Abstract
Purpose: :
Digital retinal imaging technology is being used to assess or screen patients for diabetic retinopathy. This study evaluated compliance with referral recommendations based on interpretation of images obtained with a validated diabetic retinopathy assessment program in the primary care environment.
Methods: :
Medicare patients with diabetes who had not been evaluated by an ophthalmologist in the past 12 months were imaged with the DigiScope, an Internet–based diabetic retinopathy assessment program, in the office of their primary care physician. The images were transmitted over the Internet and reviewed at a central reading center. A report was returned to the primary care physician with a recommendation for referral (non–sight threatening pathology including moderate nonproliferative diabetic retinopathy), urgent referral (sight–threatening pathology including possible macular edema, severe nonproliferative diabetic retinopathy or proliferative diabetic retinopathy), or re–evaluate in one year (no or mild nonproliferative diabetic retinopathy). Referral was also recommended for patients with unreadable images or when images could not be acquired. Follow–up was obtained to evaluate compliance with the recommendation for urgent referral for patients determined to have possible sight–threatening pathology. Time to follow–up evaluation was also assessed.
Results: :
1,453 patients with diabetes were imaged. Urgent referral for sight–threatening pathology was recommended for 52 (4%) patients. At time of follow–up, 2 patients were deceased. A visit to an ophthalmologist was confirmed for the remaining 45 (90%) patients. Mean time to evaluation by an ophthalmologist after imaging was 61 days. At least 7 patients underwent laser treatment.
Conclusions: :
Implementation of digital retinal imaging technology in the primary care setting appears to motivate previously non–compliant patients to seek ophthalmic care, as recommended based on images obtained with this diabetic retinopathy assessment program. Future studies are required to further assess and confirm clinical effectiveness of this technology in improving visual outcomes.
Keywords: diabetic retinopathy • detection • imaging/image analysis: clinical