Verbal autopsy with participatory action research (VAPAR) programme in Mpumalanga, South Africa: protocol for evaluation

Sophie Witter* (Corresponding Author), Maria Van Der Merwe, Rhian Twine, Denny Mabetha, Jennifer Hove, Gerhard Goosen, Lucia D'Ambruoso

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

7 Citations (Scopus)
15 Downloads (Pure)


Introduction There is a growing recognition of the importance of developing learning health systems which can engage all stakeholders in cycles of evidence generation, reflection, action and learning from action to deal with adaptive problems. There is however limited evaluative evidence of approaches to developing or strengthening such systems, particularly in low-income and middle-income settings. In this protocol, we aim to contribute to developing and sharing knowledge on models of building collaborative learning platforms through our evaluation of the Verbal Autopsy with Participatory Action Research (VAPAR) programme.
Methods and analysis The evaluation takes a participatory approach, focussed on joint learning on whether and how VAPAR contributes to its aims, and what can be learnt for this and similar settings. A realist-informed theory of change was developed by the research team as part of a broader collaboration with other stakeholders. The evaluation will draw on a wide variety of perspectives and data, including programme data and secondary data. This will be supplemented by in-depth interviews and workshops at the end of each cycle to probe the different domains, understand changes to the positions of different actors within the local health system and feedback into improved learning and action in the next cycle. Quantitative data such as verbal autopsy will be analysed for significant trends in health indicators for different population groups. However, the bulk of the data will be qualitative and will be analysed thematically.
Ethics and dissemination Ethics in participatory approaches include a careful focus on the power relationships within the group, such that all groups are given voice and influence, in addition to the usual considerations of informed participation. Within the programme, we will focus on reflexivity, relationship building, two-way learning and learning from failure to reduce power imbalances and mitigate against a blame culture. Local engagement and change will be prioritised in dissemination.
Original languageEnglish
Article numbere036597
Number of pages10
JournalBMJ Open
Issue number2
Early online date4 Feb 2020
Publication statusPublished - 2020

Bibliographical note

The research is supported by the Health Systems Research Initiative from Department for International Development (DFID)/ Medical Research Council (MRC)/Economic and Social Research Council (ESRC) (MR/N005597/1, MR/ P014844/1), South African Department of Science and Innovation, the University of the Witwatersrand, and the Medical Research Council, South Africa, and previously the Wellcome Trust, UK (grants 058893/Z/99/A; 069683/Z/02/Z; 085477/Z/08/Z; 085477/B/08/Z).


  • health policy
  • health services administration & management
  • international health services
  • organisation of health services
  • primary care
  • public health


Dive into the research topics of 'Verbal autopsy with participatory action research (VAPAR) programme in Mpumalanga, South Africa: protocol for evaluation'. Together they form a unique fingerprint.

Cite this