A hypothesis-free approach to identifying potential effects of relative age in school year: an instrumental variable phenome-wide association study in the UK Biobank.

Abstract

A child's relative age within their school year ("relative age") is associated with educational attainment and mental health. However, hypothesis-driven studies often re-examine the same outcomes and exposure, potentially leading to confirmation and reporting biases and missing unknown effects. Hypothesis-free outcome-wide analyses can potentially overcome these limitations. We conducted a hypothesis-free investigation of the effects of relative age within school year. We performed an instrumental variable (IV) phenome-wide association study in the UK Biobank (participants aged 40-69 years at baseline), using the PHESANT software package. We created 2 IVs for relative age: being born in September vs August (n = 64 075) and week of birth (n = 383 309). Outcomes passing the Bonferroni-corrected P value threshold for either instrument were plotted to identify a discontinuity at the school year transition. Thirteen traits associated with at least 1 of the instruments showed a discontinuity. Previously identified effects included those with a younger relative age being less likely to have educational qualifications and more likely to have started smoking at a younger age. We detected a few associations not explored by previous studies. For example, those of younger relative age had better lung function as adults. Hypothesis-free approaches could help address confirmation and reporting biases in epidemiology.