Reliability properties of the UN-moving-task, an … · 5AP-HP, Hôpital Raymond Poincaré, Service...
1
Reliability properties of the UN-moving-task, an ecological assessment of Unilateral Neglect in patients with acquired brain injury: Preliminary results 1 Sorbonne Universités, UPMC Univ Paris 06, Handicap cognitif et réadaptation (HanCRe), Paris, France; 2 Centre for Interdisciplinary Research in Rehabilitation of Greater Montreal (CRIR), Montréal QC, Canada; 3 Laboratory for Adult Development and Cognitive Aging, Department of Psychology, Concordia University, Montréal QC, Canada; 4 Institut Nazareth et Louis-Braille of the CISSS de la Montérégie-Centre, Longueuil QC, Canada; 5 AP-HP, Hôpital Raymond Poincaré, Service de MPR, Garches, France; 6 HANDIReSP, Université de Versailles Saint-Quentin, France; 7 Université de Montréal, Montréal QC, Canada; 8 Institut universitaire sur la réadaptation en déficience physique de Montréal (IURDPM) of the CIUSSS du Centre-Sud-de-l’Île-de-Montréal, Montréal QC, Canada. Frédérique Poncet 1-4 , Noémie Duclos 2,7,8 , Walter Cybis 2,4 , Philippe Azouvi 5,6 , Cyril Duclos 2,7,8 & Marie-Chantal Wanet-Defalque 2,4,7 Unilateral neglect (UN) is a frequent problem in individuals with brain injury. UN has been described as a peripersonal or extrapersonal disorder. Although many tasks assess UN in peripersonal space, neglect in extrapersonal or far space is rarely assessed. The UN-moving-task, an ecological test of UN in a real corridor and involving multi-tasking (see Figure 1), was developed and exhibits excellent discriminant validity (Poncet et al. 2012). However, other important psychometric properties of this test have yet to be studied. To further evaluate the psychometric properties of this test, this study explored reliability of the UN-moving-task. Figure 1. UN-Moving-Task INTRODUCTION OBJECTIVE RESULTS CONCLUSION Correspondence: [email protected] Preliminary results were calculated from a cohort of 29 participants recruited from the IURDPM (Gingras-Lindsay-de-Montréal), including 18 participants with stroke (12 participants with right stroke (9 men) and 6 with homonymous hemianopia; mean age 45.2 yrs., SD=18.4), and 14 control subjects (mean age 45.7 yrs., SD=18.0; 6 men). Test-retest reliability For all participants (n=29) and for the population with stroke and UN, the test-retest reliability was good (ICC =0.82 and 0.84) (table 1). For control subjects (n=14), the test-retest reliability was average (ICC=0.58). Inter rater reliability Inter rater reliability was excellent (ICC=0.99) for all participants (table 2). Patients with a unilateral hemispheric brain vascular lesion (left or right) with UN were recruited in two rehabilitation units (in Paris, France & Montréal QC, Canada). A convenience sample of healthy matched controls was recruited. In the UN-moving-task, participants were asked to go through a corridor (2.30 meters wide, 20 meters long), while pointing at each of the 18 magazine covers displayed, distributed at 3 levels of height, on the two walls. Two examiners recorded the number of omissions on each side. One OT and two physiatrists participated as raters in this study; they were familiar with the use of the UN-Moving-Task. During the task all participants wore an Eye Tracking Tobii, recording what their eye captures (i.e. magazine covers) (see Figure2). Test-retest reliability was determined based on scores obtained by the same judge in 2 evaluations 20 to 30 minutes apart, using Intra-Class Coefficients (ICC, one factor random). The re-test was performed by the participants in a second corridor of the rehabilitation unit. Inter-rater reliability was determined by comparing scores obtained by the 2 examiners in the same test, using ICC. METHOD Test-retest reliability Population ICC 95% CI All participants (n=29) 0.82 0.497; 0.923 Stroke population (n=18) 0.84 0.534; 0.945 Control subjects (n=14) 0.58 -0.561; 0.579 Inter rater reliability Population ICC 95% CI All participants (n=29) 0.99 0.992; 0.998 Stroke population (n=18) 0.99 0.994; 0.999 Control subjects 0.99 0.958; 0.996 Table 1. Omission of magazine covers. Test-retest Table 2. Omission of magazine covers. Inter rater reliability Legend : ICC: Intra Class correlation 95 CI: Confidence Internal Test-retest reliability was good for participants with stroke, unlike for control subjects. These results can probably be explained by the fact that UN is an observable deficit in attention to and awareness of one side of the field of vision. Thus, participants with UN may not be able to adapt to the test as easily as the control subjects. Inter rater reliability was excellent, which can be explained by the fact that the test is standardized, simple to implement, and its rating system is dichotomous (seen / not seen). If the results obtained from the reliability of the UN-moving-task findings are confirmed with a larger cohort (n = 50), as well as in two locations, it will be a useful clinical tool to evaluate UN on the move. Figure2. Example what eye captures with Eye Tracking Tobii
Transcript of Reliability properties of the UN-moving-task, an … · 5AP-HP, Hôpital Raymond Poincaré, Service...
ReliabilitypropertiesoftheUN-moving-task,anecologicalassessmentofUnilateralNeglectinpatientswithacquiredbraininjury:
Preliminaryresults
1SorbonneUniversités,UPMCUnivParis06,Handicapcognitifetréadaptation(HanCRe),Paris,France;2CentreforInterdisciplinary Research inRehabilitation ofGreater Montreal (CRIR),MontréalQC,Canada;3LaboratoryforAdultDevelopmentandCognitiveAging,DepartmentofPsychology,ConcordiaUniversity,MontréalQC,Canada;4InstitutNazarethetLouis-BrailleoftheCISSSdelaMontérégie-Centre,LongueuilQC,Canada;
5AP-HP,HôpitalRaymondPoincaré,ServicedeMPR,Garches,France;6HANDIReSP,UniversitédeVersaillesSaint-Quentin,France;7UniversitédeMontréal,MontréalQC,Canada;8InstitutuniversitairesurlaréadaptationendéficiencephysiquedeMontréal(IURDPM)oftheCIUSSSduCentre-Sud-de-l’Île-de-Montréal,MontréalQC,Canada.
FrédériquePoncet1-4,NoémieDuclos2,7,8,WalterCybis2,4,PhilippeAzouvi5,6,CyrilDuclos2,7,8 &Marie-ChantalWanet-Defalque2,4,7
Unilateralneglect(UN)isafrequentprobleminindividualswithbraininjury.UNhasbeendescribedasaperipersonal orextrapersonaldisorder.AlthoughmanytasksassessUNinperipersonal space,neglectinextrapersonal orfarspaceisrarelyassessed.TheUN-moving-task,anecologicaltestofUNinarealcorridorandinvolvingmulti-tasking(seeFigure1),wasdevelopedandexhibitsexcellentdiscriminantvalidity(Poncetetal.2012).However,otherimportantpsychometricpropertiesofthistesthaveyettobestudied.
Tofurtherevaluatethepsychometricpropertiesofthistest,thisstudyexploredreliabilityoftheUN-moving-task.
Figure1.UN-Moving-Task
INTRODUCTION
OBJECTIVE
RESULTS
CONCLUSION
Correspondence:[email protected]
Preliminaryresultswerecalculatedfromacohortof29participantsrecruitedfromtheIURDPM(Gingras-Lindsay-de-Montréal),including18participantswithstroke(12participantswithrightstroke(9men)and6withhomonymoushemianopia;meanage45.2yrs.,SD=18.4),and14controlsubjects(meanage45.7yrs.,SD=18.0;6men).
Test-retestreliabilityForallparticipants(n=29)andforthepopulationwithstrokeandUN,thetest-retestreliabilitywasgood(ICC=0.82and0.84) (table1).Forcontrolsubjects(n=14),thetest-retestreliabilitywasaverage(ICC=0.58).
InterraterreliabilityInterraterreliabilitywasexcellent(ICC=0.99)forallparticipants(table2).
Patientswithaunilateralhemisphericbrainvascularlesion(leftorright)withUNwererecruitedintworehabilitationunits(inParis,France&MontréalQC,Canada).Aconveniencesampleofhealthymatchedcontrolswasrecruited.IntheUN-moving-task,participantswereaskedtogothroughacorridor(2.30meterswide,20meterslong),whilepointingateachofthe18magazinecoversdisplayed,distributedat3levelsofheight,onthetwowalls.
Twoexaminersrecordedthenumberofomissionsoneachside.OneOTandtwophysiatristsparticipatedasratersinthisstudy;theywerefamiliarwiththeuseoftheUN-Moving-Task.DuringthetaskallparticipantsworeanEyeTrackingTobii,recordingwhattheireyecaptures(i.e.magazinecovers)(seeFigure2).
Test-retestreliabilitywasdeterminedbasedonscoresobtainedbythesamejudgein2evaluations20to30minutesapart,using Intra-ClassCoefficients(ICC,onefactorrandom).There-testwasperformedbytheparticipantsinasecondcorridoroftherehabilitationunit.
Inter-raterreliabilitywasdeterminedbycomparingscoresobtainedbythe2examinersinthesametest,usingICC.
METHOD
Test-retestreliability
Population ICC 95%CIAll participants (n=29) 0.82 0.497;0.923Stroke population (n=18) 0.84 0.534;0.945Control subjects (n=14) 0.58 -0.561;0.579
Interraterreliability
Population ICC 95%CIAll participants (n=29) 0.99 0.992;0.998Stroke population (n=18) 0.99 0.994;0.999Control subjects 0.99 0.958;0.996
Table1.Omissionofmagazinecovers.Test-retest Table2.Omissionofmagazinecovers.Interraterreliability
Legend:ICC: IntraClasscorrelation95CI: ConfidenceInternal
Test-retestreliabilitywasgoodforparticipantswithstroke,unlikeforcontrolsubjects.TheseresultscanprobablybeexplainedbythefactthatUNisanobservabledeficitinattentiontoandawarenessofonesideofthefieldofvision.Thus,participantswithUNmaynotbeabletoadapttothetestaseasilyasthecontrolsubjects.Interraterreliabilitywasexcellent,whichcanbeexplainedbythefactthatthetestisstandardized,simpletoimplement, anditsratingsystemisdichotomous(seen/notseen).IftheresultsobtainedfromthereliabilityoftheUN-moving-taskfindingsareconfirmedwithalargercohort(n=50),aswell asintwolocations,itwillbeausefulclinicaltooltoevaluateUNonthemove.
Figure2.ExamplewhateyecaptureswithEyeTrackingTobii
