01Purpose and principlesWhat the assessment is for and the core concepts behind it.
Growth charts convert serial measurements into an interpretable trajectory. They do not diagnose disease. A child on the 2nd centile may be healthy and genetically short, while a child near the 50th centile who has crossed down repeatedly may be unwell. Interpret position, direction, velocity, proportionality, parental target, puberty and clinical context together.
Select the correct chart before plotting. RCPCH UK–WHO 0–4 years charts cover weight, length or height and head circumference in infants and preschool children and are suitable for moderately preterm infants. The 2–18 years chart covers school-age growth, BMI, puberty, adult-height prediction and mid-parental comparison. Very preterm infants and children needing close early monitoring may require the neonatal and infant close monitoring chart. A condition-specific chart can add context, but standard UK growth charts remain useful for recognising general health change.
Obtain weight on calibrated equipment. Weigh an infant without clothing or in a clean dry nappy according to local protocol, and an older child in light clothing without shoes. Record to the precision supported by the scale and note casts, oedema, lines or clothing that distort change. Use the same equipment and conditions where feasible, but never manufacture precision by copying a previous value.
Measure supine length before age 2 with two trained people when possible: the head touches the fixed headboard, body is straight, legs extended gently and feet at right angles against the movable board. From age 2, measure standing height without shoes using a wall-mounted stadiometer, heels together, body upright and head in the Frankfort plane. Poor knee extension, hair, footwear and looking up can add centimetres.
Measure occipitofrontal head circumference with a non-stretch tape around the most prominent occiput and just above the supraorbital ridges. NICE advises 3 measurements at the same appointment when abnormal head size is suspected and plotting the longest. Repeat a surprising measurement before interpretation. Head shape, parental head size, gestational age and development influence meaning.
Calculate exact age rather than rounding to completed months, particularly in infancy. For prematurity, plot at corrected age for the period specified on the RCPCH chart; failure to correct can create false developmental and growth delay. Plot birth data following the chart's gestational-age instructions. Record that correction was used so a later clinician does not correct twice.
To plot, locate age on the x-axis, move vertically to the exact measurement and make a small cross or dot. Label unusually close points with date if ambiguity is possible. Join serial points only in chronological order. A value between printed curves has an intermediate centile; do not snap it to the nearest line. Digital centiles and z-scores improve numerical precision but are only as reliable as the input and reference selected.
Centile spaces are the bands between adjacent printed centile curves, not arbitrary percentage differences. NICE recommends concern about a fall across at least 1 weight space when birth weight was below the 9th centile, at least 2 spaces from the 9th–91st, or at least 3 spaces above the 91st. Current weight below the 2nd centile is also a concern threshold. These thresholds trigger assessment; they do not prove undernutrition.
Interpret weight in relation to length or height. A low BMI can reflect a naturally small build or undernutrition; below the 0.4th centile makes undernutrition more likely and requires assessment. A high BMI centile screens for excess adiposity but is not a direct body-fat measurement. Oedema may conceal weight loss, and rapid weight change over days usually reflects fluid or measurement conditions rather than tissue growth.
Linear growth responds more slowly than weight. Weight commonly falters first when intake is inadequate; prolonged undernutrition can then reduce height velocity. Primary endocrine disorders may preserve or increase weight relative to height while linear growth slows. Skeletal dysplasia can produce disproportion. Calculate height velocity only from reliable measurements separated by enough time—usually several months—and consider the child's age and pubertal stage.
Mid-parental height estimates genetic potential. The 2–18 chart provides a comparator; a child's height more than 2 centile spaces below the mid-parental centile can suggest undernutrition or a primary growth disorder. Parental height must be measured rather than recalled where accuracy matters. Genetic potential is a range, and illness, puberty timing and parental pathology can alter interpretation.
Head circumference trajectory must be linked to neurology and development. A small stable head in a developmentally normal child with small-headed parents differs from progressive deceleration. Rapid upward crossing with vomiting, abnormal eyes, tense fontanelle or altered consciousness suggests raised intracranial pressure. NICE recommends paediatric referral when head circumference is below the 2nd centile and assessment when it shifts across 2 or more lines.
Plotting is followed by explanation and a plan. Show parents which measurements are concerning, distinguish normal variation from uncertainty and avoid blaming language. Record values in the Personal Child Health Record and clinical record, state the chosen chart and correction, and specify the next measurement interval, clinical actions and referral threshold. Excessively frequent weighing can magnify normal fluctuation and anxiety.
Key points
- Use the current UK–WHO chart for sex and age: the detailed 0–4 years chart for most preschool children and the 2–18 years chart for school-age children and young people.
- For very preterm infants or significant early health problems, use the RCPCH neonatal and infant close monitoring chart and follow its gestational-correction instructions.
- Record the exact date, calculate exact age, mark the age on the horizontal axis and the measured value on the vertical axis, then place one precise point rather than a broad mark.
- Correct age for prematurity according to the chart: generally to 1 year after birth at 32–36 weeks and to 2 years after birth below 32 weeks.
- Measure supine length before age 2 years with an infantometer and standing height from age 2 with a calibrated stadiometer; the methods are not directly interchangeable.
- Plot weight, length or height and head circumference separately; from age 2 years calculate and plot BMI when weight or linear growth is concerning.
- A centile is a rank relative to the reference population, not a target and not the percentage of expected growth.
- Trajectory is more informative than one point: most healthy children track within a channel after early catch-up or catch-down growth.
- NICE thresholds for faltering weight depend on birth centile: crossing 1, 2 or 3 spaces is concerning from birth weights below the 9th, between the 9th–91st or above the 91st respectively.
- Current weight below the 2nd centile is a NICE threshold for concern whatever the birth weight; BMI below the 0.4th centile suggests probable undernutrition.
- Compare height with biological parental heights using the chart's mid-parental comparator; normal familial short stature should remain compatible with genetic potential and normal velocity.
- Always verify technique, equipment, transcription, chart, age and gestational correction before labelling a child with abnormal growth.
02Indications, selection and cautionsWhen it is useful, when urgency changes and important limitations.
Serial measurements follow a broadly consistent channel with appropriate velocity, proportionality and development, supporting normal variation.
Weight crosses the NICE number of centile spaces for its birth centile or falls below the 2nd centile, prompting feeding and clinical assessment.
Height velocity slows or height moves downward relative to previous measurements and genetic target, suggesting chronic systemic, endocrine or skeletal disease.
Weight-for-height, sitting-height relationship, limb proportions or head size diverges, making a single height centile inadequate.
Head circumference is extreme or crosses lines and must be interpreted with gestation, parental size, development and neurological signs.
An isolated implausible jump without clinical change often reflects age, chart, equipment, posture or transcription error and requires repeat measurement.
03Method and interpretationA systematic approach to the test and its findings.
Read from the initial assessment onwards. Tests may run in parallel in urgent care; first-line, preferred, confirmatory, definitive and gold-standard labels appear only when the chapter explicitly states them.
- 01
First-line: repeat anthropometry with calibrated equipmentFirst stepFirst line - Why
- Confirm that an apparent centile change is real before investigating disease.
- Interpretation and limitations
- Repeat weight, length or height and head circumference as appropriate using standard technique; reconcile any major discrepancy immediately.
- 02
First-line: serial UK–WHO chart reviewFirst line - Why
- Assess position, centile-space crossing and trajectory across all available reliable measurements.
- Interpretation and limitations
- Apply the correct chart, exact age and prematurity correction; interpret the whole trajectory rather than one centile.
- 03
BMI centile from age 2 years - Why
- Relate weight to height when faltering growth, thinness or excess weight is suspected.
- Interpretation and limitations
- Below the 2nd centile may be small build or undernutrition; below the 0.4th suggests probable undernutrition requiring assessment.
- 04
Mid-parental height comparison - Why
- Compare linear growth with biological genetic potential.
- Interpretation and limitations
- A child more than 2 centile spaces below the parental comparator deserves assessment for undernutrition or a primary growth disorder.
- 05
Height velocity - Why
- Detect impaired linear growth before height becomes extremely short.
- Interpretation and limitations
- Calculate from accurate measurements months apart and interpret against age and puberty; short stature with normal velocity often reflects a normal variant.
- 06
Clinical, developmental and feeding assessment - Why
- Decide whether a plotted abnormality reflects intake, disease, development, social context or safeguarding risk.
- Interpretation and limitations
- Target laboratory or specialist testing to findings; NICE advises against indiscriminate investigation in an otherwise well child with simple weight faltering.
04Clinical next stepsHow the result changes management or prompts escalation.
01Routine plottingMeasure, plot and verifyFirst stepGrowth is recorded at a scheduled review or clinical encounter.+
- 1Select the correct RCPCH chart and determine exact and corrected age before measurement.
- 2Measure with calibrated equipment, plot a precise point and compare with every reliable previous value.
- 3Explain the pattern, record chart and technique, and define whether routine or earlier review is needed.
02Unexpected pointExclude measurement error firstA new value jumps or falls implausibly from the established trajectory.+
- 1Repeat the measurement using correct technique and check scale, units, date, age, chart and transcription.
- 2Review clothing, oedema, hydration, posture, casts and whether length and standing height were mixed.
- 3If confirmed, assess the child clinically and act on the trajectory rather than averaging the abnormal point away.
03Faltering weightApply NICE thresholds and assess feedingWeight crosses the relevant spaces, is below the 2nd centile or loss is clinically concerning.+
- 1Measure length or height, calculate BMI from age 2 and review birth, illness, development and social context.
- 2Take a detailed feeding history and arrange skilled observation; consider UTI and coeliac testing when appropriate.
- 3Agree nutrition support, monitoring interval and referral criteria, avoiding coercive feeding and excessive weighing.
04Poor linear growthCompare velocity and genetic targetHeight is low, crosses down or is discordant with parental height.+
- 1Confirm standing height or length and obtain reliable serial data and measured parental heights.
- 2Assess proportion, puberty, nutrition, chronic disease, endocrine and syndromic features.
- 3Refer when growth is unexplained, velocity is slow, the child is far below target or associated red flags are present.
05Head-centile changeLink growth to neurological statusHead circumference is below the 2nd centile, unusually large or crosses at least 2 centile lines.+
- 1Repeat 3 times with correct landmarks, plot the longest and check gestational correction and parental head size.
- 2Assess fontanelle, eyes, development, tone, seizures, vomiting and consciousness.
- 3Use urgent acute assessment for neurological danger; otherwise arrange paediatric review and serial follow-up as indicated.
05Risks, monitoring and follow-upComplications, safety checks and further assessment.
- Use the NICE interval appropriate to concern: usually no more often than daily under 1 month, weekly at 1–6 months, fortnightly at 6–12 months and monthly from 1 year for faltering weight.
- When linear growth is concerning, NICE advises measuring length or height no more often than every 3 months because shorter intervals magnify measurement error.
- At every review, plot rather than list values and document whether prematurity correction and the same measurement method were used.
- Monitor feeding effectiveness, intake, symptoms, development, urine and stool alongside the curve; numerical gain without clinical progress may mislead.
- After nutritional intervention, review weight change, linear growth, dietary variety and family experience and avoid excessive catch-up that creates disproportionate weight gain.
- Escalate if the child loses weight, develops symptoms, fails to respond to a clear plan or moves further from genetic height potential.
06Special situationsVariants, exceptions and circumstances that change the usual approach.
A centile is not a grade
The 2nd and 98th centiles describe population position; either can be healthy when trajectory and clinical context are appropriate.
Velocity can reveal disease early
A child may remain within the printed range while steadily crossing downward, so serial accurate height matters more than a threshold alone.
Weight and height sequence helps
Nutritional disease often affects weight before height, whereas endocrine growth failure can slow height while relative weight is preserved.
Correction prevents false delay
Plotting a very preterm infant at chronological rather than corrected age can make normal catch-up look pathological.
Digital precision still needs technique
An exact z-score from a wrong height, date or reference chart is confidently wrong rather than clinically superior.
Explain uncertainty explicitly
Parents need to know whether the plan is reassurance, confirmation by remeasurement or investigation of a confirmed trajectory change.
07Common pitfallsFrequent interpretation and management errors.
- 01
Do not interpret a single measurement without prior trajectory and clinical context.
- 02
Do not round age to the nearest month in an infant.
- 03
Do not forget gestational correction or apply it twice.
- 04
Do not measure standing height in a child who cannot fully extend or position correctly without recording the limitation.
- 05
Do not mix recumbent length and standing height as if the methods were identical.
- 06
Do not use recalled parental heights when a genetic comparison will determine referral.
- 07
Do not call crossing one printed line ‘one centile space’ without identifying the adjacent curves actually crossed.
- 08
Do not diagnose undernutrition from low BMI without assessment, although a BMI below the 0.4th centile is strongly concerning.
- 09
Do not overweigh a stable child and react to normal fluid variation.
- 10
Do not attribute an abnormal curve to family size when velocity, proportion, development or examination is abnormal.