Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)

Graduate Group

Social Welfare

First Advisor

Phyllis Solomon


Background & Significance: Despite estimates that persons with serious mental illness (SMI) are between 2 and 8 times more likely to commit acts of violence than are members of the general population and that approximately 50% of all acts of violence by persons with SMI are against family members, the subject of family violence by persons with SMI has received little research attention. Hypothesis: After reviewing the literatures on community and family violence by persons with SMI, it was hypothesized that family violence by this population is associated with factors in four domains: 1) Perpetrator, 2) Victim, 3) Interaction, and 4) Community. Methods: A cross-sectional survey design was used. Between December 29, 2015 and April 1, 2017, 523 persons with SMI living in the U.S. completed an online survey. Respondents were recruited from a range of mental health organizations across the U.S. Respondents provided information regarding themselves, a reference relative, and the interactions they and reference relatives had with each other in the past 6 months, including possible acts of violence. The association of factors with the occurrence of violence by persons with SMI towards reference relatives was estimated with multivariate logistic regression. Results: Twelve percent (n = 13) of persons with SMI reported having committed violence towards their reference relative in the past 6 months. In the final multivariate logistic regression model, the following factors were significantly associated with violence by persons with SMI towards reference: Perpetrator—history of ever committing serious violence; Victim—age; Interaction— use of limit-setting practices by relatives, psychological abuse by persons with SMI and relatives, and violence by relatives. Conclusions: Mutual violence appears to play a considerable role in family violence by this population. Practitioners may better serve clients with SMI by offering to involve family members in their treatment and by assessing and intervening in family conflict. Interventions aimed at decreasing psychological abuse may help deescalate conflict and prevent physical violence. Interventions intended to decrease or modify limit-setting practices used by relatives towards persons with SMI may decrease the risk of family violence by this population.