In the United Kingdom, about 25% of patients with stroke are not admitted to hospital. Unfortunately many of these patients remain in the community with little or no coordinated rehabilitation. We have recently published the results of a single blind randomised controlled trial of occupational therapy for patients with stroke remaining in the community.1 We showed significant benefits across a range of outcomes at 6 months after stroke; extended activities of daily living (EADL), personal ADL, gross motor function, handicap, and carer strain. Correspondence2 suggested that these results were consistent with other similar smaller trials. In this report we present results at 1 year after stroke.

We identified patients from a community stroke register who had had a recent stroke (<1 month) and had not been admitted to hospital.1 The stroke register covered a geographical area of Nottinghamshire and southern Derbyshire and incorporated 73 general practitioner practices, covering a population of 500 000 patients. Patients were included if they fulfilled the World Health Organisation (WHO) definition of stroke and had not been admitted to hospital. Patients were excluded if they lived in a nursing or residential home, could not speak or understand English before their stroke, or had a history of dementia. Patients were randomly allocated to up to 5 months of occupational therapy treatment at home (n=94) or to no intervention (n=91). The groups were well matched for baseline demographic characteristics.1 The aim of occupational therapy was to encourage independence in personal and extended ADL. The number of visits given by the research occupational therapist to the intervention group ranged from 1 to 15, with a mean of six visits. Outcome measures were completed at 6 and 12 months after stroke. Outcome measures used at 1 year included the Barthel index, EADL scale, and the general health questionnaire (GHQ 28) for both patient and carer. All assessments were conducted in the patient's home by a blinded independent assessor.

Sixteen patients could not be assessed at 1 year: 10 had died during follow up (five in the occupational therapy group) and six withdrew from the study (four in the occupational therapy group). As the number of withdrawals from the study was small and evenly distributed between the groups, the analysis was conducted on the 147 patients who completed assessments. Demographic data at 1 year are illustrated in the table.

The primary outcome measure of the trial was the Nottingham EADL.3 Analysis of the EADL at 1 year after stroke, using the Mann-Whitney U test, showed that the group receiving occupational therapy were significantly less disabled (table).

Although it is important to ascertain whether an intervention is effective at certain time points, the above analysis does not take into account variation over time or that outcome measures at different points in time were from the same patients. We therefore applied an analysis of serial measurements. The area under the curve (AUC) was calculated as a summary measure4 for outcome assessments in both groups. There was a significant difference between the groups in favour of the treatment group on the AUC EADL scale (Mann-Whitney p=0.001, 95% confidence interval (95% CI) 7 to 28) and the AUC Barthel index (Mann-Whitney p=0.01, 95% CI 0 to 9). There were no significant differences in the AUC for the GHQ for either the patient (Mann-Whitney p=0.48, 95% CI −39 to 21) or the carer (Mann-Whitney p=0.66, 95% CI −44 to 29) .

This study indicated that patients who received occupational therapy had a greater level of independence in activities of daily living over a period of 1 year than patients who did not. Differences between the groups in terms of extended ADL were still apparent at 1 year.

The persistence of the beneficial effect of domiciliary occupational therapy adds further support to its clinical usefulness. There are grounds for establishing community occupational therapy services for patients with stroke, but further evaluation is required to confirm the generalisability of these findings and examine their economic implications.