Diabetic retinopathy (DR) is the most common cause of vision loss among working-age adults. Treatment of late-stage DR can cost as much as $28,000 per patient, while the cost of early-stage treatment is minimal.
While it is a challenging disease, 90% of DR cases can be prevented by early screening and treatment. However, because DR is asymptomatic until later stages, many patients are unaware of the disease and unmotivated to complete onerous screening processes.
In 2015, only 22-36% of Penn Medicine patients met recommended screening standards for diabetic eye care. However, according to data analysis, 8,900 diabetic patients who were non-adherent to retinal screening attended Penn Medicine appointments for other services that year.
These interactions represented an opportunity to direct non-adherent patients to retinal screening.
In partnership with a team led by Thomasine Gorry, MD, MGA, we set out to increase screening rates among this target population.
Traditional eye exams require an office visit, take considerable time, and require pupil dilation, which causes hours of disruptive blurred vision.
With this in mind, we developed Eyes on Site (EOS), a model for retinal screening that leverages non-mydriatic cameras - which do not require pupil dilation - to conduct mobile screening with remote interpretation.
We piloted the EOS model at three Penn Medicine locations where diabetic patients would be attending appointments. Patients were informed of the impromptu (and free!) screening by their provider, images were interpreted remotely, and findings were communicated to patients and providers.
A total of 125 patients were screened over the course of the pilot, and an extraordinarily high incidence of disease was uncovered. Of the patients screened:
- 47% did not know when their last screening was, or reported never having completed one
- 57% required a referral, either for baseline or detected eye disease
- 11.4% presented with evidence of diabetic retinopathy
At scale, EOS is projected to effectively identify diabetic eye disease early to avoid costly treatment and poor patient outcomes. If EOS screened 890 patients yearly, we predict 356 cases of undiagnosed disease (40%), amounting to more than $290,000 in treatment cost avoidance.