Envisioning and shaping translation of knowledge into action

(1)

ContentslistsavailableatScienceDirect

Health

Policy

jo u rn al h om ep a g e :w w w . e l s e v i e r . c o m / l o c a t e / h e a l t h p o l

Envisioning

and

shaping

translation

of

knowledge

into

action:

A

comparative

case-study

of

stakeholder

engagement

in

the

development

of

a

European

tobacco

control

tool

Robert

A.J.

Borst

a,∗

_,

_Maarten

_Olivier

_Kok

a,b

_,

_Alison

_J.

_O’Shea

c

_,

_Subhash

_Pokhrel

d

_,

Teresa

H. Jones

d

_,

_Annette

_Boaz

c

a_Erasmus_School_of_Health_Policy_&_Management,_Erasmus_University_Rotterdam,_P.O._Box_1738,₃₀₀₀_DR,_Rotterdam,_the_Netherlands

b_Amsterdam_Public_Health_research_institute,_Department_of_Health_Sciences,_Faculty_of_Science,_Vrije_Universiteit_Amsterdam,_De_Boelelaan_1085,₁₀₈₁ HV,Amsterdam,theNetherlands

c_Faculty_of_Health,_Social_Care_and_Education,_A_partnership_between_Kingston,_University_London_and_St_George’s_University_of_London,_Cranmer_Terrace, London,SW170RE,UnitedKingdom

d_Health_Economics_Research_Group,_Institute_of_Environment,_Health_and_Societies,_Brunel_University_London,_Kingston_Lane,_Uxbridge,_Middlesex,_UB8 3PH,UnitedKingdom

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received5December2018

Receivedinrevisedform18June2019 Accepted13July2019 Keywords: Stakeholderengagement Actor-scenariomapping Researchimpact Context Knowledgetranslation

a

b

s

t

r

a

c

t

Stakeholderengagementinhealthpolicyresearchisoftensaidtoincrease‘researchimpact’,buttheactive roleofstakeholdersincreatingimpactremainsunderexplored.Weexploredhowstakeholdersshapedthe translationofhealthpolicyresearchintoaction.Ourcomparativecase-studytrackedaEuropeanresearch projectthataimedtotransferanexistingtobaccocontrolreturnoninvestmenttool.Thatprojectalso aimedtoincreaseitsimpactbyengagingwithstakeholdersinfurtherdevelopingthetool.Weconducted semi-structuredinterviews,usinganactor-scenariomappingapproach.Actor-scenarioscanbeseenas relationaldescriptionsofafutureworld.Wemappedthescenariosbyaskingstakeholderstodescribe whoandwhatwouldplayaroleinthetool’sutilisation.Ourresultsshowthatstakeholdersenvisioned disparatefuturesforthetool.Somescenarioswerespeciﬁc,whereasmostweregenericprojectionsof abstractpotentialusersandresponsibilities.Weshowhowstakeholdersmobilisedelementsofcontext, suchaslegislativesupportandagriculturalpractice,thatwouldaffectthetool’suse.Weconcludethat stakeholdersshapeknowledgetranslationprocessesbycontinuouslyputtingforthexplicitorimplicit scenariosaboutthefuture.Mappingactor-scenariosmayhelpinaligningknowledgeproductionwith utilisation.Insightsintopotentialrolesandresponsibilitiescouldbefedbackinresearchprojectswith theaimofincreasingthelikelihoodthatthestudyresultsmaybeused.

1. Introduction

Thepracticeofstakeholderengagementinknowledge produc-tionisgainingincreasingtractioninresearchfundingdebates[1–3]. Oneoftheprincipalreasonsforengagingstakeholdersisthatit mightincreasethelikelihoodthatresearchoutputswillbeused [4,5].Inpractice,researchersoftenretrospectivelyattributetheuse

∗ Correspondingauthor.

E-mailaddresses:borst@eshpm.eur.nl(R.A.J.Borst),

m.kok@eshpm.eur.nl(M.O.Kok),A.Oshea@sgul.kingston.ac.uk(A.J.O’Shea), Subhash.Pokhrel@brunel.ac.uk(S.Pokhrel),teresa.h.jones@googlemail.com (T.H.Jones),A.Boaz@sgul.kingston.ac.uk(A.Boaz).

Twitter:@RAJBorst(R.A.J.Borst).

oftheirﬁndingstotheirengagementwithstakeholders[6].Others emphasisetheimportanceofprospectivelyexploringhow stake-holderengagementprocessesevolveandaffectthetranslationof knowledgeintoaction[7].Howstakeholderengagementshapes theuseofknowledge,andwhichrolesstakeholdersplayinthis, hadbeenlargelyunderexplored[8,9].

The literature suggest that stakeholder engagement affects knowledgetranslationindifferentways.First,stakeholdersmay addvaluableknowledgeandskillstotheresearchprocess[10,11]. Second,stakeholderspossessexperientialinformationaboutthe environmentinwhichtheresearchﬁndingsmightbeused.Such informationcanbeusedtoaligntheresearchprocesswiththe envi-ronmentinwhichtheresearchﬁndingscouldbeused[7].Third, bybeingengaged,stakeholdersgainabetterunderstandingofthe prospectivestudyresults.Thiswouldinformthestakeholdersof

https://doi.org/10.1016/j.healthpol.2019.07.012

(2)

thestudytakingplace,butalsoencouragesthemtothinkabout potentialuseof theresultsin practice[12,13]. Finally, engage-mentcanestablishatrust-relationshipbetweenresearchersand potentialusers. Trustis essentialfor mutualunderstandingand communicationbetweenactorsandincreasesthepresumed legiti-macyofresults[14,15].Oliver,KothariandMays[16]concludethat stakeholderengagementisgenerallyconsideredtomakeapositive contributiontoresearchprojects,butmayinducechallengesand costsaswell.Inparticular,theycallformorereﬂectiononwhento engagestakeholdersinresearchandinwhatway[16].

Scienceandtechnologystudies(STS)emergedasaconstructivist interdisciplinaryfieldinthelate1970sandisknownfor study-ingknowledgeproductionpracticesandtherolethatusersplay intheseprocesses [17–19].It seemsparticularlywellequipped toreflectontherole of stakeholderengagementin health pol-icyresearch.STSscholarsofferaconceptualisationof‘translation’ thatisdifferenttothosecommonlyusedinhealthpolicyliterature [20,21].MuchofthisconceptualisationisgroundedinwhatCallon [22]calls‘sociologyoftranslation’andwhichlaterbecameknown asactor-networktheory[23].Accordingtothisconceptualisation, knowledgetranslationcanbeseenasaprocessof(political) activ-itiesbywhichactorsactivelydisplaceandtransformknowledge [24].Fromsuchanunderstanding,translationisaboutnegotiation, transformation,andtheassociationsbetweenactorsthroughwhich networksarebuiltandextended[23].Thestrengthofthis under-standingisthatitoffersanin-depthunderstandingoftheactiverole ofpotentialknowledgeusersintranslation,theworkthatis neces-sarytomakeknowledgeusable,theroleofnon-humanactors(e.g. materialenvironments),andaspecificconceptualisationoftherole ofcontextintranslatingknowledgeintoaction[17,25].

Atheoreticalaspectthatremainsunderexploredintheliterature onstakeholderengagementishowstakeholdersthemselves envi-siontranslationofknowledgeintoaction[26].Inparticularafocus onpotentialusersandtheroletheyplay inshapingknowledge usecouldincreaseunderstandingofhowstakeholderengagement affects knowledge translation processes. Stakeholders’ perspec-tivesontranslationofstudyﬁndingsintoactioncanofferinsight intotheworldinwhichtheﬁndingsmightbeused,including nec-essaryrolesandresponsibilities.Stakeholderscanbringforward differentaccountsof thefuture world,withdifferentroles and responsibilities[27].

Toscrutinisehowstakeholderengagementinknowledge pro-ductionshapestheuseofsuchknowledgeinpractice,thisstudy prospectivelyfollowedstakeholderengagementinalargeresearch projectfundedbytheEuropeanCommission(EC).The European-study on Quantifying Utility of Investment in Protection from Tobacco(henceforth:project)centredaroundthetransferofan evidence-basedtobaccoreturn-on-investment(ROI)tool(seeBox 1).Theprojectexplicitlyplannedtoengagewithstakeholdersto increasetheproject’s‘impact’[28]. Thecase-studyathandwas partoftheparallelStakeholderEngagementinEQUIPTforImpact (SEE-Impact) study.Our aim was to envision how stakeholder engagement shapes the translation of the ROI tool into action bymappinghowstakeholdersthemselvesputforwardscenarios aboutthepotentialuseoftheROItool.Itisanticipatedthatthe ﬁndingsofthisstudywillcontributetothedevelopmentof stake-holderengagementinresearchasamethodforsupportingresearch use.

2. Methods

Forthisin-depthcase-study,wedrewondatafrom21 ethno-graphic interviews in Hungary and the Netherlands that were conductedaspartoftheSEE-Impactstudy.Thesetwocountries werepartofthefourcountries(i.e.Hungary,theNetherlands,Spain,

Box1:DescriptionoftheSEE-Impactstudyinrelation

totheEC-fundedEQUIPTproject

Studyingengagementinthedevelopmentofa

tobacco-controltool

The project under study was funded through the Euro-pean Commission’s Seventh Framework Programme. The European-studyonQuantifying UtilityofInvestmentin Pro-tectionfromTobacco(EQUIPT)wasacollaborationbetween 11 members from seven countries (i.e. Belgium, Croatia, Germany,Hungary,theNetherlands,Spain,andtheUK),and wasledbytheHealthEconomicsResearchGroup(HERG)from BrunelUniversityLondon.TheprojectcommencedinOctober 2013and ended September 2016. Their aim was to assess the “cross-context transferability of economic evidence on tobaccocontrol”whichledthemtofurtherdevelopanexisting ROItoolforuseinotherEUcountries.Aspartoftheirproject, theytestedthetoolintheNetherlands,Germany,Spain,and Hungary[28].

The existing tool had been developed in the UK by the HERG in conjunction with the National Institute for Health andCareExcellence(NICE)andisavailableontheNICE web-site(http://bit.ly/tobacco-roi).Thetoolallowsuserstocalculate savingsforeverymonetaryunitinvestedincertain tobacco-controlorsmoking cessationinterventions.Thestakeholder engagementintheEQUIPTprojectwasinformedbythe suc-cessful stakeholder engagementin theoriginal UK project. Thishad contributedto theoriginal ROItoolbecoming the NICE’ssupporttoolforEnglishlocalauthorities,which even-tuallyinformedthesmokingcessationapproachesofseveral localauthorities.

Thequalitative case-study presentedinthispaperwaspart oftheStakeholderEngagementinEQUIPTforImpact (SEE-Impact) study, funded by the Medical Research Council in the United Kingdom. SEE-Impact prospectively tracked all stakeholderengagementactivitiesinEQUIPTwiththeaimof describingtowhatextentengagementaffectsresearchimpact. TheSEE-Impactstudycollecteddatathroughliteraturereview, surveys,semi-structuredinterviews,andobservations.

andGermany)towhichtheEuropeanresearchprojectaimedto transfertheirtobaccoROItool.Forthepurposeofourstudy,we selectedHungary andtheNetherlandsascontrastingcases[29]. Thesecountrieshaveverydiversesocioeconomicandpolitical con-textsrelevanttohealthpolicy.Particularlyrelevanttothisstudy isthecountries’differenceintobaccopoliciesandsmoking preva-lence[30,31].Atthetimeofthisstudy,Hungaryhadmorestringent tobaccocontrolpoliciesthantheNetherlands,butahigher smok-ingprevalence[32,33].MoreinformationontheSEE-Impactstudy anditsmethodscanbefoundelsewhere[9].

2.1. Interviewees

Wesampledstakeholderswithdifferentlevelsofengagement. ThefirstgroupofstakeholdersconcernedpartnersoftheEQUIPT project.Thesecondgroupconsistedofactorsthatwereinvitedby theEQUIPTprojecttoprovideinputinthecontinueddevelopment ofthetool.Thefinalgroupincludedactorswhocouldhavebeen approachedbytheEQUIPTresearchers(i.e.theybelongedto sim-ilarnetworksasthesecondgroup),butwithwhomnointeraction hadoccurred.Weselectedthefinalgroupofactorsbasedontheir substantiveexperienceintobaccocontrolorhealthpolicywithin eachcountry.

2.2. Datacollectionandanalysis

Atotalof21interviewswitheightHungarianandelevenDutch stakeholderswereconducted.Theintervieweesweremostly

(3)

aca-Box 2: Description of actor-scenario mapping as an

approachtostudyingtheuseofknowledge

Actor-scenariomapping

BuildingonMichelCallon’snotionof‘actorworlds’[24],we use the concept ‘actor-scenario’ to refer to the process of actors implicitly or explicitly putting forth scenarios about practicesinafutureworld[27,34].Anactor-scenariocanbe seenasarelationaldescriptionofpotentialpractices,roles, and responsibilities. Actor-scenarios are fictive atfirst, but performativeaswellsincetheyincludedescriptionsofwhat shouldhappenforthe scenariostobeenacted[35,36].The practiceof scenario-building works asongoing negotiation processthroughwhichactorsaimtoeffectuatechange[24,37]. Researchers, for instance, constantly put forth implicit or explicitaccountsoftherolethattheirfindingsshouldplayina futureworld.

Differentactorsmayconstruct differentscenariosthat each portraytheirownrolesandresponsibilities.Somepartsofthe scenariosmightoverlap,whereasothersdiverge.Theactors thatareenrolledinthescenarioscanalsorefutetheirroleand produceadifferentscenariowithotherrolesand responsibil-ities.Oneofsuchrolesmightbereservedforknowledge,for instancetostrengthenascenarioorweakenscenariosof oth-ers[34].Mappingtheactor-scenariosofthestakeholdersinthe EC-fundedprojectmayexplicatewhothestakeholdersthink shouldusethetool,howthetoolshouldbeused,andunder which circumstances use is possible. We developed ‘actor-scenariomapping’asanapproachtoenvisionanddescribe possibletranslationsoftheROItoolintoaction.

demicsworkinginhealthpolicy,healthtechnologyassessment,or epidemiology(n=10),followedbygovernmentofﬁcialsand parlia-mentarians(n=5),andclinicians(n=4).

Weusedatheoreticalframework(seeBox2)toguidethe ethno-graphic interviews. In particular, we developed topic lists that speciﬁcallysoughttomapactor-scenariosbyaskinginterviewees tothinkofhowthetoolwouldbeusedinthefutureandwhowould playaroleinthatuse.Thisapproachaddedsomestructuretothe interviews,butallowedforasubjective,anticipatory,explorationof topicsthatdidnotdirectlyalignwiththeconceptofactor-scenarios [37].Duringdatacollection,threetopicsthatwereregularly men-tionedinthescenarioswereaddedtothelists(i.e.decentralisation ofpublicservices,earlierexperiencewithstakeholderengagement, andpolitics).Allinterviewswereaudiorecordedandthe interview-erskeptdetailednotesduringtheinterviews.Immediatelyafter eachinterview,reﬂectionalmemoswerepreparedandrecordings weretranscribedverbatim.

Theprocessofdatacollectionandanalysiswasconducted itera-tively.Thisapproachallowedtheresearcherstoidentifyemerging themes suitable for subsequent ﬁeldwork. Actor-scenario map-pingusesanabductivesequenceofanalysisthatrequiresconstant shiftingbetweentheoryand empiricalﬁndings[38]. Theaimis tosynthesisethedifferentscenariosandofferthickdescriptions ofpotentialtranslations,includingthedifferentenvisionedroles andresponsibilities.Thepotentialtranslationsinthisstudywere arrivedatthroughrepeatedin-depthcodingsessionswithallteam members.

2.3. Researchethics

The data collection of this study adhered to the Decla-ration of Helsinki and ethical clearance was obtained from KingstonUniversityLondon’sFacultyResearchEthicsCommittee (FREC2014/01/011).Accordingly,theresearchersobtainedwritten informedconsentoftheintervieweesandtheinterviewers explic-itlystatedthattheanonymisedresultswouldbepublished.

2.4. Studyschedule

ThisstudywasconductedbetweenFebruary2015andMarch 2017.ThedatacollectionwascarriedoutbetweenApril2015and September2016.

3. Howstakeholdersenvisionedthetooltobeused

Theenvisionedusesofthetoolweresituatedandshapedby local-speciﬁc dynamics and elements of context. Conventional withactor-networktheory,wewillprovideseparatedescriptive accountsofhowHungarianandDutchintervieweesenvisionedthe useofthetool.Wewillstarteachsectionwithdescribingtheroles andresponsibilitiesputforthbythestakeholders,followedbywhat thestakeholdersdescribedaspotentialenablingorconstraining elementsofcontextintheuseofthetool.

3.1. ThepotentialusersinHungary

Theactor-scenariosofHungarianstakeholderswereoftenquite similar.TheHungarianstakeholders,forexample,alldesignated theNationalFocalPointfor TobaccoControlarole asuser.The focalpoint, theydescribed,would bea suitableuserbecauseof theirexperiencewitheconomic evaluationsand embeddedness withintheofficialhealthsystem.Severaloftheintervieweesspoke ofa specificperson withinthefocalpoint. Theydescribedhow thispersoncouldusethetooltocompareinterventionsontheir cost-effectiveness,andhow“hefeedstheMinistrywithhisdata.” (Clinician1).Theyalsostressedtheimportanceofthefocalpoint beingappointed bythegovernment. This–combined withthe focalpoint’sstatusasWHOpartneroffice–wouldlegitimisetheir recommendationsamongstpolicymakers.

Somestakeholdersassigned theNationalInstituteofTB and Pulmonologyaroleintheirscenarios.Anintervieweeworkingat theNationalPublicHealthandMedicalOfﬁcer’sService(ÁNTSZ) describedthattheydidalotoftheirsmokingcessationactivities togetherwiththeNationalInstitute.Theintervieweedescribedthat theNationalInstituteisveryactiveinthisﬁeldandwouldlikely beinterestedin thetool.Whenwe askedoneoftheInstitute’s employeeswhethertheywoulduseit,theintervieweesaidthat they“would tellaboutit[thetool] and(...)wouldteachwithit.” (Clinician1)

In addition,stakeholders commonlymentioned the National HealthInsuranceFund.Whilenearlyallrespondentsassignedthe Fundsomeroleintheirscenarios,theyarticulateddifferent respon-sibilities.AnepidemiologistspokeoftheFundasthe“mostlikely user”anddescribedthattheFundcouldusethetool’soutputas “ammunitiontoargueforsomeservicestobereimbursed”.Aclinical professorarguedthattheFundmayuseanefﬁcacycomparisonof smokingcessationprogrammes,althoughthiswouldstillbe“abit furtherawayfromtheirfocus”.OthersexplainedthattheFundcould provideﬁnancialdatabutwouldotherwisenotbeinterestedin tobaccoissues.ThescenarioofaformerFundemployeeresembled scepticismabouttheFund’sresponsibilities:

“Theywould,directly,notbeinterested;evenifofﬁciallytheyneed tobeinterested.(...)Frankly,theyaregoingtohaveanew inter-ventionthatwouldneedtobereimbursed.So,theirbudgetwillbe lower.(...)Incaseitiscost-saving;thenitisﬁne.But,thatwill probablynothappen.Itwillnotbecost-saving.”(Academic1) FewintervieweesmentionedtheSecretariatofHealthasa pos-sibleuserofthetool.Thosewhodid,describedthatpolicymakers insidetheSecretariatcouldusethetooltoprioritisetheirdecisions onwhichinterventionstoimplement.

(4)

3.2. EnvisionedtranslationinHungary

Throughout the interviews in Hungary, a pattern emerged showinghowelementsofcontextwouldenableorconstrainthe potentialuseofthetool.Mostintervieweesarticulatedidentical elementsof context,commonly referring to thenewly enacted tobaccolegislationof2012.Arespondentthatwasinvolvedin writ-ingthe2012legislation,explainedthestrategicworknecessaryto establishit:

“Wecalculated:itwasDecember,theﬁrsttimethatwecouldreach theParliamentwouldbemid-April.Wedidnottrustoursystem–in awaythatthisvoicewentoutearlyonlasttime.So,wedidthe professionalwork–theplanningofthelaw–andthentheState Sec-retarydiscusseditinsidetheParliament.Whathappenedwasthat thelaw,plannedandwritten,wasgiventoParliamentwherea groupofparliamentarianssaidtogether:‘wearefromtheleading partyandwethinkitisaveryimportantpublichealthproblemin Hungary,wemustchangeit,now!’Andintwoweeks’time,itwas votedon.Thatwasprobablytheonlylaw,inthelightofyears, whereleftwingandrightwing,whateverwing,theyallvoted.And itwassomethingclosetoaninetypercentpositivevote.”(Clinician 1)

Severalrespondentsexplainedthatwiththelegislation’s enact-ment,tobaccoretail wasrestricted,smokingin conﬁnedspaces prohibited,andexcisetaxeswereincreased.Theystatedthatthere isnoneedforaROItool,asthereisbarelyroomleftforadditional interventions.

“IfyouevaluatetheactualsituationinHungary,weachieved prac-ticallyeverything.Thereisnospace.So,weareatthetop,ifrelated tolegislation.But,thereweresomeconcernsthatyourprivatecar isaconﬁnedspace.”(Clinician2)

Anotherdynamicthatsomeoftheintervieweesmentionedwas thedecentralisedand segmentedgovernment. Therespondents illustratedthatthereisatensionbetweentwoorganisationsboth operatingatthelocallevel.Ontheonehand,therearethecounty publichealthdepartments,runbythecountygovernmentoffices anddirectedbythePrimeMinister’sOffice;ontheotherhand,there arethemunicipalhealthpromotionofficesadministeredbythe ÁNTSZonbehalfoftheSecretaryofHealth.Oneoftherespondents explainedthattheseorganisations’similarresponsibilitiescause regulartensions.

“Theyareseparatedandthereareconflicts.Because,theyare work-ingonsimilarissues.Theconflictsarebecausetheydon’treallylike eachothertoworkonthesameissue.”(Governmentofficial1) Additionally,severalparticipantsdescribedthatthetool’suse mightbeconstrainedbytheprominentplacetobaccoagriculture takes in Hungary. Theydescribed that theMinistry of Agricul-turehasaprevailingroleinHungarianpolicymaking.Besides,this Ministry’smaininterestwouldbethetobaccocultivationinthe North-EasternpartofHungary.

“TheMinistryofAgriculture,forexample,isverymuchopposedto regulatingtobacco.Becausetheythinkthat,Idonotknow,these fewthousandpeopleshouldgrowtobaccoandnothingelse.Inever understoodwhynottogrowpaprikainstead,butokay.”(Academic 2)

Anotherrespondentexplainedthatitisaconcurrenceof sev-eralcircumstancesthatcomplicatesthetranslationofevidenceinto anti-tobaccopolicies.Theintervieweedescribedhowactorssuchas theeducationalsystem,soil,precipitation,temperature,andmoney positionthemselvesas“tobaccoallies”andconstraintheenactment ofanti-tobaccopolicies.

“Tobaccopolicydependspracticallyontheagriculturaltraditionof thecountry.So,youneedaspecialsoiltogrowtobacco,andthe specialcircumstancesrelatedtotemperature,precipitation,andso on.Thebestregionfortobaccoplantationsinthiscountryisthe least-developedpart,namely:TheNorth-Easternpart.”(Clinician 2)

Anelementthatappearedtobelinkedtothetobacco agricul-turewas Hungary’shistory of communism.A former politician explainedthatexcessivesmokingwasacommonhabitduringthe Sovietera.DuringserviceintheRedArmy,therespondentwould receiveadailyamountof15cigarettesregardlessofwhetherone smokedornot.ThesecigaretteswouldbesuppliedbyState-run tobaccoplantationsaspartoftheplannedeconomy.Afterthefall ofcommunism,theproprietorshipwastransferredtothecorporate tobaccoindustry.Theintervieweesaidthatitwasonlybythenthat themedicalcommunityﬁrstinitiatedananti-tobaccocommunity withthephilosophytoreducetobacco-relatedharm.

StakeholdersinHungaryregularlyspokeofthesameactorsin theirscenariosaboutthepotentialuseofthetool.Some stakehold-ersconstructedslightlydifferentscenarios.Overall,thereseemed toariseconvergenceinstakeholders’narrativesaboutenablingand constrainingdynamicsinthepotentialuseofthetool.

3.3. ThepotentialusersintheNetherlands

UnlikeinHungary,theactor-scenariosofDutchstakeholders showeddivergence.Intervieweesdescribedthatthetoolwouldnot beusedatall,orthatitsusewouldbeconstrainedbywhatwas referredtoas‘thepoliticalclimate’.Sometimesparticipants men-tionedspecificorganisations,butusuallyexpresseduncertaintyas towhethertheseorganisationswould actually usethetool. All intervieweesassigned‘policymakers’aroleintheirscenarios,but withoutspecifyingwhothisactorisinpractice.Whilesome scenar-iosweremorespecific,moststakeholdersdidnotarticulatewhat theroleandresponsibilityofpolicymakersspecificallywouldbe.

Intervieweescommonlysaid thatthemunicipalgovernment andMunicipalityHealthService(GGD)wouldplayarole.One pro-fessorinhealthpolicyexplainedthatthemunicipalgovernment mightusethetooltoguidetheirserviceprocurement.Twotobacco controlexperts,however,describedthatanti-tobaccoincentives arenot themunicipality’spriority;theirpolitical accountability causesthemtoprioritiselesssensitiveissues.Theexpertsdescribed thatthemunicipalgovernmentsdonotallocatetheGGDsany anti-tobaccorelatedtasksandthattheGGDsdonothavetheresources tocarryoutanti-tobaccoactivitiesthemselves.Eveniftheyhad, theywouldnothavetheexpertisetouseROItoolsforit,asone governancescholarexplained.

SeveralrespondentsmentionedtheNationalInstitutefor Pub-licHealthandEnvironment(RIVM)asapotentialuser.TheRIVM functionsasadvisorybody tothegovernment.Oneinterviewee explainedthattheRIVMaspotentialadministratorcouldstorethe toolandupdateitifnecessary.Twointervieweesclariﬁedthatthe RIVMusedtodeploysimilartoolstoanswertobaccocontrol ques-tionsraisedby theMinistryof Health.Anintervieweeformerly activeintobaccocontrolsharedanarticlethatshowedhowthe RIVMusedtoproduce‘scenarios’:predictionsoftheimpactthat certaincombinationsofanti-tobaccointerventionscouldhave.

“Itwasnotatoolinwhicheveryonecouldtwisttheknobs,itwas quitecomplex.(...)So,allthesescenarioswerealreadythere.(...) Because,backthen,theMinistrystillgavetheRIVMsuchorders.” (Academic3)

AnintervieweeworkingattheRIVMclaimedtorecognisethat theynolongerreceiveanyordersfromtheMinistrytoestimatethe return-on-investmentoftobaccocontrolinterventions.

(5)

“Theassumption of this European tool is that policymaking is mainlymotivatedbyrationalconsiderations;whereasinpractice, thatisobviouslynotthecase.Suchatoolcouldhelptostimulate this,thatmakessense.But,IamnotsurewhethertheRIVMwould useittoanswerquestionsoftheMinistry.Thatwouldmeanthat thereisasituationinwhichapolicymaker,atlocalornationallevel, says:‘wewanttodiscouragetheuseoftobacco,thisistheamount offunding,thesearetheconditions,nowwhatwouldbethemost efﬁcientuseofourresources?’Well,thatisalaboratorysituation thatwillneverhappeninpractice.”(Governmentofﬁcial2) Theintervieweesdisagreedontherolethatresearcherscould play.Oneacademicsaidthatacademicresearcherswouldusethe tooltoevaluatethecost-effectivenessof interventions.Someof theprojectmembersplannedforacademicstoadaptandupdate thetool. Twopublichealthacademicsspoke ofresearchersata nationalinstituteformentalhealthandaddictionusingthetoolfor monitoring.Nearlyallinterviewees,however,describedthatthis activitywouldbeabitfurtherfromtheinstitute’scorefocus–since tobaccocontrolisnotincludedintheirmandate.

Respondentsoftenarticulatedgenericideasofwhomightbe interestedorcapableinusingthetool.Occasionally,theseideas wererefutedbyotherrespondents.Overall,thereappearedtobe multipledeviatingscenariosaboutthepotentialuseofthetool.The likelihoodofthescenariostobetranslatedintoactionseemedtobe affectedbydynamicsinthecountries’context.Theelementsthat werementionedbytherespondentsareportrayedinthe subse-quentsections.

3.4. EnvisionedtranslationintheNetherlands

AnelementthatprevailedinthescenariosofDutchactorswas thepoliticalclimate.ThemajorityofDutchintervieweesreferredto twoactsofthethenMinisterofHealthin2010.Theﬁrstbeingher emphasison‘de-patronisation’withregardtoanti-tobacco legisla-tion.ThesecondactwastheMinister’srepealofthesmokingbanfor smallrestaurantsandbars.Oneintervieweeclearlyremembered theMinister’spositionontobaccocontrol:

“WewenttotheMinistryin2010andofferedtheMinisterapetition againsttobacco,withover1000signatures.So,wevisitedherand shesaid:‘well,Ireallythinkthis[smoking]isafreechoiceandIam notsofondofstatistics.’Thatiswhatshesaidinthatconversation. [raisingvoice]”(Clinician3)

Intervieweesoftenspokeoftheclosingofthenationalexpertise centreontobaccocontrolin2013.Aformeremployeeexplained thattheMinistryofHealthsuspendeditsfundingby2011. Subse-quently,thehealthfoundations,responsiblefortheotherhalfofthe funding,decidedtoindependentlyproﬁlethemselvesmoreactively ontobaccocontrol.Theex-employeeexplainedthatsomeofthe activitiesweretransferredtootherorganisations.Nonetheless,the majorityofthecentre’spromotionalactivitieswereabandonedand itremainsunclearwhoshouldﬁllthatgap.

Whiletherespondentstriedtoidentifypotentialusersofthe tool,theysaidthatitisactuallyquiteunclearwhogovernstobacco controlintheNetherlands.Onelocalgovernmentofﬁcialexplained thattheMinistryofHealth stipulatesquadrennial national pre-ventionprioritiesthatshouldguidethemunicipalgovernmentsin prioritisingatthemunicipallevel.Themunicipalitywouldthenbe ofﬁciallyresponsiblefortheexecutionofprevention.But,several intervieweesexpressedthattobaccopreventionmightnotbethe municipalities’uppermostpriority.

Whenweaskedwhetherpoliticswouldplayarole,nearlyall actorsreferredtopoliticalincentives.AMemberofParliamentfor theLabourpartydescribedthattensionswithinthethenminority cabinetwouldpreventfutureanti-tobaccointerventionsfrom

gain-ingtraction. Accordingtotheinterviewee, anti-tobaccopolicies doresonatewithintheLabourparty,butthecoalitionagreement refrainsthemfromacting.Otherinterviewees–whousedtowork ontobaccocontrolfora longerperiod– seemedsceptical:they indicatedthatParliamentarianswouldfocusonincreasingthe gov-ernmentbudgetwithinthefour-yearcycle,andtobacco-control doesnotﬁtthatagenda.

“Thecurrent political landscape is fragmentedand there is no majorityforamorestringentpolicyonsmoking.(...)Thereare actually two opposed sides: the conservative-liberalist side on whichitisafreedomofchoice,andthesocio-democraticthatsays: tobaccoisaperverseincentiveofthegovernmenttocomplement thetreasurechest.”(Parliamentarian)

Finally,therespondentsoftenspokeofarecenthistoryfullwith majorhealthsystemreforms.Anintervieweebelievedthatthese leftlittleroomforfurthertobaccocontrollegislation.The intervie-weeexplainedthattheformerMinisterofHealthimplementedthe banonsmokinginconﬁnedspacesandmajorreformsofthepublic healthlaw.Theinterviewee’spredecessorcompletelyreformedthe healthcaresystem,whereascurrentlythefocusisonredistributing powerbetweenhealthinsurersandmedicalprofessionals. Anti-smokingdidnothaveplaceonthepoliticalagenda.

4. Discussion

This study aimed to explore how stakeholder engagement in knowledge production shapestheuse ofsuchknowledge in practice. In order to explore this, we studied engagement of stakeholdersinthecontinueddevelopmentofatobaccocontrol return-on-investmenttool.Weaskedthestakeholderstoputforth anexplicitscenarioaboutwhich actorswould usethetool and underwhich circumstancesthetool couldbeused.Most stake-holdersdescribedthattheyfounditdifﬁculttoidentifypotential usersofthetool.Whilemoststakeholdersenvisionedquitea local-speciﬁcscene,theysetthestagewithexceedinglygenericpotential usersandresponsibilities.AnexamplewasthatmostDutch inter-vieweesspokeof‘policymakers’aspotentialusersofthetoolbut weremostlyunabletoidentifytheseactorsinpractice.

Ourmappingofactor-scenariosoffersthreeobservations rel-evanttostakeholderengagementinknowledgeproduction.First, wehaveintroducedaspeciﬁcunderstandingofstakeholders’rolein knowledgeproduction.Weshowedhowstakeholdershaveimplicit orexplicitunderstandingsofhow,bywhom,andunderwhich cir-cumstances,theproducedknowledgemaybeused,orwhatmakes the produced knowledge relevant and usable.By engaging the stakeholders,theserenderingsofafutureworldwillstart interact-ingwiththescenariosoftheknowledgeproducers,whothemselves inscribetheirproducedknowledgewithassumptionsaboutthe cir-cumstancesunderwhichtheproducedknowledgemaybeused [34,39].Itistheseinteractionsthatwillshapetosomeextenthow theknowledgemaybetranslatedintoaction.Thiscanbe espe-ciallychallengingwhentheactor-scenariosofstakeholdersseem todiverge,aswasthecaseintheNetherlands.Thediverging actor-scenariosmayinducedisputesoverhowthesediverseinputsof stakeholderswillbetreatedintheproductionofknowledgeorwho willbe‘theuser’oftheproducedknowledge[14,19].

Asecondobservationisthatactor-scenariomappingcan pro-videanempiricalunderstandingofwhatknowledgeuseentailsin practice.Inparticular,ourapproachreiteratesthatknowledgein itselfdoesnothaveaunivocalvalue,butrequiresactiveworkin ordertobecome‘usable’.Thisworkinvolvesa cleararticulation ofwhatrolesandresponsibilitiestheknowledgerequires.Inthe Netherlandsforexample,thestakeholdersoftheEuropeanproject triedtoidentifyanactorthatwouldgoverntobaccocontrol.

(6)

Accord-ingtotheinterviewees,suchanactorwouldbeaprerequisiteinthe useofthetool.Theyexplainedthatwithoutthisgoverningactor,it wouldbeimpossibletoactupontheknowledge.Similarly,the Hun-garianrespondentsconstitutedthefocalpointfortobaccocontrol bydescribinghowitwouldhavetobecomethekeyuserofthetool. A final observation to consider is how actor-scenario map-pingcontributes toourunderstandingof therole ofcontext in knowledgetranslation.Ourfindingsshowhowactorsconstructed contextbybringingforthascenarioofrolesandresponsibilities. Thisimpliesthatcontextisnotsomethingclearlydefinedandfixed, butinsteadreferstoafluidandcontingentnetworkofelements [40].Intheirscenarios,thestakeholdersdescribedthatsomeactors anddynamicswouldconstitute‘contexts’thatcouldenableor con-straintheuseof thetool [41]. Dutchintervieweesfor instance spokeofthethenMinisterofHealth,localgovernments,and his-toricallegislative reforms.StakeholdersinHungary would refer tothesoil,precipitation,andahistoryofcommunist-rulethatall constraintobaccocontrol.Intheirscenarios,thestakeholders con-structedafluidboundarybetweencontentandcontexttoaccount foradifferencebetweenuseandthatwhatshapespotentialuse, andsubsequentlymobilisedtheactorstheydesignatedascontext [17,42].

Severalquestionsaboutstakeholderengagementinknowledge productionremain.A firstquestionis whattheconvergence or divergenceofdifferentactor-scenariosmeans.InHungary,the sce-nariosappearedtobemorespecificand convergingthaninthe Netherlands. The data suggest that some actorsare embedded andentangledinnetworksindifferentways;potentiallyrelating moretolocal-specificactors[24].Conversely,anunderstandingof divergencemightbefoundinthegenericuserdescriptions(e.g. pol-icymaker).Shove&Ripsuggestthattheuseofsuchuniversallabels resultsinamismatchwiththerolesthatactorsconstructinpractice [43].Althoughthestakeholdersmightallconsiderthepolicymaker tobeaprimaryuser,theiridentificationsofthisactorinpractice arelikelytodiffer.Thiscontradictioncreatesaproblemof transla-tionwhenthisgenericnotionofusersisinscribedintheproduced knowledge[19].Ithasbeendescribedbeforehowthese inscrip-tionshaveperformativeeffects:theinscriptioncouldforinstance impairusebyanyoneotherthanthisgenericuser[35,44].Lastly,it isimportanttonoteherethatintheUK,wherethedevelopmentof theROItoolfirstoriginated,engagementwithstakeholders–being actorswithanexplicitstakeinboththedevelopmentanduseof thetool–hadbeenakeyelementofthesuccessoftheproject.This observationhasimplicationsforthenatureofstakeholder engage-mentinresearch–inparticularinsofarstakeholderengagement isconcernedasmeanstoincreaseresearchuse[45].

Inthisstudywedevelopedactor-scenariomappingasapproach tostudyingknowledgetranslationpractices.Thenotionof actor-scenarios has been used before and is embedded in a wider literatureonscenario-building,ﬁctivescripting,andthe construc-tionofactor-worlds[24,27,37,46].Wefurtherdevelopedthenotion ofactor-scenariosintoanapproachtoenvisionknowledge transla-tionpracticesmoregenerally.Whatdistinguishesactor-scenario mapping from other analytical approaches, including thematic analysisanda constantcomparativemethod,is itsemphasison abduction,sensitivitytothesociomaterialenvironments,andfocus onmobilisationofcontextualelements[27,38,47,48].Additionally, theapproachrecognisesthesituatednatureofboththeactorthat putsforththescenarios,andthescenariosthemselves.Anelement ofactor-scenariomappingthatneedstobedevelopedfurtheris itscapacitytoguidetranslationofknowledgeintoaction.Inother sectors,scenariosofthefutureareoftenusedtoplanoranticipate thesepossiblefutures.Whilewerecognisetheinherent contingen-ciesinknowledgetranslation,itmaybepossibletoproductively useactor-scenariomappinginexistingstakeholdermapping exer-cises.Thescenarioscouldbeusedasinputtoresearchprojects.In

theexampleoftheEuropeanresearchproject,thescenariosmay havebeenusedtoaligntheknowledgeproductionprocess(i.e. con-tinueddevelopmentofthetool)withthestakeholders’envisioned utilisationsofthatknowledge.

5. Conclusion

Ouranalysissuggeststhatengagingstakeholdersinknowledge productionshapesthetranslationofthatknowledgeintoaction in differentways. Stakeholders constantly putforth implicit or explicitscenariosaboutwhichactorsmightusetheknowledge,in whichway,andunderwhatcircumstances.Theseactor-scenarios areﬁctiveatﬁrstbuthaveaperformativeworkingaswell:through theirengagement,thestakeholders’scenarioscontributetohow theknowledgeisconstructedandthusalsowhatitsuseentails. Actor-scenario mapping mayhelp in activelyaligning research processes withthe translationsthat stakeholders envision. The assumptionsandexpectationsofroles,responsibilities,and poten-tialuse,explicatedbymappingtheactor-scenarios,couldbefed backintheresearchprojectandmighthelpinincreasingthe like-lihoodthatresultswillbeused.Additionally,ourdatacontribute toadeeperunderstandingofthe‘contextofuse’byshowinghow actorsmobiliseelementsofcontextintheirscenarios,andhowsuch elementscouldenableandconstraintheuseofknowledge.

DeclarationofCompetingInterest

None.

Acknowledgement

TheauthorswouldliketothanktheEQUIPTconsortium(http:// equipt.eu)foropeninguptoourinquiries.Inaddition,weexpress oursinceregratitudetoSteveHanneyandRolandBal–their ana-lyticalcommentstoseveraldraftsofthispaperguidedustothe currentversion.Finally,wewouldliketothanktheUnitedKingdom MedicalResearchCouncilforfundingthisproject(MR/L011123/1).

References

[1]TetroeJM,GrahamID,FoyR,RobinsonN,EcclesMP,WensingM,etal.Health researchfundingagencies’supportandpromotionofknowledgetranslation: aninternationalstudy.TheMilbankQuarterly2008;86:125–55,http://dx.doi. org/10.1111/j.1468-0009.2007.00515.x.

[2]vanBekkumJE,FergieGM,HiltonS.Healthandmedicalresearchfunding agen-cies’promotionofpublicengagementwithinresearch:aqualitativeinterview studyexploringtheUnitedKingdomcontext.HealthResearchPolicyand Sys-tems2016;14:23,http://dx.doi.org/10.1186/s12961-016-0093-4.

[3]KothariA,WathenCN.Acriticalsecondlookatintegratedknowledge transla-tion.HealthPolicy(NewYork)2013;109:187–91,http://dx.doi.org/10.1016/J. HEALTHPOL.2012.11.004.

[4]Ward V, Smith S, House A, Hamer S. Exploring knowledge exchange: a useful framework for practice and policy. Social Science & Medicine 2012;74:297–304,http://dx.doi.org/10.1016/j.socscimed.2011.09.021. [5]CavazzaM,JommiC.StakeholdersinvolvementbyHTAOrganisations:Whyis

sodifferent?HealthPolicy(NewYork)2012;105:236–45,http://dx.doi.org/10. 1016/J.HEALTHPOL.2012.01.012.

[6]BoazA,FitzpatrickS,ShawB.Assessingtheimpactofresearchonpolicy:a literaturereview.ScienceandPublicPolicy2009;36:255–70,http://dx.doi.org/ 10.3152/030234209X436545.

[7]O’BrienL,MarzanoM,WhiteRM.“Participatoryinterdisciplinarity”:towards theintegrationofdisciplinarydiversitywithstakeholderengagementfornew modelsofknowledgeproduction.ScienceandPublicPolicy2013;40:51–61, http://dx.doi.org/10.1093/scipol/scs120.

[8]BoazA,GraysonL,LevittR,SolesburyW.Doesevidence-basedpolicywork? LearningfromtheUKexperience.Evidence&Policy2008;4:233–53,http://dx. doi.org/10.1332/174426408784614680.

[9]BoazA,HanneyS,BorstR,O’SheaA,KokM,O’SheaA,etal.Howtoengage stakeholdersinresearch:designprinciplestosupportimprovement.Health ResearchPolicyandSystems2018;16:60, http://dx.doi.org/10.1186/s12961-018-0337-6.

(7)

[10]HartleyTW, RobertsonRA.Stakeholderengagement,cooperativeﬁsheries researchanddemocraticscience:thecaseofthenortheastconsortium.Human EcologyReview2006;13:161–71.

[11]Duncan S,Oliver S.Editorial: motivations for engagement. Research All 2017;1:229–33,http://dx.doi.org/10.18546/rfa.01.2.01.

[12]Baldwin S.Interactive socialscience inpractice: newapproaches to the productionofknowledgeandtheirimplications.ScienceandPublicPolicy 2000;27:183–91,http://dx.doi.org/10.3152/147154300781782048. [13]IrP,BigdeliM,MeessenB,VanDammeW.Translatingknowledgeintopolicy

andactiontopromotehealthequity:theHealthEquityFundpolicyprocessin Cambodia2000–2008.HealthPolicy(NewYork)2010;96:200–9,http://dx.doi. org/10.1016/J.HEALTHPOL.2010.02.003.

[14]JacobsonN,ButterillD,GoeringP.Developmentofaframeworkforknowledge translation:understandingusercontext.JournalofHealthServicesResearch& Policy2003;8:94–9,http://dx.doi.org/10.1258/135581903321466067. [15]DenisJL,LomasJ.Convergentevolution:theacademicandpolicyrootsof

col-laborativeresearch.JournalofHealthServicesResearch&Policy2003;8:1–6, http://dx.doi.org/10.1258/135581903322405108.

[16]OliverK,KothariA,MaysN.Thedarksideofcoproduction:dothecosts out-weighthebeneﬁtsforhealthresearch?HealthResearchPolicyandSystems 2019;17:33,http://dx.doi.org/10.1186/s12961-019-0432-3.

[17]LatourB.Scienceinaction:howtofollowscientistsandengineersthrough society,8thed.Cambridge,MA:HarvardUniversityPress;1987.

[18]SismondoS.Anintroductiontoscienceandtechnologystudies,2nded. Chich-ester,UK:Wiley-Blackwell;2010.

[19]OudshoornN,PinchT.Howusersmatter.Theco-constructionofusersand technologies.Cambridge,Massachusettsand.London,England:TheMITPress; 2003.

[20]FreemanR.Whatis“translation”?Evidence&Policy2009;5:429–47,http://dx. doi.org/10.1332/174426409X478770.

[21]BoazA,DaviesH,FraserA,NutleyS.Whatworksnow?Evidence-informed policyandpractice.Bristol,UnitedKingdom:PolicyPress;2019.

[22]CallonM.Someelementsofasociologyoftranslation:domesticationofthe scallopsandtheﬁshermenofStBrieucBay.In:LawJ,editor.Power,actionand belief:anewsociologyofknowledge?London:Routledge&KeganPaul;1986. p.196–223.

[23]LatourB.Reassemblingthesocial:anintroductiontoactor-network-theory. Oxford:OxfordUniversityPress;2005.

[24]CallonM.thesociologyofanactor-network:thecaseoftheelectricvehicle.In: CallonM,LawJ,RipA,editors.Mappingthedynamicsofscienceand technol-ogy:sociologyofscienceintherealworld.London:TheMacMillanPressLTD; 1986.p.19–35.

[25]BalR.Evidence-basedpolicyasreﬂexivepractice.Whatcanwelearnfrom evidence-based medicine? Journal of HealthServices Research & Policy 2017;22:113–9,http://dx.doi.org/10.1177/1355819616670680.

[26]WeissCH.Themanymeaningsofresearchutilization.PublicAdministration Review1979;39:426,http://dx.doi.org/10.2307/3109916.

[27]KokMO,GyapongJO,WolffersI,Ofori-AdjeiD,RuitenbergJ.Whichhealth researchgetsusedandwhy?Anempiricalanalysisof30cases.HealthResearch PolicyandSystems2016;14:36, http://dx.doi.org/10.1186/s12961-016-0107-2.

[28]PokhrelS,EversS,LeidlR,Trapero-BertranM,KaloZ,etal.EQUIPT:protocolof acomparativeeffectivenessresearchstudyevaluatingcross-context transfer-abilityofeconomicevidenceontobaccocontrol.BMJOpen2014;4:e006945, http://dx.doi.org/10.1136/bmjopen-2014-006945.

[29]YinRK.Casestudyresearch:designandmethods,5thed.ThousandOaks:SAGE Publications,Inc.;2015.

[30]OECD/EuropeanObservatoryonHealthSystemsandPolicies.Countryhealth proﬁle2017.Hungary:ParisandBrussels:OECDPublishing/European

Obser-vatory on Health Systems and Policies; 2017, http://dx.doi.org/10.1787/ 9789264283411-en.

[31]OECD/EuropeanObservatoryonHealthSystemsandPolicies.Countryhealth proﬁle2017. Netherlands:Paris andBrussels: OECDPublishing/Eruopean ObservatoryonHealthSystemsandPolicies;2017,http://dx.doi.org/10.1787/ 9789264283503-en.

[32]PaulikE,Maróti-NagyA,NagymajtényiL,RogersT,EasterlingD.Supportfor populationleveltobaccocontrolpoliciesinHungary.CentralEuropeanJournal ofPublicHealth2012;20:75–80.

[33]BosdrieszJR,NagelhoutGE,StronksK,WillemsenMC,KunstAE.Theassociation betweentobaccocontrolpolicyandeducationalinequalitiesinsmoking cessa-tionintheNetherlandsfrom1988through2011.Nicotine&TobaccoResearch 2015:1–8,http://dx.doi.org/10.1093/ntr/ntv004,0.

[34]KokMO,SchuitAJ.Contributionmapping:amethodformappingthe contri-butionofresearchtoenhanceitsimpact.HealthResearchPolicyandSystems 2012;10:21,http://dx.doi.org/10.1186/1478-4505-10-21.

[35]AkrichM.TheDe-scriptionoftechnicalobjects.In:BijkerWE,LawJ,editors. Shapingtechnology/buildingsociety:studiesinsociotechnicalchange. Mas-sachusettsandLondon,England:Cambridge;1992.p.205–24.

[36]CallonM.Writingand(re)writingdevicesastoolsformanagingcomplexity. In:LawJ,MolA,editors.Complexities:socialstudiesofknowledgepractices. DurhamandLondon:DukeUniversityPress;2002.p.191–217.

[37]deLaatB.Scriptsforthefuture.Technologyforesight,strategicevaluationand sociotechnicalnetworks:theconfrontationofscriptbasedscenarios. Univer-sityofAmsterdam;1996.

[38]TavoryI,TimmermansS.Abductiveanalysis:theorizingqualitativeresearch. ChicagoandLondon:TheUniversityofChicagoPress;2014.

[39]CallonM,LascoumesP,BartheY.Actinginanuncertainworld.Cambridge, MassachusettsandLondon,England:TheMITPress;2009.

[40]BorstRAJ,KokMO,BalR,Rotterdam,theNetherlands,Supportingtranslation ofknowledgeintobetterhealth:acriticalinterpretivesynthesisofknowledge translationandScience&TechnologyStudiesliterature;2019.

[41]RipA.Thecontextofinnovationjourneys.CreativityandInnovation Manage-ment2012;21:158–70,http://dx.doi.org/10.1111/j.1467-8691.2012.00640.x. [42]Asdal K, Moser I. Experiments in context and contexting. Science,

Technology,&HumanValues 2012;37:291–306,http://dx.doi.org/10.1177/ 0162243912449749.

[43]ShoveE,RipA.Usersandunicorns:adiscussionofmythicalbeastsininteractive science.ScienceandPublicPolicy2000;27:175–82,http://dx.doi.org/10.3152/ 147154300781781959.

[44]LawJ.OnsociologyandSTS.SociologicalReview2008;56:623–49,http://dx. doi.org/10.1111/j.1467-954X.2008.00808.x.

[45]KulchaitanaroajP,KalóZ,WestR,CheungKL,EversS,VokóZ,etal. Understand-ingperceivedavailabilityandimportanceoftobaccocontrolinterventions toinform Europeanadoption ofaUKeconomicmodel: across-sectional study.BMCHealthServicesResearch2018;18:1–12,http://dx.doi.org/10.1186/ s12913-018-2923-2.

[46]den Boer D, Rip A, Speller S. Scripting possible futures of nanotech-nologies:amethodologythatenhancesreﬂexivity.TechnologyinSociety 2009;31:295–304,http://dx.doi.org/10.1016/j.techsoc.2009.06.010. [47]StoopendaalA,BalR.Conferences,tableclothsandcupboards:howto

under-standthesituatednessofqualityimprovements inlong-termcare.Social Science&Medicine2013;78:78–85,http://dx.doi.org/10.1016/J.SOCSCIMED. 2012.11.037.

[48]OrlikowskiWJ.Sociomaterialpractices:exploringtechnologyatwork.Organ Studied2007;28:1435–48,http://dx.doi.org/10.1177/0170840607081138.