>>Adolescent Health & development

Theme leader: Kathy Kahn 

In rural South Africa, morbidity, mortality and growth failure associated with undernutrition exist alongside female obesity and emerging cardiovascular disease. By tackling proximal issues of infant growth and nutrition, and social/emotional development, we aim to impact more distal outcomes including cognitive development, educational attainment, HIV infection, adult vascular risk and economic productivity.  Click here to read a recent 8 page publication called ‘The State of Rural South Africa’s Children‘ which highlights important research findings on children and adolescents from the MRC/Wits-Agincourt Unit.


YOUth in Care: Identifying barriers to improved healthcare outcomes among young people in rural Mpumalanga Province, South Africa, March 2019
Collaboration with the MRC/Wits Developmental Pathways for Health Research Unit and the University of North Carolina at Chapel Hill.

YOUth in Care: This project aims to identify multi-level factors that influence overall health outcomes among young people ages 12 to 24 in the Agincourt Health and Socio-Demographic Surveillance System site. This will be achieved by conducting interviewer-administered surveys among nearly 700 young people in the area and linking this information to existing facility and community level data on the specific exposures of interest.  In these interviewer-administered surveys participants will be asked about their personal experiences with violence, substance use and abuse, overall emotional well-being, and healthcare accessibility.  In addition, individuals will respond to a series of questions about their experiences seeking routine healthcare services in the study site, data which will be further supported by these individuals’ existing clinical care records.  A sub-aim of this study is to more accurately characterize routine health service utilization within this highly vulnerable population so crucial intervention points can be identified in the health services seeking process and leveraged to improve service utilization among young people.  Ultimately, we intend to utilize the results of this project to intervene on persistent barriers to routine health service utilization among young people, which we hope will yield improved long-term health outcomes in the population over time.


Snacks sold at schools in the Agincourt HDSS study area

Project Ntshembo: Improving the health and nutritional status of adolescents to reduce the intergenerational risk of metabolic disease, Feb 2019
Collaboration with the MRC/Wits Developmental Pathways for Health Research Unit and Cambridge University.
Project Ntshembo: This project aims to improve adolescent nutritional status as a critical pathway to improve intrauterine and infant growth, and thereby interrupt the intergenerational transfer of metabolic disease. This will be achieved by implementing a complex community intervention that introduces community-based adolescent-focused health workers to deliver a structured two year programme to improve rural adolescent nutritional status based on extensive formative work that has been completed in the community. This has focused on physical activity, beliefs and practices regarding childbirth and the postpartum period, and the availability and distribution of food vendors. In addition, a baseline survey was completed on a cross-sectional sample of approximately 500 young women between the ages of 18 and 21 years. in order to understand their lifestyle, eating habits, physical activity level and mental health. The aim was to examine how these factors can influence their body mass and risk of developing non-communicable diseases. Other questions were asked about their household composition, education and access to health services, and measurements included weight and height to determine body mass index, their body composition and bone health using dual energy x-ray absorptiometry and ultrasound, and their metabolic disease risk by measuring biological markers such as glucose, insulin and lipids, as well as blood pressure.  This data has also been compared to similar data from an urban Soweto sample of the same age and has resulted in the publication of several international peer-reviewed manuscripts.  A pilot and feasibility study of the proposed trial procedures has just been completed in two villages in the Agincourt HDSS study area with the ultimate aim of completing a cluster-randomised trial in the future.


The relationship between chronic stressors, resilience, hypertension and cardiovascular diseases in rural adolescents, Feb 2019
Collaboration with the MRC/Wits Agincourt Health Research Unit, Wits School of Public Health and Gothenburg University.
This was a pilot project for a study which aims to test the hypothesis that high stress and low resilience to stress during adolescence lead to high blood pressure in young adulthood. Specifically, we aim to identify the long term impact of psychosocial, psychological and physiological factors associated with perceived stress and resilience during adolescence. Furthermore, we aim to identify the mechanisms through which chronic stressors and low resilience during adolescence may cause high blood pressure as a predictor for cardiovascular diseases.

The study was conducted at the Agincourt Health and socio-Demographic Surveillance System (HDSS) site. A random sample of three hundred 13-14 year old boys and girls was produced from the HDSS. We visited them at home, asked their parents for consent and invited them to participate in the study.

In order to identify psychosocial and psychological factors, we used validated questionnaires to examine family socioeconomic status, adolescents’ communication with parents, peer relations, school satisfaction, self-perceived stress, psychosomatic problems, and health behaviours.

For identification of physiological factors, participants underwent physical examinations for Body Mass Index (BMI), blood pressure and heart rate. Blood samples were taken to assess cardiovascular risk factors such as total cholesterol, HDL, triglycerides, glucose, insulin, HbA1C and biomarkers of inflammation. In order to measure long term level of cortisol, hair samples of at least 1cm length were collected from the back of the scalp.

Lessons learnt from implementing this pilot study will be used as a basis to create a new longitudinal cohort of adolescents that will be then compared to an existing cohort of the same age group in Sweden.

SARI/ROTA – Severe Acute Respiratory Infection (SARI) and Rotavirus diarrhoea surveillance, Oct 2017  
Collaboration with National Institute for Communicable Diseases and the Respiratory and Meningeal Pathogens Research Unit, Wits
Aims to describe trends in numbers of SARI and diarrhoeal cases at 4 sentinel surveillance sites. Data will inform health policy on SARI and diarrhoeal disease management, prevention and control, and assist in planning for future influenza pandemics. Project will contribute to assessment of influenza, pneumococcal conjugate and rotavirus vaccine strategies, reflecting on recent introduction of rotavirus & pneumococcal vaccines into the national Expanded Programme on Immunisation. Additional areas being explored include, the burden of hospitalized illness due to respiratory pathogens in infants Surveillance system in two district hospitals in Bushbuckridge was set up in 2009 and data collection is ongoing.

Africa Streptococcus pneumoniae is a bacteria that normally colonizes the nasopharynx. There are 93 described serotypes, which are antigenically distinct and have different potential to cause disease in the human host. Colonization of the nasopharynx is generally asymptomatic; however, carriage of invasive serotypes may lead to Invasive Pneumococcal Disease (IPD). Prior to routine immunization of children with pneumococcal conjugate vaccines (PCV), it was estimated that pneumococcal disease caused 1 million deaths annually in <5 year olds. Current

PhD student Sarah Downs taking nasopharyngeal swabs in the MRC/Wits-Agincourt Unit study area for the PCV4 study

PhD student Sarah Downs taking nasopharyngeal swabs

vaccines against pneumococcus target the more prevalent, invasive and antibiotic-resistant serotypes. The groups most at risk of developing IPD are infants aged 1 year and younger and HIV-infected women (mothers) due to pneumococcal serotypes usually causing disease in children. Childhood PCV immunization has, however, been associated with reductions in colonization by PCV-serotypes in both HIV-infected and HIV-uninfected mothers. Measurement of the impact of childhood PCV immunization on colonization among vaccinated and unvaccinated age groups, has been proposed as a proxy for determining the effect of immunization on pneumococcal disease at a community level.

Despite the success of childhood PCV immunization, the high costs of the vaccine necessitates rationalization of dosing schedules if possible. Considering the impact of childhood PCV immunization on reducing vaccine-serotype colonization in vaccinated and unvaccinated age groups, which has also translated to reduction in vaccine-serotype IPD in these groups, it might be possible to reduce the number of doses used in childhood once there is only limited circulation and transmission of the vaccine serotypes. The transitioning to a reduced dosing schedule of PCV, however, necessitates careful surveillance of the risk of exposure to vaccine serotypes during early infancy. This can be traced though serial colonization studies to best characterize the residual burden of vaccine-serotype colonization among age groups targeted for vaccination, as well as in other age groups who might serve as a reservoir for vaccine-serotype colonization which can be transmitted to young, vulnerable children. Furthermore, such surveillance could serve as an early indicator of the re-emergence of vaccine-serotype circulation and likelihood of pneumococcal disease colonization.

In South Africa, the immunogenicity of a reduced 1+1 infant PCV dosing schedule is currently being evaluated. If confirmed that the 1+1 schedule is non-inferior in immunogenicity to the 2+1 schedule, it is possible that South Africa might transition to a reduced dosing schedule. The proposed pre-requisites for such a transition, based on expert group advise convened by WHO, would however include a high PCV coverage rate (>85%), persistent reduction of >80% in prevalence of colonization by vaccine-serotypes across all age groups compared to the pre-vaccine or early PCV-era and persistent low incidence of vaccine-serotype IPD among the vaccinated and unvaccinated age groups.

The current study is being undertaken as a prospective cross-sectional survey in Agincourt (2017) and Soweto (2018). This will provide vital information on the current prevalence of VT and NVT colonization and risk of pneumococcal disease, and help inform whether South Africa is suited to transition to a 1+1 PCV dosing schedule should it be found to be non-inferior in immunogenicity compared to a 2+1 dosing schedule in the study currently underway, as well as establish a baseline of prevalence of vaccine-serotype colonization prior to possibly transitioning to a 1+1 dosing schedule. In addition to surveillance of any changes in NVT prevalence, this data will inform any subsequent interventions and strategies relating to reduction of morbidity and mortality related to pneumococcal infection.

At the end of this project we will be able to describe the overall colonization prevalence (VT and NVT) in young children (≤5 years old) and their mothers in Soweto and Agincourt and the temporal association of routine childhood PCV immunization on colonization compared to baseline data at the onset of the immunization program.

Innovations in HIV testing to enhance care for young women and their peers and partners, Oct 2017 
Sponsored by US National Institute of Health. Partner institutions: University of North Carolina Chapel Hill, MRC/Wits Rural Public Health and Health Transitions Research Unit, and University of California, San Francisco.

testing innovations

The Study proposes to improve HIV prevention and care through expanding HIV testing options to include self-testing (link to video on self testing) for young women, their peers and their sex partners, and by facilitating linkage to care.  In this study we hope to:

Aim 1: Improve Testing Uptake and Frequency – Determine whether young women offered a choice of HIV self-testing will test at a greater rate than young women offered clinic based testing only (HCT), increasing early detection.
Aim 2: Network Testing Referral – Determine whether utilizing young women’s social and sexual networks to reach peers and male sex partners with HIV self-testing or clinic-based HCT will increase uptake of testing among young people, increasing detection of previously undiagnosed infections.
Aim 3:  Enhance Linkage to Care – For those that test positive and return to the study, we will examine the uptake of linkage to care (confirmatory testing (for self-testers), CD4 testing and ART initiation) for individuals in all study arms, and explore barriers and facilitators to linkage.


Study Timeline Years
1 (2015) 2 (2016) 3 (2017) 4 (2018)
Phase 1: Formative phase                      
Obtain IRB approvals; tool development and translation X X                    
Conduct focus group discussions (FGDs)/refine education materials   X                    
Conduct observed HIV self-testing     X                  
Adapt protocol and materials for main study using FGD results     X X                
Phase 2: Conduct intervention                        
Recruitment and enrolment       X X X            
Intervention follow-up       X X X X X X      
Linkage to care       X X X X X X      
Phase 3: Post-intervention evaluation                        
Participant in-depth interviews                 X X    
Data analysis and manuscript writing   X X X

Self testing team

The Nhluvuko study team

First phase is formative and will involve two parts:
1. Conducting formative qualitative research to understand perceptions of HIV testing and HIV self-testing in the study population and
2. Conducting observed HIV self-testing to better understand any challenges with self-testing and the materials needed to make the process clear.
Second phase of the study is a randomized controlled trial where we will randomize approximately 400 young women to receive either
1. CHOICE of self-testing or clinic-based HIV Counselling and Testing (HCT) or
2. Clinic-based HCT.

Once young women have been randomized, they will be asked to recruit up to 4 peers or male sex partners to test with the method of their randomization group, at two time points (baseline and after 3 months).

Swa Koteka/HPTN 068: Effects of cash transfer for the prevention of HIV in young South African women Oct 2017   
A collaboration between The HIV Prevention Trials Network, University of North Carolina Chapel Hill, University of Witwatersrand and Wits RHI.
HPTN 068 was a Phase III research study determine whether providing cash transfers to young women and their household, conditional on school attendance, reduced young women’s risk of acquiring HIV. The overall goal of the Conditional Cash Transfer (CCT) intervention was to reduce structural barriers to education with the goal of increasing school attendance of young women, hereby decreasing their HIV risk.

A total of 2,533 young women were enrolled between March 2011 and December 2012. Participants enrolled in the study were aged 13-20 years, in high school, not married or pregnant, and resided in the Medical Research Council (MRC) / Wits University Agincourt Health and Socio-Demographic Surveillance System (AHDSS) study site.

In the trial, young women and their parent/guardian were randomized to one of two study arms: 1) a monthly cash transfer of 300 South-African rand per month conditional on 80 percent school attendance, or 2) a control arm that did not receive cash transfer.

The study found that there was no difference in HIV acquisition between the young women who received the cash transfer and those that did not. There were 59 incident HIV infections in the intervention arm and 48 in the control arm (hazard ratio 1.17 (95% CI 0.80-1.71, p=0.43)). Overall HIV incidence was 1.8% during the study, lower than the study team expected. Surprisingly, school attendance was high during the study; 95% of girls in both arms attended school, and there was no difference in school attendance between study arms. In line with the original hypothesis of the trial, young women who stayed in school and attended school more of the time across both arms of the study had a two-thirds reduced risk of acquiring HIV.

In mid-2017 the study completed post-intervention visits with the participants to understand their behavior and HIV risk after the intervention had completed. Analysis of that data is underway.

For more information on HPTN 068 and paper references, read here.


OMC mural

OMC mural in a village, painted by community members

One Man Can Community Mobilization, Oct 2017    
Sonke’s One Man Can (OMC) project in Bushbuckridge (BBR), Mpumalanga, forms part of a two-year randomised control trial (RCT) undertaken by Sonke, the University of Witwatersrand, the University of California, San Francisco (UCSF)’s Centre for AIDS Prevention Studies, and the University of North Carolina (UNC), in order to evaluate the extent to which Sonke’s OMC community mobilisation (CM) model can reduce levels of violence and ultimately reduce the levels of HIV infection amongst young women.

This groundbreaking study provided an opportunity to examine the process and effects of a CM intervention focused on young men (aged 18-35), while also providing an opportunity to examine the effects of the mobilisation intervention in general.

For this evaluation, half of the communities (11) were randomised to receive the community mobilisation interventions, while the 11 other communities are serving as comparison communities. Community mobilisation activities included workshops, door-to-door campaigns, soccer matches, street theatre, murals, video screenings, discussions led by a trained team of community mobilisers and volunteers, and a vibrant Community Action Team (CAT) model.

Two cross-section surveys were completed – baseline in 2012 and endline in 2014. Analysis of those results are underway and will be published soon. The study design and intervention paper is available here. More information about Sonke and One Man Can can be found here.

Optimising body composition in early childhood – The Ha Tlanga (We play) project,  Oct 2017 
Collaboration with the UCT/MRC Research Unit for Exercise Science and Sports Medicine amd the MRC/Wits Developmental Pathways for Health Research Unit
Higher levels of physical activity, and lower levels of sedentary behaviour have been associated with favourable measures of body composition in early childhood. This is relevant in South Africa since obesity in this age group is becoming an increasing public health concern, although it exists alongside under-nutrition amongst young South African children, especially in Body Mass Cathi draper 3rural areas. There is evidence that interventions can increase physical activity and decrease sedentary behaviour in early childhood, and while many of these interventions have been implemented in low-income communities, the majority of interventions have been implemented in high-income countries.

Very little research has been done on levels of physical activity and sedentary behaviour of young, preschool-aged children in South Africa, particularly in rural settings. This study has aimed to address this gap in the research. In 2014, we assessed physical activity (using accelerometry and direct observation) and gross motor skills of children from three independent preschools and two primary schools (Grade R learners) in Agincourt village. Focus groups were also conducted with teachers and parents from these schools, and a questionnaire was administered in 2015 to parents and caregivers to assess perceptions of their preschool children’s physical activity behaviour.

But they're still really active (Simone Tomaz)

But they’re still really active (Simone Tomaz)

In 2017, our research focus extended to look at how physical activity and gross motor skills are associated with cognitive outcomes (specifically executive function, attention) and school readiness in preschool children in Agincourt. We returned to the same three Agincourt preschools to assess the outcomes. Similar data was also collected in Soweto, Johannesburg, so that we can compare these outcomes between a rural and urban setting.  Read our recent paper Developing Intervention Strategies to Optimise Body Composition in Early Childhood in South Africa.

Based on all of these findings, along with findings from similar work conducted in Cape Town, we are busy working on theory- and evidence-based intervention strategies to increase physical activity and decrease sedentary behaviour of preschool children in urban and rural settings. Due to our more recent research, we are also looking into ways that these intervention strategies can enhance cognitive development and school readiness in young children. These strategies will involve capacity building of preschool teachers, and promote engagement with parents and caregivers.


  • Caylee J. Cook, Steven J. Howard, Gaia Scerif, Rhian Twine, Kathleen Kahn, Shane A. Norris, Catherine E. Draper. Associations of physical activity and gross motor skills with executive function in preschool children from low‐income South African settings Dev Sci. 2019;e12820

Child and Adolescent Growth Studies, June 2015 
Collaboration with the MRC/Wits Developmental Pathways to Health Research Unit (DPHRU) 
Studies seek to document and understand the double burden of undernutrition in children (particularly stunting) and overweight/obesity in adolescents, particularly girls. In 2009, fieldwork was completed on 600 adolescents, to examine the association of nutrition on body composition and metabolic disease risk. Three PhDs have been completed on this study.

Kulani participants with certificates of participation

Kulani Child Health and Resilience Project  – evaluation of Soul Buddyz/SNOC Evaluation June 2015
Collaboration with Soul City and Oxford University, UK
School-based, cluster-randomised trial to evaluate an established school-based intervention by an NGO, Soul City, to provide emotional and social support to pupils 10-12 years. Aims to enhance learners’ ability to cope and learn in an environment of chronic adversity. Baseline study in 2009 examined rates of anxiety, depression and post-traumatic stress disorder, and environmental factors (parental death or migration, poverty) associated with these symptoms. During 2010 there was ongoing monitoring of NGO intervention and the end-of-intervention survey took place in October 2010, and the report was provided to Soul City  and to the Department of Education, Mpumalanga Province in March 2013.