Gait parameter risk factors for falls under simple and dual task conditions ina FallsandBalanceResearchGroup,NeuroscienceResearchAustralia,Sydney,Australia b PrinceofWalesClinicalSchool,
Trang 1Gait parameter risk factors for falls under simple and dual task conditions in
a
FallsandBalanceResearchGroup,NeuroscienceResearchAustralia,Sydney,Australia
b
PrinceofWalesClinicalSchool,UniversityofNewSouthWales,Sydney,Australia
c
SchoolofPublicHealthandCommunityMedicine,UniversityofNewSouthWales,Sydney,Australia
ARTICLE INFO
Articlehistory:
Received13December2011
Receivedinrevisedform25May2012
Accepted26June2012
Keywords:
Gait
Dementia
Cognition
Accidentalfalls
Dualtask
ABSTRACT
Impairedgaitmaycontributetotheincreasedrateoffallsincognitivelyimpairedolderpeople.We investigatedwhethergaitundersimpleanddualtaskconditionscouldpredictfallsinthisgroup.The studysample consisted of64 communitydwelling olderpeoplewith mildto moderatecognitive impairment.Participantswalkedattheirpreferredspeedunderthreeconditions:(a)simplewalking,(b) walking while carrying a glass of water and (c) walking while counting backwards from 30 Spatiotemporal gait parameters were measured using the GAITRite1
mat Falls were recorded prospectivelyfor12monthswiththeassistanceofcarers.Twenty-two(35%)peoplefelltwoormore timesinthe12monthfollow-upperiod.Therewasasignificantmaineffectofgaitconditionanda significantmaineffectoffallerstatusformeanvaluemeasures(velocity,stridelength,doublesupport timeandstridewidth)andforvariabilitymeasures(swingtimevariabilityandstridelengthvariability) Examinationofindividualgaitparametersindicatedthatthemultiplefallerswalkedmoreslowly,had shorterstridelength,spentlongertimeindoublesupport,hadawidersupportwidthandshowedmore variabilityinstridelengthandswingtime(p<0.05).Therewasnosignificantinteractionbetweengait conditionandfallerstatusforanyofthegaitvariables.Inconclusion,dualtaskactivitiesadverselyaffect gaitincognitivelyimpairedolderpeople.Multiplefallersperformedworseineachgaitconditionbutthe additionofafunctionalorcognitivesecondarytaskprovidednoaddedbenefitindiscriminatingfallers fromnon-fallerswithcognitiveimpairment
ß 2012ElsevierB.V.Allrightsreserved
*Corresponding author at: Neuroscience Research Australia, Barker Street,
Randwick,NSW2031,Australia.Tel.:+61293991055;fax:+61293991204
** Co-correspondingauthorat:NeuroscienceResearchAustralia,BarkerStreet,
Randwick,NSW2031,Australia.Tel.:+61293991060;fax:+61293991204
E-mailaddresses:m.taylor@neura.edu.au(M.E Taylor),j.close@neura.edu.au
(JacquelineC.T.Close)
0966-6362/$–seefrontmatterß2012ElsevierB.V.Allrightsreserved
Trang 2of falls in community dwelling older adults with cognitive
impairment.
2.Methods
2.1.Participants
Sixty-fourcognitivelyimpairedparticipantswererecruitedfromanumberof
routinehealthservicesettings,communityservicesandadvertisementsinthelocal
press.Participationwasdependenton(i)havinganage60yearsorabove,(ii)living
inthecommunityand(iii)havinganidentifiedandwilling‘‘personresponsible’’
withatleast3.5hoffacetofacecontactperweek.Cognitiveimpairmentwas
defined as a Mini-Mental State Examination (MMSE)<24, Addenbrooke’s
CognitiveExamination–revised(ACE-R)<83[23]orwhereaspecialistclinician
hadmadeadiagnosisofcognitiveimpairment ordementia.Exclusioncriteria
included recent stroke (within 18months), progressive neurodegenerative
disorders(excluding dementia), insufficient English to completethe tasks or
knownendstageillness.ThestudywasapprovedbytheSouthEastSydneyHuman
ResearchEthicsCommitteeandconsentwasobtainedfromallparticipantsand
theirpersonresponsiblepriortoassessment
2.2.Assessment
Participantsandtheirpersonresponsiblewereinitiallyassessedintheirhome
environmentwithinformationobtainedondemographics,medicalandmedication
historyandfunctionalperformance
Consentingparticipantsattendedthegaitlaboratoryforthegaitassessment
TemporalandspatialgaitparametersweremeasuredwiththeGAITRite1
mat (CIRSystemsInc.,Clifton,NJ,USA).ThestandardGAITRite1
systemisa460cm matwithanactiveareaof2.23m2
containing13,824pressuresensorsarranged
inagridpatternwithaspatialresolutionof1.27cmandasamplingfrequency
of80Hz.GaitparameterswereobtainedfromGAITRiteGoldSoftwareVersion
3.3
ParticipantsperformedsixtrialsontheGAITRite1
mat,wearingcomfortable footwear,attheirpreferredwalkingspeedandwiththeirusualwalkingaidina
welllitandquietroom.Threegaitconditionswereassessed,eachwithtwotrials;
(i)simplewalking,(ii)walkingwhilecarryingaglassofwaterintheparticipants
preferredhand(filledto10mmfromrim)(Functionaldualtask)and(iii)walking
whilecountingbackwardsfrom30(Cognitivedualtask).Twoparticipantswho
mobilised witha walking framedid notperform the functional dual task
ParticipantsstartedoneandahalfmetresbeforetheGAITRite1
matandwere instructedtowalkbeyondthematcompletingafurtheroneandhalfmetres.They
wereinstructedtoconcentrateequallyontheirwalkingandfunctional/cognitive
activityunderdualtaskconditions.Tworesearchassistantswerepresentforall
gaitassessments.Footfalldatawererecordedbythecomputerisedwalkwaywith
manualcorrectionoffootfalllabellingcompletedwherenecessary.Sevengait
variableswerecalculatedusingthemeanoftwotrials;velocity(cm/s),cadence
(steps/min),stride length(cm), doublesupport time(s),stride width(cm),
coefficientofvariationofstridelength(%)andcoefficientofvariationofswing
time(%) The coefficientof variation (CoV) isa measureof variability and
isexpressedasthepercentageoftheratioofthestandard deviationtothe
mean.Dualtask costwascalculated foreachgaitvariableunderbothdual
taskconditions Thesemeasureswerechosenas theyhavepreviously been
reportedinthe literature inrelationtogaitin cognitivelyimpaired people
[3,9,15,17,18,20,21]
2.3.Fallsfollow-up Monthlyfallscalendarsandreplypaidenvelopesweregiventotheparticipants/ carerstoensureaccuratedatacollectionwithregardtofalls.Thefalldefinition articulatedtotheparticipantandcarerwas,‘‘Inthepastmonth,haveyouhadany fallsincludingasliporatripinwhichyoulostyourbalanceandlandedonthefloor
orgroundorlowerlevel?’’[24].Ifaparticipant/carerfailedtoreturnacalendar,a telephonecallwasmadetothecarer/personresponsibletoobtaintheparticipant’s fallsdata.Amultiplefallerwasdefinedassomeonewhofellatleasttwiceduring the12monthfollow-upperiod
2.4.Statisticalanalysis
Data wereanalysed usingSPSS18.0for Windows(SPSS,Inc.,Chicago, IL) Coefficientofvariation(CoV=SD/mean100)wascalculatedforstridelength (stridelengthvariability)andswingtime(swingtimevariability);theCoVforeach trialwascalculatedandthemeanoftwotrialsisreported.Forvariableswith skeweddistributions,datawerelognormalised.Dataondoublesupporttime, swingtime,stridelengthvariabilityandswingtimevariabilitywerenormalised usinglogarithmictransformationsforallconditions.Extremescoreswerecensored
atalevelof3SDfromthemeanandwereincludedintheanalysesi.e.foronecasefor thefollowingvariables:simpledoublesupporttime,functionaldualtaskswing timevariabilityandstridelengthvariability,cognitivedualtaskdoublesupport timeandstridelengthvariability;andfortwocasesforfunctionaldualtaskswing timevariability.Allparametricanalyseswereperformedwiththetransformed data.Univariateanalyseswerecarriedoutusingindependentsamplet-testsand Chisquaredtestsforcross-tabulationtablestocomparenon-multiplefallersand multiple fallers’ demographic and medical characteristics Data for each gait variablewereanalysedusingrepeatedmeasuresmultivariateANOVAwithgait condition(simplevsfunctionalvscognitive)asthewithin-subjectfactorandfaller status(0–1vs2+)asthebetween-subjectfactor.Spatiotemporalgaitmeasuresand gaitvariabilitymeasureswereanalysedseparately.DTCwascalculatedforeach variable for both functional and cognitive dual task conditions: Dual Task Cost=([dualtask simple]/simple)100
Table1
Participantcharacteristics:displayingpvaluesforunivariateanalysis
Characteristic Non-multiplefallers(<2)(n=41) Multiplefallers(2)(n=22) pvalue
Trang 3perform the secondary cognitive task and successfully walk; one
12,14,25–27] and cognitively impaired older adults [10–14,16– 19,28] with the extent of the dual task cost dependent on the
Table2
Meansandstandarddeviationsofspatiotemporalgaitparametersanddualtaskcostfornon-multiplefallersandmultiplefallers:univariateanalysiscomparingmeanscores
ofnon-multiplefallersandmultiplefallers
Gaitparameters Non-multiplefallers(<2)(n=41) Multiplefallers(2)(n=22) pvalue*
Spatiotemporalmeasures (meanSD)
Dualtaskcost(%) Spatiotemporalmeasures
(meanSD)
Dualtaskcost(%)
Velocity(cm/s)
Functionaldualtask 84.9823.28 11.0213.71 68.5628.59 15.6915.28 0.020 Cognitivedualtask 78.6727.15 17.0418.76 61.1628.39 23.6417.16 0.024 Cadence(steps/min)
Functionaldualtask 98.0712.34 3.498.86 93.0416.63 4.318.58 0.191 Cognitivedualtask 88.6917.41 12.2015.05 82.4820.71 15.2613.49 0.230 Stridelength(cm)
Functionaldualtask 103.1022.54 8.168.42 86.4525.47 12.289.89 0.012 Cognitivedualtask 104.5024.63 6.289.94 87.2025.28 9.7612.69 0.014 Doublesupporttime(s)
Functionaldualtask 0.340.11 17.6925.96 0.430.17 22.2225.69 0.021 Cognitivedualtask 0.390.22 32.6852.43 0.550.32 42.9736.75 0.027 Stridewidth(cm)
Functionaldualtask 13.063.28 3.1814.27 14.903.39 5.1816.27 0.048 Cognitivedualtask 13.433.55 5.1614.37 15.443.79 8.7314.77 0.028 Stridelengthvariability(CoV)
Functionaldualtask 3.172.05 38.3189.31 5.764.66 50.38119.96 0.027 Cognitivedualtask 3.742.83 64.61130.01 6.345.03 68.43146.57 0.009 Swingtimevariability(CoV)
Functionaldualtask 6.745.30 29.50131.39 10.657.67 50.4189.55 0.017 Cognitivedualtask 9.427.31 128.97493.20 15.0611.25 141.62341.19 0.019 Dualtaskcost=([dualtask simple]/simple)100,CoV=Coefficientofvariation(CoV=SD/mean100).Ahigherscorerepresentsbetterperformance:Velocity,cadenceand stridelength.Ahigherscorerepresentsworseperformance:Doublesupporttime,stridewidth,stridelengthvariabilityandswingtimevariability.Functionaldualtask:two participantswereunabletoperformastheymobilisedwithawalkingframe.Cognitivedualtask:onepersondeclinedtoattempt,threecountedforwardsastheywereunable
tocountbackwardsandwalkandonewalkedoffthemat,dataforonetrialwereusedforthisparticipant
*pvaluerepresentsunivariateanalysisofspatiotemporalgaitmeasurescomparingnon-multiplefallerstomultiplefallers
Trang 4counterparts [28] , suggesting that the dual task cost is greater in
condition.
Acknowledgements
References
[1]LordSR,SherringtonC,MenzHB,CloseJCT.Fallsinolderpeople:riskfactors andstrategiesforprevention, 2nded.,Cambridge:CambridgeUniversity Press;2007
[2]SheridanPL,HausdorffJM.Theroleofhigher-levelcognitivefunctioningait: executivedysfunctioncontributestofallriskinAlzheimer’sdisease.Dementia andGeriatricCognitiveDisorders2007;24:125–37
[3]Al-YahyaE,DawesH,SmithL,DennisA,HowellsK,CockburnJ.Cognitive motorinterferencewhilewalking:asystematicreviewandmeta-analysis NeuroscienceandBiobehavioralReviews2011;35:715–28
[4]Bootsma-vanderWielA,GusseklooJ,deCraenAJM,vanExelE,BloemBR, WestendorpRGJ.Walkingandtalkingaspredictorsoffallsinthegeneral population: theLeiden85-PlusStudy.JournaloftheAmericanGeriatrics Society2003;51:1466–71
[5]HermanT,MirelmanA,GiladiN,SchweigerA,HausdorffJM.Executivecontrol deficitsasaprodrometofallsinhealthyolderadults:aprospectivestudy linkingthinking,walking,andfalling.JournalofGerontologyMedicalSciences 2010;65:1086–92
[6]NordinE,Moe-NilssenR,RamnemarkA,Lundin-OlssonL.Changesin step-widthduringdual-taskwalkingpredictsfalls.GaitandPosture2010;32:92–7 [7]BeauchetO,AllaliG,AnnweilerC,BerrutG,MaaroufN,HerrmannFR,etal Doeschangeingaitwhilecountingbackwardpredicttheoccurrenceofafirst fallinolderadults?Gerontology2008;54:217–23
[8]BeauchetO,AnnweilerC,AllaliG,BerrutG,HerrmannFR,DubostV.Recurrent fallsanddualtask-relateddecreaseinwalkingspeed:istherearelationship? JournaloftheAmericanGeriatricsSociety2008;56:1265–9
[9]WebsterKE,MeroryJR,WittwerJE.Gaitvariabilityincommunitydwelling adultswithAlzheimerdisease.AlzheimerDiseaseandAssociatedDisorders 2006;20:37–40
[10]vanIerselMB,VerbeekALM,BloemBR,MunnekeM,EsselinkRAJ,OldeRikkert MGM.Frailelderlypatientswithdementiagotoofast.JournalofNeurology NeurosurgeryandPsychiatry2006;77:874–6
[11]MaquetD,LekeuF,WarzeeE,GillainS,WojtasikV,SalmonE,etal.Gait analysisinelderlyadultpatientswithmildcognitiveimpairmentandpatients withmildAlzheimer’sdisease:simpleversusdualtask:apreliminaryreport ClinicalPhysiologyandFunctionalImaging2010;30:51–6
[12]LamothCJ,vanDeudekomFJ,vanCampenJP,AppelsBA,deVriesOJ,Pijnappels
M.Gaitstabilityandvariabilitymeasuresshoweffectsofimpairedcognition anddualtaskinginfrailpeople.JournalofNeuroEngineeringand Rehabilita-tion2011;8:2
[13]CamicioliR,HowiesonD,LehmanS,KayeJ.Talkingwhilewalking:theeffectof
adualtaskinagingandAlzheimer’sdisease.Neurology1997;48:955–8 [14]GillainS,WarzeeE,LekeuF,WojtasikV,MaquetD,CroisierJL,etal.Thevalue
ofinstrumentalgaitanalysisinelderlyhealthy,MCIorAlzheimer’sdisease subjectsandacomparisonwithotherclinicaltestsusedinsingleand dual-taskconditions.Annals ofPhysicalandRehabilitationMedicine2009;52: 453–74
[15]VergheseJ,RobbinsM,HoltzerR,ZimmermanM,WangC,XueX,etal.Gait dysfunctioninmildcognitiveimpairmentsyndromes.JournaloftheAmerican GeriatricsSociety2008;56:1244–51
[16]Sheridan PL,SolomontJ, KowallN,Hausdorff JM.Influence ofexecutive functiononlocomotorfunction:dividedattentionincreasesgaitvariability
inAlzheimer’sdisease.JournaloftheAmericanGeriatricsSociety2003;51:
Trang 5Dual-taskingandgaitinpeoplewithmildcognitiveimpairment.Theeffectof
workingmemory.BMCGeriatrics2009;9:41
[18]SchwenkM,ZieschangT,OsterP,HauerK.Dual-taskperformancescanbe
improvedinpatientswithdementia:arandomizedcontrolledtrial.Neurology
2010;74:1961–8
[19]AllaliG,KressigRW,AssalF,HerrmannFR,DubostV,BeauchetO.Changesin
gaitwhilebackwardcountingindementedolderadultswithfrontallobe
dysfunction.GaitandPosture2007;26:572–6
[20]NakamuraT,MeguroK,SasakiH.Relationshipbetweenfallsandstridelength
variabilityinseniledementiaoftheAlzheimertype.Gerontology1996;42:108–13
[21]CamicioliR,LicisL.MotorimpairmentpredictsfallsinspecializedAlzheimer
careunits.AlzheimerDiseaseandAssociatedDisorders2004;18:214–8
[22]SterkeCS,vanBeeckEF,LoomanCW,KressigRW,vanderCammenTJ.An
electronicwalkwaycanpredictshort-termfallriskinnursinghomeresidents
withdementia.GaitandPosture2012;36:95–101
[23]MioshiE,DawsonK,MitchellJ,ArnoldR,HodgesJR.TheAddenbrooke’sCognitive
ExaminationRevised(ACE-R),abriefcognitivetestbatteryfordementia
screen-ing.InternationalJournalofGeriatricPsychiatry2006;21:1078–85
[24]LambSE,Jorstad-SteinEC,HauerK,BeckerC.Developmentofacommon
outcomedataset forfallinjurypreventiontrials:thepreventionoffalls
networkEuropeconsensus.JournaloftheAmericanGeriatricsSociety2005;53: 1618–22
[25]BeauchetO,DubostV,GonthierR,KressigRW.Dual-task-relatedgaitchanges
intransitionallyfrailolderadults:thetypeofthewalking-associatedcognitive taskmatters.Gerontology2005;51:48–52
[26]DubostV,KressigRW,GonthierR,HerrmannFR,AminianK,NajafiB,etal Relationshipsbetweendual-taskrelatedchangesinstridevelocityandstride timevariabilityinhealthyolderadults.HumanMovementScience2006;25: 372–82
[27]VergheseJ,KuslanskyG,HoltzerR,KatzM,XueX,BuschkeH,etal.Walking whiletalking,effectoftaskprioritizationintheelderly.ArchivesofPhysical MedicineandRehabilitation2007;88:50–3
[28]TheillN,MartinM,SchumacherV,BridenbaughSA,KressigRW Simulta-neouslymeasuringgaitandcognitiveperformanceincognitivelyhealthyand cognitivelyimpairedolderadults,theBaselmotor-cognitiondual-task para-digm.JournaloftheAmericanGeriatricsSociety2011;59:1012–8 [29]BeauchetO,DubostV,AminianK,GonthierR,KressigRW.Dual-task-related gaitchangesintheelderly,doesthetypeofcognitivetaskmatter?Journalof MotorBehavior2005;37:259–64
[30]AllanLM,BallardCG,RowanEN,KennyRA.Incidenceandpredictionoffallsin dementia:aprospectivestudyinolderpeople.PLoSOne2009;4:e5521