Background: Diabetes is a common cause of progression from chronic kidney disease (CKD) to end stage renal disease (ESRD). Progression to ESRD among racial/ethnic groups with diabetic kidney disease (DKD) is not well characterized.
Objectives: To compare time to progression to ESRD between White, Black, and Hispanic DKD patients. To describe changes in study results when using open claims data to supplement closed claims data.
Methods: A cohort analysis of adult DKD patients with newly diagnosed CKD (index) between 01/01/18-06/30/22 were identified in CHRONOS, a linked claims database. The primary analysis used closed claims to identify patients with stage 3 or 4 CKD, a history of diabetes, 12 months of continuous enrollment in the year before index, and race/ethnicity in linked SDoH data. Progression over three years after index was defined as ESRD indicated by a diagnosis, dialysis procedure, or kidney transplant procedure on a submitted claim. Patients who did not progress were censored at disenrollment from a health plan or at year 3 of follow-up. Kaplan-Meier curves describe time to progression. Hazard ratios (HRs) and 95% confidence intervals are reported from Cox Proportional Hazards models comparing time to progression between White, Black, and Hispanic patients, adjusting for demographics, baseline characteristics, and SDoH. The secondary analysis repeated the primary analysis in a cohort of patients with opens claims supplementing the outcome definition from the closed claims. All patient characteristics were defined by ICD-10-CM, CPT, and NDC codes.
Results: In the primary analysis, White (n=23,097), Black (n=4,895), and Hispanic (n=2,774) patients with DKD were 45.6, 54.1, and 46.1% female and 58 to 61 years of age, on average. Progression occurred in 5.8, 8.6, and 10.1% of White, Black, and Hispanic patients with a 3-year progression-free probability of 0.87, 0.82, and 0.79, respectively. Compared to White patients, Black [HR: 1.3 (1.2-1.5)] and Hispanic [HR: 1.6 (1.4-1.8)] patients were more likely to progress. With the inclusion of open claims in the secondary analysis, the 3-year progression-free probabilities were 0.84, 0.77, and 0.73 among White, Black, and Hispanic patients. Decreased probabilities were due to 287, 106, and 81 additional progression events identified in the open claims data. Open claims data did not meaningfully change the HRs.
Conclusions: This analysis found Black and Hispanic patients progressed to ESRD more quickly than White patients with additional outcome events in the secondary analysis leading to changes in the event-free survival without changing the relative associations from Cox Proportional Hazards models.
