The child growth planner uses the World Health Organization's 2006 growth standard from birth to age 2 and the CDC 2000 growth reference from age 2 onward. This is the combination the Centers for Disease Control and Prevention recommended for U.S. clinical use in 2010. It is the one on most pediatric charts in the United States.
The two charts were built differently and describe different things. The difference matters for how a percentile should be read.
The WHO standard
The WHO Multicentre Growth Reference Study followed about 8,500 children from birth in Brazil, Ghana, India, Norway, Oman and the United States. Children were included only if they were breastfed according to the study's feeding recommendations, born to non-smoking mothers, and raised in environments without major constraints on growth. Children in all six sites grew similarly under those conditions. The resulting curves were published in 2006 as a standard, meaning a description of how children should grow when conditions are favorable.
Because the reference children were breastfed, the WHO weight curves reflect the breastfed pattern. Weight gain is faster in the first months and slower later in the first year than formula-fed infants typically show.
The CDC reference
The CDC 2000 charts were built from five U.S. national surveys conducted between 1963 and 1994, supplemented with birth data. They describe how a cross-section of American children grew during that period, with no selection for feeding or environment. The infants in the sample were mostly formula-fed. The charts are a reference, meaning a description of what was observed, rather than a standard of what is expected under favorable conditions.
The CDC charts extend to age 20. The WHO standard extends to age 5, and a separate WHO reference covers ages 5 to 19.
How they differ
De Onis and colleagues (2007) compared the two sets of curves. In infancy, the WHO weight-for-age curves are higher in the first few months and lower after about six months than the CDC curves, reflecting the breastfed growth pattern. Length-for-age is similar between the two. A healthy breastfed infant may appear to be falling across weight percentiles on the CDC chart in the second half of the first year. The same infant often tracks a steady percentile on the WHO chart.
The CDC recommended the WHO standard for children under 2 partly for this reason. The WHO study also measured length and weight more often in infancy, which gives better-defined curves at the ages when growth is fastest.
The transition at age 2
At the second birthday, the planner switches from the WHO length-for-age and weight-for-age curves to the CDC stature-for-age and weight-for-age curves. Two things change at that point.
The measurement changes from recumbent length to standing height. Standing height is on average about 0.7 cm less than recumbent length for the same child, and both chart sets account for this. When standing height is entered for a child under 2, the planner adds 0.7 cm before comparing it with the WHO length reference.
The reference population changes. Because the WHO and CDC curves are not identical at age 2, a child's percentile can shift by a few points at the transition with no change in the child. The planner converts the WHO z-score at the transition to the corresponding CDC z-score so that a projected trajectory stays continuous. A child's reported percentile at clinic visits just before and just after the second birthday may still differ for this reason.
Corrected age for preterm birth
For a child born before 37 weeks, the planner uses corrected age through age 2, subtracting the number of weeks of prematurity from chronological age before looking up the reference. A child born at 32 weeks is compared at 8 weeks less than their calendar age. Most pediatric practice corrects for prematurity until age 2 for growth measures, after which the difference is small relative to the spread of the charts.
What the choice of chart does not change
Both charts place a child on a distribution. Neither one judges growth. The planner uses the charts to hold a child's current z-score constant and project it forward. Outside infancy, the choice between WHO and CDC affects the shape of that projection only slightly. The methodology page describes the projection method and the L, M and S parameters both charts share.