Impact of hot temperature on death in London: a time series approach

Study objective: This study investigated the relation between heat and mortality in London to determine the temperature threshold at which death rates increase and to quantify the effect of extreme temperatures on mortality. Design: Daily data on all cause mortality and temperature were obtained for a 21 year period and the relation between them investigated both graphically and by using non-parametric time series methods of analysis. Setting: Greater London. Participants: Daily mortality counts in Greater London between January 1976 and December 1996. Main results: A plot of the basic mortality-temperature relation suggested that a rise in heat related deaths began at about 19degreesC. Average temperatures above the 97th centile value of 21.5degreesC (excluding those days from a 15 day "heatwave" period in 1976) resulted in an increase in deaths of 3.34% (95% CI 2.47% to 4.23%) for every one degree increase in average temperature above this value. It was found that the 1976 heatwave resulted in a particularly large number of deaths in comparison with other hot periods. Conclusions: These results suggest that heat related deaths in London may begin at relatively low temperatures. Hot days occurring in the early part of any year may have a larger effect than those occurring later on; and analysis of separate heatwave periods suggest that episodes of long duration and of highest temperature have the largest mortality effect.