Evaluation of Safety Data from Controlled Clinical Trials: The Clinical Principles Explained
- PDF / 952,358 Bytes
- 15 Pages / 504 x 719.759 pts Page_size
- 36 Downloads / 174 Views
Drug Informarion Journal, Vol. 34, pp. 273-287. 2000
Printed in the USA. All rights reserved.
EVALUATION OF SAFETY DATA FROM CONTROLLED CLINICAL TRIALS: THE CLINICAL PRINCIPLES EXPLAINED JOHNE. GAIT,MD, SANDYSMITH,PHD, AND SHERRIL. BROWN,MD International Clinical Safety Department, Amgen Inc, Thousand Oaks, California
A comprehensive discussion is presented of the principles underlying the evaluation of
safety data from controlled clinical trials. Emphasis is placed on clinical concepts rather than statistical methodology. The essential elements of safety analysis, the critical concepts involved, and trial design issues that affect the interpretation of safety data are explained. Particular attention is paid to the importance of exposure data, the estimation of adverse event frequencies in longer-term trials, evaluation of laboratory data, and the assessment of risk factors. Key Words: Adverse event; Bias; Laboratory data; Exposure; Cumulative probability
INTRODUCTION THE ANALYSIS OF THE safety data from a clinical trial is sometimes perceived as merely a necessary task requiring little thought regarding methods of presentation and analysis or clinical interpretation of results. This is because the primary objective in most clinical trials is to demonstrate the efficacy of the trial medication; assessment of safety is a secondary objective with no hypotheses to be tested. The safety analysis may, therefore, be regarded as a simple matter of listing adverse events in descending order of frequency by treatment group, and comparing rates of occurrence of abnormal laboratory values. Often, the only safety data that are presented in clinical trial reports are those which show marked differences between the trial medication and the control. The reader is left to assume that all data
Reprint address: Dr. J. E. Gait, International Clinical Safety Department, Amgen Inc, 1, Amgen Center Drive, Thousand Oaks, California, 91320-1799. E-mail: [email protected].
relevant to safety have been assessed and there were no other findings, but the report does not actually describe what was done. In reports on trials where the medication is found to be ineffective, the data that lead to that conclusion are always presented. The purpose of this article is to provide a comprehensive discussion of what is involved in evaluating safety data, to explain the clinical thinking behind the various methods of data presentation, and to illustrate some of the pitfalls of interpretation. There are some philosophical differences in the analysis and reporting of safety data as compared to efficacy data. With efficacy data the objective is to prove a positive, that is, that the product has beneficial effects on the disease in question. What is not always so obvious is that with safety data the objective is to prove a negative, that is, that the product does not have any unacceptable tolerability or safety issues. While efficacy data relate to particular parameters in particular diseases, all available safety data are relevant regardless of the ind
Data Loading...
