In the largest cardiovascular trial of semaglutide in people without diabetes, a heart attack, stroke or cardiovascular death occurred in 6.5% of participants assigned semaglutide and 8.0% assigned placebo.

The relative reduction was 20%. The absolute difference was 1.5 percentage points over an average follow-up of 3.3 years—roughly one fewer major event for every 67 people treated over that period, using the rounded trial figures.

Those numbers are clinically useful only after the study population is named. Every participant had already experienced cardiovascular disease. Most had a previous heart attack; others had a previous stroke or symptomatic peripheral artery disease. This was secondary prevention: reducing another event in people whose starting risk was already high. It did not test semaglutide as cardiovascular prevention for otherwise healthy people interested in longevity.

What SELECT studied

SELECT was a double-blind randomised trial conducted at 804 sites in 41 countries. It enrolled 17,604 adults aged at least 45 who had a body mass index, or BMI, of 27 kg/m² or higher, established cardiovascular disease and no diabetes. Participants received once-weekly semaglutide injections, increased towards 2.4 mg, or matching placebo. Both groups continued ordinary cardiovascular care.

The background care is part of the result. At enrolment, 90.1% were taking lipid-lowering medication and 86.2% were taking an antiplatelet medicine. Semaglutide was tested as an addition to established treatment, not as a substitute for cholesterol management, blood-pressure care, smoking cessation or indicated antiplatelet therapy.

The average participant was 61.6 years old, the average BMI was 33.3, and almost three quarters were men. Two thirds had an HbA1c in the prediabetes range. Only 27.7% were women and 16% were non-White, which limits how confidently the average result can be applied across a more diverse Australian population.

The result was a composite, not a fall in every outcome

The trial’s prespecified primary outcome combined three events: cardiovascular death, non-fatal heart attack and non-fatal stroke. It occurred in 569 of 8,803 people assigned semaglutide and 701 of 8,801 assigned placebo. The hazard ratio was 0.80, with a 95% confidence interval from 0.72 to 0.90.

A composite outcome increases the number of events available for analysis, but its components still need inspection. Cardiovascular death occurred in 2.5% with semaglutide and 3.0% with placebo. That comparison did not cross the trial’s prespecified threshold for statistical superiority. Non-fatal heart attack was lower, at 2.7% versus 3.7%, while non-fatal stroke was similar, at 1.7% versus 1.9%. The component results were supportive analyses and were not adjusted for testing several outcomes.

The trial therefore supports a reduction in the combined major cardiovascular endpoint. It does not show a statistically secure reduction in cardiovascular death by itself, nor equal benefit across heart attack, stroke and death.

Weight loss may be part of the pathway, not the whole explanation

Semaglutide acts at the glucagon-like peptide-1 receptor. GLP-1 is a gut hormone involved in appetite, food intake and glucose-dependent insulin signalling. At 104 weeks, average body weight had fallen by 9.39% with semaglutide and 0.88% with placebo.

Blood pressure, waist circumference, glucose measures, lipids and C-reactive protein also moved in favourable directions. These changes offer several plausible routes to lower cardiovascular risk. The event curves separated relatively early, which has raised interest in effects beyond weight loss, including inflammation and vascular biology.

That remains a mechanism question. SELECT randomised the medicine, not the amount of weight lost or a proposed anti-inflammatory pathway. It cannot tell us how much of the cardiovascular result came from weight change, blood pressure, metabolic effects or other actions. Biomarker movement should not be presented as proof of one protective mechanism.

Tolerability changed how often treatment continued

Serious adverse events were reported in 33.4% of the semaglutide group and 36.4% of the placebo group. That broad comparison includes cardiovascular events caused by the underlying disease, so it does not mean side effects were absent.

Adverse events led to permanent treatment discontinuation in 16.6% with semaglutide and 8.2% with placebo. Gastrointestinal problems accounted for most of that difference: they led to discontinuation in 10.0% and 2.0%, respectively. Gallbladder-related disorders occurred in 2.8% versus 2.3%. Confirmed acute pancreatitis was uncommon and was not higher with semaglutide in this trial.

Follow-up lasted about 3.3 years, while cardiovascular prevention may continue for decades. SELECT cannot settle lifetime benefit, long-term tolerability, what happens after treatment stops or whether the result is worth the financial and treatment burden for a particular person. The study was funded by the manufacturer, and several authors were company employees; the randomised design remains strong, but the relationship belongs in a full reading of the evidence.

Where the Australian boundary now sits

The Therapeutic Goods Administration has registered Wegovy, the 2.4 mg semaglutide product tested in SELECT, as an addition to standard care for reducing major cardiovascular events in adults with established cardiovascular disease, BMI at least 27 and no established type 1 or type 2 diabetes.

The Australian Heart Foundation’s 2026 obesity and cardiovascular disease consensus statement gives moderate advice to consider semaglutide for adults with established atherosclerotic cardiovascular disease and BMI at least 27. That wording follows the trial boundary. It does not extend the evidence to someone with no previous cardiovascular event simply because their BMI is above 27, and it does not establish the same outcome for every medicine grouped under the GLP-1 label.

A sensible clinical discussion begins by checking whether the person resembles the population that produced the result. Is there established atherosclerotic disease? What is the risk of another event despite current treatment? Which outcome is being targeted? How would gastrointestinal effects, gallbladder history, treatment burden, access and the likelihood of staying on therapy alter the balance?

The memorable number from SELECT is not only the 20% relative reduction. It is 6.5% versus 8.0%, in people who already had cardiovascular disease, on top of standard care. That is a real cardiovascular result with a clear clinical boundary.

Sources

Evidence note: This Insight is general education, not individual medical advice. It does not recommend starting, stopping or changing semaglutide or any cardiovascular medicine. Eligibility, contraindications, adverse effects and treatment priorities require assessment by a qualified clinician.