Experience of an open referral system for stroke rehabilitation in the community

Objective: To examine referral patterns and characteristics of patients assessed by a community stroke rehabilitation service utilizing an open referral system. Design: Descriptive study. Setting: Two boroughs in a health district in South London, England. Subjects: Two hundred and seventy-six patients with stroke assessed by the service in its first 18 months of operation. Results: Patients were referred from a variety of sources including hospitals (38%), general practitioners (18%), district nurses (10%), social services (9%), self-referrals (8%), community physiotherapists (7%) and family or carers (2%). Referrals from hospital were younger (33% aged under 65 years) while referrals from community health and social services were older (44% aged 75-84 years) (chi(2) = 23, df = 9, p < 0.01). Fifty per cent of patients who had self-referred or had been referred by families or carers had longstanding stroke (> 1 year), 46% of patients from general practitioners had acute stroke (< 4 weeks) and 37% of hospital patients were seen between one and three months post stroke (chi(2) = 50.4, df = 12, p < 0.001). There was a marginally significant difference in the prevalence of incontinence and no significant differences in limb weakness, visual impairment and speech impairment. There were no significant changes in sources of referral between the first and second nine months of the service. There was a significant increase in the percentage of patients with recent stroke seen in the second nine months (chi(2) = 22.8, df = 5, p < 0.001). In this period, 36% of patients assessed were seen within four weeks of their stroke compared with 15% in the first nine months. Of patients with stroke, 8.7% were considered to have been inappropriate referrals but they were not significantly associated with any particular source of referral. Conclusions: Referrals were received from a wide variety of sources with significant differences in age and duration of stroke of patients from different referral sources. The results suggest that an open referral system may make rehabilitation accessible to a broader range of patients with stroke living in the community.