Health & Medicine · Obstetrics & Pediatrics
Growth Percentile Calculator
Calculates a child's height and weight percentile based on CDC/WHO growth standards for age and sex.
Calculator
Formula
This is the LMS method used by CDC/WHO growth charts. X is the measured value (height or weight). M is the median value for the child's age and sex (from LMS tables). L is the Box-Cox power transformation coefficient (corrects for skewness). S is the generalized coefficient of variation. \Phi is the standard normal cumulative distribution function. The result is multiplied by 100 to express percentile (0–100).
Source: Kuczmarski RJ et al. CDC Growth Charts: United States. Advance Data No. 314, National Center for Health Statistics, 2000. WHO Multicentre Growth Reference Study Group, 2006.
How it works
Growth charts are population-based reference tools that describe how children grow under normal conditions. The CDC 2000 growth charts are derived from nationally representative U.S. survey data and are the standard reference for children aged 0–20 years in the United States. The World Health Organization (WHO) charts, based on children raised in optimal conditions across six countries, are preferred for children under 2 years in many international and clinical settings. Both systems use the same underlying mathematical model.
The LMS method, developed by Cole and Green, models the distribution of measurements at each age using three parameters: L (the Box-Cox power, which normalizes the distribution), M (the median of the measurement at that age), and S (the coefficient of variation). For a measurement X, the Z-score is calculated as Z = [(X/M)^L − 1] / (L × S). When L approaches zero, the formula simplifies to Z = ln(X/M) / S. The Z-score is then converted to a percentile using the standard normal cumulative distribution function Φ(Z) × 100. A percentile of 50 corresponds to exactly the median (Z = 0).
In clinical practice, growth percentiles guide decisions about nutritional counseling, endocrinology referrals, and monitoring of chronic disease effects on development. A single measurement is less informative than serial measurements plotted over time — a consistent 25th percentile is typically normal, while crossing two major percentile lines (e.g., dropping from the 75th to the 25th) may warrant investigation. Z-scores are particularly useful for children at the extremes, as percentile differences near the tails (e.g., 0.1st vs. 1st percentile) are clinically meaningful but appear similar on charts.
Worked example
Example: 24-month-old male, height 90 cm, weight 13 kg
Step 1 — Identify LMS parameters for a 24-month male from CDC tables:
Height: L = 0.2315, M = 87.8 cm, S = 0.03030
Weight: L = 0.1040, M = 12.6 kg, S = 0.11271
Step 2 — Calculate height Z-score:
Z = [(90 / 87.8)^0.2315 − 1] / (0.2315 × 0.03030)
Z = [(1.02506)^0.2315 − 1] / 0.007014
Z = [1.00575 − 1] / 0.007014
Z ≈ 0.0575 / 0.007014 ≈ 0.82
Step 3 — Convert height Z-score to percentile:
Φ(0.82) ≈ 79.4th percentile — this child is taller than about 79% of 24-month-old boys.
Step 4 — Calculate weight Z-score:
Z = [(13 / 12.6)^0.1040 − 1] / (0.1040 × 0.11271)
Z = [(1.03175)^0.1040 − 1] / 0.01172
Z ≈ [1.00329 − 1] / 0.01172 ≈ 0.28
Step 5 — Convert weight Z-score to percentile:
Φ(0.28) ≈ 61.0th percentile — this child's weight is above average but within the normal range.
Clinical interpretation: Both measurements fall within the 5th–95th percentile range, indicating normal growth for this age and sex.
Limitations & notes
This calculator uses interpolated LMS values from published CDC tables and applies them to the nearest reference age point; minor rounding differences from official software (e.g., CDC's EpiInfo or WHO Anthro) may occur. The CDC charts are appropriate for children aged 0–20 years in the U.S.; WHO charts are generally preferred for children under 24 months internationally. The LMS model assumes the reference population is well-nourished and healthy — results should be interpreted in clinical context, not in isolation. Premature infants require corrected age adjustments. Children with skeletal dysplasias, chromosomal conditions (e.g., Down syndrome, Turner syndrome), or chronic disease may be better assessed against condition-specific growth references. This tool is for educational and screening purposes only and does not replace clinical evaluation by a qualified healthcare provider.
Frequently asked questions
What is a normal growth percentile range for children?
Most clinicians consider the 5th to 95th percentile range normal. However, context matters — a child consistently at the 3rd percentile may be following their genetic growth curve normally, while a child who drops from the 70th to the 20th percentile over a short period warrants investigation regardless of the absolute value.
What is the difference between a Z-score and a percentile?
A Z-score expresses how many standard deviations a measurement is from the mean of the reference population. A percentile tells you the percentage of the reference population that falls below that measurement. They convey the same information but Z-scores are more precise at the extremes — for example, Z = −3 corresponds to roughly the 0.13th percentile, a distinction lost on most chart scales.
Should I use CDC or WHO growth charts for my child?
The WHO charts (2006) are based on children raised in optimal conditions and are recommended for children under 2 years by the American Academy of Pediatrics and CDC for that age group. The CDC 2000 charts are the standard for U.S. children aged 2–20 years. The WHO charts tend to show breastfed infants as growing more slowly in early months compared to CDC references.
How often should growth percentiles be checked?
The American Academy of Pediatrics recommends well-child visits at 1, 2, 4, 6, 9, 12, 15, 18, and 24 months, and then annually from ages 2–21. Growth measurements should be recorded at each visit. The trend over multiple measurements is far more informative than any single data point.
What does it mean if a child's height and weight are at very different percentiles?
A significant discrepancy between height and weight percentiles — for example, height at the 50th percentile but weight at the 95th — may warrant calculation of BMI-for-age percentile and further nutritional assessment. Conversely, low weight relative to height can be a marker of acute malnutrition, while low height with appropriate weight may suggest chronic nutritional insufficiency or an endocrine disorder.
Last updated: 2025-01-15 · Formula verified against primary sources.