Abstract
OBJECTIVE: To examine the association between prenatal exposure to benzodiazepines or Z-hypnotics and a spectrum of psychiatric disorders in children. DESIGN: Population based cohort study with sibling controlled analysis. SETTING: National Health Information Database of South Korea, 2009-23. PARTICIPANTS: All liveborn children between 2010 and 2022 were followed until 2023. Pregnancies exposed to benzodiazepines or Z-hypnotics were compared with unexposed pregnancies and with pregnancies in women with previous use of these drugs (past users). MAIN OUTCOME MEASURES: Overall and 12 specific psychiatric disorders in offspring. Propensity score overlap weighting was applied to balance covariates, and hazard ratios with 95% confidence intervals were estimated using Cox proportional hazards models. Sibling controlled analyses were conducted to account for shared familial factors. RESULTS: Among 3 809 949 liveborn children, 94 482 (2.5%) were exposed to benzodiazepines or Z-hypnotics during pregnancy, 3 715 467 were unexposed, and 147 307 were born to past users. During the follow-up period, a total of 10 060, 311 997, and 15 645 events occurred in the exposed, unexposed, and past user groups, respectively. Prenatal exposure was associated with a higher risk of psychiatric disorders compared with unexposed pregnancies and past users; however, this association was attenuated in the sibling controlled analysis (hazard ratio 0.99, 95% confidence interval 0.94 to 1.04). No increased risk was observed for individual psychiatric disorders. In subgroup analyses, modestly elevated hazard ratios were observed for exposure during the second half of pregnancy (sibling controlled hazard ratios: 1.27 (0.95 to 1.71) for benzodiazepine; 1.81 (0.57 to 5.74) for Z-hypnotic), during both the first and second half of pregnancy (sibling controlled hazard ratios: 1.35 (0.93 to 1.96) for benzodiazepine; 1.44 (0.93 to 2.21) for Z-hypnotic), and for 30 or more days of Z-hypnotic exposure (sibling controlled hazard ratio 1.31, 0.96 to 1.78), although the confidence intervals in sibling analyses were wide and included the null. CONCLUSIONS: In this large population based cohort, prenatal exposure to benzodiazepines or Z-hypnotics was not associated with an increased risk of psychiatric disorders in offspring after familial factors were accounted for. However, given the modest elevations in point estimates observed in certain subgroups-particularly with benzodiazepine and Z-hypnotic exposure during the latter half of pregnancy, as well as with exposure in both early and late pregnancy, and with longer durations of Z-hypnotic use-the potential for a slightly increased risk in these specific contexts could not be ruled out.