After accounting for major health differences between semaglutide users and non-users, researchers identified a striking mortality association that now requires prospective confirmation.

Study: Semaglutide and COVID-19 mortality risk. Image Credit: Love Employee / Shutterstock

In a recent observational study published online as an unedited 'Article in Press' in the journal Scientific Reports, researchers at the University of California, San Francisco, reported that semaglutide, a glucagon-like peptide-1 receptor agonist (GLP-1 RA) used to treat type 2 diabetes and manage weight in people with obesity or overweight, was associated with a lower hazard of coronavirus disease 2019 (COVID-19)-related death, consistent with previous findings.

Analyzing medical and outpatient pharmacy claims, the researchers found that semaglutide use was associated with a 49% lower adjusted hazard of COVID-19-associated mortality, especially relevant since obesity or overweight and type 2 diabetes, the primary conditions for which semaglutide may be prescribed, have consistently been associated with an increased mortality risk. Semaglutide is established to help control blood glucose levels and reduce body weight, although the study did not measure changes in these outcomes. Instead, it identified an association with a lower hazard of COVID-19-related death across the overall insured cohort rather than specifically among people with documented SARS-CoV-2 infection.

Millions of people have experienced COVID-19, with severe illness and death occurring disproportionately among vulnerable individuals, especially among people living with diabetes or obesity. These individuals may be more susceptible to severe COVID-19 and related complications and face a higher risk of death. Emerging evidence suggests broader therapeutic potential of GLP-1 RAs beyond glucose regulation and weight management, prompting researchers to investigate their potential role in improving COVID-19 outcomes.

About the Study

In the present study, researchers investigated the association between semaglutide use and COVID-19-related mortality in a large insured population.

The team analyzed data from the Optum Labs Data Warehouse (OLDW), including outpatient pharmacy and medical claims for Medicare Advantage and commercial insurance enrollees. The analysis included people enrolled from 1 July 2021 through 30 June 2022 who also had at least 365 days of continuous enrollment before 1 July 2021. The researchers compared their results with previously reported findings from a secondary analysis of the Semaglutide Effects on Cardiovascular Outcomes in People with Overweight or Obesity (SELECT) trial.

The study sample comprised 10,109,596 individuals, with a mean age of 47 years. The team defined COVID-19-related death using discharge-status codes and an International Classification of Diseases, Tenth Revision (ICD-10) U07.1 code recorded within 30 days before death. They identified semaglutide users based on mentions of the terms ‘Ozempic’, ‘Wegovy’, or ‘Rybelsus’, or the Current Procedural Terminology (CPT) code J3590 in the pharmacy claims. They also searched for the names of other drugs and biologics that could potentially influence the results, including systemic corticosteroids (SCs), anti-cluster of differentiation 20 (anti-CD20) monoclonal antibodies, and molecules inhibiting tumor necrosis factor-alpha (TNF-α) or interleukin-6 (IL-6).

The team calculated the incidence rates (IRs) for COVID-19-related deaths per 1,000 person-years (PY) among those who used semaglutide and those who did not. Using logistic regression models, they estimated propensity scores that incorporated demographics, comorbidities, immunosuppressive medication use, COVID-19 vaccination, and prior infection, and subsequently performed inverse probability treatment weighting (IPTW). Cox regression analysis, stratified by birth year, helped them determine the hazard ratios (HRs) of COVID-19-related deaths associated with diabetes, obesity, and semaglutide use. The Cox model was additionally adjusted for diabetes and obesity because these factors remained imbalanced after weighting.

Results and Discussion

Among the study participants, the majority were White (62%) and female (51%), while the South was the most commonly represented region (42%), with a mean Charlson Comorbidity Index (CCI) of 1.12. Obesity, 15%, and diabetes, 13%, were the most prevalent comorbid conditions, while nearly one in ten participants had heart disease. Less than 8% of participants had a history of SARS-CoV-2 infection, and approximately 45% had received at least 1 dose of SARS-CoV-2 vaccine. While smoking status remained unknown for most, 90% of the participants, 7.7% of the overall cohort were classified as never smokers and 2.2% as current or former smokers.

The team found that semaglutide use was associated with a 49% lower adjusted hazard of COVID-19-related death, consistent with the estimated 34% risk reduction in the SELECT trial, while extending these findings to a large real-world cohort of commercial insurance and Medicare Advantage enrollees. The authors proposed anti-inflammatory effects, reduced airway hyperresponsiveness, and obesity treatment as possible explanations for the observed relationship. However, the study did not measure inflammation, airway function, respiratory symptoms, or exacerbations, leaving these mechanisms as hypothesis-generating and requiring further investigation.

Among semaglutide users, the team noted 192 COVID-19-related deaths in 96,842 PY, with an IR of 1.98. Among non-users, 13,591 COVID-19-related deaths were reported in 8,692,940 PY, with an IR of 1.56. Although the crude incidence rate was higher among semaglutide users, these participants were older and had substantially higher prevalences of diabetes, obesity, heart disease, and other conditions associated with mortality. The weighted and adjusted HR for COVID-19-related mortality associated with semaglutide use was 0.51 (95% confidence interval, 0.43 to 0.61; P < 0.001). Diabetes, HR 2.56, and obesity, HR 1.33, were associated with a higher hazard of COVID-19-related death, aligned with earlier evidence.

Conclusions

The findings support an association between semaglutide use and lower COVID-19-related mortality across the overall cohort and in analyses restricted to people with diabetes or obesity. Looking ahead, prospective research, including randomized controlled trials, is needed to determine whether the association is causal and clarify the underlying biological mechanisms. Future studies should also examine dose-response relationships, treatment duration, and whether semaglutide is associated with mortality from other causes.

Although the findings are encouraging, extending research to people without commercial insurance or Medicare Advantage would improve the generalizability and real-world relevance. The claims-based outcome may also have missed deaths when COVID-19 was not documented, while unmeasured confounding remains possible despite the statistical adjustments. The reported E-value of 3.31 suggests that an unmeasured confounder would need relatively strong associations with both semaglutide use and COVID-19-related mortality to fully explain the finding, but it does not eliminate that possibility. Because this observational analysis cannot establish causation, the findings should not be interpreted as evidence for prescribing semaglutide specifically to prevent COVID-19-related death.