[Effectiveness assessment in public health: conceptual and methodological foundations]

In the last few years, interest has markedly increased in evaluating health programs, especially their social utility and economic efficiency. However, consensus on key issues in evaluation, such as terminology, goals and methods is still a long way off. In this context, we review the main definitions and classifications of evaluation applied to public health programs and policies. We describe the main evaluation designs and their components, focusing on outcome evaluation. Threats to the internal validity of the results of weak evaluation designs are also discussed. The characteristics of public health interventions that limit evaluation with traditional designs are also analyzed. These limitations include the complexity of interventions, usually with multiple components, and the difficulty of forming an equivalent control group with no intervention, especially through random assignment. Finally, a two-step approach to evaluation through weak designs, which takes into account adequacy and plausibility, is described. Adequacy consists of the observation of a change in the selected indicators after the intervention, and would be sufficient to take decisions under certain conditions; at other times, plausibility would need to be analyzed, defined as attribution of the results to the program or intervention.