Long-Term Health Care Utilization Costs of GLP-1RA Use vs Bariatric Metabolic Surgery

Importance: Obesity and type 2 diabetes are major global health concerns associated with substantial clinical and economic burdens. Bariatric metabolic surgery (BMS) and glucagon-like peptide-1 receptor agonists (GLP-1RAs) are effective interventions, but their long-term impact on health care utilization costs remains incompletely understood. Objective: To compare short- and long-term health care utilization costs among patients with obesity and diabetes treated with GLP-1RAs vs those who underwent BMS. Design, Setting, and Participants: This observational, retrospective cohort study used electronic medical records from Clalit Health Services, the largest health care organization in Israel. Adults (aged ≥24 years) with obesity (body mass index [BMI] ≥30) and diabetes who underwent their first BMS or initiated GLP-1RA therapy between January 1, 2010, and December 31, 2022, were identified. Propensity score matching yielded 2721 matched pairs, who were followed up for up to 12.5 years (mean [SD], 6.5 [3.5] years), through December 31, 2023. Data cleaning and analyses were conducted from June to October 2025. Exposures: Treatment with GLP-1RAs or BMS. Main Outcomes and Measures: Health care utilization costs incurred by the health system, excluding costs of the index interventions. Attributable costs were estimated using a difference-in-differences approach and modeled with multivariable linear regression. Results: The matched cohort included 5442 adults (mean [SD] age, 51 [10] years; 3243 women [59.6%]; mean [SD] BMI, 40.5 [6.0]). Mean (SD) monthly follow-up costs were higher among patients treated with GLP-1RAs than those who underwent BMS ($415.3 [$746.7] vs $304.9 [$628.5]). The adjusted difference-in-differences method showed $109.0 higher monthly costs per patient for GLP-1RA treatment (SE = $17.0; P < .001), associated primarily with hospitalization costs ($43.9; SE = $13.3; P = .001) and non-GLP-1 medication costs ($52.5; SE = $4.4; P < .001). Most cost differences accrued within the first 4 years. BMS was associated with greater early reductions in BMI and hemoglobin A1c, whereas GLP-1RA treatment showed more modest but sustained effects. Conclusions and Relevance: This cohort study found that, among adults with obesity and diabetes, health care utilization costs were higher with GLP-1RA treatment than with BMS, excluding intervention costs, largely reflecting differences in early clinical trajectories. These findings suggest that, among patients eligible for both interventions, BMS may confer combined clinical and economic advantages over long-term follow-up.