What a Waist Measurement Adds, and What Regulators Will Not Let It Carry

Answer first: it is a tape measure, and its job is to say something weight cannot

Body weight is a single number covering fat, muscle, bone, water, and whatever is currently in a digestive tract. A waist measurement is a second number that responds to where some of that sits rather than how much of it there is.

The FDA's draft guidance on weight-reduction drugs describes why clinicians reach for it: in clinical practice, waist circumference is sometimes used as an indirect measure of visceral fat, because it is easy and inexpensive to measure. Both of those adjectives are doing real work. A tape costs almost nothing and takes seconds, which is why it can be collected on thousands of people at every visit in a trial.

What follows is what the trials in this class actually recorded, and the specific limit the agency places on what that recording is allowed to prove.

The number is in the labels, in a table with a specific name

Both approved weight products report it. In each label's clinical studies section, the tables are titled changes in anthropometry and cardiometabolic parameters, and waist circumference is the first row.

In the two main tirzepatide weight trials, measured at week 72, baseline waist circumference ran between 113.2 and 116.0 centimeters across the arms. In the trial in adults without type 2 diabetes, the placebo group's waist fell 4.0 centimeters while the arms on the drug fell 14.0, 17.7, and 18.5 centimeters, giving differences from placebo of 10.1, 13.8, and 14.5 centimeters. In the trial in adults with type 2 diabetes, the placebo group fell 3.3 centimeters against 10.8 and 13.1 on the drug.

The semaglutide injection trials report the same row. In the three trials measured at week 68, baseline waist circumference sat between 111.8 and 115.5 centimeters. Placebo groups fell 4.1, 4.5, and 6.3 centimeters. The treated groups fell 13.5, 9.4, and 14.6 centimeters, for differences from placebo of 9.4, 4.9, and 8.3 centimeters.

These are group averages of a tape measurement, reported the same way as the weight figures beside them, and they move in the same direction.

The agency refuses to let it stand for visceral fat, and gives three reasons

Here is the sentence that changes how you should read those tables. Immediately after acknowledging that clinicians use waist circumference as an indirect measure of visceral fat, the guidance says that for regulatory purposes it is not considered a surrogate for visceral fat content or metabolic abnormalities.

It then gives three reasons, and each one is checkable against your own experience of being measured. The procedure is not standardized — there is no single agreed place on a torso to put the tape, so two clinics can produce different numbers from the same body. The measurement is affected by the extent of non-visceral fat, meaning the tape cannot distinguish the fat around the organs from the fat directly under the skin. And the accuracy is reduced in patients with an index above 35, which describes a large share of the people enrolled in these very trials.

The guidance does not throw the measurement out. It says change in waist circumference may be evaluated as a secondary endpoint, and that acceptability for labeling would depend on the quality of the data. That is precisely where it ended up: reported, in a table, as a secondary measure.

Most of these waist results were not error-controlled, and the labels say which

This is the technical detail that separates a headline result from a supporting one, and both labels print it in the footnotes.

The semaglutide label footnotes its anthropometry and cardiometabolic tables with a line stating that the parameters listed were not included in the prespecified hierarchical testing. A prespecified hierarchy is the mechanism that protects a trial against the false positives you get from testing many things at once. A result outside it is a real measurement that did not go through that protection.

The footnote carries named exceptions rather than being absolute, which is itself worth noticing. In one of the newer trial pairs, the comparison between placebo and the higher amount studied is inside the hierarchy for waist circumference. In one of those two trials it is inside for blood sugar as well.

The tirzepatide label makes the same distinction with letters. Its footnote key defines one marker as a p-value below 0.001 for superiority, controlled for type 1 error rate, and another simply as not controlled for type 1 error rate. In the trial in adults without type 2 diabetes, the waist comparison at the lowest amount studied carries the uncontrolled marker while the two higher amounts carry the controlled one.

So within a single row of a single table, some comparisons went through the trial's error protection and one did not. That is the level of detail at which these labels operate, and none of it survives into a marketing claim about inches lost.

What a waist number does and does not tell you

It tells you that a body changed shape, not only mass. In these trials the waist row and the weight row moved together, which is the least surprising possible result and still worth having, because weight alone cannot distinguish a body that got smaller around the middle from one that did not.

It does not tell you what tissue moved. The agency's own objection is that the tape cannot separate visceral fat from the fat under the skin, and a number that cannot separate them cannot be evidence about either one specifically.

It does not travel well between measurements. With no standardized procedure, a number from one clinic and a number from another are not necessarily comparable, and neither is comparable to one taken at home in a different place on the torso. Inside a trial this matters less, because the same protocol and often the same staff produce both readings for the same person. Outside one, it matters a great deal.

And it does not carry a threshold in this context. The trials report change in centimeters from a baseline, not a target figure. Whether any particular measurement means anything for any particular person is a clinical question, and no published table answers it.

Where you will meet this outside a label

Two places, mainly. Progress tracking, where a program suggests a tape measure alongside a scale — a reasonable suggestion, and one the trials effectively endorse by collecting the same measurement. And marketing, where a change in inches appears as a headline result.

For the second, the useful questions are the same as for any figure in this category. Who took the measurement, at what point on the body, at the start and end of what period. Compared against what — the trials report a placebo change of 3 to 6 centimeters in groups that were also on a diet and exercise program. A change with no comparator is missing most of its meaning. And is it a mean across a group or one person's result.

The larger point is one the guidance makes and marketing never does. A measurement can be genuinely useful, collected carefully in a well-run trial, printed in an approved label, and still be barred from carrying the specific claim it looks like it supports. The tape shows a body changing. What changed inside it is a question for an instrument built to answer it.