Enhanced or Hindered Research Benefits? A Realist Review of Community Engagement and Participatory Research Practices for Non-Communicable Disease Prevention in Low- and Middle-Income Countries

"[I]n research contexts characterised by power inequalities, as is typical in global health, it is important to critically assess whether, how, why and for whom engagement or participatory approaches are beneficial or burdensome."

Community engagement in research involves building and maintaining relationships between research actors and intended beneficiaries. Participatory research similarly invites contributions to the research process itself (e.g., formulating research questions, interpreting findings), as opposed to merely treating participants as data sources. Both community engagement and participatory research are widely used and considered important for ethical health interventions and research. Based on calls to unpack their complexity and observed biases in their favour, the authors of this paper conducted a realist review to generate an understanding of how community engagement enhances or hinders the benefits of non-communicable disease (NCD) research and interventions in low- and middle-income countries (LMICs).

Realist methodologies are informed by realist philosophy of science and adopt theory-driven approaches to examine the interplay between contextual factors and mechanisms of change in producing outcomes. The researchers formulated six theories about NCD-focused community engagement and participatory research to synthesise insights about contexts, strategies, mechanisms, and outcomes. These theories were based on existing literature (17 quality-appraised articles describing 15 studies), as well as realist interviews with three researchers, one community activist, and one representative of local government with experience of research encounters.

The six theories, which are visually presented in figures 2-7 in the paper, are connected to specific contextual factors but are not exclusively tied to a single setting or intervention, capturing variation across settings. In short:

• Benefit-enhancing mechanisms include community members' agency to contribute meaningfully being realised, a shared understanding of the benefits of health promotion, communities feeling empowered, and community members feeling solidarity and unity. As an example of the latter: One of the six theories holds that, if there is NCD-related stress and stigma for which facilitated community groups develop their own solutions, then stigma and stress are reduced because community members feel solidarity and unity (M). Solidarity was a hypothesised mechanism proposed by authors reporting on a participatory learning and action (PLA) approach to diabetes prevention in Bangladesh. The participatory and empowering nature of the group approach involves addressing NCDs such as diabetes as shared challenges, rather than individual, stigmatised problems. More generally, community-based participatory approaches can generate solidarity through collective problem-solving.
• Benefit-hindering mechanisms include community members' agency remaining unrealised and participation being driven by financial motives or reputational expectations. As an example of the former: As one of the six theories elucidates, institutional barriers are often associated with the mechanism of unrealised agency being activated and community interest or preferences being overlooked. These scenarios can include ethics boards not accommodating participants' preferred involvement or researchers not having the necessary skills, support, or interest in engagement. Outcomes of these pathways may include limited future involvement in research and missed opportunities for engaged or participatory research. As this example illustrates, the review set out to understand how, why, for whom and under what circumstances community engagement and participatory research approaches work or do not work. The review captures some of the power dynamics and institutional factors involved in both facilitating or blocking beneficial outcomes.

Recommendations for each uncovered mechanism include:

1. Community members' agency to contribute meaningfully is realised: Power sharing should be pursued where feasible, welcoming views from as early as possible where the research agenda or intervention focus can still be adjusted.
2. There is shared understanding of the benefits of health promotion: Research or intervention teams should support affected communities in reaching a shared understanding about divisive or stigmatising aspects of health promotion.
3. The community feels empowered: Research teams should foster a sense of empowerment in affected communities with evidence and be available to answer health-related questions and concerns, while emphasising the communities' own expertise and agency in solving problems affecting them.
4. Community members feel solidarity and unity: Public health issues should be presented as shared challenges to be addressed together, while avoiding assumptions about "communities" defined by researchers already being cohesive or harmonious entities. Consider partnering with existing groups and self-identifying communities where relevant and possible.
5. Community members' agency remains unrealised: Research teams should be realistic and transparent about what can be accommodated, pursuing institutional flexibility where possible.
6. Participation is driven by financial motives or reputational expectations: Research teams should manage expectations as much as possible, perhaps through cultivating partnerships or relationships where non-academic actors can understand the realities and constraints of academic research. The likely risk of undue inducement needs to be taken particularly seriously whenever engaging with communities experiencing, or with a long history of, structural inequality or any form of oppression.

According to the authors, these mechanisms relate more to social dynamics than a specific health theme and can thus potentially inform health research beyond NCDs. Nevertheless, it important to note that there are unique and context-specific aspects of NCD research and interventions, and these do need to guide the approaches taken to community engagement. For example, NCD-related stigma or stress may not be relevant or be similarly expressed across settings, but researchers need to be mindful of the nuances of lived experience and cultural dimensions (e.g., gendered barriers to healthful behaviours) when it comes to NCDs.

In conclusion, this review challenges assumptions about community engagement and participatory research being solely beneficial in the context of NCD prevention in LMICs. Helpful pathways can be enhanced (maximising benefits) and harmful pathways avoided (minimising harm) by addressing institutional inflexibility and researcher capabilities, managing expectations on research, promoting solidarity in solving public health challenges, and sharing decision-making power.