← Longevity & Supplement Guides
Waist size vs BMI: what a tape measure knows that an index doesn’t
The headlines were unusually kind to a piece of string this week. Your waist size may reveal health risks BMI misses. Abdominal fat predicts heart disease risk better than BMI. Somewhere in most homes there is a tape measure that has spent a decade under the takeaway menus, and it is suddenly the most respected instrument in the house.
Two studies published this summer are doing the work behind those headlines, and both are better than the coverage suggests. But the result everyone repeated — BMI lost — is not the interesting one. The interesting one is quieter, and it points the other way from almost everything wellness culture believes about measurement.
What the two studies actually did
The first, a research letter in JAMA Network Open from a Rutgers team led by Aayush Visaria, took 1,900 American adults aged 20 to 59 from the national NHANES survey and ran a bake-off. Four ways of identifying excess body fat went in: BMI on its own, waist circumference on its own, the two together, and the elaborate new framework proposed by a Lancet commission, which stacks BMI, waist, waist-to-height and waist-to-hip into one definition. The reference standard was DXA, the scan that actually sees fat.
BMI alone had a sensitivity around 51%. Read that slowly: of the adults who met a body-fat definition of obesity, the index most of the world uses found about half. For visceral fat specifically, BMI plus a tape measure scored an AUROC of 0.785 against the four-measurement framework’s 0.717.
The second study is larger and longer. Writing in the Journal of the American College of Cardiology, Zeina Dardari and the Cross-Cohort Collaboration pooled more than 260,000 adults followed for an average of twenty years, and asked what waist circumference and waist-to-hip ratio add on top of BMI across nine cardiovascular outcomes. Among people sitting comfortably in the normal-BMI range, those with a high waist or a high waist-to-hip ratio carried roughly 15% to 50% greater risk for most of the outcomes counted. The journal’s editor-in-chief, Harlan Krumholz, put it plainly: “It is time to abandon a sole focus on body mass index.”
The buried result is that adding instruments added nothing
Here is the part nobody put in a headline. The Lancet framework is the more sophisticated tool. It takes four measurements. It was built by a commission. And a single tape measure, used once, matched it — and beat it on visceral fat.
That is an unfashionable finding. The reflex of the modern wellness industry is that more measurement produces more truth: another wearable, another panel, another composite score with a proprietary weighting. The instinct is rarely tested, because testing it is boring and selling it is not. When someone does test it, the marginal instrument usually buys less than it costs.
It is worth sitting with that before buying anything. The measurement that improved on BMI was not expensive, not novel, and not proprietary. It was the one thing in the drawer.
BMI was never wrong. It was blind.
BMI deserves a fairer hearing than it gets. It is weight divided by height squared — a formula devised in the 1830s by the Belgian statistician Adolphe Quetelet to describe populations, and pressed into service for individuals a century and a half later. As a population instrument it works fine. It costs nothing, it needs no skill, and across millions of people it tracks what it is supposed to track.
Its failure is not inaccuracy. It is blindness to shape. Weight and height cannot tell you where tissue sits, and where it sits is the part that seems to matter — fat packed deep in the abdomen around the organs behaves differently from fat under the skin of the hips. Two people with identical BMIs can be carrying entirely different bodies. BMI has no vocabulary for the difference. A tape measure has one crude word for it, and one crude word turns out to be a lot more than none.
Which is also the honest limit of the good news: a waist measurement does not see visceral fat either. It correlates with it better than a ratio of two numbers that ignore geometry. That is the whole claim, and it is enough.
The number worth remembering is half your height
A raw waist threshold in centimetres has an obvious problem — it lumps a 155 cm adult and a 190 cm adult into the same case. Dividing by height fixes most of that, which is why the UK’s National Institute for Health and Care Excellence now tells adults to keep their waist to less than half their height: a waist-to-height ratio under 0.5.
NICE splits the scale into three bands — 0.4 to 0.49 as healthy central adiposity, 0.5 to 0.59 as increased, 0.6 and above as high — and, unusually for a body-composition rule, says it can be used for both sexes and all ethnicities, including adults with a lot of muscle, provided BMI is under 35. It is also the rare official recommendation designed for people to carry out on themselves, in a bedroom, with no appointment.
The half-height rule
Pick your height. That is the waist number the guidance points at.
How to measure a waist so the number means something
Most people measure the wrong place. The waist that matters is not where trousers sit. The standard used by the WHO, NICE and the International Diabetes Federation is the midpoint between the bottom of your lowest rib and the top of your hip bone — feel for both on your side, and take the halfway point, which usually lands an inch or so above the navel.
The rest is unglamorous and decisive. Bare skin or one thin layer, never over a jumper. Stand relaxed, feet together, arms down. Breathe out gently and read the tape at the end of that breath, without pulling it tight enough to dent anything. Same time of day each time, ideally morning, before eating. Take it three times and use the middle value.
This sounds like fussiness. It is the opposite. A couple of centimetres of sloppiness is larger than a year of real change, which means bad technique doesn’t just add noise — it manufactures results in both directions. The tape measure is only cheap if you use it properly; used carelessly it is an expensive way to be misled for free.
What a waist still cannot tell you
Both studies are observational, and neither shows that moving the number moves the outcome. The JACC work reclassifies risk; it does not demonstrate that a smaller waist causes a different future. Both readings fit the data, and the honest position is that a waist measurement is a signal about the body you currently have, not a lever you have proven you can pull.
The thresholds are population averages, too. Waist cut-offs in centimetres already differ by ethnicity in most guidelines; dividing by height smooths that considerably but does not abolish it. And the number is slow — it moves over months, not mornings, so it is useless as a daily metric and will only frustrate anyone who handles it like one.
Most importantly, a waist is a single axis. It says nothing about how much air you can move, how hard you can grip, or how fast you can walk — three other numbers that predict a great deal on their own. A person can pass this test and fail all three. Read it as one line in a paragraph, not the paragraph.
Where a slow number fits in a system you actually keep
We tend to split measurements into fast and slow. Fast numbers — HRV, resting heart rate, last night’s sleep — describe the last day or two, and they are what your wearable is actually good at. Slow numbers describe the last year: grip strength, walking speed, VO₂max, and now, on this evidence, a waist.
Slow numbers need a calendar, not a dashboard. Once a quarter, same morning, same technique, written down somewhere that survives a phone upgrade. That is genuinely the whole protocol, and it is how we think about the biomarkers worth tracking inside a longevity protocol — alongside a healthy scepticism about the composite scores that promise to do the thinking for you.
The bottom line
BMI did not die this month; it was demoted, correctly, from verdict to input. The replacement is not a better algorithm or a more expensive scan. It is a tape measure, a wall, and the discipline to use both the same way twice a year.
Measure the midpoint between your lowest rib and your hip bone, after a gentle breath out. Divide by your height. If the answer starts with a 0.4, that is the band the guidance points at. If it doesn’t, that is information, not a sentence — and it belongs in a conversation with your doctor rather than a resolution made at midnight. Use numbers to correct fantasy, not to replace experience.
