Table 3. Researchers and engagement practitioners.

Researchers and engagement practitioners
Clarity around engagement purposes and outcomes: our findings suggest a need for researchers and engagement practitioners to be clearer about the purpose of their engagement efforts, both the engagement outcomes being sought and their understanding of how engagement activities will contribute to these outcomes.	•   Being explicit about the purposes and anticipated outcomes of engagement strategies can inform planning and evaluation.
Clarity around ‘terms of engagement’ for research stakeholders: researchers and engagement practitioners need to be clearer about the degree of stakeholder decision rights and control throughout the research process, and the processes of stakeholder representation that will allow clear lines of accountability.	•   Clearly distinguishing decision rights around the focus, design or implementation of particular research studies, and where there is scope to inform wider institutional policy may help to manage expectations of local research stakeholders
Early engagement to develop relationships and negotiate engagement processes: ideally both the purposes and terms of engagement will be negotiated with stakeholders. This means investing the time, space and resources to identify relevant stakeholders, and build the relationships and rapport necessary to be able to productively discuss and agree research purposes and terms of engagement. This may need to be an iterative process and develop over time.	•   An inception phase for research should focus on understanding how research affects local stakeholders and on general relationship building    •   Rather than seeing particular engagement activities as a technical input, engagement should focus on development and monitoring of relationships over time through a range of engagement activities combined and adapted as appropriate    •   The time and opportunity costs involved for community members engaging with research need to be carefully considered
Emphasis on researcher responsiveness as well as provision of accurate information on research: in engagement strategies, structured methods for listening to and responding to research stakeholder concerns may be at least as important as provision of accurate and appropriately tailored information and communication about research.	•   Engagement activities should prioritise methods for listening and responding to stakeholder concerns - including formative research, dedicated spaces for raising concerns, and structured feedback from local research staff - with information provision incorporated into these processes.
Research Centres
Engagement at research programme level: programme-wide engagement not linked to particular research studies can support the relationship building that underpins effective engagement, and provide opportunities for community and stakeholder input into research and engagement related policy and practice over time, including around fair benefits for research stakeholders at the individual, community and health infrastructure level. Research centres should be more explicit about the purposes and opportunities of such engagement.	•   Research institutions should invest in programme wide engagement not linked to individual research studies and be explicit about the purposes of providing opportunities for relationship building and stakeholder input
Institutional support for local research and engagement staff: the complex and demanding roles for research fieldworkers and engagement staff need greater recognition and requisite training, support and supervision. Research institutions policies should reflect commitment to resourcing and developing engagement capacity. Integrating engagement more thoroughly into research programme management also provides the scope for responsive research implementation.	•   Experienced based training and supportive supervision for research fieldworkers at the research institution level can help them navigate some of the relational complexities of engagement    •   Professionalisation of frontline research and engagement staff roles with related career pathways is part of building engagement capacity    •   Dedicated work with researchers on their understandings and attitudes to CE at the institutional level
Input on research priorities: research designs may put limits on how and where research protocols can be influenced, but it is often possible to consult research stakeholders on aspects of research protocols. At the same time, there can be consultation on the thematic focus of research, and the role of responsive research designs, and there is potential to link to broader processes of public consultation on research priorities as part of national research agenda setting and initiatives (like COHRED’s research fairness initiative or the European Responsible Research and Innovation RRI initiative) 2 .	•   Research priorities and designs should be framed in relation to national research agendas and opportunities for stakeholder input linked to broader public consultation on such agendas
Broader political and ethical focus for health research engagement: CE processes may be unable to fully address the differences of wealth and power between researchers and local research stakeholders, but can take considered concrete steps to mitigate them. There may also be scope to better integrate individual health studies with wider efforts to strengthen health systems, and enhance health equity (see below).	•   Compilation of existing guidance and case studies on provision of individual and community level research benefits could inform careful consideration of the opportunity costs for local research stakeholders for particular studies    •   Clear procedural standards for community engagement may help monitor levels of meaningful input and control on the part of research stakeholders    •   Guidance for Ethical Review Board processes could explicitly address the influence of wider structural factors and draw on principles of social justice
Funders and policy makers
Greater integration of health research with LMIC health systems: research institutions and initiatives often bring considerable wealth and resources to LMIC settings where health systems are underdeveloped, with impacts on health care infrastructure and personnel, levels of health care provision, and health surveillance capacity. Further attention could be given to how the health research enterprise links to health system strengthening, and more explicit ways of monitoring this relationship, which may also contribute to improved ‘host country ownership’. Such an integration could strengthen the social value of research overall, and help provide a further rationale for more institutional support for engagement infrastructure and capacity in research centres.	•   Plan health research in a way that carefully considers opportunities for strengthening local health systems and minimises any potential harms, and monitors this relationship    •   Greater attention to policy engagement by both research institutions and funder could help with the above integration    •   Core funding for health research centres over the long term facilitates programme wide engagement and integration of research plans with local health facility strengthening 3
Greater role for social science research on health and health systems to complement and inform biomedical research: social science and empirical ethics research are important for understanding prevailing health beliefs and practices and the social and cultural context. It is also important for understanding the contours of health systems and how the international governance of health and health research limits or facilitates CE. In addition, social science expertise can help understand the relationship dynamics that engagement interventions seek to facilitate, and to support their design and evaluation. Recent work on the social determinants of health and health inequalities points to the intimate relationship between the environments in which people grow, live and work, inequalities of wealth and health, the degree of control people have over the fabric of their work and civic lives, and ultimate health outcomes. Greater investment in this broader health research agenda would help to maximise the gains from biomedical research.	•   Greater investment in applied social science studies to complement the biomedical research agenda should include research on relational dynamics of engagement, anthropology of health and health systems, the social determinants of health, and international health and research governance    •   Support the strengthening of the evidence on CE through dedicated resources for formative research, evaluation or implementation studies to accompany biomedical health research
Inception phases and flexibility in research funding: the importance of relationship development and early engagement identified above suggest a need for structured inception phases for partnership development and exploration of health research priorities, with funding underpinning them. There also needs to be scope for research budgets and protocols to accommodate stakeholder insights before they are finalized in a way that may preclude further input.	•   Funders should explicitly consider processes to allow for budget and protocol flexibility in response to CE and in which circumstances these apply (such as IDRC’s Research Quality Plus tool) 4    •   Funded inception phases of research can facilitate partnership development and consultation on research focus and design (such as in NIHR Global Health Research Grants) 5