LAUSR

OBJECTIVES We aimed to investigate the association between the use of various sedative-hypnotics and the incidence of overall and individual cancers in a large, population-based, retrospective cohort study. METHODS We selected a 5% random sample of individuals aged 50 years or older from data maintained by the Korean National Health Insurance Service for the years 2002-2015, excluding individuals with a prior diagnosis of cancer and with any sedative-hypnotic use in the initial two years of follow-up, leaving 236,759 participants for the final analysis. Exposure to sedative-hypnotics was defined by type of drug, standardized to a defined daily dose, and coded as a time-varying variable. Cox proportional hazard models were applied after adjusting for sex, socio-economic status, and comorbidities. RESULTS We observed increased risk for overall cancer among men and women who used sedative-hypnotics (hazard ratio (HR) = 1.07, 95% confidence interval (CI) = 1.01-1.13 for men; HR = 1.21, 95% CI = 1.09-1.25 for women) compared with non-users after full adjustment. In the fully adjusted model, women with any sedative-hypnotic use had significantly increased risk for thyroid (HR = 1.53, 95% CI = 1.24-1.87), breast (HR = 1.29, 95% CI = 1.04-1.61), ovarian (HR = 1.65, 95% CI = 1.10-2.46), and lung cancer (HR = 1.40, 95% CI = 1.17-1.69) compared with non-users. Men with sedative-hypnotic use had increased risk for prostate (HR = 1.36, 95% CI = 1.16-1.58), brain (HR = 1.67, 95% CI = 1.04-2.69), and lung cancer (HR = 1.20, 95% CI = 1.07-1.35) compared with non-users. CONCLUSION We found a significant increase in overall cancer incidence among participants who used sedative-hypnotics, and both male and female sedative-hypnotic users had significantly increased risk for certain types of cancer.