The Waist the Scale Misses in Young People
A deep dive into one paper: what was known before, what the researchers did, what the numbers say and what they don't.
Chapters
Transcript
The script the two AI voices read from. The times are worked out from the script, not heard, so a line can start a few seconds early or late.
Read the whole episode
One in six, missed by the scale
Host 1If a young person's body mass index says they do not have obesity, can they still be carrying the kind of fat that matters for their heart?
Host 2Yes, and in this study it was common: among about sixty thousand people aged fifteen to twenty-nine in seventy countries, roughly one in six had a waist that crossed two separate warning lines while their body mass index stayed below the obesity mark. That group already had almost twice the odds of at least one early warning sign, such as high blood pressure, high blood sugar or high cholesterol, compared with young people with no sign of excess fat.
Host 1Welcome to the Hastrup Review of Science, where artificial intelligence reads real science, so you don't have to pretend you understood the paper.
Host 2The paper is in the journal E Clinical Medicine, by Setognal Aychiluhm and colleagues at Charles Sturt University in Australia and partner institutions. It asks how many young people around the world carry excess body fat once you look at the waist as well as the scale, and whether that fat already shows up in their blood pressure, blood sugar and cholesterol.
Why we measure bodies with one number
Host 1Before we get into it, where does the body mass index actually come from, and why did it become the standard way to judge weight?
Host 2The wider history is that it goes back to the Belgian statistician Adolphe Quetelet in the eighteen thirties, who noticed that weight in adults scales roughly with height squared. In the nineteen seventies the American physiologist Ancel Keys gave it the name body mass index and argued it was the best simple measure for population studies.
Host 1So it was designed for counting populations, not for telling an individual person whether their body fat is a problem?
Host 2That is a fair way to put it. It is your weight in kilograms divided by your height in metres squared, and it is cheap, quick and needs nothing but a scale and a ruler, which is why it spread into clinics and insurance tables as well.
Host 1What is the weakness people usually point to, if it is so convenient?
Host 2It cannot tell fat from muscle, and it says nothing about where the fat sits. The paper's introduction puts it the same way: someone with a normal body mass index can carry a lot of fat around the abdomen, and someone with a high one can have a fairly harmless distribution.
Host 1Why would it matter where the fat sits, if fat is fat?
Host 2Across the field, fat packed around the organs inside the abdomen, called visceral fat, is the kind most tied to insulin resistance, high blood pressure and unhealthy blood fats. Fat under the skin of the hips and thighs is, broadly speaking, less troublesome, and the waist is the cheapest available window onto the abdominal kind.
What was known before
Host 1What did researchers already know about waists and risk before this study came along?
Host 2A good deal in adults. The authors cite a United States study of more than three hundred thousand adults, where the share counted as having obesity rose from about forty-three percent to about sixty-nine percent once waist-based measures were added to body mass index.
Host 1That is a jump of more than a quarter of all adults, just by adding a tape measure. Did those extra people actually turn out to be at higher risk, or were they just relabelled?
Host 2In that same research program, the people with obesity only by waist measures had higher risks of diabetes, heart disease and death than people without obesity, so the label tracked something real. The same authors had also found, in an earlier multi-country study, that more than one in five adults with a normal body mass index had abdominal obesity.
Host 1You mentioned a Lancet Commission at the start. What is that, and what did it change?
Host 2It is an expert commission published in The Lancet Diabetes and Endocrinology in twenty twenty-five, which proposed that body mass index be used mainly for screening and for population statistics. Excess fat should then be confirmed with a waist-based measure, such as waist circumference, waist-to-height ratio or waist-to-hip ratio, or with a direct body fat scan where one exists.
Host 1And I think I have heard it split obesity into two kinds. What were they?
Host 2Clinical obesity, where the excess fat already comes with organ problems or limits on daily function, and preclinical obesity, where the fat is there and raises future risk but nothing has gone wrong yet. That second category matters a lot for young people, because most of them are not ill yet.
The open question
Host 1So what was the gap this paper set out to fill?
Host 2Almost all the evidence on confirmed excess fat came from adults or from single countries. Nobody had a picture across many countries of how common it is between fifteen and twenty-nine, how it splits between body mass index and waist, and whether it already lines up with measurable risk at that age.
Host 1Why focus on that age band in particular, rather than just everyone?
Host 2The authors argue that risk factors that appear in adolescence tend to persist into adulthood, and that fat gained around the waist in the late teens and twenties predicts later high blood pressure and diabetes even at a normal body mass index. If you want to catch things early, this is the window.
How the study worked
Host 1Where do you even get comparable body measurements from seventy countries?
Host 2From a World Health Organization program called STEPS, short for the stepwise approach to surveillance, which runs national household surveys on risk factors for chronic disease. Each survey has three steps: a questionnaire on things like smoking, diet and exercise; physical measurements of height, weight, waist and blood pressure; and a blood test for sugar and cholesterol.
Host 1Who gets picked in those surveys, is it whoever answers the door?
Host 2No, households are sampled in stages across the country, and inside each household a tablet application lists everyone eligible and picks one person at random. Weights are then applied so the results stand for the country's population, and the authors dropped any survey that covered only one province or city.
Host 1How many people and which years ended up in the analysis?
Host 2Sixty thousand six hundred and thirty-four people aged fifteen to twenty-nine, from seventy national surveys run between twenty ten and twenty twenty-three, with pregnant women excluded. About half came from Africa, and only about four and a half percent were aged fifteen to seventeen.
Host 1Now the key part: how exactly did they decide someone had excess fat, if not by the scale alone?
Host 2They used three routes, following the Commission's idea. You counted if your body mass index was in the obesity range and at least one waist measure was high, or if both waist measures were high regardless of your body mass index, or if your body mass index alone was very high, around forty or more.
Host 1What are the two waist measures, in plain terms?
Host 2The first is waist circumference, measured with a tape: high meant ninety-four centimetres or more for non-Asian men, ninety for Asian men, and eighty centimetres for women. The second is waist-to-height ratio, your waist divided by your height, and the line was half: a waist more than half your height counted as high.
Host 1So if I am one metre eighty tall, my waist would need to stay under ninety centimetres to be under that line?
Host 2Exactly right. The appeal of that ratio, as many researchers in the field see it, is that the same simple rule of thumb, keep your waist under half your height, works across heights without a table.
Host 1Did they treat a fifteen-year-old the same as a twenty-five-year-old?
Host 2Not for body mass index. For fifteen to seventeen they used the World Health Organization's growth reference, which compares a teenager with others of the same age and sex, and for adults they used a lower obesity line of twenty-seven and a half for Asian participants and thirty for others.
Host 1And the heart and blood sugar side, how did they decide someone had a warning sign?
Host 2Three markers. High blood pressure meant a reading of one hundred and forty over ninety or more, averaged from the second and third of three readings, or taking blood pressure medicine; diabetes meant fasting blood sugar of seven millimoles per litre or more, or taking glucose-lowering medicine; and high cholesterol meant total cholesterol of five millimoles per litre or more, or taking cholesterol medicine.
Host 1How did they combine those three into a single picture of risk?
Host 2They counted them for each person, from zero to three, and then looked at two outcomes: having at least one marker, and having two or more. Anyone missing any of the three was left out of those particular analyses.
Host 1Earlier you described this as a kind of meta-analysis. If they had all the individual people, why pool countries like separate studies?
Host 2For the headline prevalence, they estimated a figure for each country and then combined them with a random-effects model, which assumes countries genuinely differ rather than all sharing one true value. That stops the big African surveys from swamping the result, and it gives wider, more honest uncertainty when countries disagree.
Host 1And for the links with risk, what kind of model did they use?
Host 2Logistic regression, which estimates odds ratios while adjusting for other factors: age, sex, education, work, country income, physical activity, sitting time and smoking. For the question of who has excess fat, they also let each country have its own baseline level.
Host 1Before the results, can you remind me what an odds ratio of two actually means?
Host 2It means the odds of having the outcome are twice as high in one group as in the comparison group. Odds are not quite the same as risk, and when an outcome is fairly common an odds ratio makes the gap look somewhat bigger than the plain difference in percentages would.
A word from our sponsor
Host 2This episode is brought to you by Lego bricks. By stepping on a Lego brick barefoot in the dark, you agree to the following terms.
Host 1The brick will list every eligible member of your household and select one of you at random for the next step.
Host 2Each step is recorded as the mean of the second and third steps, with three minutes of rest provided in between.
Host 1The Lego Group has not paid for this, has not been asked, and is presumably busy building something else.
Host 2Readings from your hallway are pooled with those of sixty-nine other hallways using a random-effects model.
Host 1After enough steps, the brick becomes nationally representative and is recognised as the seventh region of the World Health Organization.
What they found
Host 1So, the headline: how many of these young people had excess fat by this definition?
Host 2The pooled estimate was thirty-three point seven percent, with a ninety-five percent confidence interval from about twenty-nine to thirty-eight percent. So roughly one in three young people aged fifteen to twenty-nine.
Host 1Is one in three consistent everywhere, or does that average hide big differences between places?
Host 2It hides enormous differences. Country figures ran from six point six percent in Ethiopia to about seventy-seven percent in Tonga and the Cook Islands, and by region from about nineteen percent in South-East Asia and twenty-one percent in Africa to about forty-eight percent in the Eastern Mediterranean and the Western Pacific.
Host 1Does it follow how rich a country is?
Host 2Broadly yes: about twenty-two percent in low-income countries rising to about fifty-two percent in high-income ones. After adjustment, young people in high-income countries had nearly six times the odds of excess fat compared with those in low-income countries, though the interval on that is wide, from about three to eleven.
Host 1Now the part from the opening. How did the waist and the scale actually split?
Host 2About thirteen percent had obesity by body mass index, but thirty-eight percent crossed at least one waist line, and twenty-seven percent crossed both. Sorted into groups that do not overlap, fifteen point seven percent had both waist measures high without an obesity-range body mass index, twelve percent had a high body mass index plus a high waist or a very high body mass index, and only zero point eight percent had a high body mass index with no high waist.
Host 1So if I have this right, almost everyone the scale flags is also flagged by the tape, but the tape flags a whole extra group the scale misses?
Host 2That is right, and it is the core finding. The scale on its own added almost nobody the waist did not already catch, while the waist added a group larger than the scale's whole obesity group.
Host 1And that extra waist-only group, were they actually less healthy, or just bigger around the middle?
Host 2Compared with people with no sign of excess fat, they had almost twice the odds of at least one risk marker, an odds ratio of one point nine two, and almost four times the odds of two or more, three point nine three. In plain numbers, about four and a half percent of them had two or more markers, against about one and a half percent of the no-fat group.
Host 1How does that compare with the people whose body mass index was high as well?
Host 2That group was worse again: about three times the odds of any marker and nearly seven times the odds of two or more, with almost ten percent carrying two or more. Even people who crossed just one waist line had modestly raised odds, so the picture is a gradient rather than a cliff.
Host 1Who was most likely to have excess fat, apart from living in a richer country?
Host 2Older participants, women and the physically inactive. Twenty-five to twenty-nine year-olds had almost four times the odds of fifteen to seventeen year-olds, women had about four times the odds of men overall and about nine times in Africa, and inactive people had about twenty percent higher odds.
Host 1Women four times as likely seems huge. Could part of that come from the waist line for women being set at eighty centimetres, compared with ninety-four for most men?
Host 2The paper does not test that, so it is a reasonable question rather than an answer. The cut-offs are sex-specific by design, meant to reflect where risk starts for each sex, but how much of the gap is biology, behaviour or the choice of lines is something this study cannot separate.
Host 1You mentioned education as one of the adjusted factors. Did it point the same way everywhere?
Host 2No, and that was one of the more interesting details. In Africa, South-East Asia and the Western Pacific, having no formal schooling went with lower odds of excess fat, while in the Eastern Mediterranean and the Americas it went with higher odds.
Host 1Why would being less educated protect you in one place and harm you in another?
Host 2The authors point to what is called the reversal hypothesis from the wider obesity literature: in poorer settings, less schooling often means manual work and less access to energy-dense food, while in richer settings the cheap, energy-dense food reaches the less educated first. It is an interpretation, not something the data here prove.
Why it matters
Host 1Stepping back, what is the principle here, beyond the numbers?
Host 2That a single number built from weight and height misses a large share of young people whose fat sits where it seems to matter most, and that this group already shows early signs of risk. A tape measure costs almost nothing, so the practical point is that waist measures belong next to body mass index in surveys and checkups.
Host 1Who would actually change what they do because of this?
Host 2Mainly the people who run national health surveys and set screening rules, who could report waist-based figures alongside body mass index for young people. The authors also suggest the STEPS surveys add short questions on symptoms and daily function, so that clinical and preclinical obesity could one day be told apart at the population level.
Host 1And for an ordinary listener in their twenties, is there anything to take from it?
Host 2Within the limits of this study, it suggests that a normal body mass index is not the whole story, and that the half-your-height rule for the waist is a reasonable check to know about. It does not show that shrinking the waist lowers these risk markers, and the authors say plainly they cannot tell how well prevention would work.
The honest caveats
Host 1What can we not conclude from this, given it is a snapshot?
Host 2Cause and effect, first of all. Each person was measured once, so the study shows that waist-based fat and risk markers go together, not that one causes the other, and the authors note that unmeasured factors such as genetics and early life could play a part.
Host 1Is a tape measure around the waist actually a good measure of the fat inside the abdomen?
Host 2It is an indirect one. The surveys had no body scans, so nobody here measured visceral fat directly; waist and waist-to-height ratio are stand-ins that the wider field considers practical but imperfect.
Host 1You said the blood tests came from many different countries. Were they all done the same way?
Host 2No, and that is a real weakness. Some countries used finger-prick devices in the field and others sent venous blood to a central laboratory, and the authors applied no correction between them, while glucose and cholesterol were also not available everywhere, which shrank some analyses.
Host 1Is there anything about the statistics that might make the results look more certain than they are?
Host 2One technical point: the information on which neighbourhood clusters people were sampled from was not consistently available, so each person was treated as their own cluster. Over the field as a whole, that kind of shortcut tends to make confidence intervals narrower than they should be, though the authors do not estimate by how much.
Host 1And how well does this represent the whole world's young people?
Host 2Not perfectly. Half the sample is from Africa, several high-income countries are not in STEPS at all, and fifteen to seventeen year-olds were only four and a half percent of the sample, with none from the European or American surveys.
Host 1One thing that struck me: you said the warning-sign definitions were the adult ones, such as one hundred and forty over ninety. Are those the right lines for a fifteen-year-old?
Host 2For teenagers, clinical practice usually uses lower, age-based blood pressure thresholds, so using adult lines probably undercounts risk in the youngest group rather than inflating it. Since that group is small here, it mostly affects how far the findings can be stretched down to younger adolescents.
Host 1And the exercise and diet answers, those were just what people said about themselves?
Host 2Yes, self-reported, so they can be off through poor memory or a wish to give the right answer. That makes the modest link with inactivity, an odds ratio of about one point two, the softest of the main results.
What to watch next
Host 1So if I put it together: a large, broad snapshot showing the waist catches many young people the scale misses, with early risk attached, but no proof of cause and some rough edges in the data?
Host 2That is a fair summary. The prevalence and the classification results are the sturdiest part; the size of the risk links is less certain because of the mixed blood-testing methods and the single time point.
Host 1What would settle the bigger question of whether this waist-only group goes on to get sick?
Host 2Following young people over years, measuring waist and body mass index at the start and seeing who develops diabetes or heart disease. Studies with body scans would also show how well the tape tracks the fat inside the abdomen at these ages.
Host 1Is anything likely to change in how surveys or clinics measure young people?
Host 2Worth watching is whether the World Health Organization and national programs start reporting waist-based obesity figures for young people alongside body mass index, as the Lancet Commission urged. If future STEPS rounds add questions on symptoms and function, the clinical versus preclinical split could finally be measured across countries.
Sign-off
Host 1Is there a single sentence you would leave someone with?
Host 2In seventy countries, one in six young people had a waist the tape measure flagged and the scale did not, and they already carried more early risk than those with no sign of excess fat. It is a snapshot, not a verdict, but it is a strong argument for measuring the waist.
Host 1Every weekend this same feed carries the roundup, the week's new papers in one short episode. Thanks for listening to the Hastrup Review of Science.
More from this week
- Sprint Caffeine, Lost Lean Mass and a Too-Eager Atlantic CollapseEpisode 14 · 10 min
- Caffeine on the Blocks, How Much Makes Sprinters FasterDeep dive · 18 min
- Weight-Loss Drugs and the Lean Mass That Goes With the FatDeep dive · 21 min
- Which Cholesterol Drugs Prevent a First Heart AttackDeep dive · 16 min
- Do Climate Models Give Up on the Atlantic Too Easily?Deep dive · 21 min
- Four Layers of Nickel, Superconducting Without the SqueezeDeep dive · 20 min