Seasonal variation in mortality in Scotland

Abstract

Background. Seasonal patterns in mortality have been recognized for many years. This study assesses seasonal variation in mortality in Scotland between 1981 and 1993 and considers its association with socioeconomic status and outdoor temperature. Methods. Lagged Poisson regression analysis of numbers of deaths and average weekly temperature with adjustment for serial autocorrelation and influenza epidemics. Results. There was significant seasonal variation in weekly death rates with a difference of about 30% between a summer trough and a winter peak. This variation was principally attributable to respiratory disease, cerebrovascular disease and coronary artery disease. Seasonal variation in mortality fell from around 38% in 1981-1983 to around 26% in 1991-1993. There was no clear evidence of a relationship between socioeconomic status and seasonal mortality, however the extent of the fall in seasonal variation was greater in deprived areas than in affluent areas. Overall, a 1°C decrease in mean temperature was associated with a 1% increase in deaths one week later. The lag in this relationship varied by cause of death and underlying temperature. Conclusions. Seasonal variations in mortality and the relationship between temperature and mortality are a significant public health problem in Scotland. It is likely that the strength of this relationship is a result of the population being unable to protect themselves adequately from the effects of temperature rather than the effects of temperature itself.