The methodology of the IDEFICS (Identification and prevention of dietary- and lifestyle-induced health effects in children and infants) study raises a number of important ethical questions. Many of these are already well recognised in ethical guidelines that uphold principles of individual and parental consent, confidentiality and scientific review. There are, however, wider issues that require ethical reflection. In this paper, we focus on a set of problems surrounding the evaluation of complex social interventions, and argue that comprehensive and objective evaluation is a much more ethically charged aim than it may first appear. In particular, we contend that standard scientific measures-of body size and biomarkers-convey only part of the story. This is partly because, when we intervene in communities, we are also concerned with complex social effects. These effects are made even more complex by contemporary social anxieties about fat and physical appearance, as well as about the safety and security of children. Such anxieties increase the risk of undesirable side effects that are themselves difficult to gauge. In the face of these and other complexities, we argue that the evaluation of interventions should involve a strong ethical dimension. First, it must include-as does the IDEFICS study-consideration of the opinions of the people affected, who are subjected to interventions in ways that necessarily go beyond individual consent. Second, we suggest that interventions might also be assessed by how much they empower people-and especially those persons, such as children, who are otherwise often disempowered.