The effect of melatonin supplement use on pubertal timing: target trial emulation in the Adolescent Brain Cognitive Development study.

Abstract

Sustained melatonin supplement use may delay pubertal onset, but evidence is limited. In the Adolescent Brain Cognitive Development study, we assessed if melatonin use for 38 months affected the timing of pubertal onset in males (N=3,134) and menarche in females (N=4,424). Pubertal outcomes were parent-reported using the Pubertal Development Scale. We emulated sequential target trials to evaluate the effect of initiating and continuing melatonin supplement use. Findings were contrasted with comparisons of ever- to never-users and initiators to non-initiators using Cox models. In sequentially emulated trials with 1,037,709 person-months among males and 1,818,084 person-months among females, there were 1,872 and 3,377 instances of melatonin use initiation among 186 males and 333 females, respectively. Observational analogues of intention-to-treat and per-protocol effects from target trial emulation showed no difference in risk of pubertal onset in males (end-of-follow-up per-protocol RR=1.00, 95%CI=0.87,1.13) or menarche in females (RR=0.93, 95%CI=0.80,1.07). By contrast, Cox models suggested a delay in menarche among 335(7.74%) melatonin ever-users after adjustment for child, family, and neighborhood characteristics collected at the enrollment visit (HR=0.80, 95%CI=(0.69,0.94)). In large samples, rigorous causal analyses that aligned eligibility criteria with treatment initiation and adjusted for time-varying confounding showed no effect of melatonin supplement use on pubertal timing.