Semaglutide use was associated with a lower risk of COVID-19-related mortality in a large retrospective cohort study published in Scientific Reports. In an analysis of more than 10 million commercially insured and Medicare Advantage beneficiaries, investigators found that patients receiving semaglutide had a 49% lower hazard of COVID-19-related death compared with patients without semaglutide use following adjustment for demographic characteristics, comorbidities, immunosuppressive medication use, prior COVID-19 infection, vaccination status, and residual imbalance in age, diabetes, and obesity. The investigators wrote that the findings corroborate observations from the Semaglutide Effects on Cardiovascular Outcomes in People with Overweight or Obesity trial and that the observed association merits further investigation of semaglutide's impact on other causes of death.

“[T]hese results support an association in real-world data and merit further investigation of the impact of semaglutide on other causes of death,” wrote lead study investigator Emily Tang, MAS, of the F.I. Proctor Foundation at the University of California in San Francisco, and colleagues.

Investigators analyzed deidentified medical and outpatient pharmacy claims from the Optum Labs Data Warehouse. Eligible patients were continuously enrolled between July 1, 2021, and June 30, 2022, with at least 365 days of continuous enrollment prior to July 1, 2021. Semaglutide exposure was identified using the Current Procedural Terminology code J3590 and pharmacy claims for semaglutide products. COVID-19-related mortality was defined by discharge status codes and an International Classification of Diseases, 10th Revision diagnosis code for COVID-19 recorded within 30 days prior to death. To address measured confounding, the investigators used inverse probability of treatment weighting based on propensity scores that incorporated demographic characteristics, comorbidities, immunosuppressive medication use, COVID-19 vaccination, and prior COVID-19 infection. Cox proportional hazards models stratified by birth year were then used to evaluate the association between semaglutide use and COVID-19-related mortality, with additional adjustment for age, diabetes, and obesity because these variables remained imbalanced following weighting.

The study included 10,109,596 patients with a mean age of 47 years, of whom 51% were female. Among patients exposed to semaglutide (n = 104,878), 192 COVID-19-related deaths occurred during 96,842 person-years of follow-up. Among patients without semaglutide exposure, 13,591 COVID-19-related deaths occurred during 8,692,940 person-years. Although the unadjusted incidence rate of COVID-19-related death was higher among semaglutide users than among nonusers, adjusted analyses showed that semaglutide use was associated with a 49% lower hazard of COVID-19-related mortality. Diabetes and obesity remained associated with a higher risk of COVID-19-related mortality following adjustment, and the association between semaglutide use and lower mortality was also observed in subgroup analyses of patients with diabetes and patients with obesity. The investigators reported that the E-value suggested an unmeasured confounder would need to have a strong association with both semaglutide use and COVID-19-related mortality to explain the observed association.

The investigators noted several limitations, including the retrospective observational design, reliance on ICD-10 codes and discharge status data to identify COVID-19-related deaths, lack of information on semaglutide dose and duration of therapy, and limited generalizability to populations outside commercial insurance and Medicare Advantage plans. They concluded that the findings support an association between semaglutide use and lower COVID-19-related mortality in real-world data and merit further investigation of semaglutide's impact on other causes of death.

Disclosures: Nisha R. Acharya, MD, MS, reported a consulting relationship with Roche. The other investigators declared no competing interests. The study was supported by a core grant from the National Eye Institute to the Department of Ophthalmology at the University of California in San Francisco, and by an unrestricted grant from the Research to Prevent Blindness Foundation. The funding organizations did not participate in the study design, data collection, analysis, interpretation, manuscript preparation, or the decision to submit the study for publication.

Source: Scientific Reports