Measures of Health and Function that Predict Future Falls...6 Measures of Health and Function that Predict Future Falls 131 6.1 Introduction Falls are one of the biggest problems which

6 Measures of Health and Function that Predict Future Falls

Matthew O’Connell, Rose Anne Kenny and Orna Donoghue

ContentsKey Findings �� 130

6�1 Introduction �� 131

6�2 Prevalence of falls �� 132

6.3 Profileofindividualsreportingrecurrentfalls �� 1356�3�1 Co-morbidities and medications �� 1356�3�2 Physical health and function�� 1366�3�3 Cognitive and mental health and function �� 1376�3�4 Sensory function �� 140

6.4 Identifyingthemostimportantpredictorsofrecurrentfalls �� 144

6.5 Identifyingthemostimportantpredictorsofinjuriousfalls �� 145

6�6 Conclusion �� 146

References �� 148

The Irish Longitudinal Study on Ageing

130


Key Findings

• Fallsareassociatedwithmanynegativeoutcomessuchasinjury,disability,hospitalisationandreducedqualityoflife,thereforetheyareamajorburdenforolderadultsandamajorchallengeforthehealthcareservices.

• Fallsarecommonincommunity-dwellingadultsaged50yearsandoverinIreland,withalmost2in5reportingafallduring4yearsoffollow-upand1in5reportingrecurrentfallsandinjuriousfalls.

• Theprevalenceoffallsishigherinwomenthanmenandincreaseswithageinboth.Twooutoffivewomenaged75yearsandoverreportedrecurrentfallsduring4yearsoffollow-upwhileasimilarnumbersustainedafall-relatedinjuryrequiringmedicalattention�

• Olderadultswhoreportrecurrentfallsorinjuriousfallsdisplaypoorerindicatorsofphysical,cognitiveandmentalhealthandfunctioncomparedtonon-fallers.

• Manyofthemostimportantriskfactors,e.g.unsteadinesswhenwalking,depressivesymptoms,non-cardiovascularconditions,fearoffallingandhavingorthostatichypotension,aremodifiableandcanbetreatedandimproved.

• Fallsassessmentsshouldberoutinelyconductedinolderadultstoidentifyriskfactorsandcausesoffallssothatappropriatemanagementandfallpreventionstrategiescanbeimplemented.


131

6.1 Introduction

Fallsareoneofthebiggestproblemswhichoccurinadvancingyears,becauseoftheirhighfrequencyandadversehealthconsequences.Olderadultsareparticularlyvulnerabletobothfallsandtheassociatednegativeconsequencesoffallsastheydisplayage-relatedphysiologicalchangessuchaspoorerbalanceandslowerreactiontimes.Inadditiontheyaremoresusceptibletoinjuryduetoanincreasedprevalenceofclinicalconditionssuchasosteoporosis,andtheyrecoverlesswellafterinjury,thereforeplacingtheminadeconditionedstateforlonger(1).Fallscanleadtofearoffallingandactivityrestriction,disability,hospitalisationandreducedqualityoflife.AnEconomicBurdenofIllnessStudyestimatedthatfall-relatedinjuriesinolderpeoplecost€402milliontotheIrisheconomyin2006andwouldincreaseto€922-1077millionby2020intheabsenceofaNationalFallandFracturePreventionStrategybeingimplemented(2).Thisisaconservativeestimatewhichdoesnotincludethewidereconomicimpactincurredbycarersandthelongertermconsequencesoutwithinjurysuchassubsequenthomecare,medicationsandinstitutionalisation costs�

Manyaccidentalfallsoccurpartlyduetoenvironmentalfactorse.g.trippingoveranunevenpavement;however,intrinsicfactorsrelatingtotheindividualalsoplayarole.Previousresearchhasidentifiedalargenumberofriskfactorsforfallsincludingpriorhistoryoffalls;chronicconditionssuchasarthritisorurinaryincontinence;deficitsingaitorbalance;reducedmusclestrength;poorvision;depressivesymptoms;dizziness;takingfourormoremedicationsandimpairedcognition(1,3).Mostfallsarenotduetojustoneriskfactor;insteadmanyinteractwitheachotherandtheriskoffallingincreaseswithanincreasingnumberofriskfactors.However,themostimportantrisksappeartobeimpairmentsingait,balanceandmusclestrength(1).

Despitethesepreviousstudies,thereremainsalackofinformationonboththeimpactoffallsandthemostimportantriskfactorsatthepopulationlevelinIreland.Throughitsextensivesurveydatacollectionandstate-of-thearthealthassessment,TILDAcollectssomeofthemostcomprehensiveinformationonfallsriskfactorsavailableinternationally.

Inthischapter,theaimistodescribethefrequencyoffallsincommunity-dwellingolderadultsinIrelandoverthefirstthreewavesofdatacollectionandsubsequently,toestablishaprofileoffallersinrelationtobothself-reportedandobjectivemeasuresofhealthandfunction obtained at baseline (Wave 1)� These measures can be roughly divided into chronicconditionsandmedications,physicalhealthandfunction,sensoryfunction,cognitiveandmentalhealth.Wethenidentifythefactorsthatarethemostimportant


132

predictorsoffuturefallsandexamineiftheseriskfactorsaremodifiableinanefforttobestinformpolicymakersandclinicalpracticeinIreland.

Inthisanalysis,informationisobtainedfromboththeinterviewandhealthcentreassessmentcomponentsofTILDA,thereforeonlythosewhoparticipatedinbothcomponentsatbaselinewereeligibleforthisanalysis(n=5,035).Oftheseparticipants,4,700participatedinWave2(including12proxyinterviews)and4,400participatedinWave3(including32proxyinterviews).Proxyinterviewswerecompletedbyafamilymember/closefriendiftheparticipantwasunabletoparticipatethemselvesduetophysicalorcognitivereasons.Participantsweredividedintothreeagegroups(50-64years,65-74yearsand75yearsorover)whichreflecttheiragesatWave1.

6.2 Prevalence of falls

Duringeachinterview,participantswereaskediftheyhadfalleninthepastyearand/orsincetheirlastinterviewandiftheyhadrequiredmedicalattentionasaresultofanyofthesefalls.AtWave1,thisinformationreflectsanindividual’shistoryoffallsinthepastyear.FallsreportedatWaves2and3wereusedtodefinethevariousfallsoutcomes(anyfalls,recurrentfalls,injuriousfalls)occurringafterthisbaselinemeasurement(Table6.1).

Table 6.1: Definition of falls outcomes

Outcome Definition

Any Fall AnyfallreportedinthelastyearORsincetheparticipant’slastinterview�

Recurrent Falls TwoormorefallsreportedatWave2orWave3orbothORasinglefallreportedatbothWave2andWave3.

Injurious Falls Anyfallinwhichaninjuryseriousenoughtorequiremedicalattention was sustained at Wave 2 or Wave 3�

Table6.2showstheprevalenceoffalls,recurrentfallsandinjuriousfallsatWaves1,2and3�

Table 6.2: Prevalence of recurrent falls and injurious falls at Waves 1, 2 and 3

Wave 1 Wave 2 Wave 3 Wave 2 and 3

N 4,788 4,696 4,398 4,789

Any fall (%) 21�1 23.9 25�8 37�3

Recurrent falls (%) 7�7 9.9 11�1 19.1

Injurious falls (%) 7�7 10�2 12�1 18�4


133

AtWave1,21%ofparticipantsreportedanyfallintheyearpriortointerviewwhilethisfigureincreasedto24%atWave2and26%atWave3.WhenconsideringfallswhichoccurredatWaves2and3,overonethird(37%)ofparticipantsexperiencedafallduringthisfollow-upperiod.AtWave1,8%ofparticipantsreportedrecurrentfallsinthepastyearandthisincreasedto10-11%atWaves2and3.Overthefouryearsoffollow-up(i.e.Waves2and3),19%ofparticipantsexperiencedrecurrentfalls.Theprevalenceofinjuriousfallswasverysimilartorecurrentfallsatallwaves.Fortheremainderofthischapter,wepresentthecombinedfigureforfallsoccurringduringfollow-upi.e.basedonfalls occurring at Wave 2 and Wave 3�

Womenweremorelikelytoreportanyfallsduring4yearsfollow-upcomparedtomen(42%versus32%).Theprevalenceincreasedwithageinbothgendersfrom29%and34%(inmenandwomenaged50-64years)to42%and60%(inmenandwomenaged75yearsandolder)(Figure6.1).ApproximatelyoneinfiveolderadultsinIreland(17%ofmen;21%ofwomen)experiencedrecurrentfallsinthelastfouryears.Recurrentfallsweremorecommoninwomenaged65-74yearsand75yearsandoldercomparedtowomenaged50-64years(Figure6.2),howeverthisrelationshipwithagewasweakerinmen.Injuriousfallswerealsomorecommoninwomen(23%)comparedtomen(13%)butincreasedwithageinbothgenders(Figure6.3).Forexample,18%ofwomenaged50-64yearsreportedaninjuriousfallcomparedto38%ofthoseaged75yearsandover.

Figure 6.1: Prevalence of any fall at Wave 2 and Wave 3, by age group and sex


134

Figure 6.2: Prevalence of recurrent falls at Wave 2 and Wave 3, by age group and sex

Figure 6.3: Prevalence of injurious falls at Wave 2 and Wave 3, by age group and sex


135

Asshownabove,fallscanbedefinedinanumberofways.However,fallsareoftensubjecttorecallbiasleadingtoinaccuraciesinreporting.Inaddition,asinglefallcanoftenresultfromanisolatedeventratherthananumberofunderlyingcauses.Consequently,wehavechosentofocusonthemoreseriousfallsoutcomesofrecurrentfallsandinjuriousfallsfortheremainderofthischapter.

6.3 Profileofindividualsreportingrecurrentfalls

Self-reportedmeasuresweretypicallyobtainedduringtheinterviewwhileobjectivemeasureswereobtainedduringthehealthassessmentwhichtookplaceinhealthcentresinDublinandCork.Thishealthassessmentincludedtestsofcardiovascularfunction,cognitivefunction,mobility,vision,bonehealthandanthropometry.

6.3.1 Co-morbidities and medications

Participantswereaskedtoindicateifadoctorhadevertoldthemthattheyhadanyofthefollowingcardiovascularconditions(highbloodpressure,angina,heartattack,heartfailure,diabetes,stroke,mini-stroke,highcholesterol,heartmurmurandheartrhythm)andnon-cardiovascularconditions(chroniclungdisease,asthma,arthritis,osteoporosis,cancer,Parkinson’sdisease).Thenumberofconditionsineachcategorywasobtained.Participantsalsolistedallmedicationstakenregularlywithpolypharmacydefinedasfiveormore medications�

Havingahighburdenofcardiovascularornon-cardiovascularconditionswasstronglyassociatedwithrecurrentfalling.Approximatelytwiceasmanyrecurrentfallershad3ormorecardiovascularconditionsor2ormorenon-cardiovascularconditions,comparedtonon-recurrentfallers(Table6.3).Similarly,polypharmacywasalmosttwiceascommonamongrecurrentfallers.Thisdifferencewasmostpronouncedinwomenaged50-64years(10%innon-recurrentfallersversus25%inrecurrentfallers).

Table 6.3: Co-morbidities and medications in non-recurrent fallers and recurrent fallers

Co-morbidities and medicationsNon-recurrent fallers Recurrent fallers

n% (95% CI) % (95% CI)

≥3cardiovascularconditions 9.6 (8.5-10.9) 18�3 (15.3-21.9) 4,788

≥2non-cardiovascularconditions 9.9 (8�8-11�1) 18.9 (15�7-22�5) 4,788

Polypharmacy 18�2 (16.7-19.9) 32�3 (28�2-36�8) 4,766

Note:CI=confidenceinterval


136

6.3.2 Physical health and function

TheTILDAhealthassessmentincludesanumberofhighqualityandnovelassessmentsofphysicalhealthandfunction.Walkingspeedwasmeasuredusingacomputerisedmat(activearea4.88m)withembeddedpressuresensors(GAITRite®,CIRSystemsInc,NewYork,USA).Participantsstartedandstopped2.5mbeforeand2mafterthemat;averagewalkingspeedwasobtainedacrosstwotrials.GripstrengthisaproxyforoverallbodystrengthandwasmeasuredusingahandheldBaselinedynamometer.Theaverageoftwomeasurements was obtained for the dominant hand�

OneofthemostinnovativetestsincludedinTILDAiscontinuousbloodpressuremonitoringduringactivestanding.AFinometerdevicewasusedtomeasurechangesinbloodpressurewhilestandingupfromalyingposition.OrthostaticHypotension(OH)isdefinedaslowbloodpressurethroughoutthesecondminuteofstanding(4).Participantsalsounderwentaquantitativeheelboneultrasound(AchillesHeelUltrasound,Lunar,Madison,USA)toassessbonehealthandtheriskofosteoporoticfracture.Thisdeviceprovidesanindexofbonestiffnessandwhilethisisnotdiagnostic,anindividualisconsideredtohaveosteoporosis,osteopeniaornormalbonedensityifthestiffnessindexis<65%,65-86%,or>86%respectively(5).

HeightandweightweremeasuredusingstandardequipmentandusedtocalculateBodyMassIndex(BMI).PhysicalactivitywasmeasuredusingtheshortformInternationalPhysicalActivityQuestionnaire(IPAQ),inwhichthenumberofdaysandhoursspentwalkinganddoingvigorousandmoderateintensityactivitiesinthelastweekwasrecorded(6).SpecificcriteriawereusedtoclassifyparticipantsintoHigh,ModerateorLowActivitylevels (7)�

Recurrentfallersintheolderadultpopulationhadslowerwalkingspeedandlowergripstrengthcomparedtonon-recurrentfallers(Table6.4).Orthostatichypotensionwasalmosttwiceascommonamongrecurrentfallers,whilelowphysicalactivitywasslightlymorecommon.Bonehealthwassimilarinrecurrentfallersandnon-recurrentfallers.Levelsofobesitywerealsosimilarexceptwhenlookingatadultsaged50-64yearswhererecurrentfallersweremorelikelytobeobese(39%)comparedtonon-recurrentfallers(33%).Unsurprisingly,ahistoryoffallingisstronglyrelatedtofuturefallsas40%ofrecurrentfallersatWaves2and3reportedafallintheyearpriortotheirWave1interviewcomparedto17%ofnon-recurrentfallers.


137

Table 6.4: Physical health and function measures in non-recurrent fallers and recurrent fallers

Physical health measures Non-recurrent fallers Recurrent fallers n

Mean(95%CI)

Walkingspeed(cm/s) 134�4 (133�5-135�3) 123�8 (121�3-126�2) 4,714

Gripstrength(kg) 26�5 (26�1-26�8) 23�7 (22�8-24�5) 4,732

%(95%CI)

OrthostaticHypotension 5�5 (4�7-6�5) 9.6 (7�0-13�2) 4,255

Bone health

Normal 50�6 (48�7-52�4) 46�1 (42�0-50�2)

Osteopenia 40�3 (38�6-42�2) 41.9 (37�6-46�3) 4,760

Osteoporosis 9.1 (7.9-10.4) 12�0 (9.3-15.5)

Body Mass Index

Underweight 0�5 (0�3-0�8) 0�5 (0�2-1�3)

Normal 22�0 (20�5-23�6) 25�4 (21.8-29.4) 4,780

Overweight 44�5 (42�7-46�3) 38�6 (34.4-42.9)

Obese 33�1 (31�4-34�8) 35�5 (31.7-39.4)

Physical activity level

Low 28�4 (26�6-30�4) 34�7 (30.7-39.0)

Moderate 35�8 (34�0-37�7) 36�4 (32�6-40�3) 4,748

High 35�7 (33�6-37�5) 28.9 (25�3-32�8)

Fallin12monthspriortoWave1 16�5 (15�1-18�0) 40�3 (36�2-44�6) 4,786

OsteopeniaandosteoporosiswerebasedontheStiffnessIndexobtainedfromthequantitativeheelultrasound.


6.3.3 Cognitive and mental health and function

GlobalcognitionwasassessedusingtheMontrealCognitiveAssessment(MoCA)whichassessesabilityacrossmultipledomainsofcognitivefunction:memoryrecall,visuospatialability,executivefunction,attention,language,andorientationtotime/place.Executivefunctionallowsanindividualtoplan,organizeandcompleteatask.Itisoftenassociatedwithbalanceandwalking-basedtasksandthereforepoorexecutivefunctionisariskfactorforfalls.ItwasassessedusingtheColourTrailsTestwhichhastwocomponents,bothof which involve connecting a series of numbered and coloured circles� The difference betweenthetwocomponents(ColourTrailstimedifference)reflectsexecutivefunction(8).


138

Concentrationandprocessingspeedwereassessedusingacomputer-basedchoicereactiontimetest.Participantspressedabuttononthekeyboarduntilanon-screenstimulusappeared(YES/NO);thentheyreleasedthebuttonandpressedthecorrespondingYES/NObuttononthekeyboard.Totalresponsetimeisthetimetakenfromwhenthestimulusappeared,fortheparticipanttoreleasethebuttonandpressthetargetbuttononthekeyboard.

RecurrentfallersintheolderadultpopulationhadlowerscoresonMoCA(lowerglobalcognitivefunction)andalargerColourTrailstimedifference(poorerexecutivefunction)(Table6.5).Therewerenogenderdifferencesinglobalcognition,executivefunctionorreactiontime,howevertherewasaclearassociationbetweenincreasingageandpoorerfunction in all three cognitive tests in both recurrent fallers and non-recurrent fallers� The relationshipbetweenageandColourTrailtimedifferenceisshowninFigure6.4.

Table 6.5: Cognitive function in non-recurrent fallers and recurrent fallers

Cognitive functionNon-recurrent fallers Recurrent fallers

n% (95% CI) % (95% CI)

MoCA (max score 30) 24�7 (24.6-24.9) 24�1 (23�7-24�4) 4,772

Colour trail time difference (s) 56�3 (55�1-57�6) 61�3 (58�2-64�4) 4,723

Choice reaction time (ms) 844�2 (829.0-859.5) 891.6 (859.0-924.1) 4,694



139

Figure 6.4: Executive function (indicated by colour trails time difference) in non-recurrent fallers and recurrent fallers, by age group

Depressivesymptomswereassessedusingthe20-itemCentreforEpidemiologicalStudiesDepression(CES-D)scale,wherethemaximumscoreis60(9).Participantsscoring≥16wereclassifiedashavingmoderateorpotentiallyclinicallyrelevantdepressivesymptomswhilethosescoring8-15wereclassifiedashavingmildorsub-thresholddepressivesymptoms.Fearoffallingwasidentifiedbyasking“Areyouafraidoffalling?”andifso,towhatextent(somewhatafraidorverymuchafraid).

Participantswhoreportedrecurrentfallsatfollow-upweremorelikelytoreportpotentiallyclinicallyrelevantdepressivesymptomsandtobesomewhatorverymuchafraidoffallingcomparedtonon-recurrentfallers(Table6.6).Depressivesymptomsdidnotvarybyagegrouporsexinrecurrentfallers,howeverbeingsomewhatafraidoffallingwasreportedmorefrequentlyinwomen(36%versus18%inmen)andintheolderagecategories(41%infallersaged≥75yearsversus20%infallersaged50-64years).


140

Table 6.6: Depressive symptoms and fear of falling in non-recurrent fallers and recurrent fallers

Mental healthNon-recurrent fallers Recurrent fallers

n% (95% CI) % (95% CI)

Depressivesymptoms(%)

None 73�4 (71�6-75�2) 62�8 (58�5-66�8)

Mild (sub-threshold) 16.9 (15�5-18�4) 19.3 (16�0-23�1) 4,788

Moderate(potentiallyclinically relevant)

9.7 (8�5-11�0) 17.9 (14�8-21�5)

Fear of falling (%)

None 81�5 (79.6-83.3) 63�2 (58�5-67�7)

Somewhat afraid 14�8 (13�3-16�6) 28�4 (24�0-33�2) 4,787

Very much afraid 3�6 (2�8-4�8) 8�3 (6�1-11�3)


6.3.4 Sensory function

Duringtheinterview,participantsratedtheirsteadinesswhenwalking(10)withunsteadinessindicatedifparticipantsreportedbeingSlightlysteady,SlightlyunsteadyorVeryunsteady(asopposedtoVerysteady).Participantsalsoself-ratedtheirvisionandhearingwithanswersdichotomisedintotwocategories:Excellent/VeryGood/GoodandFair/Poor.OneparticipantwhoreportedbeinglegallyblindwasincludedintheFair/Poorvision category�

Two measures of visual function were assessed during the health assessment� Visual acuity,whichistheacutenessorclearnessofvision,wasassessedusingaLogMAR(MinimalAngleofResolution)chartwhichisoftenusedbyoptometristsandGPs.Participantsstood4metresawayfromthechartandcompletedthetestwitheacheye,usingcorrectiveglasses/lensesifrequired.ThelogMARscoreforthebesteyewasconvertedintoaVisualAnalogScore(VAS)wherehigherscoresrepresentbettervisualacuity(11).Contrastsensitivityistheabilitytodistinguishanobjectfromthebackgroundinlowcontrastconditionse.g.atdusk.ItwasmeasuredusingaFunctionalVisionAnalyser(StereoOptical,Chicago,IL,USA)undermesopic(3cd/m2)backgroundilluminationconditionsinnon-glareandglareconditions.Thetestprovides10scoresobtainedatdifferentspatialfrequencies(1.5,3,6,12and18cyclesperdegree,cpd).Sharpimageswithsmalldetailscontainhighspatialfrequencyinformationwhilecoarserimageswith


141

blurreddetailscontainlowspatialfrequencyinformation.Wesubsequentlyusedfactoranalysistoidentifytwodistinctfeaturesforeachindividualcorrespondingtotheir“highfrequency”and“lowfrequency”contrastsensitivity(12).Eachfactorscorewasdividedintoquartileswiththelowestquartilerepresentingpoorercontrastsensitivity.

Overall,recurrentfallersintheolderadultpopulationinIrelandweremorelikelytoreportunsteadinesswhenwalkingcomparedtonon-recurrentfallersalthoughtherewasnodifference in self-rated vision or hearing (Table 6�7)�

Table 6.7: Sensory function in non-recurrent fallers and recurrent fallers

Sensory functionNon-recurrent fallers Recurrent fallers

n % (95% CI) % (95% CI)

Self-rated unsteadiness 14�7 (13�1-16�4) 37�4 (33�0-42�1) 4,785

Self-ratedvision(poor/fair) 8�8 (7�7-10�1) 12�4 (9.8-15.7) 4,788

Self-ratedhearing(poor/fair) 14�5 (13�2-16�0) 19.0 (15�8-22�8) 4,788

Visual acuity score 96.0 (95.7-96.4) 95.1 (94.2-95.9) 4,771

Contrast sensitivity (no glare)

1.5cpd 36�1 (35.4-36.9) 34�2 (32�4-36�0)

3cpd 66�6 (65�4-67�8) 62�4 (59.7-65.0)

6cpd 31�7 (30�8-32�7) 27�6 (25.7-29.5) 4,561

12cpd 6�5 (6�1-6�8) 5�3 (4�4-6�3)

18cpd 0�7 (0�7-0�8) 0�6 (0�4-0�7)

Contrast sensitivity (glare)

1.5cpd 37.9 (37�2-38�6) 33�8 (32�3-35�2)

3cpd 68.9 (67�6-70�2) 62�6 (59.8-65.3)

6cpd 32�1 (31�1-33�1) 27�8 (25.9-29.6) 4,553

12cpd 6.9 (6�5-7�3) 5�1 (4�4-5�7)

18cpd 0�8 (0.7-0.9) 0�5 (0�4-0�6)



142

Oldermenandwomenwhowererecurrentfallerswereequallyaslikelytoreportunsteadinesswhilewalking(36%versus39%)andfair/poorvision(14%versus11%),howeverwomenwerelesslikelytoreportfair/poorhearingthanmen(14%versus26%).Ingeneral,unsteadinessbecamemoreprevalentwithincreasingageespeciallyinrecurrentfallers(Figure6.5),althoughtherelationshipwasweakerforfair/poorvisionandhearing.

Figure 6.5: Self-reported unsteadiness while walking in non-recurrent fallers and recurrent fallers, by age group


143

AtmostspatialfrequenciesofvisualfunctionassessedinTILDA(i.e.acrossthespectrumrepresentingsharpdetailedimagestocoarserimages),contrastsensitivitywaslowerinrecurrentfallerscomparedtonon-recurrentfallersbuttherewasnodifferenceinvisualacuity(Table6.7).However,therewereclearageandsexdifferenceswithlowervisualacuity and contrast sensitivity observed with increasing age and in women (Figure 6�6)�

Figure 6.6: Visual acuity in non-recurrent fallers and recurrent fallers, by age group and sex


144

6.4 Identifyingthemostimportantpredictorsofrecurrentfalls

AllofthevariablesusedabovetoprofilefallerswereincludedinamultivariatemodeltoidentifythemostimportantpredictorsofrecurrentfallsatWave2and/orWave3.Thesewereidentifiedasprevioushistoryoffalling,fearoffalling,self-reportedunsteadinesswhilewalking,andnon-cardiovascularchronicconditions(Table6.8).TheRelativeRisks(RR)presentedinthetablecanbeinterpretedasthechangeinriskassociatedwithhavingtheriskfactormentioned.Forexample,participantswhoreportedunsteadinesswhilewalkinghada1.5times,or50%,increasedriskofreportingrecurrentfallscomparedtothosewhodidnotreportunsteadinesswhilewalking.Itshouldbenotedthatthisrelativelysimplemodellingapproachdoesnottakeaccountofpotentiallyimportantmodifyingeffectsofdifferentvariablesinthemodel,norofpotentialcausalpathwaysbetweenvariables.FutureanalysesfromTILDAwillusemoresophisticatedstatisticalmethodstobetterunderstandhowtheseriskfactorsmayvarydependingontheirinteractionswithcertaincharacteristicse.g.sex,agegroup,depressivesymptoms.

Table 6.8: Indicators of health and function which predict recurrent falls at Wave 2 and/or Wave 3

RR (95% CI)

Self-reportedunsteadinesswhenwalking 1�50 (1�22-1�85)

1 non-cardiovascular condition 1�47 (1.21-1.79)

≥2non-cardiovascularconditions 1�55 (1.21-1.97)

Somewhat afraid of falling 1�31 (1.07-1.59)

Very afraid of falling 1�50 (1.13-1.99)

Previous history of falls 2�03 (1�72-2�38)

Fullmodelincludesage,sex,BMI,previoushistoryoffalls,numberofcardiovascularconditions,numberofnon-cardiovascularconditions,orthostatichypotension,osteopenia/osteoporosis,gripstrength,usualwalkingspeed,self-ratedunsteadinesswhenwalking,self-ratedhearing,self-ratedvision,visualacuity,contrastsensitivity,MontrealCognitiveAssessment,colourtrailstimedifference,choicereactiontime,depressivesymptoms,fearoffallingandpolypharmacy(n=3,858).


145

6.5 Identifyingthemostimportantpredictorsofinjuriousfalls

Theprofileofinjuriousfallerswasverysimilartorecurrentfallers.Theyalsodisplayedpoorerindicatorsofhealthandfunctionacrossmultipledomainsincludingphysical,cognitiveandmentalhealth.Wetheninvestigatedthepredictorsofinjuriousfalls,againusingamultivariateanalysis.Themostimportantvariablesthatpredictinjuriousfallsatfollow-upareage,beingfemale,aprevioushistoryoffalls,havingorthostatichypotension,clinicallyrelevantdepressivesymptoms,osteopeniaandself-reportedunsteadinesswhenwalking(Table6.9).

Table 6.9: Indicators of health and function which predict injurious falls at Wave 2 and/or Wave 3

RR (95% CI)

Orthostatichypotension 1�56 (1�22-2�00)

Self-reportedunsteadinesswhenwalking 1�32 (1�06-1�64)

Clinicallyrelevantdepressivesymptoms 1�35 (1�05-1�74)

Osteopenia 1�26 (1�05-1�50)

Age(per1yearincrease) 1�02 (1�01-1�03)

Female sex 1�54 (1�27-1�87)

Previous history of falls 1�82 (1�54-2�15)

Fullmodelincludesage,sex,BMI,previoushistoryoffalls,numberofcardiovascularconditions,numberofnon-cardiovascularconditions,orthostatichypotension,osteopenia/osteoporosis,gripstrength,usualwalkingspeed,self-ratedunsteadinesswhenwalking,self-ratedhearing,self-ratedvision,visualacuity,contrastsensitivity,MontrealCognitiveAssessment,colourtrailstimedifference,choicereactiontime,depressivesymptoms,fearoffallingandpolypharmacy(n=3,858).


146

6.6 Conclusion

Theburdenoffallsishighinadultsaged50yearsandoverinIreland,withalmost40%reportingatleastonefallduring4yearsfollow-up.Importantly,approximatelyhalfoffallersornearly20%ofthepopulationreportedrecurrentfalls,awellknownriskfactorforsubsequentinjuryandhospitalisation.Asimilarproportionexperiencedafallthatresultedinaninjuryseriousenoughtorequiremedicaltreatment.Ingeneral,olderadultswhoreportrecurrentfallsorinjuriousfallsdisplaypoorerindicatorsofphysical,cognitiveandmentalhealthandfunctioncomparedtonon-fallers.Themostimportantfactorsthatpredictrecurrentfallsareahistoryofpreviousfalls,self-reportedunsteadinesswhenwalking,anincreasingnumberofnon-cardiovascularconditions,andbeingsomewhatafraidoffalling.Injuriousfallerssharemanyoftheseriskfactors(historyofpreviousfalls,self-reportedunsteadinesswhenwalking),howeverbeingolder,female,havingpotentiallyclinicallyrelevantdepressivesymptoms,reducedbonestrength(osteopenia)andhavingorthostatichypotensionarealsoimportantpredictorsinthisgroup.

Theimportanceofaprevioushistoryoffallshighlightstheneedforearlymonitoringandinterventiontoaddressanyexistingriskfactors.Mostfallsoccurwhilewalking(13),thereforeunsteadinesswhilewalkingpresentsanopportunitytoidentifyanunderlyingdifficultyorperceptionthatsomethingisnotquiteright,evenifthisisnotvisuallyobvious.Depressivesymptomsarecommoninolderadultsandtheyhaveacomplexrelationshipwithfalls.Bothsharecommonriskfactors;depressivesymptomsandconsequenttreatmentwithantidepressantscanleadtoafall,whilefallscanalsoresultinincreaseddepressivesymptoms(14).Orthostatichypotensionisalsocommoninolderadultsandmaycausefallsandfaintsthroughdecreasesinbrainbloodflowwhenstandingup.

Fallshaveveryseriousconsequences,therefore,itisimportanttoidentifysomeonewithahighriskoffallsandsubsequentlyimplementanappropriatefallpreventionprogrammetoreducetheirriskfactorswherepossible.ThecurrentAmericanGeriatricSocietyandBritishGeriatricSocietyGuidelinesforFallPrevention(15)recommendthatalladultsaged65yearsandovershouldbeaskedaboutfallsinthepast12monthsanddifficultieswithgaitorbalance,atleastonceperyear.Thosewhohavefallenpreviouslyorwhoreportdifficultieswithgaitorbalanceshouldthenundergoacomprehensivefallsassessmenttoidentifytheneedforaninterventiontoreducetheirfallrisk.Ideally,thiswouldbeimplementedattheprimarycareleveltopromoteaproactiveapproachthatfocusesonearlydetectionofriskratherthanareactiveapproachafterafallhasoccurred.Luckily,manyoftheriskfactorshighlightedabovearemodifiableandcanatleastbeimprovedifnot removed altogether�


147

Itisimportanttohighlightthatfallsandphysiologicaldeclinearenotaninevitableconsequenceofageing.Themosteffectivefallpreventionstrategiesaremultifactorialandmayincludeinterventionstoimprovemusclestrength,gaitandbalance;reducemedications(wherepossible);manageposturalhypotension;managefootproblemsandfootwear;adaptormodifythehomeenvironmentandsupplementwithVitaminD(AmericanGeriatricSociety/BritishGeriatricSocietyGuidelines,2011).Physicalexerciseisoneofthemosteffectivemeansofimprovingwalking,balanceandstrengthwhilealsoprovidingarangeofotherphysical,cognitive,psychologicalandsocialbenefits(16,17,18).Assumingthatitisadaptedtoeachindividual’sneeds,itshouldberecommendedasstandardforallindividuals,youngandold.However,Rosehighlightedthatthemosteffectivetype,intensityanddoseofexerciseisnotyetclearandthereforeislikelytovarywithdifferentlevelsoffallrisk(19).

Giventhechangingdemographics,fallsrepresentamajorchallengeforthehealthcareservices.In2008,theHealthServiceExecutive,DepartmentofHealthandChildrenandNationalCouncilonAgeingandOlderPeoplepublishedaStrategytoPreventFallsandFracturesinIreland’sAgeingPopulation.Thisreliesonidentificationofriskfactors,managementofspecificconditionsandappropriateinterventiontohelppeopletoreducetheirriskoffalls.TheNationalFallsandBoneHealthProjectAFFINITY(ActivatingFallsandFracturePreventioninIrelandTogether)wassetupin2013toimplementthisstrategy(20).Howevertodate,fallriskassessmentisnotroutinelycarriedout,thereforemanyolderadultswhowouldbenefitfrominterventionand/ormodificationofriskfactorsarenotbeingflaggedforfollow-up.

Morerecently,theNationalClinicalProgrammeforOlderPeoplewhichisajointinitiativebetweentheHSEClinicalStrategyandProgrammesDivisionandtheRoyalCollegeofPhysiciansofIreland,havehighlightedtheroleofthecomprehensivegeriatricassessment.Thisreviewsanolderperson’smedicalconditions,mentalhealth,functionalcapacityandsocialcircumstanceswiththeintentionofdevelopingandimplementinganintegratedassessment,interventionandreviewofanolderperson’sindividualneeds(21).Thisencompassesmanyofthesameteststhatareusedinfallsassessmentandtherefore,thereisalotofscopetoidentifyfallriskusingthisapproach.

Insummary,fallsandfall-relatedinjuriesarecommonamongolderadultsinIrelandandtherearemultiplehealth-relatedfactorsthatmayincreaseanindividual’sriskoffalls.Excitingly,manyofthefactorsidentifiedarepotentiallymodifiable,pavingthewayforfutureinterventionsandfallspreventioninitiatives.


148

References

1� RubensteinLZ.Fallsinolderpeople:epidemiology,riskfactorsandstrategiesforprevention.AgeAgeing.2006;35Suppl2:ii37-ii41.

2� HealthServiceExecutive,NationalCouncilonAgeingandOlderPeople,DepartmentofHealthandChildren.StrategytoPreventFallsandFracturesinIreland’sAgeingPopulation.ReportoftheNationalSteeringGrouponthePreventionofFallsinOlderPeopleandthePreventionandManagementofOsteoporosisthroughoutLife,June2008�

3� TinettiME.Preventingfallsinelderlypersons.NEnglJMed.2003;348(1):42-49.

4� FinucaneC,O’ConnellMD,FanCW,SavvaGM,SoraghanCJ,NolanH,CroninH,KennyRA.AgerelatedNormativechangesinphasicorthostaticbloodpressureinalargepopulationstudy:FindingsfromtheIrishLongitudinalStudyonAgeing(TILDA).Circulation.2014;130:1780-1789

5� VarennaM,SinigagliaL,AdamiS,GianniniS,IsaiaG,MaggiS,etal.Associationofquantitativeheelultrasoundwithhistoryofosteoporoticfracturesinelderlymen:theESOPOstudy.OsteoporoInt.2005;16(12):1749-1754.

6� CraigCL,MarshallAL,SjostromM,BaumanAE,BoothML,AinsworthBE,etal.Internationalphysicalactivityquestionnaire:12-countryreliabilityandvalidity.MedSciinSportsExerc.2003;35(8):1381-95.

7� IPAQ.GuidelinesforthedataprocessingandanalysisoftheInternationalPhysicalActivityQuestionnaire2005.

8� D’EliaLF,SatzP,UchiyamaCL,WhiteT.ColorTrailsTest.Professionalmanual.Odessa,FL:PsychologicalAssessmentResources,1996.

9. RadloffLS.TheCES-Dscale:Aself-reportdepressionscaleforresearchinthegeneralpopulation.AppliedPsychologicalMeasurement.1977;1:385-401.

10� ClarkDO,CallahanCM,CounsellSR.,Reliabilityandvalidityofasteadinessscore.AmGeriatrSoc.2005;53(9):1582-6.


149

11� ColenbranderA.Visualstandards,aspectsandrangeofvisionlosswithemphasisonpopulationsurveys.ReportpreparedfortheInternationalCouncilofOphthalmologyatthe29thInternationalCongressofOphthalmology.Sydney,2002.

12� DonoghueOA,RyanH,DugganE,FinucaneC,SavvaGM,CroninH,LoughmanJ,KennyRA.Relationshipbetweenfearoffallingandmobilityvarieswithvisualfunctionamongolderadults.GeriatrandGerontolInt.2014;14(4):827-36

13� MackenzieL,ByesJ,HigginbothamN.Aprospectivecommunity-basedstudyoffallsamongolderpeopleinAustralia:Frequency,circumstances,andconsequences.OTJR:Occupation,Participation,andHealth.2002;22(4):143-152.

14� IaboniA,FlintAJ.Thecomplexinterplayofdepressionandfallsinolderadults:aclinicalreview.AmJGeriatrPsychiatry2013;21:484-492.

15� SummaryoftheUpdatedAmericanGeriatricsSociety/BritishGeriatricsSocietyclinicalpracticeguidelineforpreventionoffallsinolderpersons.AmGeriatrSoc.2011;59(1),148-157�

16� ColcombeS,KramerAF.Fitnesseffectsonthecognitivefunctionofolderadults:ameta-analyticstudy.PsycholSci.2003;14:125-130.

17� DonoghueO,O’ConnellM,KennyRA.WalkingtoWellbeing:PhysicalActivity,SocialParticipationandPsychologicalHealthinIrishadultsaged50yearsandOlder.Dublin:TheIrishLongitudinalStudyonAgeing,2016.

18� WorldHealthOrganisation.GlobalRecommendationsonPhysicalActivityforHealth.2010�

19.RoseDJ.Preventingfallsamongolderadults:No“onesizesuitsall”interventionstrategy.RehabilResDev.2008;45(8):1153-1166.

20� NationalFallsandBoneHealthProjectAFFINITY(ActivatingFallsandFracturePreventioninIrelandTogether).Availableat:http://www.affinityfallsbonehealth.ie/index.html�

21� NationalClinicalProgrammeforOlderPeople(2016).ComprehensiveGeriatricAssessmentAvailablefrom:https://www.hse.ie/eng/about/Who/clinical/natclinprog/olderpeopleprogramme/resources/ComprehensiveGeriatricAssessmentSummary.pdf


150

Documents

Measures of Health and Function that Predict Future Falls...6 Measures of Health and Function that Predict Future Falls 131 6.1 Introduction Falls are one of the biggest problems which