The study hypothesized that the training would be associated with increased participant rating of self-efficacy in identifying and helping partner violence victims; that participants would demonstrate an increased endorsement of the role of health care providers and settings in helping partner violence victims; that participants would report increased comfort with making appropriate community referrals to help partner violence victims; and that training effects would be moderated by prior training in partner violence and by prior experience with helping partner violence victims. These hypotheses were tested before and after training and at a 6-month follow-up. The training program was the Family Peace Project, which was modified to meet specific local needs. The goals of this training are to provide participants with a common knowledge base about the definitions, dynamics, incidence, and prevalence of partner violence in medical settings. These goals are achieved primarily through didactic lecture and facilitated discussion. The training also facilitates the development of participant attitudes about the importance of viewing partner violence as both a public health and an individual medical problem that deserves the attention of medical professionals through the implementation of screening and intervention skills. Participants completed the Health Care Provider Survey on Intimate Partner Violence both before and after the training. At 6-month follow-up, one-third of the participants (n=277) were randomly recruited to complete the survey to assess long-term changes in self-efficacy and attitude toward screening. A total of 124 participants responded to the follow-up survey. The findings show that all four of the hypotheses were supported through the 6-month follow-up. The findings suggest that different departments within a health care system may have different training needs that must be assessed and addressed to optimize training effects. The study may also have implications for health care system administration, which must approve universal training programs for health care providers. Based on trainees' prior training and experience, it may be possible to develop and implement different, possibly shorter training models, so the training may be more cost efficient. Study limitations and future research are discussed. 6 tables and 42 references