Age-related Macular Degeneration (AMD) has long been thought to be more frequent in non-Hispanic whites than in blacks in the United States. This was based on observations reported in clinical case series. The use of new AMD classification schemes and severity scales in large population-based studies has confirmed that non-Hispanic whites have higher prevalences of late AMD, both neovascular AMD and geographic atrophy, than in non-Hispanic blacks and Hispanics. For example, in the Salisbury Eye Examination (SEE), the prevalence of neovascular AMD (1.7% vs 1.1%) and geographic atrophy (1.8% vs 0.3%) was higher in whites than in blacks. Age-specific prevalence of late AMD was consistently lower in Mexican Americans in the Los Angeles Latino Eye Study (LALES) than in whites in the Beaver Dam Eye Study (BDES) (eg, 70-79 years, 1.5% vs 3.7%; ≥80 years, 8.5% vs 9.5%). In the Multi-Ethnic Study of Atherosclerosis (MESA), the prevalence of late AMD was higher in Chinese Americans than in whites. Similar risk factors, for example older age, family history, and smoking, were associated with AMD regardless of race/ethnicity.

There are few population-based studies describing the incidence or progression of AMD in different racial/ethnic groups. These studies, using newer standardized grading schemes, have shown that while younger non-Hispanic blacks and Mexican Americans manifest early AMD with almost the same prevalence as in whites, they are less likely to develop late AMD than non-Hispanic whites. In the SEE, blacks were more likely than whites to develop early AMD after 2 years of follow-up examination but had similar risk of progression to more severe AMD. Over a 9-year period in the population-based Barbados Eye Study, the overall incidence rate of early AMD in blacks was 13% and that of late AMD was 0.7%, significantly lower than found in whites in the BDES, a white population of similar age. In the LALES, the 4-year incidence of both early (7.5%) and late (0.2%) AMD was significantly lower than the estimated incidence in non-Hispanic whites in the BDES. When the data were age-standardized, the overall incidence in Mexican Americans in the LALES population remained lower for early AMD (0.95 vs 1.73) and late AMD (0.11 vs 0.18) than in whites in the BDES.

Recognizing the need for more data regarding AMD in different racial/ethnic groups, VanderBeek and associates, in this issue of the Journal , report differences in the prevalence and incidence of nonexudative and exudative AMD between whites and non-Hispanic blacks, Hispanics, and Asian Americans using health care claims data from a large, national managed care network. While adjusting for confounders (but not smoking status), they found at age 60 years that non-Hispanic blacks had a 25% lower prevalence of having nonexudative and 30% lower prevalence of having exudative AMD coded in their medical claims record compared to whites. At age 80 years, the comparable prevalences were 44% and 55% that of whites. Similar comparisons for Hispanics demonstrated at age 60 years a similar risk for nonexudative AMD to that in whites but a 28% higher risk of exudative AMD than whites, while at age 80 years there was an 18% reduced lower risk for nonexudative AMD and a similar risk of incident exudative AMD in Hispanics compared to that found in whites. Compared to whites, Asian Americans showed a 28% increased risk for nonexudative AMD at 60 years but a 46% decreased risk for exudative AMD at age 80 years. The authors are well aware of the significant limitations in using claims data to estimate prevalence and incidence in making these and other estimates that appear in their paper. Readers of this and other epidemiologic papers must be cautious when interpreting the comparisons between racial/ethnic groups; studies differ in methods of ascertainment of the study cohort, in participation, and in ascertainment and definitions of the presence and severity of AMD.

Assuming that the data in the current paper reflect accurate estimates of the prevalence of early and late AMD, the reasons for the differences among racial/ethnic groups are not known. Jampol and Tielsch speculated that melanin, “acting as a free radical scavenger or in some other way, may protect the pigment epithelium, Bruch’s membrane, choroid, or the outer retina from degeneration changes predisposing the patient to choroidal neovascularization.” More recently, differences in distributions of deleterious and protective genes for AMD among the different racial/ethnic groups suggest that these may explain the differences in risk of developing AMD. In the MESA, whites, blacks, and Hispanics who were homozygous for the CFH Y402H CC variant genotype had the highest prevalences of early AMD compared to those with the CFH Y402H TT wild genotype. The CFH Y402H CC variant genotype distributions varied among these racial/ethnic groups and explained some of the difference in early AMD prevalence for Chinese and for Hispanics, but not for blacks. It was postulated that differences in distributions of other protective (CFH-related 1/3c deletions, C2/BF haplotype) and deleterious genotypes (eg, ARMS2) and their interactions with environmental/host factors (eg, smoking, inflammation, oxidative stress, diet) among the different racial/ethnic groups would further explain these differences. In the LALES, the homozygous CFH Y402H CC variant polymorphism was present in only 3% of the study population compared to 9% to 21% of non-Hispanic whites. Even within a racial/ethnic group (eg, Mexican American), there is great diversity in racial/ethnic origins and likely differences in gene distributions attributable to this heterogeneity.

What do these data tell us about AMD? They suggest that genetic, and possibly genetic and environmental, interactions play a role in the pathogenesis of AMD. Understanding these differences may indicate the possibility of successful preventive approaches for this disease. Also, late AMD, while less common in blacks and Hispanic whites than non-Hispanic whites, is still frequent enough to warrant identifying those at greatest risk of progression, and they, too, would likely benefit from interventions to reduce noxious exposures (eg, cessation of smoking) and use of potential preventive therapies (eg, use of antioxidant zinc multivitamins) in those at highest risk of progression to late AMD.