A study from Washington University School of Medicine, published in BMJ Medicine, found that stopping GLP-1 drugs such as Ozempic, Wegovy, Mounjaro and Zepbound may mean more than regaining weight. Researchers followed 333,687 U.S. veterans with type 2 diabetes for three years; those who stopped treatment for two years had a 22% higher risk of heart attack, stroke or death than patients who continued. Published on March 18, the study drew renewed attention this week alongside estimates that one in eight U.S. adults has used these medicines.
A stepwise pattern. A six-month break was associated with a 4% higher risk, a one-year break with 14%, and a two-year break with 22%. By contrast, those who continued treatment for three years had an 18% lower risk than the sulfonylurea comparison group—equivalent to about four fewer cardiac events per 100 people over three years. Those who stopped and later restarted partly regained protection: the reduction was 12%, rather than 18%. Lead author Ziyad Al-Aly said: “It takes years to build protection, and half as long to lose it. Stopping leaves a lasting mark.”
Why patients stop. In this study, 26% of users discontinued treatment and 23% had a break longer than six months. A separate study of more than 157,000 people newly prescribed semaglutide found that 49% stopped within the first year. Reasons include cost, side effects such as nausea and vomiting, and supply problems. Al-Aly describes the pattern as a “metabolic whip”: after treatment stops, weight, blood pressure, cholesterol and inflammation can rebound together.
What the study cannot show. As an observational study, it cannot prove cause and effect. People who stopped may already have been less healthy or less likely to follow treatment; researchers used a “target trial emulation” approach to reduce, but not eliminate, this bias. The cohort consisted mainly of male U.S. veterans, so the same figures may not apply to people without diabetes using the drugs for weight loss. The findings are not a reason to stop or change treatment without medical advice.
Patient count, follow-up period, risk figures, discontinuation rates, Al-Aly quotations, journal and publication date: WashU Medicine, Becker’s, Axios, CNBC, ScienceDaily and Fox News reports. The 49% discontinuation figure comes from a separate study reported by NewsNation. The study was published in March and recirculated in September; the estimate of “four fewer events per 100 people” relies on a single secondary source. The full paper was not reviewed; comparable data for people without diabetes taking the drugs for weight loss are not available here.