[Skip to Content]
[Skip to Content Landing]
Views 83
Citations 0
Comment & Response
April 29, 2019

Proper or Improper Use of the Crossover Design in Clinical Trials in Chronic Sciatica?—Reply

Author Affiliations
  • 1Department of Pharmacy, Medical Services Group, The Townsville Hospital, Douglas, Townsville, Queensland, Australia
  • 2College of Medicine and Dentistry, James Cook University, Douglas, Townsville, Queensland, Australia
  • 3Department of Neurosurgery, Institute of Surgery, The Townsville Hospital, Douglas, Townsville, Queensland, Australia
  • 4Public Health & Topical Medicine, James Cook University, Douglas, Townsville, Queensland, Australia
JAMA Neurol. 2019;76(6):731-732. doi:10.1001/jamaneurol.2019.0873

In Reply We have read with great interest the letter by Rato et al regarding our study1 that investigated the effect of pregabalin and gabapentin on pain intensity in adults. We thank the authors for their insight into the methodological aspects of the crossover design used.

First, the crossover design may provide guidance regarding the implications of any potential need to substitute one drug for another. For example, our trial design may determine whether the adverse effects and/or efficacy experienced with one drug is also observed with the other (ie, in the same patient, in close temporal succession after crossover). This is important because despite a lack of evidence-base, many formulatory regulation authorities worldwide typically favor one drug for subsidy rather than the other.2 This hinders interchange wherever the favored drug is either ineffective or not tolerated. We agree that chronic sciatica (CS) intensity fluctuates over time and noted treatment duration as a limitation in the article.1 We also acknowledged that a major limiting factor when using a crossover design is the mitigation or nullification of carryover effects. A common approach is to model the carryover effect and use it to adjust the treatment estimate. Such approaches, while statistically elegant, are based on assumptions that can rarely be justified in practice.3,4

×