Cigarette smoking and pancreatic cancer risk in 83 500 Norwegian men and women

Abstract

Objective: To investigate the association between cigarette smoking and the risk of developing pancreatic cancer and if the risk differed for male- and female smokers.

Methods: For our statistical analysis, using data from the Norwegian Counties Study, Cox proportional hazards model was used to obtain age-adjusted and multivariate hazard ratios with 95% confidence intervals for obtaining risk estimates for smoking and levels of smoking exposure. The multivariate analysis included covariates; age, education level, body mass index and physical activity level.

Results: In multivariate analysis, the overall smoking associated risk of pancreatic cancer compared to never smokers for both sexes were similar to that of the age adjusted analysis. For the exposure variables (initiation-age, consumption, duration and cumulative amount) total years of smoking indicates the strongest association for male smokers (HR = 2.82, 95% CI = 1.87-4.24). For female smokers, early smoking initiation (≤ 19 years) gave the strongest association (HR = 3.28, 95% CI = 2.22-4.82). Female smokers showed linear trends across the different exposure categories (Ptrend < 0.05) Male smokers, had no observed linear trend for any of the exposure categories.

Conclusion: Our findings conclude that there is a significant increased risk of pancreatic cancer associated with cigarette smoking. Further, we find that female smokers have a dose-response relationship between smoking and pancreatic cancer risk, which is not evident for men.