Abstract P164: Seasonal patterns of mortality from cardiovascular diseases in Panama.

Abstract

Seasonal patterns of cardiovascular (CV) mortality have been reported in many regions of the world, but mostly in temperate zones. These cycles and seasonal patterns of mortality have been identified by looking at mortality rate series over extended periods of time and by filtering its stochastic components. The reasons for these variations have been attributed to multiple causes. Empirical observations by health care worker have suggested that in Panama there are seasonal changes of CV disease mortality but this phenomenon had not been studied. The speculation is that this may be related to variation in the incidence of respiratory diseases, weather changes or to population activity patterns associated with holidays. Our hypothesis for this study is that in Panama mortality related to CV disease has a seasonal pattern. We compiled from the National Registry of Mortality all deaths that occurred in the country from cardiovascular diseases (n=36,145) and diabetes (n=7,076) for the years 2001–2009. A monthly time series of mortality was calculated and its components were decomposed utilizing moving averages. Elements, like cycle, seasonal occurrence and stochastic components, were separated. We also developed a seasonal index of mortality for these diseases. For the same years, we performed a similar analysis for all deaths attributed to cancer and external causes. Analysis of these mortality series revealed seasonal cycles for total cardiovascular mortality and diabetes mortality. Death from cancer and external causes also showed seasonal cycles. CV mortality and diabetes mortality peaked at the beginning of the third trimester of the year and were lowest during the first trimester. For the month of July, the CV mortality seasonal index showed a score of 108, and an average number of CV deaths of 364. The lowest seasonal index score for CV mortality was 87 in February, with an average number of CV deaths of 296. The cycles observed for cancer and deaths from external causes did not match the cycles observed for CV diseases and diabetes. We conclude that there is a seasonal pattern of CV mortality in Panama. The drivers of these cyclical changes of CV mortality have not yet been identified, but we are evaluating the relationship of acute respiratory illnesses, population activity patterns and weather changes to these seasonal mortality variations.