Impact and sustainability evaluation of the KaziKidz physical activity and health promotion intervention in public primary schools in South Africa

Background: Non-communicable diseases (NCDs) are the leading cause of death and disability worldwide, disproportionately affecting low- and middle-income countries (LMICs). Behavioural risk factors, notably insufficient physical activity, contribute significantly to cardiometabolic changes associated with NCDs. Extensive research has demonstrated the prevalence of these risk factors in children and adolescents and their tracking into adulthood. Yet, evidence from sub-Saharan Africa is scarce. Early identification of cardiovascular risk and promotion of healthy, active lifestyles from a young age are vital to prevent NCDs later in life. This is especially relevant in socioeconomically disadvantaged communities in LMICs, including South Africa, where health inequalities and physical activity barriers are prevalent. In such settings, schools acquire particular importance for shaping physical activity and health habits among children. Therefore, there is a pressing need for context-specific, comprehensive, cost-effective interventions to strengthen physical and health education in South African under-resourced schools. However, ensuring the sustainment and real-world effectiveness of such interventions poses a substantial challenge and warrants thorough evaluation.
Aims: This PhD thesis aimed to evaluate the KaziKidz physical activity and health promotion programme, a sub-initiative from the KaziBantu project, during its sustainability phase in public primary schools from South Africa. The specific objectives included assessing the identification of physical activity and cardiovascular risk as well as their association in South African children, examining short-term intervention effects on NCD risk factors and their long-term development in high-risk children, exploring stakeholder acceptability and perceived feasibility, and evaluating the programme’s continuity in participating schools.
Methods: KaziKidz was implemented in the context of the KaziBantu randomised controlled trial across eight quintile 3 schools in low-income areas near Gqeberha, South Africa. Four schools served as controls, and four implemented the intervention receiving teaching materials. One intervention school was additionally supported by teacher workshops, one by a physical education coach, and a third one by both. Intervention’s efficacy was assessed on children’s (n = 961) self-reported and accelerometer-captured physical activity, body-mass-index, blood pressure, serum lipids, and plasma glucose. Quantitative measurements were gathered twice, at baseline (January 2019) and after a 20-week intervention period (October 2019). A follow-up mixed-methods study was conducted two years’ post-intervention (October 2021) during the programme’s sustainability phase. Quantitative data were collected from a sub-cohort of high-risk children (n = 280), who presented at least one NCD risk factor, namely overweight, elevated blood pressure, dyslipidaemia, or hyperglycaemia. Qualitative assessments encompassed semi-structured interviews with school principals (n = 3) and focus group interviews with teachers (n = 16) and caregivers (n = 16). The implementation science RE-AIM framework guided the comprehensive evaluation of KaziKidz using a mixed-methods convergent design.
Results: Divergent results were observed between instruments assessing paediatric hypertension and physical activity and their association with cardiovascular risk, demonstrating the need for evidence-based, context-appropriate methodologies for identifying risk in South African children. In terms of short-term intervention efficacy, KaziKidz positively impacted moderate-to-vigorous physical activity levels among girls at school and improved mean arterial pressure of children as well as cholesterol and glucose levels of non-high-risk children. Nevertheless, research is needed to determine the benefit of specific intervention components for distinct subgroups and further enhance physical activity levels for all children inside and outside of school. School staff and caregivers emphasised the programme’s significance for children, but called for increased involvement and external support to deliver and sustain the programme in schools. The evaluation of the intervention’s sustainability was hampered by the Coronavirus Disease 2019 (COVID-19) pandemic and closure of schools, resulting in the discontinuation of KaziKidz. Furthermore, a decline of health outcomes among high-risk children was observed in the long-term. Accounts from school personnel suggested that, under normal circumstances, maintaining the programme without additional support would have been challenging.
Conclusion: Public schools are pivotal platforms for ensuring quality physical education, providing physical activity opportunities for all children, and promoting healthy behaviours in disadvantaged communities in South Africa. KaziKidz’s holistic approach holds promise for improving risk factors for NCDs, but ensuring intervention continuity is imperative for sustaining long-lasting health benefits. In-service training should enhance programme ownership among nonspecialist teachers and be supported by implementation coaching, school administration engagement, and political commitment. Evaluating the feasibility of these measures and the effectiveness and fidelity of KaziKidz’s delivery upon resumption is recommended to refine the programme and guide its scale-up.