Lecture in French
Public health interventions, aimed at improving the health of populations, are preventive interventions (primary, secondary or tertiary prevention) that often incorporate an educational component according to the principles of health promotion (1). These interventions are often described as complex because, in general (i) they include several components that can interact with each other, (ii) they act at different levels and target different targets, which can make their implementation difficult, (iii) they interact with the context and environment in which they are put in place (2). As a result, the evaluation of their effectiveness presents a certain number of particularities which will be evoked through examples. Firstly, the use of the reference methodology derived from drug research, namely the controlled trial with individual randomization, is often impossible because of technical and / or ethical difficulties or the political context. The need to use alternative designs, while at the same time ensuring that potential biases are mastered, is an ongoing challenge. In addition, evaluating the results of the intervention is certainly necessary, but never sufficient. To be able to conclude in terms of causality of the effect obtained, it is essential to associate an evaluation of the implementation process (3) in order to answer the following questions: (i) to what extent the evaluated intervention has been implemented as part of the study as planned? and, (ii) to what extent other concurrent interventions could not explain at least part of the result? Finally, if the intervention evaluated is effective, it is also necessary to conclude in terms of possible generalization of the large-scale intervention. To do this, the evaluation of effectiveness must also involve (i) an evaluation of the feasibility of routine intervention, and (ii) an assessment of the acceptability of the intervention by the target population.