Background: Few studies have investigated cumulative effects and mortality displacement of short-term air pollution exposure on deaths; therefore, uncertainty remains regarding its public health significance.
Methods: We obtained air pollution and daily cause-specific mortality data of Nanjing from January 2004 to December 2019, covering a period of 16 years. We performed a time-series analysis with single-day, 2-day moving average, and distributed lag models, respectively, to estimate the effects of PM2.5, PM10, NO2, and SO2 exposure on total cardiovascular disease, ischaemic heart disease (IHD), and cerebrovascular disease (CBVD) mortality. Distributed lag models were used to assess the roles of cumulative exposure and mortality displacement.
Results: Cumulative effect estimates for 0-7 lag days were more considerable than estimates for single-day lags and 2-day moving average lag. The cumulative effect estimates for PM10, NO2 and SO2 on total cardiovascular and CBVD mortality became essentially zero within 30 days, which suggested the existence of mortality displacement. But the cumulative effect estimates for PM2.5 and SO2 on IHD mortality remained elevated and statistically significant within 27 (2.11%; 95% CI: 0.12, 4.27%) and 22 (2.63%; 95% CI: 0.39, 4.91%) days, respectively, which suggested the absence of mortality displacement.
Conclusions: Our results indicated that risk assessment based on single-day or 2-day moving average exposure rather than cumulative exposure likely underestimate the adverse effects of air pollution. The cumulative PM2.5 and SO2 exposure for nearly a month may have adverse effects on IHD mortality.
Keywords: Air pollution; Cardiovascular mortality; Distributed lag model; Fine particulate matter; Mortality displacement.
Copyright © 2020 Elsevier Ltd. All rights reserved.