FRI638 Glycemic And Cost Outcomes In Hispanic/Latino People With Type 2 Diabetes Initiating Dulaglutide Or Basal Insulin In The US

Abstract

Abstract Disclosure: M. Hoog: Employee; Self; Eli Lilly & Company. Stock Owner; Self; Eli Lilly & Company. J. Maldonado: Employee; Self; Eli Lilly & Company. Stock Owner; Self; Eli Lilly & Company. R. Wangia-Dixon: Employee; Self; Optum Inc. Other; Self; Eli Lilly & Company. R. Halpern: Employee; Self; Optum Inc. Other; Self; Eli Lilly & Company. E. Buysman: Employee; Self; Optum Inc. Other; Self; Eli Lilly & Company. G. Gremel: Employee; Self; Optum Inc. Other; Self; Eli Lilly & Company. A. Huang: Employee; Self; Tigermed. Other; Self; Eli Lilly & Company. M. Konig: Employee; Self; Eli Lilly & Company. Stock Owner; Self; Eli Lilly & Company. Hispanic/Latino (H/L) adults in the US have a greater prevalence of type 2 diabetes (T2D) and diabetes-related complications compared to non-H/L adults. Optimal glycemic control remains a challenge in H/L adults with T2D, resulting in poor health outcomes and a substantial comorbidity burden. This study compared glycemic and cost outcomes among H/L adults with T2D who initiated treatment with dulaglutide (DU) or basal insulin (BI). This retrospective cohort study used claims data from the Optum Research Database from July 2014 - March 2022 (commercial and Medicare Advantage). H/L patients with ≥1 pharmacy claim for DU or BI during the identification period (IDP), January 01, 2015 - March 31, 2021, were included. Index date was the date of first claim for DU or BI during the IDP and cohorts were assigned based on index date medication. Patients were required to have continuous enrollment during the 6-month baseline and 12-month follow-up periods; ≥1 claim with baseline T2D diagnosis; ≥1 glycated hemoglobin (HbA1c) test result each in the baseline period and 4-12 months after index date; and no injectable T2D medications at baseline. Cohorts were propensity-score matched on demographic and baseline clinical characteristics. Measures of glycemic control included change in HbA1c from baseline to 12-month follow-up, which was modelled using ordinary least squares regression (OLS) on the matched cohorts, and proportion of patients achieving the HbA1c target of <7.0%. Bootstrap sampling with 5,000 samples was used to compare total costs per 1% change in HbA1c between cohorts. The propensity-score matched cohorts comprised of 2,872 patients with 1,436 patients in each cohort (DU and BI). Most demographic and baseline characteristics were balanced between the cohorts except for age, income, office visit counts, and medical costs. Mean [standard deviation (SD)] age in the DU and BI cohorts was 59.8 [12.8] and 61.4 [12.9] years, respectively. About half of the matched population was male (DU: 54.3%; BI: 51.4%). At 12 months, the mean change from baseline in HbA1c was significantly greater in the DU cohort than BI (DU: -1.4% [1.9]; BI: -0.9% [2.1]; p<0.001) and the total cost per 1% change in HbA1c was significantly lower in the DU cohort than BI (DU: $13,768; BI: $19,128; p<0.001). The proportion of patients achieving HbA1c <7.0% at 12 months was significantly higher among patients initiating DU than BI (DU: 37.9%; BI: 22.1%; p<0. 001). Results from OLS showed that the DU cohort had a larger decrease in HbA1c compared to BI (estimate: -0.57; p<0.001). Patients initiating DU had twice the odds of achieving HbA1c <7.0% compared to BI (odds ratio: 2.19; p<0.001). DU demonstrated better glycemic outcomes and lower costs per 1% HbA1c reduction among H/L adults with T2D compared with those initiating BI. DU may be an effective therapeutic option for this vulnerable population and may help with achieving improved T2D outcomes. Presentation: Friday, June 16, 2023