TY - JOUR T1 - Fantoms JF - Archives of Disease in Childhood - Fetal and Neonatal Edition JO - Arch Dis Child Fetal Neonatal Ed SP - F79 LP - F79 VL - 92 IS - 2 AU - Martin Ward Platt Y1 - 2007/03/01 UR - http://fn.bmj.com/content/92/2/F79.abstract N2 - Neonatologists unfamiliar with the arcana of bronchopulmonary dysplasia (BPD) could be forgiven for being surprised at the intrusion of sex hormones into a world dominated by cytokines, modes of ventilation, vitamin supplementation and corticosteroids. Yet the rationale for supplementing preterm infants’ nutrition with estradiol and progesterone is there, and Trotter et al have done a randomised controlled trial to test the effectiveness of these two hormones empirically, when given together, in reducing the risk of BPD. Unfortunately the messages we take away from this paper are not entirely the ones the authors intended. First, we see how a power calculation does not necessarily translate into a study with sufficient power, if the attrition of subjects from the intervention arm is not adequately allowed for. Second, we see how an underpowered pragmatic randomised controlled trial can nevertheless be salvaged by the use of multivariable analysis, and that this is particularly powerful when the non-intervention arm is not contaminated by deliberate or inadvertent administration of the agents under test. Finally we can conclude that since there was … ER -