What BMI Is Considered Skinny for Men and Women?

A BMI below 18.5 is classified as “underweight” by the World Health Organization, and that threshold applies equally to men and women. In everyday language, this is the number most people mean when they ask what counts as “skinny.” But that single cutoff hides a lot. The same BMI reflects very different amounts of body fat depending on whether you are male or female, how old you are, and your ethnic background, which means 18.5 is a useful starting point but a poor finishing line for understanding what a low weight actually means for your body.

The Standard Underweight Cutoff

BMI is calculated by dividing your weight in kilograms by the square of your height in meters. The WHO brackets are the same for both sexes: below 18.5 is underweight, 18.5 to 24.9 is normal weight, 25 to 29.9 is overweight, and 30 or above is obese. Within the underweight range there are further subdivisions. A BMI of 17 to 18.49 is considered “mild” thinness, 16 to 16.99 is “moderate,” and anything below 16 is “severe.” Most clinical guidelines use 18.5 as the cutoff when screening for malnutrition risk in adults, regardless of sex.

So if you are a 5’10” man weighing about 129 pounds, or a 5’5″ woman weighing about 111 pounds, you are right at the underweight line. Drop a few pounds below those figures and you would be classified as underweight. These are rough landmarks; the formula is continuous, so there is no cliff edge at 18.5 where everything suddenly changes.

Why the Same Number Means Different Things for Men and Women

Even though the official cutoff is identical, a BMI of, say, 19 does not represent the same body composition in a man as in a woman. At any given BMI, women carry significantly more total body fat than men across the entire adult lifespan. That difference holds after accounting for age and persists across ethnic groups.¹ Women also tend to store more of their fat under the skin rather than around internal organs, while men at the same BMI carry proportionally more visceral fat.²

In practical terms, this means a man at a BMI of 18.5 is likely leaner in terms of body fat percentage than a woman at the same BMI. A woman with a BMI of 19 might have a perfectly healthy amount of body fat, while a man at 19 could also be fine but is closer to the functional floor where low fat reserves start to matter. The universal cutoff of 18.5 does not capture this nuance. Some researchers have argued for sex-specific thresholds, but no major health agency has adopted them so far.

Health Risks of Being Underweight

Being below a BMI of 18.5 is not simply a cosmetic category. A large prospective study following over 48,000 men and women found that underweight men had roughly two and a half times the risk of dying from any cause compared to those in the normal-weight range.³ Interestingly, that particular study did not find a statistically significant increase in mortality for underweight women, though other research does show elevated risk for both sexes. A study of elderly men and women in Norway found that for every 2.5-unit drop in BMI below 25, mortality rose by about 20 percent, with no significant difference between men and women.⁴ A large Korean cohort confirmed a J-shaped pattern: the lowest death rates were among people with a BMI between 23 and 25, and risk climbed steeply at lower values.⁵

Beyond mortality, underweight status is linked to a U-shaped pattern of infection risk. Both underweight and obese adults tend to get infections more often than those in the middle of the BMI range.⁶ A Danish study tracking women over 12 years found that those with a BMI below 18.5 had roughly four times the hospitalization rate for upper respiratory infections and about three times the rate for fungal infections compared to normal-weight women.⁷ The immune system needs adequate energy and micronutrients to function, and chronic energy deficiency compromises that supply.

Bone and Reproductive Health at Low Body Weight

For women, one of the most clinically important consequences of being too thin is the loss of menstrual periods, a condition called functional hypothalamic amenorrhea. When the body senses it does not have enough energy coming in, it dials down reproductive hormones. This happens in eating disorders, overtraining, and high-stress situations, and the common thread is relative energy deficiency.⁸ The consequences go beyond fertility. Low bone density is extremely common in these women, raising their risk of stress fractures now and fragility fractures later in life. Younger women who never reach their peak bone mass because of prolonged underweight face compounding skeletal consequences for decades.

Hormonal treatments like estrogen replacement have limited success at rebuilding that bone. The most effective strategy is increasing caloric intake enough to gain weight and restore menstrual cycles.⁹ This finding underscores that the problem is fundamentally about energy availability, not just hormone levels. Men at very low body weight face their own endocrine disruption: testosterone drops, which also accelerates bone loss and reduces muscle mass, though the research base here is thinner than for women.

Athletes who maintain low body weight are a particularly at-risk group. The syndrome known as Relative Energy Deficiency in Sport affects both sexes and can cause gastrointestinal problems, cardiovascular dysfunction, hormonal suppression, psychological disturbances, and impaired bone health, all of which ultimately hurt the athletic performance they were trying to optimize.¹⁰

When “Skinny” Is Just How You Are Built

Not everyone with a BMI below 18.5 is malnourished or at risk. Some people are constitutionally thin, meaning they have been naturally slender their entire lives without restricting food or exercising excessively. Research comparing constitutionally thin individuals with people who have anorexia nervosa reveals that these are profoundly different conditions despite overlapping BMI numbers. Constitutionally thin people tend to be underweight but not underfat. Their body fat percentage is closer to normal because they also have lower lean mass. Their hormones related to growth, reproduction, and stress are normal, whereas those hormones are disrupted in anorexia.¹¹

The appetite-signaling hormones also tell a different story. Constitutionally thin people have higher levels of a satiety hormone called peptide YY, which may help explain why they feel full easily. Their ghrelin, the “hunger hormone,” is actually lower than in people with normal weight, which is the opposite of what you see in anorexia.¹² Their resting metabolic rate, when adjusted for lean mass, tends to be similar to normal-weight controls and higher than in anorexia, suggesting their metabolism runs at a normal pace rather than being suppressed by starvation.¹³

The heritability data reinforces this. Persistent healthy thinness has a heritability of about 28 percent, meaning genes account for a meaningful share of why some people stay thin throughout life. That figure is comparable to the heritability of severe obesity, which sits around 32 percent.¹⁴ If you have always been slender, your periods are regular (if applicable), your energy levels are good, and you eat normally, a BMI slightly below 18.5 does not necessarily signal a health problem.

The BMI Cutoff Changes with Age

The standard 18.5 threshold was developed primarily from data on younger and middle-aged adults. For older people, the picture shifts. Body composition changes with aging: you lose muscle and bone mass and tend to gain fat, so a given BMI number in a 70-year-old represents a very different mix of tissue than the same number in a 30-year-old. Research in elderly populations consistently finds that being slightly “overweight” by standard BMI criteria is associated with better survival than being in the normal range.

A study of older adults in India proposed age-specific cutoffs that look dramatically different from the WHO categories. For people aged 60 to 74, the underweight threshold dropped to 17.4 or below, and the upper end of “normal” stretched all the way to 28.8. For those 75 and older, the underweight cutoff was even lower at 13.3, with “normal” reaching up to 22.8.¹⁵ These cutoffs were better at predicting cardiometabolic outcomes than the standard WHO values for those age groups. A separate review of geriatric assessment data found that the worst outcomes in older adults appeared at BMIs below 25, suggesting the “normal” floor for older people should be substantially higher than 18.5.¹⁶

The upshot: if you are over 65 and your BMI is 20, that would be “normal” by the standard chart but may actually put you in a higher-risk category. Older adults who lose weight unintentionally should take it seriously, even if they remain technically above 18.5.

Ethnicity Matters More Than Most People Realize

The 18.5-to-25 “normal” range was established largely using data from white European populations. For many Asian populations, the relationship between BMI and health risk is shifted to the left, meaning disease shows up at lower BMI values. A WHO expert consultation concluded that substantial health risks for diabetes and cardiovascular disease appear at BMIs well below 25 in Asian populations, though the exact cutoff varies between groups, ranging from about 22 to 25 depending on the specific population studied.¹⁷

This has implications on the “skinny” end too. If the overweight threshold for some Asian populations is lower, then the healthy range narrows, and what looks like a perfectly normal BMI of 24 in a European might already carry elevated metabolic risk in a South Asian or Filipino individual. U.S. data confirmed this: Vietnamese, Korean, Filipino, and South Asian Americans had significantly higher diabetes rates at BMIs of 23 to 25 compared to non-Hispanic white Americans at the same BMI.¹⁸ When interpreting your own BMI, your ethnic background is a relevant variable that the standard chart ignores.

Why BMI Can Be Misleading at Any Weight

BMI tells you about the ratio of weight to height. It says nothing about what that weight is made of. A person with substantial muscle mass can have a BMI in the “normal” or even “overweight” range while having very low body fat. Conversely, someone with little muscle and a modest amount of fat can land at a “normal” BMI while actually carrying an unhealthy proportion of body fat. Using BMI as the sole measure of body composition lumps together people with very different amounts of fat and very different risks of death.¹⁹

This limitation cuts both ways. On the low end, a man with a BMI of 19 who has good muscle tone and low visceral fat is in a very different physiological state than a man at the same BMI who has lost muscle through illness or inactivity. On the high end, the phenomenon of “normal-weight obesity” describes people whose BMI falls in the healthy range but whose body fat is excessive. These individuals face elevated risks for metabolic problems including high blood sugar, high blood pressure, low HDL cholesterol, and cardiovascular disease.²⁰ A BMI in the “normal” bracket does not guarantee metabolic health any more than a BMI below 18.5 guarantees illness.

The Psychological Side of Low Body Weight

Sustained underweight status does not just affect the body. A scoping review of starvation’s effects in humans found that adults with low body weight commonly experience depression, anxiety, reduced cognitive functioning, impaired sleep, social withdrawal, and diminished quality of life.²¹ These are not exclusive to people with eating disorders; chronic energy deficit from any cause can produce them. In children, cognitive impairments were the most frequently reported effect.

These psychological consequences create a feedback loop. Depression and anxiety reduce appetite, which worsens energy deficit, which deepens the psychological effects. For someone trying to gain weight, simply being told to “eat more” may not be enough. Addressing the mental health dimension is often necessary before sustained weight gain becomes possible.

Social Perceptions of Thinness

How other people interpret a low BMI depends heavily on context. Research has shown that people judge underweight women differently depending on the perceived reason for their thinness. When told an underweight woman’s low weight was due to an eating disorder, study participants rated her as more personally responsible for her condition and ascribed more negative characteristics to her than when the same weight was attributed to illness or heredity. The study used over 500 participants viewing photographs of the same objectively underweight woman with different backstories, and the attribution reliably shifted both blame and character judgments.

This matters because the social experience of being “skinny” is not uniform. A man or woman who is naturally thin may face unsolicited comments, concern, or assumptions about disordered eating. These social pressures can themselves become a source of stress, particularly for constitutionally thin individuals who eat normally and are otherwise healthy. The cultural fixation on BMI as a single number can paradoxically create anxiety in people who fall outside the “normal” range on either end, even when their actual health is fine.

Gaining Weight When You Need To

For people who are medically underweight and want to gain, the approach depends on the underlying cause. If the issue is insufficient caloric intake relative to expenditure, the solution is straightforward in principle: eat more, with an emphasis on calorie-dense, nutrient-rich foods. Resistance training helps ensure that added weight includes muscle rather than just fat. A pilot study of bodybuilders found that combining resistance training with a caloric surplus produced greater gains in muscle mass than resistance training with more moderate intake, though the surplus group also gained more body fat.²² For non-athletes, even modest resistance exercise combined with higher protein and calorie intake can shift body composition in a healthier direction.

For women with functional hypothalamic amenorrhea, weight gain is not optional if they want to protect their bones. As noted earlier, hormonal therapy alone has limited effectiveness; the evidence points clearly toward increased caloric intake leading to weight gain and menstrual resumption as the primary strategy.⁹ If you are underweight because of an underlying medical condition like hyperthyroidism, celiac disease, or chronic infection, treating the root cause is the first step. Piling on calories without addressing why you are losing weight in the first place rarely works well.

One thing worth flagging: the goal is not necessarily to reach a BMI of 22 or some other midpoint number. For constitutionally thin people, a BMI of 17.5 or 18 may represent their body’s natural settling point, and forcing it higher through chronic overeating is neither comfortable nor necessarily beneficial. A doctor who evaluates blood work, bone density, hormonal status, and overall function can help determine whether a low BMI is a problem to solve or simply how you are built.

Where the 18.5 Number Came From

The modern term “body mass index” was coined in 1972 by the physiologist Ancel Keys, who built on the work of the Belgian statistician Adolphe Quetelet. Quetelet developed the original index in the nineteenth century as part of his effort to describe the physical characteristics of an “average man” using statistical analysis. Keys analyzed samples of several thousand men and provided evidence to support using the weight-over-height-squared formula as a rough index of body fatness for population-level studies. The scale was never designed as a diagnostic tool for individuals; it was a blunt population-level instrument that became far more widely applied than its creators intended. Insurance companies had already been using weight-for-height tables since the 1950s to set premiums, and BMI eventually replaced those tables as a simpler standard. Understanding this origin helps explain why the cutoffs feel arbitrary: they are statistical conventions, not biological constants, and the boundaries between “normal” and “underweight” are drawn where increased risk becomes statistically detectable in large populations, not where your individual body switches from healthy to unhealthy.