The early life course-related traits with three psychiatric disorders: A two-sample Mendelian randomization study.

Objectives: Several studies have indicated a potential association between early life course-related traits and neurological and psychiatric disorders in adulthood, but the causal link remains unclear.

Methods: Instrumental variables (IVs) that have been shown to be strongly associated with exposure were obtained from summary data of genome-wide association studies (GWASs). Four early life course-related traits [i.e., birthweight (BW), childhood body mass index (BMI), early body size, and age at first birth (AFB)] were used as exposure IVs to estimate their causal associations with three neurological and psychiatric diseases [i.e., Alzheimer's disease (AD), major depressive disorder (MDD), and attention-deficit hyperactivity disorder (ADHD)]. Four different statistical methods, i.e., inverse-variance weighting (IVW), MR-Egger (MRE), weighted median (WM), and weighted mode (Wm), were performed in our MR analysis. Sensitivity analysis was performed by using the leave-one-out method, and horizontal pleiotropy was assessed using the MR-PRESSO package.

Results: There was evidence suggesting that BW has a causal effect on AD (OR = 1.05,  = 1.14E-03), but this association was not confirmed multivariable Mendelian randomization (MVMR) (OR = 0.97, 95% CI 0.92-1.02,  = 3.00E-01). A strong relationship was observed between childhood BMI and ADHD among both sexes; a 1-SD increase in BMI significantly predicted a 1.46-fold increase in the OR for ADHD ( = 9.13E-06). In addition, a similar relationship was found between early life body size and ADHD (OR = 1.47,  = 9.62E-05), and this effect was mainly driven by male participants (OR = 1.50,  = 1.28E-3). Earlier AFB could significantly predict a higher risk of MDD (OR = 1.19,  = 1.96E-10) and ADHD (OR = 1.45,  = 1.47E-15). No significant causal associations were observed between the remaining exposures and outcomes.

Conclusion: Our results reveal the adverse effects of childhood obesity and preterm birth on the risk of ADHD later in life. The results of MVMR also show that lower BW may have no direct relationship with AD after adjusting for BMI. Furthermore, AFB may predict a higher risk of MDD.