The Relationship Between Health Insurance Status and Diabetic Retinopathy Progression

Abstract

Listen

Translate article

ObjectiveTo determine if baseline diabetic retinopathy severity mediates the relationship between health insurance status and diabetic retinopathy progression. DesignRetrospective cohort study Subjects716 patients aged ≥18 years with a diagnosis of type 1 or 2 diabetes mellitus, and a diagnosis of non-proliferative diabetic retinopathy (NPDR) were identified from the electronic health record (EHR) of a tertiary academic center between 6/2012 and 2/2022. MethodsNPDR severity at baseline was the proposed mediator in the relationship between insurance status and PDR progression. Logistic regression was used to determine the association between insurance status and NPDR severity at baseline, and Cox proportional hazards regression was used to assess the association between insurance status and time to PDR progression. To analyze the mediation effect of NPDR severity at baseline, a counterfactual approach, which decomposes a total effect into a natural direct effect and a natural indirect effect was applied. Main Outcome MeasuresTime to progression from first NPDR diagnosis to first PDR diagnosis. ResultsOf the 716 patients, 581 (81%) had Medicare or private insurance, 107 (15%) had Medicaid, and 28 (4.0%) were uninsured at their baseline eye visit. Uninsured or Medicaid patients had a higher proportion of moderate or severe NPDR at their baseline eye visit, and a higher proportion of progression to PDR. After adjusting for confounders and NPDR severity at baseline, uninsured patients had significantly greater risk of progression to PDR compared to Medicare/private insurance patients (HR 2.63, 95% CI 1.10 – 6.25). Medicaid patients also had an increased risk of progression to PDR compared to Medicare/private insurance patients, although not statistically significant (HR 1.53, 95% CI 0.81 – 2.89). NPDR severity at baseline mediated 41% of the effect of insurance status (uninsured vs Medicare/private insurance) on PDR progression. ConclusionsUninsured patients were more likely to have an advanced stage of NPDR at their baseline eye visit and were at significantly greater risk of progression to PDR compared with Medicare or privately insured patients. Mediation analysis revealed that differences in baseline NPDR severity by insurance explained a significant proportion of the relationship between insurance status and diabetic retinopathy progression.