Cost of a Community-Based Diabetic Retinopathy Screening Program

The prevalence of diabetes in the U.S. has steadily increased over the last few decades (1). Although adults with diabetes are at increased risk for ocular disease and blindness, primarily related to diabetic retinopathy (2), many are not evaluated for retinopathy. Barriers to vision-screening adherence among adults with diabetes include the perception—and often the reality—that examinations are expensive and a poor understanding of the increased risk of vision loss (3). The goal of this study was to determine the costs of a community-based program to screen adults for diabetic retinopathy.

In this study, free ocular health screening was offered over a 17-month period at a community health center in a low socio-economic area. Fundus photographs were taken, uploaded to a HIPPA-compliant secure Web site, and reviewed for any clinical pathology and specifically for diabetic retinopathy. We collected information on all costs of the screening program. Costs were assessed in three ways in a base model: the costs per participant for 1) screening, 2) detection of ocular abnormalities, and 3) referrals for diabetic retinopathy treatment. Sensitivity analyses were conducted varying the wages of ocular technicians and estimates of time devoted to research versus screening.

Overall, 607 adults with diabetes were screened. The sample was 65.6% female, 45.5% non-Hispanic black or Haitian, and 52.2% Hispanic. The average age was 55.8 years (SD 9.2). Most (78.4%) did not have insurance, 45.2% had not had an eye exam in the last 2 years, and 10.9% reported never having an eye exam. Based on the reading of the fundus photographs, 61.8% were identified with some clinical pathology and 24.4% were referred with diabetic retinopathy. More Hispanics (12.3%) than non-Hispanic blacks (9.5%) and more males (14.9%) than females (8.8%) had never had an eye exam. The uninsured were much less likely to have had previous screening than those with insurance (5.3 vs. 12.5%).

Over the 17 months of the program, the total cost was $91,294, with staff time ($53,388) being approximately 58% of the total cost of the screening intervention. Start-up costs were 31% of the program costs.

In our base model, the cost per screening was $116, the cost per ocular abnormality identified was $188, and the cost per diabetic retinopathy referral was $478 (Fig. 1). In sensitivity analyses, costs were reduced by 7–23% as wages from national databases for ocular technicians were less than the wages paid to our staff and our original estimates of time spent on research versus the screening program were conservative.

Very few retinopathy screening studies, including those on the cost of screening (4,5), have been conducted in community settings. The incidences of ocular abnormalities and diabetic retinopathy in this low socioeconomic population were very high and the cost of community-based screening was slightly over $100 per individual screened and less than $500 for preliminary identification of diabetic retinopathy. In addition, many patients with diabetes had never had an eye exam. Our findings illustrate the tremendous need for and great potential of community-based diabetic retinopathy surveillance programs.