This study examined a sample of 965 couples who attended mediation in Pima County, Arizona, for the purposes of resolving custody and parenting disputes. Based on a semi-structured clinical interview with each parent individually, mediators identified IPV in 59 percent of cases. This study concluded that mediators identified many, but not all, cases of self-reported IPV, and it suggests that classification based on a non-systematic, semi-structured interview may not be the best method for identifying IPV. This study also found that safety accommodations were provided most often for the participants who specifically requested it; about 19 percent of all couples received at least one accommodation, despite a much higher prevalence of IPV overall. Regarding the link between IPV and custody agreements, higher levels of IPV were associated with a lower likelihood of a full or partial agreement being reached. Also, mediated agreements rarely restricted parenting or contact between parents, even in cases with the highest levels of IPV. Among the study's recommendations are that mediators, judges, and court staff be trained in the dynamics of IPV and that a systematic assessment of it be adopted for divorce mediation. Researchers also proposed an exploration of a hybrid mediation/arbitration model that could force creation of binding agreements. Online access to the full report is provided.