BMJ2013;347:f5264doi:10.1136/bmj.f5264(Published10September2013) Page1of12 Research B M J RESEARCH : firs t p u b lis h e d a s 1 0 Effectiveness of interdisciplinary primary care .1 1 3 6 approach to reduce disability in community dwelling /b m frail older people: cluster randomised controlled trial j.f5 2 6 4 OPENACCESS on 1 0 S Silke F Metzelthin scientific researcher1, Erik van Rossum lecturer in innovations in care for frail ep te elderly2,LucPdeWitteprofessoroftechnologyincare3,AntoniusWAmbergenstatistician4,Sjoerd m b O Hobma general practitioner5, Walther Sipers geriatrician6, Gertrudis I J M Kempen professor of er 2 social gerontology1 01 3 . D 1DepartmentofHealthServicesResearch-FocusingonChronicCareandAgeing,CAPHRISchoolforPublicHealthandPrimaryCare,Maastricht ow University,POBox616,6200MDMaastricht,Netherlands;2CentreofResearchonAutonomyandParticipation,ZuydUniversityofAppliedSciences, nlo POBox550,6400ANHeerlen,Netherlands;3CentreofResearchonTechnologyinCare,ZuydUniversityofAppliedSciences;4Departmentof a d MethodologyandStatistics,CAPHRISchoolforPublicHealthandPrimaryCare,MaastrichtUniversity;5DepartmentofGeneralPractice,CAPHRI ed SchoolforPublicHealthandPrimaryCare,MaastrichtUniversity;6OrbisMedicalCentre,POBox5500,6130MBSittard,Netherlands fro m h ttp Abstract ConclusionsThisstudyfoundnoevidencefortheeffectivenessofthe ://w ObjectiveToevaluatewhetheraninterdisciplinaryprimarycareapproach PoCapproach.Thestudycontributestotheemergingbodyofevidence w forcommunitydwellingfrailolderpeopleismoreeffectivethanusual thatcommunitybasedcareinfrailolderpeopleisachallengingtask. w.b careinreducingdisabilityandpreventing(further)functionaldecline. Moreresearchinthisfieldisneeded. m DesignClusterrandomisedcontrolledtrial. TrialregistrationCurrentControlledTrialsISRCTN31954692. j.co m Setting12generalpracticesinthesouthoftheNetherlands Introduction o/ n PInadritciactiopra)nwtesr3e4in6cflruadileodl;d2e7r0pe(7o8p%le)(sccoomrepl≥e5teodnthGerostnuindgy.enFrailty Icnhaolulernaggeesinignshoecailetthyc,acraer.e12foErvoidldeenrcpeesoupglgeeisstosntehaotfctohmegmreuantietsyt 15 J a InterventionsGeneralpracticeswererandomisedtotheintervention basedcareincomparisonwithinstitutionalisationmayachieve nu orcontrolgroup.Practicesinthecontrolgroupdeliveredcareasusual. betteroutcomesatlowercostsandispreferredbyolderpeople ary Practicesintheinterventiongroupimplementedthe“PreventionofCare” themselves.34Consequently,anincreasingdemandexistsfor 2 (PoC)approach,inwhichfrailolderpeoplereceivedamultidimensional innovativeinitiativestoprovidecosteffectivecommunitybased 02 3 assessmentandinterdisciplinarycarebasedonatailormadetreatment care.56InmostWesterncountries,suchastheUnitedKingdom b planandregularevaluationandfollow-up. andtheNetherlands,generalpractitionershaveacentralposition y g u MainoutcomemeasuresTheprimaryoutcomewasdisability,assessed intheprovisionofcommunitycare,astheyaregatekeepersto e s at24monthsbymeansoftheGroningenActivityRestrictionScale. specialisedandhospitalcare.7IntheUK,generalpractitioners t. P Secondaryoutcomesweredepressivesymptomatology,socialsupport havebeenrequiredsince1990toofferanannual ro interactions,fearoffalling,andsocialparticipation.Outcomeswere multidimensionalassessmenttotheirpatientsaged75yearsand te measuredatbaselineandat6,12,and24months’follow-up. over.8Inaddition,generalpractitioners’geographicalproximity cte d Results193olderpeopleintheinterventiongroup(sixpractices) toolderpeopleandtheirintenseandlonglastingrelationship b receivedthePoCapproach;153olderpeopleinthecontrolgroup(six withtheirpatientsmaycontributetoeffectivecareinolder y c people.7However,communitybasedcareoffrailolderpeople o practices)receivedcareasusual.Follow-upratesforpatientswere91% p (2n4=m31o6n)thast.sMixixmeodnmthosd,e8l6m%u(lntil=e2v9e8l)aanta1ly2semsosnhthosw,eadndno78si%gn(infi=ca2n7t0)at ihdseecafihlntahelcdleaanresgindneigfef.dicFsu,ralwtiylhoioclrdhdeoerfppteeennodpleelnaedchytaovineditmshaeubleitlxiipteylce.u9-at1in1oDdnicosoafmbdipaliiltleyyxis yright. differencesbetweenthetwogroupswithregardtodisability(primary activitiesthatareessentialforindependentliving.9Asdisability outcome)andsecondaryoutcomes.Pre-plannedsubgroupanalyses isconsideredadynamicprocess,olderpeoplecanrecovertoa confirmedtheseresults. Correspondence to: S F Metzelthin [email protected] No commercial reuse: See rights and reprints http://www.bmj.com/permissions Subscribe:http://www.bmj.com/subscribe BMJ2013;347:f5264doi:10.1136/bmj.f5264(Published10September2013) Page2of12 RESEARCH lessdisabledornon-disabledstate.12Regardless,preventive andsystematicpolicyforthedetectionandfollow-upoffrail actionshavetobetakentoimprovetheabilitiesoffrailolder olderpeopletotakepartinthestudy.Intotal,24practiceswere B peopletoremainathomeaslongaspossible.13 interested,ofwhichwerandomlyselected12forthestudyon M J Dedxiastnaebinëisllisivtyaeniodnvcceoorvlmlieemawguuonefitsyedxdiidswtiaenlnglianirngrtaeftrriavvieelnortelidvoeinerswpfeotoorppplrreeo.v1v4eiTdnhetieoannof tfyhoeecaurbssa)es.siTsohonoftshaeecwiorhmcoopmwutmeerruegnteeitnrymedriawntaeeldlllyilniisgltl,ofrwfaenilrueomlcdboeenrrfspi.naTetidheentotsstbu(e≥ddy7,0 : first pu identifiedinterventions,mostofwhichwereinthefieldof hadseverecognitiveorpsychologicalimpairments,orwere blis comprehensivegeriatricassessmentandphysicalexercise unabletocommunicateinDutchwereexcludedonthebasisof h e programmes,showedalargediversityintermsofcontent, theadviceofthegeneralpractitioner.Theremainingotherolder d a disciplinesinvolved,duration,intensity,andsetting.Onlya people(n=3498)inthe12practicesreceivedapostal s smallnumberhaveshownbeneficialeffectswithregardto questionnaire,includingtheGroningenFrailtyIndicator.18In 10 disability,andmoststudiesdidnotreportonanylongterm theliterature,ascoreof4orhigher(range0-15)isproposedas .11 effects.14Onthebasisofthisreview,theauthorssuggestedthat thecut-offpointformoderatelytoseverelyfrailolderpeople.18 36 communitycareinterventionsforfrailolderpeopleshouldbe However,thisstudyfocusedonpeoplewhowereconsiderably /bm cinodnidvuidcuteadlisbeydaanssienstsemrdeisnctsipalnindairnyteprrvimenatiroyncsa(rteaitleoarmmiandveoclavrien)g, fdreavile,lorapnignigndgisfraobmilitpyetoopdleiswabhloedhaovldeearnpeinocprleea.sTehderriesfkorfeo,rolder j.f52 6 selfmanagementsupport,engagementinmeaningfulactivities, peoplewhosignedtheinformedconsentformandhada 4 casemanagement,andlongtermfollow-up.Inaneffortto GroningenFrailtyIndicatorscoreof5orhigherwereincluded on reducedisabilityandprevent(further)functionaldeclinein inthestudy.Forpracticalreasons,therecruitmentoffrailolder 10 communitydwellingfrailolderpeople,wehavecombinedthese peopletookplaceinthreecycles.Thefirstcyclestartedin S e ealpepmroeancths.inTthoisonaepparpoparcohacfohc:uthsees“oPnrebvoetnhtioolndeorfpCeaorpel”e(wPoitCh)an DMeacrecmhb2e0r1200.0T9h,ethinetesrevceonntdioinnaFnedbrtuhaercyo2ll0e1c0ti,oannodftdhaettahairldsoin ptem increasedriskfordevelopingdisabilityandolderpeoplewho tookplaceinthreecycles.Allincludedolderpeoplegavewritten be arealreadydisabled.15Apreviouspilotstudy(n=41)usingthe informedconsentbeforecollectionofthebaselinemeasure. r 2 0 PoCapproachhasshownpromisingresults.15Olderpeople 1 appreciatedtheattentiontheygotandfeltsupportedinreaching Intervention 3. D theirgoalsandinhandlingfuturedisability.Healthcare Intheinterventiongroup(sixpractices),frailolderpeople ow professionalsreportedthattheapproachprovidedauseful receivedthePoCapproach.Thegeneralpractitionerandpractice nlo structureforgeriatricprimarycare.Inaddition,theapproach nursebuiltthecoreteamoftheinterdisciplinarycareapproach. a d stimulatedinterdisciplinarycollaboration,afocusonmeaningful In2001theprofessionofpracticenursewasintroducedinthe ed activities,andselfmanagementsupport.15However,the NetherlandstoreducetheworkloadofDutchgeneral fro effectivenesswithregardtodisabilityandvariousrelated practitioners,whoarethegatekeepertospecialisedandhospital m otohunitsvctoarmriiaoelustsohpainsavtnieeosntttyigleeatvtbeeeltehoneustetcufofdemiceetdisv.CceonomensspseaoqrfueedthnwetlyiPt,howCuesaucpaoplnrcdoauarcceth.e1d6 cgmaearnene.ar1g9alePmprareacnctittc,itemionenunertrsa,eloshnoefadtlietshneaswesoervrpkicr,eeuvsn,edanestisroetnshs,emcsheurnpotensriovcficsfiaroraneiloofldtheer http://w w Wechoseaclusterrandomiseddesignforpracticalreasonsand people,andcareoffamilieswithyoungchildren.20Withinthe w toavoidcontaminationbias.17 PoCapproach,thegeneralpractitionerandpracticenurse .bm cooperatecloselywithoccupationalandphysicaltherapists.If j.c Methods needed,otherinpatientandoutpatienthealthcareprofessionals, om Study design suchasapharmacistorageriatrician,areinvolvedaswell. o/ n ThePoCapproachaimstoreducedisabilityandprevent(further) 1 Wedidatwoarmclusterrandomisedcontrolledtrialamong12 functionaldeclinebyusingasixstepapproach(fig1⇓).21After 5 J generalpracticesinthesouthoftheNetherlands.Beforethe thepostalscreeningforfrailtyusingtheGroningenFrailty an screeningprocedureforidentifyingfrailolderpeoplestarted, Indicator(step1),frailolderpeopleandtheirinformalcaregiver, ua wsixerparnadctoimcelsytoallcoocnattienduesicxapreraacstiucseusatlo.BtheefoProeCraanpdpormoaicshatiaonnd, iafmavualitliadbimlee,nrescioenivaelaashsoemssemveinstitfboycuthsienpgraocntiecxeisntuinrsgepwrohboldemoess ry 20 2 thepracticeswerepre-stratifiedintofourstratabasedonnumber inperformingdailyactivitiesandonriskfactorsfordisability 3 ofolderpatients(<350versus≥350patients)andlocation(urban (step2).Thefocusisonactivitiesthataremeaningfultothe by versusruralarea).Weassumedthatgeneralpractitioners olderperson.Examplesofmeaningfulactivitiesaregardening, gu workinginapracticewithalargenumberofolderpatientshave visitingfamily/friends,readingabook,takingawalk,playing es moreexperiencewithgeriatriccareandthatolderpeopleliving games,andjoiningreligiousactivities.Afterthehomevisit,the t. P inaruralareareceivemoresupportfromtheinformalcare generalpractitionerandpracticenursediscusswhether ro systemthandothoselivinginanurbanarea.Westratifiedthe additionalassessmentsbyotherinpatientoroutpatienthealthcare tec practicesinpairsandusedacomputergeneratedrandomisation professionalsareneeded.Onthebasisoftheassessmentphase, ted listtorandomisethemintoeithertheinterventionorcontrol apreliminarytreatmentplanisformulated(step3),eitherina by group.Topromoteextrapolationoftheresults,practicesinan bilateralmeeting(generalpractitionerandpracticenurse)orin c o urbanareawithalargenumberofolderpeoplehadtwicethe anextendedteammeetingconsistingofageneralpractitioner, py chanceofbeingallocatedtotheinterventiongroupthandid practicenurse,occupationalandphysicaltherapist,and,if rig practicesintheotherthreestrata.Theclusterrandomised necessary,otherhealthcareprofessionals. ht. controlledtrialwasperformedasplanned.Moredetailsofthe Duringasecondhomevisitbythepracticenurse(step4),afinal studydesignhavebeenpublishedelsewhere.16 treatmentplanisformulated,includingalistofgoals,strategies, andactionsthatmeettheolderperson’sneeds.Dependingon Participants theselfmanagementskillsandpreferencesoftheolderperson, WeinvitedallgeneralpracticesintheregionofSittard(the strategiesandactionsareeitherfocusedontheolderpersonor Netherlands)anditssurroundingareathathadnocurrentactive moreon(supportof)thesocialandphysicalenvironment.On No commercial reuse: See rights and reprints http://www.bmj.com/permissions Subscribe:http://www.bmj.com/subscribe BMJ2013;347:f5264doi:10.1136/bmj.f5264(Published10September2013) Page3of12 RESEARCH thebasisofthe5AsBehaviouralChangeModel,22and alimitof25%missingvalues.Multilevelanalysesarequite motivationalinterviewingtechniques,23thepracticenurse robustagainstmissingvaluesatthemeasurementlevel. B encouragesactiveinvolvementindecisionmakingand Therefore,weneededatleastthebaselinemeasurementand M J eosldtaebrlpisehressonaacnodopthereaitnivfeorwmoarlkcinargegreivlaetri.oSnsuhbispeqwuiethnttlhye,tfhraeil oinnethoeuatnoafltyhsreese.fFoollroawll-uapnamlyesaessu,rewmeeunstesdtoainstcalnuddaerodldmeordpeelople : firs treatmentstarts(step5).Theinterventionprotocoloffers includingsixindependentvariables.Wecorrectedoutcome t p u recommendationsandguidelinesfortheexecutionofthe estimatesofthemultilevelanalysesforage,sex,educational b lis treatmentplan.Forexample,atoolboxofinterventionsis level,significantdifferencesatbaseline(frailtyanddisability), h e availablethatfocusesonfivetopics:“enhancingmeaningful andthebaselinestatusoftheoutcomevariable(inthecaseof d a activities,”“dailyphysicalactivity,”“socialnetworkandsocial secondaryoutcomes)byincludingthesevariablesascovariates s activities,”“adaptingtheenvironment,activities,orskills,”and ineachmodel.Weobtainedinsightintotheeffectivenessofthe 10 “stimulatinghealth.”Thepracticenurseisalsothecasemanager PoCapproachincomparisonwithusualcareatvarious .1 1 and,alongwiththefrailolderpersonandtheinformalcaregiver, follow-uptimesbyexaminingfixedeffectsforgroupbytime 3 6 regularlyevaluatestheachievementofgoals,theimplementation interaction.Weevaluatedthetrendintimebyremovingthe /b m ofoflslotrwatienggiepseriniodda(islytelpif6e),.aTnhdethperonfeeesdsiofonralssuipnpvoorltviendthaere ionntleyrafcixtieodnetfefremcts(gfororugprobuyptaimndet)imfroem.Inthaefemwodimelpaunteddtedsattiansgets, j.f52 6 updatedabouttheprogressandtheagreementsmade.Thebox varianceofpracticeiteratedtozero.Consequently,weexamined 4 illustrateshowtheapproachworks. inabasicmodelofdisability,includingonlybaselinestatusof on Theremainingsixpractices(controlgroup)continuedtodeliver disabilityasacovariate,whetherpracticehadaneffecton 10 careasusual. outcomes.Theanalysesofthebasicmodelwithandwithout S e practiceasarandomeffectshowedthattheresultswerehighly p Measurements similarforthetwoanalyses.Therefore,wedecidedtoexclude tem practiceasanextralevel. b e Wemeasureddatafortheeffectivenessanalysisatthelevelof Wedidseveralsubgroupanalyses.Firstly,wedividedolder r 2 thepatientatbaselineandafter6,12,and24monthsbyusing 0 peopleintheinterventiongroupintotwosubgroupsonthebasis 1 postalquestionnairesandtelephoneinterviews.Whereasolder 3 peopleandhealthcareprofessionalswereawareoftheallocated oftheirexposuretothePoCapproach.Wecomparedolder . D peoplewhoreceivedonlyassessment(s)(exposuregrouplow) o arm(interventionorcontrol),outcomeassessorswerekept w withthosewhoreceivedinterventions,follow-upvisits,orboth n blindedtotheallocation. lo (exposuregrouphigh).Wetestedfixedeffectsforexposure a d groupbytimeinteractionsforsignificance.Inaddition,wedid e Outcome measures d pre-plannedsubgroupanalysesforthepotentialeffectmodifiers fro Wemeasuredtheprimaryoutcome,disability,at24monthsby baselinestatusoffrailtyandmastery.16Wecreatedtwogroups m meansoftheGroningenActivityRestrictionScale.24Thisisan foreacheffectmodifierbasedonthemedianscores:lowfrailty h easytoadminister,comprehensive,reliable,hierarchical,and (GroningenFrailtyIndicatorscore5-6)versushighfrailty(score ttp validmeasureforassessingdisabilityinolderpeople.Itconsists 7-14)andlowmastery(PearlinMasteryScalescore23-32) ://w w oftwosubscales.Thefirstsubscaleisaboutactivitiesofdaily versushighmastery(score10-22).Again,wetestedfixedeffects w living(11items),andthesecondsubscalerelatestoinstrumental foreffectmodifierbygroupinteractionsforsignificance. .b m activitiesofdailyliving(sevenitems).Thescoresforthetotal Thesamplesizecalculationwasbasedonourprimaryoutcome j.c scalerangefrom18to72,withhigherscoresindicatingmore o (disability).Onthebasisofapowerof80%andanαof0.05 m (addnisedapbsroeilscistiyaio.l2nf4uAsnuscbtdsioicsnaalibenigoli,ft2y5thwiesesHctrhooosnpsgeitladylerAperlneaxstesieidvtyteoaspynsmdyDpchteoopmlroeagstiosciloaolngy (ltehtwaesortet2qa.ui0lierpdeodtiensstastminopgnl)et,hasenizdGerawonnaeisnx8gp0eencpteAerdcgttirrvoeiauttypmR(e1en6st0tdriiicnfftietoornetanSlc)c.eaolef,2a4t on 15/ Scale),26socialsupportinteractions(SocialSupport J Accountingforadropoutrateof30%andaclustereffectof a List—Interactionversion),27fearoffalling(ShortFallsEfficacy 1.73(intraclasscorrelationcoefficient0.05),16assumingequal nu Scale—International),28andsocialparticipation(Maastricht clustersizes,thefinalsamplesizehadtobe180pergroup(360 ary SocialParticipationProfile,subscaleA)29assecondary intotal).16WeusedthesoftwarepackageSPSSforWindows, 20 outcomes.Inaddition,weusedthePearlinMasteryScaleto 2 version20.0,forallstatisticalanalyses. 3 determinethefeelingsofcompetenceandcontrolinolder b y people,30feelingscrucialforselfmanagementandcoping,31 Results gu whichbelongtotheimportantunderlyingmechanismsofthe e s PoCapproach. Weallocated12generalpracticesatrandomtothecontrol(six t. P practices)orinterventiongroup(sixpractices).Halfofthe ro Statistical analysis practiceshadlessthan350patientsandhalfhadatleast350 te c Weuseddescriptivetechniquestodescribethestudygroups. patients.Inaddition,sixpracticeswerelocatedinanurbanarea ted Wecomparedbaselinevariablestodetectdifferencesbetween andsixinaruralarea.Theseclustercharacteristicswereequally b y theinterventionandcontrolgroupsatthestartofthestudy. distributedamongthegroups.Asshowninfigure2⇓,3498 c o Becauseoftheclusterrandomiseddesignofthestudyincluding communitydwellingolderpatients(≥70years)ofthe12 p y threelevels(generalpractices,participants,andrepeated practicesreceivedthescreeningquestionnaire.Theresponse rig measurements),weappliedamixedmodelmultilevelanalysis. ratewas80%(n=2790).Non-respondersweresignificantly ht. Weanalysedtheprimaryandsecondaryoutcomes,measured youngerthanresponders(meanage76.75v77.62years; atthelevelofthepatient,accordingtotheintentiontotreat P<0.05),andslightlymorenon-respondersweremen(42.9%v principle.Weimputedmissingvaluesatthelevelofthescale 39.1%;P=0.07).Olderpeoplewhocompletedthequestionnaire bymeansofmultipleimputations.Webasedthemaximum andwerewillingtoparticipateinthestudy(n=1101)were numberofmissingvalueswithinascaleontheguidelinesgiven significantlyfrailerthanrespondentswhocompletedthe bythedevelopers.Ifnoguidelineswereavailable,weaccepted questionnairebutdeclinedparticipation(n=1634)(meanscore No commercial reuse: See rights and reprints http://www.bmj.com/permissions Subscribe:http://www.bmj.com/subscribe BMJ2013;347:f5264doi:10.1136/bmj.f5264(Published10September2013) Page4of12 RESEARCH Casesummary B M AKisa75yearoldwomanlivingindependentlyinasmallflat.Shehasfourchildren,wholiveinthesamecity.Herhusbandhasbeenliving J iAnKarneucersivinegdhaolmetetefrofrrotwmohyeeragresn.eralpractitioner,whoaskedhertofillintheGroningenFrailtyIndicator(step1).Shehadatotalfrailtyscore : firs of7,andthepracticenursecalledhertoofferahomevisitforamultidimensionalassessment(step2).Theassessmentfocusedonexisting t p problemsinperformingdailyactivitiesandriskfactorsfordevelopingdisability.ThepracticenursealsodiscussedwithAKherindividual u neexpeedrsieanncdedgoparolsbalenmdshweritmhocotivoaktiniogn,tsohompapkinegc,haanndgevissiitninhgehrelirfeh.uAsKba’snmdoinstthimepnourrtsainntgghooamlwe.aTshtoeslatasytoinfdtheepseendweanstlpyainrtihcuelrahrloymmee.aSnhinegful blis h toher.Sheoftenfeltexhaustedandhadafearoffalling.Inaddition,memorydeficitsaffectedherparticipationindailylife.Herchronic e diseases(diabetesandheartfailure)wereundercontrol. d a Afterthehomevisit,thepracticenursediscussedtheresultsoftheassessmentwiththegeneralpractitioneranddecidedtoreferAKtoa s geriatricianforfurthercognitiveassessment.Inaddition,theyagreedthatinvolvinganoccupationaltherapistandphysiotherapistwouldalso 1 0 beuseful,asAKhadproblemswithdailyactivities.Aninterdisciplinaryteammeeting,consistingofthegeneralpractitioner,practicenurse, .1 occupationaltherapist,andphysiotherapisttookplacetoformulateapreliminarytreatmentplanbasedontheresultsoftheassessment 1 (step3).Theassessmentofthegeriatricianshowednosignsofdementia.Regardingherfearoffalling,theteamassumedthatAKneeded 3 6 tochangeherattitudesandselfefficacybeliefswithregardtofalling,leadingtowardsimprovedparticipationindailyactivitiessuchas /b shoppingandvisitingherhusband.Anincreaseinphysicalactivitywassupposedtopositivelyaffectherfearoffallingaswell.Inaddition, m sAifmteprlethsetrtaetaemgiemseaentidnga,ftehwephrealpcitnicgeanidusrsweevriesitdeidscAuKssaegdationhtoelfpinAalKisewitthhectoreoakitnmgeanntdplhaann(dsltienpg4h)e.rWmheicmhotroyodlbeofixciptsa.rtscouldbeusedwas j.f52 alsodiscussed.ForthetreatmentofAKthetoolboxes“adaptingtheenvironment,activities,orskills”and“dailyphysicalactivity”werechosen 6 4 (step5).Duringthetreatment,thepracticenursevisitedAKfourtimestoevaluatetheachievementofgoalsandtheimplementationof o strategiesindailylife(step6).Fourmonthslater,duringthelastvisit,AKreportedthatshehadfewerproblemswithcookingandvisiting n herhusband.Shehadincreasedherphysicalactivityindailylifeandhadlessfearoffalling.However,thestrategieslearntforhandlingher 1 memorydeficitswerestilldifficulttoapplyindailylife.Afewhelpingaidsandastoolplacedinthekitchenhelpedhertocookmoreefficiently. 0 ThepracticenursewillvisitAKeverysixmonthstofollow-upwithher. S e p te onGroningenFrailtyIndicator3.64v2.96;P<0.05).Several Secondary outcomes m b participantsdeclinedparticipationandthequestionnairewas e not(completely)filledin(n=55),sowewerenotabletoobtain Table3⇓showstheresultsofthesecondaryoutcomes.Again, r 2 wefoundnosignificantgroupbytimeinteractioneffectsofthe 0 frailtyscoresandhavenoinformationabouttheleveloffrailty. 1 interventiongrouponanyoftheseoutcomes. 3 Oftheolderpeoplewhowerewillingtoparticipate,34% . D (n=179)inthecontrolgroupand38%(n=214)inthe o Subgroup analyses w interventiongroupwerefrailaccordingtotheirfrailtyscore n lo (score≥5).Ofthe393olderpeoplewhowereeligibleforthe Thefixedeffectsforexposuregroups(lowversushigh)bytime a d study(werefrailandgavewritteninformedconsent),47were interactionswerenotsignificant(P>0.05).Wefoundno e d notincludedinthestudy,astheyhadnotcompletedthebaseline significant(P>0.05)mediatingeffectsforahigherlevelof fro measurement(fig2⇓).Finally,346olderpeoplewereincluded masteryoralowerleveloffrailty(datanotshown). m inthestudy,193(56%)ofwhomreceivedthePoCapproach. h T(nh=e1m99e)awneargeefoemfpaalert,i4ci9p%an(tns=w1a7s0)77w.e2r(eSlDivi5n.g1)alyoenaer,s,an5d8%58% Discussion ttp://w Ourstudyhasprovidednoevidencefortheeffectivenessofa w (n=202)hadalowlevelofeducation. w proactiveprimarycareapproach,consistingofa .b Wefoundsignificantdifferencesbetweeninterventionand multidimensionalassessmentwithinterdisciplinarycarebased m controlgroupparticipantswithregardtofrailty(Groningen onatailormadetreatmentplanandregularevaluationand j.co FrailtyIndicator)anddisability(GroningenActivityRestriction m follow-up,amongfrailolderpeople.Wefoundnosignificant Scale)scores.Theinterventiongroupparticipantswere differencesbetweentheinterventiongroupandthecontrolgroup o/ n significantlyfrailer(score7.13v6.72;P<0.05)andmore (careasusual)withregardtodisability(primaryoutcome)or 1 disabled(score33.09v30.58;P<0.05).Allothercharacteristics oursecondaryoutcomes:depressivesymptomatology,social 5 J weresimilarbetweenthegroupsatbaseline(table1⇓).Intotal, a supportinteractions,fearoffalling,andsocialparticipation. n 76olderpeoplewerelosttofollow-upduringthetrial, Pre-plannedsubgroupanalysesconfirmedtheseresults. ua significantlymoreofthemintheinterventiongroup(26%v ry 2 17%;P<0.05). Strengths and weaknesses of study 02 3 Primary outcome Thestrengthsofthisclusterrandomisedtrialincludealong by follow-upperiodwithrelativelyfewmissingdataandhigh g All12clusters,consistingof310frailolderpeoplewitha follow-uprates.Inaddition,weusedoutcomemeasureswith ue s bmaesaesliunreemdiesnatbsi,liwtyersecoinreclaunddedatinletahsetomniexeodutmoofdtherlemeufoltlilloewve-lup goodpsychometricproperties,whichwereassessedbyblinded t. P datacollectors.Thisstudyalsohassomeweaknesses.Firstly, ro analyses.Withregardtodisability,weidentifiednosignificant significantbaselinedifferencesexistedbetweentheintervention te c differencebetweenthecontrolandinterventiongroupsat24 andcontrolgroupswithregardtofrailtyanddisability,andthe te months’follow-up.Wefoundnosignificantgroupbytime samplesizedistributionwasskewed.Thesedifferenceswerea d b interactioneffectsforthetotalGroningenActivityRestriction y resultoftheclusterrandomiseddesignofthestudy,whichisa c Scalescoresorfortheactivitiesofdailylivingandinstrumental o commonapproachwiththiskindofinterventiontoavoid p ainctteivraitciteisonoftedramilyfrloivmintghesumbosdcaelle,wsceotreesst.eAdfttheertrreemndovfionrgtitmhee. cdoifnftearmenicneastiionnobuirasa.n17aAlylstehso,utghhiswmeaaydsjtuislltehdavfoerabffaescetleindeour yrigh Bothgroupsincreasedsignificantly(P<0.05)indisabilityover t. findingstosomeextent.Secondly,significantlymore aperiodof24months,butnosignificantdifferencesbetween participantswerelosttofollow-upintheinterventiongroup thegroupswithrespecttotheirincreaseexisted.Table2⇓gives thaninthecontrolgroup(26%v17%).Wecannotfullyexplain asummaryoftheseresults. thisfinding,butolderpeopleintheinterventiongroupwere significantlymorefrailanddisabledthanthoseinthecontrol group,whichmighthaveaffectedthecompletionrate.Thirdly, No commercial reuse: See rights and reprints http://www.bmj.com/permissions Subscribe:http://www.bmj.com/subscribe BMJ2013;347:f5264doi:10.1136/bmj.f5264(Published10September2013) Page5of12 RESEARCH thePoCapproachwasevaluatedinareallifesettingin12 group,insufficientimplementationofthePoCapproach,and generalpractices.Althoughwedidacomprehensiveprocess currenthealthcaredeliveryintheNetherlands. B evaluationalongsidethetrial,32wehavelimitedinsightinto Firstly,thePoCapproachfocusesonfrailolderpeople M J woirlrhdeaestrphepaceptoipvpeelenoewfdethirneeipprrasattciuetdincytespfoaofrrttishceeivpireartgiaeolnnre.erCaasloonpnsrsae.cqPtuiateirnotitnclyeipr,amtiankging (shGcigorhroeenriinnthgoaeunnrtfhFraraatiliilnstayamIcnopdmleicp(aamtroaerbanslec3os2are.m0≥,pSl5eD).(≥T171h0.e2y)bewaasaressl)isnouefbtdshtieasanDbtiuialtilctlyhy : first pu adistinctionbetweenusualcareactivitiesandcontactsrelated generalpopulation(mean24.9,SD9.3).41Someofthe blis tothePoCapproachwasdifficult,resultinginanoverlapin participantsinourstudymayhavebeentoofrail,assome h e timespentindeliveringusualcareversusthenewapproach. previousreviewsinthefieldofpreventivehomevisiting d a Also,practicenurseshadtroubleindeterminingaclearendpoint programmessuggestthatinterventionsmaybemoreeffective s ofthePoCapproach,becauseolderpeopleremainedpatients inlowrisk,non-disabledolderpeople.2533Thisisinlinewith 10 oftheirgeneralpractitionersafterthePoCapproachhadbeen amorerecentreviewreportingthatfrailolderpeoplehaveto .11 delivered,resultingincontinuousmonitoringofolderpeople. beidentifiedatarelativelyearlystagewhennegativehealth 36 Inaddition,olderpeoplewerereferredtootherhealthcare outcomescanstillbeavoided.42Incontrast,practicenursesand /bm pmruocfehstsiimoneawlsaassswpeenllt.iAnsdaelrievseurlitn,gwtehedoPonCotakpnporwoaecxhaacntldyhhooww gmeennetriaolnperdatchtaittioanlaerrgseinntuemrvbieewroefdpdaurrtiincigptahnetspwroecreesisnetvhaeliuravtiieown j.f52 6 forlongtheseactivitieswerecontinued.However,theprocess noteligibleforthePoCapproach,astheyhadhardlyany 4 evaluationshowedthatslightlymorethanonethirdofthe disabilityintermsofactivitiesofdailylivingandinstrumental on participantsintheinterventiongroup(34%)hadonlythe activitiesofdailyliving.32Ineffortstoreducedisabilityand 10 multidimensionalassessmentconductedbythepracticenurse prevent(further)functionaldecline,whicholderpeoplewould S e dautariinlogramnaindietiatrlehaotmmeenvtisfiotl.lTohweerdembyaiunpintgooflidveerfpoelloopwle-urepceviivsietds bisesnteilflitntohtecmleoasr.tfrominterventionssuchasthePoCapproach ptem bythepracticenurse. b Comparison with other studies Sinetceorvnednlyti,otnheprportoocceoslswevearelunaotitoinmsphloewmeednttehdatassopmlaenpnaerdts.32oTfthhee er 20 1 problemanalysisandthedevelopmentofapreliminarytreatment 3 Duringthepastdecades,muchresearchtargetingcommunity plan(step3)wasoftennotdoneinabilateralmeetingoran . D o dwelling(frail)olderpeoplehasbeendone,withmanystudies extendedteammeeting,andonlyhalfofthetreatmentplans w inthefieldofpreventivehomevisitingprogrammes.Since werediscussedwiththefrailolderperson(step4).Also,the nlo 2000,severalmeta-analyses,systematicreviews,andliterature toolboxpartswerenotfrequentlyusedinthetreatmentphase ad reviewshavebeenpublished.121333-37Thestudiesevaluateda (step5),andtheextentofevaluationandfollow-up,especially ed rangeofinterventions(suchasmultidimensionalgeriatric amongthehealthcareprofessionals,waslimited(step6).32 fro assessment,careplanning,organisationandmonitoring,health Insufficientimplementationisawellknownproblem,especially m promotion,selfmanagementsupport,nursingservices,and inthefieldofpreventiveandbehaviouralchangeinterventions.43 http r(egfeenreraralslptoraoctthiteironseerrvs,icneusr)sceas,rraileldieoduptrboyfevsasiroionuaslsp).roTfheessaiiomnaolsf Dsoumriengpatrhtesopfrotcheesisnetevravleunattiioonn,pprrootfoecsosliownearles(mtoeon)titoimneedthat ://w w theseinterventionsistoproactivelydetectmodifiableriskfactors consumingordifficulttoapply32;thismayhavebeenareason w andworseninghealthconditionstoreduceorpreventdisability, forinsufficientimplementation.43Inaddition,professionals .bm healthcareuse,andrelatedcosts.Resultsregardingthe expressedaneedformoretrainingonthejobandmore j.c effectivenessoftheseinterventionshavebeeninconsistentand opportunitiestoexchangeexperienceswitheachother. om conflicting.Afewstudieshaveshownfavourableeffectson Educationandexperienceofprofessionalsandtheintensityof o/ disability.Forexample,Bernabeiandcolleaguesdida providedtrainingactivitiesarestronglyrelatedtobeneficial n 1 randomisedcontrolledtrialshowingthatamodelofintegrated outcomes.12Despiteanextensivedevelopmentperiodanda 5 J careandcasemanagementhadfavourableeffectsondisability comprehensivetrainingprogrammewithregardtothe a n incommunitydwellingolderpeople.38Serviceswereprovided interventionprotocol,weprobablyfailedinproviding u a bythegeneralpractitionerandacommunitygeriatricevaluation professionalswiththenecessarycompetenciesandfeasibletools ry unit,consistingofageriatrician,asocialworker,andseveral toapplyrathercomplexconcepts,suchasinterdisciplinary 20 2 nurses.Gillandcolleagueshavereportedasuccessful collaboration,tailormadecare,andselfmanagementsupport, 3 randomisedcontrolledtrialevaluatinganintenseexercise intodailypractice.32Forexample,thedevelopmentof by programmeforphysicallyfrailolderpeoplelivinginthe individualisedgoals,aprerequisitefortailormadecareandself gu community.39Theprogrammeisbasedontheoutcomesofan managementsupport,isachallengingtask,aspatientstendto es extensiveassessment;itfocusesontheindividualneedsofolder adoptapassiveroleingoalsetting.44Encouragementofactive t. P peoplebutalsotakestheirenvironmentalconditionsinto involvementisevenmoredifficultwitholderpeopleowingto ro account.Moststudies,however,reportednooronlymodest highlyprevalentcognitiveimpairments,communication tec effectsoftheirinterventions.Also,thelargesttrialinthisfield, difficulties,andcomorbiditiesandassuchrequiresauniqueset te d byFletcherandcolleagues,comparingdifferentstrategiesfor ofcompetencies.45Goalidentificationtools,suchasthe b y assessment(targetedversusuniversal)andevaluationand CanadianOccupationalPerformanceMeasureorGoal c o management(primarycareversusmultidisciplinarygeriatric AttainmentScaling,4647maybeusefulintheprocessofgoal p y team)inmorethan40000olderpeople,didnotresultin setting.Inaddition,moreattentionhastobepaidtothe rig convincingeffectsoradequateevidencethatonestrategyis implementationofevaluationandfollow-upactivities,asa ht. betterthananother.40 minimumintensityandlengthoffollow-upisneededtoreach favourableeffects.2534 Meaning of study: explanations and clinical Thirdly,standardhealthcaredeliveryintheNetherlandsis implications alreadyatarelativelyhighlevel.Nearlyallpeoplearecovered Besidesthemethodologicaldrawbacks,someotherexplanations byhealthcareinsurance,healthcareiseasilyaccessible,andits forthelackofeffectsarepossible.Theserelatetothetarget qualityisoftenconsideredtobegood.48Moreover,thecontrast No commercial reuse: See rights and reprints http://www.bmj.com/permissions Subscribe:http://www.bmj.com/subscribe BMJ2013;347:f5264doi:10.1136/bmj.f5264(Published10September2013) Page6of12 RESEARCH betweenthePoCapproachandcareasusualwasprobablytoo 1 BergmanH,BélandF,PerraultA.Theglobalchallengeofunderstandingandmeeting theneedsofthefrailolderpopulation.AgingClinExpRes2002;14:223-5. smalltodetectsubstantialeffects.Thenon-effectiveresultsof 2 ColemanEA.Challengesofsystemsofcareforfrailolderpersons:theUnitedStatesof B thisstudyandthecomplexityofeffectiveinterventions(or Americaexperience.AgingClinExpRes2002;14:233-8. M ecloenmcleunstiso)ndseasbcroiubtewdhinicthhesplietceirfaitcusretramteegainesthwatodurldawreinsugltina 34 WPUlnooieoteglhdaJS,ntBdalrteaerzsSila,KnC,daHCmuatpcnbhaeidsllaoT.n,NBH,EimKnagmclezJlosMrteoeiwndsD2k0iU0J.,3CD;3oa4sl9bts:y7o6Df8Mh-7e,5aG.ltohldcsamreithadCmHin,iesttraalt.ioEnffeinctthoef J: firs surpluseffectinwhichtargetgroupisdifficult. preventiveprimarycareoutreachonhealthrelatedqualityoflifeamongolderadultsat t p riskoffunctionaldecline:randomisedcontrolledtrial.BMJ2010;340:c1480. u b Future research 5 KofeiantfionrgmNal,nOettfwinoorwksskoifPfr,aWilseenngieorrsC:,aFcaassteJ,foDrecrakrseenneLtw.Uonrkdse.rAsgtaenindginSgothce20ca0r3in;2g3c:1a1p5a-c2i7ty. lish 6 Markle-ReidM,BrowneG,WeirR,GafniA,RobertsJ,HendersonSR.Theeffectiveness e Althoughthisstudyhasnotshownanybeneficialeffectsofa alitnedraetuffricei.eMnceydoCfahroemRee-bsaRseevd2n0u0rs6i;n6g3:h5e3a1l-th69p.romotionforolderpeople:reviewofthe d as proactiveprimarycareapproach,includingamultidimensional 7 StijnenMMN,Duimel-PeetersIGP,JansenMWJ,VrijhoefHJM.Earlydetectionofhealth 1 assessmentandinterdisciplinarycarebasedonatailormade p[Gro]OblLeDm:sdienspigonteonftiaallloynfgraitiulcdoinmaml,quunaitys-i-dewxeplelirnimgeolndtearlspteuodpyl.eBbMyCgeGneerriaaltprr2a0c1ti3c;e1s3—(7p):r1o-j1e0ct. 0.1 treatmentplanandregularevaluationandfollow-up,infrail 1 8 FletcherAE,JonesDA,BulpittCJ,TullochAJ.TheMRCtrialofassessmentand 3 olderpeople,itaddstotheevidencebaseforclinicaldecision managementofolderpeopleinthecommunity:objectives,designandinterventions 6 [ISRCTN23494848].BMCHealthServRes2002;2(1):21. /b makingandfutureresearchregardingcommunitybasedcare m 9 FriedLP,FerrucciL,DarerJ,WilliamsonJD,AndersonG.Untanglingtheconceptsof forsuchpeople.Thepublicationofnon-effectivestudiesis disability,frailty,andcomorbidity:implicationsforimprovedtargetingandcare.Gerontol j.f5 highlyrelevanttopreventanoverestimationofthebenefitsof 10 MAaBrikolleS-RceiMideMd,SBcrio2w0n0e4;G5.9C:2o5n5c-6e3p.tualizationsoffrailtyinrelationtoolderadults.JAdv 264 interventionsandawasteofhealthcareresources.Thisstudy Nurs2003;44:58-68. o contributestotheemergingbodyofevidencethatmoreresearch 11 PmeelaLsiuttreinlgREo,faScchounucrempat.nJsNMuJt,rEHmeamltehloAtgVinognk20M0H9;,1V3e:3rh9a0a-4r.HJJ.Frailty:definingand n 1 isneededtoimprovetheeffectivenessofinterventionsforfrail 12 LiebelDV,FriedmanB,WatsonNM,PowersBA.Reviewofnursehomevisiting 0 S olderpeople. interventionsforcommunity-dwellingolderpersonswithexistingdisability.MedCareRes e Rev2009;66:119-46. p 13 HallbergIR,KristenssonJ.Preventivehomecareoffrailolderpeople:areviewofrecent te Wethanktheparticipantsinthestudy;theparticipatinghealthcare casemanagementstudies.JClinNurs2004;13(6B):112-20. m b p(MroEfeMsIsCio):nAanlsi;tathLeegCteenntbreerfgo,rADlfaotnasaSncdhIrnoftoernm,aatniodnMMaarlnèangeeRmoennnter;the 14 D2p0rae1nv0iee;n7lst:3Rd7i,s-5Ma5be.itliztyelitnhifnraSil,cVoamnmRuonsistyu-mdwEe,llDinegWolidtteerLp,eVrsaonndse:nanHoevuevrevlieWw..InEtuerrvJeAngtioeninsgto er 20 15 DanielsR,vanRossumE,MetzelthinS,SipersW,HabetsH,HobmaS,etal.Adisability 1 membersoftheresearchgroup:LiloCrasborn(MCCOmnes),Simone 3 Denis(MCCOmnes),MarlouWolters(MCCOmnes),HerbertHabets p2r0e1v1e;n2t5io:9n6p3r-o7g4r.ammeforcommunity-dwellingfrailolderpersons.ClinRehabil . D (OrbisMedicalCentre),andRamonDaniels(ZuydUniversityofApplied 16 MetzelthinSF,vanRossumE,deWitteLP,HendriksMR,KempenGI.Thereductionof ow disabilityincommunity-dwellingfrailolderpeople:designofatwo-armclusterrandomized n Sciences,Heerlen);theirresearchassistants:FloorKoomen,Ine controlledtrial.BMCPublicHealth2010;10:511. lo Hesdahl,andAstridDello;andtheirsponsors:theNetherlands 17 CampbellMK,ElbourneDR,AltmanDG,CONSORTGroup.CONSORTstatement: ad extensiontoclusterrandomisedtrials.BMJ2004;328:702-8. e OrganisationforHealthResearchandDevelopment(ZonMw),the 18 SteverinkN,SlaetsJPJ,SchuurmansH,vanLisM.Measuringfrailty:developmentand d Hague,theNetherlands,andtheNetherlandsOrganisationforScientific testingoftheGroningenFrailtyIndicator(GFI).TheGerontologist2001;41(1):236-37. fro 19 DerckxE.Eerstenursepractitionersmetdifferentiatiehuisartsenzorg.Tijdschriftvoor m Research(NWO),theHague,theNetherlands. Verpleegkundigen2006;3:26-30. h Ccoonncteripbtuiotonrsa:nSdFdMes,iEgnvRo,ftLhPedsWtu,dayn.SdFGMIJ,MEKvRw,eLrPedrWes,pAoWnsAib,SleOfoHr,tWheS, 2201 VDMaaenndiWe2l0se1eR2l.;C2F,5rSa(sicluhepelpdrlse1rHl)y:,S—T1iim2d-em7n.etirfmicaatniosnAa.nHdeadlitshacbailirteyipnrethveenNteiothneinrlapnrdims.aJryAcmarBeo[aPrhdDFam ttp://w andGIJMKwereinvolvedintheanalysisandinterpretationofthedata. Thesis].MaastrichtUniversity,2011. w 22 GlasgowRE,EmontS,MillerDC.Assessingdeliveryofthefive‘As’forpatient-centred w SFMcreatedthefirstdraftofthispaper.Theotherauthorscommented counseling.HealthPromotInt2006;21:245-55. .b onitandapprovedthefinalversion.Allauthorshadfullaccesstoallof 23 MillerWR,RollnikS.Motivationalinterviewing:preparingpeopleforchange.Guildford m thedataandcantakeresponsibilityfortheintegrityofthedataandthe Publications,2002. j.c 24 KempenGIJM,MiedemaI,OrmelJ,MolenaarW.Theassessmentofdisabilitywiththe o accuracyofthedataanalysis.SFMistheguarantor. GroningenActivityRestrictionScale:conceptualframeworkandpsychometricproperties. m Funding:ThisresearchisfundedbytheDutchNationalCareforthe SocSciMed1996;43:1601-10. o/ 25 StuckAE,WalthertJM,NikolausT,BulaCJ,HohmannC,BeckJC.Riskfactorsfor n ElderlyProgrammebyTheNetherlandsOrganisationforHealth functionalstatusdeclineincommunity-livingelderlypeople:asystematicliteraturereview. 1 ResearchandDevelopment(ZonMw311070301).Itisaninitiativeof 26 SSpoicnhSocvieMnePd,1O99rm9;e4l8J:,4S45lo-e6k9e.rsPPA,KempenGIJM,SpeckensAEM,vanHemertAM. 5 Ja theDutchMinistryofHealth,WelfareandSporttoimprovethequality AvalidationstudyoftheHospitalAnxietyandDepressionScale(HADS)indifferentgroups n u ofcareforfrailolderpeople(www.nationaalprogrammaouderenzorg. ofDutchsubjects.PsycholMed1997;27:363-70. a 27 KempenGIJM,vanEijkLM.ThepsychometricpropertiesoftheSSL12-l,ashortscale ry nl).OpenaccessofthispublicationwasfinancedbytheNetherlands formeasuringsocialsupportintheelderly.SocIndicRes1995;35:303-12. 2 OrganisationforScientificResearch(NWO). 28 KempenGIJM,YardleyL,vanHaastregtJCM,ZijlstraGAR,BeyerN,HauerK,etal.The 02 ShortFES-I:ashortenedversionofthefallsefficacyscale-internationaltoassessfearof 3 Competinginterests:AllauthorshavecompletedtheICMJEuniform falling.AgeAgeing2008;37:45-50. b disclosureformatwww.icmje.org/coi_disclosure.pdf(availableon 29 MsoacrisalGpMarJti,cKipeamtiopnenpGroIfJilMe:,dPeovsetlMopWmMen,tParonodtcIMlin,imMeetsrtiecrpsrIo,pvearntieEsijkinJTolMde.rTahdeuMltsaawsittrhicaht y gu requestfromthecorrespondingauthor)anddeclare:thisresearchis chronicphysicalillness.QualLifeRes2009;18:1207-18. e fundedbytheDutchNationalCarefortheElderlyProgrammebyThe 3301 PKeeamrplinenLIG,SIJcMh.oPorleerveCn.tTieheenstoruucdteurrewoorfdceonp:inmgo.gJeHlijekahlethdeSnovcoBoerhzaevlfm19a7n8a;g1e9m:2e-2n1t.. st. P NetherlandsOrganisationforHealthResearchandDevelopment;no Epidemiologischbulletin2006;41(2):27-32. ro financialrelationshipswithanyorganisationsthatmighthaveaninterest 32 MetzelthinSF,DanielsR,VanRossumE,CoxK,HabetsH,deWitteLP,etal.Anurse-led te interdisciplinarycareapproachtopreventdisabilityamongcommunity-dwellingfrailolder c inthesubmittedworkinthepreviousthreeyears;nootherrelationships people:alarge-scaleprocessevaluation.IntJNursStud2013;50:1184-96. te d oractivitiesthatcouldappeartohaveinfluencedthesubmittedwork. 33 ElkanR,KendrickD,DeweyM,HewittM,RobinsonJ,BlairM,etal.Effectivenessof b homebasedsupportforolderpeople:systematicreviewandmeta-analysis.BMJ y Ethicalapproval:ThestudywasapprovedbytheMedicalEthical 2001;323:719-25. c CommitteeoftheMaastrichtUniversity/AcademicHospitalMaastricht 34 StuckAE,EggerM,HammerA,MinderCE,BeckJC.Homevisitstopreventnursing op homeadmissionandfunctionaldeclineinelderlypeople:systematicreviewand y intheNetherlandsin2009(MEC09-3-067).Betweenrandomisation meta-regressionanalysis.JAMA2002;287:1022-8. rig andbaselinemeasurement,allincludedparticipantsgavewritten 35 BoumanA,vanRossumE,NelemansP,KempenGIJM,KnipschildP.Effectsofintensive h homevisitingprogramsforolderpeoplewithpoorhealthstatus:asystematicreview. t. informedconsent. BMCHealthServRes2008;8:74. Datasharing:Additionaldatafromthestudydatabaseareavailableon 36 HussA,StuckAE,RubensteinLZ,EggerM,Clough-GorrKM.Multidimensionalpreventive homevisitprogramsforcommunity-dwellingolderadults:asystematicreviewandmeta requestfromthecorrespondingauthorat analysisofrandomizedcontrolledtrials.JGerontolABiolSciMedSci2008;63:298-307. [email protected] 37 VanHaastregtJC,DiederiksJP,vanRossumE,deWitteLP,CrebolderHF.Effectsof preventivehomevisitstoelderlypeoplelivinginthecommunity:systematicreview.BMJ level,asparticipants’consenttosharetheirdatawasnotobtained. 2000;320:754-8. No commercial reuse: See rights and reprints http://www.bmj.com/permissions Subscribe:http://www.bmj.com/subscribe BMJ2013;347:f5264doi:10.1136/bmj.f5264(Published10September2013) Page7of12 RESEARCH Whatisalreadyknownonthistopic B M Comparedwithinstitutionalisation,communitybasedcareofolderpeoplemayachievebetteroutcomesatlowercostsandisalso J pVraerfieorursedcobmymoludneirtypeboapseledthinetemrsveelnvteiosnsaimingatreductionofdisabilityhavebeendevelopedduringthepastdecades : firs However,onlyasmallnumberofinterventionshaveshownbeneficialeffectsondisability,andmoststudiesdidnotreportonthelong t p u termeffects b lis Whatthisstudyadds h e d Noevidencewasfoundfortheeffectivenessofaproactiveprimarycareapproach,includingamultidimensionalassessment, a interdisciplinarycarebasedonatailormadetreatmentplan,andregularevaluationandfollow-up,infrailolderpeople s 1 Effectiveeducationandtrainingprogrammesareneededtoprovidehealthcareprofessionalswithadequatecompetenciesandfeasible 0 toolstofacethechallengesofcommunitybasedcareinfrailolderpeople .1 1 3 Thenon-effectiveresultsofthisstudymeanthatmoreresearchisneededtooptimisetheeffectivenessofcommunitybasedinterventions 6 forfrailolderpeople /b m j.f5 38 BernabeiR,LandiF,GambassiG,SgadariA,ZuccalaG,MorV,etal.Randomisedtrial 46 LawM,BaptisteS,McCollM,OpzoomerA,PolatajkoH,PollockN.TheCanadian 2 ofimpactofmodelofintegratedcareandcasemanagementforolderpeoplelivinginthe OccupationalPerformanceMeasure:anoutcomemeasureforoccupationaltherapy.Can 6 4 community.BMJ1998;316:1348-51. JOccupTher1990;57:82-7. o 39 GillTM,BakerDI,GottschalkM,PeduzziPN,AlloreH,ByersA.Aprogramtoprevent 47 KiresukTJ,ShermanRE.Goalattainmentscaling:ageneralmethodforevaluating n functionaldeclineinphysicallyfrail,elderlypersonswholiveathome.NEnglJMed comprehensivecommunitymentalhealthprograms.CommunityMentHealth 1 40 F20le0tc2h;3e4r7A:1E0,6P8ri-c7e4.GM,NgESW,StirlingSL,BulpittCJ,BreezeE,etal.Population-based 48 1S9m6e8u;1ld:e4r4s3E-5S3T.F,vanHaastregtJCM,AmbergenT,Uszko-LencerNHKM,Janssen-Boyne 0 S e multidimensionalassessmentofolderpeopleinUKgeneralpractice:acluster-randomised JJJ,GorgelsAPM,etal.Nurse-ledself-managementgroupprogrammeforpatientswith p factorialtrial.Lancet2004;364:1667-77. congestiveheartfailure:randomizedcontrolledtrial.JAdvNurs2010;66:1487-99. te 41 DanielsR,VanRossumHIJ,BeurskensA,VandenHeuvelW,DeWitteL.Thepredictive m validityofthreeself-reportscreeninginstrumentsforidentifyingfrailolderpeopleinthe Accepted:15August2013 b e 42 cPoijmpemrsuEni,tyF.eBrrMeiCraPI,uSbtleichoHuewaelthrC2D01A2,;N1i2e:u6w9.enhuijzenKrusemanAC.Thefrailtydilemma: r 2 Citethisas:BMJ2013;347:f5264 0 reviewofthepredictiveaccuracyofmajorfrailtyscores.EurJInternMed2012;23:118-23. 1 43 GTylapsegsoowfeRvEid,eEnmcemnoenesdKeMd..AHnonwucRaenvwPeuibnlcicreHaesaeltthra2n0s0la7t;i2o8n:4o1f3re-3se3a.rchintopractice? ThisisanOpenAccessarticledistributedinaccordancewiththeCreativeCommons 3. D AttributionNonCommercial(CCBY-NC3.0)license,whichpermitsotherstodistribute, 44 SiegertRJ,TaylorWJ.Theoreticalaspectsofgoal-settingandmotivationinrehabilitation. o DisabilRehabil2004;26:1-8. remix,adapt,builduponthisworknon-commercially,andlicensetheirderivativeworks wn 45 ParsonsJGM,ParsonsMJG.Theeffectofadesignatedtoolonperson-centredgoal ondifferentterms,providedtheoriginalworkisproperlycitedandtheuseis lo identificationandserviceplanningamongolderpeoplereceivinghomecareinNewZealand. non-commercial.See:http://creativecommons.org/licenses/by-nc/3.0/. a HealthSocCareCommunity2012;20:653-62. de d fro m h ttp ://w w w .b m j.c o m o/ n 1 5 J a n u a ry 2 0 2 3 b y g u e s t. P ro te c te d b y c o p y rig h t. No commercial reuse: See rights and reprints http://www.bmj.com/permissions Subscribe:http://www.bmj.com/subscribe BMJ2013;347:f5264doi:10.1136/bmj.f5264(Published10September2013) Page8of12 RESEARCH Tables B M J Table1|Baselinecharacteristicsofparticipantsincontrolgroupandinterventiongroup.Valuesarenumbers(percentages)unlessstated : firs otherwise t p u Characteristics Control(n=153) Intervention(n=193) b lis Mean(SD)age(years) 76.80(4.92) 77.49(5.28) h e d Femalesex 93(61) 106(55) a s Livingalone 80(52) 90(47) 1 0 Loweducation 94(61) 108(56) .1 1 Mean(SD)scores: 36 /b GARStotal 30.58*(10.62) 33.09*(11.52) m GARSADLscale 16.54*(5.35) 17.97*(6.14) j.f5 2 GARSIADLscale 14.03(5.86) 15.12(5.96) 6 4 MSPP-CP-D 1.90(1.63) 1.63(1.48) o n MSPP-CP-F 0.46(0.44) 0.36(0.35) 10 MSPP-FSP-D 0.73(0.88) 0.610.84 S e p MSPP-FSP-F 0.45(0.63) 0.38(0.55) te m ShortFES-I 12.38(4.72) 13.24(5.39) b e HADS-D 6.69(4.35) 6.54(3.77) r 2 0 SSL-I12 27.46(6.06) 27.17(6.30) 1 3 GFI 6.72*(1.71) 7.13*(1.89) . D o PMS 21.41(4.25) 21.97(4.01) w n lo GARS=GroningenActivityRestrictionScale(rangetotalscale18-78,rangeactivitiesofdailyliving(ADL)scale11-44,rangeinstrumentalADL(IADL)scale7-28; a d higherscoresindicatemoredisability);MSPP=MaastrichtSocialParticipationProfile;MSPP-CP-D=MSPPconsumptiveparticipation,diversityscore(range0-7; e d higherscoreindicatesmorediverseconsumptiveparticipation);MSPP-CP-F=MSPPconsumptiveparticipation,frequencyscore(range0-3;higherscoreindicates fro morefrequentconsumptiveparticipation);MSPP-FSP-D=MSPPformalsocialparticipation,diversityscore(range0-7;higherscoreindicatesmorediverseformal m socialparticipation);MSPP-FSP-F=MSSPformalsocialparticipation,frequencyscore(range0-3;higherscoreindicatesmorefrequentformalsocialparticipation); h SShcaolret—FEdSep-Ir=eSshsoiorntFsaulblsscEaffleica(rcaynSgeca0le-2—1,Inhtiegrhneartisocnoarle(rianndgiceat7e-s28m;ohrieghdeerpsrecossreivien)d;iScaStLe-sI1m2o=rSeofceiaalrSoufpfaplloinrtgL);isHt—ADInSte-Dra=cHtioosnpvitearlsAionnxi(eratyngaend12D-e4p8r;ehsisgihoenrscore ttp://w w indicatesmoresocialsupport);GFI=GroningenFrailtyIndicator(range0-15;higherscoreindicatesmoreseverefrailty);PMS=PearlinMasteryScale(higherscore w indicateslessowncontrol). .b m *Significantdifferences:P<0.05. j.c o m o/ n 1 5 J a n u a ry 2 0 2 3 b y g u e s t. P ro te c te d b y c o p y rig h t. No commercial reuse: See rights and reprints http://www.bmj.com/permissions Subscribe:http://www.bmj.com/subscribe BMJ2013;347:f5264doi:10.1136/bmj.f5264(Published10September2013) Page9of12 RESEARCH Table2|Multilevelanalysesfordifferencesbetweencontrolgroup(CG)andinterventiongroup(IG)forprimaryoutcomemeasuresat6, 12,and24months’follow-up(n=310) B M J Mea6nm(oSnDt)hs’follow-upMean Me1a2nm(SoDn)ths’follow-upMean Me2a4nm(SoDn)ths’follow-upMean : firs difference* difference* difference* t p u (95%CI);P (95%CI);P (95%CI);P b Outcome CG IG value CG IG value CG IG value lis h GARS 30.16 32.83 0.41(−0.80to 30.81 33.08 0.47(−0.81to 31.50 34.39 1.18(−0.35to ed (10.07) (10.98) 1.62);0.51 (10.29) (11.34) 1.76);0.47 (10.92) (11.58) 2.71);0.35 a s GARS 16.17(5.13) 17.54(5.82) 0.25(−0.44to 16.30(5.31) 17.81(5.90) 0.59(−0.14to 16.73(5.73) 18.31(5.82) 0.77(−0.05to 1 0 ADL 0.94);0.48 1.33);0.11 1.59);0.07 .1 1 GARS 14.00(5.51) 15.29(5.92) 0.17(−0.63to 14.51(5.69) 15.28(6.03) −0.12(−0.93to 14.77(5.86) 16.08(6.35) 0.40(−0.54to 3 6 IADL 0.97);0.67 0.68);0.76 1.34);0.41 /b m GARS=GroningenActivityRestrictionScale(rangetotalscale18-78;higherscoresindicatemoredisability);GARSADL=GroningenActivityRestrictionScale—activities j.f5 ofdailyliving(ADL)subscale(rangetotalscale11-44;higherscoresindicatemoredisability);GARSIADL=GroningenActivityRestrictionScale—instrumental 2 6 ADLsubscale(rangetotalscale7-28;higherscoresindicatemoredisability). 4 o *Adjustedforage,sex,education,andsignificantdifferencesatbaseline(frailtyanddisability). n 1 0 S e p te m b e r 2 0 1 3 . D o w n lo a d e d fro m h ttp ://w w w .b m j.c o m o/ n 1 5 J a n u a ry 2 0 2 3 b y g u e s t. P ro te c te d b y c o p y rig h t. No commercial reuse: See rights and reprints http://www.bmj.com/permissions Subscribe:http://www.bmj.com/subscribe BMJ2013;347:f5264doi:10.1136/bmj.f5264(Published10September2013) Page10of12 RESEARCH Table3|Multilevelanalysesfordifferencesbetweencontrolgroup(CG)andinterventiongroup(IG)forsecondaryoutcomemeasuresat 6,12,and24months’follow-up B M J Me6anm(oSnDt)hs’follow-upMean Me1a2nm(SoDn)ths’follow-upMean Me2a4nm(SoDn)ths’follow-upMean : firs difference* difference* difference* t p u (95%CI);P (95%CI);P (95%CI);P b Outcome CG IG value CG IG value CG IG value lis h MSPP-CP-D 2.10(1.63) 1.92(1.57) −0.06(−0.21to 2.10(1.64) 1.73(1.45) −0.22(−0.48to 1.94(1.70) 1.61(1.33) −0.13(−0.43to ed (n=310) 0.09);0.71 0.03);0.09 0.16);0.38 a s MSPP-CP-F 0.46(0.40) 0.40(0.36) 0.00(−0.04to 0.45(0.40) 0.35(0.32) −0.05(−0.11to 0.44(0.45) 0.33(0.31) −0.04(−0.11to 1 0 (n=310) 0.03);0.96 0.01);0.12 0.04);0.32 .1 1 MSPP-FSP-D 0.69(0.88) 0.64(0.79) 0.03(−0.05to 0.73(0.91) 0.60(0.81) −0.06(−0.14to 0.71(0.87) 0.58(0.77) −0.04(−0.12to 3 6 (n=310) 0.10);0.73 0.02);0.43 0.04);0.57 /b m MSPP-FSP-F 0.41(0.59) 0.39(0.55) 0.03(−0.02to 0.43(0.62) 0.35(0.51) −0.03(−0.08to 0.45(0.64) 0.34(0.51) −0.05(−0.11to (n=310) 0.08);0.52 0.01);0.47 0.00);0.31 j.f5 2 HADS-D 5.82(3.88) 5.72(3.49) −0.11(−0.80to 5.68(3.92) 6.36(4.13) 0.78(0.04to 6.10(3.78) 5.97(4.18) −0.07(−0.90to 64 (n=305) 0.58);0.76 1.53);0.04 0.77);0.87 o n SSL-I12 26.94(5.53) 27.03(6.36) 0.18(−079to 27.27(6.54) 27.10(6.09) −0.12(−1.22to 27.35(6.27) 26.76(5.98) −0.29(−1.37to 1 0 (n=312) 1.15);0.71 0.99);0.84 0.79);0.60 S e ShortFES-I 12.37(4.90) 12.66(5.25) −0.67(−1.48to 12.15(5.24) 13.42(5.43) 0.34(−0.54to 12.95(5.29) 13.73(5.75) −0.04(−1.01to p (n=306) 0.14);0.11 1.22);0.44 0.93);0.94 te m b MSPP=MaastrichtSocialParticipationProfile,MSPP-CP-D=MSPPconsumpativeparticipationdiversityscore(range0-7;higherscoreindicatesmorediverse e consumptiveparticipation);MSPP-CP-F=MSPPconsumptiveparticipation,frequencyscore(range0-3;higherscoreindicatesmorefrequentconsumptive r 2 0 participation);MSPP-FSP-D=MSPPformalsocialparticipation,diversityscore(range0-7;higherscoreindicatesmorediverseformalsocialparticipation); 13 MSPP-FSP-F=MSPPformalsocialparticipation,frequencyscore(range0-3;higherscoreindicatesmorefrequentformalsocialparticipation);HADS-D=Hospital . D o AnxietyandDepressionScale—depressionsubscale(range0-21,higherscoreindicatesmoredepressive);SSL-I12=SocialSupportList—Interactionversion w (range12-48;higherscoreindicatesmoresocialsupport);ShortFES-I=ShortFallsEfficacyScale—International(range7-28;higherscoreindicatesmorefearof nlo falling). a d *Adjustedforage,sex,education,significantdifferencesatbaseline(frailtyanddisability),andbaselinestatusoutcomemeasure. ed fro m h ttp ://w w w .b m j.c o m o/ n 1 5 J a n u a ry 2 0 2 3 b y g u e s t. P ro te c te d b y c o p y rig h t. No commercial reuse: See rights and reprints http://www.bmj.com/permissions Subscribe:http://www.bmj.com/subscribe