Article

OBJECTIVES: To explore the relation between a computer adaptive functional
cognitive questionnaire and a performance-based measure of cognitive instrumental
activities of daily living (C-IADL) and to determine whether the Montreal
Cognitive Assessment (MoCA) at admission can identify those with C-IADL
difficulties at discharge. DESIGN: Prospective cohort study. SETTING: Acute
inpatient rehabilitation unit of an academic medical center. PARTICIPANTS:
Inpatients (N=148) with a diagnosis of stroke (mean age, 68y; median, 13d
poststroke) who had mild cognitive and neurological deficits. INTERVENTIONS: Not
applicable. MAIN OUTCOME MEASURES: Admission cognitive status was assessed by the
MoCA. C-IADL at discharge was assessed by the Executive Function Performance Test
(EFPT) bill paying task and Activity Measure of Post-Acute Care (AM-PAC) Applied
Cognition scale. RESULTS: Greater cognitive impairment on the MoCA was associated
with more assistance on the EFPT bill paying task (rho=-.63; P<.01) and AM-PAC
Applied Cognition scale (rho=-.43; P<.01). This relation was nonsignificant for
higher MoCA scores and EFPT bill paying task scores. The AM-PAC Applied Cognition
scale and the EFPT bill paying task had low agreement in classifying functional
performance (Cohen's kappa=.20). A receiver operating characteristic curve
identified optimal MoCA cutoff scores of 20 and 21 for classifying EFPT bill
paying task status and AM-PAC Applied Cognition scale status, respectively. For
values above 20 and 21, sensitivity increased whereas specificity decreased for
classifying functional deficits. Approximately one third of the participants
demonstrated C-IADL deficits on at least 1 C-IADL measure at discharge despite
having a MoCA score of >/=26 at admission. CONCLUSIONS: Questionnaire and
performance-based methods of assessment appear to yield different estimates of
C-IADL. Low MoCA scores (<20) are more likely to identify those with C-IADL
deficits on the EFPT bill paying task. The results suggest that C-IADL should be
assessed in those who have mild or no cognitive difficulties at admission.
CI - Copyright (c) 2017 American Congress of Rehabilitation Medicine. Published by
Elsevier Inc. All rights reserved.

Langue : ANGLAIS