While a large body of research has been generated to explain wife assault, relatively few studies have focused on predicting wife assault. The current literature review looks at the empirical research for evidence of predictors of wife assault recidivism, lethal wife assault, and wife assault onset, as well as any evidence on recidivism following wife assault treatment. The main empirical findings indicate that the following variables are risk factors for wife assault recidivism: young age, low socioeconomic status, a history of marital conflict and abuse, and antisocial personality and conduct, including substance abuse. Next, various structured clinical risk assessments and actuarial risk assessments are reviewed for their use in predicting wife assault onset and recidivism. In the prediction of lethal wife assault, the authors suggest the use of actuarial assessments of wife assault risk combined with a focus on the strongest correlates of lethal assault. Future research should continue to make improvements on existing risk assessments and public policy should focus on more intensive risk management efforts targeted at high-risk offenders. Glossary, appendix, references