Richard J Cook1, Nancy M Heddle. 1. Department of Statistics and Actuarial Science, University of Waterloo, Waterloo, Ontario, Canada. rjcook@uwaterloo.ca
Abstract
BACKGROUND: Several randomized trials of platelet (PLT) products have been conducted with different study designs, endpoints, and analyses. The purpose of this article is to discuss methodologic issues in the design and analysis of PLT transfusion trials evaluating pathogen reduction technology and make recommendations for the conduct of future trials. STUDY DESIGN AND METHODS: Six randomized clinical trials of pathogen-inactivated PLT products are reviewed and associated methodologic issues are discussed. RESULTS: The variation in the trial designs, outcomes, and methods of analysis suggest the need to harmonize the way trials of pathogen-reduced PLT products are conducted to facilitate comparisons between studies and the synthesis of results. Recommendations are made with this goal in mind and to increase the rigor and relevance of findings from future trials. CONCLUSIONS: Future randomized trials of pathogen-reduced PLT products should be based on a clearly stated hypothesis driven by an important research question, a design that is optimal for the research question, outcomes that relate to the research question, clearly defined observation periods, and statistical analyses that lead to valid tests of these hypotheses and associated estimates of treatment effect.
BACKGROUND: Several randomized trials of platelet (PLT) products have been conducted with different study designs, endpoints, and analyses. The purpose of this article is to discuss methodologic issues in the design and analysis of PLT transfusion trials evaluating pathogen reduction technology and make recommendations for the conduct of future trials. STUDY DESIGN AND METHODS: Six randomized clinical trials of pathogen-inactivated PLT products are reviewed and associated methodologic issues are discussed. RESULTS: The variation in the trial designs, outcomes, and methods of analysis suggest the need to harmonize the way trials of pathogen-reduced PLT products are conducted to facilitate comparisons between studies and the synthesis of results. Recommendations are made with this goal in mind and to increase the rigor and relevance of findings from future trials. CONCLUSIONS: Future randomized trials of pathogen-reduced PLT products should be based on a clearly stated hypothesis driven by an important research question, a design that is optimal for the research question, outcomes that relate to the research question, clearly defined observation periods, and statistical analyses that lead to valid tests of these hypotheses and associated estimates of treatment effect.