Determinants of body mass index during early life: findings from an exposome-wide association study with follow-up replication and Mendelian randomization analyses

Background
Studies of factors influencing body mass index (BMI) mostly focus on hypothesized/established factors. We aimed to explore potential factors influencing BMI across childhood and adolescence using a hypothesis-free exposome-wide association study (ExWAS), with external replication and Mendelian randomization (MR).

Methods
The ExWAS was done in the Avon Longitudinal Study of Parents and Children (ALSPAC) using univariable regressions to estimate prospective associations of 2,523 exposures with BMI in infancy (up to 12 months, N = 12 761), early-childhood (13–60 months, N = 12 138), middle-childhood (61 months to 12 years, N = 9,105) and adolescence (13–18 years, N = 7379). Associations passing Bonferroni correction were followed-up with confounder-adjusted regression in ALSPAC. Replication of confounder-adjusted associations were explored in eight independent cohorts and/or MR analyses in ALSPAC.

Results
The ExWAS identified 8, 16, 66 and 82 exposures prospectively associated with BMI in infancy, early-childhood, middle-childhood, and adolescence, respectively. Of these, 8, 11, 42, and 26 remained associated following confounder adjustment. Replicated associations included maternal BMI, smoking and fat intake; only maternal smoking was supported by MR. Associations of more TV viewing time with higher BMI at one or more ages was supported by MR showing more time spent watching TV in early-childhood leads to higher BMI in middle-childhood. It was not possible to explore replication or MR for confounder-adjusted associations of offspring emotion, sociability, and peer problems with subsequent BMI.

Conclusions
Our ExWAS suggests more TV viewing time in early-childhood increases BMI in mid-childhood. Novel potential effects of social, emotional and peer exposures in childhood need replication and assessment of causality.