A Case-Crossover Study on the Heat Waves and Daily Death Counts

Abstract

ISEE-0919 Objective: To explore the effects of heat wave to daily death counts in Beijing. Methods: The data of death was from the Beijing Center for Disease Control and Prevention (CDC). The data of weather was provided from Institute of Urban Meteorology, CMA, Beijing. Case-crossover design was used to study the impact of heat wave to death. All deaths in summer from Jan. 1, 1999 to Jun. 30, 2000 were cases. The controls were the 7th day before death to analyze the lag days, duration of the effects, and the hazard. Results: There were five heat waves during the study period. The first heat wave sustained 9 days and the maximum temperature was 38.8°C with low humidity. The OR for the non-accidental death, cardiovascular disease and cerebrovascular disease were 1.442 (χ2 = 55.970, P < 0.001), 1.384 (χ2 = 9.499, P = 0.002) and 1.776 (χ2 = 32.438, P < 0.001), respectively. The second heat wave lasted 3 days, with the maximum temperature of 36.8°C and higher humidity than the first. The OR for the non-accidental death was 1.200 (χ2 = 2.730, P = 0.098). The third heat wave continued 7 days, with the maximum temperature of 41.5°C, and higher humidity than the first. The OR for all non-accident death, the daily death counts caused by cardiovascular disease, cerebrovascular diseases were 2.135 (χ2 = 215.082, P < 0.001), 2.613 (χ2 = 84.840, P < 0.001) and 2.317 (χ2 = 63.532, P < 0.001). The fourth last 3 days, with the maximum temperature of 39.6°C, low humidity. The OR for the non-accidental death and cerebrovascular diseases were 1.464 (χ2 = 4.529, P = 0.033) and 2.429 (Exact P < 0.064). The fifth heat wave lasted for 4 days. The maximum temperature was 37.4°C, with low humidity during the period. The OR for daily death counts caused by all non-accidental and cardiovascular disease were 1.525 (χ2 = 6.040, P = 0.014) and 2.333 (χ2 = 5.625, P = 0.018), respectively. Conclusion: There were significant increase for daily death counts in Beijing during the heat wave and there was lag days usually. The counts of non-accidental death was generally a time lag of 2∼3 days. Cardiovascular disease and cerebrovascular disease deaths usually increase after 2∼4 days. The increasing of air temperature was greater at the beginning of heat wave, the lag time was shorter. Fluctuations in daily maximum air temperature affected the death of cardiovascular disease death. The temperature rose higher after the beginning of the heat wave, the effects to cerebrovascular disease deaths was greater.