부산대학교 도서관

Background We investigated whether the incidence of death following myocardial infarction (MI) varied by season and latitude in the Swedish population. Methods We studied deaths following MI from January 1987 to December 2009, using the Swedish National Cause of Death Register. County of residence was used to determine latitude and population density. An extension of Poisson regression was used to study the relationship between risk of death following MI with age, latitude, time (from 1987), population density and calendar days. Results Over the study period, there was a secular decrease in the incidence of MI-related death. In men, MI-related death incidence increased by 1.3% [95% confidence interval (CI) = 1.1–1.5] per degree of latitude (northwards). In women, MI-related death incidence increased by 0.6% (95% CI = 0.4–0.9) per degree of latitude. There was seasonal variation in the risk of MI-related death with peak values in the late winter and a nadir in the summer months in both the north and the south of Sweden. Findings were similar with incident MI as the outcome. Conclusions The incidence of MI-related death varied markedly by season and latitude in Sweden, with summer months and more southerly latitude associated with lower rates than winter months and more northerly latitude.