APHRC’s Two Decades Expose Where Co-creation Works and Where It Breaks Down
Co-creation has become one of global health’s most attractive promises, yet practice often stops short of the claim. A reflective synthesis from the African Population and Health Research Center (APHRC) traces two decades of work across research, policy engagement, capacity strengthening, and rights-based advocacy, and the pattern is blunt: co-creation was real where knowledge was being made, but far weaker where authority, money, and institutional control were in play.
The paper defines co-creation across four dimensions: shared agenda setting, joint knowledge production, power sharing, and community leadership and ownership. It then separates co-creation from consultation, stakeholder engagement, and partnership, treating those as different grades of participation rather than interchangeable labels. A second analytic axis follows the journey from research to policy to advocacy, with attention to the condition of the transitions between those segments. Six cases were examined through that lens, spanning evidence generation, policy processes, capacity strengthening, and rights-based advocacy.
One sentence captures the central fault line: co-creation was strongest where ideas were exchanged and weakest where decisions were controlled.
Shared Agendas, Uneven Authority, and No Control Over Funding
Across the six cases, agendas were substantively shared in four, which suggests a meaningful degree of collaboration at the front end of the research process. Yet financial resources were shared in none of the cases, and that absence matters because funding control often determines who sets priorities, who can sustain participation, and whose labor is treated as optional. In practice, co-creation can look robust while still leaving the deepest levers of power untouched.
The paper’s graded assessment makes that imbalance visible. Joint knowledge production appears feasible within institutional research settings, where collaboration can be built into design, fieldwork, analysis, and interpretation. Power sharing is harder to realize because it requires more than inclusive meetings or advisory input; it demands structural redistribution of authority. Community leadership and ownership were substantive only in rights-based advocacy, a finding that points to a domain-specific ceiling on how far participation can travel.
That pattern is clinically and politically relevant for global health teams that use co-creation language without testing what kind of decision-making actually shifts. A project may invite community voices and still preserve a conventional hierarchy over budgets, timelines, publication decisions, and policy access. The result is a familiar asymmetry: communities help produce the evidence, but institutions retain control over the process and the output.
Breakpoints at the Research-Policy-Advocacy Interface
The sharpest failure did not occur inside individual projects. It appeared at the transitions between research, policy engagement, and advocacy, where co-creation was neither designed, evaluated, nor recorded.
Those transition points are easy to miss because they sit between formal stages. Research teams may document participation in study design, and advocacy teams may document community mobilization, but the handoffs between those domains often remain invisible. In the APHRC synthesis, those gaps mattered more than isolated shortcomings inside any single case. When a project moves from evidence generation to policy dialogue, or from policy engagement to advocacy, the social logic changes, the power dynamics change, and the risks of dilution or exclusion increase.
This is where many co-creation efforts lose continuity. A participatory study can generate credible findings, but if the move into policymaking is managed by a narrower set of actors, the original community contribution can disappear from the decision chain. Likewise, rights-based advocacy can be vibrant at the community level while the policy interface remains disconnected from the people whose experiences produced the demand for action. The synthesis argues that these transitions deserve the same design attention as recruitment, data collection, or dissemination planning.
Rights-Based Advocacy Carries the Strongest Community Leadership
Among the six cases, rights-based advocacy was the only domain in which community leadership and ownership were substantively realized. That distinction is not cosmetic. Advocacy grounded in rights often requires lived experience, local legitimacy, and moral authority in ways that conventional research processes do not. Community actors are not simply sources of data in that setting; they become central agents in defining the problem, framing the claim, and pushing for change.
By contrast, evidence generation and policy engagement were more likely to preserve institutional control even when they included community input. The paper’s framework makes this visible by separating participation into distinct dimensions rather than collapsing everything into a single label. A project can share an agenda without sharing authority. It can include communities in knowledge production without transferring ownership. It can engage policy actors without building a true bridge to advocacy.
That distinction has practical consequences for how funders, implementers, and researchers judge success. If co-creation is measured only by the presence of engagement activities, it will look more successful than it is. If it is assessed by whether communities help decide, direct, and own the work, the picture becomes harder, but more honest.
Governance, Funding, and Evaluation Need to Be Written Into the Design
The paper closes with practice guidance drawn from this institutional experience, emphasizing governance, funding, community roles, transitions, and assessment. The logic is straightforward: co-creation cannot be left to goodwill alone. It needs explicit structures that define who decides, who pays, who leads, and how the project moves from one phase to the next.
Assessment is part of that structure. Without a way to record the quality of transitions, the analysis of power sharing remains incomplete. Without a way to track funding distribution, the claim of shared ownership may conceal dependency. Without a way to define community roles in advance, participation can drift into consultation disguised as partnership.
The synthesis also stays careful about what it cannot establish. It is a single-institution insider account, which means it cannot prove that the same patterns hold everywhere. It also cannot settle whether other organizations, in different political or funding environments, would produce different distributions of power or more seamless transitions. Even so, the internal evidence is strong enough to expose a practical truth: co-creation is not one thing, and it is not equally achievable across all phases of a project.
For global health research, the hardest problem may not be getting communities to the table. It is deciding whether the table itself is shared, who owns the chair, and what happens when the discussion moves into the next room.