What Are Baby Growth Charts Actually Based On?

Explanation

A growth chart looks like a fact about children. It is actually the output of one specific study, with a specific design, and knowing that design changes how much weight a single dot on it deserves.

The WHO Child Growth Standards were built from the WHO Multicentre Growth Reference Study (MGRS), carried out between 1997 and 2003. WHO had reviewed the growth reference in use since the late 1970s and concluded it did not adequately represent early childhood growth; the World Health Assembly endorsed building a new one in 1994.

WHO Child Growth Standards

Supporting evidence

Feature of the study	Detail
Data collection period	1997–2003
Children enrolled	8,440
Countries	Brazil, Ghana, India, Norway, Oman, USA
Feeding	Healthy breastfed infants and young children
Design	Longitudinal birth to 24 months, plus cross-sectional 18 to 71 months

The standards derived from it cover length/height-for-age, weight-for-age, weight-for-length/height, BMI-for-age, head circumference-for-age, arm circumference-for-age, subscapular and triceps skinfold-for-age, motor development milestones, and growth velocity for weight, length and head circumference.

What the numbers mean

Three design choices in that table do most of the work.

It is a standard, not a reference. The children were selected for conditions recommended for healthy growth. So the curves describe how children grow when circumstances are favourable, a prescriptive model rather than describing how a given population did in fact grow. That is the opposite of an average.

All the infants were breastfed. Deliberately: the earlier reference was built largely on formula-fed infants, which produced curves that made normal breastfed growth patterns look like faltering.

Six countries, on purpose. The study was designed around the finding that, given good conditions, early linear growth is remarkably similar across ethnic and geographic backgrounds. Six sites was a test of that, not a sampling shortcut.

And one thing the design does not do: it does not make any individual child's chart position a judgement. A percentile is a description of one measurement compared with that model at that moment. It carries no information on its own about whether a particular baby is thriving.

What parents can observe

What is genuinely informative at home is the sequence of a child's own measurements, not where any one of them falls on a printed curve:

• The same measurement repeated over time, taken the same way.
• The interval between measurements, because growth is uneven and a short interval exaggerates noise.
• How the measurement was taken. Length in particular is difficult to measure accurately in a baby who will not lie straight; a centimetre of technique can look like a change in growth.

This is also why Lybadora plots a baby's own logged measurements as they were recorded, with no reference curves, no percentile bands and no interpretation. The curve is a clinical tool used with clinical measurements; a home log is a record of what you measured.

What may explain a pattern

A measurement that looks out of line with the ones before it has a long list of ordinary explanations before any clinical one: a different scale, clothes on instead of off, a different person measuring, a squirming baby, a different time of day, or simply a longer gap than usual since the last reading.

Two consistent measurements taken carefully say more than five casual ones.

What parents can do

Record measurements with the date, and note who took them and how. Bring the sequence to appointments rather than a single number, and let the clinician do the plotting, they have the calibrated equipment, the full history and the training to read a trajectory rather than a point.

If a measurement surprises you, the first useful step is almost always to measure again carefully rather than to interpret the first one.