Overview

A cluster randomised controlled trial quantitatively and qualitatively examined the impact of three different delivery modalities on programme implementation within the intervention group. This included 1) geographical location (rural vs. peri-urban), 2) group composition (existing vs. newly formed groups), and 3) facilitator experience (experience vs. novice).

Context

Evidence that built our understanding of the prevalence and prevention of violence against children in low- and middle-income countries remained sparse. Evidence of effective, culturally relevant, and locally grounded VAC prevention interventions was even less.

The Parenting for Respectability (PfR) Programme in Uganda represented a home-grown, culturally-relevant parenting intervention that aimed to prevent violence against children. It had been disseminated widely in Uganda, and both government and NGOs had expressed interest to scale it. However, two key uncertainties remained: the optimal way to scale up the intervention in a ‘real-world setting’, and whether the evidence of effectiveness was confirmed through a more rigorous, experimental, evaluation.

In 2019, the Ugandan Parenting for Respectability Implementation Science Evaluation (UPRISE) project, was launched to address the first question. The current study aimed to address the second.

Study objectives

The study tested the effectiveness of the PfR programme, in modifying key outcomes in parent-child relationships and relationships between partners, which contributed to violence against children. The trial was powered to detect meaningful change in four primary outcomes: parent-reported and child-reported harsh parenting and partner conflict. Secondary outcomes assessed included positive parenting, parental support for children’s education, child behaviour problems, parent/child mental health problems, material provision for children (all parent- and child-report), and inequitable gendered socialization.

The study examined the cost effectiveness of PfR using incremental cost-effectiveness analyses (CEA) based both on the primary outcomes of harsh parenting and partner conflict as well as on disability-adjusted life years (DALYs).

The study examined how three implementation variables – rural vs peri-urban locality, previously established groups vs new groups, and professional vs non-professional facilitators – affected participation, programme fidelity, and quality of delivery (measured quantitatively).

The study examined the impact of implementation variables – participant engagement and quality of delivery by facilitators – on primary outcomes of parent- and child-report of harsh parenting and partner conflict. This project also examined associations between baseline characteristics and participant engagement to further understand potential barriers to participation and whether there were particularly vulnerable families that had greater challenges in attending the programme.

The study qualitatively investigated five elements of implementation: (i) what training facilitators needed (length, follow-up, who delivered training, location, etc.); (ii) what supervision facilitators needed (frequency, by whom, nature of feedback, etc.); (iii) targeting of PfR at the most vulnerable families and how this could be done; (iv) disseminating PfR’s messages beyond those participating in group sessions to operate at a community, as well as individual, level; (v) differences between Wakiso and Gulu Districts in facilitative and hindering contextual factors. This project explored how these affected participation, programme fidelity, quality of delivery, participant response and community-wide impact.

Project setting

The research took place in Wakiso and Gulu, two districts in Uganda.