Abstract
BACKGROUND: Studies evaluating depression's role in lung cancer risk revealed contradictory findings, partly because of the small number of cases, short follow-up periods, and failure to account for key covariates including smoking exposure. We investigated the association of depressive symptoms with lung cancer risk in a large prospective cohort over 24 years while considering the role of smoking. METHODS: Women from the Nurses' Health Study completed measures of depressive symptoms, sociodemographics, and other factors including smoking in 1992 (N = 42 913). Depressive symptoms were also queried in 1996 and 2000, whereas regular antidepressant use and physician-diagnosed depression were collected starting in 1996. Multivariable Cox regression models estimated hazard ratios (HRs) and 95% confidence intervals (CIs) of lung cancer risk until 2016. RESULTS: We identified 1009 cases of lung cancer. Women with the highest v. lowest level of depressive symptoms had an increased lung cancer risk (HR(sociodemographics-adjusted) = 1.62, 95% CI 1.34-1.95; HR(fully-adjusted) = 1.25, 95% CI 1.04-1.51). In a test of mediation, lifetime pack-years of smoking accounted for 38% of the overall association between depressive symptoms and disease risk. When stratifying by smoking status, the elevated risk was evident among former smokers but not current or never smokers; however, the interaction term suggested no meaningful differences across groups (p = 0.29). Results were similar or stronger when considering time-updated depression status (using depressive symptoms, physician diagnosis, and regular antidepressant use) and chronicity of depressive symptoms. CONCLUSIONS: These findings suggest that greater depressive symptoms may contribute to lung cancer incidence, directly and indirectly via smoking habits, which accounted for over a third of the association.