BMI matters, but less than most people think. It’s a useful starting point for spotting weight-related health risks across large populations, yet it misses critical details about any single person’s body. The number on its own can’t tell you how much of your weight is muscle versus fat, where your body stores that fat, or whether your blood sugar and cholesterol are actually in a healthy range. Understanding what BMI can and can’t do helps you put your own number in perspective.
What BMI Actually Measures
BMI is a ratio of your weight to your height, nothing more. It divides your weight in kilograms by your height in meters squared. The standard categories for adults 20 and older break down like this:
- Underweight: below 18.5
- Healthy weight: 18.5 to 24.9
- Overweight: 25.0 to 29.9
- Class 1 obesity: 30.0 to 34.9
- Class 2 obesity: 35.0 to 39.9
- Class 3 (severe) obesity: 40.0 and above
These thresholds were designed as a screening tool. They flag people who may be at higher risk so a doctor can dig deeper. BMI was never intended to be a diagnosis on its own.
Where BMI Gets It Right
At the population level, BMI tracks meaningfully with mortality risk. A massive meta-analysis pooling data from 239 prospective studies across four continents found that the lowest risk of death from any cause falls in the 20.0 to 25.0 range. Once BMI climbs above 25, risk rises in a clear, graded pattern. Compared to people in the 22.5 to 25.0 range, those with a BMI of 27.5 to 30 had a 20% higher risk of death, those with class 1 obesity (30 to 35) had a 45% higher risk, and those with class 3 obesity (40 and above) had a 176% higher risk. Being significantly underweight carries danger too: a BMI below 18.5 was associated with a 51% increase in mortality.
These are not small numbers. For identifying broad risk categories in millions of people, BMI works. It’s cheap, fast, and requires no equipment beyond a scale and a tape measure. That practical simplicity is genuinely valuable in public health.
Why BMI Fails Individuals
The problems start when you treat BMI as a precise health verdict for one person. Because it measures relative weight, not body composition, it can’t distinguish between someone carrying 30 extra pounds of muscle and someone carrying 30 extra pounds of fat. A study of adolescent athletes found that 62% of those classified as obese by BMI were false positives. When their body fat was measured directly with skinfold testing, they weren’t obese at all. Their muscle mass had inflated the number.
The reverse problem is just as serious. Roughly one in four people with a normal-weight BMI are what researchers call “metabolically unhealthy normal weight.” A study of over 37,000 adults found that 23.3% of those in the healthy BMI range already met criteria for metabolic problems like elevated blood sugar, high blood pressure, or abnormal cholesterol. Their BMI looked fine. Their health did not. This is the pattern sometimes called “skinny fat,” where a person carries too little muscle and too much visceral fat without it showing up on the scale.
Fat Location Matters More Than Total Weight
Not all body fat is equally dangerous. Fat stored around your organs, known as visceral fat, drives inflammation and metabolic disease far more aggressively than fat stored under the skin on your hips or thighs. BMI cannot tell the difference.
An analysis of nearly 388,000 participants in the UK Biobank found that waist-to-hip ratio was a stronger and more consistent predictor of death from any cause than BMI. Using a genetic analysis method that can help establish cause and effect, the researchers concluded the link between waist-to-hip ratio and mortality is likely causal, not just a correlation. In practical terms, two people with identical BMIs can have very different risk profiles depending on where their body stores fat. A tape measure around the waist can sometimes tell you more than a BMI calculator.
BMI Thresholds Don’t Fit Every Ethnicity
The standard BMI cutoffs were developed primarily from data on White European populations. They don’t translate cleanly to other groups. South Asian populations, for example, develop type 2 diabetes at significantly lower BMIs. A large population-based study in England found that the diabetes risk a White person faces at a BMI of 30 kicks in at a BMI of just 23.9 for South Asian individuals. That’s a gap of more than six BMI points.
This has real consequences. A South Asian person with a BMI of 26 might be told they’re only mildly overweight by standard charts, while their actual metabolic risk is equivalent to someone with clinical obesity. The UK’s National Institute for Health and Care Excellence now recommends using a BMI of 27.5 as the action threshold for South Asian and Chinese populations, but even that may be too generous given the data.
The Rules Change as You Age
For adults over 65, the standard BMI categories appear to be miscalibrated. A study tracking older adults found the lowest mortality risk at a BMI of 27.1, which falls squarely in the “overweight” category by standard definitions. The sweet spot where death risk from all causes stayed within 10% of the minimum ranged from 24.2 to 30.1. Being below 21.1 was significantly more dangerous than being at 28.
This likely reflects the fact that older adults lose muscle mass naturally, and carrying some extra weight provides a buffer during illness, surgery, or falls. The WHO cutoff of 25 for overweight may simply be too low for this age group. If you’re over 65 and your doctor is focused solely on getting your BMI below 25, it’s worth a conversation about what the evidence actually shows for your age.
Better Ways to Assess Your Health
BMI is one data point in a much larger picture. Waist circumference gives a rough estimate of visceral fat. A waist measurement above 40 inches for men or 35 inches for women signals elevated risk regardless of BMI. Waist-to-hip ratio narrows it further, and the research suggests it outperforms BMI for predicting both cardiovascular events and overall mortality.
A newer metric called the Body Roundness Index (BRI) uses waist and hip measurements to estimate body shape in a way that’s independent of height. It predicts visceral fat more accurately than BMI alone. In one validation study, waist and hip measurements together explained roughly twice as much of the variation in visceral fat as BMI did in men. The tradeoff is that BRI requires a more complex calculation, which is why it hasn’t replaced BMI in routine practice yet.
The most telling indicators of metabolic health, though, aren’t body measurements at all. Blood pressure, fasting blood sugar, cholesterol ratios, and markers of inflammation paint a far more detailed picture than any number derived from a scale. The 2025 clinical guidelines from the American Association of Clinical Endocrinology explicitly moved away from a singular focus on BMI, emphasizing that treatment decisions should target actual health complications rather than a weight category.
BMI is a rough filter, not a final answer. It can tell you that something might deserve a closer look. It cannot tell you whether you’re healthy.