Presentation (CTI)

OBJECTIVES: Over the past decade, clinical guidelines and real-world practice have increasingly positioned glucagon-like peptide-1 receptor agonists (GLP-1 RAs) earlier in the management of type 2 diabetes, reflecting benefits beyond glycemic control. Concurrently, baseline HbA1c levels reported in diabetes trials and observational cohorts have declined over time, suggesting earlier treatment escalation in routine care. We evaluated temporal trends in HbA1c levels before and after GLP-1 RA initiation from 2015-2024.
METHODS: We conducted a retrospective cohort study using U.S. administrative claims linked to electronic medical record laboratory data. Among 60,362,590 adults with type 2 diabetes (ICD-10: E11*) identified between 2015 and 2024, only patients initiating a GLP-1 RA after their diabetes index date were included. HbA1c values were identified using LOINC codes, with valid laboratory results available for 1,611,196 patients. Mean HbA1c was calculated in 3-month increments from 18 months pre-index through 18 months post-index. Annual trends in mean HbA1c were evaluated using linear regression with calendar year as a continuous predictor.
RESULTS: Approximately 16% of patients with type 2 diabetes initiated a GLP-1 RA following their diabetes index date. Mean HbA1c at GLP-1 initiation declined from 8.52% in 2015 to 7.33% in 2024, corresponding to an average annual decrease of 0.13%. Similar downward trends were observed in the 6- and 18-month pre-index periods, indicating progressively earlier initiation over time. Despite lower baseline HbA1c, post-initiation glycemic improvements remained stable across calendar years, with an average reduction of approximately 1.1% percentage points at 6 months post-index and sustained improvement through 18 months.
CONCLUSIONS: From 2015 to 2024, GLP-1 RAs were initiated at lower baseline HbA1c, yet the post-initiation glycemic benefit remained constant. This change in treatment initiation relative to HbA1c levels impacts real-world effectiveness and economic evaluations. Patient-level data are needed to assess variability in treatment response and long-term outcomes.