AI system boosts diabetic eye exam completion rates among youth with diabetes
Click Here to Manage Email Alerts
Key takeaways:
- All participants in the intervention arm completed their diabetic eye exams vs. 22% in the control arm.
- The 6-month follow-through completion rates were 64% and 22%, respectively.
An autonomous artificial intelligence system increased diabetic eye exam completion rates in youth with diabetes and could help close the care gap in underserved communities, according to a study published in Nature Communications.
“While the introduction of telemedicine over the last two decades has improved screening and facilitated early detection of diabetic eye disease [DED], the development of diagnostic autonomous artificial intelligence (AI) systems for diagnosing DED has ushered in the next chapter of DED screening,” Risa M. Wolf, MD, from Johns Hopkins School of Medicine, and colleagues wrote.
To determine whether autonomous AI could improve completion rates for diabetic eye exams, researchers conducted a parallel, randomized trial of 164 racially and ethnically diverse youth (mean age, 15.2 years; 58% girls) with type 1 and type 2 diabetes.
Participants in the intervention arm underwent a 5- to 10-minute autonomous AI system (IDx-DR, Digital Diagnostics) diabetic eye exam without dilation, while the control group received a standard-of-care diabetic eye exam from an eye care provider.
According to results, all participants in the intervention arm completed their exams, corresponding to a primary care gap closure rate of 100%, compared with 18 of 82 in the control arm, for a primary gap closure rate of 22%.
Further, 25 participants in the intervention arm received a “DED present” result and were referred for care, with 16 completing a visit with an eye care provider within 6 months for a follow-through completion rate of 64%. Conversely, 18 participants in the control arm visited an eye care provider, resulting in a follow-through completion rate of 22%.
“The results of the ACCESS trial confirmed our hypothesis that autonomous AI increases diabetic eye exam completion rates and closes this care gap in a racially and ethnically diverse population of youth with diabetes, compared to standard of care,” Wolf and colleagues wrote. “This result held true despite augmenting the standard of care referral in the control arm with deliberate education for the patient and caregiver regarding the importance of diabetic eye exams.”