Letters Section Editor: Jody W. Zylke, MD, Senior Editor.
Author Affiliations: Department of Otolaryngology, Naval Medical Center Portsmouth, Portsmouth, Virginia (Dr Gallagher); Department of Otolaryngology, Naval Medical Center San Diego, San Diego, California (Dr Brigger); and Department of Pediatric Otolaryngology, Massachusetts Eye and Ear Infirmary, Boston (Dr Hartnick; email@example.com).
In Reply: In noninferiority trials, the margin of noninferiority must be determined based on both statistical reasoning and clinical judgment.1 We think both criteria were satisfied in our study. The chief concern by Dr Czarnetzki and colleagues was that the same noninferiority threshold of 5% was applied to all bleeding end points and that this level was too high for level III bleeding. These concerns were at the center of our decision to choose level III bleeding rates as our primary end point and to power the study based on these rates and not on the comparison between the means of all types of bleeding events.
Gallagher TQ, Brigger MT, Hartnick CJ. Bleeding Risk and Dexamethasone Use in Children Undergoing Tonsillectomy—Reply. JAMA. 2013;309(5):437-438. doi:10.1001/jama.2012.113521