GLP-1s After 65: The Risks That Differ, and What Monitoring Should Look Like

Why age changes the calculation

Three differences compound. Older adults typically have less lean mass to begin with, so a given percentage loss costs more functionally. Age-related bone density decline means additional loss lands on a lower baseline. And appetite suppression in someone already eating less can shade into inadequate intake more quickly.

None of these makes GLP-1 therapy inappropriate. Obesity in this group carries its own serious risks, and the SUMMIT trial found benefit specifically in obesity-related heart failure with preserved ejection fraction. They make it a treatment that needs monitoring rather than a prescription that needs filling.

What the lean-mass evidence actually shows

Body-composition analyses put the lean-mass share of total weight lost on high-efficacy GLP-1s at roughly 20% to 40%, with reviews of older adults clustering at the higher end.

A phase 2 trial presented at ADA 2026 combining semaglutide with bimagrumab, a myostatin inhibitor, reduced that fraction from about 21% to 7%. That is the first strong evidence the ratio is pharmacologically modifiable rather than fixed.

What the label already permits

Slower escalation is not a departure from the label. Zepbound directs increases in 2.5 mg increments no sooner than every four weeks, based on tolerability and response — a minimum interval, not a schedule. It names 5, 10 and 15 mg as recommended maintenance dosages, and nothing requires reaching the top.

A programme escalating on a fixed calendar without assessment is not following the label's logic. For an older patient that difference matters more than it does for anyone else.

The access change that makes this urgent

Federal law excludes weight-loss drugs from Medicare Part D. The GLP-1 Bridge sidesteps that as a demonstration programme, giving eligible enrollees certain GLP-1s at a flat $50 monthly copay from 1 July 2026 to 31 December 2027.

That is a large, sudden expansion of access in precisely the population where the monitoring questions are hardest. The programme covers the medication; it does not itself provide the baseline scan, the dietetic input or the resistance training the clinical literature keeps pointing to.

Six questions before starting after 65

1. What will you measure before I start, and what will you re-measure?
2. Given my other medications, what interactions matter?
3. How fast do you plan to escalate, and what would make you slow down?
4. What are the signs I am losing weight too quickly or eating too little?
5. What is the plan for maintaining muscle, and who helps me with it?
6. Who do I contact between appointments, and how fast do they respond?

A prescriber who answers all six is treating this as a clinical process. A platform that cannot answer them is selling a subscription.

What is not established

Long-term functional outcomes in older adults on these drugs are not well characterised. Fracture risk over years has not been established. And whether slower escalation preserves function — as opposed to composition — has not been tested in a trial.

We would rather name those gaps than imply a completeness the evidence has not reached.

The nutrition problem nobody has solved

The mechanism that produces weight loss also reduces the volume of food a person wants to eat. Advice to increase protein intake therefore competes directly with the drug's primary effect — which is a practical problem rather than a theoretical one, and it is sharper in an older patient whose baseline intake was already lower.

This is one of the clearer arguments for dietetic input alongside prescribing, and one of the clearer gaps in telehealth models that supply medication without it. It is worth asking any provider whether dietetic support is included in the price, billed separately, or simply absent. In our pricing dataset, laboratory work and dietetic input are the two fields we most often cannot establish from a published page.

Where a programme includes unlimited dietitian access as part of a membership, that membership is buying something real. Whether it is worth the fee is a judgement, but it is not the same as a membership that buys only platform access.

What routine monitoring would look like if it existed

Baseline and periodic body-composition assessment. A functional measure such as grip strength, which is cheap, quick and almost never done. Attention to intake adequacy rather than only to weight. Bone density measurement where risk factors are present. And escalation decisions informed by those measures rather than by a calendar.

Very little of this happens in practice, in any care model — not in telehealth, and frequently not in primary care either. Naming it is useful precisely because it lets you ask for it. A prescriber who takes the list seriously is running a clinical process; one who treats it as unusual has told you what monitoring you will receive.

The asymmetry worth noticing is that the drug is easy to obtain and the monitoring is not. That is the opposite of how a chronic-disease treatment should be delivered, and it is the structural criticism of this market that has the most force.