Is BMI Accurate? What Changed in 2025 (and What Didn't)

This article is for informational purposes only and does not constitute medical advice. It explains how the medical definition of obesity changed. It cannot tell you your diagnosis, and nothing here is a reason to change how you take a medication. Always consult your doctor or pharmacist before making any changes to your medication routine.

For adults, BMI still counts, but it was demoted. In January 2025 a global commission reclassified BMI as a screening tool rather than a diagnosis, and said confirming obesity needs a waist-based measurement too. Children and teens have their own separate criteria.

Smaller than "BMI is garbage." Bigger than most articles admit.

The question behind the question

Most people who search this are not running a research study. They got a number, and it did not match what they see or how they feel. A lifter with a BMI of 27. A runner told she is "overweight." Someone whose doctor mentioned a weight-loss medication after glancing at a chart.

The usual answer you will find is that BMI cannot tell muscle from fat. True, and about ten years out of date. The actual news is that the medical definition of obesity was rewritten, and BMI's job inside that definition got much narrower.

What the 2025 commission actually said

In January 2025, a commission of 58 experts led by Francesco Rubino of King's College London published new diagnostic criteria in The Lancet Diabetes & Endocrinology. It is worth reading what they wrote rather than what headlines said about it.

They did not retire BMI. Their own text calls it "a valuable screening tool to help identify subjects with potential excess/abnormal adiposity".

What they narrowed is where BMI alone is allowed to be the answer. BMI-only measures, the commission wrote, should be used "only as a surrogate measure of health risk at a population level, for epidemiological studies or for screening purposes."

Population level. Not you, standing on a scale.

For an individual, the commission requires confirming excess body fat one of three ways:

1. Direct body fat measurement, such as a DEXA scan or bioimpedance
2. At least one waist-based measurement in addition to BMI
3. At least two waist-based measurements, regardless of BMI

The waist-based options are waist circumference, waist-to-hip ratio, or waist-to-height ratio. The commission also states that "age-, gender-, and ethnicity-appropriate cut-off points should be used." That last clause matters more than it looks, and we will come back to it.

The two words that changed everything: clinical and preclinical

Here is the part that reshapes real decisions.

The commission split obesity into two states:

Preclinical obesity is excess body fat with organs still working normally, plus a higher risk of problems later.

Clinical obesity is excess body fat that is already affecting how your body works. Diagnosing it takes confirmed excess fat plus evidence of reduced organ or tissue function, or real limits on daily activities like walking, dressing, or bathing.

So obesity stopped being a number and became a number plus a functional assessment. No calculator on the internet, including ours, can tell you which state you are in. That takes a physical exam, lab work, and a doctor.

Your diagnosis barely moves. Your treatment conversation might.

This is the finding almost nobody has written about for a general reader.

Researchers ran the new criteria against national US health survey data from 2021 to 2023, published in Clinical Obesity. Three numbers came out:

• 96.8% of people with a BMI over 30 also had an elevated waist measurement. Adding the waist step reclassified almost nobody.
• Overall obesity prevalence stayed at about 39.6%, essentially unchanged.
• But 41.3% of people who currently fit the indication for obesity medication do not have clinical obesity under the new framework.

Read that last one carefully, because it is easy to overstate. It does not say those people lose access to medication. It says they land in the preclinical group. The commission's own guidance there is that preclinical obesity does not generally require drug treatment when the risk of progressing to clinical obesity is judged low, while people at higher risk should still have access to it.

In plain terms: for a large group of people, an automatic yes becomes a conversation.

If that is the part you came for, the labels themselves are worth reading, because the BMI number people quote for these medications is not in the indication at all.

Why the standard calculator can be wrong for you specifically

Remember that clause about "ethnicity-appropriate cut-off points." It is doing quiet work.

The American Medical Association adopted a policy in June 2023 noting that BMI "loses predictability when applied on the individual level," and that the data behind it came primarily from "previous generations of non-Hispanic white populations."

The AMA policy is narrower than it is usually reported. It does not ban BMI from medicine. What it specifically says is that BMI "should not be used as a sole criterion to deny appropriate insurance reimbursement." It also lists what should sit alongside BMI: visceral fat, body composition, relative fat mass, waist circumference, and genetic or metabolic factors.

So two separate bodies, two years apart, landed on the same instruction. One number is not enough, and the cutoffs should not be identical for everyone.

The waist number that goes with your BMI

If you only add one thing to your BMI, add a tape measure. Where you put it changes the number, so the method matters. The NHLBI instructions are to stand up, place the tape around your middle just above your hipbones, and take the reading just after you breathe out. Measuring at your belt line or at your narrowest point will give you a different figure, and the thresholds below assume the hipbone method. The reference table below comes from the NHLBI clinical guidelines published in 1998, and it is still the grid your doctor is most likely using. That gap is part of the story: the working chart is 27 years older than the definition it sits under.

These categories describe associated disease risk for groups of adults, not a prediction about any one person, and they are not the pediatric criteria. The columns are the point. Two people with the same BMI of 32 can sit in different risk rows depending on one tape measure. The NHLBI also notes that a larger waist can signal risk even when BMI looks normal, which is exactly the case the old single number missed.

What our calculator does differently

Most BMI calculators do the 1830s arithmetic and stop. Ours asks for the second measurement too, then reads it two ways.

It divides your waist by your height, which is one of the three waist-based options the commission accepts. That ratio is scored against the bands in NICE guideline NG246, which sets 0.4 to 0.49 as healthy central adiposity, 0.5 to 0.59 as increased, and 0.6 or more as high. NICE states those bands hold for both sexes and all ethnicities, including adults with high muscle mass, for a BMI under 35. It also maps your waist and BMI together onto the NHLBI grid above.

There is deliberately no ethnicity switch. The commission asks for ethnicity-appropriate cutoffs, but no primary source publishes a clean set of them to swap in. The 2004 WHO expert consultation proposed action points along the BMI continuum at 23.0, 27.5, 32.5 and 37.5 for Asian populations while stating that the available data do not indicate a single clear cut-off point for all Asians. So the calculator reports that as a continuum rather than inventing a category swap it cannot source.

What it will not do is tell you that you have obesity, clinical obesity, or that you qualify for any medication. Those are diagnoses, and the 2025 criteria are explicit that diagnosis requires evaluating organ function. A web form cannot check your liver panel.

If you are already taking medication for this

A lot of readers arrive here after a prescription, not before one. If that is you, the framework shift is less about your label and more about what gets watched over time.

The whole logic of clinical obesity rests on organ function, and organ function is what daily medication is usually protecting. That is where consistency stops being an abstract virtue. Metformin has meal-timing rules that differ by formulation, and long-term use comes with a nutrient level worth monitoring. Weekly injectables carry their own timing anxiety and a specific window when a dose is missed, whether that is Wegovy or Ozempic. Even a stomach bug changes the rules, which is why sick-day guidance exists. And if weight-loss surgery is part of your history, absorption itself can shift, which is what happens with thyroid medication after a gastric bypass.

Pillo is built for the boring part of that. Its alarms keep going until you actually mark the dose taken, instead of one polite notification that disappears while your hands are full. You can track more than one medication on separate schedules, and manage doses for the people you care for from your own phone.

Download Pillo on Google Play

FAQ

Is BMI accurate for an individual person?

Not on its own. The 2025 Lancet commission classifies BMI as a screening measure appropriate at a population level, and requires either a direct body fat measurement or at least one waist-based measurement before confirming excess body fat in an individual. The AMA reached a similar conclusion in 2023.

Why does my BMI say I am overweight when I do not look it?

BMI uses only height and weight, so it cannot separate muscle from fat, and it does not account for where fat sits on your body. This is a known limitation, which is why the current criteria pair BMI with a waist-based measurement rather than treating BMI as the final word.

Is a BMI of 26 something to worry about?

A BMI of 26 falls in the overweight range on the NHLBI table, but the risk row it lands in depends on your waist measurement. The same BMI can carry "increased" or "high" associated risk depending on that second number. Your doctor is the right person to interpret the pair.

What is the difference between clinical and preclinical obesity?

Preclinical obesity means excess body fat while organs still function normally, with elevated future risk. Clinical obesity means excess body fat that is already causing reduced organ or tissue function, or significant limits on daily activities. The distinction determines whether the guidance points toward routine treatment or an individual risk and benefit discussion.

Is BMI still used by doctors?

Yes. The 2025 criteria keep BMI as the first step and call it a valuable screening tool. What changed is that BMI alone no longer finishes the job for an individual patient. Your doctor is meant to pair it with a waist-based measurement or a direct body fat measurement before confirming excess body fat.

What is a healthy waist measurement?

The NHLBI risk table treats waist above 40 inches in men and above 35 inches in women as the threshold where associated risk moves up a row. Measure standing, with the tape just above your hipbones, right after breathing out. The 2025 criteria also call for cutoffs appropriate to age, gender, and ethnicity, so a single universal number is not the whole picture.

Did the new criteria change who can get obesity medication?

The criteria are diagnostic, not insurance rules, so nothing changed automatically. The Clinical Obesity analysis found that 41.3% of people who fit current medication indications would be classified as preclinical rather than clinical obesity. For that group the commission advises that drug treatment is not generally required when progression risk is low, with access preserved for those at higher risk. What that means for any specific prescription is a decision for you and your prescriber.

A note on how this is discussed

The commission itself spends pages on weight bias and stigma, and that is worth carrying into any conversation about these numbers. A BMI reading is a screening result, not a verdict on a person. If a number has ever been used to dismiss a symptom you raised, that experience is documented in the literature, and it is a reason the criteria changed.

This article is for informational purposes only and does not constitute medical advice. Always consult your doctor or pharmacist before making any changes to your medication routine. Consult your doctor or pharmacist for advice specific to your medications.