This chapter outlines selected evidence that public health education and communication have affected important health behavior. Three questions are addressed: What programs are public health communication interventions? What are their effects on behavior? Under what conditions are such programs effective? This is an exemplary rather than a comprehensive review that should not be read as suggesting that such efforts always or usually have such effects. The nature of the available evidence precludes any such conclusion. Rather, it shows that at least in some cases there have been effects. The investigation begins by contrasting the Stanford Five City Project and the National High Blood Pressure Education Program. The first is a landmark, a well-controlled, carefully developed program that may have had only small or no effects on behavior. The second is a national multicomponent program with an uncontrolled evaluation that probably produced a massive reduction in stroke mortality. It goes on to review other apparently successful cases (e.g., AIDS campaigns in Netherlands and Switzerland, Reye's Syndrome, Sudden Infant Death Syndrome, smoking, immunization in the Philippines). The chapter closes with an analysis of the conditions of success: when the recommended behavior is supported by the health system, when much of the audience is reached repeatedly, when sponsors expect the process of change to reflect slow social norm change rather than individual effects of direct message exposure, and when new behaviors fit easily with the existing pattern of behavior.