Clinical research on pharmaceutical and environmental agents has traditionally focused on "what goes in" and "what comes out;" i.e., if a drug or other substance is administered to a population at some chosen dose(s), what (mean) response will be seen, and how does it compare to those for other doses or agents? However, the path from "dose" to "response" for any given individual following administration ordinarily involves a complex set of processes that may vary considerably from subject to subject and that play a key role dictating in the effects of an agent on an individual, and hence on the intended population. Until recently, many statisticians have been largely unaware of the implications of pharmacokinetics (PK), "what the body does to the drug," and pharmacodynamics (PD), "what the drug does to the body," for efficacy and clinical evaluation of drugs and other substances. Elucidation of the role of these processes at the individual and population level over the past 25 years has evolved largely through an amalgam of mathematical and hierarchical statistical modeling, making it one of the true "success stories" of statistics in which statistical considerations have played and continue to play a fundamental role in guiding scientific understanding. In this talk, I will review the history of this success story and the role of statistics in particular. Along the way, I will describe the basic principles underlying PK/PD and the benefits of understanding of these principles, statistical models and advances that have arisen in this area, recent attempts to exploit mathematical/statistical PK/PD modeling to improve clinical trial design, and more generally the emerging interest in integration of statistical and mathematical modeling in the study of complex disease processes for which the success of PK/PD modeling is a motivation.
Return to Biostatistics Working Group