Growth Z-Score Calculator
Calculate a child's growth z-score from age, sex, height, weight, or BMI, with WHO/CDC references, percentile gates, and preterm-age correction.{{ summary.title }}
| Metric | Value | Copy |
|---|---|---|
| {{ row.label }} | {{ row.value }} |
| {{ header }} | Copy |
|---|---|
| {{ cell }} |
| Priority | Recommendation | Rationale | Timeline | Copy |
|---|---|---|---|---|
| {{ row.priority }} | {{ row.recommendation }} | {{ row.rationale }} | {{ row.timeline }} |
Introduction:
Pediatric growth charting turns a child's measurement into a comparison with children of the same age and sex in a reference population. The raw number still matters, but its meaning changes with age. A 12 kg toddler, a 12 kg infant, and a 12 kg school-age child do not occupy the same place on a growth curve.
A growth z-score, also called a standard deviation score or SDS, reports that place as distance from the reference median. A value near 0 is close to the middle of the reference curve. Negative values sit below the median, positive values sit above it, and larger absolute values mean the measurement is farther from the middle for that age and sex.
Percentiles express the same curve position in rank form. The 75th percentile means the value is higher than the reference value for about 75 out of 100 children of the same age and sex. It is not a percentage of health, body fat, height potential, or nutrition status.
| Term | What it tells you | Common misread |
|---|---|---|
| Z-score or SDS | Distance from the reference median in standard deviation units. | It is sometimes mistaken for a diagnosis or a target. |
| Percentile | Rank on the same reference curve. | It is sometimes read as a health score or percentage of normal. |
| Reference family | The chart population and age range used for comparison. | Switching references can look like sudden growth change. |
| Trend | How repeated points move over time. | A single point is sometimes overread without the growth record. |
Reference choice is one of the biggest reasons two correct chart readings can differ. In United States practice, WHO standards are recommended from birth to 2 years, and CDC charts are recommended from age 2 through 20 years. The handoff at 24 months also coincides with a change from recumbent length to standing height, so a child's apparent percentile can move even when the child is growing normally.
Measurement technique matters as much as the formula. Recumbent length is usually longer than standing height, small age errors can move an infant to a different lookup row, and corrected age can be appropriate for some preterm infants during early growth monitoring. BMI-for-age adds another caution because pediatric BMI is read against age-and-sex curves, not adult BMI categories.
Growth z-scores are screening and review aids. They can highlight a point that deserves remeasurement, closer follow-up, or comparison with the child's earlier chart, but they do not diagnose disease, malnutrition, obesity, or healthy growth by themselves. Clinical decisions need the full growth history, measurement quality, family context, symptoms, and local clinical guidance.
How to Use This Tool:
Set the charting context first, then enter the measurement path that matches the growth metric.
- Choose
Sexand enterAgefrom0to240months. Decimal months are useful near month boundaries and the 24-month reference switch. - Select
Growth metric. UseLength or height for agefor linear growth,Weight for agefor body weight relative to age, orBMI for agewhen body mass index is the value being charted. - Enter the measurement in the visible unit controls. Height accepts
cmorin, weight acceptskgorlb, and BMI useskg/m^2. - For height, set
Measurement modetoAuto from age,Recumbent length, orStanding height. Auto uses length before 24 months and standing height from 24 months onward. - For BMI, choose
Derive from height and weightunless a known pediatric BMI is already available. Direct BMI skips the height and weight checks. - Leave
Reference setonAutofor the default WHO-before-24-months and CDC-from-24-months path, or force WHO or CDC when a charting workflow requires it. WHO lookup is limited to60months here, and CDC BMI starts at24months. - Turn on
Correct age for pretermonly for an eligible infant. The calculation subtracts prematurity weeks from lookup age during the first24months and records aCorrected-age adjustmentinSDS Snapshotwhen it applies. - Use
Advancedonly when you need custom SDS review bands.Watch thresholdaccepts1.0to3.0;Severe thresholdaccepts1.5to4.5and is kept above the watch value. - Read the top z-score brief and
SDS Snapshotfirst. OpenReference Gatesfor equivalent SD-lane measurements,Interpretation Notesfor follow-up cues, andSDS Lane Chartto see the current value against nearby median, watch, and severe lanes.
Interpreting Results:
Start with Z-score, Percentile, and Interpretation. The z-score gives distance from the median, the percentile gives rank, and the interpretation label applies the WHO or CDC rule used for that run.
Action band is a separate custom SDS check. It uses the selected watch and severe thresholds, so it can disagree with the WHO or CDC interpretation when those rules use different boundaries. A child can be in an obesity screening range by CDC BMI rules while still sitting inside the custom watch band if the z-score has not crossed the chosen watch threshold.
Use Reference Gates when a result is near a boundary. That table shows the measurement values at -3, -2, -1, median, +1, +2, +3, and the configured watch and severe lanes. Boundary checks are especially important near the 5th, 85th, 95th, and P95 markers.
A high or low result should not be overread from one visit. Recheck the age, sex, units, length-versus-height technique, BMI entry path, reference family, and preterm age setting before using the result for follow-up planning.
Technical Details:
Growth references use age- and sex-specific curves rather than one fixed adult-style threshold. For a selected metric, the reference row provides a median and curve shape for that age and sex. The measurement is standardized against that row, then converted into a z-score and percentile.
The LMS method is the main calculation. L describes the Box-Cox power used to handle skew, M is the median, and S is the coefficient of variation. This lets the same equation handle measurements whose spread changes with age.
Formula Core:
For a reference-aligned measurement X, the LMS equation calculates the SDS value z. When L is effectively zero, the logarithmic form is used.
BMI runs can either use a supplied BMI or derive it from height and weight before the reference lookup.
| Symbol or value | Meaning | Unit or display |
|---|---|---|
X |
Measurement after any length/height alignment or BMI derivation. | cm, in, kg, lb, or kg/m^2 display. |
L, M, S |
Reference curve values for the selected age, sex, metric, and chart family. | Lookup row values; M carries the measurement unit. |
z |
Standard deviation distance from the median. | Shown to two decimals. |
| Percentile | Standard-normal percentile equivalent of the z-score. | Shown to one decimal and bounded from 0.1 to 99.9. |
A simple median substitution gives X / M = 1, which makes the LMS numerator 0 and returns z = 0. A value above the median returns a positive z-score; a value below the median returns a negative z-score. The exact distance changes with age, sex, metric, reference family, and the local curve shape.
Reference and Adjustment Rules:
| Area | Rule used | Why it matters |
|---|---|---|
| Age range | Age accepts 0 to 240 months. |
Age chooses the reference row and can shift infant results quickly. |
| Auto reference | WHO is used below 24 months; CDC is used from 24 months onward. |
The 24-month change can alter both percentile and label. |
| Forced WHO | WHO lookup stops at 60 months. |
Older children need a CDC path for this calculation. |
| Forced CDC BMI | CDC BMI-for-age starts at 24 months. |
Under-2 BMI should be read cautiously and usually belongs on a WHO path. |
| Corrected age | Prematurity weeks are subtracted from lookup age only before or at 24 months. |
Comparisons across visits need the same age basis recorded. |
| Length/height technique | 0.7 cm is added or subtracted when recumbent length and standing height need alignment. |
A small technique difference can move a borderline stature result. |
Classification Rules:
| Path | Boundaries | Interpretation note |
|---|---|---|
| WHO height-for-age | z < -3 severe low, -3 ≤ z < -2 low, -2 ≤ z ≤ 3 expected, z > 3 tall. |
Uses SDS boundaries directly. |
| WHO weight-for-age | z < -3 severe underweight, -3 ≤ z < -2 underweight, -2 ≤ z ≤ 2 expected, 2 < z ≤ 3 high, z > 3 very high. |
Weight-for-age does not show body proportionality by itself. |
| WHO BMI-for-age | z < -3 severe wasting, -3 ≤ z < -2 wasting, -2 ≤ z ≤ 1 expected, 1 < z ≤ 2 risk of overweight, 2 < z ≤ 3 overweight, z > 3 obesity. |
Under-2 BMI is a pediatric growth reference, not an adult BMI category. |
| CDC height or weight | percentile < 5 low/short, 5 ≤ percentile < 95 typical, percentile ≥ 95 high/tall. |
CDC height and weight labels use percentile gates. |
| CDC BMI-for-age | percentile < 5 underweight, 5 to below 85 healthy weight, 85 to below 95 overweight, percentile ≥ 95 obesity, and BMI ≥ 120% of P95 severe obesity screening range. |
CDC BMI runs report P95 and 120% of P95. |
| Custom action band | |z| ≥ severe is priority, |z| ≥ watch is watch, and smaller absolute z-scores stay inside the watch band. |
These bands are user-set SDS review lanes, not a replacement for WHO or CDC labels. |
CDC public BMI guidance also uses BMI ≥ 35 kg/m^2 as an alternative severe-obesity criterion for children and adolescents. The CDC BMI label here follows the P95-based marker shown in Reference Gates, so very high adolescent BMI values should be reviewed with the full CDC charting guidance and clinical context.
Accuracy Notes:
Growth z-scores are informational estimates based on reference charts and the entered values. They are not a substitute for pediatric assessment, diagnosis, treatment decisions, or local clinical protocols.
- Repeat measurements when the result is near an SD lane, percentile gate, P95 marker, watch threshold, or severe threshold.
- Keep the same reference family, age-correction choice, and measurement technique when comparing repeat visits.
- Check unit conversions, especially inches to centimeters, pounds to kilograms, and BMI derived from height squared.
- Pair weight-for-age with length/height, BMI, clinical history, nutrition context, and the child's longitudinal chart before drawing conclusions.
- Use corrected age only when it fits the infant's history and the comparison method being used by the care team.
Worked Examples:
Routine height placement. A female child aged 36 months with standing height 95 cm and Reference set on Auto uses the CDC path. SDS Snapshot shows Z-score 0.27, Percentile 60.7 percentile, and Interpretation Typical stature-for-age range. Reference Gates gives a watch lane of about 86.18 cm to 101.96 cm, so the result sits inside the configured watch band.
Preterm infant weight check. A male infant entered at 10 chronological months, 7.5 kg, 32 weeks gestational age, and Correct age for preterm on has 8 prematurity weeks removed. SDS Snapshot records Age used for lookup as 8.16 months, Z-score -1.31, and Percentile 9.5 percentile. The label remains Expected weight-for-age range, but the low-side placement should be compared with the infant's prior measurements.
CDC BMI boundary. A male child aged 120 months with direct BMI 27 kg/m^2 uses CDC BMI-for-age on Auto. The result shows Z-score 2.23, Percentile 98.7 percentile, Percent of CDC 95th percentile 122.1%, and Interpretation Severe obesity screening range. Because the value is above 120% of P95, the P95 marker rather than the ordinary percentile alone drives that label.
Reference recovery. If a toddler under 24 months is set to CDC BMI, an error states that CDC BMI-for-age starts at 24 months. Switch Reference set back to Auto or WHO, then rerun the BMI path and check Reference set used before interpreting the output.
FAQ:
Is a z-score the same as a percentile?
No. The z-score gives distance from the reference median in standard deviation units. The percentile converts that same curve position into a rank.
Why can WHO and CDC results change at 24 months?
The reference family changes, standing height replaces recumbent length in many workflows, and some cutoff values differ. Keep the same reference when comparing visits unless your charting workflow requires the switch.
When should corrected age be used?
Use it only when preterm age correction is appropriate for the child. The calculation applies it before or at 24 chronological months by subtracting weeks short of 40 weeks from the lookup age.
Why does length versus height change the result?
Recumbent length is usually longer than standing height. When the entered technique and reference basis differ, a 0.7 cm alignment keeps the measurement on the right basis.
What should I do when a reference error appears?
Check the age and reference choice first. WHO is limited to 60 months here, CDC BMI starts at 24 months, and all measurement fields must be positive numbers.
Glossary:
- Z-score
- Distance from the reference median in standard deviation units.
- SDS
- Standard deviation score, another name for a growth z-score.
- LMS
- The reference method that uses a Box-Cox power, median, and spread value for each lookup row.
- Reference median
- The middle value on the selected age-, sex-, metric-, and reference-specific curve.
- Corrected age
- Age adjusted for prematurity before the reference lookup, used during the early correction window.
- P95
- The age- and sex-specific 95th percentile BMI value used in CDC BMI-for-age screening output.
- Action band
- The custom watch or severe SDS lane based on the selected absolute z-score thresholds.
References:
- WHO Child Growth Standards, World Health Organization.
- What Growth Charts Are Recommended?, CDC, March 14, 2025.
- Using WHO Growth Standard Charts, CDC, March 20, 2024.
- Summary and References: Using BMI-for-Age Growth Charts, CDC, January 30, 2025.
- Anthropometric Measurements, American Academy of Pediatrics, February 13, 2026.