Air pollution, blood pressure, and their long-term associations with mortality.

This article addresses the importance of blood pressure as a covariate in studies of long-term associations between air quality and mortality. We focus on a cohort of about 50,000 U.S. veterans who had been diagnosed as hypertensive at some time and whose survival rates were predicted by blood pressure (BP) and ambient air quality, among other factors. The relationship between BP and air quality is considered by reviewing the literature, by deleting variables from the proportional hazards regression model, and by stratifying the cohort by diastolic blood pressure (DBP) level. The literature review shows BP to be an important predictor of survival and finds small transient associations between air quality and BP that may be either positive or negative. The regression model sensitivity runs showed that the associations with air pollution are robust to the deletion of the BP variables, for the entire cohort. For stratified regressions, the confidence intervals for the air pollution-mortality associations overlap for the two DBP groups. We conclude that associations between mortality and air quality are not mediated through blood pressure, nor vice versa.