Medical disclaimer. This article explains how growth charts are built and read. It is general information, not medical advice, and it cannot assess your baby. If you are worried about your baby's weight, feeding, alertness or nappies, contact your pediatrician, health visitor or midwife — the same day if your baby seems unwell.
Want the number for your own baby? The free baby percentile calculator scores weight, length or head circumference against WHO's tables for your baby's exact age in days, and lists WHO's month-by-month values for boys and girls. This article explains how to read what it tells you.
Baby growth chart percentiles rank your baby against other babies of the same age and sex. That is all they do. The 30th percentile means roughly 30 healthy babies out of 100 are smaller and 70 are bigger. It is not a score, not a grade, and not a forecast of how tall your child will be. Half of all perfectly healthy babies sit below the 50th line, because that is how a middle works.
What the number is useful for is comparison over time. One dot on a chart is a snapshot. The line those dots make is the actual clinical information.
Where the curves come from
The curves on your baby's chart are not averages someone eyeballed. They come from the WHO Multicentre Growth Reference Study, which collected growth data from approximately 8,500 children in Brazil, Ghana, India, Norway, Oman and the USA between 1997 and 2003. WHO's stated aim was a single international standard describing physiological growth, and "to establish the breastfed infant as the normative model for growth and development."
The word standard is doing the work there. The CDC's 2010 growth-chart recommendation draws the distinction plainly: "Whereas the WHO charts are growth standards, describing the growth of healthy children in optimal conditions, the CDC charts are a growth reference, describing how certain children grew in a particular place and time." The same document records that "100% of the reference population of infants were breastfed for 12 months and were predominantly breastfed for at least 4 months," and lists maternal smoking during pregnancy or lactation among the exclusion criteria.
WHO publishes separate tables for boys and girls, and no combined table. Sex is not a nicety on a growth chart; it changes the curve.
WHO under two, CDC after two
If you are in the US, your baby's first two years are plotted on WHO curves and the years after that on CDC curves. That split comes from that same recommendation, which advises clinicians to "use the 2006 WHO international growth charts, rather than the CDC growth charts, for children aged <24 months," while "the CDC growth charts should continue to be used for the assessment of growth in persons aged 2–19 years."
The reasoning matters for parents. The CDC's own charts are a reference — a record of how a particular group of children grew. And, in the CDC's words, "healthy breastfed infants typically gain weight faster than formula-fed infants in the first few months of life but then gain weight more slowly for the remainder of infancy." Read against a reference chart, that normal slowdown can look like faltering growth; the CDC warns that clinicians unfamiliar with the breastfed growth pattern "might inappropriately recommend that mothers supplement breastfeeding with formula." Read against the WHO standard, it looks like exactly what it is.
In the UK, the UK-WHO charts published by the RCPCH are likewise based on the WHO Child Growth Standards, and are described as suitable for moderately preterm infants born at 32 to 36 weeks.
What "off the chart" actually means
On WHO charts, the flag is 2 standard deviations either side of the median. The CDC explains that these cutoffs "correspond to the 2.3rd and 97.7th percentiles" and are printed on the WHO charts as the 2nd and the 98th percentile. A measurement below the 2nd or above the 98th is the point at which a clinician looks harder. It is a prompt to investigate, not a verdict.
Note what that implies: a baby at the 4th percentile is inside the normal range. So is a baby at the 96th. Roughly one healthy baby in fifty will land below the 2nd line simply because the line has to have someone under it.
The curve beats any single point
The CDC is explicit that "growth monitoring is based on a series of accurate measurements over time," and that the series "accounts for both short- and long-term conditions and provides context for interpreting an individual measurement."
The CDC also puts weight on technique: "accurate measurements and recording of information is critical for growth charts to be an effective screening tool." The three measurements are not equally easy to take:
- Weight — a baby on a calibrated scale, with the same clothing off each time.
- Length — under two it is measured lying down (recumbent length), which needs the baby held straight and both ends read. A squirming newborn is the obvious difficulty.
- Head circumference — this one depends on the tape sitting at the widest point. A tape that slips produces a jump that looks alarming and is not real.
Which is why the CDC's instruction is that "health care providers should interpret any changes in a child's percentile classification with caution." A single "she dropped from the 50th to the 25th" reading is a reason to measure again, not a verdict.
The NHS frames the same idea in plain terms: a baby's weight and height may not follow a centile line exactly and "may go up or down by 1 centile line, but it's less common for them to cross 2 centile lines" — and if that happens, talk to a health visitor. The NHS also notes that most babies are at, or above, their birthweight by 3 weeks.
Three measurements tell three different stories
The WHO charts the CDC recommends under two carry four indicators, not one: weight-for-age, length-for-age, weight-for-length and head-circumference-for-age. They answer different questions, and the interesting information often sits in the relationship between them.
A baby at the 15th percentile for both weight and length is small and proportionate — the two measurements agree with each other. A baby at the 85th for length and the 10th for weight is a different picture, and weight-for-length is the chart that makes that difference visible. Head circumference is followed on its own curve, because it can move independently of the other two. Which pattern a particular baby is showing, and what it means, is a question for the person doing the measuring.
Corrected age: plotting a preterm baby honestly
A baby born at 32 weeks was eight weeks early — the AAP defines weeks preterm as gestational age at birth subtracted from 40. Plot her at 16 weeks of calendar age against other 16-week-olds and she will look tiny, because you are comparing her with babies who had eight more weeks of gestation.
The fix is corrected age. The AAP puts it simply: "Begin with your baby's actual age in weeks (number of weeks since the date of birth) and then subtract the number of weeks your baby was preterm." At 16 weeks old, our 32-weeker has a corrected age of 8 weeks, and 8 weeks is the age she is compared against. The AAP says that during the first 2 years, using corrected age gives a better idea of when a baby should reach common developmental goals; how long a chart keeps correcting is a question for your clinician, so ask at the next visit.
This comes up often for twins and other multiples: the NHS defines premature labour as labour before the 37th week and notes that "twins and triplets are often born prematurely." Two babies on two curves is normal — both landing on the same curve was never the goal.
When to ask, and when to wait
Worth raising at the next visit:
- Crossing two centile lines in either direction — the NHS says one line is common, two much less so, and to talk to a health visitor if it happens
- A measurement below the 2nd or above the 98th percentile, the two cutoffs printed on the WHO charts
- Birthweight not regained by 3 weeks, or a large weight loss — the NHS says a midwife or health visitor will support you in either case
- Any percentile change you want explained. The NHS weighs a baby more often than the standard schedule "if you ask for it or if there are concerns about their health or growth"
Worth contacting a clinician today, regardless of any percentile — the NHS lists these among the signs of dehydration in a baby that need urgent advice:
- Fewer wet nappies than usual
- A soft spot on the head that sinks inwards (a sunken fontanelle)
- Few or no tears when they cry
- Being unusually drowsy or irritable
The same goes for a baby who is refusing feeds or who simply seems unwell to you. Percentiles are a background instrument; behaviour is the foreground one. A thriving baby on the 5th percentile is doing better than a listless baby on the 60th.
How ProbaBaby helps
ProbaBaby's baby tracking module records weight, length and head circumference, and scores each one against WHO's own per-day L/M/S tables — generated from WHO's anthro reference data, covering days 0 to 730 for both sexes, with WHO's LMS transform and its documented tail adjustment for weight. Charts draw the P3, P15, P50, P85 and P97 curves, and lookups run on age in days, so a measurement taken on day 23 is scored against day 23 rather than rounded to a month. Because WHO publishes no combined table, the app requires the baby's sex rather than guessing one. For a baby born preterm, growth percentiles are looked up on corrected age until 24 months once gestational weeks at birth are recorded. When it is time to bring all of this to an appointment, the pediatrician visit summary exports as a PDF and shares by QR code, pulling growth, milestones, notes and mom wellbeing into one document.