A new method to estimate the temperature-CVD mortality relationship.

Abstract

Few detailed, individual-focused studies have researched the added effect of temperature on cardiovascular disease (CVD), particularly in China. Moreover, no prior studies have explored the exposure-response relationship among all populations and different sub-sociodemographic groups. A distributed lag nonlinear model (DLNM) was applied to evaluate the adverse health effects of temperature on CVD mortality for all populations and different sub-sociodemographic groups (by age, sex, educational level, living arrangement, and occupation) in Beijing. Based on the exposure-response relationships, firstly, we proposed a new model (COCKTAIL, Code Of Climate Key To An Ill) for revealing the split-and-merge relationships of the temperature-CVD mortality curve. This method could be used to apply the CVD deaths in a studied area to forecast the exposure-response relationships in the same area in the future. Secondly, this is the most detailed study to analyze the relationship between temperature and CVD mortality for different subgroups among the existing researches for developed and developing countries. We found that the cold temperature (at - 14°C) was the risk factor for people with low socioeconomic status, especially for single people (including unmarried, divorced, and widowed), for indoor workers, and for people with low education, compared with the minimum mortality temperature, with a cumulative increase of 3.9 (80%CI, 2.9-5.4), 3.8 (80%CI, 2.8-5.1), and 4.5 (80%CI, 3.1-6.3) times respectively. Meanwhile, the hot temperature (at 35°C) was the risk factor for CVD death, with a cumulative increase of 2.6 (80%CI, 2.0-3.4) for females, and 3.1 (80%CI, 2.4-4.2) for single people. The varying CVD vulnerability in terms of CVD mortality among various groups may assist governments in preparing health resources and taking measures to prevent or reduce temperature-related deaths.