Ischemic Heart Disease Mortality and Diesel Exhaust and Respirable Dust Exposure in the Diesel Exhaust in Miners Study

Abstract

Diesel exhaust is a suggested risk factor for ischemic heart disease (IHD), but evidence from cohorts using quantitative exposure metrics is limited. We examined the impact of respirable elemental carbon (REC), a key surrogate for diesel exhaust, and respirable dust (RD) on IHD mortality, using data from the Diesel Exhaust in Miners Study in the United States. Using data from a cohort of male workers followed from 1948-1968 until 1997, we fitted Cox proportional hazards models to estimate hazard ratios for IHD mortality for cumulative and average intensity of exposure to REC and RD. Segmented linear regression models allowed for nonmonotonicity. Hazard ratios for cumulative and average REC exposure declined relative to the lowest exposure category before increasing to 0.79 and 1.25, respectively, in the highest category. Relative to the category containing the segmented regression change points, hazard ratios for the highest category were 1.69 and 1.54 for cumulative and average REC exposure, respectively. Hazard ratios for RD exposure increased across the full exposure range to 1.33 and 2.69 for cumulative and average RD exposure, respectively. Tests for trend were statistically significant for cumulative REC exposure (above the change point) and for average RD exposure. Our findings suggest excess risk of IHD mortality in relation to increased exposure to REC and RD.

Figure 1.

The shape of the exposure-response relationship between cumulative exposure to respirable elemental carbon (REC) and average intensity of respirable dust (RD) and ischemic heart disease among male workers first employed after the start of dieselization in the Diesel Exhaust in Miners Study Cohort, United States, beginning during 1948–1968 and completing follow-up in 1997. A) Hazard ratios (solid symbol) and 95% confidence intervals for categories of exposure to cumulative REC (μg/m³-years), adjusted for time-weighted average intensity of RD (mg/m³) and the fitted segmented linear regression model for cumulative REC (solid line). B) Hazard ratios (solid symbol) and 95% confidence intervals for categories of time-weighted average intensity of (RD (mg/m³), adjusted for exposure to cumulative REC (μg/m³-years) and the fitted linear regression model for average RD (solid line). Because reference categories included nonexposed and exposed workers, the regression model included adjustment to pass through the inverse-variance weighted mean hazard ratio for all categories.

Figure 2.

The shape of the exposure-response relationship between cumulative exposure to respirable dust (RD) and average intensity of respirable elemental carbon (REC) and ischemic heart disease among male workers first employed after the start of dieselization in the Diesel Exhaust in Miners Study Cohort, United States, beginning during 1948–1968 and completing follow-up in 1997. A) Hazard ratios (solid symbol) and 95% confidence intervals for categories of exposure to time-weighted average REC (μg/m³), adjusted for cumulative RD (mg/m³-years). B) Hazard ratios (solid symbol) and 95% confidence intervals for categories of exposure to cumulative RD (mg/m³-years) adjusted for time-weighted average REC (μg/m³) and the fitted linear regression model cumulative RD (solid line). Because reference categories included nonexposed and exposed workers, the regression model included adjustment to pass through the inverse-variance weighted mean hazard ratio for all categories.