Using some light-hearted 'stereotypes' and informal examples, we will explore some fundamental differences between biostatisticians and pharmacometricians in drug development roles. These include visibility and familiarity to clinical teams, history in the pharmaceutical industry, types of models, and communication of results. Kowalski (1) identified several points of tension and skepticism that has in the past hindered successful collaboration between these two quantitative disciplines. Some of these sources of conflict include the distinction between having a good fitting model versus a model with demonstrated predictive performance, the use of multiple comparisons (stats) versus dose- or exposure-response (pharmacometrics) to make dosing decisions, and pharmacometricians' frequent use of exploratory data analyses to draw confirmatory conclusions. While many of these issues do indeed cause friction, it is important to recognize that pharmacometricians and biostatisticians are both, at their core, modelers; partnering together as a quantitative team is critical. We will make some recommendations on how these two disciplines can cooperate and leverage their complementary assets and skillsets to strengthen the quality of drug development decision making.
1. Kowalski KG. My Career as a Pharmacometrician and Commentary on the Overlap Between Statistics and Pharmacometrics in Drug Development, Stat Biopharm Res. 2015; 7(2):148-59.