Influence of temperature to the short-term effects of various ozone metrics on daily mortality in Suzhou, China

Epidemiological studies have reported that both ozone (O3) and temperature are associated with daily mortality. However, few studies have explored the interactive effect between O3 and temperature on mortality. Moreover, whether different O3 metrics influence this interaction is still unknown. This study used a Poisson generalized additive model to examine the interactive effects of daily average temperature and O3 on mortality in Suzhou between 2008 and 2009. The results showed that the O3 effects were stronger in cold season and low temperature days. On days with low temperatures (0th–25th percentile), an interquartile range increment in 1-h maximum O3 corresponded to a 15.96% (95% CI: 1.52, 32.45) increase in total mortality and an 18.77% (95% CI: 0.09, 40.95) increase in cardiovascular mortality. On days with normal temperatures (26th–75th percentile) or high temperatures (76th–100th percentile), the estimates were only 2.04% (95% CI: −5.94, 10.69) or 1.65% (95% CI: −3.21, 6.76) for total mortality, and 7.82% (95% CI: −15.21, 37.10) or 6.14% (95% CI: −17.35, 36.31) for cardiovascular mortality. Using an 8-h maximum metric yielded similar interactive effects between ozone and temperature. For total and cardiovascular mortality, evidence indicates that low temperature has a stronger modifying effect on the 1-h maximum and the maximum 8-h average O3 concentrations than the 24-h average O3 concentrations. The interaction pattern of O3 with temperature was less sensitive at different cut-offs for temperature levels and including co-pollutants in the models. Our findings provide evidence that low daily average temperature significantly modifies the acute effects of ozone on daily mortality for both the 1-h maximum metric and the maximum 8-h average metric in Suzhou, China.