|
|||||
|
|
||||||
Journal of Clinical Oncology, Vol 23, No 25 (September 1), 2005: pp. 6267 © 2005 American Society of Clinical Oncology. DOI: 10.1200/JCO.2005.01.6154
In Reply:Department of Medicine and Department of Clinical Epidemiology and Biostatistics, McMaster University; and the Juravinski Cancer Centre, Hamilton, Ontario, Canada A meta-analysis is best performed with data from randomized trials.1,2 The pooling of results by Bentzen and Yarnold does not correct for any potential biases in the three case series because of the non-random allocation of data between sequential versus concurrent tamoxifen and radiotherapy. Nonetheless, their analysis identifies the limited power of the three studies and supports the need for a large randomized controlled trial to address this issue. Authors' Disclosures of Potential Conflicts of Interest The authors indicated no potential conflicts of interest. REFERENCES
1. Oxman A, Cook D, Guyatt G: Users' guides to the medical literature VI: How to use an overview. JAMA 272:1367-1371, 1994
2. Kinz R, Oxman A: The unpredictability paradox: Review of empirical comparisons of randomised and non-randomised clinical trials. BMJ 317:1185-1190, 1998
Related Correspondence
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||
|
Copyright © 2005 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|