Leveraging Artificial Intelligence to Prevent Vision Loss From Diabetes

Multicenter National Parallel Cluster Randomized Controlled Superiority Trial Comparing an Artificial Intelligence-Based Screening Strategy to Usual Care for Improving Eye-Care Follow-Up Among Patients With Diabetes (AI-BRIDGE Trial)

This study aims to investigate whether a novel artificial intelligence based screening strategy (AI-Based point of caRe, Incorporating Diagnosis, SchedulinG, and Education or AI-BRIDGE), which allows primary care providers to screen patients for vision-threatening diabetic eye disease in the primary care clinic, improves screening and follow-up care rates across race/ethnicity groups and reduces racial/ethnic disparities in screening.

This is a multicenter clinical trial and University of Wisconsin is the coordinating center of the study.

A stepped-wedge cluster randomized clinical trial will be conducted. The investigators will evaluate the effectiveness of two standard diabetic retinopathy screening strategies at primary care clinics; (1) AI-based eye screening program called AI-BRIDGE, eye photos of the patients will be obtained in the primary care clinic by trained clinic staff. Images will be reviewed using autonomous artificial-intelligence (AI) algorithm (Digital Diagnostics). Patients with referrable diabetic retinopathy are detected within minutes and patients with referrable disease will be assisted with scheduling an in-person follow-up eye care visit (2) usual care screening, primary care providers refer patients with diabetes to an eye care provider for an in-person dilated eye exam.

After adapting AI-BRIDGE protocols to clinics and training of clinic personnel, stepped wedge randomized clinical trial begins with sites transitioning from usual-care to AI-BRIDGE in 4 steps.

Primary Objective:

• Compare the proportion of patients, by race and ethnicity, who follow-up with recommended eye care in the AI-BRIDGE and usual-care arms within 6 months of the recommendation.

Secondary Objectives:

• Compare the difference in proportion of White vs Hispanic and White vs Black patients who get screening in the AI-BRIDGE and usual-care arms within 6 months of the recommendation.
• Compare proportion of patients, by race and ethnicity, who receive eye screening in the AI-BRIDGE and usual-care arms within 6 months of the recommendation.

Inclusion Criteria:

• Diagnosed with type 1 or 2 diabetes
• No known diabetic eye disease
• Medicaid as their primary insurance
• Not had an eye exam in the prior year

Exclusion Criteria: