When semaglutide's full range of cardiovascular, kidney, and metabolic benefits are counted, only 11 patients with obesity and heart disease need treatment for 4 years to prevent one adverse event — far better than the headline NNT of 58 for heart events alone.
NNT = 11 at 4 yearsWhen cardiovascular events, hospitalizations, kidney outcomes, and diabetes prevention are all counted, only 11 patients with obesity and heart disease need semaglutide treatment for 4 years to prevent one adverse outcome
What the researchers found
When semaglutide's benefits are measured only by its primary endpoint of major adverse cardiovascular events (MACE), 58 patients need treatment for 4 years to prevent one event (NNT=58). But when the analysis includes hospitalizations, coronary revascularization, non-cardiovascular death, new diabetes onset, and kidney outcomes, only 11 patients need treatment for 4 years to prevent one adverse outcome (NNT=11).
The composite risk reductions were 20% for MACE alone, 20% for the extended composite, and 41% when cardiorenal-metabolic outcomes were included. At just 1 year, the broadest composite NNT was already 20, meaning one patient benefits for every 20 treated.
Why it matters
Number needed to treat (NNT) is how payers, insurers, and health systems decide whether a drug is worth its cost. Semaglutide's headline NNT of 58 for MACE alone might not impress decision-makers, but when its broader benefits — fewer hospitalizations, less kidney disease, prevention of new diabetes — are included, the NNT drops to just 11. This analysis makes a powerful economic and clinical case that semaglutide delivers far more value than its primary trial endpoint suggests, which could influence insurance coverage and prescribing decisions for millions of patients.
The numbers in context
n=17,604 · 39.8 months mean follow-up · MACE risk reduction: 20% · cardiorenal-metabolic risk reduction: 41% · NNT at 4 years: 58 (MACE), 25 (extended), 11 (broadest) · NNT at 1 year: 125 (MACE), 49 (extended), 20 (broadest)
How the study worked
Secondary analysis of the SELECT trial, a randomized, double-blind, placebo-controlled phase III cardiovascular outcomes trial. 17,604 patients with overweight/obesity and established cardiovascular disease (without diabetes) received once-weekly subcutaneous semaglutide or placebo for a mean of 39.8 months. NNTs were calculated for three progressively broader composites: 3-point MACE (CV death, non-fatal MI, non-fatal stroke), an extended composite adding hospitalizations, revascularization, and non-CV death, and a cardiorenal-metabolic composite adding HbA1c ≥6.5% and a nephropathy composite.
Who was studied
17,604 adults with overweight or obesity and established cardiovascular disease without diabetes, from the SELECT trial
What this study cannot tell us
This is a secondary analysis, not the primary SELECT trial itself. The broadest composite (NNTCKM) includes outcomes of varying clinical severity, so a prevented case of new-onset diabetes is counted equally with a prevented heart attack. The SELECT trial excluded patients with diabetes, so these NNTs apply only to the overweight/obese cardiovascular disease population without diabetes. Combining multiple endpoints improves NNT by definition, so the improvement partly reflects methodology rather than additional drug efficacy.
How to read the evidence
This study earns a strong evidence grade as a secondary analysis of the SELECT trial — one of the largest and most rigorous cardiovascular outcomes trials ever conducted for a GLP-1 drug, enrolling 17,604 patients in a randomized, double-blind, placebo-controlled design.
When this study was published
Published in 2025, this analysis builds on the SELECT trial results and is directly relevant to current policy debates about GLP-1 drug coverage for cardiovascular indications.
The bigger picture
The SELECT trial was already landmark evidence that semaglutide reduces cardiovascular risk in people with obesity. This NNT analysis reframes those results in terms that health economists and insurance companies use to make coverage decisions. An NNT of 11 is exceptionally favorable — comparable to or better than statins for some populations. As payers and governments debate whether to cover expensive GLP-1 drugs for cardiovascular protection (not just diabetes or weight loss), this analysis provides ammunition for broader coverage. It also sets a precedent for evaluating other GLP-1 drugs using composite endpoints that capture their multi-system benefits.
Questions still open
- Will these NNT figures influence insurance coverage decisions for semaglutide in cardiovascular indications?
- How do semaglutide's composite NNTs compare to established cardiovascular drugs like statins or SGLT2 inhibitors?
- Would similar composite NNT analyses for tirzepatide show even better numbers given its dual receptor mechanism?
Common questions
What is 'number needed to treat' and why does it matter?
Why does the NNT change so much depending on what outcomes are counted?
Read the original research
The Composite Number Needed to Treat for Semaglutide in Populations with Overweight or Obesity and Established Cardiovascular Disease Without Diabetes.
Advances in therapy, 42(5), 2513-2525
Citation
Lübker, Christopher; Bhavsar, Jigish; Duque do Vale, Ruben; Emerson, Scott S; Nørtoft, Emil; Plutzky, Jorge; Roberts, Geraint; Tarp, Jens Magelund; Lincoff, A Michael. (2025). The Composite Number Needed to Treat for Semaglutide in Populations with Overweight or Obesity and Established Cardiovascular Disease Without Diabetes.. Advances in therapy, 42(5), 2513-2525. https://doi.org/10.1007/s12325-025-03176-w