Introduction Recreational physical activity (PA) has been associated with reduced lung cancer risk. Occupational-related PA may be one of the most important sources of PA, but the few studies that have examined the occupational PA-lung cancer relationship have produced inconsistent result.

Objective To investigate the association between occupational PA and lung cancer risk.

Methods A case-cohort study was nested among four cohorts of the Canadian Partnership for Tomorrow’s Health Project. Through linkage of each cohort to their respective provincial cancer registry, 453 incident lung cancer cases diagnosed between 2009 and 2016 were identified; a random sub-cohort of 2,435 participants was sampled at baseline. Data on the longest-held job and risk factors were gathered via a harmonized questionnaire. Using a database generated by our team on the energy expenditures associated with almost 3,600 job titles, metabolic equivalent of tasks (METs) were assigned to the longest-held job of each participant. In total, 335 cases (139 men; 196 women) and 2,320 sub-cohort members (879 men; 1,441 women) with assignable job codes and MET values were included in this analysis. Multivariable logistic regression models, with normal or robust variance estimators, were used to estimate study-specific odds ratios (ORs) and 95% confidence intervals (CIs) for the relationship between occupational PA (categorized in tertiles) and lung cancer risk, while adjusting for smoking and lung cancer risk factors identified using directed acyclic graphs. The ORs were then pooled using a random-effects model.

Results Compared with low occupational PA as reference, subjects with medium and high occupational PA had ORs of 0.64 (95% CI: 0.38–1.07) and 0.73 (95% CI: 0.39–1.34), respectively. The ORs were similar by sex and by smoking status (never vs. ever smokers).

Conclusion Our findings suggest that, like recreational PA, occupational PA may protect against lung cancer risk.