The number on her chart said “overweight.” The patient sitting across from me was a former Division I athlete. Resting heart rate in the low 50s, glucose in the optimal range, body fat percentage in the lean category. By BMI, she had been counseled for years to lose weight. By every other marker that mattered, she was metabolically thriving.

That contrast is one of the most common reasons people leave a clinic visit confused, frustrated, or quietly ashamed about a number that was never designed to describe them as individuals.

Body Mass Index has earned a strange dual life in modern medicine. It is everywhere, and on its own, it is a poor measure of any one person’s health. Both can be true. The work of being an informed patient (or an informed clinician) is knowing which job BMI is actually good at, and which jobs it keeps getting handed that it cannot do well.

What BMI is, and what it isn’t

BMI is a ratio: weight in kilograms divided by height in meters squared. It was developed in the 1830s by a Belgian astronomer and statistician named Adolphe Quetelet to study populations, not individuals. He was clear about that distinction. The math behaves reasonably well for groups, and is a much rougher tool when applied to one human standing in front of you.

BMI does not measure fat, where fat sits in the body, muscle, bone density, or fluid. It does not adjust for age, sex distribution patterns, ethnicity, or fitness. It collapses everything that makes a body what it is into a single ratio. That is not a flaw of the math. It is what the math was built for.

Why your clinician still uses it at every visit

Even with all of those caveats, BMI keeps showing up on every visit summary. There is a system reason for that.

In American medicine, BMI is wired into how care gets delivered, paid for, and approved. Diagnosis codes for overweight and obesity are tied to BMI thresholds. Insurance coverage for certain medications, surgical procedures, and even some preventive screenings is gated by BMI. Hospital quality metrics track it. Risk calculators feed it into algorithms that determine who is offered which test.

BMI is, in other words, a billing and coding instrument that was also pressed into service as a clinical risk tool. The first job is largely administrative. The second is where things get complicated.

In June 2023, the American Medical Association formally adopted new policy recognizing BMI’s significant limitations and recommending it no longer be used as a standalone measure of health. The AMA pointed to BMI’s roots in data collected from earlier generations of mostly non-Hispanic white populations, its poor performance across different bodies, and its history of being used for racist exclusion. The recommendation: pair BMI with measures of visceral fat, body composition, waist circumference, and metabolic markers.

Where the field is moving: the 2025 Lancet Commission

A larger shift followed in January 2025, when The Lancet Diabetes & Endocrinology published a Commission led by Professor Francesco Rubino and over 50 international experts. Their conclusion was direct. BMI alone is not enough to diagnose obesity in any one person, and it should be used mainly for tracking populations, screening, and research, not as a verdict on an individual’s health.

The Commission also introduced two new terms that patients are starting to hear in clinics and online, often without much explanation:

Preclinical obesity. A person carries extra body fat, but the body is still working well. There is no current illness caused by the extra fat. The future risk of conditions like type 2 diabetes, heart disease, and certain cancers is higher than average.
Clinical obesity. The extra body fat is already causing problems, such as organ strain or limits on daily activity. In this case, obesity is treated as a chronic illness in its own right, not just a risk factor for future problems.

To place anyone in either group, the Commission recommends confirming excess body fat with a measurement that actually looks at the body, not just height and weight. That can be a waist circumference, a waist-to-hip or waist-to-height ratio, or a body composition scan such as DEXA. The Commission also notes that healthy ranges for these measurements differ by ancestry, which the standard U.S. chart does not currently reflect.

What this comes down to in practice: the Lancet Commission is doing at the diagnostic level what the AMA did at the policy level. Both are pulling BMI out of the role of solo decision-maker and asking it to share the work with measurements that actually reflect where fat sits and what it is doing. The categories may sound clinical, but the underlying message is simple. The number on the chart is a starting point, not a diagnosis.

That guidance matters. It will also take time to filter into every clinic, every electronic health record, and every insurance approval. In the meantime, the BMI number is still on the page.

How BMI shows up in my own practice

To be direct: BMI is not a measure I find clinically useful for any individual patient, for the reasons above and the ones still to come. It is also a measure I am required to record, because the diagnosis codes, insurance approvals, and access pathways described above are gated by it. A patient who needs a particular workup, medication, or procedure may need a particular BMI on the chart to qualify. That is the system we work in.

So BMI gets recorded because the rules require it. It is not a tool I use to judge a person’s health, monitor their progress, or define what success looks like. The conversation in the exam room is built around everything the number cannot see.

That is also why BMI is the wrong tool for the rest of your life. Watching your BMI shift by a fraction of a point will tell you almost nothing useful about your body, your metabolic health, or where your effort is paying off. Better tools exist for those questions, and they are coming up next.

Four places BMI fails the individual

1. It cannot see body composition

A six-foot, 200-pound person can be a powerlifter with single-digit body fat or a sedentary office worker with significant adiposity and low muscle mass. Same BMI, wildly different metabolic and longevity profiles.

Muscle is denser than fat. People with high muscle mass land in higher BMI categories without higher health risk. People with low muscle mass and elevated body fat (a pattern sometimes called sarcopenic obesity) can sit in a “normal” BMI range while carrying the metabolic signature of obesity. BMI cannot distinguish between them.

2. It cannot see where the fat is

Where fat lives matters more than how much there is. Visceral adipose tissue, the fat stored deep in the abdominal cavity around organs, behaves very differently than subcutaneous fat (the kind that sits under the skin on the hips, thighs, and buttocks). Visceral fat is metabolically active, inflammatory, and far more strongly linked to insulin resistance, cardiovascular disease, and type 2 diabetes than subcutaneous fat.

Two people with identical BMIs can have very different ratios of visceral to subcutaneous fat, and therefore very different health risk. A waist circumference, a waist-to-height ratio, or a body composition scan can see the difference. BMI cannot.

3. It does not adjust for age

The BMI category that minimizes mortality risk shifts with age. A 2014 meta-analysis of 32 studies and nearly 200,000 community-dwelling older adults found that, after age 65, being in the “overweight” range was not associated with increased mortality. Being in the lower end of the “normal” range actually carried higher risk, likely reflecting unintentional weight loss, sarcopenia, and frailty. Mortality risk did not begin to rise meaningfully until BMI exceeded 33.

The BMI threshold that flags risk in a 35-year-old is not the same threshold that flags risk in a 75-year-old. A standard chart treats them as equivalent.

4. It does not adjust for ethnicity

The thresholds used in the United States were built largely on data from non-Hispanic white populations. A WHO expert consultation published in The Lancet in 2004 concluded that for many Asian populations, cardiometabolic risk begins climbing at BMIs well below the standard 25 kg/m² cutoff. A South Asian individual with a BMI of 24 may carry the same diabetes and cardiovascular risk as a non-Hispanic white individual with a BMI of 28 or higher.

Ethnicity-specific thresholds are now used by some health systems. The standard U.S. chart is not one of them.

The two phenotypes BMI completely misses

Two terms have appeared in the research lit...

Zero Longevity

Message History

Delivery option(s)