The purpose for this trial is to test the effectiveness of the first interactive internet-based safety decision aid on abused women's exposure to repeat intimate partner violence (IPV) and mental health outcomes. An improved safety decision-making process (e.g., knowing the advantages and disadvantages of the relationship, having enough information to make a decision) will increase safety-seeking behaviors which in turn will decrease exposure to repeat IPV and improve mental health outcomes. Findings from the development and initial test of our computerized safety decision aid suggests that it offered abused women privacy to consider personalized safety options, informed them about danger in their relationship and would be used again if they had access to it through a safe internet site.

The investigators are conducting a five year experimental trial in four states (Oregon, Maryland, Missouri, and Arizona) to address the following aims:

1. Test the effectiveness of an interactive internet-based safety decision aid on abused women's safety seeking behaviors and exposure to violence compared to women assigned to control websites. The investigators hypothesize that at three, six, and 12 months post-baseline the intervention group will have increased safety seeking behaviors and reduced IPV exposure in comparison to the control group.
2. Test the effectiveness of an interactive internet-based safety decision aid on abused women's mental health compared to women assigned to control websites. The investigators hypothesize that at three, six, and 12 months post-baseline the intervention group will have improved mental health in comparison to the control group.
3. Test if the effect of an interactive internet-based safety decision aid on abused women's mental health and exposure to violence is mediated by the safety decision making process and safety-seeking behaviors. The investigators hypothesize that the intervention group will have a better decision making process and have greater safety seeking behavior over the year in comparison to the control group, and that this better decision process and increased safety seeking behavior will mediate improvement in mental health and exposure to violence at 12 months post baseline.

This study will provide much needed new information about safety planning's impact on making difficult safety decisions, exposure to violence, and mental health effects.

Intimate partner violence (IPV) is well established as a widespread problem with important negative physical and mental health, social and cost consequences for victims, their families and the community. Beyond the physical conditions associated with IPV, research has consistently demonstrated a strong association between experiencing IPV and increased rates of anxiety, depression, low-self esteem, substance abuse and suicidality. Increasing abused women's safety and reducing ongoing exposure to violence is key to minimizing such consequences.

Building from the empowerment framework, safety planning interventions focus on protecting women from exposure to repeat violence. However, despite the significant evidence of the negative outcomes of IPV and the complex individual and community factors that influence safety, little is known about what interventions improve the safety and mental health outcomes of abused women. While experimental trials are beginning, to date there has not been a trial to examine the effectiveness of safety planning, a cornerstone of IPV interventions, on exposure to repeat violence, safety, and mental health outcomes for women. An internet-based safety decision aid may assist abused women in weighing risks and benefits of safety options, assessing danger, and identifying the factors that are most important in their safety decisions. Further, it provides personalized links with community mental health and domestic violence resources, which are invaluable in assisting women in planning for safety, thereby reducing the immediate and long terms risk of negative mental health consequences of violence.

Women who consent to participate in this study and who have access to a safe computer (e.g. at home, work, community agency, friend/family) will be randomized to the intervention or control group. If a woman is randomized into the intervention group, she will complete the internet-based safety decision aid program. The decision aid program will be located on a confidential password protected and secured study website and will take approximately 60 minutes to complete. Participants will first be asked demographic and relationship questions as well as questions about their ability to make decisions about safety in their relationship. Next, participants will be asked to make a series of comparisons to determine the importance of specific factors such as: keeping the abuse private, safety of self, family and children, feelings for partner, resources such as housing and employment and maintaining norms within family, to safety decisions. Participants will then be asked questions specific to risk factors for violence and safety seeking behaviors in their relationship. After completion, the safety decision aid generates results to provide an individualized safety plan to the participant. Women randomized into the control group will access the confidential and password protected control website which includes a brief assessment of risk factors and their history of violence, and will provide a list of local resources, but not an individualized safety plan. Both intervention and control group participants will then complete a series of instruments to measure intimate partner violence, mental health outcomes, and safety seeking behaviors. Each participant will have a skilled and trained research assistant available by phone or email to assist them in understanding questions, interpreting the results, developing strategies for safety and providing referrals for community-based services.

Follow up sessions on the study website will be conducted at 3, 6, and 12 months post-baseline. Participants will be encouraged by the RA through phone or email contact to access the password-protected website to complete the internet session and follow-up assessment questions. Post-baseline questions will be the same as the baseline, but will focus on outcomes since the previous internet session.

Inclusion Criteria:

• Female
• English or Spanish speaking
• 18-64 years of age
• Report current physical, or sexual intimate partner violence or emotional abuse within the context of threats of physical or sexual intimate partner violence
• Live in one of the target states (Oregon, Arizona, Missouri, or Maryland)
• Express comfort with and access to a safe computer with internet

Exclusion Criteria:

• Male
• Younger than 18 years of age
• Older than 64 years of age
• Cannot read/speak English/Spanish
• Live outside targeted 4 states
• No access to safe computer with internet
• Uncomfortable with computers
• Does not report past year violence from an intimate partner