The Prevalence of Diabetic Retinopathy in American Indians/Alaska Natives and non-Indigenous Americans: A Systematic Review and Meta-Analysis.

Abstract

Topic: To estimate the prevalence of diabetic retinopathy (DR) in adult American Indians / Alaska Natives (AI/AN) and non-AI/AN patients with diabetes.

Clinical relevance: Although diabetes mellitus is more prevalent among AI/AN patients compared to non-AI/AN patients, the evidence is inconsistent regarding whether AI/AN patients have a higher prevalence or severity of diabetic retinopathy.

Methods: We searched Ovid MEDLINE, EMBASE, and Web of Science Databases from inception until February 23 2025. We included primary studies evaluating the prevalence of DR in Americans with diabetes.The prevalence of a) DR, b) diabetic macular edema (DME), c) proliferative diabetic retinopathy (PDR), d) vision-threatening diabetic retinopathy (VTDR, including DME, PDR, or severe non-proliferative diabetic retinopathy), e) PDR complications were estimated. Meta-analyses were performed using Freeman Tukey double arcsine transformations and random-effects modelling. The Joanna Briggs Institute (JBI) Appraisal Checklist for Prevalence Studies was used to assess risk of bias. GRADE guidelines were used to assess certainty of evidence.

Results: Overall, 53 studies of 10 070 617 individuals were included. In the AI/AN and non-AI/AN groups respectively, the pooled prevalence of DR is estimated to be 21% (95% CI: 13%, 30%; GRADE: Low), and 20% (95% CI: 16%, 25%; GRADE: Low), respectively. The prevalence of PDR is estimated to be 3% (95% CI: 1%, 6%; GRADE: Low), and 2% (95% CI: 1%, 4%; GRADE: Low), respectively. The prevalence of DME is estimated to be 3% (95% CI: 2%, 4%; GRADE: Low) and 3% (95% CI: 2%, 4%; GRADE: Low), respectively; and the prevalence of VTDR is estimated to be 3% (95% CI: 1%, 7%; GRADE: Low) and 5% (95% CI: 4%, 7%; GRADE: Low), respectively. High-quality evidence is lacking. Comparative analysis demonstrated there may be no difference in the rate of DR between AI/AN and non-AI/AN patients (OR: 0.67; 95% CI: 0.31, 1.48; GRADE: Low).

Conclusion: There appears to be no appreciable difference in the prevalence of DR between AI/AN and non-AI/AN patients, albeit evidence is limited by heterogeneity of studies. The high disease burden highlights that public health strategies are needed equally for AI/AN as well as non-AI/AN patients.