Objective: To undertake a detailed analysis of therapy provided in a multicentred randomized controlled trial of activities of daily living (ADL) and leisure (TOTAL), testing the hypothesis that specific interventions given in the trial affected specific aspects of outcome.
Subjects: Three hundred and nine stroke patients who had been randomly allocated to receive either occupational therapy aimed at ADL activities (n = 156) or leisure (n = 153).
Measures: Number, duration and type of activity undertaken per patient. Barthel Index, Extended Activities of Daily Living Scale (EADL) and Nottingham Leisure Questionnaire (NLQ) six months after entry to the study.
Method: Activities that had been used in treatment were coded and categorized. Frequently used activities identified. These activities were matched to items from the six-month outcome measures. Patient independence in these outcome items was compared between the leisure and ADL groups.
Results: Three hundred and nine therapy record forms were returned. Patients received a median of ten sessions with a median duration of 55 minutes. The ADL group received significantly more, mobility training, transfer training, cleaning, dressing, cooking and bathing training (chi-squared, p < 0.05). Sport, creative activities, games, hobbies, gardening, entertainment and shopping were used significantly more in the leisure group (chi-squared, p < 0.05) than the ADL group. Fifteen items from the outcome measures were identified as specific to these interventions. There were no statistically significant differences in outcome on these 15 items between the ADL and leisure groups (chi-squared, p > 0.05).
Conclusions: We found no evidence that specific ADL or leisure interventions led to improvements in specific relevant outcomes. We believe that these findings should prompt a review of the relationship between process and outcome of occupational therapy.