GLP-1 Drug Liraglutide's Heart Benefits from Clinical Trial Apply to Real-World Veterans with Diabetes | Research

The takeaway

The cardiovascular benefits of liraglutide observed in the LEADER trial generalize to US veterans with diabetes — a high-risk population underrepresented in the original trial — with potentially even greater reductions in heart events and death.

MACE HR 0.74 in veterans
Liraglutide's estimated cardiovascular benefit was even larger in the VA population (26% MACE reduction) than in the original LEADER trial (13% reduction), suggesting high-risk patients may benefit most.

What the researchers found

Using transportability analysis — a statistical method that re-weights clinical trial data to match a real-world population — researchers estimated what would happen if the LEADER trial had enrolled VA veterans instead of its original participants.

For major adverse cardiovascular events (MACE): the transported hazard ratio was 0.74 (95% CI 0.61-0.90) in the VA-weighted analysis versus 0.87 (0.78-0.97) in the original LEADER trial, suggesting a 26% risk reduction in veterans compared to 13% in the trial.

For all-cause mortality: HR was 0.71 (0.57-0.89) in the VA-weighted analysis versus 0.85 (0.74-0.97) in LEADER — a 29% reduction versus 15%.

The 3-year absolute risk difference for MACE was 2.0% in the VA-weighted analysis versus 1.6% in LEADER. Results were robust across all sensitivity analyses.

Why it matters

US veterans are a high-risk population for cardiovascular disease — they tend to be older, have more comorbidities, and are predominantly male compared to clinical trial populations. Despite this higher risk, they've been underrepresented in GLP-1 outcome trials. This analysis provides the evidence needed to support broader GLP-1 prescribing within the VA healthcare system, potentially benefiting millions of veterans with diabetes.

How the study worked

Transportability analysis integrating data from the LEADER randomized clinical trial (n=9,336) with real-world data from 357,075 VA patients with diabetes (2015-2023). The researchers used augmented inverse probability weighting after balancing baseline characteristics between LEADER participants and trial-eligible veterans. The primary outcomes were MACE (composite of non-fatal MI, non-fatal stroke, and cardiovascular death) and all-cause mortality. Multiple sensitivity analyses varied the VA cohort composition and balancing variables.

What this study cannot tell us

This is a statistical modeling study, not a new clinical trial in veterans. The analysis assumes that the treatment effect observed in LEADER can be appropriately re-weighted to the VA population, which relies on the assumption that all relevant effect modifiers are captured in the available data. The VA population is predominantly male, limiting applicability to female veterans. The study uses a preprint (medRxiv), so it has not yet undergone peer review. The approach estimates what would have happened, not what actually happened in treated VA patients.

How to read the evidence

This is a high-quality transportability analysis combining randomized trial data (LEADER, a landmark trial) with a large real-world cohort (357,075 VA patients). The statistical methodology is rigorous with multiple sensitivity analyses. However, it is a modeling study, not a direct trial in veterans, and is currently available only as a preprint awaiting peer review.

When this study was published

Published as a 2025 preprint on medRxiv. As it has not yet been peer-reviewed, the findings should be interpreted with appropriate caution, though the methodology and data sources are well-established.

The bigger picture

There's growing recognition that clinical trial results may not apply equally to all populations. Transportability analysis is an emerging statistical method that bridges this gap, allowing healthcare systems to make evidence-based decisions for their specific patient populations. For GLP-1 drugs specifically, this study strengthens the case that cardiovascular benefits may be even greater in higher-risk populations than what the pivotal trials showed — a finding with implications for treatment guidelines, insurance coverage, and public health policy.

Questions still open

Would the even larger cardiovascular benefits suggested for veterans be confirmed in an actual VA-based randomized trial or real-world effectiveness study?
Do these transported benefits apply equally to newer GLP-1 agonists like semaglutide, or are they specific to liraglutide?
Could transportability analysis help identify other populations where GLP-1 drugs might provide disproportionately large cardiovascular benefits?

Common questions

What is a transportability analysis?

It's a statistical method that takes results from a clinical trial and estimates what those results would look like in a different population. Since clinical trials often enroll patients who are younger, healthier, or more diverse than specific real-world groups, transportability analysis adjusts the trial data to account for the demographic and clinical differences. In this case, it re-weighted the LEADER trial results to estimate liraglutide's effects as if the trial had enrolled VA veterans.

Why might liraglutide work even better in veterans?

Veterans with diabetes tend to have higher baseline cardiovascular risk than typical clinical trial participants. When a treatment reduces relative risk by a certain percentage, the absolute number of events prevented is greater in a higher-risk population — meaning more heart attacks and strokes avoided per person treated. The higher HR reduction (0.74 vs. 0.87) may also reflect that liraglutide's benefits are amplified in people with more cardiovascular risk factors.

Read the original research

Real-world cardiovascular effects of liraglutide: transportability analysis of the LEADER trial.

medRxiv : the preprint server for health sciences

View on PubMed ↗Publisher’s paper ↗

Citation

Josey, Kevin; Liu, Wenhui; Warsavage, Theodore; Medici, Morten; Kvist, Kajsa; Derington, Catherine G; Reusch, Jane E B; Ghosh, Debashis; Raghavan, Sridharan. (2025). Real-world cardiovascular effects of liraglutide: transportability analysis of the LEADER trial.. medRxiv : the preprint server for health sciences. https://doi.org/10.1101/2025.05.12.25327466