Sick population--treated population: the need for a better definition. The VALIDATA Group.

There are many questions concerned with therapy and its application. Depending on the perspective of the study there can be several "populations" which, when considered individually, may give different, or even inconsistent conclusions. These populations are: the sick population, the therapist's target population, the eligible population for a study, the study population, the treatment target population, and the treatment distribution population. Semantic precision, giving a better definition of the populations, is an essential prerequirement in order to tackle scientifically all the facets of assessment of therapy. Accurate definition will help in the study of areas as varied as the methodology, the therapeutic studies, the validity of the recommendations made, the impact of these recommendations on prescription, and finally, the consequences of the prescription in terms of public health (clinical and economical). Several examples are given to show the relationship between these populations and the importance of an accurate definition for each. Guidelines for the identification of these populations are provided. In this overall approach to therapy, epidemiology is an essential, complementary tool to clinical trials.