Copyright 2002 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2002American Medical Association
In Reply: As we noted, we stopped our study after enrollment of 50% of the planned sample size because there was a negligible likelihood of showing a benefit to weekly courses of corticosteroids in our sample. Based on our findings after evaluation of 308 women (relative risk [RR], 0.98; 95% confidence interval [CI], 0.67-1.44) and also 502 women (RR, 0.80; 95% CI, 0.59-1.10), there was only a 7% probability that we would observe a significant difference in composite morbidity rates, if we recruited our planned sample size. We do not agree with Dr Jenkins and colleagues that our event rates were changing dramatically over time or that our assumptions in performing our conditional power analysis were false. Our trend analysis, based on cohorts of 100 patients each, is presented in Table 1. No clear pattern emerged. Jenkins et al emphasize the uncertainty surrounding the use of inferential statistics in a population sample. However, with a larger sample, preliminary trends suggesting a benefit might have been reversed. We agree that our decision to prematurely terminate the study resulted in decreased statistical power to determine the effect of weekly courses.
Guinn DA, Atkinson MW, Sullivan L, et al. Are Weekly Courses of Antenatal Steroids Beneficial or Dangerous?—Reply. JAMA. 2002;287(2):189–190. doi:10.1001/jama.287.2.189-JLT0109-2-4
Customize your JAMA Network experience by selecting one or more topics from the list below.