Effectiveness of the SAFE eHealth Intervention for Women Experiencing Intimate Partner Violence and Abuse: Randomized Controlled Trial, Quantitative Process Evaluation, and Open Feasibility Study

Intimate partner violence and abuse (IPVA) is a pervasive societal issue that impacts many women globally. Web-based help options are becoming increasingly available and have the ability to eliminate certain barriers in help seeking for IPVA, especially in improving accessibility. This study focused on the quantitative evaluation of the SAFE eHealth intervention for women IPVA survivors. A total of 198 women who experienced IPVA participated in a randomized controlled trial and quantitative process evaluation. Participants were largely recruited on the internet and signed up through self-referral. They were allocated (blinded for the participants) to (1) the intervention group (N=99) with access to a complete version of a help website containing 4 modules on IPVA, support options, mental health, and social support, and with interactive components such as a chat, or (2) the limited-intervention control group (N=99). Data were gathered about self-efficacy, depression, anxiety, and multiple feasibility aspects. The primary outcome was self-efficacy at 6 months. The process evaluation focused on themes, such as ease of use and feeling helped. In an open feasibility study (OFS; N=170), we assessed demand, implementation, and practicality. All data for this study were collected through web-based self-report questionnaires and automatically registered web-based data such as page visits and amount of logins. We found no significant difference over time between groups for self-efficacy, depression, anxiety, fear of partner, awareness, and perceived support. However, both study arms showed significantly decreased scores for anxiety and fear of partner. Most participants in both groups were satisfied, but the intervention group showed significantly higher scores for suitability and feeling helped. However, we encountered high attrition for the follow-up surveys. Furthermore, the intervention was positively evaluated on multiple feasibility aspects. The average amount of logins did not significantly differ between the study arms, but participants in the intervention arm did spend significantly more time on the website. An increase in registrations during the OFS (N=170) was identified: the mean amount of registrations per month was 13.2 during the randomized controlled trial and 56.7 during the OFS. Our findings did not show a significant difference in outcomes between the extensive SAFE intervention and the limited-intervention control group. It is, however, difficul