OR28-5 Widening Economic Disparity in Rates of Readmissions Following Hospitalizations for Hyperglycemic Emergencies Among US Adults over a Decade.

INTRODUCTION: Hyperglycemic emergencies, including diabetic ketoacidosis (DKA) and hyperglycemic hyperosmolar state (HHS), are common reasons for hospitalization among adults with diabetes. Household income is a known determinant of health and impacts access to medications and health services. Outpatient services and medications impact readmission rates as medication nonadherence and inadequate follow-up are common underlying precipitants of DKA and HHS. This study outlined the trends in readmissions following initial hospitalization for hyperglycemic emergencies. METHODS: This was a longitudinal trend analysis of the National Readmission Database (NRD), which is the largest readmission database in the US. The study period was from 2010 to 2018, and involved adults with DKA and HHS as the principal discharge diagnosis. We excluded elective and December admissions. Using the unique visit linkage variable within the NRD, we estimated the 30-day all-cause readmission rate (30ACR) stratified by the median household income for the patient's zip code quartile (MHOI). Outcomes included comparing trends in 30ACR in DKA and HHS between the low-income quartile (LIQ) and high-income quartile (HIQ) as well as trends in 30-day DKA and HHS specific readmission rates between the LIQ and HIQ. We estimated the estimated annual percent change (APC) via logistic trend regression adjusted for age categories, sex, and MHOI. RESULTS: From 2010 to 2018, there was a trend towards increasing readmission rates overall. Among DKA hospitalizations, patients in the LIQ had a higher 30ACR compared to the HIQ (17.1% vs 12.3% in 2010 and 21.7% vs 15.3% in 2018). The LIQ patients also had a higher DKA-specific readmission rate over the period (7.2% vs 5.1% in 2010 and 7.6% vs 10.4% in 2018). The adjusted APC for DKA 30ACR was 1.6%, with the overall odds of readmission higher in the LIQ (adjusted odds ratio [aOR]: 1.50, 95% CI: 1.42 - 1.58, p<0.001) compared to the HIQ. Among HHS hospitalizations, patients in the LIQ had a higher 30ACR compared to the HIQ (12.2% vs 10.4% in 2010 and 14.8% vs 12.4% in 2018). The LIQ patients also had a higher HHS-specific readmission rate over the period (0.8% vs 0.6% in 2010 and 1.2% vs 1.1% in 2018). The adjusted APC for HHS 30ACR was 1.9%, with the overall odds of readmission higher in the LIQ (adjusted odds ratio [aOR]: 1.13, 95% CI: 1.02 - 1.26, p=0.025) compared to the HIQ. CONCLUSION: There seem to be rising readmission rates following hyperglycemic emergency hospitalizations with significant disparity based on median household income. This difference is was higher DKA. Though the study design limits the elaboration on specific income-related factors that are at play, these findings establish the need for further exploring the barriers that predispose patients to readmissions to decrease healthcare utilization costs. Presentation: Tuesday, June 14, 2022 10:45 a.m. - 11:00 a.m.