Abstract
PURPOSE: To investigate the association of subjects with refractive error and diabetic retinopathy (DR) in the United States comparing results between different race groups. METHODS: All data were derived from National Health and Nutrition Examination Survey (NHANES) from 2005 to 2008. The data were divided into four groups (emmetropia, mild myopia, high myopia, hypertropia) according to the spherical equivalent (SE), and those who met the enrollment conditions were selected as the study subjects. Multivariable logistic regression analysis was used to evaluate the relationship between refractive error and diabetic retinopathy risk. RESULTS: A total of 1317 participants were included in the study, including 331 participants with diabetic retinopathy, and 986 without diabetic retinopathy. After adjustment for potential confounders, subjects with high myopia were associated with a lower risk of diabetic retinopathy. The odds ratio (OR) was 0.44, 95% confidence interval (CI): (0.20-0.96), P-value = 0.040 in the multivariate regression analysis. Subgroup analyses showed that subjects with high myopia in the non-Hispanic Black group were associated with decreased odds of diabetic retinopathy. (OR was 0.20, and 95% CI: 0.04-0.95, P-value = 0.042). CONCLUSION: The results show that high myopia is associated with diabetic retinopathy in diabetic patients.