GLP-1 Medication and Heart Disease Risk in Women

The trial behind the headline, and who was actually in it

The claim that a GLP-1 "reduces cardiovascular risk" traces back mostly to one landmark trial: SELECT, which randomized over 17,600 adults with overweight or obesity and existing cardiovascular disease to semaglutide or placebo and found a real, statistically significant reduction in cardiovascular death, non-fatal heart attack, or non-fatal stroke — a 20% relative risk reduction over roughly three years of follow-up¹. That result is genuine and important. What gets left out of most coverage is who was in the trial: SELECT's own published baseline-characteristics paper reports the study population was 72.5% male² — meaning fewer than three in ten participants were women. Women were underrepresented, but the investigators also analyzed cardiovascular outcomes by sex; that analysis is more informative than enrollment proportions alone. ⁷

This is not unique to SELECT — it's a pattern

This gap isn't a flaw specific to one study; it's a documented pattern across cardiovascular research generally. An analysis of women's participation in cardiovascular clinical trials from 2010 to 2017 found consistent underrepresentation relative to the actual burden of cardiovascular disease in women³, and a 2024 follow-up analysis found the same underrepresentation persisting, with trial design characteristics still shaping how many women get enrolled⁴. This matters beyond statistics: heart disease in women often presents differently than the "textbook" pattern established from largely male study populations — narrative reviews of ischemic heart disease in women describe both delayed diagnosis and different symptom patterns as a consequence of decades of cardiology research being disproportionately built on male subjects⁵⁶. A trial that is roughly three-quarters male sits squarely inside that broader pattern, not outside it.

Why this is a genuinely women's-health question, not a technicality

Cardiovascular disease is the leading cause of death in women, not a secondary concern behind breast cancer or other conditions more commonly associated with women's health — a fact both major reviews cited above open with, because it is so often underestimated³⁵. The investigators' 2024 post hoc analysis included 4,872 women, or 27.7% of the randomized population. For major cardiovascular events, the hazard ratio was 0.84 in women (95% confidence interval 0.66–1.07) and 0.79 in men (0.70–0.90). The sex-by-treatment interaction P value was 0.63, with no detected difference in treatment effect by sex. ⁷

The estimate in women was less precise, and its confidence interval included no difference from placebo. That does not establish that semaglutide fails to benefit women: the investigators reported consistent cardiovascular effects across sexes. Underrepresentation remains a limitation, while the interaction analysis does not prove that the effects are identical. ⁷

What this means in practice

SELECT supports cardiovascular risk reduction with semaglutide in the population it studied: adults with overweight or obesity and established cardiovascular disease, without diabetes. The analysis by sex does not support excluding women from that finding. It also does not establish the same benefit for every woman considering a GLP-1. ¹⁷ Your cardiovascular risk factors (blood pressure, lipids, family history) still need individual assessment rather than assuming a GLP-1 alone rebalances them. This ties directly into the broader risk picture our perimenopause and menopause guide covers, since cardiovascular risk rises sharply for women after menopause as estrogen's protective effects fade — timing that overlaps heavily with when many women start a GLP-1 in the first place. It's also worth knowing that SELECT tested semaglutide specifically; our semaglutide versus tirzepatide comparison covers what is and isn't established for each molecule individually, since cardiovascular outcomes data doesn't automatically transfer between them. This guide is educational only and not medical advice.