CBPR Partnership Agreements: Principles, Governance and Equitable Authorship

Community-based participatory research (CBPR) is judged, in practice, on whether its equitable-partnership claims are backed by real governance instruments — not on how the study describes itself in its introduction section. This guide covers the practical side: what belongs in a CBPR partnership agreement, how a community advisory board’s authority should actually be structured, how to write authorship criteria that hold up before data collection starts, and how to handle data-ownership terms. For the definition of CBPR itself, its nine founding principles, and how it differs from participatory action research (PAR) and community-engaged research more broadly, see CASRAI’s CBPR dictionary entry.

From principle to instrument

Israel, Schulz, Parker and Becker’s nine CBPR principles (1998) — recognizing community as a unit of identity, building on community strengths, facilitating equitable partnership across all phases, fostering co-learning, integrating knowledge with action, attending to social inequalities, working cyclically and iteratively, addressing health ecologically, and disseminating findings to all partners — are widely cited and just as widely under-implemented, because they describe an orientation rather than a set of enforceable commitments. A funder, an IRB, or a skeptical community partner cannot verify an orientation. What they can verify is a signed agreement, a named advisory body with defined authority, and an authorship policy that predates the data. Each principle below maps to a concrete instrument, not a restatement of the principle itself:

• Equitable partnership across all phases → a written partnership agreement or memorandum of understanding (MOU) specifying who has decision authority at each phase (question formulation, instrument design, recruitment, analysis, dissemination), not just at recruitment and dissemination, which is where token “community involvement” usually stops.
• Co-learning and capacity building → budget line items and named roles for training community members as research staff or co-investigators, not an unfunded aspiration in the narrative.
• Integrating knowledge and action for mutual benefit → a benefit-sharing agreement specifying how findings, products, or any commercial value flow back to the community, distinct from a one-time thank-you acknowledgment.
• Cyclical, iterative process → a renewal or review clause in the agreement itself, since a single static MOU signed at grant award rarely survives a multi-year project unchanged.
• Disseminating findings to all partners → an authorship and co-authorship policy, addressed in detail below, plus a community-review step before external release of results.

A study that satisfies none of the instruments above but describes itself as CBPR in a grant narrative is functioning as community-engaged research at best — a real distinction, not a technicality, and reviewers and community partners increasingly know the difference.

The partnership agreement: what actually belongs in it

A CBPR partnership agreement is not a boilerplate MOU restating the study’s aims. It is a governance document, ideally drafted jointly and signed before data collection begins, that should address:

• Roles and decision authority — who has final say (not just consultative input) at each project phase, and what happens when the community partner and the research team disagree.
• Scope, duration and renewal — what the agreement covers, its term, and the process for revisiting it as the project evolves (principle: cyclical and iterative).
• Budget control — whether the community organization holds or co-holds funds, receives a subcontract, or is paid a fee-for-service; each has different implications for who is accountable to the funder and who can direct spending.
• Data ownership and access — addressed in its own section below, but the agreement is where this is actually specified, not left implicit.
• Authorship and publication rights — the authorship criteria the partnership will use (see below), who reviews manuscripts before submission, and whether the community partner can require changes or withhold approval to publish.
• Ethics oversight — how IRB review and informed consent processes are handled, and whether a tribal or community-level review body has an independent approval role alongside the IRB.
• Dispute resolution and exit terms — what happens if the partnership needs to end before the study does; without this, disputes tend to get resolved unilaterally by whichever party controls the funds.

None of this needs to be lengthy or legally exotic — a two- or three-page MOU covering these points, reviewed and re-signed at defined intervals, does the governance work that a narrative description of “community partnership” cannot.

The community advisory board: real authority, not a rubber stamp

A Community Advisory Board (CAB) is the most common governance mechanism for translating the partnership agreement into ongoing oversight, but a CAB’s value is entirely a function of what authority it actually has, not its existence. Questions worth answering explicitly in the CAB’s charter:

• Does the CAB approve protocol changes and instrument revisions, or is it consulted after the fact?
• Does the CAB have sign-off authority on how results are framed and disseminated, including in press releases and public-facing summaries, or only on the peer-reviewed manuscript?
• How are members selected, and by whom — is composition controlled by the research team (a common source of token boards) or by the community partner organization?
• What are the terms, turnover, and compensation for CAB members’ time?

A CAB that reviews recruitment flyers but has no role in analysis or dissemination decisions is providing patient and public involvement (PPI)-level input, which is legitimate on its own terms, but it is not the equitable, shared decision-making CBPR claims. The distinction matters for funders and reviewers increasingly asking projects to document CAB authority specifically, not just CAB membership.

Equitable authorship criteria

Authorship in any biomedical or health-sciences publication is still governed by the ICMJE’s four-part test: substantial contribution to conception/design or data acquisition/analysis/interpretation; drafting or critically reviewing the work; final approval of the published version; and agreement to be accountable for the work’s accuracy and integrity. ICMJE explicitly leaves the judgment of who meets that test to the author group itself, not to journal editors — which is exactly where a CBPR partnership needs to do deliberate work up front, because the test’s language was not written with a community co-researcher’s typical contributions in mind.

The practical problem: activities central to a CBPR study — leading recruitment within a trusted community network, providing the cultural or linguistic framing that makes an instrument valid, running member-checking sessions that materially change how findings are interpreted, or reviewing a draft for community-relevant accuracy — can each satisfy ICMJE’s test (they are substantial contributions to data acquisition/interpretation, or to critical review), but partnerships often default to acknowledging community contributors instead of naming them as authors, either from unexamined habit or because the contribution wasn’t framed against the test at all. The fix is procedural, not aspirational:

• Write the authorship criteria into the partnership agreement before data collection, mapping the ICMJE test explicitly onto the roles the community partnership will actually perform, so the question isn’t relitigated manuscript by manuscript under submission-deadline pressure.
• Distinguish authorship from acknowledgment deliberately, and document why a given contributor lands on one side of that line — “the community wasn’t as involved in analysis” is a defensible reason if true; “we didn’t think to ask” is not.
• Decide author order and corresponding-author responsibilities as part of the same up-front process, since these carry real career and credibility weight and are harder to negotiate fairly after a paper is drafted.
• Use CRediT-style role documentation to make community contributions visible even where a contributor’s overall involvement doesn’t clear the full ICMJE bar for authorship — see CASRAI’s CRediT for non-traditional contributors guide, which maps CRediT’s fourteen roles specifically against community co-researchers, PPI partners, and other non-traditional contributors, and names where the standard roles fit cleanly (investigation, data curation, validation) versus where they don’t (conceptualization, methodology, project administration usually require genuine co-design to justify).

For the general mechanics of writing contribution statements once authorship is settled, see CASRAI’s author contribution statement guide.

Data-ownership and data-sharing agreements

Data ownership is the clause most often left implicit in CBPR partnerships, and the one most likely to cause conflict once a study ends and secondary-use or data-sharing decisions come up. Models in practice range across a spectrum:

• Sole institutional ownership — the university or research institution retains ownership and controls all downstream access; common by default (because it requires no explicit negotiation) but the weakest fit for CBPR’s equitable-partnership standard.
• Joint ownership — both the institution and the community organization hold rights, with data-sharing and secondary-use decisions requiring agreement from both.
• Community ownership with research-use license — the community or community organization owns the data outright and grants the research team a license scoped to the current study’s stated purposes, with any further use requiring renewed permission.

Whichever model applies, the agreement should specify: who can access raw data after the study period ends; whether secondary analyses require the community partner’s approval or only notification; how data are stored and by whom; and what happens to identifiable or community-level data if the partnership dissolves. Where the research involves an Indigenous community’s data specifically, these terms intersect with Indigenous Data Sovereignty and Indigenous Data Governance principles, and consent processes should be evaluated against the Free, Prior and Informed Consent (FPIC) standard rather than treated as satisfied by a standard IRB consent form alone; any data touching a community’s traditional knowledge carries additional governance obligations beyond a conventional research data-ownership clause. These terms belong in the same document as — or are explicitly cross-referenced from — the study’s data management plan, so that data-governance commitments made to the community partner and the DMP submitted to a funder don’t silently diverge.

Recognizing a partnership that isn’t equitable in practice

A few patterns reliably indicate that “CBPR” describes an aspiration rather than an operating structure:

• Community involvement concentrated at recruitment and a final dissemination event, with no role in question formulation, instrument design, or analysis.
• A CAB that reviews materials but has no documented approval authority over protocol changes or how results are publicly framed.
• An authorship decision made after a manuscript draft exists, rather than criteria agreed before data collection.
• No written agreement at all — verbal understanding of “partnership” that isn’t specific enough to be checked against the study’s actual conduct.
• Data-ownership and secondary-use terms left unaddressed until a dispute forces the question.

None of these individually disqualifies a study from calling itself community-engaged, but each is a gap between how CBPR is typically claimed and how it is structurally verified — and the gap is exactly what funders, IRBs, and community partners increasingly ask to see closed with a document, not a description.

Frequently Asked Questions

What is a CBPR partnership agreement?

A written document, ideally an MOU signed before data collection begins, that specifies decision authority at each project phase, budget control, data ownership and access, authorship criteria, ethics oversight, and how disputes or an early end to the partnership are handled. It operationalizes CBPR’s equitable-partnership principle into terms that can actually be checked against the study’s conduct.

Who should be listed as an author on a CBPR publication?

The same ICMJE four-part test that governs any biomedical or health-sciences authorship decision applies: substantial contribution, drafting or critical review, final approval, and accountability. The CBPR-specific step is mapping that test explicitly onto community co-researcher roles — leading recruitment, providing cultural/linguistic framing, running member-checking, reviewing drafts — before data collection, rather than defaulting those contributors to an acknowledgment without checking whether their actual involvement meets the test.

Does a community advisory board have real decision-making power?

It depends entirely on the CAB’s charter. A CAB that only reviews recruitment materials after the fact is providing PPI-level input, not CBPR’s equitable shared decision-making. A CAB with documented sign-off authority over protocol changes and how results are disseminated is functioning at the standard CBPR claims. This should be specified explicitly, not assumed from the board’s existence.

Who owns the data collected in a CBPR study?

There is no single default — models range from sole institutional ownership to joint ownership to community ownership with a research-use license to the research team. The partnership agreement should specify the model explicitly, along with who can access the data after the study ends and whether secondary use requires the community partner’s approval.

How is CBPR different from participatory action research (PAR) or community-engaged research?

PAR is the broader methodological tradition of cyclical inquiry-and-action; CBPR is its health-research-specific application, formalized through Israel et al.’s nine principles; community-engaged research is a still-broader umbrella that may or may not reach CBPR’s specific standard of equitable, shared decision-making across all phases. See CASRAI’s CBPR dictionary entry for the full comparison.