BMI and health: it’s complicated

Forget the scales: why waist size may matter more than BMI as we age

For decades, body mass index (BMI) has been the standard shorthand for a healthy weight. But a new study of nearly 7,000 older adults suggests that, once you’re past 65, it may be telling only part of the story, and in some cases, the wrong story.

The study results seem, at first glance, to contradict each other. Men classed as “overweight” by BMI had a lower risk of death than men with a BMI in the “normal” range. Women who were overweight had a 27% lower risk. Looking at waist circumference, however, showed a different picture.
 

A confusing set of results

The study, published on 19 August in the Journal of the American Geriatrics Society, followed adults aged 65 and over for up to 14 years, tracking how their BMI and waist circumference changed over time and how those changes related to their risk of dying during the study period. The data came from a large, nationally representative US study funded by the National Institutes of Health.

The results seem, at first glance, to contradict each other. Men classed as “overweight” by BMI (25–29.9) had a 46% lower risk of death than men with a BMI in the “normal” range. Women who were overweight had a 27% lower risk. For men and women with class I or II obesity (BMI 30 to just under 40), the picture was mixed – lower risk for men, no significant difference for women – and the highest BMI category, class III obesity (40+), showed no significant difference in either sex.

Waist circumference told a different story entirely. Men with a waist over 40 inches (101.6cm) and women with a waist over 35 inches (89cm) had a 24% higher risk of early death than those below those thresholds, and that held true even after accounting for BMI.

Put simply: by BMI alone, carrying some extra weight looked protective. By waist size, it looked risky. Both can’t be the full picture.
 

Not a green light to gain weight

It’s worth being clear about what this study does not show. It’s an association, not proof that gaining body fat helps people live longer, and the researchers were explicit that their findings shouldn’t be read as evidence that excess fat is protective.

Part of what is probably going on is reverse causation: serious illnesses such as cancer and heart disease often cause weight loss before they’re diagnosed, which can make lower weight appear to be linked to earlier death when the illness is really driving both. There’s also survivorship bias: someone who reaches their late 70s or 80s despite living with obesity for years may simply be a different, hardier case than someone who developed obesity-related disease and died younger. And then there’s BMI itself, which may just not be measuring what we think it’s measuring.
 

Why BMI becomes less reliable with age

BMI is calculated from height and weight alone, so it can’t distinguish between muscle and fat. Two people of identical height and weight could have very different bodies – one carrying more muscle, the other more fat – and BMI would score them exactly the same.

That blind spot matters more with age. Muscle mass naturally declines as we get older, a process called sarcopenia, and older adults can also lose height and shift where they store fat. So a higher BMI in later life may sometimes reflect better-preserved muscle rather than excess fat, while someone with a “normal” BMI could have lost significant muscle while still carrying more fat than is healthy. Neither would show up correctly on the BMI scale alone.
 

Why the waist matters

Fat stored around the abdomen – visceral fat – sits deep around the internal organs, and it behaves differently to fat stored elsewhere on the body. It’s linked to inflammation, insulin resistance and a higher risk of cardiovascular disease and diabetes. Waist circumference is a simple, low-tech way of estimating how much of it someone is carrying, which is why a growing number of health experts now recommend using it alongside BMI, waist-to-hip ratio, or other measures of body composition, rather than relying on BMI in isolation.
 

 

The other end of the scale: being underweight

The study also flagged a particularly striking finding: older adults classed as underweight had close to double the risk of early death of those with a “normal” BMI. That fits with what’s already known about frailty: older adults rely on muscle strength for everyday tasks, mobility and reducing fall risk, so unintentional weight loss driven by muscle loss can leave someone more vulnerable even as their BMI moves into a “healthier”-looking range.

Here too, cause and effect cut both ways. Being underweight is often a consequence of poor health rather than a cause of it. Chronic illness can drive weight and muscle loss, and unexplained weight loss in an older person is usually worth flagging to a doctor rather than treating as good news.
 

The takeaway

None of this means BMI is useless, or that carrying more weight is automatically fine in later life. What it does suggest is that BMI alone is too blunt a tool to tell the whole story for older adults, and that a simple waist measurement, taken alongside it, may give a more useful picture of risk.

For anyone caring for an older relative, or supporting an older patient, the message is the same: a “normal” BMI shouldn’t be read as reassurance on its own, a “high” one shouldn’t be read as alarm on its own, and unexplained weight loss deserves attention whatever the numbers say.

Sourced from a study in the Journal of the American Geriatrics Society (19 August 2026) and commentary from Dr Leana Wen, emergency physician and clinical associate professor at George Washington University, speaking to CNN (external link will open in a new browser tab or window).
 

Article ends – click to return to top or check out related resources below

Further reading and resources

A Word (Or More) About BMI

Obesity and weight loss eligibility

Make sure you don’t miss out on news from Independent Living

return to top