A patient in their seventies will have their blood pressure taken at nearly every appointment and their A1c checked once or twice a year. Both are measured routinely because both predict how that patient will fare. Muscle mass predicts the same things, arguably more directly, falls, hospitalizations, loss of independence, yet almost nobody measures it.
Sarcopenia, the progressive loss of muscle mass and strength that comes with age, has been a formally recognized disease with its own diagnosis code since 2016. Expert groups in Europe, Asia and the US have each published criteria for identifying it, and it predicts falls, fractures, loss of independence and mortality. Yet if you ask a primary care team how they screen for it, the honest answer is usually that they do not. Grip strength is in the guidelines, though it rarely makes it into the visit. The reference standard is a DEXA scan, which means a referral, a facility and a cost. In practice that gives a patient a single reading every few years, usually because somebody was asking about bone density rather than muscle. The field understands that muscle matters. What it lacks is a practical way to watch it.
What GLP-1s did to the argument
For most of the last decade this was a geriatrics conversation. That has changed.
Body composition data from the GLP-1 trials shows that somewhere between a quarter and two fifths of the weight lost is muscle rather than fat in most analyses, higher in some individual trials and considerably lower in others. The range is wide, and anyone quoting a single figure from it is overselling the evidence. The width of the range is itself the argument. When the same drug class can take a quarter or nearly half of a patient's weight loss out of muscle depending on the patient and the dose, the only way to know which patient you have in front of you is to measure.
Withings Health Solutions supports healthcare professionals who manage more than 500,000 patients in obesity and weight management programs. A growing share of the people starting these medications are in their sixties and seventies.A 68-year-old losing 18% of body weight is on paper a program success. If much of that came out of muscle, the program has traded one risk for another, and neither weight nor BMI can tell you whether it did. That is the same reasoning behind the Lancet Commission's 2025 definition of clinical obesity, which moved the diagnosis toward body composition rather than weight alone. The clinical thinking has moved ahead of the measurement infrastructure.
One measurement cannot show a direction
Muscle loss accumulates gradually, at one to two percent a year, invisibly, until it surfaces as a fall or a hospitalization. No individual reading, however precise, can tell you which way a patient is heading. This is the same point we have made in heart failure, where daily weight is the oldest signal there is and works precisely because it is daily. Any single reading is noise. Three months of them together give a care team a trend worth acting on.
So the useful question is what a patient's muscle mass has been doing for the last six months, and whether that direction is acceptable given what we are treating them for. Answering it calls for something a patient will actually stand on, in their own bathroom, most mornings, for years.
Which requires being honest about what a connected scale can do. These devices read body composition by sending a small electrical signal through the body, and while they line up closely with DEXA on average, that average conceals real variation for any given person on any given day. They should not be used to put a patient above or below a diagnostic threshold. Their strength is detecting change in the same patient on the same device over time, which is the measurement DEXA is worst at providing. The gold standard stays where it is, and a connected scale covers the long stretch between scans.
Nobody is paid to notice
This isn't a knowledge gap. It's an incentive gap. Blood pressure and A1c are measured routinely, partly because they matter clinically, and partly because they are built into quality measures and reimbursement codes, the systems that determine where a care team's attention goes. Muscle loss has a diagnosis code and almost nothing built around it, so a team that screens for it is doing so without any structural support or incentive to.
The treatment side has never been the obstacle. Strength training and adequate protein intake are well established; what has been missing is a reliable way to identify which patient needs them, and when. That data can now come from a connected scale used at home. The capability to monitor muscle loss cheaply and continuously, in a patient's own home, exists today. It is largely going unused.
Go back to that patient in their seventies. Their blood pressure gets taken at every visit because the system is built to notice it. Their muscle mass could be tracked with the same regularity, on a device they already own, at a fraction of the cost of a DEXA referral. The only thing standing between where care is today and where it could be is the decision to start measuring.