BMI is one of the most widely used health metrics in the world — and one of the most misunderstood. Critics call it outdated and misleading. Defenders point to decades of epidemiological evidence. The truth is more nuanced: BMI is a genuinely useful tool in the right context, and a genuinely unreliable one in the wrong context.
This guide explains exactly where BMI works, where it fails, and what to use instead when it falls short.
Sedentary or lightly active adults of average build, large-scale population health research, identifying clear cases of underweight or severe obesity, and initial screening before more detailed assessment.
At the population level, BMI correlates strongly with body fat percentage, metabolic disease risk, cardiovascular outcomes, and all-cause mortality. Dozens of large-scale studies — tracking hundreds of thousands of people over decades — consistently show that people with BMI above 30 have meaningfully higher rates of type 2 diabetes, heart disease, and premature death.
This is why health systems worldwide use BMI as a screening tool. It's inexpensive, requires no equipment, and identifies risk at a population scale. For the average adult who doesn't do significant resistance training, BMI is a reasonable first-pass estimate of health risk.
When BMI works vs when it misleads
A rugby player, powerlifter, or dedicated gym-goer may have a BMI of 27–30 — classified as "overweight" or "obese" — despite having low body fat and excellent metabolic health. BMI sees weight relative to height; it cannot tell whether that weight is muscle or fat.
This is one of the most cited criticisms of BMI, and it's valid. Research suggests that around 20–25% of people classified as "obese" by BMI are metabolically healthy, while a significant proportion of those in the "normal" BMI range have unhealthy metabolic profiles.
Standard BMI thresholds were derived primarily from studies of European populations. Research consistently shows that Asian populations develop metabolic disease at lower BMI values — approximately 3–5 BMI points lower. A BMI of 24 may be associated with low health risk in a European adult and significantly elevated risk in an East or South Asian adult.
The WHO recommends adjusted thresholds for Asian populations: overweight begins at BMI 23.0 (not 25.0), and obesity at 27.5 (not 30.0). Use our Asian BMI Calculator for ethnicity-adjusted results.
Two people can have identical BMIs — say, both at 25 — but completely different health risk profiles depending on where their fat is stored. Visceral fat (fat around the abdominal organs) is far more metabolically dangerous than subcutaneous fat (fat under the skin). BMI cannot differentiate between these two types.
A person with a "normal" BMI of 23 but significant abdominal fat may have higher cardiovascular risk than someone at BMI 27 with fat distributed primarily on the hips and thighs.
As people age, they naturally lose muscle mass and gain fat — even at stable body weight. A 65-year-old with a BMI of 22 typically has significantly more body fat than a 25-year-old at the same BMI. Some research suggests that slightly higher BMI (22–27) may actually be protective in older adults, who face greater risk from being underweight than overweight.
Women naturally carry a higher percentage of body fat than men at the same BMI — a biological difference related to hormonal function, not health risk. The same BMI represents different body compositions in men and women, which the standard thresholds don't fully account for.
| Metric | What it measures | Healthy target |
|---|---|---|
| Waist circumference | Abdominal fat directly | Women <80 cm, Men <94 cm |
| Waist-to-height ratio | Visceral fat relative to height | Below 0.5 for all adults |
| Body fat % | Proportion of fat vs lean mass | Women 20–31%, Men 8–19% |
| Fasting blood glucose | Metabolic function directly | Below 5.6 mmol/L (100 mg/dL) |
| Blood pressure | Cardiovascular risk directly | Below 120/80 mmHg |
For most people, adding waist circumference to BMI gives a significantly more complete picture of metabolic health risk — and requires nothing more than a tape measure.
Yes — with appropriate context. BMI is a useful starting point for most adults. It's free, takes seconds, and identifies the extremes (clearly underweight or clearly obese) reliably. The problems arise when people treat it as a precise individual diagnosis rather than a rough population-level screen.
⚠️ BMI is a screening tool, not a diagnosis. Individual health risk depends on family history, lifestyle, blood markers, and many other factors. Consult your healthcare provider for a complete assessment.
Next step
Get personalised results based on your own numbers — instant, no sign-up required.