Retinopathy is more common, but age-related macular degeneration carries more visual impairment: a cross-sectional study of US adults graded from the same fundus photographs

Background Retinopathy and age-related macular degeneration (AMD) are the most common findings on fundus photographs of older adults, yet their frequency, correlates and visual consequences have been estimated in separate samples. We compared them in a national survey that graded both from the same images. Methods We analyzed 5593 adults aged >=40 years in the US National Health and Nutrition Examination Survey 2005-2008 with photographs gradable for both conditions. We estimated survey-weighted prevalence, fitted identically specified logistic models for retinopathy (participants with diabetes) and AMD (full sample), and compared presenting visual impairment (better-eye acuity worse than 20/40) and self-rated vision across four groups: neither condition, retinopathy only, AMD only, and both. Results Retinopathy was more prevalent than AMD (9.54% [95% confidence interval (CI) 8.52-10.56] vs 6.52% [5.54-7.50]); AMD overtook it only from age 70 years. In participants with diabetes, retinopathy was associated with HbA1c (odds ratio [OR] 1.54 per percentage point; 1.33-1.78) and male sex (1.54; 1.10-2.17). AMD was associated with age (OR 1.09 per year; 1.07-1.11) and was less common in non-Hispanic Black adults (0.48; 0.33-0.71). Presenting impairment affected 4.91% of participants with neither condition, 7.09% with retinopathy only, 14.57% with AMD only and 26.15% with both. After age adjustment, impairment remained associated with AMD only (OR 2.23; 1.54-3.23) and both conditions (4.09; 2.28-7.37) but not retinopathy only (1.32; 0.91-1.90). Neither retinopathy nor diabetes was associated with AMD. Self-rated vision agreed poorly with measured impairment (kappa = 0.15): only 12% of impaired participants rated their vision poor or very poor. Conclusions Retinopathy is the more frequent retinal finding in US adults aged >=40 years, but AMD carries more of the measured visual impairment, independently of age. Grading diabetic screening photographs for AMD, and measuring acuity rather than asking about vision, would reach more people who see poorly. Keywords Retinopathy; Diabetic retinopathy; Age-related macular degeneration; Visual impairment; Self-reported vision; Prevalence; NHANES; Cross-sectional studies