The short-term effect of ambient ozone on mortality is modified by temperature in Guangzhou, China Tao Liu a, b, 1 , Tian Tian Li c, 1 , Yong Hui Zhang b , Yan Jun Xu b , Xiang Qian Lao d , Shannon Rutherford e , Cordia Chu e , Yuan Luo a, b , Qi Zhu a, b , Xiao Jun Xu b , Hui Yan Xie a, b , Zhao Rong Liu f , Wen Jun Ma a, b, * a Guangdong Institute of Public Health, Guangzhou, China b Center for Disease Control and Prevention of Guangdong Province, Guangzhou, China c Institute of Environmental Health and Related Product Safety, Chinese Center for Disease Control and Prevention, Beijing, China d School of Public Health and Primary Care, Faculty of Medicine, The Chinese University of Hong Kong, China e Center for Environment and Population Health, School of Public Health, Griffith University, Australia f College of Environmental Sciences and Engineering, Peking University, Beijing, China highlights < Ambient ozone had independent effects on non-accidental mortality in cold season. < The effects of ozone lasted longer in cold season and low temperature days. < Mortality displacement was observed in days with low temperature. < The risk assessment might be underestimated by using single-day exposure model. article info Article history: Received 22 March 2012 Received in revised form 2 July 2012 Accepted 6 July 2012 Keywords: Ambient ozone Non-accidental mortality Lag effect Time-series study Mortality displacement abstract Effects of ozone on mortality have been widely assessed in developed countries but rarely in developing countries, and the effects of season and temperature on these associations remain unclear. The present study aimed to explore the modifying effects of temperature on the association of ozone with mortality, and to examine the lag effect structure in Guangzhou, China. Daily non-accidental mortality, air pollution and meteorological data from January 1st 2006 to December 31st 2008 in Guangzhou were collected. Gener- alized additive models (GAM) and distributed lag models (DLM) were used to estimate the excess risk (ER) of ozone on daily mortality and capture the lag effect structure. Results revealed that, in cold season, an increment in the ozone concentration of 10 mgm 3 was associated with a 0.87% (95% CI: 0.06 to 2.29%) and 3.34% (95%CI: 1.36e5.35%) increase of mortality for lag0 and lag0e6, respectively. However, no significant effects were observed in the warm season. Daily average temperature and ozone concentration had interactive negative effects on non-accidental mortality in the cold season and at daily average tempera- tures in the 0e25th percentile, and mortality displacement was also found in these conditions. Therefore, the short-term effect of ambient ozone on mortality is modified by temperature in Guangzhou, China. Ó 2012 Elsevier Ltd. All rights reserved. 1. Introduction Concomitant with rapid industrialization and urbanization in China are alarming increases in the number of motor vehicles in most Chinese mega-cities. Air pollution sources in these cities have gradually changed from conventional coal combustion to a mixture of coal combustion and motor vehicle emissions. Large emissions of motor vehicle exhaust compounds such as nitrogen oxides (NO x ), volatile organic compounds (VOCs), and carbon monoxide (CO) can produce ambient ozone through photochemical reactions. In recent decades, China has become one of the countries with the highest Abbreviations: GAM, generalized additive models; DLNM, distributed lag non- linear models; DLM, distributed lag models; ER, excess risk; CI, confidence interval; NO x , nitrogen oxides; VOCs, volatile organic compounds; O 3 , ozone; PM 10 , particulate matter with an aerodynamic diameter less than 10 mm; PM 2.5 , partic- ulate matter with an aerodynamic diameter less than 2.5 mm; TM, daily mean temperature; RH, relative humidity; CER, cumulative excess risk; RR, relative risk. * Corresponding author. Guangdong Institute of Public Health, Guangzhou, China. Tel.: þ86 20 84458530; fax: þ86 20 89104239. E-mail addresses: mwj68@vip.tom.com, mwj68@tom.com (W.J. Ma). 1 These authors contributed equally to this work. Contents lists available at SciVerse ScienceDirect Atmospheric Environment journal homepage: www.elsevier.com/locate/atmosenv 1352-2310/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.atmosenv.2012.07.011 Atmospheric Environment 76 (2013) 59e67