The 1998 British Crime Survey of households in England and Wales estimated that the incident rate for violence per 10,000 nurses was 1,797, placing this occupation at the second highest risk for assault after security and protective services workers. The Safer Surrey Hospital Initiative (SSHI) was developed to reduce violence against staff in five NHS Trusts in Surrey. The evaluation examined whether the SSHI's main intervention, a conflict-management training package for staff, met its goal of equipping staff with the skills and knowledge to manage actual and potential violent encounters, as well as the identification of lessons learned for achieving reductions in workplace violence in general. The post-training evaluation survey consisted mainly of precoded questions that addressed experiences of violence before and since training, the recall of elements of the training package, and perceptions of the usefulness of violence training in NHS Trust settings. A simple random sample of 600 trained staff were selected from 3 of the 5 participating Trusts. The final valid sampling frame consisted of 510 staff. For data protection reasons, evaluation researchers were not permitted to compare levels of violence and reporting patterns over the same time period for trained and untrained staff. Although trained staff reported a subsequent reduction in the number of incidents of all forms of violence, since no data were available from untrained staff, there was no way of knowing if this reduction was distinctive for trained staff. Since the evaluation found that the more complex advice given in the training tended to be forgotten after 9 months, this article advises that training packages should take into account the limits of recall as instructional material becomes more complex. 3 tables, 2 figures, and 12 references