Lung cancer and cardiovascular disease mortality associated with ambient air pollution and cigarette smoke: shape of the exposure-response relationships

Abstract

Background: Lung cancer and cardiovascular disease (CVD) mortality risks increase with smoking, secondhand smoke (SHS), and exposure to fine particulate matter < 2.5 μm in diameter (PM₂.₅) from ambient air pollution. Recent research indicates that the exposure-response relationship for CVD is nonlinear, with a steep increase in risk at low exposures and flattening out at higher exposures. Comparable estimates of the exposure-response relationship for lung cancer are required for disease burden estimates and related public health policy assessments.

Objectives: We compared exposure-response relationships of PM₂.₅ with lung cancer and cardiovascular mortality and considered the implications of the observed differences for efforts to estimate the disease burden of PM2.5.

Methods: Prospective cohort data for 1.2 million adults were collected by the American Cancer Society as part of the Cancer Prevention Study II. We estimated relative risks (RRs) for increments of cigarette smoking, adjusting for various individual risk factors. RRs were plotted against estimated daily dose of PM₂.₅ from smoking along with comparison estimates for ambient air pollution and SHS.

Results: For lung cancer mortality, excess risk rose nearly linearly, reaching maximum RRs > 40 among long-term heavy smokers. Excess risks for CVD mortality increased steeply at low exposure levels and leveled off at higher exposures, reaching RRs of approximately 2-3 for cigarette smoking.

Conclusions: The exposure-response relationship associated with PM₂.₅ is qualitatively different for lung cancer versus cardiovascular mortality. At low exposure levels, cardiovascular deaths are projected to account for most of the burden of disease, whereas at high levels of PM₂.₅, lung cancer becomes proportionately more important.

Figure 1

Adjusted RRs [with 95% confidence intervals (CIs)] of lung cancer mortality (A) and IHD, cardiovascular, and cardiopulmonary mortality (B) plotted over estimated daily exposure of PM_2.5 (milligrams) and increments of cigarette smoking relative to never smokers (cigarettes/day). Diamonds represent comparative mortality risk estimates (with 95% CIs) for PM_2.5 from air pollution from the comparative studies (Dockery et al. 1993; Laden et al. 2006; Miller et al. 2007; Pope et al 1995, 2002, 2004). Stars represent comparable pooled RR estimates (with 95% CIs) associated with SHS exposure from comparative studies (Teo et al. 2006; U.S. Department of Health and Human Services 2006). The dotted lines represent the nonlinear power function fit through the origin and the estimates (including active smoking, SHS, ambient PM_2.5). Estimated doses from different increments of active smoking are dramatically larger than estimated doses from ambient air pollution or SHS; therefore, associations at lower exposure levels (due to ambient air pollution and SHS) are shown as insets with a magnified scale.