Abstract
This paper is concerned with some of the statistical issues that may be important in comparative advertisements based on data derived from adequate and well-controlled clinical investigations. Special attention is given to: (1) The appropriateness of the design; (2) the choice of the statistical model; (3) the completeness of information; (4) the completeness of follow up; (5) the choice of a comparison group; (6) the consideration of statistical power; (7) the method of handling covariables; (8) the stopping procedures; and (9) the issue of multiplicity. This paper also discusses the need for a common protocol, a consideration of interaction between clinics in multiclinic studies. The paper illustrates these statistical issues with the aim of making the reader aware of their importance with regard to comparative advertisements.
Get full access to this article
View all access options for this article.