The Lancet Planetary Health (Jan 2018)

Ambient carbon monoxide and cardiovascular mortality: a nationwide time-series analysis in 272 cities in China

  • Cong Liu, MPH,
  • Peng Yin, PhD,
  • Renjie Chen, PhD,
  • Xia Meng, PhD,
  • Lijun Wang, MPH,
  • Yue Niu, MPH,
  • Zhijing Lin, MPH,
  • Yunning Liu, MPH,
  • Jiangmei Liu, MPH,
  • Jinlei Qi, MPH,
  • Jinling You, MPH,
  • Haidong Kan, PhD,
  • Maigeng Zhou, PhD

DOI
https://doi.org/10.1016/S2542-5196(17)30181-X
Journal volume & issue
Vol. 2, no. 1
pp. e12 – e18

Abstract

Read online

Background: Evidence of the acute health effects of ambient carbon monoxide air pollution in developing countries is scarce and mixed. We aimed to evaluate short-term associations between carbon monoxide and daily cardiovascular disease mortality in China. Methods: We did a nationwide time-series analysis in 272 major cities in China from January, 2013, to December, 2015. We extracted daily cardiovascular disease mortality data from China's Disease Surveillance Points system. Data on daily carbon monoxide concentrations for each city were obtained from the National Urban Air Quality Real-time Publishing Platform. City-specific associations between carbon monoxide concentrations and daily mortality from cardiovascular disease, coronary heart disease, and stroke were estimated with over-dispersed generalised linear models. Bayesian hierarchical models were used to obtain national and regional average associations. Exposure–response association curves and potential effect modifiers were evaluated. Two-pollutant models were fit to evaluate the robustness of the effects of carbon monoxide on cardiovascular mortality. Findings: The average annual mean carbon monoxide concentration in these cities from 2013 to 2015 was 1·20 mg/m3, ranging from 0·43 mg/m3 to 2·45 mg/m3. For a 1 mg/m3 increase in average carbon monoxide concentrations on the present day and previous day (lag 0–1), we observed significant increments in mortality of 1·12% (95% posterior interval [PI] 0·42–1·83) from cardiovascular disease, 1·75% (0·85–2·66) from coronary heart disease, and 0·88% (0·07–1·69) from stroke. These associations did not vary substantially by city, region, and demographic characteristics (age, sex, and level of education), and the associations for cardiovascular disease and coronary heart disease were robust to the adjustment of criteria co-pollutants. We did not find a threshold below which carbon monoxide exposure had no effect on cardiovascular disease mortality. Interpretation: This analysis is, to our knowledge, the largest study done in a developing country, and provides robust evidence of the association between short-term exposure to ambient carbon monoxide and increased cardiovascular disease mortality, especially coronary heart disease mortality. Funding: Public Welfare Research Program.