Mortality from Particulate Matter 2·5 in India: National Prospective Proportional Mortality Study
58 Pages Posted: 18 Oct 2019More...
Background: Direct studies on the extent to which long-term exposure to ambient particulate matter (PM) of 2·5 µm diameter causes adult mortality in India are lacking, despite over 99% of Indians being exposed to levels that the World Health Organisation (WHO) considers unsafe.
Methods: We conducted a prospective proportional mortality study within the Million Death Study (MDS) to provide the first-ever direct national estimates of chronic respiratory disease, heart attack and stroke mortality from exposure to particulate matter (PM2·5) in India from 2001-13. We calculated odds ratios (ORs) for three-year annual median PM2·5 exposures measured 1-3 years earlier (and the closest three-year range or 4-6 years earlier) in about 14 000 small villages or urban blocks, adjusted for individual age, year of death, smoking status, rural/urban residency, and area-level female illiteracy. ORs adjusted for spatial clustering.
Findings: PM2.5 exposure levels increased from 2000 to 2014, particularly in North India. Among 77 976 cases and 38 283 injury controls at ages 15-69 years, the excess mortality risks per 10 unit µg/m3 increase in PM2·5 exposure were 2% and 3% for chronic respiratory disease and heart attack, respectively, but 22% for stroke in the northeast states. Risks were similar for each sex, above 70 years and in various sensitivity analyses, including in areas reporting lower solid fuel use. Adjustment for spatial clustering reduced the risks for chronic respiratory and heart attack, but not for stroke.
Interpretation: If these modest risks are causal, then PM2·5 exposure accounted for 225 000 deaths from respiratory and vascular disease at ages 15-69 years in 2015 in India, but with considerable uncertainty (range 16 000 to 571 000 deaths). The modest excess risks suggest suitable caution in interpreting the direct MDS evidence, and even more caution in extrapolating the WHO's current model-based risks to India.
Funding Statement: Supported by the Indian Council of Medical Research, Canadian Institutes of Health Research Foundation grant (FDN 154277), the University of Toronto, the National Institutes of Health grants (1R01TW007939).
Declaration of Interests: The authors declare no competing interests.
Ethics Approval Statement: Not required.
Keywords: particulate matter, PM 2.5, mortality, epidemiology, geospatial
Suggested Citation: Suggested Citation