Early increases in ischaemic heart disease mortality dissociated from and later changes associated with respiratory mortality after cold weather in south east England

Abstract

Study objective: To identify the time courses and magnitude of ischaemic heart (IHD), respiratory (RES), and all cause mortality associated with common 20-30 day patterns of cold weather in order to assess links between cold exposure and mortality.

Design: Daily temperatures and daily mortality on successive days before and after a reference day were regressed on the temperature of the reference day using high pass filtered data in which changes with a cycle length < 80 days were unaffected (< 2%), but slower cyclical changes and trends were partly or completely suppressed. This provided the short term patterns of both temperature and mortality associated with a one day displacement of temperature. The results were compared with simple regressions of unfiltered mortality on temperature at successive delays.

Study population and setting: Population of south east England, including London, over 50 years of age from 1976-92.

Main results: Colder than average days in the linear range 15 to 0 degrees C were associated with a "run up" of cold weather for 10-15 days beforehand and a "run down" for 10-15 days afterwards. The increases in deaths were maximal at 3 days after the peak in cold for IHD, at 12 days for RES, and at 3 days for all cause mortality. The increase lasted approximately 40 days after the peak in cold. RES deaths were significantly delayed compared with IHD deaths. Excess deaths per million associated with these short term temperature displacements were 7.3 for IHD, 5.8 for RES, and 24.7 for all cause, per one day fall of 1 degree C. These were greater by 52% for IHD, 17% for RES, and 37% for all cause mortality than the overall increases in daily mortality per degree C fall, at optimal delays, indicated by regressions using unfiltered data. Similar analyses of data at 0 to -6.7 degrees C showed an immediate rise in IHD mortality after cold, followed by a fall in both IHD and RES mortality rates which peaked 17 and 20 days respectively after a peak in cold.

Conclusion: Twenty to 30 day patterns of cold weather below 15 degrees C were followed:(1) rapidly by IHD deaths, consistent with known thrombogenic and reflex consequences of personal cold exposure; and (2) by delayed increases in RES and associated IHD deaths in the range 0 to 15 degrees C, which were reversed for a few degrees below 0 degree C, and were probably multifactorial in cause. These patterns provide evidence that personal exposure to cold has a large role in the excess mortality of winter.