11.2 Growth Charts & Z-Score Interpretation

Standards Versus References: WHO and CDC

The single most-tested chart-selection fact: use the WHO growth standards for children 0-24 months and the CDC growth references for ages 2-20 years. This is the joint recommendation of the CDC and the American Academy of Pediatrics (AAP).

The distinction between a standard and a reference is conceptual and examinable. The WHO Multicentre Growth Reference Study enrolled healthy, predominantly breastfed infants raised under optimal conditions in six countries, so the WHO charts describe how children SHOULD grow - a prescriptive standard. The CDC charts were built from US survey data collected in the 1960s-1990s, including many formula-fed infants, so they describe how US children DID grow - a descriptive reference. Because breastfed infants gain weight more slowly than formula-fed infants in the second half of infancy, plotting a breastfed baby on the CDC charts can falsely suggest faltering weight; the WHO charts avoid this misclassification.

Percentiles and Z-Scores

A percentile ranks a child among 100 peers: the 25th percentile means 25% of the reference population falls below. A z-score (standard deviation score) expresses how many standard deviations a value lies from the median: z = (value - median) / SD. Useful conversions: the 50th percentile = z 0; the 2.3rd percentile is about z -2; the 97.7th percentile is about z +2; z -3 is about the 0.1st percentile and z +3 about the 99.9th.

Z-scores are preferred for children at the extremes and for serial tracking, because percentiles compress at the tails - a child cannot be plotted below the 0.1st percentile, but a z-score of -3.5 quantifies exactly how far below the median the child is and whether the gap is narrowing with treatment. The AND/ASPEN malnutrition criteria (11.4) are written entirely in z-scores.

The Key Indices and What Each One Tells You

• Weight-for-age: reflects body mass relative to age; influenced by both linear growth and adiposity. Sensitive to acute and chronic undernutrition but cannot distinguish them.
• Length/height-for-age: reflects long-term, cumulative linear growth; a low value signals chronic undernutrition or chronic disease.
• Weight-for-length (under 2 years) / BMI-for-age (2 years and older): reflects weight relative to linear size - the index of wasting when low and overweight/obesity when high.
• Head circumference-for-age (under 2-3 years): screens for abnormal brain growth; microcephaly and macrocephaly cutoffs are commonly z below -2 and above +2.

Threshold Definitions to Memorize

Interpreting Trends

Single points are snapshots; growth velocity and trajectory tell the story. In the first 24 months many healthy infants cross percentile lines as they shift from their birth size (driven by the intrauterine environment) toward their genetic channel - this normal canalization appears as catch-up growth in small-for-gestational-age infants or catch-down growth in large babies, and usually settles by about 2 years. After age 2, children should track along a stable channel. Red flags include sustained deviation across two or more major percentile lines after age 2, a plateau in length, decelerating height velocity (roughly <5 cm/year from age 4 to puberty), or any downward crossing accompanied by inadequate intake or illness. Always interpret a suspicious point by first re-checking the measurement and the plot before diagnosing.

Worked Plotting Examples

Example 1: A 9-month-old breastfed girl, weight 7.2 kg, length 68 cm. Correct index set: WHO weight-for-age, length-for-age, weight-for-length, and head circumference-for-age. If weight-for-age plots at about the 9th percentile on the CDC chart but near the 25th on the WHO chart, the WHO value is the one to use - do not diagnose faltering from the CDC chart.

Example 2: A 6-year-old boy with BMI 14.0 kg/m^2, BMI-for-age z-score -2.3. Interpretation: BMI z below -2 = wasting by anthropometric criteria; on the AND/ASPEN framework this single data point supports at least moderate malnutrition (z -2 to -2.9). The chart tells you the severity band; the full assessment tells you the cause.

Limitations

Growth charts describe populations, not individual diagnoses - a z-score of -2.1 is a screening and classification signal, and diagnosis requires the complete nutrition assessment (intake, history, physical exam, labs, functional status). Charts assume accurate measurement and correct age and sex entry. Finally, standard charts do not apply cleanly to preterm infants, genetic syndromes, or non-ambulatory children - those situations need the corrected ages and condition-specific charts covered in 11.3.