Can differences in management processes explain different outcomes between stroke unit and stroke-team care?

Background Stroke units reduce mortality and dependence, but the reasons are unclear. We have compared differences in management and complications of patients with acute stroke who were admitted to a stroke unit or to a general ward as part of a previously reported randomised trial. Methods 304 patients had been randomly assigned to stroke units (n=152) or to general wards supported by a specialist stroke team (152). We used a structured format to gather prospective data on the frequency of prespecified interventions in each of the major aspects of stroke care. Observations were undertaken daily for the first week and every week for the next 3 months by independent observers. The effect of differences in management on outcome at 3 months was assessed with the modified Rankin score, dichotomised to good (0–3) and poor (4–6) outcome. Findings Patients in the stroke unit were monitored more frequently (odds ratio 2·1 [1·3–3·4]) and more patients received oxygen (2·0 [1·3–3·2]), antipyretics (6·4 [1·5–27·5]), measures to reduce aspiration (6·0 [2·3–15·5]), and early nutrition (14·4 [5·1–40·9]) than those in general wards. Complications were less frequent in patients in the stroke unit than those in general wards (0·6 [0·2–0·7]), with fewer patients having progression of stroke, chest infection, or dehydration. Measures to prevent aspiration, early feeding, stroke unit management, and frequency of complications independently affected outcome. Interpretation Differences in management and complications between the stroke unit and general wards differ substantially, even when specialist support is provided. Such differences could be responsible for the more favourable outcome seen in patients on stroke units than those on general wards.