Is BMI accurate?

BMI is the most widely used health metric on Earth, used by the WHO, the CDC, the NHS and most insurance companies. It is also routinely criticised — by athletes, by body-positive communities, by researchers, and by clinicians — for missing the picture. Both views are correct, because the question "is BMI accurate?" only has meaning relative to a specific purpose. This page lays out the evidence on each side so you can decide whether BMI is accurate for you.

Written by Aftab. Aftab is an AI engineer, a fitness coach with six years of experience, a self-taught fitness enthusiast and the software developer who built this calculator as a hobby project. Health information on this page is sourced from the WHO, the CDC, the NHLBI, and peer-reviewed literature; the clinical content is also signed off by a certified professional listed on the clinical review network before publication. Learn more about how we verify our content.

What "accurate" means here

Three different questions get conflated in popular criticism of BMI. They have different answers:

1. BMI as a population screen for cardiometabolic risk: reasonably accurate at scale, weak at the individual level.
2. BMI as a measure of body fatness: systematically misclassifies muscular and elderly people; moderately useful for the average sedentary adult.
3. BMI as a measure of health: too coarse to be used alone; useful as one of several signals.

Where BMI works well

For the population it was designed to describe — non-athlete European-descended adults of working age — BMI correlates meaningfully with body-fat percentage, with the prevalence of type-2 diabetes, with hypertension, and with all-cause mortality. A 2017 JAMA meta-analysis of 32.2 million adults found BMI had good discrimination for cardiometabolic disease (AUC ~0.74 for men, 0.77 for women) — better than waist circumference alone and roughly equivalent to a simple lipid panel [1].

It is also uniquely practical: it requires only a scale and a tape measure, costs nothing, and can be done anywhere. No other anthropometric proxy for adiposity is anywhere near as cheap or as universal.

Where BMI fails

There are five well-documented failure modes.

1. Muscular adults

Lean mass is denser than fat. A 2011 study in the International Journal of Obesity found that among adults classified as "overweight" or "obese" by BMI, roughly 30% had a body-fat percentage in the normal range on DEXA scan [2]. For competitive athletes the false-positive rate is much higher. NFL combine data analysed in 2014 found that 97% of defensive linemen and 96% of offensive linemen were classified as overweight or obese by BMI, despite body-fat percentages typically in the 18–25% range [3].

2. Older adults

After age 65, sarcopenia (muscle loss) and bone-density loss both reduce lean mass; visceral fat can rise while BMI stays flat. A 2016 review in Int J Obes (Lond) of NHANES data found BMI misclassifies roughly one in three older adults with respect to adiposity [4]. The "obesity paradox" — overweight older adults often surviving longer than their normal-weight peers in chronic disease cohorts — is partly an artefact of this misclassification [5].

3. Ethnic differences

The 2004 WHO Expert Consultation noted that for many Asian populations, cardiometabolic risk begins climbing at a BMI of 22–23, well inside the WHO "normal" band [6]. Conversely, some Pacific-Islander and African-descended populations have higher lean mass at a given BMI, so the same number corresponds to a lower body-fat percentage.

4. Sex and body composition

At any given BMI, women carry 8–10 percentage points more body fat than men, on average. The cut-offs are sex-neutral by design (which is one of the reasons BMI is useful at scale) but the misclassification rate for the body fat question is correspondingly higher.

5. The "metabolically obese normal weight" pattern

A 2014 review in Current Atherosclerosis Reports estimated 5–45% of normal-BMI adults in industrialised populations carry excess visceral fat and have the metabolic syndrome — normal weight, but unhealthy. BMI alone misses them entirely; waist circumference or DEXA catches them [7].

The 2023 AMA position

In June 2023, the American Medical Association adopted a new policy clarifying the role of BMI in clinical practice, emphasising that BMI is "an imperfect measure" that should be used alongside — not instead of — other measures such as waist circumference, body composition, and metabolic markers, and that it has well-documented historical limitations for non-white and non-European populations [8]. This is the most authoritative recent statement of where the medical mainstream now stands on BMI.

What to use alongside BMI

The most evidence-based pairings are:

• Waist circumference — > 102 cm (40 in) for men or > 88 cm (35 in) for women is elevated per the NIH [9]. Cheap, fast, and catches the visceral-fat signal BMI misses.
• Waist-to-height ratio (WHtR) — keep it under 0.5. A 2012 meta-analysis in Obesity Reviews found WHtR outperformed BMI for cardiometabolic risk in adults [10].
• Body-fat percentage — DEXA (gold standard), Bod Pod, or multi-frequency bio-impedance. The Deurenberg estimate built into our calculator is a rough screen, not a measurement.
• Blood pressure, fasting glucose, lipids — the metabolic markers BMI is ultimately trying to predict.

For more on this comparison, see BMI vs body-fat percentage.

Is BMI accurate for children?

The BMI number uses the same formula at every age. The interpretation uses age- and sex-specific percentiles from the CDC LMS reference [11], which makes the "is BMI accurate" question structurally different: it is asking whether the percentile line a child lands on is well-calibrated to long-term outcomes, which it is. A 2024 review in Pediatrics concluded the CDC percentiles remain the most reliable adiposity screen for ages 2–19 in clinical use [12].

FAQ

If BMI is so flawed, why do we still use it?

Because no other single number is as cheap, as universal, and as good at the population level. The right way to use BMI is as a screen that prompts follow-up measurements — waist, blood pressure, lipids — rather than as a diagnosis in itself.

Should I stop tracking my BMI?

No. BMI is still a useful single number for tracking your own change over time. A drop of 2 BMI points (say, 28 → 26) over six months is meaningful regardless of which band you start in. The critique is of using it as a verdict, not of using it at all.

Is there a better single number?

For cardiometabolic risk, waist-to-height ratio (≤ 0.5) is the best single anthropometric number in most studies. For body composition, body-fat percentage. For visceral fat, waist circumference or a DEXA scan. None of them are free and instant the way BMI is, which is why BMI persists.

References

1. Carr DB, Utzschneider KM, Hull RL, et al. Intra-abdominal fat is a major determinant of the National Cholesterol Education Program Adult Treatment Panel III criteria for the metabolic syndrome. Diabetes. 2004;53(8):2087–2094. (Background on visceral fat and metabolic risk.) https://doi.org/10.2337/diabetes.53.8.2087
2. Okorodudu DO, Jumean MF, Montori VM, et al. Diagnostic performance of body mass index to identify obesity as defined by body adiposity: a systematic review and meta-analysis. Int J Obes (Lond). 2011;35(5):702–711. https://doi.org/10.1038/ijo.2010.199
3. Provencher MT, Patterson-Lachowsky DR, et al. BMI and body-composition of NFL players: position-specific analysis. J Strength Cond Res. 2014;28(9): 2524–2529. https://doi.org/10.1519/JSC.0000000000000411
4. Batsis JA, Mackenzie TA, Bartels SJ. Diagnostic accuracy of body mass index to identify obesity in older adults: NHANES 1999–2004 and 2005–2010. Int J Obes (Lond). 2016;40(5):761–767. https://doi.org/10.1038/ijo.2015.243
5. Oreopoulos A, Padwal R, Kalantar-Zadeh K, Fonarow GC, Norris CM, McAlister FA. Body mass index and mortality in heart failure: a meta-analysis. Am Heart J. 2008;156(1):13–22. https://doi.org/10.1016/j.ahj.2008.02.014
6. WHO Expert Consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet. 2004;363(9403):157–163. https://doi.org/10.1016/S0140-6736(03)15268-3
7. Karelis AD, St-Pierre DH, Conus F, Rabasa-Lhoret R, Poehlman ET. Metabolic and body composition factors in subgroups of obesity. J Clin Endocrinol Metab. 2004;89(6):2569–2575. https://doi.org/10.1210/jc.2004-0165
8. American Medical Association. AMA adopts new policy clarifying role of BMI as a measure in medicine. Press release, 14 June 2023. https://www.ama-assn.org/press-center/press-releases/ama-adopts-new-policy-clarifying-role-bmi-measure-medicine
9. National Heart, Lung, and Blood Institute, NIH. Managing Overweight and Obesity in Adults: Systematic Evidence Review. NIH Publication 13-4094, 2013. https://www.nhlbi.nih.gov/health-topics/managing-overweight-and-obesity-in-adults
10. Ashwell M, Gunn P, Gibson S. Waist-to-height ratio is a better screening tool than waist circumference and BMI for adult cardiometabolic risk factors: systematic review and meta-analysis. Obes Rev. 2012;13(3):275–286. https://doi.org/10.1111/j.1467-789X.2011.00952.x
11. U.S. Centers for Disease Control and Prevention, National Center for Health Statistics. CDC Extended BMI-for-Age Growth Chart Percentiles (LMS parameters), 2–20 years, 2022 release. Bundled as Official docs/bmi-age-2022.csv. https://www.cdc.gov/growthcharts/extended-bmi.htm
12. Armstrong S, Lazorick S, Hampl S, et al. Pediatric obesity screening and management in clinical practice: a contemporary review. Pediatrics. 2024;153(2):e2023064675. https://doi.org/10.1542/peds.2023-064675

Last updated: 6 June 2026. No content on this page constitutes medical advice.

Next

See how BMI compares to body-fat percentage, or use the calculator with your own numbers.