Share this article on:

Socioeconomic Position and Lung Cancer Risk: How Important is the Modeling of Smoking?

Matukala Nkosi, Thomasa,b; Parent, Marie-Élisea,b,c; Siemiatycki, Jacka,b,c; Rousseau, Marie-Claudea,b,c

doi: 10.1097/EDE.0b013e31824d0548

Background: Although it has been reported that low socioeconomic position (SEP) is associated with lung cancer, the extent to which this reflects SEP differences in cigarette smoking is unclear. We investigated how various modeling approaches for smoking might influence this observed association.

Methods: We used data from a case-control study conducted in Montreal, Canada (1996−2002), comprising 1203 subjects with incident lung cancer and 1513 population controls. SEP was measured by census-based and self-reported income, residential value, education level, and occupational class. Odds ratios (ORs) and 95% confidence intervals (CIs) were estimated using logistic regression, and Akaike's Information Criterion (AIC) was used to compare model fit.

Results: Associations were observed between SEP indicators and lung cancer, but gradually disappeared with more comprehensive adjustment for smoking. For comparisons of the highest to lowest categories of census-based income, the OR for lung cancer was 0.58 (95% CI = 0.32−1.05) when adjusting only for smoking status (never, former, current), but 0.97 (0.51−1.86) when adjusting for smoking status, cigarette-years, and time since cessation. For comparisons of highest to lowest levels of education, the ORs for lung cancer were 0.50 (0.38−0.65) and 0.76 (0.57−1.02), when making the least and most comprehensive adjustments for smoking, respectively. Similarly, comparing highly skilled with unskilled manual workers, the ORs were 0.78 (0.54−1.12) and 1.00 (0.68−1.47), respectively. With thorough smoking adjustment, associations between SEP indicators and lung cancer virtually disappeared, and SEP did not improve model fit.

Conclusions: Previously reported associations of SEP with lung cancer may be attributable to incomplete adjustment for smoking. Our findings underline the importance of adjusting for several dimensions of smoking behavior to make correct inferences.

Supplemental Digital Content is available in the text.

From the aDepartment of Social and Preventive Medicine, Université de Montréal, Montréal, QC, Canada; bEpidemiology and Biostatistics Unit, INRS-Institut Armand-Frappier, Laval, QC, Canada; and cResearch Center of the Université de Montréal Hospital Center, Montréal, QC, Canada.

Submitted 13 July 2010; accepted 10 November 2011; posted online 13 March 2012.

Supported by research and personnel support grants from Health Canada, the Canadian Cancer Society, the Institut de recherche en santé et sécurité au travail du Québec, the Fonds de la recherche en santé du Québec (FRSQ), and the Canadian Institutes of Health Research (CIHR). M.C.R. was the recipient of a New Investigator Award from the CIHR and a Salary Award from the FRSQ. M.É.P. is the recipient of a Senior Investigator award from the FRSQ. J.S. holds the Guzzo-Cancer Research Society Chair in Environment and Cancer. The authors reported no other financial interests related to this research.

Supplemental digital content is available through direct URL citations in the HTML and PDF versions of this article ( This content is not peer-reviewed or copy-edited; it is the sole responsibility of the author.

Correspondence: Marie-Claude Rousseau, Epidemiology and Biostatistics Unit, INRS-Institut Armand-Frappier, 531 boulevard des Prairies, Laval (Québec), Canada H7V 1B7. E-mail:

© 2012 Lippincott Williams & Wilkins, Inc.