Customised birthweight standards are widely recognised to improve the prediction of adverse perinatal outcomes compared with conventional birthweight-for-gestational-age charts. However, their apparent benefits are more likely to have been derived from their incorporation of intrauterine-based (EFW) reference values at preterm ages than their adjustment for maternal characteristics. Although maternal characteristics are able to explain population-level differences in birthweight, they are not strong enough predictors for individual-level prediction of birthweight. With maternal characteristics accounting for only a small per cent of the total factors influencing birthweight, the best estimate of an infant's birthweight remains close to the population average, explaining the ineffectiveness of adjusting for maternal characteristics. Given that customised percentiles are also unable to distinguish between pathological and physiological influences of maternal characteristics on birthweight, customising birthweight percentiles for maternal characteristics has little justification.
© 2010 Blackwell Publishing Ltd.