Implementing a Digital System of Parenting Support Within Government Services in South Africa: Protocol for a 1) Time Series Trial of ParentText and 2) Randomized Controlled Trial of an Additional Programme Module for Child Safety

ParentText, a WhatsApp-based parenting support programme, is designed as a universal intervention to improve parent-child relationships (module 1) and promote child safety (module 2). The SWIFT study is an explanatory sequential time series design with an embedded randomised controlled trial. The time-series design aims to assess whether a digital parenting support programme, ParentText, can be embedded effectively into the regular services offered by the Department of Health and Wellness in the Western Cape, South Africa, and to explore factors affecting parent engagement. The embedded randomised controlled trial tests whether the second module affects sexual violence reduction against children. Participants include caregivers 18 years and older, living in South Africa, and caring for at least one child aged 2 to 17 years.

The primary outcome measures of the study are both implementation 1): Assessing which of four approaches to advertising the ParentText programme leads to the highest rates of programme uptake, engagement, and completion; and programmatic: Testing the effectiveness of the additional Child Safety Module in 2) - reducing child sexual violence vulnerability, 3) - increasing child sexual violence risk planning and 4) - reducing child sexual violence exposure.

The secondary outcomes assess changes in programme users' self-reported parenting behaviour in terms of 1) positive involvement, 2) positive parenting, 3) parent mental health, 4) parent support of education, 5) caregiver child/ adolescent communication, 6) online safety, 7) harsh parenting, 8) child or adolescent behavioural problems; and assessing users' experience of the 9) accessibility of the chatbot, 10) acceptability of the chatbot; and 11) adolescents' report of their sexual violence risk and safety planning communication..

In the first period of the time series design, participants will be recruited via posters, which provide a WhatsApp number for the programme, at their local clinic. Once they join, they will be provided with information on the study and be invited to consent to participate. If they consent, they will complete a baseline survey and then be able to move on to the first 5-day module. On consent to the study, participants will be placed randomly into either the group that receives the second module (intervention group) or a control group that will receive the module only after the study is completed. Both the intervention group and the control group receive the one-month and three-month surveys.

In the second, third, and fourth time periods, everything remains the same except the encouragement method. In the first time period, only posters were used for recruitment. Within the second period, one month later, recruitment will be via both posters and additional encouragement from clinic staff. Clinic staff will encourage either verbally or by providing the caregiver with a flyer during their clinic visit. Throughout both these recruitment periods (1) and (2), caregivers who have completed the programme can share it on social media by posting predesigned tiles that include the joining information for the chatbot. Should our ideal sample size (4,000 participants) not be reached through these mechanisms, there will be an additional time period (3-4 weeks; time period three) during which the study team will advertise directly on Facebook. Finally, should our sample size not be reached in this third period, there will be a fourth time period of advertising that offers ZAR50 (approximately USD3) in airtime to those who sign up. This ZAR50 will be provided in sums of: ZAR15 for completing each of the baseline and one-month questionnaires, and ZAR20 for completing the three-month questionnaire.

Once the three-month questionnaires are completed, participants will also be invited to a focus group discussion or telephonic interviews. Participants will be purposively selected from the group of participants who consent to the qualitative data collection phase.

The focus groups will be conducted with participants residing in the Western Cape province, who completed one or both modules, and will discuss topics such as their experience of the ParentText programme, and the encouragement.

For the telephonic interviews, participants will be purposely selected for interviews, looking at a) caregivers' experience of the child safety module, b) reasons for engaging/not engaging with the programme, and c) which referral services they accessed.

Further, adolescents (over the age of 12 years) will be recruited via their caregivers - upon completion of the 3-month survey, parents or legal guardians will be asked to provide consent for the team to contact their adolescent to undertake a telephonic survey and, in some cases, a qualitative interview with them. Teens will complete informed assent.