Older adults are represented in the major GLP-1 trials but are not the population those trials were designed around. Age-stratified subgroup analyses provide the available evidence, and reading them requires understanding what a subgroup analysis can and cannot establish.
Key Takeaways
- Age-stratified subgroups in the major trials generally show efficacy consistent with the overall population.
- Subgroup analyses are typically not powered for definitive conclusions and are best read as consistency checks rather than independent findings.
- The clinically distinctive concerns in older adults — sarcopenia, bone density, falls, polypharmacy — were not primary endpoints in any registration trial.
- Trial inclusion criteria systematically exclude the frailest older adults, so the studied population is healthier than the clinical population.
- The SELECT cardiovascular outcomes trial included a substantial older cohort and is among the more informative sources for this group.
How to read a subgroup analysis
Pre-specified subgroup analyses partition the randomized population by a baseline characteristic and estimate the treatment effect within each stratum. Randomization holds within subgroups defined at baseline, so confounding is not the main threat. Power is.
A trial powered for its overall population is generally underpowered within any stratum. That produces two failure modes: a real effect can fail to reach significance in a subgroup purely because of sample size, and apparent differences between subgroups can arise from noise. The statistically appropriate question is whether there is evidence of interaction — a genuine difference in effect by age — rather than whether each subgroup independently reaches significance.
For age subgroups across the GLP-1 obesity programme, the general pattern reported has been consistency of effect rather than evidence of interaction.
The inclusion-criteria problem
This is the more consequential limitation, and it is not a statistical one. Trial protocols exclude participants with significant renal or hepatic impairment, recent cardiovascular events, certain malignancies, and other conditions whose prevalence rises steeply with age. Some trials impose upper age limits directly.
The result is that a 72-year-old in a registration trial is systematically healthier than a typical 72-year-old in clinical practice. Subgroup results describe the fitter tail of the older population. Extrapolation to frail older adults, or to those on extensive polypharmacy, is extrapolation beyond the studied population — not a conclusion the data supports.
| Concern in older adults | Addressed in registration trials? | Evidence status |
|---|---|---|
| Weight-loss efficacy | Yes — via age subgroups | Broadly consistent with the overall population |
| Cardiovascular outcomes | Yes — SELECT and related programmes | Substantial older representation |
| Sarcopenia and functional decline | No — not a primary or key secondary endpoint | Evidence gap |
| Bone mineral density and fracture | No | Evidence gap |
| Falls risk | No | Evidence gap |
| Drug–drug interactions in polypharmacy | Limited — excluded populations | Extrapolation required |
| Frailty and nutritional adequacy | No — frail participants largely excluded | Evidence gap |
SELECT and the cardiovascular question
The SELECT trial evaluated semaglutide 2.4 mg in adults with overweight or obesity and established cardiovascular disease, without diabetes, with major adverse cardiovascular events as the primary endpoint. Results were published in the New England Journal of Medicine in 2023 and reported a significant reduction in the primary composite endpoint versus placebo.
SELECT is relevant here for two reasons. Its enrolment criteria required established cardiovascular disease, which skews the population older and sicker than the STEP obesity trials. And its endpoint is a hard clinical outcome rather than a surrogate. For the older adult with established cardiovascular disease, it is the closest available evidence to their situation.
The concerns the trials did not measure
Sarcopenia. Age-related muscle loss proceeds independently of weight change. Superimposing substantial weight reduction — a portion of which is lean tissue in every weight-loss modality — onto an already declining trajectory is the concern most frequently raised by geriatric clinicians. No registration trial measured muscle mass or function as an endpoint in an older subgroup.
Bone. Weight loss is associated with bone mineral density reduction in the general weight-loss literature, and older adults start from a lower baseline. Fracture outcomes were not a designed endpoint.
Nutritional adequacy. Appetite suppression in a population already at risk of inadequate protein and micronutrient intake is mechanistically concerning. It was not systematically assessed.
Polypharmacy. Delayed gastric emptying can alter the absorption profile of concurrently administered oral medications. Older adults take more of them, and the excluded-population problem means the trials contain few participants with complex regimens.
Frequently Asked Questions
Is efficacy reduced in older adults?
Age-stratified subgroup analyses have generally shown effects consistent with the overall trial populations, without evidence of a clear age interaction. These analyses are consistency checks rather than independently powered findings.
Is there an upper age limit for these medications?
Labelling and clinical guidance vary by jurisdiction and product. The more meaningful constraint is that trial populations thin out and become progressively less representative at older ages.
What about muscle loss specifically in older adults?
This is the clearest evidence gap. No registration trial used muscle mass or physical function as an endpoint in an older subgroup, so recommendations rest on extrapolation from the general sarcopenia and weight-loss literature.
Does the SELECT result apply to older adults without cardiovascular disease?
SELECT enrolled participants with established cardiovascular disease by design. Applying its findings to older adults without that history extends beyond the enrolled population.
References
- Lincoff AM et al. Semaglutide and Cardiovascular Outcomes in Obesity without Diabetes. New England Journal of Medicine, 2023 (SELECT).
- Wilding JPH et al. Once-Weekly Semaglutide in Adults with Overweight or Obesity. New England Journal of Medicine, 2021 (STEP 1).
- Villareal DT et al. Weight Loss, Exercise, or Both and Physical Function in Obese Older Adults. New England Journal of Medicine, 2011.
- Batsis JA, Villareal DT. Sarcopenic obesity in older adults: aetiology, epidemiology and treatment strategies. Nature Reviews Endocrinology, 2018.
Comparing providers, not just trials?
Our homepage keeps a maintained side-by-side grid of GLP-1 telehealth providers with current pricing, formats and disclosures. That is where the provider comparisons live — this research desk stays citation-first.
Open the provider comparison grid