Growth & Height

Z-score (Standard Deviation Score, SDS)

A z-score shows how many standard deviations a measurement is from the mean. It is the more precise, statistical sibling of the percentile, preferred at the extremes.

5 min read

What is a z-score?

A z-score (standard deviation score, SDS) is a number saying how many “standard deviations” a child’s measurement is from the mean of the same-age, same-sex population. z = 0 is exactly average, positive values are above and negative below. For example, a child with a height z-score of +1 is one standard deviation taller than average, while −1.5 is markedly short.

A standard deviation is a statistical unit measuring how spread out values are in a population. Because the z-score uses this unit, it translates different measurements — height, weight, BMI — into a common, comparable language.

Difference from percentile

A z-score and a percentile carry the same information but serve different purposes. Percentiles are more intuitive for families (“25 of 100 children”); z-scores are statistically more precise. The difference is clearest at the extremes: both the 1st and the 0.1st percentile may look like “below the 3rd percentile”, while their z-scores (e.g. −2.5 vs −3.5) clearly show the serious difference between them.

Rough equivalents: z = 0 → 50th percentile; z = −2 → about the 2.3rd; z = +2 → about the 97.7th. So in clinical follow-up, especially for very short or very tall children, the z-score is preferred.

How is it interpreted?

Most healthy children fall between −2 and +2 SDS. Values outside this range do not mean “abnormal” but warrant a closer look: below −2 points toward short stature/low weight, above +2 toward tall stature/high weight. As with percentiles, it is not a single z-score that matters but its trend over time and its fit with parental heights.

Reading height and weight z-scores together is also valuable. Their consistent, stable course is reassuring; a clear gap opening between them may raise nutritional or hormonal questions.

Why z-scores are valuable in follow-up

The strongest use of the z-score is tracking change over time numerically. If a child’s height z-score drifts over months from, say, −1 to −2, that is a clear “downward” signal; the same change can be harder to catch via percentiles. So the z-score is a practical tool for catching growth problems early and for following treatment response.

Because z-scores put different measurements on a common scale, they make it easy to assess height, weight and BMI side by side. For example, a stable height z-score while weight z-score rises rapidly may indicate the weight-for-height balance is shifting. Still, the z-score is a screening and monitoring tool; it does not diagnose on its own and is always interpreted with growth velocity and the clinical picture.

Frequently asked questions

My child’s z-score is below −2 — does it mean illness?

No, not directly. −2 SDS is a statistical threshold; some children below it are perfectly healthy with familial short stature. What matters is assessing it together with growth velocity, parental height and the trend.

Why is the z-score sometimes used instead of percentiles?

At the extremes, percentiles lose discriminating power (e.g. the 1st and 0.1st look like “below 3rd” yet differ greatly). The z-score keeps that difference numerically, so it is more precise in clinical follow-up.

What does a change in z-score show?

A marked fall or rise in the z-score over time can indicate the child is leaving their own growth channel and warrants review. Small fluctuations, though, are normal.

Related terms

This glossary entry is for information only and is not medical advice. Consult your paediatrician or the relevant specialist for diagnosis and treatment.