BMI is not the final word on health risk
A large Danish cohort study has quietly dismantled one of medicine's most relied-upon shortcuts: the assumption that the Body Mass Index tells us, with any real precision, who is at risk of dying early. Tracking more than 85,000 adults over five years, researchers found that the gravest danger lay not in excess weight but in its absence — with underweight individuals facing nearly triple the mortality of their heavier peers — while those in the overweight and mildly obese ranges showed no meaningful increase in death risk at all. The finding invites a deeper reckoning with how we measure human health, and whether a single ratio of height to weight can ever bear the weight we have placed upon it.
- A five-year Danish study of over 85,000 adults recorded 7,555 deaths and found that being underweight nearly tripled the risk of premature death — a figure that cuts against decades of clinical messaging.
- The data exposed a troubling blind spot: people classified as overweight or mildly obese showed no elevated mortality, while those at the low end of the 'healthy' BMI range died at higher rates than those at the high end of it.
- Severe obesity — BMI 40 and above — did double mortality risk, but the broad middle ground of excess weight proved far less lethal than standard guidelines have long implied.
- Researchers point to two fractures in the BMI framework: reverse causality, where illness causes weight loss before death, and the metric's fundamental inability to distinguish dangerous visceral fat from metabolically inert fat elsewhere in the body.
- The emerging consensus among scientists is that cardiorespiratory fitness and body composition — what the body is made of and what it can do — are far stronger predictors of longevity than any number a scale can produce.
A Danish research team led by Sigrid Bjerge Gribsholt at the Steno Diabetes Center Aarhus has spent five years following more than 85,000 adults, and what they found challenges the foundational logic of how medicine thinks about body weight and death. Among the 7,555 people who died during the study period, the starkest risk did not belong to the heaviest participants — it belonged to the thinnest. Those with a BMI below 18.5 faced nearly three times the mortality of people at the upper end of the normal weight range, the group the researchers used as their reference point.
Presented at the European Association for the Study of Diabetes annual meeting, the findings revealed a pattern that defies the intuitive assumption that health risk climbs steadily with weight. People classified as overweight — BMI between 25 and 30 — and even those with mild obesity showed no increase in five-year mortality compared to the reference group. The picture did darken again at the extremes: severe obesity, defined as BMI 40 or above, more than doubled death risk, and the 35–40 range carried a modest 23 percent increase. But the broad middle ground of excess weight appeared, by raw mortality measures, surprisingly safe.
Even within the normal BMI range itself, a gradient emerged. Those at the lower end of normal died at higher rates than those at the upper end — a detail that further erodes confidence in the category's meaning. The researchers offer two explanations: that reverse causality distorts the data, since people already ill tend to lose weight before dying, and that BMI is simply too crude a tool, measuring only height and weight while ignoring where fat is stored and what the body's metabolic state actually is.
The distinction is not trivial. Visceral fat packed around internal organs behaves very differently from subcutaneous fat at the hips and thighs. Two people with identical BMI scores can carry radically different health risks depending on their muscle mass, fat distribution, and cardiovascular fitness. What the study ultimately argues — and what a growing body of research supports — is that cardiorespiratory fitness and body composition are far better guides to how long someone will live than any single number a scale and a formula can produce.
A Danish research team following more than 85,000 adults over five years has arrived at a finding that upends conventional health wisdom: being significantly underweight carries nearly three times the risk of premature death compared to people in the upper ranges of what doctors call a normal weight. The study, led by Sigrid Bjerge Gribsholt at the Steno Diabetes Center Aarhus, tracked 7,555 deaths during the observation period and suggests that the relationship between body weight and longevity is far messier than the standard Body Mass Index categories suggest.
The research, presented at the European Association for the Study of Diabetes annual meeting, challenges the assumption that health risk increases steadily as weight goes up. When researchers used people at the upper end of the normal BMI range as their reference point, they found something unexpected: individuals classified as overweight (BMI 25-30) and even some with mild obesity (BMI 30-35) showed no increase in mortality during the five-year window. The danger, it turns out, lies at the other end of the scale. Those with a BMI below 18.5—the threshold for underweight—faced dramatically elevated death rates.
The picture does darken again at the extremes of obesity. People with severe obesity, defined as a BMI of 40 or higher, had more than double the mortality risk of the reference group. Those in the BMI range of 35 to 40 showed a more modest increase of about 23 percent. But the middle ground—from overweight through mild obesity—appeared surprisingly safe by comparison, at least in terms of raw mortality numbers over this timeframe.
Even within the supposedly healthy normal range, the data revealed a gradient. People at the lower end of normal weight (BMI 18.5-25) had higher mortality than those at the upper end of that same category. The researchers propose two explanations for this pattern. One is reverse causality: people who lose weight due to underlying illness appear in the data as underweight but are actually sick, skewing the numbers. The other is that BMI itself is too blunt an instrument. It measures only height and weight, ignoring where fat is stored and what the body's metabolic state actually is.
This distinction matters enormously. Visceral fat—the kind that accumulates around organs deep in the abdomen—wreaks metabolic havoc in ways that subcutaneous fat stored in the hips and thighs does not. Two people with identical BMI numbers can have radically different health profiles depending on their muscle mass, their fat distribution, and their cardiovascular fitness. The researchers emphasize this point: BMI is not, as some have treated it, the final word on health risk.
Other recent research cited by experts outside this study suggests that cardiorespiratory fitness and body composition—the actual ratio of muscle to fat, and where that fat lives—may be far better predictors of how long someone will live than BMI alone. The implication is clear: a single number on a scale, plugged into a formula, cannot capture the complexity of human health. What matters is not just how much you weigh, but what you're made of and what your body can do.
Citas Notables
The relationship between weight and health is more complex than often assumed, and the lowest mortality risk may not fall clearly within the traditional healthy BMI range— Study findings presented at European Association for the Study of Diabetes
Visceral fat accumulated around organs has far more damaging metabolic effects than subcutaneous fat in hips or thighs, meaning two people with the same BMI can have very different risk profiles— Study researchers
La Conversación del Hearth Otra perspectiva de la historia
Why does being underweight carry such a high mortality risk? That seems counterintuitive.
The study doesn't fully explain the mechanism, but the researchers point to reverse causality—people often lose weight because they're already sick. So the underweight category captures people with undiagnosed illness, not necessarily people who are healthy at a low weight.
So the BMI categories themselves might be misleading?
Exactly. BMI is just height divided into weight. It tells you nothing about muscle versus fat, or where the fat is stored. Someone muscular might have a high BMI but be very healthy. Someone with the same BMI could be mostly fat around their organs, which is metabolically toxic.
If BMI is so flawed, why do doctors still use it?
It's simple, cheap, and quick. You don't need expensive imaging or fitness tests. But this study suggests that simplicity comes at a cost—it misses the actual picture of health.
What should people actually pay attention to instead?
The research points toward fitness and body composition. Can you climb stairs without getting winded? Do you have muscle mass? Where is your fat stored? Those things predict longevity better than a BMI number.
Does this mean mild obesity is actually safe?
Not safe—the study shows severe obesity still carries real risk. But mild overweight didn't increase mortality in this five-year window. That doesn't mean it's harmless long-term; it just means the picture is more nuanced than BMI suggests.