Statistics versus statistical science in the regulatory process.

This paper reviews the established practice of providing evidence to regulatory authorities about the claimed properties (such as efficacy and safety) of new pharmaceutical products. The established conventions and procedures are contrasted with scientific concepts and principles. The following issues are discussed: (a) recruitment of subjects and its connection to treatment heterogeneity; (b) the measurement process and the handling of missing data; (c) data transformation and the use of generalized linear models; (d) model selection and model checking; (e) the 'cult of the single trial' and the use of prior information; and (f) hypothesis testing and the P-value culture.