Do babies grow the same way in the womb around the world?
Even in carefully selected healthy pregnancies, fetal growth differed markedly between ten countries, so a single global growth chart can misclassify babies as too small or too large.
Source
The World Health Organization Fetal Growth Charts: A Multinational Longitudinal Study of Ultrasound Biometric Measurements and Estimated Fetal Weight
Study at a glance
- Design
- Cohort — Prospective multinational longitudinal cohort: low-risk women recruited in the first trimester in ten countries had standardised serial ultrasound biometry at roughly 4-weekly visits; percentiles built with quantile regression and tested for effects of country, fetal sex and maternal characteristics.
- N
- N=1362 · 1,362 women contributing ultrasound data to the growth curves, from 1,439 enrolled (52 withdrew consent); 8,203 scan sessions were analysable.
- Population
- Healthy pregnant women aged 18-40 with BMI 18-30, singleton pregnancies, no smoking or chronic disease, living below 1,500 m in Argentina, Brazil, D. R. Congo, Denmark, Egypt, France, Germany, India, Norway and Thailand.
- Outcome
- Percentile reference curves for estimated fetal weight and biometric measures (biparietal diameter, head and abdominal circumference, femur and humerus length), and the influence of country, fetal sex and maternal factors on them.
Structured fields used in claim comparison tables when every cited study has a complete layer.
What they did
WHO recruited low-risk pregnant women early in pregnancy in ten countries in Africa, Asia, Europe and South America, confirmed gestational age by ultrasound, and scanned each fetus about every four weeks with identical machines and trained sonographers. From these repeated measurements they calculated estimated fetal weight and built percentile charts with quantile regression, a method that makes no assumption about the shape of the distribution. They then tested how much country, fetal sex and maternal age, height, weight and parity shifted the percentiles.
What they found
Country significantly affected every estimated-weight percentile even after adjusting for maternal characteristics and fetal sex. Near term the 10th percentile of estimated weight was about 3,400 g in Norway but 2,700 g in India, against a global value of about 3,100 g, so the same fetus could be 'small' in one country and normal in another. Male fetuses were a few percent heavier than female ones, and older, taller, heavier and parous mothers had somewhat larger fetuses. The weight distribution became increasingly skewed towards heavier fetuses late in pregnancy, and removing pregnancies with complications barely changed the charts.
The limits
What it doesn't show
Ten sites, one per country, are a thin sample of global diversity, especially for Africa, and the site samples may not represent their national populations. The study was not designed to explain why countries differ, so it cannot separate genetic, dietary, socioeconomic or measurement contributions, and ethnicity was only partly recorded. Gestational age appeared on screen during scans, which could in principle have influenced care, and pooled charts only partly reflect each country, so the authors advise testing and possibly customising them locally.
Key terms
- Estimated fetal weight (EFW)
- A fetal weight calculated from ultrasound measurements of head, abdomen and femur using a formula (here Hadlock's).
- Quantile regression
- A regression method that models chosen percentiles directly, without assuming the data follow a normal distribution.
- Prescriptive standard versus reference chart
- A standard describes how fetuses should grow under ideal conditions; a reference describes how a given population actually grows.
- Percentile
- The value below which a given percentage of measurements fall; fetuses below the 10th percentile are often labelled small for gestational age.
- Parity
- Whether a woman has given birth before; parous women tended to have heavier fetuses.
- Customised growth chart
- A chart adjusted for factors such as country, fetal sex or maternal size to better fit the individual pregnancy.
Flashcards
0 of 10 answers reviewed
Research intelligence for this paper
See its role on concept claims, tensions it is part of, placement history, and related discoveries.
Quiz yourself
What statistical method was used to build the WHO fetal growth percentiles?
Common questions
Why not exclude pregnancies that developed complications?
The charts are meant for clinical use in real populations, and removing complications or the smallest babies would artificially shift the lower percentiles; the authors also showed that excluding them made little difference.
How does this study differ from the Intergrowth-21st project?
Intergrowth-21st used a strictly prescriptive approach and did not publish estimated fetal weight; this WHO study reports EFW charts and emphasises that growth varies between populations.
Should clinicians use the WHO charts everywhere?
The authors say they can be used where no local data exist, but performance should be checked locally and percentiles adjusted or customised if they misclassify fetuses.
More on Maternal health