Primary Care Screening Flops in Detecting Diabetic Eye Disease

Fundoscopic examination in a primary care setting was no boon for the detection of diabetic retinopathy, a retrospective cohort study showed.

Among 59 patients with diabetes who underwent both a primary care exam and a professional eye exam by either an ophthalmologist or optometrist, primary care exams had a sensitivity of 0% (95% CI 0.0-14.9) for detecting disease and an accuracy of 62.7% (95% CI 50.0-73.9), reported Anthony N. Kuo, MD, of Duke University in Durham, North Carolina, and colleagues in JAMA Network Open.

“These findings suggest that fundoscopy may be a suboptimal method to screen for diabetic retinopathy in primary care,” the authors wrote, adding that “research to facilitate implementation of more effective primary care-based diabetic retinopathy screening strategies is needed.”

“Although prior studies have suggested that training may improve primary care professionals’ accuracy in diabetic retinopathy screening, the cost-effectiveness of doing so at a population level needs to be compared with other population-based screening strategies, such as teleretinal imaging programs and artificial intelligence-based diagnostics,” they noted.

Only 12% of the 767 patients included in the analysis received a fundoscopy exam at a primary care visit. Certain factors were associated with a greater likelihood of such an exam. For example, every 10 years longer that the primary care providers had been in practice was tied with a 26% higher odds of their patient receiving fundoscopy at least once during the year (adjusted OR 1.26, 95% CI 1.01-1.59, P=0.04).

On the other hand, being seen by a nurse practitioner at a primary care office — rather than a physician — was tied with a 77% lower odds of receiving fundoscopy (aOR 0.23, 95% CI 0.04-0.79, P=0.049).

That being said, there weren’t any specific clinical factors or patient demographics that were associated with a primary care provider performing fundoscopy, such as HbA1c level, age during their first encounter, insulin use, diabetes type, sex, race, ethnicity, insurance type, and report of an eye concern at any primary care encounter during the year.

In addition, the location of the primary care clinic and the number of ophthalmologists in the area were also not linked to the likelihood of a primary care provider performing fundoscopy.

These associations were all adjusted for rural clinic location, clinic location in a primary care shortage area, and documentation of an up-to-date eye examination by a primary care provider in the study year.

For this study, patients with type 1 or type 2 diabetes were seen at one of 28 clinics within the Duke Health primary care network in 2019. Median patient age was 64, and half were women. Among the 767 randomly selected patients, 40.1% didn’t have a professional eye care visit nor did they receive a fundoscopic examination from their primary care provider.

Most of these patients had commercial insurance (47.1%), followed by Medicare (25.9%) or Medicare Advantage (23.9%). Only 2.5% were insured with Medicaid.

The study’s retrospective review of electronic health records was a limitation, Kuo and colleagues acknowledged. Furthermore, all patient encounters only occurred at the same North Carolina healthcare system, thus limiting the generalizability of the findings.