Implementing Artificial Intelligence to Prevent Vision Loss From Diabetes

Implementing Artificial Intelligence (AI) to Prevent Vision Loss From Diabetes

This pragmatic clinical trial is being conducted to test the effectiveness of AI in improving screening and follow-up eye care compared to usual-care among patients with diabetes across 4 primary care clinics. This is an autonomous AI-based screening to detect diabetic eye disease at primary care visits.

Study Overview

• Status: Enrolling by invitation
• Conditions: Diabetes; Vision Disorders
• Intervention / Treatment: Other: AI; Other: Usual Practice

Detailed Description

Four clinics from two health systems will be recruited, with two clinics from each health system and randomly assigned to either usual-care or the AI Intervention with 2 clinics assigned to each arms. Beginning in month 3, a 6-week baseline period will be conducted in all clinics, followed by a 4-month intervention period for the two clinics assigned to AI.

Primary Objective:

Compare the odds of patients, who get eye screening in the AI and usual-care arms within 5 months of the recommendation

Secondary Objective:

1. Compare the odds of patients, who completed follow-up with recommended eye care in the AI and usual-care arms within 5 months of the recommendation
2. Compare the odds of patients in different demographic groups who receive eye screening in the AI and usual-care arms within 5 months of the recommendation

• Study Type: Interventional
• Enrollment (Estimated): 1700
• Phase: Not Applicable

Contacts and Locations

Study Locations

• United States
  • Wisconsin
    • Madison, Wisconsin, United States, 53792
      • University of Wisconsin

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: Yes

Description

Clinic Inclusion Criteria:

• Serve at least 267 patients with diabetes during the study period
• No point-of-care screening system in use for diabetic eye disease
• Agree to share limited identifiers data as requested

Patient Inclusion Criteria:

• Age 22 years or older
• Diagnosis of type 1 or 2 diabetes
• No known diabetic eye disease
• No diabetic eye exam in the past 12 months

Exclusion Criteria:

• Have a documented eye exam in the electronic health record within 12 months of the date of the primary care visit.
• Contraindication includes diagnosed with macular edema, severe non-proliferative retinopathy, proliferative retinopathy, radiation retinopathy, or retinal vein occlusion.
• Pregnant

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Health Services Research
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: None (Open Label)

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Active Comparator: Usual Care; Patients with diabetes will follow the clinic's usual practice, in which the primary care provider recommends an annual screening eye exam for patients with diabetes. This requires the patient to make a separate visit to see an eye care provider. Clinic staff will provide scheduling assistance per the standard scheduling procedure for the clinic.	Other: Usual Practice; PCP recommends annual vision screening, a separate visit entirely
Experimental: AI Intervention; AI intervention includes (1) acquisition of eye photos and (2) autonomous (i.e. without human oversight) AI-based identification of referrable or non-referrable eye disease at the primary care clinic.	Other: AI; AI-based eye screening program

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Time Frame
Proportion of patients who get eye screening in the AI and usual-care arms within 5 months of the recommendation	up to 5 months

Secondary Outcome Measures

Outcome Measure	Time Frame
Proportion of patients, who completed follow-up with recommended eye care in the AI and usual-care arms within 5 months of the recommendation	up to 5 months
Proportion of patients in different demographic groups who receive eye screening in the AI and usual-care arms within 5 months of the recommendation	up to 5 months

Collaborators and Investigators

• Sponsor: University of Wisconsin, Madison
• Investigators: Principal Investigator: Roomsa Channa, MD, University of Wisconsin, Madison

Study record dates

Study Major Dates

• Study Start (Actual): February 15, 2026
• Primary Completion (Estimated): February 1, 2027
• Study Completion (Estimated): February 1, 2027

Study Registration Dates

• First Submitted: April 23, 2026
• First Submitted That Met QC Criteria: April 23, 2026
• First Posted (Actual): April 30, 2026

Study Record Updates

• Last Update Posted (Actual): April 30, 2026
• Last Update Submitted That Met QC Criteria: April 23, 2026
• Last Verified: April 1, 2026

More Information

Terms related to this study

• Keywords: Artificial Intelligence; AI; pragmatic; diabetic eye disease
• Additional Relevant MeSH Terms: Neurologic Manifestations; Endocrine System Diseases; Nervous System Diseases; Metabolic Diseases; Glucose Metabolism Disorders; Eye Diseases; Sensation Disorders; Pathological Conditions, Signs and Symptoms; Nutritional and Metabolic Diseases; Signs and Symptoms; Diabetes Mellitus; Vision Disorders
• Other Study ID Numbers: 2025-1248; SMPH | Ophthal and Visual (Other Identifier: UW Madison); Protocol Version 4/2/24 (Other Identifier: UW Madison)

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: YES
• IPD Sharing Supporting Information Type: STUDY_PROTOCOL; SAP; ICF

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No