Article Text

G324(P) Does head size or centile shift predict later neurodevelopmental problems?
  1. CM Wright1,
  2. A Emond2
  1. 1PEACH Unit, Glasgow University, Glasgow, UK
  2. 2Child Health, Bristol University, Bristol, UK


Background and methods Head circumference is widely measured in child health surveillance, but there is a lack of evidence on its predictive validity. We aimed to assess the relationship between extremes of head size and centile shifting with later neurodevelopmental problems, using the Avon Longitudinal Study of Parents and Children (ALSPAC), a UK prospective population based cohort study.

Measures head circumference measured by service health visitors at 2, 9 and 18–24 months and retrieved from the Avon child health computer; WISC IQ measured in research clinics at age 8; neurodevelopmental (ND) diagnostic codes at age 11 for all children assessed for a statement of special educational needs (SSEN) or admitted with a relevant disorder, generated from retrospective notes review.

Analysis All measures converted into WHO Z scores, internally standardised and 2–3 Z scores per child averaged (summary Z score). Extreme head size defined as head <2nd centile (–2SD) or > 98th (+2SD). Centile shift defined as difference between successive Z scores of >1 centile space (0.667SD).

Results Of the 9279 children with a summary Z score, 1.3% had heads <2nd and 1.3% >98th centile; 390 children (4%) had a ND diagnosis and /or low IQ (NDIQ) and 227 (3%) had SSEN, while 4.6% had one or both. Children with small heads had increased relative risk (2.3) of NDIQ and SSEN (5.6) compared to children with heads within the normal range (p < 0.001). However only 3% of children with NDIQ and 7% with SSEN had small heads and 85% of children with small heads had neither NDIQ or SSEN. Children with NDIQ and SSEN showed slightly slower mean head growth between 8 weeks and 9 months, but were no more likely to show a substantial shift in head centile. On a single measure at age 9 months a small head identified only 5% of those with future NDIQ or SSEN and 8 times as many false positives.

Conclusions Even with repeated measures and a perfectly fitted growth standard, head measurement is not discriminating for ND problems, which calls into question its value as a universal surveillance tool.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.