Dr Nishioka expresses concern about the interpretation of results from our study of corticosteroid use in immunocompetent children with and without complicated varicella.1 We highlighted the 95% confidence interval (0.2-16.9) around the point estimate (odds ratio, 1.6) as a more appropriate and significantly lower clinical estimate of risk than that previously published.2
Only 2 case subjects and 1 control subject with recent corticosteroid drug exposure were found in a review of 1079 medical charts; clearly, corticosteroid drug use in the group of immunocompetent children with varicella is a rare event. Furthermore, complicated varicella in itself is uncommon. Given this rarity, degree of risk is very much a relative issue. Using the method of Schlesselman3 for power estimation with unequal case-control ratio, the power to detect a relative risk of 12 or greater was 80%3; the power to detect a relative risk of 2 was 6%. These
Patel H, Macarthur C, Johnson D. Corticosteroids and Risk of Complicated Varicella-Reply. Arch Pediatr Adolesc Med. 1996;150(12):1314. doi:10.1001/archpedi.1996.02170370092021
Customize your JAMA Network experience by selecting one or more topics from the list below.