Community-Based Participatory Research: Who Decides the Question

Community-based participatory research is not a gentler way to run a study. It is a claim about authority: that the people who will live with a finding have standing to decide what gets asked in the first place.

The American human-subjects system was built to answer a different question. The Public Health Service syphilis study at Tuskegee followed Black men in Macon County, Alabama from 1932 to 1972 and withheld treatment that became available partway through. Congressional hearings followed the study's exposure in 1972, and the National Commission that resulted produced the Belmont Report in 1979 — respect for persons, beneficence, justice — which remains the moral architecture underneath the regulations every investigator works within.

Belmont protects the individual. It requires that the person in front of you understand what will be done to their body and agree to it. It has nothing to say about who chose the question.

That gap is not a technicality. It is where most of the remaining harm in community research now lives.

A study can satisfy every requirement of informed consent and still take from a community something the community never offered. Every form signed, every risk disclosed, every review board box checked, and the research question still arrived from outside, addressed a problem the investigators found interesting, and produced a paper nobody in the community will ever read. Consent governs the body. It does not govern the agenda.

Barbara Israel and colleagues at the University of Michigan gave the corrective its working definition in a 1998 review in the Annual Review of Public Health. Community-based participatory research is an equitable partnership in which community members participate in every phase — question formulation, design, data collection, interpretation, dissemination — and in which findings return to the community in a form the community can use. Meredith Minkler and Nina Wallerstein assembled the field's standard text around that definition in 2003, and it has been through multiple editions since.

The phrase that carries the weight is every phase, and the phase that decides everything is the first one.

Most academic centers practice a version that starts later. The aims are written. The budget is built. The methods section is drafted and the power calculation is done. Then someone books a conference room, invites four people from a neighborhood organization or a tribal health board, calls them a community advisory board, and asks for help with recruitment. That is not partnership. That is subcontracting the hardest part of the work, which is earning access, after every intellectual decision has already been made.

Where the Method Came From

Kurt Lewin published "Action Research and Minority Problems" in the Journal of Social Issues in 1946. The argument was that research and action belong in the same cycle, conducted with the group whose problem it is rather than about them. Lewin was a social psychologist working on intergroup conflict, and he was proposing something structurally strange to the science of his period: that the people being studied were competent participants in the study's design.

Paulo Freire pushed the argument further in Pedagogy of the Oppressed, published in Portuguese in 1968 and in English in 1970. Freire's target was what he called the banking model of education, in which the teacher deposits knowledge into a passive student. Against it he set dialogue, and against the category of subject he set co-investigator. The people whose lives are being analyzed are already analysts of those lives. A method that treats them as data is throwing away the analysis.

Orlando Fals Borda took this into research practice in Colombia. Working with peasant organizations, he developed investigación acción participativa, and the volume he edited with Muhammad Anisur Rahman in 1991, Action and Knowledge: Breaking the Monopoly with Participatory Action-Research, named the target in its title. Fals Borda's claim was that the university's exclusive right to say what counts as a finding is a political arrangement rather than an epistemic necessity.

North American public health synthesized these lines into something fundable. Caroline Wang and Mary Ann Burris published the photovoice method in Health Education and Behavior in 1997, in which participants photograph their own conditions and hold the interpretive frame over what the photographs mean. Israel's group built the Detroit urban research partnership. Wallerstein and Duran argued in a 2010 American Journal of Public Health supplement that participatory design is not only more just but more likely to produce interventions that survive contact with the setting. The Patient-Centered Outcomes Research Institute, created by federal statute in 2010, wrote stakeholder engagement into its funding conditions.

The lineage matters because it tells you what the method is for. It did not come out of a methods department. It came out of organizing.

What the Havasupai Case Established

Around 1990, researchers from Arizona State University collected blood samples from members of the Havasupai Tribe, whose land sits at the bottom of the Grand Canyon, for a study of type 2 diabetes. The tribe had asked for help with diabetes. That was the arrangement as the tribe understood it.

The samples were subsequently used in research on schizophrenia, on consanguinity, and on population migration, including work bearing on the tribe's own account of its origin. Tribal members learned of the additional studies years later. Litigation followed. In 2010 the Arizona Board of Regents settled, compensated the plaintiffs, returned the remaining samples, and the tribe barred the university's researchers from its land.

The consent forms had contained broad language about the study of behavioral and medical disorders. Whether individual consent was technically satisfied is a question lawyers argued at length. It is also the wrong question. The tribe's objection was never that individuals had been deceived about a needle. It was that a community authorized one inquiry and a different inquiry was conducted with what the community gave. No mechanism existed for a community to hold that authority, or to withhold it, or to revoke it, because Belmont never conceived of the community as a party to the agreement.

What followed the case is the beginning of that mechanism. Tribal research review boards now hold genuine approval authority over work conducted on tribal land and with tribal members, and their review asks questions no university board asks: what does the nation get, who holds the data, what happens to the samples when the study ends. The CARE Principles for Indigenous Data Governance, released by the Global Indigenous Data Alliance in 2019, set out collective benefit, authority to control, responsibility, and ethics as governance requirements. They were written deliberately as a complement to the FAIR principles, which optimize data for reuse and say nothing at all about who governs the reuse.

Linda Tuhiwai Smith, the Māori scholar whose Decolonizing Methodologies appeared in 1999, made the broader version of this argument. Research as a practice arrived in Indigenous communities alongside colonization, carried its assumptions in with it, and cannot be repaired by making the researchers more polite. It has to be re-founded on different terms about who authorizes inquiry.

Medicine tends to file all of this under ethics, which in practice means under compliance. That filing is a mistake.

Why This Is a Methods Argument

Sandra Harding published Whose Science? Whose Knowledge? in 1991 and named the concept that makes participatory design a rigor claim rather than a politeness claim. Conventional objectivity works by stripping out the investigator's position. Harding's objection is that this fails on its own terms, because the assumptions that survive peer review are precisely the ones the entire peer community shares and therefore cannot see. Removing the individual investigator's bias leaves the field's collective bias completely untouched.

Her alternative, strong objectivity, starts inquiry from the lives of people outside the dominant position. Not because their view is purer. Because it makes visible the background assumptions that the dominant position holds invisibly.

Issue 012 of this publication argued that the standpoint tradition Risjord introduced to nursing — Harding, Haraway, Collins — carries methodological consequences and not only philosophical ones. Community-based participatory research is one of those consequences. It is what strong objectivity looks like when it is written into a protocol and a budget.

The practical form of the argument is unglamorous, and it arrives first as a measurement problem. Issue 009 of this publication argued that clinical plans routinely ignore the determinants of adherence and then attribute the failure to the patient rather than to the plan. The research version of that error is the more durable one, because it gets frozen into an instrument. A team that scores a low-income clinic population on a knowledge questionnaire will find a knowledge deficit, because a knowledge deficit is the only thing a knowledge questionnaire can find. The pharmacy that closes at five, the bus route that does not connect the clinic to the pharmacy, and the monthly cost the patient never raised do not show up as low scores. They do not show up at all. Nobody who lived with those constraints was in the room where the instrument was chosen.

That is not a values failure. It is measurement error, introduced at the design stage, invisible to every statistical check downstream of it.

Nursing was positioned to receive this argument in a way medicine was not. Doctoral training in nursing has treated qualitative and mixed methods as first-class designs for fifty years, and nursing PhD students defend qualitative dissertations without apologizing for them. Margarete Sandelowski's methodological writing is the kind of internal rigor argument a discipline only produces when it takes a method seriously enough to police it.

A practice lineage runs alongside the research one. Lillian Wald opened the Henry Street Settlement in New York in 1893 and built public health nursing around clinicians who lived in the neighborhood they served, which put the definition of the problem inside the neighborhood rather than at the hospital. Madeleine Leininger's ethnonursing, which Issue 019 of this publication took up as a method and not a sensibility, made the same move inside a research design: it establishes a community's own account of health before it proposes that anything be changed. Both were built by nurses, and both are what medicine is reaching for when it says community engagement. The vocabulary transferred. The method did not.

This is not a claim that nurses run community research well and physicians run it badly. Plenty of nurse-led studies are as extractive as anything medicine has produced. The claim is narrower and I think it holds: the methodological vocabulary that participatory research requires was already sitting in nursing's doctoral curriculum, and it was not sitting in medicine's.

What It Costs

The method is slow in a way the funding structure punishes. Building the relationship that makes a genuine partnership possible takes years before an aim is written, and no grant mechanism pays for those years. The extractive design is not just easier. It is the one the funding cycle was built to reward.

Participatory studies also rarely produce the effect estimate that changes a guideline. A community-designed intervention tested in one setting with a sample the setting can supply is not the trial that moves a recommendation, and pretending otherwise does the method no favors. The randomized trial is not the adversary here. A trial answers whether an intervention works. It cannot tell you whether that intervention was the one that mattered. Those are different questions and they require different designs, and a research program that runs only the first kind will keep producing precise answers to questions nobody in the affected community would have asked.

The hardest problem is internal. A community is not one thing. An advisory board can reproduce the community's own hierarchy exactly, so that whoever already holds standing speaks for people who do not, and the study arrives at consensus that was never contested because the contesting parties were never at the table. Patricia Hill Collins's matrix of domination, which Issue 012 took up, operates inside a community as much as between a community and a university. I do not know how to solve that. What can be done is smaller: name in the methods section whose voice the board carried and whose it did not, so the limitation is on the page where it can be argued with.

The method demands listening, and listening is the part that is easiest to counterfeit. A signature, a quarterly advisory meeting, a line in the acknowledgments — all of it can be produced without a single research decision changing hands. The test is not about tone or intention, and it is not hard to apply. It sits on the aims page. Who wrote it, when, and who had the standing to say no.

About the author

Peter Schindler, MD, PhD is an Assistant Professor of Medicine and Associate Program Director of the Community Health Center Family Medicine Residency Program at the University of Nebraska Medical Center. He practices at Winnebago Comprehensive Health System, OneWorld Community Health Center, and Nebraska Medicine. He completed a Primary Care Research Fellowship at McGill University and holds a BSN from the University of Wisconsin-Oshkosh, an MS and PhD in nursing from Emory University's Laney Graduate School, a Diploma in Tropical Medicine from the Liverpool School of Tropical Medicine, and an MD from the Medical College of Wisconsin-Green Bay. He is board certified by the American Board of Family Medicine. The Interprofessional publishes new essays every week at the intersection of medicine, nursing, and the clinical knowledge that lives between them.

Disclaimer. The views in The Interprofessional are Peter Schindler's own and do not represent the official positions of the University of Nebraska Medical Center, Winnebago Comprehensive Health System, OneWorld Community Health Center, Nebraska Medicine, or any other affiliated institution.