Body mass index, earnings and partnership: genetic instrumental variable analysis in two nationally representative UK samples

Amanda Hughes,Yanchun Bao,Melissa Smart,Meena Kumari
DOI: https://doi.org/10.1101/608588
2019-04-16
Abstract:Abstract In high-income countries there is an established link between high body mass index (BMI) and low income, but the direction of this association is unclear. Recent analyses in a large UK population using genetically-instrumented BMI supported a causal influence of BMI on household income, educational attainment and job class. Since analyses were based on an age-restricted and relatively wealthy population, it is unclear whether results are generalizable, and limited income data precluded decomposition of household income effects into own-income and partnership effects. Investigation is therefore warranted in more representative UK populations where associations may differ, and where individual and partner-based mechanisms can be studied separately. Data came from two nationally-representative samples, the UK Household Longitudinal Survey (UKHLS) and the English Longitudinal Study of Ageing (ELSA). Analysis was conducted in each sample, with results then pooled by meta-analysis. We used externally-weighted polygenic scores based on the latest genome-wide association study for BMI to examine the influence of genetically-instrumented BMI on earnings, probability of employment, job class conditional on working, likelihood of partnership, and partner’s earnings. A one-unit (kg/m 2 ) increase in genetically-instrumented BMI was associated with a roughly 9% decrease in own monthly earnings (pooled coefficient: 0.91, CI:0.86, 0.97) and lower probability of employment (OR: 0.89, CI:0.83, 0.96) or having a university degree (OR: 0.95, CI:0.90, 0.99). Employed individuals with higher genetically-instrumented BMI were less likely to have professional or managerial occupations (OR: 0.91, CI:0.86, 0.96). No associations were seen with partnership. A one-unit increase in BMI was associated with a 5% decrease in partners’ earnings, but estimates were imprecise (pooled coefficient: 0.95, CI:0.88,1.01). Results are consistent with a negative influence of body mass index on a range of labour market and educational outcomes for both men and women. Key Messages Higher genetically-instrumented BMI was associated with lower earnings and odds of working Also with lower odds of holding a managerial/professional occupation or a degree No associations were seen with probability of cohabiting partnership
What problem does this paper attempt to address?