One-size-fits-all fetal growth charts used in the NHS to monitor babies' growth before birth often misclassify babies as being either too small or too large, which can lead to missed cases at risk of stillbirth or unnecessary interventions, finds a study of over 3 million NHS births across England published by The BMJ today.
Fetal growth restriction (when a baby’s growth in the womb is slower than expected) increases risk of adverse pregnancy outcome, so spotting this during pregnancy and before the baby is born is essential for the safety of mother and baby. Its lack of recognition during pregnancy is the most frequent cause of avoidable stillbirth.
A variety of fetal weight charts are in use across hospitals in the NHS to define small and large for gestational age babies, including Hadlock, Intergrowth-21st (IG21), World Health Organisation (WHO), and Fetal Medicine Foundation (FMF) charts, all of which are not adjustable, and the customised GROW charts which can be personalised to each individual mother.
Researchers at the Perinatal Institute in Birmingham compared rates of babies identified as too small or too large according to different charts in 3.2 million births between 2015 and 2025 in 38 of the 42 NHS integrated care boards (ICBs) in England.
They found that using a customised growth chart that adjusts for the mother's characteristics such as weight and ethnic origin, which affect the usual growth of the baby in the womb, provides more accurate and consistent data, improving safety and helping midwifery and medical staff to better identify babies requiring extra care during pregnancy.
Without adjustment for such factors, the rates of babies too small or too large varied widely between charts because of differing methods used to produce them. The ‘universal’, unadjustable charts were mostly derived from other countries, while the GROW standard was based on a UK-wide population.
As an example, the average rate of babies identified as being small for gestational age (below 10th centile) at term (37+weeks) ranged from 4.8% with Intergrowth to 17.2% with WHO and FMF charts, and 12.3% with customised GROW charts.
With the unadjustable charts, wide variation was also observed with the same chart in different ICBs, because of local population differences. This was not seen with GROW charts because they were adjusted for normal variation.
These are observational findings, and the researchers acknowledge that they used birthweight rather than ultrasound-estimated weight to assess chart performance, but this allowed all cases to be included regardless of whether they had growth scans during pregnancy.
However, the study was based on routinely collected information covering 90% of NHS areas in England, providing greater confidence in their conclusions. The authors call for urgent standardisation of growth charts used in the NHS.
The potential consequences of using inappropriate growth charts were illustrated in a recent BMJ Analysis by the Perinatal Institute on the need for the NHS to improve prevention of avoidable perinatal deaths.
The authors argue that individual NHS trusts tend to see rare but catastrophic outcomes which may result from local protocols and practice too late. They call for a co-ordinated and coherent programme across the NHS which recognises the importance of standardised growth assessment, and propose the establishment of real-time national oversight to monitor quality and safety in this essential area of maternity care.
The BMJ
Observational study
People
Designation of small for gestational age according to seven fetal growth charts in England’s National Health Service: population based cohort study of 3.2 million births
8-Jul-2026
All authors have completed the ICMJE uniform disclosure form at www.icmje.org/disclosure-of-interest/ and declare that they are employees of the Perinatal Institute, a not-for-profit social enterprise that developed the GROW charts used in the study and provide them to NHS trusts as part of its GAP programme. The authors have had no support from, or financial relationship with, any organisation that might have interest in the submitted work; and no other relationships or activities that could appear to have influenced the submitted work.