Nonmydriatic fundus camera for diabetic retinopathy screening in a safety net hospital: effectiveness, prevalence, and risk factors.

Abstract

To evaluate diabetic retinopathy (DR) prevalence and risk factors, and the effectiveness of nonmydriatic fundus camera as a screening tool for the detection of DR, in a safety net hospital. This was a retrospective, population-based, cross-sectional study. Diabetic patients, referred by their primary care physicians to a DR community screening program, were included. A Topcon TRC NW-6S camera was used to obtain 45-degree digital color fundus images. Images were interpreted by retina specialists using a quality rating system. Patients with retinal findings or unreadable photographs were referred for a complete examination. Outcome measures were attendance rates, photograph quality, DR prevalence, and associated risk factors. A total of 948 diabetic patients were sent for camera screening, with an attendance rate of 65.6%, which increased during the study period. The mean age was 55.8 ± 11.6 years, the majority (56.9%) were Hispanic, and 43.5% were uninsured. Overall photograph quality rating was relatively high, with 81.7% graded as good or fair. Thirty photographs (2.9%) were unreadable. The prevalence of newly diagnosed DR was 11.1%. Independent DR-associated risk factors included Hispanic race (odds ratio [OR] = 2.29), lack of health insurance (OR = 2.49), longer duration of diabetes (OR = 1.07), higher HbA1c levels (OR = 1.19), presence of diabetic complications (OR = 2.93), and lack of previous eye examination (OR = 13.22). Nonmydriatic fundus camera is an effective and feasible screening tool for the early detection of DR in a safety net institution. It should be considered in areas with limited access to health care to improve quality of care and potentially reduce vision loss rates.