Abstract
Purpose :
To compare the rate of diabetic retinopathy (DR) screening before and after implementation of an automated program designed to capture the dilated eye exam in an electronic health record (EHR).
Methods :
A retrospective chart review of all patients in an integrated health system with a DR screening requirement. An EHR automated program was created to capture the dilated eye exam performed for diabetic patients at a university eye center in an integrated health system. The program was designed to automatically satisfy a “checklist” for the DR screening requirement, which is listed for each diabetic patient in their EHR medical record. An EHR database query was then performed retrospectively to identify active health system patients with a DR screening requirement, whether a dilated eye exam was performed at the eye center, and screening status (satisfied vs not satisfied). Annual rates of screening satisfaction before and after implementation of the automated program were calculated and compared using χ2 test.
Results :
In the year prior to implementation of the automated program, 4155 subjects had a dilated eye exam for diabetes at the eye center but only 36.8% (1529) of these patients had their DR screening status “satisfied” in the EHR via manual entry by their physician. In the year following implementation, 4198 subjects had a dilated eye exam for diabetes at the eye center and 100% of these patients had their DR screening status “satisfied” via automated capture (P≤.0001). When looking at total number of diabetic patients in the integrated health system (those seen and not seen at the eye center) 18% (3091/17613) had their DR screening correctly reported prior to implementation versus 31% (5599/18290) in the year following implementation (P≤.0001).
Conclusions :
An automated workflow in an EHR can improve capture rate of DR screening exams in an integrated health system, especially for patients who receive their eye care within the system. It is harder to assess rates of diabetic eye screening in patients whose care is fragmented across disparate medical facilities.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.