By Daniel Diaz-Gil, MD· April 2026 · 18 min read
Growth monitoring is a fundamental component of pediatric care, serving as an indicator of overall health and nutritional status. Accurate growth assessment requires selecting the appropriate growth chart for the child's age, feeding method, and any underlying conditions. The 2026 AAP Clinical Practice Guideline on Faltering Weight has introduced standardized z-score-based criteria and recommends replacing the term "failure to thrive" with "faltering weight."
The WHO growth charts are growth standards that describe how healthy children should grow under optimal environmental and health conditions. The reference population consisted of infants from six countries who were exclusively breastfed for at least 4 months and predominantly breastfed through 12 months.
The CDC growth charts are a growth reference that describes how children in the United States actually grew during a specific time period (1963-1994). The reference population included both breastfed and formula-fed infants.
The practical difference is clinically significant:
Using the wrong reference can lead to unnecessary investigation for nonexistent pathology or, conversely, missed identification of true growth concerns.
The CDC, NIH, and AAP recommend:
When using WHO growth charts to screen for abnormal growth, the 2.3rd and 97.7th percentiles (±2 standard deviations) are recommended as thresholds, rather than the 5th and 95th percentiles.
A 2025 study found that the abrupt switch from WHO to CDC charts at age 2 years causes clinically important changes in z-scores in children with stable growth patterns. The mean BMI-for-age z-score dropped by 0.59 at the transition, and 28.3% of children showed a drop in BMI z-score greater than 1.0. The authors developed "gradual transition" charts that use a weighted average from ages 2 to 5 years, which may reduce overidentification of slow weight gain.
The 2026 AAP guideline recommends using z-scores rather than percentiles for growth assessment:
A z-score of 0 corresponds to the 50th percentile; a z-score of −2 corresponds to the 2.3rd percentile.
The 2026 AAP/NASPGHAN Clinical Practice Guideline replaces "failure to thrive" with "faltering weight" and provides standardized diagnostic criteria:
Faltering weight is defined as any of the following:
Not all children meeting these criteria have undernutrition. Children born large for gestational age commonly exhibit early declines in z-scores as they regress toward their genetic potential. This is physiologic, not pathologic. Similarly, children with genetic conditions may have different growth potentials.
The guideline recommends against routine diagnostic testing in the initial workup for faltering weight in children without specific signs, symptoms, or findings that would prompt focal evaluation.
Growth velocity, the rate of change over time, is often more informative than a single percentile measurement. Height deviating more than two major percentile lines represents abnormal growth compared with peers.
Growth velocity should be evaluated using measurements every 3 to 6 months in children older than 6 months; shorter intervals may be appropriate for younger children and specific clinical situations. Use the Growth Velocity calculator to track height velocity accurately.
A child growing consistently at 5 cm/year and tracking the 10th percentile likely represents normal genetic variation. By contrast, a child previously growing at 8 cm/year who decelerates to 3 cm/year represents a significant problem, even if the child remains at the 25th percentile. The deceleration in velocity is the critical red flag.
When weight velocity declines but height velocity is preserved, consider inadequate intake or malabsorption. When height velocity declines, consider systemic illness or endocrine disorders.
Preterm infants require age correction for accurate growth assessment. Corrected age accounts for the time the infant would have remained in utero:
Corrected age = Chronologic age − (40 weeks − Gestational age at birth)
A 2025 study of extremely and very preterm infants found that using chronological age resulted in misclassification of up to 72.9% as stunted and 89.8% as underweight at term-equivalent age. The study concluded that age correction is required for all growth measures through 36 months of corrected age for extremely (<28 weeks) and very preterm (28 to <32 weeks) infants.
The AAP recommends:
The INTERGROWTH-21st Preterm Postnatal Growth Standards provide an alternative approach, with curves that overlap with WHO Child Growth Standards by 64 weeks postmenstrual age (6 months corrected age) without requiring adjustment.
Children with genetic syndromes have inherent differences in growth potential. Plotting them on standard charts may incorrectly suggest pathology. Condition-specific growth charts are available for:
Down syndrome: The Down Syndrome Growing Up Study (DSGS) charts (2015) are based on 1,520 measurements from 637 U.S. participants and show marked improvements compared with older charts. A 2026 Italian study developed Mediterranean-specific charts, finding statistically significant differences from U.S. charts in weight and BMI distribution. Use the Down Syndrome Growth calculator.
Turner syndrome: Syndrome-specific charts account for the characteristic short stature and growth pattern. Use the Turner Syndrome Growth calculator.
Achondroplasia: The AAP 2020 health supervision guidelines include syndrome-specific charts for height, weight, and head circumference. Additional charts for body proportions (sitting height, arm span, leg length) are available.
Prader-Willi syndrome, Noonan syndrome, cri du chat syndrome: Condition-specific charts are available, though based on smaller data samples.
Using both syndrome-specific and standard charts in parallel can help identify unexpected growth problems superimposed on the underlying condition.
In the first 6-12 months of life, normal infants frequently cross percentiles as they regress toward their genetic growth potential. An infant born large due to maternal diabetes may track the 95th percentile at birth but settle to the 50th percentile by 6 months. This is physiologic.
Children born with anthropometric z-scores greater than 0 (>50th percentile) are much more likely to exhibit declines in z-scores during the first 2 years of life.
Findings that warrant concern include:
Sex-adjusted midparental height calculations should be performed to evaluate growth velocity and compare predicted adult height with genetic potential:
A child's predicted adult height (estimated by projecting the current growth curve percentile into adulthood) should be compared with midparental height to help distinguish normal variants from pathologic stature.
Bone age compared with chronologic age helps narrow the differential diagnosis in children with short or tall stature: