Koch G G, Davis S M, Anderson R L
Department of Biostatistics, University of North Carolina, Chapel Hill, USA.
Stat Med. 1998;17(15-16):1675-90. doi: 10.1002/(sici)1097-0258(19980815/30)17:15/16<1675::aid-sim970>3.0.co;2-c.
Regulatory success for new medical products requires convincing statistical results from planned analyses for relevant and reliable data from well-designed clinical studies. Statistical methodology has a critical role for such success through strengthening the robustness of study findings to sources of bias, variability and spurious events. Areas of particular importance for statistical methodology include the structure for study design, the schedule and procedures for data collection, and plans for primary data analyses. This paper discusses some challenging statistical issues for these areas and suggests potential strategies for addressing them. The objective of these strategies as well as other methodological advances is the minimization of ambiguity in the findings of confirmatory studies. In this way, their effective use can enhance regulatory success.
