ABSTRACT
<p><b>OBJECTIVE</b>To evaluate the impacts of air temperature on years of life lost (YLL) among the residents in Guangzhou and Zhuhai, Guangdong province.</p><p><b>METHODS</b>Daily mortality and meteorology data in Guangzhou and Zhuhai were collected, and distributed lag non-linear model (DLNM) was used to evaluate the cumulative and delayed effects of daily air temperature on YLL of total non-accident mortality. The accumulative effect of air temperature on mortality under the extreme high temperature (0-1 days) and extreme low temperature (0-13 days) situation in Guangzhou and Zhuhai were analyzed respectively.</p><p><b>RESULTS</b>The average YLL was 1 928.0 in Guangzhou and 202.5 in Zhuhai. The exposure-response functions seemed to be non-linear. The hot effect seemed to be acute and reached the peak at the same day, while the cold effect reached the peak at 5(th) days and lasted for about two weeks. Low temperature had stronger gross effect than high temperature had. The cold effect among males was greater than that among females in Guangzhou. The hot/cold effect on YLL was greater in people aged ≥ 65 years than in people aged < 65 years and in people suffering from respiratory disease than in people suffering from cardiovascular disease in both Guangzhou and Zhuhai.</p><p><b>CONCLUSION</b>The effects of high and low temperatures on YLL were obvious, and the impact of low temperature was greater. The elderly and people suffering from respiratory disease or cardiovascular disease are the vulnerable populations.</p>