Comparative effectiveness of telemedicine to detect diabetic retinopathy in American Indians and Alaska Natives

BACKGROUND: Diabetic retinopathy is the leading cause of new cases of blindness among U.S. adults aged 20–74. Both diabetes and diabetes-related blindness disproportionately affect American Indians and Alaska Natives (AI/AN). Early detection and treatment have been shown to prevent vision loss, but multiple barriers to receiving annual eye exams exist in the AI/AN population, including limited access to eye care providers in rural areas. METHODS: The researchers partnered with two tribal health clinics in Oregon and Kansas to focus on screening for diabetic retinopathy. Participants were randomly assigned to either telemedicine (using digital images of the retina captured with a non-mydriatic camera) or traditional surveillance (annual eye exams in an eye care provider's office). Experts used a validated scale to grade diabetic retinopathy from Stage 0 (no retinopathy) to Stage 5 (proliferative diabetic retinopathy). RESULTS: The proportion of those that received a screening exam within one year of enrollment was higher in the telemedicine group than the traditional surveillance group (94% vs. 44%, p<0.001). The prevalence of diabetic retinopathy at baseline was 25%, with severity graded as “none” in 75%, “mild non-proliferative” in 15%, “moderate non-proliferative” in 7%, “severe non-proliferative” in 1%, “very severe non-proliferative” in 0%, and “proliferative” in 2%. Diabetic retinopathy was associated with higher mean systolic blood pressure (p=0.013), non-white primary ethnicity (p=0.019), higher mean HgA1c (p<0.001), and longer duration of diabetes (p<0.001). CONCLUSION: Data from this research may contribute to changes in national and state guidelines in the delivery of diabetic eye exams.