Does higher education protect against obesity? Evidence using Mendelian randomization

Objectives. The aim of this explorative study was to examine the effect of education on obesity using Mendelian randomization. Methods. Participants (N = 2011) were from the on- going nationally representative Young Finns Study (YFS) that began in 1980 when six cohorts (aged 30, 33, 36, 39, 42 and 45 in 2007) were recruited. The average value of BMI (kg/m(2)) measurements in 2007 and 2011 and genetic information were linked to comprehensive register based information on the years of education in 2007. We first used a linear regression (Ordinary Least Squares, OLS) to estimate the relationship between education and BMI. To identify a causal relationship, we exploited Mendelian randomization and used a genetic score as an instrument for education. The genetic score was based on 74 genetic variants that genome- wide association studies (GWASs) have found to be associated with the years of education. Because the genotypes are randomly assigned at conception, the instrument causes exogenous variation in the years of education and thus enables identification of causal effects. Results. The years of education in 2007 were associated with lower BMI in 2007/2011 (regression coefficient (b) = -0.22; 95% Confidence Intervals [CI] = -0.29,-0.14) according to the linear regression results. The results based on Mendelian randomization suggests that there may be a negative causal effect of education on BMI (b = -0.84; 95% CI = -1.77, 0.09). Conclusion. The findings indicate that education could be a protective factor against obesity in advanced countries. (C) 2017 Elsevier Inc. All rights reserved.