Customised birthweight models: do they increase identification of at-risk infants?

Aim The study aims to describe the cohort of women and babies who are classified as small-for-gestational age (SGA) at term by both an Australian customised birthweight model (CBM) and a commonly used population-based standard, and to investigate and compare the utility of these models in identifying babies at risk of experiencing adverse outcomes Methods Routinely collected data on 54 890 singleton-term births at the Mater Mothers' Hospitals, Brisbane, with birthweight less than 4000 g between January 1997 and December 2008, was extracted. Each birth was classified as SGA (<10th centile) or not SGA by either and/or both methods: population-based standards (SGApop) and CBM (SGAcust). Babies classified as SGApop, SGAcust or SGAboth were compared with those not classified as SGA by both methods using relative risk and 95% confidence interval, and those only classified as SGAcust were compared with those only classified as SGApop. Maternal demographics, maternal risk factors for fetal growth restriction, pregnancy and labour complications and adverse neonatal outcomes are reported. Results  A total of 4768 (8.7%) births were classified as SGApop, while 6479 (11.8%) were SGAcust of whom 4138 (63.9%) were also classified as SGApop. Maternal risk factors such as smoking and hypertension were statistically higher for the SGAcust group when compared with SGApop. For the majority of adverse neonatal outcomes, a trend was noted to increased identification using the CBM. Conclusion The CBM provides a modest improvement when compared to a population-based standard to identity term infants at birth who are at risk of adverse neonatal outcomes.