Cost-effective screening for diabetic retinopathy using a nonmydriatic retinal camera in a prepaid health-care setting.

OBJECTIVE

To assess the efficacy of using a nonmydriatic Polaroid retinal camera as a method for screening diabetic patients for treatable diabetic retinopathy.

RESEARCH DESIGN AND METHODS

All 522 diabetic patients followed in a health maintenance organization-affiliated diabetes program had retinal photos taken. Compliance with the routine referral to one of two retinal specialists (the examiners) was 74%. The results from the examiners were compared with the results of the reader of the retinal photos.

RESULTS

Sensitivity was 100% and specificity was 82% for the diagnosis of serious diabetic retinopathy (preproliferative or proliferative retinopathy or macular edema) by the examiners compared with the diagnosis of any diabetic retinopathy by the reader. No patient had serious diabetic retinopathy inside or outside the photographic field that was missed because all patients with serious diabetic retinopathy showed some diabetic retinopathy within the photographic field. The reader tended to underrate the severity of the diabetic retinopathy, but when the reader diagnosed serious diabetic retinopathy, it was always present on exam.

CONCLUSIONS

The nonmydriatic retinal camera is easy to use, inexpensive, and can be used as part of a general diabetes exam, independent of a physician, in patients who should, but may not, be referred to an ophthalmologist. Any patient with abnormal findings on photos should be referred to an ophthalmologist, and any patient with findings of serious diabetic retinopathy on the photos should be referred immediately for possible laser therapy.