Explore this article's sources with AI
Follow PeptideMethods on Google
Semaglutide in Very Old Adults
A matched study linked semaglutide to better daily function and heart-failure health status in adults aged 80 or older, without proving causation.
Why we wrote this. A large association in adults aged 80 or older needs to be separated from what a matched observational design can prove.
In this article (5 sections)
A 2026 real-world study associated once-weekly semaglutide use with better daily-living performance and heart-failure health status after 24 months in adults aged 80 or older who also had type 2 diabetes and obesity. After propensity score matching, each group included 232 people. The odds of improvement were higher in the semaglutide group for both outcomes[1][2]. The result is encouraging, but this was an observational comparison rather than a randomized trial. It cannot establish that semaglutide caused the differences.
What the study compared
The investigators studied routine-practice patients aged at least 80 who had heart failure plus type 2 diabetes and obesity. One group received once-weekly semaglutide. The comparison group did not receive a glucagon-like peptide-1 receptor agonist. The co-primary outcomes were improvement in activities of daily living, measured with the Barthel Index, and improvement in heart-failure health status, measured with the Kansas City Cardiomyopathy Questionnaire, at 24 months[1].
Those measures address different parts of life with heart failure. The Barthel Index assesses independence in basic daily activities. The Kansas City Cardiomyopathy Questionnaire records the patient's experience of symptoms, physical limitations, social limitations, and quality of life. The PubMed abstract does not state the score thresholds used to define improvement. It also omits the starting scores and absolute score changes in either group.
The authors used propensity score matching before comparing outcomes, leaving 232 patients in each group[1]. Matching can make treatment groups more similar on recorded characteristics. It cannot balance factors that were not measured or were recorded incompletely. That distinction is especially important in very old adults. Treatment selection may reflect frailty and kidney function, cognitive status and caregiver support, plus other medicines and a clinician's judgment about whether semaglutide is suitable.
What the researchers reported
Semaglutide use was associated with greater odds of improved activities of daily living than the matched control condition. The reported odds ratio was 4.03, with a 95% confidence interval from 1.82 to 7.31. For improved heart-failure health status, the odds ratio was 4.65, with a 95% confidence interval from 1.92 to 7.71. Both comparisons had a reported p value below .01[1]. These are relative odds, not the percentage of patients who improved.
An odds ratio can look large when the underlying event is uncommon, and it does not tell a reader the absolute difference without the event rates in both groups. The abstract does not provide those rates. It also does not report mean changes in the two questionnaire scores. The finding is therefore best described using the published odds estimates, without translating them into an unsupported number of patients helped.
The abstract also reports reductions in heart-failure and all-cause hospitalization, plus cardiovascular death in the semaglutide group. It does not give event counts or effect estimates. Confidence intervals and the statistical plan for those outcomes are also absent[1]. Semaglutide use was also associated with improved glycemic control and lower body weight. Negative correlations linked body weight and NT-proBNP levels with the daily-living and health-status measures, but a correlation does not identify which change produced another.
Why the design limits the conclusion
People were not randomly assigned to semaglutide. A clinician and patient decided whether to use it in ordinary care. That creates confounding by indication and by treatment selection. People who receive one therapy can differ from those who do not in ways that affect function or symptoms and downstream hospitalization or survival. Propensity matching reduces imbalance only for variables included in the model.
The 24-month follow-up is useful because daily function and heart-failure health status matter over time. It also leaves room for changes in other diabetes and heart-failure medicines, rehabilitation, nutrition, illness severity, adherence, or access to care. The abstract does not explain how those time-varying factors were handled, how much outcome data were missing, or whether treatment changes occurred during follow-up.
The enrolled population was narrow. Every patient was at least 80 and had heart failure plus type 2 diabetes and obesity. The result should not be extended to younger adults, people without diabetes, people without obesity, or every heart-failure subtype. Nor does a group average predict whether one older person will gain function or tolerate treatment.
What this means for patients
This paper adds a clinically relevant signal in a group often underrepresented in trials. Daily independence and patient-reported health status can matter as much as a laboratory value. Still, the study does not turn an association into a treatment instruction. Readers looking at semaglutide regulation and access should keep approval and individual suitability separate from this new observational evidence.
For an adult in their eighties with heart failure plus diabetes and obesity, a clinician would need to weigh current symptoms and frailty, kidney function and nutrition, plus other medicines and treatment goals. The study abstract does not provide adverse-event or discontinuation results. It also omits dose details and changes in lean mass. Those omissions prevent a balanced estimate of benefit and harm from the abstract alone.
What we do not yet know
We do not know the absolute proportion of patients who improved or the average number of Barthel or Kansas City Cardiomyopathy Questionnaire points gained. Nor do we know whether the changes exceeded accepted thresholds for an important individual difference. The abstract does not report how outcomes differed by heart-failure subtype and frailty, or by sex, kidney function, or baseline functional status.
A randomized trial in very old adults could better test causation, but it would also need enough follow-up. It should measure function and symptoms, hospitalizations and survival, plus adverse events and discontinuation. Until then, this matched study supports a hypothesis: semaglutide use may be linked to better functional and health-status outcomes in a selected older population. It does not prove the medicine produced those outcomes.
Medical disclaimer: This article is for educational and journalistic purposes only and does not constitute medical advice. Peptides discussed may be classified as prescription medicines or research chemicals depending on your jurisdiction. Always consult a qualified healthcare professional before using any peptide product. PeptideMethods.com does not sell, distribute, or facilitate the sale of any peptide product.
Frequently asked
What did the semaglutide study find in adults aged 80 or older?
Semaglutide use was associated with greater odds of improved daily-living performance and heart-failure health status at 24 months. The observational design cannot prove that semaglutide caused the differences.
How many patients were included?
After propensity score matching, the study included 232 patients treated with semaglutide and 232 controls who did not receive a GLP-1 receptor agonist.
Was this a randomized semaglutide trial?
No. It was a real-world observational study. Propensity matching can reduce measured differences between groups, but it cannot remove unmeasured confounding or establish causation.
Did the study report semaglutide adverse events in very old adults?
The PubMed abstract did not provide adverse-event or discontinuation results. That missing information limits any assessment of the balance between possible benefit and harm.
Sources
No revisions yet. First published .