodel isstillcomplexfroman interdisciplinary (here:healthcarevs. informationsystems)domainperspectiveanddifficultsinceitneedstobecompleteforsuccessfulimplementation(Mans2010;BergandToussaint2003).

Meanwhile,standardprocessmodelinglanguageslikeBusinessProcessManagementNotation(BPMN)havebecomewidespreadalternativesforconceptualmodelingandanal‐ysesofworkflows(Peleg2011;BeckerandJaniesch2008;Petersenetal.2010).However,theauthorshaveonlybeenawareeitherofusagecontexts instationaryscenariosornotspecificallyforambulantcare(Pryssetal.(2010),Pryssetal.(2011)analyzevariousfielddomainsatonce).CorrespondingtoR1,weproposetheuseofBPMNsolelyfortheseenvi‐ronments:thislanguageissupposedtobeeasilyunderstandableevenforusersinexperi‐enced with or beginners in modeling (OMG 2006), such as field service providers likecommunitynursesorparamedics.Atthesametime,it isadvocatedandpreferred,e.g.toUML, for servicemodeling (HeßandMeis2011).Therefore,weuseBPMN tobridgebe‐tweenconceptualizationandimplementationoftheprocessesfortheusersneedingtoun‐derstandthem.Workflowrepresentationthusenhancedmakestreatmentsmoretranspar‐ent for stakeholders other than the aforementioned ones: this would concern patients,theirrelativesand–asasynergy–designersofotherdomain‐specificsupportingsystemsrespectivelysolutions(Kaiseretal.2011).Thatiswhyanintuitivenotationmattersespe‐ciallyformodelrepresentation.

3.3 Alternative Toolset Options

Theoutcomesof carecanbeenhancedwhen implementingworkflows into (clinical) in‐formationsystemssothattheyareenabledtoprovideadvancedmechanismsasrequiredbyindividualdecisionsupportorcaredelivery(Kaiseretal.2011).AsPryssetal.empha‐size, the IT infrastructure for such an approach must be user‐friendly for proper con‐figurationofprocesses,services,devicesorapplications(2011).

Todate,atleast70differenttoolssupportBPMN(OMG2012)thusfulfillingR1.Con‐sideringfinancialrestrictions(R10)andthedemandedmobility(R9)asparamount,cost‐/time‐saving(R8)andwebbrowser‐basedtoolsdeployableonsmallportabledevices likesmartphonesorpad/tabletPCs(Breitschwerdtetal.2011)werelookedfor.

Yanetal.(2011)recommendOryxassuchaBPMNtool.Sincenolongeronlinetoday(seewww.oryx‐project.org),weanalyzedbothSignavio(academic.signavio.com)originat‐ingfromOryx(KunzeandWeske2010)andits‘derivate’Activiti(activiti.org).NoneoftheothertoolsanalyzedinYanetal.(2011)iscapabletosupportweb‐basedmodeling.Sowesearchedthegiven70incomprehen‐sivemannerforanyotherfulfillingthatcharacteristicandidentifiedsuitesofOracleBPMSuite(www.oracle.com/us/technologies/bpm/suite/),Rigrr (rigrr.rapilabs.com), Inubit (www.inubit.com/en/inubit‐suite.html), IYOPRO(www.iyopro.com),Cordys(www.cordys.com/bpms‐business‐process‐management‐suite),EMC Documentum xCP (www.emc.com/products/detail/software/xcp‐business‐process‐management.htm), SkeltaBPM.NET(www.skelta.com/products/bpm/overview.aspx),ProcessMaker(www.processmaker.com)andjBossBRMS (www.redhat.com/products/jbossenterprisemiddleware/business‐rules/).However,mostof themhavenotbeensup‐portinguseof their full functionalitiesonmobiledevices, so far; theyonlyofferprocess



design,notexecutionviabrowser,too(seeTable2withfeaturesrelevantfortherequire‐mentsidentified).

Activiti comprisesallnecessarycomponents toserveasacomprehensivePAIS itselfandrunsonmobileiOSdevices.Signaviodoesthelatter,too,butisamodelingtoolinthefirstplace.Incomparison,itdoesnotrequireanyuser‐driveninstallationandadditionallyoffersaroleconceptwithinviteoptions(cp.R7),print‐to‐PDF(cp.R6),commentingfea‐tures,automatedversioningandglossaryaswellassyntaxchecking.Thesefeaturesmakeitsuperiortoother,costliertoolsrunningonmobiledeviceslikefromCordysandInubit.Requirements2,4,5and11arerathernotaffectedbythetoolchosenandwerethereforenotconsideredinthissection.

Table2.Featuresofbrowser‐basedBPMNtoolsforPAISrealizationfieldenvironment

4 Evaluation

4.1 Case Study in Ambulant End‐of‐Life Care Scenario

Wehaveevaluatedourapproachbymeansofstudyingausecaseinthecontextofanin‐ternationalprojectintegratingbothresearchersfromInformationSystemsandHealthSci‐encedomains.Australiancommunitynursingproviderswereinvolvedasprospectiveus‐ersfromamobileservicefield.Sincepartofanintercontinentalinitiative,follow‐upsandwork(e.g.theworkflowmodeling)betweenoraftermeetingsonlyhappensviaphoneormostlyvirtuallybecauseoftimezonedifferencetoEurope.

Palliative care representsa complexdomaindealingwith individualphysical, social,psychologicalandspiritualneedsofpeopleintheprocessofdyingfromalife‐limitingcon‐

Feature

Signavio

Activiti

Rigrr

Cordys/Inubit

EMC/

Oracle

IYOPRO

ProcessMa‐

ker/jBoss

BRMS

Skelta

Workflowengine/supportsprocessexecutionasinstances(cp.R3)

no yes no yes yes yes;butnotinfreever‐sion

noornotviabrowser

notviabrowser

Clientinstallation(s)required(cp.R8)

no yes no yes yes yes yes yes

Specifiedcompati‐blewithmobileplatforms(cp.R9)

iOS iOS no yes no no no no

Freeware(cp.R10),notconsideringsupport

onlyforre‐search

yes yes no No yes(forfreeversion)

yes no



dition (Moutthamet al. 2012; vonReibnitz andNussbaumer2011; Sadeghi et al. 2011).Recognizing death as the final stage of life, paramount goal is integrating holistic ap‐proaches toassessandensuretheconcernedpersons’qualityof life.Thiscomprisese.g.supportingthedyingandtheirrelativesaswellasprovidingrelieffromsymptoms/pain(KralikandvanLoon2011,pp.370‐371).

Per year, about 140,000Australians (most of them70 years or older) decease. Thedemographicallyagingpopulation(KralikandvanLoon2011,p.63:25%shareofthepop‐ulationolderthan65yearsby2040)andpatientssurvivingtheirchronicillnesseslongerwill increasetheneedfor(palliative)careadditionally(Moutthametal.2012;Gianchan‐dani2011).Eventodate,specialistend‐of‐lifecarecannotbeprovidedineverycase.Thatis why Australian initiatives, especially Palliative Care Australia(www.palliativecare.org.au), try to integrate it into generalist community nursing. Thistargetsthateachdying individualcanbe lookedafterbyprimarycareproviders,usuallycommunitynursesnottrainedforend‐of‐lifesituations,withinputfromspecializedstaffwhen/ ifnecessary.Affiliatedservices requiredhave tobeaddressedbydevelopmentofnovelprovisionmodels(KralikandvanLoon2011,pp.370‐371),toensurequalityespe‐cially fornursing inpatients’homes(Moutthametal.2012;SegalandLeach2011).Thissettingisinterestingfortworeasons:1.regularcommunitynursesneedtodeliverpallia‐tivecarebecauseofalackofcorrespondingspecialistnurses,2.everyhealthcareprofes‐sional is likely to face end‐of‐life situations (KralikandvanLoon2011,pp.373‐374). Inanycase,thecaregiversrequireadditionalinputtoreachanadequateskilllevel.

Also,(community)palliativecareisacollaborativeusagescenariowhereIT‐basedso‐lutionscanprovideasignificantdealofhelp(Johnstonetal.2012;Martinezetal.2009),notonlyininpatient(Ashetal.2012;Hongetal.2009)butevenmoreinremote(McCalletal.2008;Pitsillideset al.2006)outpatient settings (Sadeghi et al. 2011;Wälivaaraet al.2009).Australiaisavastlandandcommunitynurseseveninmetropolitanregionshavetotake rides from one patient to another for hours and use advanced mobile IT likesmartphonesandpadsonaregularbasis.Moutthametal.recommendedmobiledevicesaswellasuseofprocessandguidelinemodelsinthisenvironment,butnotconsideringcol‐laborative conception (2012). However, goal of this project had been to jointly modelworkflowsanddevelopaPAISforgeneralistcommunitynursestogettoknow,initiateandcontinueend‐of‐life careasper theLiverpoolCarePathway for theDyingPatient (Eller‐shaw andWilkinson 2003) (LCP; seewww.mcpcil.org.uk/liverpool‐care‐pathway/): thatmeansprovidinginformationanddecisionsupportonportabledevices,e.g.concerningdi‐agnosisofdying, guidelines for symptommanagement,painassessmentprompt,nauseaandvomitingpromptaswellasguidingthroughdatalikevitalsignstobecollected.

4.2 Results

Inordertochallengethefindingsofchapter3wecross‐checkedthemagainsttheafore‐mentionedproject.Theyprovednotonlymostlyvalidthroughout iterativepersonalandvirtualcollaborationwithpalliativecareprovidersincludinggeneralistcommunitynurses,specialistpalliativenurses,theirmanagementandITadministratorsoverseveralmonths.Wealsoassessedtheminaworkshopwithexpertsforinformationsystemdevelopment,nursingandITinfrastructure.

Thismostlyconfirmedtherequirements,butwithinterdependentpreferencesonthecase‐specificsignificanceforrealization(seeTable3).Someaspectsweremorenegligible,e.g.R11,thanothers.Alreadyforthemodelingphase,awebbrowserbasedtoolrequiringbasicallynotanyinstallationhadbeenmandatory.TheOryx‐basedtoolswerepreferredto



theothersbecauseof their iOScompatibility.Activitiwith itspartlycomplex installationandmulti‐componentusagewasdenied,however.Therefore,Signaviowasusedformodel‐ing and commenting, even though not providing direct process execution as instances.ThusrequiringadditionaldevelopmenteffortstorealizethePAIS,aworkflowenginetobeconnectedandaway to transfer theprocessmodels therehad tobecomeupwith.ThiswasachievedbymeansofanexportfromSignavioviabpmn.xmlformatforfurtherhan‐dlinginthePAIS(seeFig.1forscreenshotsofsamplerealizationasan‘app’).

Table3.Evaluationifrequirementsinambulantscenario

No. Enduserrequirements Applicationsuccessfulincase

R1 Intuitivelyunderstandablerepresen‐tationofprocessesandinstancesistobesupported.

Yes,BPMNapprovedasprocessrepresen‐tation.

R2 Processesinsystemmustbecompli‐anttomedicalguidelines,pathways,etc.

Yes,ismaintainedbyuseofexpert‐consentedLCP

R3 Attachingdatatoexecutedprocessinstancesforpost‐processing,esp.viafree‐textentryfields,needstobepossible.

Yes,requested,butnotpossiblewithselect‐edtool,additionaldevelopmentforrealiza‐tionhastofollow.

R4 Dataisexchangedcompatiblywithstandardinterfaces

Yes,datacreatedwith/processedbysystemshouldbeHL7‐compliantforlatercouplingwithexistingclinicalinformationsystem.

R5 Systemconceptionfollowsdatasecu‐rityandprivacyrequiredbypatient,caregiverandlaw.

Yes,Australianlegislationneedstobeap‐plied;ifuseconsideredforothercountrypartneringintheprojectthataspecthastoberevisited.

R6 Thesystemenablesartifact‐based(back‐)officeprocessing.

Yes,‘print‐to‐PDF’functionalityrequired(featuredbySignavioduringmodeling,tobeimplementedinPAIS).

R7 Easyinvite‐functionalityforsharingpatient‐relateddatawithauthorizedpersonsmustbeprovided.

Yes,invitingcolleaguesandpatientrelativesrequired.Userandroleconceptexistinse‐lectedtool,additionaldevelopmentforreali‐zationinworkflowenginehastofollow.

R8 Theinstallationandupdateofthesoftwareiseasyandfast.

Yes,theusersdislikeinstallingsoftwarere‐spectivelysystemsmoreoftenoratall.De‐liveryviamostcommon‘app’storeswouldbewelcomed.

R9 Thesystemrunsbestpossiblyplat‐form‐independentonstationaryandmobiletouch‐screendevices.

Yes,especiallyAppleiOSoperatingsystemshouldbesupportedforusers’mobiledevic‐es.

R10 Thesolutionneedstobelowcostsincefundingcriticalinthedomain,especiallyitsambulantscenarios.

Yes,thesolutiontobedevelopedcannotbefundedadditionallybytheprovider.Onlyfreewaretoolswiththenlimitedfunctionali‐tycanbeselected.

R11 Asupportandmaintenanceconceptisrequired.

Yes,ahandovertoproviders’ITserviceforpost‐livesupportispursued,buttofulfillR10oflimitedsignificance.



5 Discussion and Outlook

Wehaveelicitedrequirements,recommendedBPMNasacorrespondingprocessmodelingnotationincludingcorrespondingtoolsanddevelopedarepresentationapproach.

Thatwaspracticallychallengedbyapalliativecommunitynursingcaseaccordingtoaclinicalguideline.Thedegreeofintegrationofusersandtheirspecificneedswhilegener‐atingthesolutions,especiallymodelingtheworkflows,hasbeenadvancedregardingPAISconceptioninambulantscenarios.Someoftherequirementscouldbeaddressedbymeansofmobileapplicationsystemsrunningfreewareprocessmanagementtoolsforbothmod‐elinganddevelopingaPAIS.

Accordingtodesignscience,theevaluationshowsthatourapproachhasprovidedin‐sighttoaproblemoffieldservicedeliveryandsomemeasurestoaddressit.However,itisratheraqualitative‐argumentativeexperiencereportthananempiricalstudyfordemon‐stratingitspracticalusefulness.Morecaseswillbeneededtoproveitsoverallvalidity,pri‐oritizerespectivelydiscoveradditionalrequirementsormoresophisticatedwaystorepre‐senttheprocessesandtheirinstanceswithadequatetoolsupport.

Predefinedworkflowsrepresentsignificantroutinestofollow.However,theydependinteractivelyonsituation,mightalwaysbedecontextualized,incomplete,quicklyoutdatedandhave to be correctly interpreted (Berg andToussaint 2003). If this kind of process‐orientedsupportisprovidedviamobileITsystems,acceptanceofthesystemamongstus‐erslikecommunitynursesneedstobedetermined(Zhangetal.2010).Sameappliestotheclients’sidewhoaresuspiciousoftoolsforcorrectprovisionofservice(LenzandReichert2007)andwanttobeinvolvedmoreaboutproviders’ITuseintheirhomes(Wälivaaraetal.2009).Additionally,apositiveimpactontheoutcomeshastobeproven.Helpfulprovesthatmanyfieldserviceprovidersdoalreadyusemobiledevicesforprofessionalreasons:aPAISdeployedonthatcouldhelpwithlearninganddeliveringcertainservices(LenzandReichert2007)especiallyinundersuppliedmoreruralorvastareas.

References

Akesson,K.;Saveman,B.;Nilsson,G.(2007):Healthcareconsumers’experiencesofinfor‐mationcommunicationtechnology–asummaryofliterature.InternationalJournalofMedicalInformatics76(9):633–645

Ammenwerth, E.; Breu, R.; Paech, B. (2010): User‐Oriented Quality Assessment of IT‐SupportedHealthcareProcesses–APositionPaper.In:ProHealth2009Proceedings,617–622

An,Y.;Dalrymple,M.;Rogers,M.;Gerrity,P.;Horkoff,J.;Yu,E.(2009):Collaborativesocialmodelingfordesigningapatientwellnesstrackingsysteminanurse‐managedhealthcarecenter.In:DESRIST2009Proceedings,Article2

Ash,J.;Sittig,D.;Guappone,K.;Dykstra,R.;Richardson,J.;Wright,A.;Carpenter,J.;McCul‐len,C.;Shapiro,M.;Bunce,A.;Middleton,B.(2012):Recommendedpracticesforcom‐puterized clinical decision support and knowledgemanagement in community set‐tings:aqualitativestudy.BMCMedicalInformaticsandDecisionMaking12:6

Bates, D.; Bitton, A. (2010): The future of health information technology in the patient‐centeredmedicalhome.HealthAffairs29(4):614–621

Becker,J.;Janiesch,C.(2008):RestrictionsinProcessDesign:ACaseStudyonWorkflowsinHealthcare.In:ProHealth2007Proceedings,323–334

BenDhieb,A.;Barkaoui,K.(2012):OntheModelingofHealthcareWorkflowsUsingRecur‐siveECATNets.In:BPMWorkshops2011Proceedings,99–107



Berg,M.;Toussaint,P.(2003):Themantraofmodelingandtheforgottenpowersofpaper:asociotechnicalviewonthedevelopmentofprocess‐orientedICTinhealthcare.In‐ternationalJournalofMedicalInformatics69(2‐3):223–234

Birkhofer,A.;Deibert,S.;Rothlauf,F. (2007):Criticalsuccessfactorsformobilefieldser‐viceapplications:Acaseresearch.In:Wirtschaftsinformatik2007Proceedings,291–308

Breitschwerdt, R.; Robert, S.; Thomas, O. (2011): Mobile Application Systems for HomeCare:RequirementsAnalysis&UsagePotentials.In:AMCIS2011Proceedings,Paper152

Chatterjee,S.;Chakraborty,S.;Sarker,Sa.;Sarker,Su.;Lau,F.(2009):Examiningthesuc‐cessfactorsformobileworkinhealthcare:Adeductivestudy.DecisionSupportSys‐tems46(3):620–633

Ellershaw,J.;Wilkinson,S.(2003):Careofthedying.OxfordUniversityPress,NewYorkFrank,U.(2010):OutlineofaMethodforDesigningDomain‐SpecificModellingLanguages.

ICBResearchReport,vol.42,UniversityofDuisburg‐EssenGianchandani,E.(2011):Towardsmarterhealthandwell‐being:an implicitrole fornet‐

workingandinformationtechnology.JournalofInformationTechnology26(2):120–128

Heß, M.; Meis, J. (2011): Entwurf ausgewählter Spracherweiterungen zur Ressour‐cenmodellierung in Pflegedienstleistungsmodellen. In: Wirtschaftsinformatik 2011Proceedings,99–108(inGerman)

Hevner,A.(2007):Thethreecycleviewofdesignscienceresearch.ScandinavianJournalofInformationSystems19(2):87–92

Hevner, A.;March, S.; Park, J.; Ram, S. (2004): Design science in information system re‐search.MISQuarterly28(1):75‐105

Hong,H.;Kim,I.;Lee,S.;Kim,H.(2009):AdoptionofaPDA‐basedhomehospicecaresys‐temforcancerpatients.Computers,Informatics,Nursing27(6):365–71

Houy,C.;Fettke,P.;Loos,P.;vanderAalst,W.;Krogstie,J.(2011):BusinessProcessMan‐agementintheLarge.Business&InformationSystemsEngineering3(6):385–388

Jha, A.; Doolan, D.; Grandt, D.; Scott, T.; Bates, D. (2008): The use of health informationtechnology in seven nations. International Journal of Medical Informatics 77(12):848–854

Johnston,B.;Kidd,L.;Wengstrom,Y.;Kearney,N.(2012):AnevaluationoftheuseofTele‐healthwithinpalliativecaresettingsacrossScotland.PalliativeMedicine26(2):152–161

Jun,G.;Morrison,C.;O’Loughlin,C.;Clarkson,P. (2012):WhichDiagramsandWhen? In:DiagramsInternationalConference2012Proceedings,340–342

Kaiser, K.; Seyfang, A.;Miksch, S. (2011): Identifying Treatment Activities forModellingComputer‐Interpretable Clinical Practice Guidelines. In: KR4HC 2010 Proceedings,114–125

Kralik,D.,vanLoon,A.(2011):CommunityNursinginAustralia.Wiley,MiltonKunze,M.;Weske,M. (2010): Signavio‐OryxAcademic Initiative. In:BPM2010Proceed‐

ings,6–10Lee, T. (2007): Nurses’ experiences using a nursing information system. Computers, In‐

formatics,Nursing25(5):294–300.Legner,C.;Nolte,C.;Urbach,N.(2011):EvaluatingMobileBusinessApplicationsinService

andMaintenanceProcesses.In:ECIS2011Proceedings,Paper247Lenz,R.;Reichert,M.(2007):ITsupportforhealthcareprocesses–premises,challenges,

perspectives.Data&KnowledgeEngineering61(1):39–58



Mans,R.;vanderAalst,W.;Russell,N.;Bakker,P.;Moleman,A.(2010):Process‐AwareIn‐formationSystemDevelopmentfortheHealthcareDomain.In:BPMWorkshops2009Proceedings,635–646

Martinez, I.;Escayola, J.;Martinez‐Espronceda,M.;Serrano,L.;Trigo, J.; Led,S.;Garcia, J.(2009):ImplementationexperiencesofISO/IEEE11073standardappliedtonewusecasesfore‐healthenvironments.In:EMBC2009Proceedings,1679–1682

McCall,K.;Keen, J.; Farrer,K.;Maguire,R.;McCann,L.; Johnston,B.;McGill,M.; Sage,M.;Kearney,N.(2008):Perceptionsoftheuseofaremotemonitoringsysteminpatientsreceiving palliative care at home. International Journal of Palliative Nursing 14(9):426–431

Mouttham, A.; Kuziemsky, C.; Langayan, D.; Peyton, L.; Pereira, J. (2012): Interoperablesupport for collaborative, mobile, and accessible health care. Information SystemsFrontiers14(1):73–85

OMG(2006):BusinessProcessModelingNotationSpecification.OMGFinalAdoptedSpeci‐fication

OMG(2012):BPMNImplementers,http://www.bpmn.org/Österle,H.;Becker,J.;Frank,U.;Hess,T.;Karagiannis,D.;Krcmar,H.;Loos,P.;Mertens,P.;

Oberweis,A.;Sinz,E.(2010):Memorandumondesign‐oriented informationsystemsresearch.EuropeanJournalofInformationSystems20(1):7–10

Pedersen,R.;Ellingsen,G.;Monteiro,E.(2011):TheStandardizedNurse:MissionImpossi‐ble?In:IFIPWG8.22011Proceedings,168–178

Peleg,M. (2011):TheRoleofModeling inClinical InformationSystemDevelopmentLifeCycle.MethodsofInformationinMedicine50(1):7–10

Petersen,S.;Bach,G.;Svarlein,A.(2010):PatientCareacrossHealthCareInstitutions:AnEnterpriseModellingApproach.In:IFIPWG8.12010Proceedings,91–105

Picot,A.;Schmid,M.(2009):MobilisierungvonWertschöpfungsprozessendurchinnovati‐veundsichereInformationstechnologie.LMU,Munich(inGerman)

Pitsillides, A.; Pitsillides, B.; Samaras, G.; Dikaiakos, M.; Christodoulou, E.; Andreou, P.;Georgiadis,D.(2006):DITIS:acollaborativevirtualmedicalteamforhomehealthcareofcancerpatients.In:M‐Health.EmergingMobileHealthSystems.Springer,Berlin

Pryss,R.;Tiedeken, J.;Kreher,U.;Reichert,M.(2011):TowardsFlexibleProcessSupportonMobileDevices.In:CAiSEForum2010Proceedings,150–165

Pryss, R.; Tiedeken, J.; Reichert,M. (2010):Managing Processes onMobileDevices: TheMARPLEApproach.In:CAiSEForum2010Proceedings,Paper15

Reichert,M. (2011):What BPM Technology Can Do for Healthcare Process Support. In:AIME2011Proceedings,2–13

Rouse,W.(2009):Engineeringperspectivesonhealthcaredelivery:Canweaffordtechno‐logical innovation in healthcare? Systems Research and Behavioral Science 26(5):573–582

Rügge,I.(2003):MobileLösungenfürmobileTätigkeiten.In:MoCoMed2003Proceedings,101–107

Rügge,I.(2007):MobileSolutions.Dissertation,UniversityofBremen.DUV,WiesbadenSadeghi, P.; Benyoucef,M.; Kuziemsky, C. (2011): Amashup based framework formulti

levelhealthcareinteroperability.InformationSystemsFrontiers14(1):57–72Schoenbaum,S.;Gottlieb,L.(1990):Algorithmbasedimprovementofclinicalquality.Brit‐

ishMedicalJournal301(6765):1374–1376Segal, L.; Leach,M. (2011): An evidence‐based healthworkforcemodel for primary and

communitycare.ImplementationScience6:93



Sherry,J.;Ratzan,S.(2012):MeasurementandEvaluationOutcomesformHealthCommu‐nication:Don’tWeHaveanAppforThat?JournalofHealthCommunication17(1):1–3

Tachakra,S.;Wang,X.;Istepanian,R.;Song,Y.(2003):Mobilee‐health:theunwiredevolu‐tionoftelemedicine.TelemedicineJournalande‐Health9(3):247–257

Tong,A.;Sainsbury,P.;Craig,J.(2007):Consolidatedcriteriaforreportingqualitativere‐search (COREQ): a 32‐item checklist for interviews and focus groups. InternationalJournalforQualityinHealthCare19(6):349–57

vandeKar,E.;denHengst,M.(2009):Involvingusersearlyoninthedesignprocess.Elec‐tronicMarkets19(1):31–42

Varshney, U. (2007): PervasiveHealthcare andWirelessHealthMonitoring.MobileNet‐works&Applications12(2‐3):113–127

vonReibnitz,C.;Nussbaumer,G. (2011):VernetzteStrukturen inderambulantenPflegeamBeispielHomecare.HBScience.2(1):13(inGerman)

Wälivaara,B.;Andersson, S.;Axelsson,K. (2009):Viewson technology amongpeople inneedofhealthcareathome.InternationalJournalofCircumpolarHealth68(2):158–169

Walter, P. (2009):Modellierung technischer Kundendienstprozesse desMaschinen‐ undAnlagenbausalsBestandteilhybriderProdukte.In:DLM2008,129–145.Physica,Ber‐lin(inGerman)

Winston, P. (1980): Learning and reasoning by analogy. Communications of the ACM23(12):689–703

Yan,Z.;Reijers,H.;Dijkman,R.(2011):AnEvaluationofBPMNModelingTools.In:BPMN2010Proceedings,121–128

Zhang,H.;Cocosila,M.;Archer,N.(2010):Factorsofadoptionofmobileinformationtech‐nologybyhomecarenurses.Computers,Informatics,Nursing28(1):49–56

Living Lab Business Process Management e.V. Universität Osnabrück Katharinenstraße 3 49074 Osnabrück www.living-lab-bpm.de ISSN 2193-777X

Collaborative Design of Process-Aware Information Systems ...

Documents

Transcript of Collaborative Design of Process-Aware Information Systems ...