Fried phenotype of frailty: cross-sectional comparison of three ...

The stages of frailty, based on the Fried criteria, were defined as follows: a score of 0 means that a person is robust or not frail. Persons ... Skiptomaincontent Advertisement SearchallBMCarticles Search DownloadPDF Researcharticle OpenAccess Published:09July2015 Friedphenotypeoffrailty:cross-sectionalcomparisonofthreefrailtystagesonvarioushealthdomains LindaP.M.OphetVeld1,2,ErikvanRossum1,2,GertrudisI.J.M.Kempen2,HenricaC.W.deVet3,KlaasJanHajema4&AnnaJ.H.M.Beurskens1,5  BMCGeriatrics volume 15,Article number: 77(2015) Citethisarticle 26kAccesses 60Citations 24Altmetric Metricsdetails AbstractBackgroundThepopulationageinginmostWesterncountriesleadstoalargernumberoffrailolderpeople.Thesefrailpeopleareatanincreasedriskofnegativehealthoutcomes,suchasfunctionaldecline,falls,institutionalisationandmortality.Manyapproachesareavailableforidentifyingfrailtyamongolderpeople.ResearchersmostoftenuseFriedandcolleagues’descriptionofthefrailtyphenotype.Theauthorsdescribefivephysicalcriteria.Otherresearcherspreferacombinationofmeasurementsinthesocial,psychologicaland/orphysicaldomains.Theaimofthisstudyistodescribethelevelsofsocial,psychologicalandphysicalfunctioningaccordingtoFried’sfrailtystagesusingalargecohortofDutchcommunity-dwellingolderpeople.MethodsTherewere8,684community-dwellingolderpeople(65+)whoparticipatedinthiscross-sectionalstudy.BasedonthefiveFriedfrailtycriteria(weightloss,exhaustion,lowphysicalactivity,slowness,weakness),theparticipantsweredividedintothreestages:non-frail(score0),pre-frail(score1–2)andfrail(score3–5).Thesestageswererelatedtoscoresinthesocial(socialnetworktype,informalcareuse,loneliness),psychological(psychologicaldistress,mastery,self-management)andphysical(chronicdiseases,GARSIADL-disability,OECDdisability)domains.Results63.2 %oftheparticipantswasnon-frail,28.1 %pre-frailand8.7 %frail.Whencomparingthethreestagesoffrailty,frailpeopleappearedtobeolder,weremorelikelytobefemale,weremoreoftenunmarriedorlivingalone,andhadalowerlevelofeducationcomparedtotheirpre-frailandnon-frailcounterparts.Thedifferencebetweenthescoresinthesocial,psychologicalandphysicaldomainswerestatisticallysignificantbetweenthethreefrailtystages.Themostpreferablescorescamefromthenon-frailgroup,andleastpreferablescoreswerefromthefrailgroup.Forexampleuseofinformalcare:non-frail3.9 %,pre-frail23.8 %,frail60.6 %,andGARSIADL-disabilitymeanscores:non-frail9.2,pre-frail13.0,frail19.7.ConclusionWhenolderpeoplewerecategorisedaccordingtothethreefrailtystages,asdescribedbyFriedandcolleagues,therewerestatisticallysignificantdifferencesinthelevelofsocial,psychologicalandphysicalfunctioningbetweenthenon-frail,pre-frailandfrailpersons.Non-frailparticipantshadconsistentlymorepreferablescorescomparedtothefrailparticipants.ThisindicatedthattheFriedfrailtycriteriacouldhelphealthcareprofessionalsidentifyandtreatfrailolderpeopleinanefficientway,andprovideindicationsforproblemsinotherdomains. PeerReviewreports BackgroundAnageingsocietyisacommonphenomenon.TheincreasingproportionofolderpeopleinmostWesterncountriesleadstoalargernumberofpeoplewhoareoldandfrail.Thesefrailolderpeopleareatanincreasedriskofnegativehealthoutcomes,suchasfunctionaldecline,falls,institutionalisationandmortality[1].Overthelastdecade,theinterestinfrailtyhasgrown[2].Themainreasonisthebeliefthatearlyidentificationofthoseatriskcouldhelptodelayorpreventtheadverseoutcomesoffrailty.Despiteconsiderableresearchonfrailty,thereisstilldebateonthenature,definition,prevalence,andthecharacteristicsofolderpeopleinvariousfrailty‘stages’[1,3].Threemainapproachestoconceptualisefrailtyhavebeendistinguished.Oneapproachconsidersfrailtytobeadeclineinphysicalfunctioning.Thefrailtyphenotype,asdescribedbyFriedandcolleagues,isbasedonfivepre-definedphysicalfrailtycriteria,whicharewellknownandmostfrequentlyusedbyresearchers[4,5].Anotherapproachistolookatfrailtyastheaccumulationofdeficitsacrossvariousdomains(e.g.cognition,physicalfunctioning,self-ratedhealth,smokinghistory,andlaboratoryresults).TheFrailtyIndex,developedbyRockwoodandcolleagues,isoftenusedforthisapproachanditischaracterisedbytheuseofanon-fixedsetofclinicalconditionsanddiseases[6,7].Athirdapproachalsoassumesthatmultipledomains(social,psychological,physical)areinvolvedintheconceptoffrailty,withresearchersusingapre-definedsetofquestionsrelatedtoeachdomain(e.g.TilburgFrailtyIndicator,GroningenFrailtyIndicator)[8,9].Eachapproachhasitsadvantagesanddisadvantages.Inthepresentstudy,wearelookingforabriefandsimpletool(i.e.aself-reportquestionnairewithalimitednumberofitems)thatisfeasibleforuseinlargepopulationsofcommunity-dwellingolderpeople.TheFriedfrailtycriteriaseemtoreflectsuchatool.AlthoughtheFriedcriteriawereoriginallynotdevelopedasaself-reportquestionnaire,researchersnowadaysoftenuse(partly)modifiedquestionnairesthatarebasedonthefrailtyphenotype(e.g.Barretoandcolleagues,Macklaiandcolleagues)[10,11].Thefivefrailtycriteriaareweightloss,exhaustion,lowphysicalactivity,slownessandweakness.Thesumscoreofthesefivecriteriaclassifiespeopleintooneofthreefrailtystages(orgroups):notfrail(score0),pre-frail(score1–2)andfrail(score3–5).FriedandcolleaguesdescribedthecharacteristicsofthesethreegroupsusingacohortofUnitedStatescitizens.Thetrendwasthatfrailpeoplewereolder,morelikelytobefemale,sufferedfrommorediseases(exceptcancer),reportedhigherratesofdisability,werelesseducated,hadlowerincome,wereinpoorerhealth,hadmorecognitiveimpairmentsandexperiencedhigherlevelsofdepressivesymptomscomparedtotheirpre-frailandnon-frailcounterparts[5].Resultsfromthepre-frailpeoplewereintermediate,fallingbetweenthescoresofthefrailandnon-frailpeople(exceptforcancer).Inaddition,outcomesoftheSurveyofHealth,AgingandRetirementinEurope(SHARE),whichalsousedtheFriedcriteriatoassessfrailtyinpopulationsfrom11Europeancountries,showedthatfrailpeopleweremorelikelytobefemaleandreportmoredisabilityproblemscomparedtotheirpre-frailandnon-frailcounterparts[11].Theparticularcharacteristicsofinterestinbothaforementionedstudiesweredemographics,andaspectsinthephysicaldomain,aswellas(chronic)diseases.StudiesfromBandeen-Rocheandcolleagues[12],Bleandcolleagues[13]andCawthonandcolleagues[14]alsousedthefiveFriedfrailtycriteriatodifferentiatebetweengroups,focusingonsimilarcharacteristicsofinterest.Itisstillunclearwhetherthislimitedscopeissufficientforidentifyingdifferentprofilesoffunctioningoffrail,pre-frailandnon-frailolderpeople.Levelsofsocialandpsychologicalfunctioningmightalso,forexample,playanimportantroleinthedevelopmentoffrailty.AdditionalknowledgeregardingwhethersuchsocialandpsychologicalfactorscouldaddtothediscriminativepowerofthethreeFriedfrailtystageswillbeveryusefulforbothhealthcareprofessionalsandresearchers.Upuntilnow,thepsychologicalandsocialfactorsrelativetothefrailtystageshavenotbeenextensivelystudied.Ifthesestagesalsoshowvariationsinthesedomains,thiscouldhelphealthcareprofessionalsinefficientlyidentifyingandtreatingfrailolderpeople.Ifapatientis(pre-)frailaccordingtotheFriedcriteria,itcouldalertthemtotheexistenceofproblemsinotherdomainsaswell.Moreover,asthenumberofitemsoftheFriedfrailtycriteriaislimited,theuseofthisshortinstrumentismuchmoreefficientthanmanyotherfrailtymeasures.Theaimofthisstudyistodescribethelevelsofsocial,psychologicalandphysicalfunctioningaccordingtothethreeFriedfrailtystagesusingalargecohortofDutchcommunity-dwellingolderpeople.Wealsostudiedpossiblegenderdifferencesintheselevelsoffunctioning.MethodsAcross-sectionalstudywasconductedamongcommunity-dwellingolderpeopleinLimburg,aprovinceinthesouthernpartofTheNetherlands.ThemedicalethicalcommitteeAtrium-Orbis-Zuydapprovedthisstudy(12-N-129).SelectionofthestudypopulationwasmadefromtheHealthMonitor,anextensivepostalgeneralhealthquestionnairewhichissenteveryfouryearsbytheCommunityHealthServicetoalargesampleofcommunity-dwellingpeopleintheNetherlands[15].StudypopulationForthemeasurementusingtheHealthMonitorinLimburg,duringthefallof2012,56,000peopleaged55 yearsandoverwereselected.Selectionwasrandomforallagegroups,exceptforthoseover75 years.Thispopulationwasoverrepresentedinthesampleinordertoobtainsufficientdataamongtheoldestagegrouplivingathome.Peoplelivinginneighbourhoodswithalowsocioeconomicstatuswereoverrepresentedaswell.Respondentswereaskedtogivetheirconsentforusingtheirdataforourstudy.TheresponseratefortheHealthMonitorwas54 %(n = 30,130).Oftherespondents,13,521gavepermissionfortheuseoftheirdatainourstudy.Theselectionwasalsorestrictedtothosewhowere65 yearsandolder,becausethisistheagegroupinwhichtheFriedcriteriawereoriginallydeveloped[5].Afterexcludingthequestionnairesthatwerefilledoutbyapersonotherthantheaddresseeandthosequestionnaireswithasignificantamountofmissingdata,atotalof8,684peopleparticipatedinourstudy.MeasurementsTheHealthMonitoriscomprisedofabroadrangeofquestions.Inadditiontodemographiccharacteristics(age,gender,maritalstatusandlevelofeducation),questionsincludedtheFriedfrailtycriteria,(chronic)diseases,useofhealthcareservices,useofinformalcareanditemsaboutsocial,psychologicalandphysicalfunctioning.FriedfrailtycriteriaFriedandcolleaguesdevelopedfivecriteria(weightloss,exhaustion,lowphysicalactivity,slownessandweakness)tobeusedforidentifyingfrailolderpeople[5].Incontrastwiththeoriginalcriteria,wereplacedthetwophysicalmeasurementsofslownessandweaknessbyquestions.Weightlosswasmeasuredusingthequestion:“Inthelastyear,haveyoulostmorethan4.5kilogramsunintentionally?(i.e.notduetodietingorexercise)”.ThisquestionisthesameasproposedbyFriedandcolleagues,onlypoundswerereplacedbykilograms.Thiscriterionwasmetwhentheparticipantanswered“yes”.ExhaustionwasmeasuredusingtwoquestionsfromtheCenterforEpidemiologicStudiesDepression(CES-D)scale:“Howoftendidyoufeelthateverythingyoudidwasaneffort?”and“Howoftendidyoufeelthatyoucouldnotgetgoing?”[16,17].ThesequestionsarethesameasproposedbyFriedandcolleagues.Responseoptionswereslightlydifferent:“always,mostofthetimes,sometimes,occasionally,never”,comparedto“rarelyornoneofthetime(<1 day),someoralittleofthetime(1–2days),amoderateamountofthetime(3–4days),mostofthetime”inFried’sversion.Thiscriterionwasmetwhenparticipantsanswered:“alwaysormostofthetimes”toatleastoneofthetwoquestions.LowphysicalactivitywasnotmeasuredbyusingtheMinnesotaLeisureTimeActivityQuestionnaire,asproposedbyFriedandcolleagues.Instead,aslightlyadjustedversionoftheShortQuestionnairetoAssessHealth-enhancingphysicalactivity(SQUASH)wasused[18].Participantshadtoanswerquestionsabouthowmanytimesaweektheyspenttimewalking,cycling,gardening,doingoddjobsorexercising/playingsports.Foreachactivity,theyhadtoreporthowmuchtimetheyspentengagedinthatactivityoneachoccasion.Kilocaloriesperweekwerecalculated.Theresultswerestratifiedbygenderandcomparedwiththecut-offvaluesasdescribedbyFriedandcolleagues(men383 kcal/week,women270 kcal/week).Ifapersonusedfewerkcalsperweekthiscriterionwasmet.Slowness/walktimewasmeasuredusingthequestion:“Canyoureachtheothersideoftheroadwhenthelightturnsgreenatazebracrossing?”Wedevelopedthisquestionourselves.Iftheparticipantchoseanyreplyotherthan“yes,withoutanytrouble”,thecriterionwasmet.Weakness/gripstrengthwasmeasuredbyaskingthequestion:“Doyouexperiencedifficultiesindailylifebecauseoflowgripstrength?”ThisquestionwasderivedfromtheTilburgFrailtyIndicator[8].Iftheparticipantanswered“yes”,thecriterionwasmet.Thestagesoffrailty,basedontheFriedcriteria,weredefinedasfollows:ascoreof0meansthatapersonisrobustornotfrail.Personswithascoreof1or2areatintermediateriskforadverseoutcomesorareconsideredtobepre-frail.Ascoreof3–5indicatesthatsomeoneisfrail[5].PerceivedhealthandhealthcareuseOnequestionwasaskedregardingperceivedhealth:“Howwellisyourhealthingeneral?”Thequestioncouldbeansweredona5-pointLikertscalewithanswerchoicesrangingfrom“verygood”to“verypoor”.Theuseofhealthcareserviceswasmeasuredbyreportinganycontactwithageneralpractitionerwithinthelasttwomonths.Theparticipantsalsohadtoprovidedetailsregardingthehealthcareprofessionaltheyhadcontactedoverthepasttwelvemonths.Thehealthcareproviderswerealreadyspecified:medicalspecialist,dietician,occupationaltherapist,physiotherapist,homecare(nursingcareandhouseholdcare)andsocialworker.SocialdomainWengerandcolleaguesdevelopedan8-itemquestionnaireregardingsocialnetwork[19].Thescoresdividedpeopleintofivetypesofsupportnetworks:familydependent,locallyintegrated,localself-contained,widercommunityfocused,andprivaterestricted.Thefamilydependentandprivaterestrictedsupportnetworksarecharacterisedbyalimitednumberofpeoplethatcouldprovidesupport.Thelocallyintegratedandwidercommunity-focusedsupportnetworksarelargernetworks.Wengerandcolleaguesfoundthatthesenetworktypeswereconsistentwiththeavailabilityofinformalsupportandtheuseofhealthcareservices[19].Inaddition,onequestionwasaskedabouttheuseofinformalcareoverthepast12 months.LonelinesswasmeasuredbyusingtheDeJong-GierveldLonelinessScale[20].Thisisan11-itemscale,withquestionssuchas“Imisshavingareallyclosefriend”,whichallowstheparticipantstochoosefromthreeanswerchoices:“yes”,“moreorless”or“no”.Ahigherscoreindicatesmorefeelingsofloneliness.PsychologicaldomainThe10-itemKesslerPsychologicalDistressScale(K-10)wasusedtomeasurepsychologicaldistress[21].Thisquestionnaireiscomprisedofquestionssuchas:“Duringthelastfourweeks,abouthowoftendidyoufeeldepressed?”Thefive-categoryresponsescalerangedfrom“allofthetime”(score5)to“never”(score1).Ahighertotalscoreindicatedhigherlevelsofpsychologicaldistress.MasterywasassessedbyusingPearlinandSchooler’sinstrument[22].Sevenstatements,suchas:“Ihavelittlecontroloverthethingsthathappentome”,areansweredusinga5-pointscale,rangingfrom“Itotallyagree”to“Itotallydisagree”.Thehigherthetotalscore,themoretherespondentthinksthatlife-chancesareunderone’sowncontrol.Self-managementwasmeasuredusingtheshortversionoftheSelf-ManagementAbilityScale(SMAS-S)[23].TheSMAS-Sconsistsofsixthree-itemsubscales(takinginitiative,investmentbehaviour,variety,multifunctionality,self-efficacyandpositiveframeofmind),whichreflectcoreabilitiestoformtheconstructofself-managementofwell-being[24].Responseoptionswereslightlyadjustedsothateveryquestionhadsixpossibleanswers.Therefore,thefinalscoresrangefrom1to6,withahigherscoreindicatingmoreself-managementabilities.PhysicaldomainChronicdiseasesweremeasuredbyaskingparticipantswhetherornottheysufferedfromoneormoreofthefollowingchronicdiseases:diabetes,stroke/cerebralhaemorrhage/cerebralinfarction,myocardialinfarction,othercardiacdiseases,cancer,asthma,chronicobstructivepulmonarydisease(COPD),hiporkneearthrosis,chronicjointinflammation,orbackproblems(incl.hernia).Forcancerandmyocardialinfarctiontheparticipantshadtoreportiftheyeverhadthediseases.Foralloftheotherdiseases,theyhadtoreportwhethertheysufferedfromthediseaseoverthepasttwelvemonths.IADL-disability(Instrumentalactivitiesofdailyliving)wasmeasuredusingaseven-itemsubscalefromtheGroningenActivityRestrictionScale(GARS)[25,26].Thesubscaleiscomprisedofquestions,suchas“Canyoufullyindependentlypreparedinner?”Theitemswereansweredonafour-pointscale,withanswersrangingfrom“Yes,withoutanydifficulty”to“No,onlywithsomeone’shelp”.Scoresrangefrom7to28points,withahigherscoreindicatingahigherlevelofIADL-disability.PhysicallimitationswereassessedusingtheOrganizationforEconomicCooperationandDevelopment(OECD)long-termdisabilityquestionnaire[27].Inthisstudy,weusedasix-itemversion,asusedbytheCommunityHealthService.Thisversioniscomprisedofquestionsaboutproblemswithhearing,vision,bending,andwalking400metres.Thenumberofitemsthatpeopleindicatedasproblematicwereusedforanalysis.StatisticalanalysisThecentralfocusofthisstudywastodescribethelevelsoffunctioningacrossvariousdomains.Descriptivestatisticswereusedtopresentdemographiccharacteristicsofthestudypopulationandthelevelsoffunctioning.AssociationsbetweenscoresinthethreehealthdomainsandthefrailtystageswereanalysedusingKendall’staufornominalandordinalvariables,andanalysesofvariance(ANOVA)forallothervariables(P