We have developed evidence on levels, causes and circumstances of deaths, and on how social and health systems settings contribute to outcomes. Working in the areas of under-5 mortality, water, alcohol and drug abuse, non-communicable diseases and HIV/AIDS, our process provides local data, rigorously validated at community and health systems levels addressing priority issues and developing action plans.
Alcohol and drug abuse

Alcohol and other drug (AOD) abuse is a growing problem in Mpumalanga Consumption is generally hazardous, associated with risky sexual behaviours, crime, violence, adverse effects on health behaviours, care seeking, utilisation and treatment adherence, with implications for HIV infection, transmission, and conditions including heart disease, cancers, stroke and seizures. Working with local communities, we developed evidence on the health and social impacts of, and local priorities to address, AOD abuse in Mpumalanga.


Lack of access to safe water poses significant risks to health and wellbeing. We worked with local communities in the MRC Wits/Agincourt Health and Socio-Demographic Surveillance System (HDSS) to develop routine surveillance data with local knowledge for action on priority health issues. In 2017, lack of water was nominated by communities as a priority topic. We held a series of community discussions on the drivers, impacts and needs around lack of water, and developed visual evidence on this topic.

Under-5 mortality

We worked with local communities in the MRC/Wits Agincourt Health and Socio–Demographic Surveillance System (HDSS) to gain local knowledge on deaths in children under–5 years. The aim was to systematise local voices to inform action through knowledge partnerships. Participants identified a range of influences related to an overburdened and under staffed clinics in the district health system and fundamental root causes related to social contexts e.g. inadequate housing, lack of clean, safe water, unemployment and poverty.


We gained insights into the perspectives of rural communities on HIV-related mortality including using the Photovoice (participatory photography) method. Addressing social practices of blame, stigma and mistrust around HIV-related mortality may be a useful focus for policy and planning. Research that engages communities and authorities to coproduce evidence can capture these practices, improve communication and build trust.

Non-communicable diseases (NCDs)

Predicted to be the leading cause of death in South Africa by 2030, NCDs

form part of a complex and dynamic burden of avoidable mortality. Most

NCD deaths are preventable and occur among people who are socially

disadvantaged. We combined routine data on NCD mortality with local

knowledge gained through community research for more complete bio-social

understandings of NCDs.