Using couple-level patterns of intimate partner violence to predict divorce outcomes

With the high rates of marriages ending in divorce, and many experiencing intimate partner violence (IPV), it is important to understand divorce process outcomes for couples alleging IPV. There are several proposed couple-level typologies of IPV, but few have strong empirical support. Our previous work tested one typology using a data-driven clustering technique (latent class analyses). The utility of any of these typologies in predicting divorce mediation process, postdivorce and law enforcement outcomes has not been tested. The current study using archival data investigates these outcomes in relation to our data-driven IPV couple types from our previous work. This is the only study to test the utility of a set of dyadic-data-driven IPV couple types with a second method, a mediator's dichotomous decision regarding whether IPV is present in a case, using a large (N = 965) epidemiological sample of matched husband-wife data. Results indicate the latter method is more useful in predicting the status of mediation agreements; however, the IPV couple types provide a more nuanced understanding of the associations with specific status of mediation agreements, postdecree hearings and orders, and increased contact with law enforcement. This study assists mental health professionals, court personnel, and researchers to understand how patterns of IPV differentially impact outcomes of divorce mediation and postdivorce process. To reduce the costly use of precious judicial and family court resources, courts need to again consider establishing policies of early identification and differentiated case management (or triage) of identifiable couples alleging high levels of IPV victimization.