Does a Weight-Loss Drug Prevent Heart Attacks? Reading the SELECT Trial

Yes, but only for a narrow group, and only for the reasons the trial was built to test. In the SELECT trial, once-weekly semaglutide lowered major cardiovascular events by roughly a fifth in adults who already had heart disease and carried extra weight but did not have diabetes. The finding is credible because it comes from counting real events, heart attacks, strokes, and cardiovascular deaths, not from a number on a scale. What the trial does not do is promise that a weight-loss injection protects the heart of an otherwise healthy person.

Key points#

• SELECT enrolled 17,604 adults with prior heart disease, a body-mass index of 27 or higher, and no diabetes.
• The primary outcome (cardiovascular death, nonfatal heart attack, or nonfatal stroke) fell about 20 percent with semaglutide versus placebo.
• The absolute gap was near 1.5 percentage points over about three and a half years, roughly one event avoided per 67 people treated.
• The result applies to secondary prevention (people who already had an event), not to weight loss in the general population.
• Benefit came with real harms: gastrointestinal side effects drove twice the discontinuation rate.

Why this trial was different#

For years, the heart data on this class of medicine came almost entirely from people with type 2 diabetes. That left a loophole in interpretation: any benefit to the heart could be chalked up to better blood sugar rather than to the drug itself. SELECT closed the loophole by enrolling only participants without diabetes. If the medicine still protected hearts here, the protection had to come from somewhere other than glucose control. That design choice is what makes the trial worth reading closely, and it is the first large study of its kind to return a positive answer in a population free of diabetes.

Published in the New England Journal of Medicine in November 2023 by A. Michael Lincoff and colleagues, SELECT (Semaglutide Effects on Cardiovascular Outcomes in People With Overweight or Obesity) followed 17,604 adults aged 45 and older for a mean of about 40 months. Everyone had survived a prior heart attack or stroke or lived with symptomatic peripheral artery disease, and everyone carried a body-mass index of at least 27. Half received weekly injectable semaglutide, half a matching placebo.

The number that matters, and the one that sells#

Coverage of these drugs almost always leads with weight. Weight is a surrogate endpoint: a convenient stand-in that we hope tracks the outcomes patients actually live and die by. The trouble with surrogates is that a drug can move them handsomely and still fail to help, or even do harm. Picture a thermostat reading that looks perfect while the furnace behind the wall is failing. SELECT was built to check the furnace, not the display. It counted strokes and heart attacks directly, and that is why its evidence carries a different kind of weight than a study of pounds lost.

Here is what the counting showed. A primary event occurred in 6.5 percent of the semaglutide group against 8.0 percent on placebo: a hazard ratio of 0.80 (95 percent confidence interval, 0.72 to 0.90; P less than 0.001). That is a 20 percent relative reduction and an absolute difference of about 1.5 percentage points, which works out to roughly one event prevented for every 67 people treated over nearly three and a half years. Break the composite apart and the picture is honest but uneven: nonfatal heart attack fell most clearly (hazard ratio near 0.72), while cardiovascular death and nonfatal stroke leaned the same direction without reaching firm ground on their own. The bundle is convincing; not every strand inside it is.

So was it the weight loss?#

It is the obvious story, and the trial will not quite tell it. Participants on semaglutide shed about 9.4 percent of body weight versus 0.9 percent on placebo, and a later Nature Medicine analysis found that this loss kept building for more than a year and held for up to four. Yet the event curves began parting company early, before much weight had come off. The drug also nudged blood pressure down, cooled inflammatory markers, and reduced the onset of new diabetes. As the American College of Cardiology summary put it, the mechanism is most likely multifactorial. Crediting the scale alone reads more into the data than the data support.

Who the answer is for#

A trial result belongs to the people who were actually in it. SELECT enrolled middle-aged and older adults with established cardiovascular disease and excess weight but no diabetes. That is secondary prevention: treating people who have already had an event to prevent the next one. If you have a high BMI and clean coronary arteries, the study says nothing dependable about you, and nothing about using the drug for cosmetic weight loss. Stretching this finding to primary prevention, or to the general consumer market, is exactly the leap the enrollment criteria rule out.

Benefit, with a bill attached#

None of this arrived free. Adverse events leading to permanent discontinuation hit 16.6 percent in the semaglutide group versus 8.2 percent on placebo, driven mostly by gastrointestinal intolerance, per the NEJM report. The trial was funded by the drug's manufacturer, an arrangement that careful readers note rather than dismiss. So the finding is genuine and also modest: a 20 percent cut in hard events for a high-risk group, bought with real side effects and real dropout. Both things are true at once.

The habit worth carrying past this one molecule is simple. When a product is sold to you on how much weight it removes, the sharp question is whether anyone has shown it moves the outcomes that weight is only supposed to predict. SELECT did that work for one drug in one narrow group. Separating a hard endpoint from a flattering surrogate, and asking precisely whom the data cover, is the discipline to bring to your next headline.