Shifting obesity treatment paradigms: Trends of glucagon‐like peptide‐1 receptor agonists and bariatric surgery in the United States

1. BACKGROUND

Two‐thirds of adults in the United States have overweight or obesity, placing them at significant risk for debilitating chronic diseases. ¹ It is estimated that over half of the US population will have obesity by 2030. ² Guidelines for the treatment of obesity recommend first‐line treatment with lifestyle interventions with reduced calorie intake, physical activity and behavioural therapy. While this approach can be effective in some people, even if effective initially, weight loss via lifestyle interventions can be difficult to maintain, despite intensive counselling and monitoring. ³ Bariatric surgery is an effective obesity treatment, but it is associated with a risk of short‐and long‐term gastrointestinal complications. ⁴

Glucagon‐like peptide‐1 receptor agonists (GLP‐1 RAs) have emerged as a promising therapeutic option, demonstrating efficacy in improving glycaemic control and promoting weight loss. ⁵ , ⁶ , ⁷ GLP‐1 RAs and glucose‐dependent insulinotropic polypeptide (GIP)/GLP‐1RAs (e.g., tirzepatide) vary in molecular structure and receptor binding affinity, ⁸ potentially leading to differences in effectiveness, tolerability, discontinuation rate and adverse outcomes among individual drugs. These medications were introduced as antidiabetic drugs but later approved for chronic weight management, with the first approval by the Food and Drug Administration (FDA) in December 2014. Recent large observational studies suggest that bariatric surgery and GLP‐1 RAs are associated with reductions in adverse cardiovascular outcomes. ⁹ , ¹⁰ Despite the potential benefits of the two approaches to curb the obesity pandemic, data on their real‐world use remain limited. This study aimed to quantify the population‐level displacement effect of GLP‐1 RA adoption on bariatric surgery utilisation.

2. METHODS

This retrospective cohort study analysed de‐identified, patient‐level healthcare claims data from MarketScan IBM from January 1, 2015, to December 31, 2022. Estimates for bariatric surgery from the period prior to the start of our study (2011–2014) were obtained from a previously published study that used the same data source. ¹¹ The current cohort consisted of patients who (a) initiated treatment with GLP‐1 RAs (i.e., liraglutide, semaglutide) or GIP/GLP‐1RAs (e.g., tirzepatide) and had a prior obesity diagnosis or (b) had documented procedure codes for bariatric surgery. We defined bariatric surgery using the same diagnosis and procedure codes used by Alalwan et al. ¹¹ to enhance estimate comparability. GLP‐1 RA exposure was ascertained using outpatient pharmacy prescription data. We focused on GLP‐1 RAs and GIP/GLP‐1RAs approved by the FDA for weight loss, including liraglutide, semaglutide and tirzepatide. We described baseline demographic and clinical characteristics across study years for users of GLP‐1 RAs and bariatric surgery. Temporal trends in utilisation were assessed using the Cochran‐Armitage test, which is appropriate for evaluating trends in proportions across ordered groups. The primary analysis was stratified by age and sex, given prior evidence of demographic differences in obesity treatment uptake. Bonferroni correction was applied to account for multiple comparisons. We conducted sensitivity analyses to assess the robustness of our findings, including: (1) requiring at least two GLP‐1 RA prescriptions to address the possibility of primary nonadherence and (2) excluding patients with a diagnosis of type 2 diabetes to reduce the likelihood that prescriptions were issued for glycaemic control rather than weight management. The study was approved by the Vanderbilt University Medical Center Institutional Review Board.

3. RESULTS

We identified a total of 299 755 patients who underwent bariatric surgery between 2011 and 2022 (Figure 1). Baseline characteristics remained relatively stable across the study period (Tables S1 and S2). The annual number of bariatric surgeries declined by 65%, from 40 818 procedures in 2011 to 14 184 in 2022 (P for trend <0.001). Importantly, this downward trend began well before the introduction of GLP‐1 RAs as anti‐obesity agents. The most pronounced decline occurred between 2011 and 2015, with procedures decreasing by approximately 47% during that period (from 40 818 to 21 798). We observed a 49% decline in bariatric surgery procedures between 2016 and 2020 (from 24 018 to 12 365), followed by a 15% rebound in 2021, with 14 172 procedures performed. In contrast, the annual use of GLP‐1 RAs increased by 1451% from 29 421 in 2016 to 129 608 in 2022 (p‐value for trend <0.001). Among these, semaglutide was the most commonly used agent by 2022, with 91 849 prescriptions for semaglutide compared with 22 243 for liraglutide and 15 516 for tirzepatide.

FIGURE 1.

Trend of utilisation of GLP‐1 RAs and bariatric surgery at the national level between 2011 and 2022. GLP‐1 RAs, glucagon‐like peptide‐1 receptor agonists. A total of 3964 users of GLP‐1 RAs underwent a bariatric surgery procedure, median time from GLP‐1 RAs initiation to bariatric surgery procedure 505 days. (25th percentile, 224 days, 75th percentile 1001 days). *The number of bariatric surgery procedures for years 2011‐2014 was obtained from a study by Alalwan et al. that used MarketScan IBM data. Approval dates for GLP‐1 RAs including liraglutide, semaglutide and tirzepatide are 2010, 2017 and 2022 for type 2 diabetes. The same medications were approved for weight loss in 2014, 2021 and 2023.

Figures S1a,b demonstrate that individuals aged 35–55 years consistently accounted for the highest proportion of GLP‐1 RA and bariatric surgery use across all study years. However, we observed a notable increase in utilisation among individuals under 35 years of age over time. For GLP‐1 RAs, the proportion of users under age 35 tripled from 5.1% in 2015 to 15.3% in 2021 and 2022. Similarly, for bariatric surgery, this age group increased from 24.0% in 2015 to 36.8% in 2021 before a slight decline in 2022. These trends suggest growing adoption of both treatment strategies among younger individuals, even as middle‐aged adults (35–55 years) remained the predominant users throughout the study period. Utilisation of both treatments has increased among females over time (Figure S2).

Of GLP‐1 RA users, only 3964 (1.1%) proceeded to bariatric surgery. The median duration between the initiation of GLP‐1 RA therapy and bariatric surgery was 505 days (25th percentile, 224 days; 75th percentile, 1001 days). Results from stratified analysis indicate that GLP‐1 RAs were less often utilised among individuals 35 years or under when compared to bariatric surgery (Figure 2). Females were more likely to use GLP‐1 RAs and bariatric surgery in comparison to males. Results from sensitivity analyses showed consistent trends in GLP‐1 RA uptake and bariatric surgery utilisation when requiring two or more GLP‐1 RA prescriptions and when excluding individuals with type 2 diabetes.

FIGURE 2.

Overall utilisation of GLP‐1 RAs and bariatric surgery at the national level by age and sex. Using pooled data across the study years. GLP‐1 RAs, glucagon‐like peptide‐1 receptor agonists.

4. DISCUSSION

This population‐based analysis in the United States highlights a significant shift in obesity treatment, marked by a 65% reduction in bariatric surgery, accompanied by a 14‐fold increase in the utilisation of GLP‐1 RAs. Unlike the study by Lin et al., ¹² which examined trends over a shorter period (2022–2023), our research analysed utilisation patterns over a full decade. Consistent with their findings, we observed a marked increase in GLP‐1 RA use and a modest decline in bariatric surgery utilisation, the latter of which began before the introduction of GLP‐1 RAs. Although the recent decline in bariatric surgery rates may appear to correspond with the increased uptake of GLP‐1 RAs, our data indicate that this trend began prior to the approval of GLP‐1 RAs for obesity management. Multiple factors likely contributed to this decline, including limited insurance coverage, ¹³ and changes in referral practices by healthcare providers. ¹⁴ Although a few observational studies attempted to compare the two approaches for obesity treatment on cardiovascular outcomes, these studies had inherent limitations—lack of randomisation and thus potential unmeasured confounding variables—or compared bariatric surgery patients to diabetes patients, making it difficult to make definitive conclusions until head‐to‐head randomised clinical trials become available. However, the benefits of bariatric surgery underscore the sustained clinical value of bariatric surgery, particularly among patients with more severe obesity, and suggest that it is unlikely to be replaced entirely by GLP‐1 RA therapies.

Although a small portion of GLP‐1 RA users opted later for surgery, the long‐term interplay between surgical interventions for obesity and these medications remains uncertain, particularly given reports of high discontinuation rates for GLP‐1 RAs and the risk for rapid weight gain. Additionally, the lower uptake of GLP‐1 RAs among patients under 35 years, compared to those selecting surgery, points to potential disparities in treatment preferences and access based on age. As GLP‐1 RAs emerge as promising treatment alternatives for obesity, it is essential to evaluate their long‐term effectiveness and safety to inform treatment selection in high‐risk populations. The relatively high uptake of bariatric surgery among individuals under 35 years of age likely reflects the interplay of several factors, including the need for durable, long‐term weight loss solutions; the high discontinuation rates associated with GLP‐1 RAs, which are often followed by weight regain; and the well‐established cost‐effectiveness and sustained metabolic benefits of bariatric surgery. These observations are further supported by recent data showing a 15% increase in adolescent bariatric surgery utilisation between 2021 and 2023, ¹⁵ in alignment with the 2023 American Academy of Paediatrics Clinical Practice Guidelines, endorsing bariatric surgery as a safe and effective treatment option in this population. ¹⁶

This study provides the most up‐to‐date data on the utilisation of bariatric surgery and GLP‐1 RAs using a nationally representative sample of patients with commercial or Medicare Supplemental insurance across the United States. However, several limitations should be noted. First, although we restricted the cohort to patients with a prior diagnosis of obesity, misclassification is possible if GLP‐1 RAs were prescribed primarily for type 2 diabetes rather than weight management. In the absence of clinical notes, we were unable to confirm treatment intent, although our study results remained consistent when excluding patients with prior diabetes diagnosis. Second, we may have underestimated GLP‐1 RA use if prescriptions were filled through specialty pharmacies or obtained via self‐pay, which are not fully captured in claims data. Third, the estimates are based on patients captured in the IBM MarketScan database; shifts in the underlying population over time, due to changes in contributing insurance plans, could affect the observed trends. Fourth, bariatric surgery was identified using ICD procedure codes, which may be subject to misclassification, although we relied on previously validated coding algorithms. Fifth, although patients undergoing bariatric surgery likely have higher baseline body mass index (BMI) than those initiating GLP‐1 RAs, we were not able to ascertain baseline BMI since it is not routinely captured in administrative claims databases. Lastly, the findings are generalisable only to individuals with commercial or Medicare Supplemental insurance and may not reflect trends in populations with other types of coverage or the uninsured.

FUNDING INFORMATION

None.

CONFLICT OF INTEREST STATEMENT

GKD receives funding from the National Institutes of Health and the American Society of Haematology. GKD received honoraria from Valley Health Winchester Medical Center, the University of Florida and the BMS‐Pfizer Alliance for educational talks unrelated to this work. JMS has nothing to disclose. CMS receives funding from the National Institutes of Health.

PEER REVIEW

The peer review history for this article is available at https://www.webofscience.com/api/gateway/wos/peer-review/10.1111/dom.16600.

Supporting information

Table S1. Demographics and clinical characteristics of individuals with obesity who were new users of GLP‐1 RAs across study years.

Table S2. Demographics and clinical characteristics of individuals with obesity who underwent bariatric surgery across study years.

Figure S1. Trends in the use of GLP‐1 RAs and bariatric surgery by age groups.

Figure S2. Trends in the use of GLP‐1 RAs and bariatric surgery among females.

Dawwas GK, Samuels JM, Stein CM. Shifting obesity treatment paradigms: Trends of glucagon‐like peptide‐1 receptor agonists and bariatric surgery in the United States. Diabetes Obes Metab. 2025;27(10):6023‐6026. doi: 10.1111/dom.16600

DATA AVAILABILITY STATEMENT

Data is not publicly available.