A digital parenting intervention with intimate partner violence prevention content: A pilot pre-post study in Jamaica and South Africa (Preprint)

BACKGROUND Intimate partner violence (IPV) and violence against children (VAC) are global issues with severe consequences. Intersections shared by the two forms of violence have led to calls for joint programming efforts that prevent both IPV and VAC. Parenting programmes have been identified as a key entry point for addressing multiple forms of family violence. Building on the IPV prevention material that has been integrated into the parenting programme ParentText, a digital parenting chatbot, this pilot study seeks to explore parents’ engagement with the IPV prevention content in ParentText, and explore preliminary impacts on IPV. OBJECTIVE This study aimed to assess parents’ and caregivers’ level of engagement with the IPV prevention material in the ParentText chatbot and explore preliminary indications of the chatbot’s impact on experiences and perpetration of IPV, attitudes toward IPV, and gender-equitable behaviours following the intervention. METHODS Caregivers of children 0-18 years were recruited through convenience sampling by research assistants in Cape Town in South Africa and by UNICEF Jamaica staff in three parishes in Jamaica. Quantitative data from women (N=47) and, in South Africa, men (N=21) caregivers were collected electronically via weblinks sent to users’ phones using Open Data Kit. The primary outcome was IPV experience (women) and perpetration (men), with secondary outcomes including gender-equitable behaviours and attitudes toward IPV. Descriptive statistics were used to report sociodemographic characteristics and engagement outcomes. Linear, Poisson, and logistic regression models were used to investigate potential changes in IPV outcomes between pre- and post-test. RESULTS The average daily interaction rate with the programme was 0.57 and 0.59 interactions per day for women and men in South Africa, and 0.21 for women in Jamaica. The rate of completion of at least one IPV prevention topic was 25% (5/20) for women and 5% (1/20) for men in South Africa, and 21% (6/28) for women in Jamaica. Exploratory analyses detected significant pre-post reductions in overall IPV experience among women and in men’s overall harmful IPV attitudes. CONCLUSIONS This pilot study is the first to our knowledge that investigates user engagement with and indicative outcomes of a digital parenting intervention with integrated IPV prevention content. Study findings provide valuable insights into user interactions with the chatbot and shed light on challenges related to low levels of chatbot engagement. Indicative results which suggest reductions in IPV and improvements in attitudes after the programme are promising, yet, modest. Further research using a randomized controlled trial is warranted to establish causality. CLINICALTRIAL The trial is embedded within a larger study registered with ClinicalTrials.gov (NCT05003518) on 12/08/2021.