FDA Cardiovascular Indication Expansion and Dispensing of Semaglutide/Tirzepatide in CVD Patients With Overweight/Obesity

In March 2024, the U.S. Food and Drug Administration (FDA) approved a new indication for semaglutide to reduce cardiovascular risk in adults with established cardiovascular disease (CVD) and overweight or obesity.^1,2 This expansion followed the SELECT (Semaglutide Effects on Cardiovascular Outcomes in People with Overweight or Obesity) trial, which demonstrated cardiovascular benefit among adults with overweight/obesity and without diabetes.² Whether, and how quickly, this evidence and regulatory change have translated into clinical practice remains uncertain.

By analyzing dispensing data, this study examined trends in the first-time dispensing of semaglutide and tirzepatide among adults with CVD and overweight or obesity. It focused on immediate changes following FDA approval and hypothesized that, although use increased, overall uptake remained substantially below the level expected if all eligible patients were treated— representing a missed opportunity, particularly among those at the higher end of cardiovascular risk.

What is the clinical question being addressed?

How did the FDA’s March 2024 cardiovascular risk-reduction indication for semaglutide influence real-world first-time dispensing for semaglutide and tirzepatide among U.S. adults with established cardiovascular disease and overweight/obesity?

What is the main finding?

In a national cohort of more than 2 million eligible adults observed from January 2021 through May 2025, the U.S. Food and Drug Administration approval was associated with an immediate 23% increase in first-time semaglutide dispensing, followed by a gradual decline in subsequent months. No corresponding immediate change was observed for tirzepatide, which instead demonstrated a steady, sustained rise in first-time dispensing over time. Nevertheless, the overall proportion of eligible adults receiving either drug remained far below the level expected if all who could benefit were treated.

Methods

Data sources

We used Truveta, a continuously updated, linked database of deidentified electronic health records from >30 U.S. health systems. The data set includes demographics, medical encounters, diagnoses, and medication records. Prescription dispensing data, covering fills within and outside Truveta’s network, were obtained via third-party linkages. This study used only deidentified patient records and did not require Institutional Review Board approval.

Study population

For each month (January 2021-May 2025), we constructed a cohort of adults (≥18 years) who had: 1) ≥1 outpatient encounter in the current month and ≥1 encounter in the prior 12 months (“baseline”); 2) a subsequent encounter after the current month (to ensure adequate capture of prescription dispensing); 3) body mass index (BMI) ≥27 kg/m² based on the most recent measurement from baseline or current month; and 4) established CVD documented at baseline or in the current month, defined by ICD-10 codes for myocardial infarction (I21∗, I22∗), ischemic stroke (I60∗-I63∗, I69∗), or peripheral artery disease (I70∗). Individuals were excluded if they had any prior glucagon-like peptide-1 (GLP-1) dispensing or a diagnosis of type 1 diabetes, type 2 diabetes, or gestational diabetes recorded at baseline or in the current month.

Measurements

First-time semaglutide and tirzepatide dispensing was identified using brand names and RxNorm or National Drug Codes. We assessed age (18-34, 35-49, 50-64, and ≥65 years), sex (female, male, and unknown), race (Asian, Black, White, other, and unknown), ethnicity (Hispanic, non-Hispanic, and unknown), and BMI categories (overweight 27-29.9 kg/m²; obesity class 1: 30-34.9 kg/m²; class 2: 35-39.9 kg/m²; class 3: ≥40 kg/m²; and unknown).

Statistical analysis

Descriptive statistics summarized monthly incidence rates (per 100,000 eligible persons) of first-time semaglutide and tirzepatide dispensing. March 2024, the FDA announcement month, was excluded to minimize transitional prescribing bias. To assess the impact of the FDA cardiovascular indication, interrupted time series Poisson regression models with a log link and an offset for the monthly eligible population evaluated changes in level (immediate) and slope (trend) of dispensing before (June 2021-February 2024) and after (April 2024-May 2025) the approval. Wald CIs were used to estimate incidence rate ratios (IRRs). Models adjusted for age, sex, race, ethnicity, and BMI categories, with additional stratified analyses conducted. All analyses were performed in R (version 4.5.1).

Results

Across the study period, 2,099,582 adults with a BMI ≥27 kg/m² and established CVD met the eligibility criteria (44.5% females; 72.3% aged ≥65 years), of whom 16,920 (0.81%) received a first-time prescription and 12,380 (0.59%) were dispensed either semaglutide or tirzepatide between 2021 and 2025.

Semaglutide dispensing was initially low but rose steadily from 13.0 per 100,000 persons in January 2021 to 153.0 per 100,000 by May 2023, followed by a subsequent decline to 65.2 per 100,000 around December 2023, coinciding with publication of the SELECT trial. Following FDA approval, the incidence increased from 78.4 per 100,000 persons in February 2024 to 132.0 and 151.0 per 100,000 in April and May 2024, respectively, followed by a gradual decline to 123.0 per 100,000 in May 2025 (Figure 1A). The interrupted time series analysis demonstrated a significant immediate increase in semaglutide dispensing associated with the FDA approval (IRR: 1.23; 95% CI: 1.14-1.32), with a modest downward monthly trend thereafter (IRR: 0.94; 95% CI: 0.92-0.95) (Figure 1B). Patterns were similar across age, sex, race/ethnicity, and BMI subgroups (Figure 1B).

Figure 1.

Trends in Semaglutide and Tirzepatide Dispensing and Interrupted Time-Series Analysis

Interrupted time series (ITS) Poisson regression models (B) with a log link and an offset for the monthly eligible population assessed changes in first-time semaglutide dispensing before (June 2021-February 2024) and after (April 2024-May 2025) the FDA cardiovascular indication (A). For each subgroup, 3 incidence rate ratios (IRRs) are reported: 1) the preapproval monthly trend, indicating whether dispensing was increasing or decreasing; 2) the immediate level change, comparing the month after approval to the month before; and 3) the postapproval monthly trend, showing whether dispensing continued to rise or decline. IRRs >1 indicate an increase in dispensing, whereas IRRs <1 indicate a decrease. All models were adjusted for age, sex, race, ethnicity, and BMI category, and Wald CIs were used for inference. CIs for trends before and after expansion were plotted but not visible at this scale because they were extremely narrow. CVD = cardiovascular disease; FDA = U.S. Food and Drug Administration.

Tirzepatide did not exhibit a significant immediate level change after the FDA’s cardiovascular indication for semaglutide (IRR: 1.11; 95% CI: 0.96-1.29). However, monthly incidence rose steadily from 4.1 per 100,000 in January 2023 to 92.3 per 100,000 by May 2025. In the postapproval period, tirzepatide exhibited a gradual and sustained increase in incidence, in contrast to semaglutide’s pronounced initial rise followed by a decline.

Discussion

In this large, retrospective analysis, the FDA’s expanded cardiovascular indication for semaglutide was followed by a rapid rise in first-time dispensing, reflecting swift translation of evidence into clinical practice. Despite this increase, overall uptake remained limited relative to the eligible population,³ suggesting persistent barriers to use. As new evidence within this drug class and additional regulatory updates emerge, dispensing trends are likely to continue fluctuating, despite well-established cardiovascular prevention benefits.

Our findings add to the growing literature by characterizing recent trends in semaglutide/tirzepatide dispensing among adults with established CVD and overweight/obesity—the population at the highest cardiometabolic risk and most likely to benefit from these agents. A prior analysis using the Epic Cosmos data set reported that among >39 million eligible adults, only ∼2% had received prescriptions for semaglutide or tirzepatide for obesity treatment through October 2024.³ Leveraging clinically rich Truveta Data, our study extends these observations into 2025 and reveals a similar gap: Despite expanding evidence and regulatory support, those most likely to benefit from GLP-1–based therapies remain under-represented in their use.

A notable strength of our analysis is the inclusion of dispensing rather than prescription data. This distinction is critical, as high-cost medications—despite being evidence-based and guideline-endorsed—may be prescribed but never filled due to affordability or patient preference. Our findings therefore better reflect the real-world medication use and identify the subset of patients actually receiving therapy.

The divergent postapproval trends—declining semaglutide dispensing and rising tirzepatide use—are likely multifactorial. Although the GLP-1 class is increasingly recognized for its cardiometabolic benefits, they may still be perceived primarily as antiobesity medications rather than cardiovascular prevention therapies. Although tirzepatide lacks definitive evidence for CVD risk reduction, its superior weight-loss efficacy—demonstrated in the SURMOUNT-5 trial—may influence recent prescribing behavior.⁴ Observational and clinical experiences suggesting greater metabolic benefits with tirzepatide could also play a role. A recent post hoc analysis reported greater reductions in predicted 10-year CVD risk with tirzepatide than semaglutide, although confirmatory outcome trials are needed to sufficiently claim class effect.⁵

Our analysis has limitations. These include potential residual confounding from unmeasured factors such as marketing and payer policy changes, incomplete capture of out-of-network care, and the absence of direct measures of affordability or prescriber intent. Furthermore, individuals with sparse health care utilization might have been excluded, which could limit generalizability. We did not assess 6- or 12-month persistence; thus, findings reflected initiation and might overestimate sustained use. Finally, we did not evaluate postprescription cardiovascular outcomes, which will be an important area for future research.

Conclusions

Semaglutide dispensing increased sharply following the FDA approval, whereas tirzepatide showed a steady upward trajectory; however, overall uptake remained low compared with the size of the eligible population. These trends suggest that despite clear cardiometabolic benefits, access and implementation barriers persist, underscoring the need for policies that ensure equitable and prioritized use among those at the highest cardiovascular risk.

Funding support and author disclosures

Ms Cartwright and Drs Rodriguez, Do, Gratzl, and Stucky are employed by Truveta Inc; and report holding equity in Truveta. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.