Mendelian randomization study of birth weight and risk of psychiatric disorders later in life

Abstract

Low birth weight has been associated with a higher risk of psychiatric disorders later in life. The underlying causal mechanisms of this relationship are however not clear. In this study, we investigate whether variation in fetal growth has a direct causal effect on mental health. Using birth weight as a proxy measure for fetal growth, we first assessed associations between observed birth weight and later diagnosis of psychiatric disorders in the Danish iPSYCH and ANGI-DK cohorts. Next, we constructed a polygenic score for birth weight based on >1 million variants and tested for association with psychiatric disorders. Finally, using 86 single-nucleotide polymorphisms with robust fetal-only genetic associations with birth weight, we assessed the causal relationship of genetically mediated fetal growth and psychiatric disorders using Mendelian randomization analyses. We found that higher observed birth weight was associated with lower risk of several psychiatric disorders. Polygenic score analyses supported this pattern for attention deficit/hyperactivity disorder, where an increase of one standard deviation in the score for birth weight corresponded to an odds ratio of 0.85 (95% confidence interval 0.79-0.92, P=6.27x10-5). However, one- and two-sample Mendelian randomization analyses did not indicate a direct causal relationship between the birth weight of children and their risk of psychiatric disorders. In conclusion, our study does not support a direct causal effect of fetal growth (as proxied by birth weight) on the risk of psychiatric disorders later in life, suggesting that the observed association between birth weight and mental health is likely to be caused by other factors.