Prevalence of Eye Disease Related to Diabetes Remains High in United States

The prevalence of eye disease due to diabetes remains high in the United States, according to a study published in JAMA Ophthalmology. The new estimates from the study can help inform public health resource allocations and interventions for populations at the highest risk, according to the study authors.

There are more than 37 million people living in the United States that have diabetes and are at risk for eye disease, with diabetic retinopathy (DR) being the leading cause of blindness in working-age adults. The Eye Diseases Prevalence Research Group (EDPRG) found that 40.3% of adults with diabetes had DR and 8.2% had vision-threatening DR (VTDR) in 2004.

New composite estimates are important to determine changes in population demographics and diabetic prevalence. This study aimed to provide new prevalence estimates based on sociodemographic groupings.

Meta-regression methods were used to estimate the national prevalence of DR and VTDR. A 5-year age group was used to estimate prevalence, along with a nondifferentiated sex and gender measure and a composite indicator of race and ethnicity. Race was split into 4 groupings of Hispanic, non-Hispanic Black, non-Hispanic White, and all other races and ethnicities.

The meta-regression included the published results of population-based studies, National Health and Nutrition Examination Survey data from 2005 to 2008, Medicare Part B fee-for-service claims from 2019, and medical claims to commercial insurers found in the 2016 IBM MarketScan Commercial Research Database. The NHANES was missing 20.1% of DR data and 20.4% of VTDR data.

Population-based studies were included if they were published after 2000 and presented primary results; 5 of these studies in adults and 2 in children were included. County-level estimates were derived from county-level random effects and national estimates. Prevalence of diabetes by age, sex and gender, and race were estimated by analyzing the NHANES survey from the 2017 to March 2020 cycle. Census data from 2021 was used to estimate prevalence counts of DR and VTDR.

The analysis estimated 9.60 million people living with DR (95% CI, 7.90-11.55), which suggests a prevalence rate of 26.43% (95% CI, 21.95-31.60) in people with diabetes. The estimation accounted for 5.56 individuals who had male sex and gender (95% CI, 4.38-6.98) and 4.04 million individuals who had female sex and gender (95% CI, 3.24-5.01) living with DR. The rate of DR was 15.7% (95% CI, –7.0 to 45.2) higher in men than women when using an age and race standardized rate.

Hispanic (29.21%; 95% CI, 22.13%-38.26%) and Black (34.39%; 95% CI, 26.09%-43.75%) individuals had higher rates of DR compared with White individuals (24.40%; 95% CI, 18.38%-31.56%) The standardized prevalence of DR was 21.90% (95% CI, –17.00% to 79.30%).

The prevalence of DR increased with age before declining in the oldest age groups, with those younger than 25 years having a prevalence of 13.01% (95% CI, 7.83%-25.68%),those aged 40 to 64 years having a prevalence of 27.15% (95% CI, 21.71%-33.65%), and those aged 80 years and older having a prevalence of 24.81% (95% CI, 17.41%-33.62%). Prevalence rates of DR ranged from 20.84% (95% CI, 17.45%-24.81%) in Nevada to 31.31% (95% CI, 25.14%-38.30%) in Massachusetts.

The estimation for prevalence of VTDR in the United States in 2021 was 1.84 million people, which came to a prevalence rate of 5.06% (95% CI, 3.90%-6.57%) in people with diabetes. The researchers estimated that 1.05 million individuals of male sex and gender (95% CI, 0.78-1.41) were living with VTDR. The prevalence was higher in Hispanic (7.14%; 95% CI, 4.86%-10.32%) and Black (8.66%; 95% CI, 5.91%-12.17%) individuals compared with White individuals (3.55%; 95% CI, 2.22%-5.41%).

Prevalence of VTDR increased by age until individuals were older adults, with prevalence for those younger than 25 years being 1.41% (95% CI, 0.80%-2.23%), prevalence for those aged 40 to 64 years at 5.27% (95% CI, 3.69%-7.41%), an prevalence for those aged 80 years and older found to be 3.48% (95% CI, 2.17-5.20).

The NHANES data was used for the basis of the estimates but was last collected from 2005 to 2008. Data that was missing was imputed by the researchers, which yielded higher estimates for DR and VTDR. Medicaid and Medicare Part B claims were used to estimate county variation in prevalence but this assumes that the claims corresponded to total population prevalence.

The researchers concluded that there were an estimated 9.60 million people who had DR and 1.84 million people who had VTDR. These estimates could “inform the allocation of public health resources and interventions to communities and populations at highest risk, such as expanding teleretinal imaging to improve DR screening in counties with the highest prevalence.”

Reference

Lundeen EA, Burke-Conte Z, Rein DB, et al. Prevalence of diabetic retinopathy in the US in 2021. JAMA Ophthalmol. Published online June 15, 2023. doi:10.1001/jamaophthalmol.2023.2289