The first model involves public health campaigns that identify and address the underlying causes of a problem. Although experts do not agree on causation of domestic violence and several different theories exist, these theories share some common beliefs that can serve as a foundation for prevention strategies. The second public health model can be useful for identifying opportunities for domestic assault prevention along a continuum of harm. This continuum has three parts: (1) primary prevention to reduce the incidence of the problem before it occurs, (2) secondary prevention to decrease the prevalence after early signs of the problem, and (3) tertiary prevention to intervene once the problem is already clearly evident and causing harm. Examples of primary prevention include school-based programs that teach students about domestic violence and alternative conflict resolution skills, as well as public education campaigns. Secondary prevention programs could include home visiting for high-risk families and community-based programs on dating violence for adolescents referred through child protective services. Tertiary prevention includes the man targeted intervention programs already in place. Early evaluations of existing prevention programs are promising, but results are still preliminary and programs remain small, locally based, and scattered throughout the United States and Canada. What is needed is a broadly based, comprehensive prevention strategy with sound research and evaluation, adequate public backing, and a policy of zero tolerance for domestic violence. Case examples, photographs, and 46 references (Author abstract modified)