Comparing participation outcome over time across international stroke cohorts: outcomes and methods.

Abstract

To enable a direct comparison of participation levels in the first year post-stroke, assessed by different outcome measures internationally. Two prospective stroke cohort studies following persons from stroke onset to 12 months post-stroke. Community. Persons with stroke, not living at a nursing home, from Australia (START-PrePARE; n = 100) and the Netherlands (Restore4stroke; n = 395). Not applicable. Activity Card Sort-Australia (ACS-Aus) and Utrecht Scale for Evaluation of Rehabilitation-Participation (USER-P). Activity domains were matched across measures to find common denominators and original scoring methods were recoded, hereby enabling a direct comparison of retained activities. Ninety-one (START-PrePARE) and 218 (Restore4stroke) persons with stroke were included for analyses. No major differences in background characteristics were observed between the cohorts; the Dutch cohort suffered from slightly more severe stroke. A higher level of participation was observed (radar charts) in the first months post-stroke for the Australian cohort than in the Dutch cohort, especially for unpaid work (p < .003). At 12 months post-stroke, participation levels were similar, without significant differences in retained activities using the defined common denominators (p > .003). An international comparison of actual activities that persons re-engage in in the first year post-stroke was achieved using a new method and recoding of data. High levels of participation were observed in both cohorts. Unpaid work showed different frequencies at 2-3 months, contributing to different trajectories over time across cultures. Important insights were gained. Although valuable information is inevitably lost with recoding, the approach may assist future studies on the harmonization of data across cohorts, particularly for one of the key outcomes of stroke: participation.