Building a rondavel of support: the development and pilot randomised control trial of a parenting programme to reduce the risk of child maltreatment in low-income families with children aged 3 to 8 years in South Africa

Background: In high-income countries, parenting programmes have been shown to be effective in reducing the risk of child maltreatment. However, there is limited evidence of their effectiveness in low- and middle-income countries. This thesis focuses on the development and pilot randomised controlled trial (RCT) of a parenting programme to reduce the risk of child maltreatment in low-income families with young children in Cape Town, South Africa. Methods: This thesis is comprised of three studies submitted as papers for publication. The first study focused on the development of an evidence-informed, locally relevant parenting programme for families with children aged three to eight years in Cape Town, South Africa. Intervention development took place over three stages: (a) identification of core intervention components common in evidence-based parenting programmes; (b) formative evaluation using qualitative in-depth interviews and semi-structured focus groups with South African practitioners and low-income parents; and (c) integration of evidence-based approaches and local contextual issues to develop the intervention structure, protocols, and manual. The second study used a pilot RCT (N = 68 parent-child dyads) to examine the evaluation feasibility and initial effects of the parenting programme developed during the first study in Cape Town – the Sinovuyo Caring Families Programme. Parents of children aged three to eight years with clinical levels of parent-reported child behaviour problems were randomly allocated to either a 12-session, group-based parenting programme or a wait-list control group. Primary outcomes included parent-report and observational assessments of positive parenting, harsh parenting, and child behaviour problems; secondary outcomes included parent-report of parent depression, parenting stress, and social support. Assessments occurred at baseline and immediate post-test (i.e., 3-months after baseline). The third study was a mixed-methods process evaluation assessing the feasibility of the parenting programme based on three theoretical dimensions: participation, implementation, and acceptability. Quantitative data included attendance registers, fidelity checklists, satisfaction surveys, and parent-report of engagement in home practice activities. Qualitative data included post-programme focus groups with community facilitators (n = 8), individual interviews with a randomly selected group of parents (n = 15), transcripts from parenting sessions, and minutes from supervision sessions with facilitators. Results: In the first study, the formative evaluation suggested that many evidence-based parenting programme components and approaches were compatible with the local cultural context. These included managing child behaviour problems, learning effective discipline strategies, building positive parent-child relationships, and reducing parenting stress. Findings also suggested that programmes may benefit from including additional content on keeping children safe in violent communities, communicating about HIV/AIDS and poverty, involving fathers and alternative caregivers, and incorporating cultural values of social responsibility and respect. The pilot RCT in the second study showed high levels of study recruitment and retention, outcome measurement reliability and response rates, and a minimal effect of clustering due to delivering the intervention in groups of parents. Analyses showed moderate intervention effects for parent-report of increased positive parenting and observations of improved child-led play. However, observational assessments also found reduced frequency of positive child behaviour in the treatment group in comparison to controls. In the third study, quantitative results showed high levels of programme acceptability, implementation, and participation. Thematic analysis of qualitative data identified seven themes related to feasibility: (a) receptivity to strengthening existing parenting practices, (b) initial resistance to new parenting skills, (c) contextualising content within a cultural framework, (d) reinforcing implementation fidelity and improving quality of delivery, (e) challenges delivering content on nonviolent discipline, (f) supporting participant involvement, and (g) engagement in a collaborative learning approach. Conclusion: This thesis is the first in sub-Saharan Africa to use a systematic approach to develop and rigorously pilot a parenting programme to reduce the risk of maltreatment against young children in low-resource settings. Initial results indicate that a parenting programme derived from evidence-based approaches is feasible, culturally acceptable, and has the potential to reduce the risk of child maltreatment by improving positive parenting behaviour. Further intervention development and testing is necessary to strengthen core programme components and determine programme effectiveness.