What Is an Unhealthy BMI? Categories and Health Risks

A BMI below 18.5 or at 30 and above is generally considered unhealthy for adults, though the risks at each end of the scale look quite different and the number itself misses important context. BMI, which is simply your weight in kilograms divided by the square of your height in meters, was originally developed in the 1830s by Belgian mathematician Adolphe Quetelet and didn’t even get its modern name until 1972.¹ It remains the standard screening tool, but the relationship between that single number and actual health is far messier than the tidy category labels suggest.

The Standard BMI Categories

For adults, the widely used classification breaks down like this:

• Underweight: below 18.5
• Normal weight: 18.5 to 24.9
• Overweight: 25.0 to 29.9
• Obese (Class I): 30.0 to 34.9
• Obese (Class II): 35.0 to 39.9
• Obese (Class III): 40.0 and above

These cutoffs are used globally to classify body size, estimate disease risk, and shape public health policy.² The “unhealthy” label most commonly applies to both tails: underweight and obese. But the overweight category sits in a surprisingly ambiguous zone, and even within the obese range, Class I carries very different risks from Class III.

How BMI Relates to Dying Early

Large-scale studies tracking millions of people find that the relationship between BMI and mortality is not a straight line. Instead, it follows a J-shaped or U-shaped curve, meaning that both very low and very high BMIs are associated with a greater chance of dying earlier, while a broad range in the middle carries the lowest risk. A systematic review and meta-analysis covering 230 cohort studies and over 30 million participants found the lowest mortality risk at a BMI of roughly 23 to 24 in never-smokers who were otherwise healthy, shifting slightly lower (around 20 to 22) in studies with more than 20 years of follow-up.³

Another meta-analysis of 97 cohorts similarly described a U-shaped pattern, but found the lowest mortality in the 25 to 30 range for the general population, with the sharpest jump in risk appearing above a BMI of 35.⁴ The discrepancy between these two studies matters. When researchers carefully exclude smokers and people who may have lost weight due to undiagnosed illness, the optimal BMI tightens toward the normal-weight range. When the analysis includes everyone, reverse causation and confounding can shift the sweet spot upward. The practical takeaway is that a BMI in the low-to-mid twenties appears safest for long-term survival in otherwise healthy people, while BMIs above 35 carry consistently elevated risk regardless of how you slice the data.

Cardiovascular Risks of a High BMI

Heart disease and stroke are the health outcomes most strongly linked to excess weight. Clinical and epidemiological research has tied obesity to coronary heart disease, heart failure, high blood pressure, stroke, atrial fibrillation, and sudden cardiac death.⁵ These are not minor statistical associations; they represent the leading causes of death in most developed countries.

The stroke connection, in particular, has been studied over long time horizons. A Norwegian study tracking participants over 42 years found that people who were overweight throughout adulthood had roughly a 29% higher risk of ischemic stroke compared to those who stayed at a normal weight.⁶ That risk held even after adjusting for other factors like smoking and blood pressure, suggesting the effect of long-term elevated weight is independent of those other hazards.

Diabetes and Metabolic Risk

Type 2 diabetes is one of the most dose-dependent consequences of rising BMI. A Japanese study found that people with a BMI between 25 and 27.4 had about three times the risk of developing type 2 diabetes compared to those below 25, and the risk climbed further beyond that threshold, independent of blood pressure, metabolic profiles, and insulin resistance.⁷ At the extreme end, data from an American cohort showed that at a BMI of 40 or above, men had about four and a half times the risk and women about three and a half times the risk of needing insulin therapy, compared to normal-weight individuals.⁸

The relationship works in the other direction too. Underweight individuals have roughly half the risk of developing type 2 diabetes compared to those at normal weight, a finding that has been replicated in both men and women.⁹ So while a lower BMI protects against diabetes, it creates an entirely different set of problems.

The Dangers of Being Underweight

A BMI below 18.5 is often overlooked when people talk about “unhealthy” weight, but it carries real risks. The most well-documented is weakened bones. Low body weight is closely tied to malnutrition, which leads to bone deterioration and osteoporosis. It is also associated with sarcopenia, the loss of muscle mass and strength, which increases the likelihood of falls and fractures.¹⁰ Women with low BMI are particularly vulnerable: research has found that the change in osteoporosis risk associated with a single BMI unit (roughly five to eight pounds) is greater than most other modifiable risk factors for the disease.¹¹

Among men, the picture is similarly stark. Those with a BMI below 18.5 have about five times the odds of osteoporosis compared to men at normal weight, while underweight women face about three and a half times the odds.⁹ Beyond fracture risk, being underweight is also linked to weakened immune function, fertility problems, and a higher mortality rate in the large cohort studies discussed earlier. When researchers look at all-cause mortality, a BMI at the bottom of the scale is often as dangerous as a BMI well into the obese range.

Cancer Risk and Obesity

Obesity increases the risk of several common cancers. An estimated 4 to 8% of all cancers are attributed to obesity, with the strongest links to post-menopausal breast cancer, colorectal cancer, endometrial cancer, kidney cancer, esophageal cancer, pancreatic cancer, liver cancer, and gallbladder cancer.¹² The mechanisms vary by cancer type but generally involve chronic inflammation, elevated hormone levels, and insulin resistance that create an environment more favorable to tumor growth. This is one of the less-discussed consequences of obesity, partly because the lag time between weight gain and cancer diagnosis can stretch over decades.

The Obesity Paradox

Here is where things get uncomfortable for anyone looking for a clean answer. In certain populations with established chronic diseases, a higher BMI appears to be associated with better survival, not worse. This has been termed the “obesity paradox,” and it shows up most prominently in heart failure research. In outpatients with established heart failure, higher BMIs were associated with lower mortality risks: overweight and obese patients had lower death rates than those at a normal weight.¹³ Mild to moderate obesity in chronic heart failure patients has been associated with substantially improved survival compared to normal-weight patients.¹⁴

But this finding is contested. When researchers carefully matched patients for indicators of disease severity, the apparent advantage of higher BMI disappeared. BMI no longer predicted survival as either a continuous or categorical variable once disease severity and other confounders were accounted for.¹⁵ The most likely explanation is that normal-weight patients with heart failure tend to be sicker in ways that BMI doesn’t capture: they may have more advanced disease, less muscle mass, or other wasting conditions. The paradox is probably less about excess weight protecting you and more about BMI being a crude measure that misreads what is happening in seriously ill people.

Why BMI Misclassifies So Many People

BMI tells you nothing about what your body is made of. It cannot distinguish between muscle and fat, and it says nothing about where your fat is stored. A well-trained athlete with low body fat can easily be classified as overweight by BMI standards.¹⁶ Conversely, a sedentary person with very little muscle mass can fall into the “normal weight” category while carrying a metabolically dangerous amount of fat around their organs.

This is why fat distribution matters enormously. Fat stored around the abdominal organs, called visceral fat, is far more strongly associated with type 2 diabetes and coronary artery disease than fat stored in the hips and thighs. That association holds even after controlling for age and total body fat.¹⁷ Data from the Framingham Heart Study confirmed that visceral fat remains more strongly linked to an unfavorable metabolic profile than other fat depots, even after accounting for standard body-size measurements like BMI.¹⁸ Someone with a normal BMI but a thick layer of visceral fat can be at serious metabolic risk. Researchers have described this phenotype as “metabolically unhealthy normal weight,” and although the long-term outcomes are still being studied, the concept highlights how misleading a normal BMI can be for some people.¹⁹

Ethnicity Changes the Risk Thresholds

The standard BMI categories were developed primarily from data on white European populations, and they don’t apply uniformly across ethnic groups. South Asian, Chinese, and Arab populations tend to develop metabolic disease at lower BMIs than white populations do. A large study of adults in England found that the BMI at which South Asian individuals develop type 2 diabetes at the same rate as white people at a BMI of 30 is only about 24. For Chinese populations, the equivalent threshold is roughly 27, and for Arab populations it is about the same.²⁰ Black populations, by contrast, had a threshold closer to 28, suggesting a somewhat different risk profile.

This discrepancy starts early in life. Research in children and adolescents found that Asian youth develop comparable levels of truncal fat, abnormal blood sugar, and unfavorable lipid profiles at BMI percentiles roughly 5 to 9 points lower than their non-Asian peers.²¹ Several organizations now recommend lower BMI cutoffs for obesity in Asian adults, using thresholds of 25 or 27.5 instead of the standard 30.²² If you are of South Asian or East Asian descent, a BMI that looks reassuringly “normal” by standard categories may actually place you in a higher risk zone than you think.

The BMI Sweet Spot Shifts With Age

In older adults, the relationship between BMI and health looks meaningfully different from what holds in middle age. A meta-analysis of people aged 65 and older found that the lowest mortality risk sat between a BMI of 24 and 31, with the very lowest point at about 27 to 28. People in this age group with a BMI between 21 and 22 had about a 12% higher mortality risk than those around 23 to 24, and those at a BMI of 20 or below faced at least a 28% higher mortality risk.²³ A systematic review similarly concluded that the optimal BMI range for the lowest mortality in elderly people falls in the overweight or mildly obese range, suggesting that standard guidelines about overweight being a major risk factor don’t apply cleanly to this population.²⁴

The reasons for this are partly biological and partly statistical. Older adults lose muscle mass naturally, so a person who drops into the “normal” BMI range in their seventies may have done so by losing protective lean tissue rather than harmful fat. Extra weight may also provide metabolic reserves during illness or surgery, and it likely offers some protection against the fracture risk that makes hip breaks so deadly in older populations. One Turkish study found the optimal BMI cutoffs for favorable geriatric health outcomes to be around 27 to 28 for men and 31 to 32 for women.²⁵

BMI Categories Work Differently for Children

Children and teenagers use a completely different BMI classification system. Instead of fixed cutoffs, pediatric BMI is plotted against age- and sex-specific growth charts. Underweight is defined as below the 5th percentile, normal weight ranges from the 5th to just below the 85th, overweight is from the 85th to below the 95th, and obesity is at or above the 95th percentile. Severe obesity is defined as a BMI at or above 120% of the 95th percentile value.²⁶ This percentile approach exists because children’s body composition changes dramatically as they grow, so a fixed number would be meaningless without age context. It does mean there’s a bumpy transition point at age 20 when pediatric and adult classification systems swap, and a teenager who is classified as normal weight by percentile can cross into overweight by adult BMI standards overnight simply by aging into the adult system.

BMI in Pregnancy

Pre-pregnancy BMI has significant implications for both the mother and baby. Obese women face higher risks of miscarriage, gestational diabetes, preeclampsia, blood clots, cesarean delivery, and wound infections. Their babies are at greater risk of stillbirth, congenital anomalies, prematurity, and abnormally large birth weight.²⁷ The preeclampsia risk alone is striking: for each increase in BMI of 5 to 7 units, the risk of developing preeclampsia roughly doubles.²⁸

Being underweight before pregnancy carries its own hazards. Research has found that underweight women had about three times the odds of preterm birth compared to normal-weight women, and somewhat surprisingly, they also had nearly three times the odds of gestational diabetes. Overweight women, meanwhile, had about five and a half times the gestational diabetes risk and more than nine times the risk of gestational hypertension.²⁹ The fact that both ends of the BMI spectrum create problems in pregnancy echoes the U-shaped mortality curve seen in the general population.

Alternatives and Complements to BMI

Given its blind spots, you might wonder whether better measures exist. The most promising and simplest is the waist-to-height ratio: just divide your waist circumference by your height. The rule of thumb is to keep your waist measurement below half your height. Meta-analyses in adults have found the waist-to-height ratio to be slightly better than BMI at predicting cardiovascular and metabolic risk, and it has the advantage of not needing age- or sex-specific cutoffs.³⁰ A systematic review in people with diabetes found that BMI still showed the strongest overall relationship with estimated cardiovascular risk, with the waist-to-height ratio close behind.³¹

More precise tools like DEXA scans or MRI can directly measure body fat and its distribution, but they are expensive, time-consuming, and impractical for population-wide screening.³² BMI persists not because it is accurate for any individual person, but because it is free, fast, and reasonably informative across large groups. The best practical approach is probably to use BMI as a starting point and supplement it with a waist measurement, especially if your BMI falls in the borderline zones.

Weight Stigma and What the Number Doesn’t Tell You

There’s a real human cost to reducing health to a single number. Weight stigma, the social discrimination against people perceived as overweight, causes psychological, physical, and socioeconomic harm.³³ People who experience weight-based discrimination are more likely to avoid medical care, which ironically worsens the very health outcomes that BMI is supposed to help predict. Clinicians who rely too heavily on BMI can miss genuinely metabolically unhealthy patients at a normal weight while over-alarming healthy patients who carry more muscle or have different body proportions. BMI is a screening tool, not a diagnosis. It flags people who may benefit from further evaluation, including metabolic blood work, blood pressure checks, and an honest conversation about how their body actually functions. It was never meant to be the final word on anyone’s health.