Health education can have unexpected, or even adverse, effects, particularly when it addresses multiple health risks on the one hand, and a single risk factor linked to several different conditions on the other. Some unexpected effects may result from preventive information that is either too abundant or too narrowly focused. In the first case, the information may foster a sense of relativism regarding a particular health risk, or be used by certain interests to oppose public health objectives. When the information is too targeted, it can fuel discriminatory reactions toward at-risk groups, which may result in a denial of the risk among those who do not consider themselves affected by this labeling. Furthermore, preventive measures sometimes have an unexpected effect—or no effect at all—because the targeted behavior serves a functional purpose for the individual. When seeking to prevent risky behavior, one must therefore consider the threats it helps avoid (concurrent risks), as well as other risky behaviors that achieve the same goals (substitutable risks). To understand and avoid these various pitfalls, we must acknowledge that the public holds knowledge and beliefs, often rooted in daily experience and therefore difficult to challenge. We must also contextualize the targeted risky behavior by identifying its function, competing risks, and substitutable risks, which requires deepening the dialogue with the social sciences. In the future, the need for a comprehensive approach to health issues will become increasingly apparent, as the prevention of behavioral risks increasingly focuses on multiple risk factors for a single condition as well as on the concurrent prevention of multiple conditions. [author’s abstract]