Seasonal patterns of incidence and case fatality of stroke in Malmo, Sweden: The STROMA study

Background: Studies on the temporal variation in stroke incidence have reported inconsistent results. Few have studied the temporal variations in case fatality. No study on incidence and case fatality of stroke by season in Sweden has been found. This study explores the weekly, monthly and seasonal variations in incidence and 28-day case fatality of stroke. Methods: A total of 7,129 patients with first-ever stroke during the period 1989-1999 were retrieved from the Stroke Register of Malmo, Sweden. chi(2) test was performed to test the seasonal differences and Poisson regression analysis was used to calculate the case fatality ratios adjusted for sex and age. Results: The stroke cases were on the whole randomly distributed over the study period of 4,017 days. Incidence of all types of stroke, intracerebral hemorrhage and subarachnoid hemorrhage showed no variation by season, month or weekday, but incidence of cerebral infarction for the total population was higher in autumn and winter. Twenty-eight-day case fatality (930 fatal events, 13%) clustered significantly over the study period. Winter emerged as the peak season among men (12.5%), women (17.2%) and total population (15.1%). No consistent variation of incidence and case fatality of stroke by month or weekday was found. Conclusion: Case fatality after stroke demonstrates a seasonal variation with a peak in winter. Incidence of stroke showed no consistent association with season, month or weekday. Copyright (C) 2005 S. Karger AG, Basel.