Automated Disease Detection and the Demand for Human Experts: The Case of Diabetic Retinopathy

Nearly four in ten persons with diabetes currently fail to undergo recommended annual retinal exams, resulting in tens of thousands of cases of preventable blindness annually in the United States. Current retinal exams take about two hours to allow time for pupil dilation, consultation with the ophthalmologist or optometrist, the patient’s recovery, and documentation of the results in the patient’s medical chart or electronic health record.

As the current availability of ophthalmologists and optometrists is inadequate to screen all patients at risk every year, automated screening systems deployed in primary care settings and even in patients’ homes could fill the current gap in supply. Compact retinal imagers have been prototyped that allow patients to perform retinal scans on their own in less than 2 minutes. So-called “smart” retinal imagers are equipped with image-analysis algorithms that evaluate the camera-generated digital images for retinal abnormalities, which, in addition to retinal disease, may eventually aid in the early detection of cardiovascular disease, macular degeneration, glaucoma, malaria, and Alzheimer’s disease. These patient-operated retinal imagers do not require the labor-intensive interpretation of medical images and data entry and automatically flag patients for a follow-up assessment by a physician.

To examine the potential impact of an automated disease-detection system on physician productivity, health care cost, and patient health, we estimated the likely change in the number of office visits and cases of retinal disease that are detected and treated.

Under the conservative assumption that the adoption of automated detection systems left the proportion of persons with diabetes undergoing screening unchanged at its current level of 63%, the number of referrals to ophthalmologists would drop by more than half from 16.25 to 7.57 million, while the proportion of patients requiring treatment who receive it would decline minimally from 57% to 56%. Automated detection systems would thus allow patients to avoid every other office visit and yield a significantly higher rate of detecting and treating diabetic retinopathy per dilated retinal examination.

The ease of use, minimal time requirement, and widespread availability of retinal imagers might raise screening rates and thus help detect additional cases of retinal disease that require treatment, in turn reducing the number of diabetic patients who develop complications such as blindness. By raising the number of retinopathy cases detected per office visit, and coupled with the rising prevalence of diabetes in the United States, increased screening and disease detection rates would largely sustain the demand for ophthalmologist services.