Genetically Predicted Higher Educational Attainment Decreases the Risk of COVID-19 Susceptibility and Severity: A Mendelian Randomization Study

Abstract

Background: Prior observational studies indicated that lower educational attainment (EA) is associated with higher COVID-19 risk, while these findings were vulnerable to bias from confounding factors. We aimed to clarify the causal effect of EA on COVID-19 susceptibility, hospitalization, and severity using Mendelian randomization (MR). Methods: We identified genetic instruments for EA from a large genome-wide association study (GWAS) (n = 1,131,881). Summary statistics for COVID-19 susceptibility (112,612 cases and 2,474,079 controls), hospitalization (24,274 cases and 2,061,529 controls), and severity (8,779 cases and 1,001,875 controls) were obtained from the COVID-19 Host Genetics Initiative. We used the single-variable MR (SVMR) and the multivariable MR (MVMR) controlling intelligence, income, body mass index, vigorous physical activity, sedentary behavior, smoking, and alcohol consumption to estimate the total and direct effects of EA on COVID-19 outcomes. Inverse variance weighted was the primary analysis method. All the statistical analyses were performed using R software. Results: Results from the SVMR showed that genetically predicted higher EA was correlated with a lower risk of COVID-19 susceptibility [odds ratio (OR) 0.86, 95% CI 0.84-0.89], hospitalization (OR 0.67, 95% CI 0.62-0.73), and severity (OR 0.67, 95% CI 0.58-0.79). EA still maintained its effects in most of the MVMR. Conclusion: Educational attainment is a predictor for susceptibility, hospitalization, and severity of COVID-19 disease. Population with lower EA should be provided with a higher prioritization to public health resources to decrease the morbidity and mortality of COVID-19.

Keywords: COVID-19; Mendelian randomization; educational attainment; severity; susceptibility.

Figure 1

The SVMR forest plot of EA on COVID-19 susceptibility using SNPs from (A) all the cohorts and (B) all the cohorts excluding 23andMe. COVID-19, coronavirus disease 2019; MR, Mendelian randomization; OR, odds ratio; SNPs, single-nucleotide polymorphisms; IVW, inverse-variance weighted; SVMR, single-variable MR; EA, educational attainment.

Figure 2

The MVMR forest plot of EA on COVID-19 susceptibility controlling intelligence, income, body mass index, vigorous physical activity, sedentary behavior, smoking, and alcohol consumption separately. COVID-19, coronavirus disease 2019; OR, odds ratio; MVMR, multivariable Mendelian randomization; EA, educational attainment.

Figure 3

The SVMR forest plot of EA on COVID-19 hospitalization using SNPs from (A) all the cohorts and (B) all the cohorts excluding 23andMe. COVID-19, coronavirus disease 2019; MR, Mendelian randomization; OR, odds ratio; SNPs, single-nucleotide polymorphisms; IVW, inverse-variance weighted; SVMR, single-variable MR; EA, educational attainment.

Figure 4

The MVMR forest plot of EA on COVID-19 hospitalization controlling intelligence, income, body mass index, vigorous physical activity, sedentary behavior, smoking, and alcohol consumption separately. COVID-19, coronavirus disease 2019; OR, odds ratio; MVMR, multivariable Mendelian randomization; EA, educational attainment.

Figure 5

The SVMR forest plot of EA on COVID-19 severity using SNPs from (A) all the cohorts and (B) all the cohorts excluding 23andMe. COVID-19, coronavirus disease 2019; MR, Mendelian randomization; OR, odds ratio; SNPs, single-nucleotide polymorphisms; IVW, inverse-variance weighted; SVMR, single-variable MR; EA, educational attainment.

Figure 6

The MVMR forest plot of EA on COVID-19 severity controlling intelligence, income, body mass index, vigorous physical activity, sedentary behavior, smoking, and alcohol consumption separately. COVID-19, coronavirus disease 2019; OR, odds ratio; MVMR, multivariable Mendelian randomization; EA, educational attainment.