BACKGROUND: In patients with acute stroke, systemic thrombolysis needs to be administered within 3 hours of symptom onset. The aim of the present study was to reduce prehospital delays in a population-based intervention. METHODS: We performed a cluster-randomized trial with 48 zip code areas as cluster units in the catchment area of 3 inner-city hospitals in Berlin, Germany. The primary end point was time between symptom onset and hospital admission. The intervention consisted of an educational letter indicating stroke symptoms and emphasizing the importance of calling the emergency medical services. We additionally included a bookmark and sticker with the emergency medical services' telephone number. We fitted a series of log-normal survival regression models (time to admission) with frailty terms shared by inhabitants of the same zip code area. RESULTS: A total of 75 720 inhabitants received the intervention. Between 2004 and 2005, 741 patients with cerebrovascular events were admitted from the control areas (n = 24) and 647 from the intervention areas (n = 24). A prehospital time of 2 hours or less and 3 hours or less was achieved by 22% and 28% of patients, respectively, in the control group compared with 26% and 34%, respectively, in the intervention group. In the log-normal model, time to hospital was reduced by 27% in the intervention group in women (acceleration factor, 0.73; 95% confidence interval, 0.58-0.94), while no significant effect was found in men. CONCLUSIONS: The population-based intervention was effective in reducing prehospital delays in women but not in men. Future research should focus on the potential transferability of the intervention, its sustainability, and sex-specific impact.
BACKGROUND: In patients with acute stroke, systemic thrombolysis needs to be administered within 3 hours of symptom onset. The aim of the present study was to reduce prehospital delays in a population-based intervention. METHODS: We performed a cluster-randomized trial with 48 zip code areas as cluster units in the catchment area of 3 inner-city hospitals in Berlin, Germany. The primary end point was time between symptom onset and hospital admission. The intervention consisted of an educational letter indicating stroke symptoms and emphasizing the importance of calling the emergency medical services. We additionally included a bookmark and sticker with the emergency medical services' telephone number. We fitted a series of log-normal survival regression models (time to admission) with frailty terms shared by inhabitants of the same zip code area. RESULTS: A total of 75 720 inhabitants received the intervention. Between 2004 and 2005, 741 patients with cerebrovascular events were admitted from the control areas (n = 24) and 647 from the intervention areas (n = 24). A prehospital time of 2 hours or less and 3 hours or less was achieved by 22% and 28% of patients, respectively, in the control group compared with 26% and 34%, respectively, in the intervention group. In the log-normal model, time to hospital was reduced by 27% in the intervention group in women (acceleration factor, 0.73; 95% confidence interval, 0.58-0.94), while no significant effect was found in men. CONCLUSIONS: The population-based intervention was effective in reducing prehospital delays in women but not in men. Future research should focus on the potential transferability of the intervention, its sustainability, and sex-specific impact.