Interpersonal violence against women and children has increasingly been recognized as a public health priority in humanitarian emergencies. However, because the household is generally considered a private sphere, violence between family members remains neglected. A systematic literature review was conducted to identify predictors of household violence in humanitarian emergencies. PubMed, Web of Science, and Scopus were searched from January 1, 1998, to February 16, 2016. A predictor was defined as any individual, household, or community-level exposure that increases or decreases the risk associated with physical, sexual, or emotional interpersonal violence between two or more people living together. All studies reporting on quantitative research were eligible for inclusion. Results were analyzed using qualitative synthesis. Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines were followed as applicable. The search strategy resulted in 2,587 original records, of which 33 studies met inclusion criteria. Thirty-two of the 33 studies used a cross-sectional design. This was the first known systematic review of predictors of household violence in humanitarian settings. The household framework drew attention to several factors that are associated with violence against both women and children, including conflict exposure, alcohol and drug use, income/economic status, mental health/coping strategies, and limited social support. There is a need for longitudinal research and experimental designs that can better establish temporality between exposures and household violence outcomes, control for confounding, and inform practice. In the interim, programmers and policy makers should try to leverage the predictors identified by this review for integrated violence prevention and response strategies, with the important caveat that ongoing evaluation of such strategies is needed.
Keywords: child abuse; children exposed to domestic violence; domestic violence; predicting domestic violence; war.