Assessing diabetic eye screening rates in a rural, multi-payer health system

Abstract

Purpose : Although yearly diabetic eye screening is a widely-used measure of health quality, records are often incomplete due to lack of communication between eye and primary care providers. Obtaining accurate screening rates is important for health systems to measure the impact of quality improvement initiatives. We hypothesized that a brief telephone survey combined with medical records review could assess diabetic eye screening rates and patient perceptions of diabetic eye disease in a rural, multi-payer health system.

Methods : We conducted a 3-week telephone survey (June to July 2017) at Mile Bluff Medical Center in Mauston, WI. Patients had a diagnosis of either type 1 or 2 diabetes mellitus and had seen a primary care provider at Mile Bluff within the past 3 years (at the Main or one of 4 Outreach clinics). A random-number generator in Microsoft Excel (Microsoft, Inc., Redmond, WA) was used to select patients for telephone calls within the sampling frame. Minimum sample sizes were calculated (95% confidence level, 10% confidence interval). Medical records were requested and reviewed to confirm screening. The diabetic eye screening rate was defined as the proportion with documented screening within the past 12 months.

Results : The survey response rate was 82.9% and 87.2% for the Main and Outreach clinics, respectively. All sample-size minimums were achieved (n=92 Main; n=269 Outreach). Most patients (90.9%) believed that diabetes can lead to vision problems and many (76.7%) self-reported diabetic eye screening within the past two years. However, 12.4% of patients who self-reported screening were inaccurate upon medical records review. Among the records received, only 9.9% were diagnosed with diabetic retinopathy. Diabetic eye screening rates were 62% for the Main and 56% for the Outreach clinics. Screening rates based on medical records review were commensurate with those from a local private insurer.

Conclusions : A brief telephone survey combined with medical records review was effective for obtaining diabetic eye screening rates in a rural, multi-payer health system. Low screening rates were consistent with national averages despite patient knowledge of vision loss related to diabetes. The accuracy of patient self-reported diabetic eye screening may be limited. Medical records review showed a low prevalence of diabetic retinopathy among those adherent with screening.

This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.