Individual-level and community-level effect modifiers of the temperature–mortality relationship in 66 Chinese communities

OBJECTIVES: To examine the modification of temperature-mortality association by factors at the individual and community levels. DESIGN AND METHODS: This study investigated this issue using a national database comprising daily data of 66 Chinese communities for 2006–2011. A ‘threshold-natural cubic spline’ distributed lag non-linear model was utilised to estimate the mortality effects of daily mean temperature, and then examined the modification of the relationship by individual factors (age, sex, education level, place of death and cause of death) using a meta-analysis approach and community-level factors (annual temperature, population density, sex ratio, percentage of older population, health access, household income and latitude) using a meta-regression method. RESULTS: We found significant effects of high and low temperatures on mortality in China. The pooled excess mortality risk was 1.04% (95% CI 0.90% to 1.18%) for a 1°C temperature decrease below the minimum mortality temperature (MMT), and 3.44% (95% CI 3.00% to 3.88%) for a 1°C temperature increase above MMT. At the individual level, age and place of death were found to be significant modifiers of cold effect, while age, sex, place of death, cause of death and education level were effect modifiers of heat effect. At the community level, communities with lower socioeconomic status and higher annual temperature were generally more vulnerable to the mortality effects of high and low temperatures. CONCLUSIONS: This study identifies susceptibility based on both individual-level and community-level effect modifiers; more attention should be given to these vulnerable individuals and communities to reduce adverse health effects of extreme temperatures.