A pilot health school programme has been set up on an experimental way since January 1984 in a few professional and high schools of the Canton of Vaud. The paper presents the methodological tools that will allow an evaluation of various aspects of this experience, which has lasted for three years. The process will be evaluated among three populations: the pupils, the teachers and the members of the health team. The results will be evaluated in two fields: the screening of hypertension and tuberculosis on one side and the outcome of individual counseling on the other side. The paper illustrates both the utility and the difficulties linked to the setting up of such an evaluation process: it underlines the complexity of choosing indicators that are useful accessible and acceptable among the huge number of available data.