JMIR Publications

ABSTRACT

Background:

Screening for diabetic retinopathy is important for the prevention of blindness among the adult population. Currently, patients with diabetes require a referral from their primary care physician to see an ophthalmologist for their annual eye exam, which can be an added inconvenience. As such, there is a need for alternative screening strategies within an outpatient network. The use of a telemedicine platform in a primary care network serves as a novel strategy to increase diabetic retinopathy screening rates. LifeBridge Health operates two Track 1 Accountable Care Organizations with a combined attribution of approximately 28,000 patients. Many value-based care and pay for performance programs use diabetic retinopathy screening rate as a quality measure. In order to provide better access to diabetic retinopathy screening for our patients, three specialized cameras were placed in three primary care practices in October 2017 as part of a pilot program. The online Intelligent Retinal Imaging Systems (IRIS) platform was utilized as a secure data warehouse of images that could be interpreted remotely by an ophthalmologist within the LifeBridge Health network for the diagnosis of diabetic retinopathy or detecting other types of pathology (e.g. macular edema).

Objective:

The aim of this retrospective descriptive study was to examine if a telemedicine platform can be used to increase diabetic retinopathy screening rates in the primary care setting.

Methods:

Three distinct datasets corresponding with three time periods were examined for this study. Pre-post comparison examined screening rates from all practices from January 2018 – December 2018 to those of January 2017 – December 2017. The pilot program dataset examined screening rates in the practices before and after the implementation of the IRIS cameras in October 2017. Aggregate diagnostic data from the IRIS online dashboard from October 2017- December 2019 was also examined to determine the benefit of the IRIS platform since the initial implementation.

Results:

Pre-post comparison screening rates showed mean screening rates of 38.5% and 47.2%, respectively, indicating an 8.7% improvement in screening. The pilot program showed improved screening rates at each outpatient practice with the implementation of the IRIS cameras. Aggregate data since the implementation of the IRIS cameras showed that, of the 1213 patients who were screened, approximately 17.1% (n=207 patients) were diagnosed with diabetic retinopathy and an additional17.7% (n=215 patients) were suspected of having some form of other pathology. 10.1% (n=123 patients) were also suspected to be at risk for imminent vision loss.

Conclusions:

This retrospective descriptive study suggests that a telemedicine platform can be used to improve diabetic retinopathy screening rates in the primary care setting within a large healthcare system.