Greenness, air pollution, and mortality risk: A U.S. cohort study of cancer patients and survivors

Carver J Coleman; Ray A Yeager; Daniel W Riggs; Nathan C Coleman; George R Garcia; Aruni Bhatnagar; C Arden Pope

doi:10.1016/j.envint.2021.106797

Greenness, air pollution, and mortality risk: A U.S. cohort study of cancer patients and survivors

Environ Int. 2021 Dec:157:106797. doi: 10.1016/j.envint.2021.106797. Epub 2021 Jul 29.

Authors

Carver J Coleman¹, Ray A Yeager², Daniel W Riggs³, Nathan C Coleman⁴, George R Garcia¹, Aruni Bhatnagar³, C Arden Pope⁵

Affiliations

¹ Department of Economics, Brigham Young University, Provo, UT 84602, United States.
² Department of Environmental and Occupational Health Sciences, University of Louisville, Louisville, KY 40292, United States.
³ Department of Medicine, University of Louisville, Louisville, KY 40292, United States.
⁴ Bates White LLC, 2001 K St NW, North, Tower Suite 500, Washington, DC 20006, United States.
⁵ Department of Economics, Brigham Young University, Provo, UT 84602, United States. Electronic address: cap3@byu.edu.

PMID: 34332301
DOI: 10.1016/j.envint.2021.106797

Abstract

Background: Several studies suggest that living in areas of high surrounding greenness may be associated with a lower cardiopulmonary mortality risk. However, associations of greenness with specific causes of death in cancer patients and survivors has not been examined and it is unknown whether this relationship is affected by area levels of fine particulate matter air pollution (PM_2.5). This study evaluated associations between greenness and PM_2.5 on causes of death in a large, U.S.-based cohort of cancer patients and survivors.

Methods: Surveillance, Epidemiology and End Results (SEER) data were used to generate a cohort of 5,529,005 cancer patients and survivors from 2000 to 2016. Census-tract Normalized Difference Vegetation Index (NDVI) during May-October from 2003 to 2016 was population-weighted to act as a county-level greenness measure. County-level PM_2.5 exposure was estimated from annual concentrations averaged from 1999 to 2015. Cox Proportional Hazards models were used to estimate the association between greenness, PM_2.5, and cause-specific mortality while controlling for age, sex, race, and other individual and county level variables.

Findings: An IQR increase in greenness was associated with a decrease in cancer mortality for cancer patients (Hazard ratio of 0.94, 95% CI: 0.93-0.95), but not for cardiopulmonary mortality (0.98, 95% CI: 0.96-1.00). Inversely, an increase in 10 μg/m³ PM_2.5 was associated with increased cardiopulmonary mortality (1.24, 95% CI: 1.19-1.29), but not cancer mortality (0.99, 95% CI: 0.97-1.00). Hazard ratios were robust to inclusion of PM_2.5 in models with greenness and vice versa. Although exposure estimates were constant over most stratifications, greenness seemed to benefit individuals diagnosed with high survivability cancers (0.92, 95% CI: 0.90-0.95) more than those with low survivability cancers (0.98. 95% CI: 0.96-0.99).

Interpretation: Higher levels of greenness are associated with lower cancer mortality in cancer patients. The evidence suggests minimal confounding between greenness and PM_2.5 exposures and risk of mortality.

Keywords: Air pollution; Cancer; Greenness; PM(2.5); Survival.

Publication types

Research Support, N.I.H., Extramural
Research Support, U.S. Gov't, Non-P.H.S.

MeSH terms

Air Pollutants* / analysis
Air Pollution* / analysis
Air Pollution* / statistics & numerical data
Cohort Studies
Environmental Exposure / analysis
Environmental Exposure / statistics & numerical data
Humans
Neoplasms*
Particulate Matter / analysis
Survivors

Substances

Air Pollutants
Particulate Matter