The authors respond:
Thank you for the opportunity1 to further discuss the potential in our study for residual confounding by smoking and misclassification of exposure.2 Although these concerns1 have been raised regarding many epidemiologic studies of air pollution, it is unlikely that they are responsible for the positive associations we reported.
Numerous studies3 have found that self-reported measures produce valid estimates of smoking behavior. We found negative correlations between all smoking variables (ie, smoking pack-years, years since cessation, and residential and occupational second-hand smoke exposure) and air pollution exposures, which limits any positive bias in our results. While the potential for response and recall bias exists in all case-control studies, this population-based study has a relatively high response rate for cases (62%) and controls (67%). Furthermore, we found no difference between cases and controls in the completeness of the self-reported residential histories that we used to assign air pollution exposures.
Our long-term exposure assessment approach represents clear improvements over past studies. Specifically, we included complete residential histories over a 20-year period and applied multiple spatiotemporal models of PM2.5, NO2, and O3. While some degree of exposure misclassification is present, this error is likely nondifferential and thus would produce bias toward (rather than away from) the null. As reported,2 the increased lung cancer odds ratio for NO2 exposures derived from fixed-site monitors likely represents contributions from PM2.5 due to the high correlation of these two pollutants. Furthermore, all other sensitivity analyses using various spatiotemporal models revealed consistent associations. Thus, our study offers a useful contribution to the epidemiologic evidence regarding air pollution exposure and lung cancer incidence.
While we made no claims of a causal association in our article, we concur with recent commentaries4 and systematic reviews and meta-analyses5,6 that the current weight of evidence supports an association of PM2.5 and NO2 exposures with lung cancer incidence. Whether these associations are causal is the focus of the upcoming International Agency for Research on Cancer monograph evaluating the carcinogenicity of ambient air pollution.7 We do note, however, that the Environmental Protection Agency’s Integrated Science Assessment8 mentioned by Drs. Sax and Goodman concluded that the evidence is “suggestive of a causal relationship between long-term exposures to PM2.5 and cancer.” We also highlight that NO2 itself is not likely to be responsible for the increase in lung cancer risk but rather is a marker for other traffic-related carcinogens.
School of Population and Public Health, University of British Columbia, Vancouver, BC, Canada, email@example.com
Paul A. Demers
Occupational Cancer Research Centre, Cancer Care Ontario, Ontario, Canada
Kenneth C. Johnson
Department of Epidemiology and Community Health, University of Ottawa, Ottawa, Ontario, Canada
Richard M. Carpiano
Department of Sociology, University of British Columbia, Vancouver, BC, Canada
School of Population and Public Health, University of British Columbia, Vancouver, BC, Canada
1. Sax S, Goodman JE. Residential air pollution and lung cancer [letter]. Epidemiology. 2014;25:159
2. Hystad P, Demers PA, Johnson KC, Carpiano RM, Brauer M. Long-term residential exposure to air pollution and lung cancer risk. Epidemiology. 2013;24:762–772
3. Patrick DL, Cheadle A, Thompson DC, Diehr P, Koepsell T, Kinne S. The validity of self-reported smoking: a review and meta-analysis. Am J Public Health. 1994;84:1086–1093
4. Fajersztajn L, Veras M, Barrozo LV, Saldiva P. Air pollution: a potentially modifiable risk factor for lung cancer. Nat Rev Cancer. 2013;13:674–678
5. Chen H, Goldberg MS, Villeneuve PJ. A systematic review of the relation between long-term exposure to ambient air pollution and chronic diseases. Rev Environ Health. 2008;23:243–297
6. Health. Effects Institute. 2010. Traffic-Related Air Pollution: A Critical Review of the Literature on Emissions, Exposure, and Health Effects. Boston, MA The Health Effects Institute Available at: http://pubs.healtheffects.org/view.php?id=334
8. U.S. EPA. . Integrated Science Assessment for Particulate Matter (Final Report). 2009 Washington, DC U.S. Environmental Protection Agency EPA/600/R-08/139F;