BMI Percentile Calculator (Ages 2-20)

Why children use percentiles instead of fixed BMI cut-offs

For adults aged 20 and over, a single set of BMI thresholds applies regardless of age. Children and teenagers are still growing, and the normal amount of body fat changes considerably with age and differs between boys and girls. A BMI of 18 in a 4-year-old would signal very different health status than the same BMI in a 14-year-old. The CDC solved this by producing BMI-for-age percentile charts based on a nationally representative sample, showing how any child's BMI compares to peers of the same age and sex. A percentile tells you what fraction of the reference population had a lower BMI: a child at the 60th percentile is heavier than 60 percent of same-age, same-sex children in the reference group.

The CDC LMS method and what it means

The CDC growth charts use the LMS (Lambda-Mu-Sigma) statistical method devised by Cole and Green. Three parameters describe the distribution of BMI at each exact age and sex: L (lambda) corrects for skewness using a Box-Cox power transformation, M (mu) is the median BMI at that age and sex, and S (sigma) is the coefficient of variation (a normalized measure of spread). The z-score formula is z = [(BMI / M)^L - 1] / (L x S). Once the z-score is known, it converts to a percentile via the standard normal cumulative distribution function. This approach handles the fact that BMI distributions are not perfectly symmetric, especially at the tails, so a z-score of +1.645 corresponds precisely to the 95th percentile rather than being an approximation.

Above the 95th percentile: the CDC extended charts

The LMS model was fitted to data collected between 1963 and 1994 and was never intended to describe the extreme upper tail. Extrapolating it past roughly the 97th percentile compresses z-scores, so two children with very different BMIs can end up with nearly the same number. In December 2022 the CDC released the extended BMI-for-age growth charts to fix this. Data from youth at or above the 95th percentile in national surveys were modelled as a half-normal distribution with a sex- and age-specific scale parameter, sigma. This calculator follows the CDC method exactly: at or below the 95th percentile it uses the original LMS parameters, and above it the percentile is 100 x [1 - 0.05 x 2 x (1 - Phi((BMI - P95) / sigma))], where P95 is the 95th percentile BMI for that age and sex. That is why a child well into the obesity range gets a meaningful percentile and z-score instead of a value pinned near 99.9.

How to interpret the results

The CDC defines four weight-status categories based on percentile ranges: below the 5th percentile is underweight, 5th to below 85th is healthy weight, 85th to below 95th is overweight, and 95th or above is obesity. A further category of severe obesity applies when a child's BMI is at or above 120 percent of the 95th percentile for their age and sex, or at or above 35 kg/m^2, whichever is lower. These categories are starting points for clinical conversation. A single measurement is less meaningful than the trend across multiple well-child visits; a child consistently in the healthy range who starts crossing percentile lines upward deserves attention even if they remain below the 85th percentile cut-off.

Limitations and when to seek clinical advice

BMI-for-age does not distinguish muscle from fat. A muscular athlete may have a high percentile without excess fat, and an older child who has lost muscle through illness may appear in the normal range despite elevated body fat. The CDC reference charts were developed from data collected between 1963 and 1994, and population BMI distributions have shifted since then. Height and weight measurement errors, especially in shoes or heavy clothes, can shift results noticeably. For children close to a category boundary, clinical judgment about growth trajectory, dietary patterns, physical activity, family history, and additional measurements such as waist circumference is far more informative than the number alone. Always share these results with a pediatrician or family physician rather than acting on them independently.