What is the association between the different components of stroke rehabilitation and health outcomes?

To describe the rehabilitation input stroke patients received from health professionals during the early post stroke period and to explore possible associations between health outcomes and these rehabilitation inputs.

Community-based study with prospective identification of stroke patients from a random sample of 24 general practices stratified by geographical area and practice size.

Ayrshire and Arran Health Board area, West of Scotland.

All physiotherapy, occupational therapy, dietetics, podiatry, speech and language therapy and community nursing inputs given to stroke patients in the course of normal treatment were recorded.

Barthel Index and Medical Outcomes Study Short Form-36 (SF-36). Both recorded at one, three and six months post stroke.

Of the 152 people providing data, more had received physiotherapy than any other rehabilitation input at all three follow-ups (81%, 47%, 39%), with occupational therapy being the next most common service (65%, 44%, 25%). Amount of rehabilitation input was significantly negatively correlated with health outcomes measured at each discrete time point: those patients with the poorest outcomes received greatest input. However, regression analysis of change in outcome scores showed that increasing amounts of rehabilitation input were significantly associated with a reduction in disability, particularly between one and three months post stroke.

Not only have we shown that those stroke patients with poorest outcomes received most rehabilitation input, but, from analysis of the individual rehabilitation inputs, we have identified some rehabilitation inputs that significantly predict improved outcomes. This suggests that there would be merit in further investigation of these associations.