Health-related effects of education level: a Mendelian randomization study

Background A deeper understanding of the causal links from education level to health outcomes may shed a light for disease prevention at a novel and efficient perspective. Methods We conducted a wide-angled Mendelian randomization to disentangle the causal role of education level from intelligence for 31 health outcomes and explore to what extent body mass index and smoking mediate the associations. Univariable and multivariable Mendelian randomization analyses were performed. Results Genetically higher education level was associated with lower risk of major mental disorders and most somatic diseases independent of intelligence. The intelligence-adjusted odds ratios for each additional standard deviation of education (4.2 years) were 0.48 (0.37, 0.62) for suicide attempts, 0.50 (0.36, 0.68) for large artery stroke, 0.51 (0.42, 0.63) for heart failure, 0.52 (0.42, 0.65) for lung cancer, 0.45 (0.33,0.61) for rheumatoid arthritis, and 0.48 (0.43, 0.55) for type 2 diabetes. Higher education level adjusted for intelligence was additionally associated with lower risk of insomnia, major depressive disorder, stroke, coronary artery disease, breast cancer, ovarian cancer and gout but with higher risk of obsessive-compulsive disorder, anorexia nervosa, bipolar disorder and prostate cancer. Moreover, higher education level was associated with modifiable health-related risk factors in a favorable manner. Adjustment for body mass index and smoking attenuated the associations between education level and several outcomes, especially for type 2 diabetes and heart failure. High education level exerts causal protective effects on major somatic diseases. Conclusions These findings emphasize the importance of education to reduce the burden of common diseases.