Statistical power in 'Prophylactic surfactant nebulisation for the early aeration of the preterm lung: a randomised clinical trial

Gaertner et al1 published a randomised clinical trial of preterm infants comparing, as the primary outcome, the difference in end-expiratory lung impedance from birth to 30 min after birth (EELI30min) for infants randomised to receive positive distending pressure, either alone (standard care) or with surfactant nebulisation (SN). The study sample size of 16 patients per treatment arm was calculated to achieve 80% statistical power to detect a mean difference in treatments’ EELI30min of 11 AU/kg at the 5% significance level. An SD of 11 AU/kg from pilot measurements was used for this sample size calculation. The EELI30min SDs observed in the study were not reported, but the IQRs are consistent with SDs much larger than 11 AU/kg. Although the sample size calculation was based on a two-sample t-test which assumes EELI30min is normally distributed, medians and IQRs were reported because the assumption of normality was violated, which typically...
Source: Archives of Disease in Childhood - Fetal and Neonatal Edition - Category: Perinatology & Neonatology Authors: Tags: PostScript Source Type: research