BMI has been the standard weight-status metric for decades. It's the number on every doctor's chart, every fitness app, every "what's a healthy weight for my height?" article. But it's also been criticized for as long as it's existed — and in 2026, those criticisms are well-supported by research.
This guide is an honest, evidence-based look at when BMI works, when it fails, and what to use instead. The answer isn't "BMI is useless." It's "BMI is one signal among several — useful in some contexts, misleading in others."
Why BMI is still used
Before we get to the limits, it's worth understanding why BMI has stuck around for 30+ years despite its critics. Five reasons:
- It is cheap. No equipment, no blood draw, no imaging. Just height and weight.
- It is fast. A calculation takes seconds.
- It correlates with cardiometabolic risk at the population level. Large epidemiological studies repeatedly show a J-shaped or U-shaped curve between BMI and mortality.
- It is comparable across studies and decades. Researchers can track changes in populations over time.
- It is simple enough for individuals to calculate. No clinical visit required.
These reasons explain why BMI survives — but they also explain the limits. A metric designed for population-level screening gets used (often implicitly) as if it were a personal diagnosis. That's where the trouble starts.
Where BMI works well
BMI is a reasonable signal for:
- Population screening — public health agencies use it to estimate the prevalence of underweight, overweight, and obesity in a population.
- Average adults with average body composition — for someone who is neither very muscular nor very sedentary, BMI tracks cardiometabolic risk reasonably well.
- Trend monitoring — watching your BMI change over months or years gives a directional signal about weight trends.
- Quick screening in clinical settings — when time is short and a more detailed body-composition assessment isn't available.
Where BMI fails
BMI fails — sometimes badly — for several specific groups. If you fall into one of these, treat your BMI number with extra caution.
1. Athletes and muscular individuals
This is the most common critique, and it's well-founded. Muscle is denser than fat, so a muscular person weighs more per unit of height than a sedentary person of the same height. A lean, muscular bodybuilder at 15% body fat can easily read as overweight or obese by BMI despite being metabolically healthy.
Example: A 6'0" (183 cm) male bodybuilder at 15% body fat weighs about 95 kg (209 lb) — a BMI of 28.3, technically overweight. The same man at 30% body fat would weigh about 109 kg (240 lb) — a BMI of 32.5, obese. The first is metabolically healthy; the second is not. BMI cannot distinguish them.
For a more honest read, see our [Athlete BMI calculator](/calculators/bmi-athletes), which adds body-fat percentage and muscle-mass context.
2. The elderly
Older adults lose muscle mass (sarcopenia) and often gain visceral fat, even at stable weight. A "healthy" BMI of 22 in a 75-year-old may reflect lost muscle and increased fat — a different metabolic profile than the same BMI in a 30-year-old. Mortality research suggests older adults fare better with a BMI slightly above the standard healthy range. See our [Senior BMI calculator](/calculators/bmi-seniors).
3. Asian, Black, and Hispanic populations
Asian populations develop cardiometabolic complications at lower BMIs than European-origin populations. The WHO has issued lower cutoffs (18.5–22.9 healthy, 23–27.5 overweight, 27.5+ obese) for Asian adults. See our [Asian BMI calculator](/calculators/bmi-asian).
Black populations tend to have higher bone density and lean mass at a given BMI, which may make BMI slightly more accurate for them in some respects. However, the relationship between BMI and cardiometabolic risk can differ — research is ongoing.
Hispanic populations show similar BMI–risk patterns to European-origin populations in most studies, though some research suggests waist circumference may matter more. No consensus on adjusted BMI cutoffs exists yet.
4. Pregnant or breastfeeding women
BMI does not apply during pregnancy — gestational weight gain guidelines are based on pre-pregnancy BMI, not on current weight. Breastfeeding changes body composition in ways BMI cannot capture. Use pre-pregnancy BMI as the baseline, and consult your OB or midwife for personalized guidance.
5. People with very different body shapes
Two people with identical BMI can have meaningfully different body compositions — different waist circumference, different limb-to-trunk ratios, different visceral fat. BMI is a single number; bodies are multidimensional.
What BMI misses
The biggest blind spot: BMI tells you nothing about where the weight is. Visceral (abdominal) fat is far more dangerous than subcutaneous (limb) fat. A person with a 34-inch waist and BMI 26 has meaningfully higher cardiometabolic risk than a person with a 30-inch waist and BMI 28 — even though the second has a higher BMI.
BMI also misses:
- Cardiorespiratory fitness. A "fit and fat" person consistently has lower mortality than an "unfit and thin" person, even at the same BMI.
- Metabolic health. Blood pressure, fasting glucose, HbA1c, and lipid panel all predict risk far better than BMI alone.
- Body composition. Two people with BMI 27 can have 15% vs. 35% body fat — the latter has substantially higher risk.
Better alternatives (and when to use them)
Five commonly used alternatives — each with their own strengths and weaknesses.
1. Waist circumference
What it measures: abdominal fat, which correlates more strongly with cardiometabolic risk than total weight.
How to use it: measure at the level of your navel, relaxed. Above 40 in (102 cm) for men or 35 in (88 cm) for women is considered elevated risk.
Best for: everyone. Cheap, fast, more informative than BMI on its own.
2. Waist-to-height ratio
What it measures: abdominal fat relative to height. Stronger predictor than BMI for cardiometabolic risk in adults.
How to use it: divide waist circumference by height. Below 0.5 is generally considered low risk; 0.5–0.6 is elevated; above 0.6 is high.
Best for: everyone; particularly useful across ethnicities and body types.
3. Body-fat percentage
What it measures: the actual proportion of your weight that is fat.
How to measure it:
- DEXA scan — clinical gold standard. ~1% accuracy. Costs $50–150 out of pocket.
- Bioelectrical impedance (BIA) — scales and hand-held devices. Convenient but accuracy varies (3–5% error).
- Skinfold calipers — requires a trained operator; accuracy depends heavily on technique.
- Hydrostatic weighing — historically a gold standard, less common now.
Best for: athletes, anyone whose BMI doesn't match their lived experience.
4. DEXA scan
What it measures: total body composition — fat mass, lean mass, bone mass — plus visceral fat estimate. Clinical gold standard.
How to get one: many sports medicine clinics, radiology centers, and some gyms offer DEXA scans for $50–150 per session.
Best for: detailed body composition assessment, tracking changes over time.
5. Bloodwork + blood pressure
What it measures: actual cardiometabolic risk — not a body composition proxy.
- Fasting glucose
- HbA1c (3-month blood sugar average)
- Lipid panel (LDL, HDL, triglycerides)
- Blood pressure (systolic / diastolic)
- Optional: hs-CRP (inflammation marker), liver enzymes (NAFLD screening)
Best for: everyone — this is the most clinically meaningful set of risk indicators.
A practical framework
How to use BMI in 2026 without over-relying on it:
- Calculate your BMI for context — use our calculator above or any tool.
- Measure your waist circumference. This is more informative on its own than BMI alone.
- Calculate waist-to-height ratio. Aim for under 0.5.
- Get bloodwork. Fasting glucose, HbA1c, lipid panel, blood pressure — annually for most adults.
- Assess cardiorespiratory fitness. How easily you can walk, climb stairs, exercise. This predicts mortality independently of BMI.
- Talk to a clinician about your personal risk profile — family history, medications, existing conditions all matter.
BMI is the start of that conversation, not the end of it.
The "obesity paradox" and other complications
One more nuance worth knowing: in some clinical contexts — heart failure, advanced kidney disease, certain cancers — patients with higher BMI have better outcomes than patients with lower BMI. This is called the "obesity paradox."
It doesn't mean being overweight is healthy. It means BMI is too blunt to tell the whole story, and in some sick populations, the protective effects of lean mass and energy reserves matter more than the metabolic costs of extra weight.
The bottom line
BMI in 2026 is what it has always been: a useful, imperfect, population-level screening tool. Use it as one input among several. Pair it with waist circumference, body composition where possible, bloodwork, and a clinician's assessment. For athletes, muscular individuals, older adults, and Asian populations, BMI is meaningfully incomplete — use the adjusted calculators we link to throughout this article.