Smoking at age 18–20 and suicide during 26 years of follow-up—how can the association be explained?

Authors:
Open Access
Publisher website
Google Scholar
Abstract
Objectives Several studies have shown a relation between smoking and suicide. It is not clear if this relation should be considered causal, or if other risk factors for suicide, left unmeasured in many studies, might explain the smoking–suicide association. The aim of this cohort study was to analyse the role of smoking and other potential risk factors measured in adolescence on subsequent suicide. Methods Information on smoking, and other potential risk factors from childhood and adolescence was collected among 49 323 men, born 1949–1951, at conscription for compulsory military training in the years 1969–1970. Mortality data were obtained from the Swedish cause of death register for the years 1971–1996. Results There was a clear trend of increasing suicide risk with increasing intensity of smoking reported at conscription. Those smoking >20 cigarettes/day had a strongly increased relative risk of suicide (OR = 3.03, 95% CI: 1.72, 5.34) during the first 13 years of follow-up. From 14 to 26 years after conscription, the risk was only slightly weaker (OR = 2.53, 95% CI: 1.36, 4.72). When we adjusted for other risk factors for suicide measured at the conscription examination (psychiatric diagnosis, parental divorce, low emotional control, medication for nervous problems, contact with police and childcare, heavy alcohol consumption, drug use, and education) in a multivariate analysis, the trend with smoking intensity disappeared, and there was no longer an increased relative risk for the heavy smokers (follow-up years 1–13: OR = 0.98, 95% CI: 0.53, 1.82; years 14–26: OR = 1.31, 95% CI: 0.91, 1.87). Conclusion The increased risk of suicide among smokers was almost entirely explained by an increased prevalence of heavy alcohol consumption and low mental well-being among smokers. That is, the association between smoking and suicide is probably due to confounding by these other factors. These results do not support the hypothesis that tobacco consumption itself is a risk factor for suicide.

This publication has 14 references indexed in Scilit: