Copyright 2003 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2003
The article comparing tonsillectomy by plasma-mediated ablation with monopolar electrosurgery was described as reporting a prospective, randomized, blinded comparison.1 In reviewing the statistical analysis several questions arose. Patients were randomized between the 2 groups but the process of randomization was not described. Although it stated that 60 patients were needed in each group for statistical significance, it is unclear on what basis that number was chosen. To determine P values, it is necessary to first determine the rate at which a chosen variable or complication will occur. Moreover, it was decided that a hemorrhage rate greater than 2.7% in the plasma-mediated ablation group would result in the study being stopped because 2.7% was the typical postoperative hemorrage rate in the institution. However, the study was stopped when 1 author had 2 patients with airway complications, and the typical incidence of that complication rate in the institution was not stated. This is another variable that needs to be factored in when determining sample size. Surgical efficacy, estimated blood loss, surgical time during tonsillectomy, morphine use, postoperative pain, and recovery scores after tonsillectomy were other measured outcomes. Once again, sample size is determined on an incidence rate.
Yaremchuk K. Tonsillectomy by Plasma-Mediated Ablation. Arch Otolaryngol Head Neck Surg. 2003;129(4):498–499. doi:10.1001/archotol.129.4.498-a
Customize your JAMA Network experience by selecting one or more topics from the list below.