Abstract: ObjectiveTo investigate the effect of ambient ozone on mortality risk and its temperal-spatial distribution in the Pearl River Delta cities, and further explore modification factors of the variation in the distribution. MethodsA time-series study design was employed. Cities of Guangzhou, Dongguan, Foshan, Jiangmen, and Zhaoqing were selected as study settings. Daily non-accidental mortality data from 2013 to 2014 in each city were collected from Guangdong Provincial Center for Disease Control and Prevention. The contemporaneous daily ambient ozone concentration and meteorological indictors were obtained from Guangdong Environmental Monitoring Center and Guangdong Meteorological Service, respectively. A generalized additive model was used to estimate the excess risk (ER) of mortality for each 10μg/m3 increment in average ambient ozone concentration during Lag 02 days in the full year, cold season (November to April), and warm season (May to October) in each city. Then a Meta-analysis was used to quantitatively combine the ERs in each city. Finally, a Meta-regression model was used to explore the effects of city specific ambient ozone level, elderly population proportion (>65 years), and average income level on the ER in the cities.ResultsFrom 2013 to 2014, the average daily nonaccidental deaths were 136.2, 29.5, 9.3, 77.2, and 18.9 and the average concentrations of atmospheric ozone were 57.0, 71.9, 53.9, 53.9, and 64.6 μg/m3 in cities of Guangzhou, Dongguan, Foshan, Jiangmen, and Zhaoqing, respectively. In the full year, a 10μg/m3 increment in ambient ozone concentration was associated with an increment in ER by 0.37% (95%CI：0.07% - 0.66%), 0.44% (95%CI：-0.11% -0.99%), -0.05% (95%CI：-1.08% - 0.99%), 0.21% (95%CI：-0.23% - 0.66%), and 0.65% (95%CI: -0.09% - 1.40%) of total mortalities in Guangzhou, Dongguan, Foshan, Jiangmen, and Zhaoqing, respectively. The combined ER in the five cities was 0.35% (95%CI：0.14% -0.56%). The pooled ER in the cold season (ER=0.51%，95%CI：0.20% - 0.83%) was larger than that in the warm season (ER=0.22%，95%CI：-0.07% - 0.51%). Meta regression analysis showed that the city specific ER was positively related to the annual average ozone concentration and elderly population proportion, but negatively related to average income level although without statistical significance.ConclusionThe ambient ozone in the Pearl River Delta region could increase the mortality risk of population with a higher effect in the cold season. The health effects of ozone largely differentiated between cities. The modification effects of ambient ozone level, proportion of the elderly, and income level on the ER are still not clear, which needs further study.