Is BMI Accurate? What It Gets Right and Where It Fails

Is BMI accurate?

BMI is accurate enough for sorting large groups of people by size, and much less accurate at telling one person how much fat they carry. The CDC calls BMI a screening measure that "does not distinguish between fat, muscle, and bone mass," and a 2025 Lancet commission of 58 experts said it should be used for population-level risk, research and screening, not as an individual measure of health.

So the number isn't useless. It's a starting point that needs company, usually a tape measure.

How is BMI calculated?

BMI is your weight divided by your height squared: kg ÷ m² in metric, or pounds ÷ inches² × 703 in US units (StatPearls). If you'd rather not do the arithmetic, you can calculate your BMI and healthy weight range with our free calculator.

A worked example: someone 5 ft 9 in (1.75 m) who weighs 170 lb (77 kg) has a BMI of about 25.1. Metric: 77 ÷ (1.75 × 1.75) = 25.1. US: 170 ÷ (69 × 69) × 703 = 25.1.

That's the whole formula. There's no input for muscle, waist, age or sex, which is where most of its problems come from.

What do the BMI categories mean?

For adults, the CDC categories match the WHO definitions: overweight from 25, obesity from 30. In the UK, NICE uses lower thresholds for groups whose health risk tends to rise at a lower BMI.

The NICE column comes from its 2025 guideline.

What is BMI good for?

Speed, cost and consistency. You need a scale and a height, and the result is the same whoever takes it. The CDC describes BMI as "moderately to strongly associated" with other body fat measures, which is why researchers can compare obesity rates across countries and decades with it.

It also rarely raises a false alarm. A 2010 meta-analysis of 25 studies and about 32,000 people found that a BMI of 30 or more almost never labelled someone obese who didn't have excess body fat when measured directly (97% specificity).

The trouble is the other direction.

Why is BMI inaccurate for some people?

The same meta-analysis found standard BMI cutoffs missed about half of the people who did have excess body fat. You can sit in the "healthy" range and still carry more fat than is good for you. The reverse happens too.

Muscle. BMI counts every pound the same. A strength athlete can be classed as overweight with little body fat, as the NHS points out, and NICE tells clinicians to interpret BMI with caution in very muscular adults.

Sex. In the HERITAGE Family Study, women's body fat measured by underwater weighing was about 10.4 percentage points higher than men's at the same BMI. The adult categories don't change by sex, so the same number means different things.

Age. The HERITAGE analysis and an earlier study of 1,626 adults both found that age, along with sex and ethnicity, changed how much fat a given BMI predicted.

Ethnicity. A WHO expert consultation found that many Asian people face high risk of type 2 diabetes and heart disease below the standard overweight cutoff of 25, with risk thresholds ranging from 22 to 25 across populations. That's the basis for the lower NICE cutoffs.

Where the fat sits. BMI says nothing about distribution. Fat around the middle is the kind NICE links to type 2 diabetes, high blood pressure and heart disease, and two people with the same BMI can carry very different amounts of it.

What should you measure alongside BMI?

Your waist, turned into a waist-to-height ratio. It's free, takes a minute and picks up the abdominal fat BMI can't see.

NICE recommends it alongside BMI for adults with a BMI under 35:

• 0.4 to 0.49: healthy
• 0.5 to 0.59: increased health risk
• 0.6 or more: further increased risk

NICE says these cutoffs work for both sexes, all ethnicities and people with high muscle mass, which covers several of BMI's weak spots at once. The short version: keep your waist to less than half your height.

To measure, find the bottom of your ribs and the top of your hips, wrap the tape midway between them, breathe out normally and read it. Divide by your height in the same units. Our 5 ft 9 in example with a 34-inch waist gets 34 ÷ 69 = 0.49.

The Lancet commission goes further. It recommends confirming excess fat with a direct body fat measurement where available, or at least one body measurement such as waist circumference, waist-to-hip ratio or waist-to-height ratio. Above a BMI of 40, it says excess fat can be assumed.

How should you use BMI if you track your weight?

As a slow-moving reference point, not a daily verdict. BMI moves only when your weight does, so it inherits all the scale's day-to-day noise. At 5 ft 9 in, an ordinary 2-pound overnight swing shifts BMI by about 0.3.

Working it out from a smoothed average gives a steadier number; trend weight vs scale weight explains why that's a better input than any single weigh-in. Periscope's Reports cards, for example, show BMI calculated from your trend weight once you add your height. If you're weighing up apps that do this, we've compared BMI tracker apps for iPhone.

Add a monthly waist measurement and you have a much fuller picture than BMI alone. And if your BMI or waist ratio lands in a higher-risk range, ask a doctor about blood pressure, cholesterol and blood sugar, the other factors the CDC says to consider alongside BMI.

Sources

1. CDC. About Body Mass Index (BMI).
2. CDC. Adult BMI Categories.
3. WHO. Obesity and overweight fact sheet.
4. Zierle-Ghosh A, Jan A. Physiology, Body Mass Index. StatPearls. 2023.
5. Rubino F, et al. Definition and diagnostic criteria of clinical obesity. Lancet Diabetes Endocrinol. 2025.
6. NICE. Overweight and obesity management (NG246): identifying and assessing overweight, obesity and central adiposity.
7. NHS. Obesity: overweight and obesity in adults.
8. Okorodudu DO, et al. Diagnostic performance of body mass index to identify obesity as defined by body adiposity. Int J Obes. 2010.
9. Jackson AS, et al. The effect of sex, age and race on estimating percentage body fat from body mass index: the Heritage Family Study. Int J Obes Relat Metab Disord. 2002.
10. Gallagher D, et al. Healthy percentage body fat ranges: an approach for developing guidelines based on body mass index. Am J Clin Nutr. 2000.
11. WHO Expert Consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet. 2004.