Policy-driven changes in the health risk of PM2.5 and O3 exposure in China during 2013–2018

China issued a series of control measures to mitigate PM2.5 pollution, including long-term (i.e., Air Pollution Prevention and Control Action Plan, APPCAP) and short-term (emergency measures in autumn and winter) acts. However, the O3 concentration increased significantly as PM2.5 levels sharply decreased when these measures were implemented. Therefore, the policy-driven positive/negative health effects of PM2.5/O3 need to be comprehensively estimated. The health impact function (HIF) is applied to evaluate the health burden attributable to long- and short-term PM2.5 and O3 exposure. The results show that the PM2.5 concentration decreased by 42.95% in 74 cities, whereas O3 pollution is increased by 17.56% from 2013 to 2018. Compared with 2013, the number of premature deaths attributable to long- and short-term PM2.5 exposure decreased by almost 5.31 × 104 (95% confidence interval [CI]: 2.87 × 104–4.71 × 104) (10.13%) and 3.00 × 104 (95% CI: 1.66 × 104–4.39 × 104) (72.49%), respectively, in 2018. In contrast, O3-attributable deaths, increased by 1.98 × 104 (95% CI: 0.31 × 104–3.59 × 104) (130.57%) and 0.91 × 104 (95% CI: 0.50 × 104–1.33 × 104) (76.16%) for long- and short-term exposure, respectively. The number of avoidable deaths attributed to PM2.5 reduction is larger than the level of premature deaths related to increasing O3. Although annual mean PM2.5 concentrations have fallen rapidly, the benefits of reducing long-term exposure are limited, whereas the deaths associated with acute exposure decrease more significantly due to the reduction of heavy-pollution days by implementing emergency measures. The results show appreciable effectiveness in protecting human health and illustrate that synchronous control of PM2.5 and O3 pollution should be emphasized.