By Katrina Messiha.
How can we ensure that the many people whose lives are shaped by homelessness, migration, poverty, trauma, mental illness, caring responsibilities, social isolation and other contributors to marginalisation are adequately represented and well engaged with in relevant research?
This is important because if some lives are missing from the evidence base, they may also be missing from the services, policies and practices built upon it. But what happens when researchers try to engage people whose previous encounters with healthcare, welfare, housing, immigration or other public systems, including research, may have been difficult, exhausting or even harmful?
The usual language used to describe such people is “hard to reach,” “seldom heard” or “difficult to engage.” These phrases may be commonly used and well intended, but they can hide a pressing question: Where does trust already exist and how can research begin there?
This i2Insights contribution locates trust as practical, relational and situated. It is built on learnings from a series of workshops focused on brain health (ie., engaging with stakeholders at elevated risk of dementia) in the United Kingdom, and has broader relevance for other research areas. I examine how research is often experienced, share insights from discussions with relevant communities and related stakeholders, and provide key lessons for researchers.
When research feels like scrutiny
A research invitation does not arrive in isolation. It is interpreted through people’s prior experiences of being believed or dismissed, helped or scrutinised, respected or diminished.
For researchers, this should give us pause. Methods that appear routine to us may not feel routine to the person being invited to take part. An interview may be intended as a respectful conversation, but may be experienced as another interrogation. A survey may be a data collection tool, but may resemble yet another eligibility form. An assessment may be scientifically necessary, but may feel like an examination in which failure carries stigma.
For those who have experienced fragmented care, long waits and processes that left them feeling scrutinised rather than supported, interviews, surveys and examinations could themselves be triggering, particularly for people already exhausted by repeated institutional questioning.
The topic of the research may also carry fear, stigma, shame and uncertainty. If we take dementia for example, it may be linked with weakness, embarrassment or loss of dignity. It may also be linked with loss of transport, digital exclusion, unstable housing, language barriers, fear of services or concern about how personal information will be used.
This means that the conditions under which those questions are asked matter.
What marginalised communities and supporters say
Communication, cultural sensitivity, dignity, personalisation, transparency, community integration, social value and meaningful incentives are conditions for building trust.
Trust exists in community cafés, familiar organisations, faith spaces, shared meals, reliable intermediaries and places where people are known as more than patients, participants or cases.
“Friendly places” are also an important way of maintaining contact with people whose circumstances may change frequently.
People living with instability and marginalisation may need researchers to adapt to their lives, rather than expecting their lives to fit the administrative needs of a study. For example, for someone experiencing homelessness, insecure housing, trauma, caring responsibilities or poor mental health, a formal appointment system may not be the most realistic or humane way to stay connected.
Lessons for researchers
Trust-building is not a soft addition to rigorous research. It is part of rigour when the aim is to understand lives that routine systems fail to capture. If participation is only feasible for people who are comfortable with institutional language, forms, appointments, professional hierarchies and repeated questioning, then the resulting evidence base will remain partial.
Inclusion has to begin before recruitment. Research teams should work with trusted community organisations before protocols are fixed. Community spaces should not be treated only as access routes, but as places where knowledge, relationships and legitimacy already exist.
Funding should include time for relationship-building, fair payment, accessibility, emotional support, shared feedback and sustained communication.
Methods should be co-developed so that interviews, surveys and assessments do not unintentionally reproduce forms of scrutiny that people already experience elsewhere.
Researchers should also report stakeholder engagement honestly. It may not be enough to state that communities were merely consulted. We should explain what people were able to influence in the research, what changed because of their engagement, what could not be changed and why. Without this transparency, participation risks becoming a procedural claim as opposed to a substantive shift in how knowledge is produced.
Conclusion
It is clear that trust matters. Not as a slogan, but as a condition for meaningful participation, better evidence and more humane policy and services. Trust often already exists, but not always where researchers first look for it. Researchers should therefore start not by asking why people are absent, but by asking where people already feel safe enough to be present.
What is your experience in working with marginalised communities? Do you have other lessons to share? Can you provide examples of effective implementation of the lessons provided here?
To find out more:
Messiha, K., Thomas, N., Brayne, C., Agnello, D. M., Delfmann, L. R., Giné-Garriga, M., Lippke, S. and Downey, J. (2025). Grey literature scoping review: A synthesis of the application of participatory methodologies in underrepresented groups at an elevated risk of dementia. BMC Medical Research Methodology, 25, 122: 1-13.
Websites for more information about the project context:
- Messiha, K. and Thomas, N. (n.d). Critical realist insights: Navigating interpretation and experience of co-development work in the ComPHAD project. Cambridge Public Health website. (Online): https://www.cph.cam.ac.uk/research/life-course-and-ageing/comphad-community-and-public-health-approaches-dementia-research-0.
- Cambridge Public Health (n.d). ComPHAD: Community and public health approaches to dementia research. Cambridge Public Health website. (Online): https://www.cph.cam.ac.uk/research/life-course-and-ageing/comphad-community-and-public-health-approaches-dementia-research.
- Thomas, N. (2024). ComPHAD: Community and Public Health Approaches to Dementia Research website. Padlet: Visual collaboration for creative work for education website. (Online): https://padlet.com/thenicolethomas/comphad-community-and-public-health-approaches-to-dementia-r-ic976x3mpievatcb.
Use of Artificial Intelligence (AI) Statement: Artificial intelligence was not used in the development of this i2Insights contribution or the work on which the contribution is based. (For i2Insights policy on artificial intelligence please see https://i2insights.org/contributing-to-i2insights/guidelines-for-authors/#artificial-intelligence.)
Biography: Katrina Messiha MSc is a Marie Skłodowska-Curie PhD Fellow in ‘Theory-based Principles for Co-creation in Public Health’ at Amsterdam University Medical Center, in the Netherlands. Alongside her doctoral work, she has contributed to policy, training and knowledge-exchange activities with organisations including EuroHealthNet, the University of Cambridge and the World Health Organization.
Katrina, this is beautifully articulated. I can feel, even between the lines, the care you are pointing toward in how we conduct our research.
In conversations about trust, I sometimes sense that there is a philosophical gap we haven’t fully addressed. It often feels as though trust is being defined as a form of approval. As in, “I will trust you if I see that you take my needs into account and respond to them”. This understanding has strong roots in the idea that greater inclusivity leads to greater trust.
But if we define trust slightly differently: as a relational stance in which one accepts vulnerability based on the belief that the other will act with integrity, care, or competence, even without full certainty. This second understanding still requires the criteria you highlight (inclusion, honest reporting, and genuine transparency), but it also recentres responsibility and accountability for care within the specific contexts we are examining.
This opens up further layers in the powerful question you raised: “Researchers should therefore start not by asking why people are absent, but by asking where people already feel safe enough to be present.”
It invites us to ask, too: Do people want to carry the intellectual and professional labour of compensating for the lack of responsibility and accountability they encounter in doctors, policymakers, and other institutional actors?
Thank you for this
Dear Hussein,
Thank you very much for this thoughtful and generous response. I really appreciate the way you have extended the point, particularly the distinction you draw between trust as approval and trust as a relational stance involving vulnerability, integrity, care and competence.
I agree that this distinction is important. If trust is understood only as something generated through inclusion, we risk placing too much emphasis on participation itself, without sufficiently attending to the responsibilities of those who hold institutional, professional or epistemic power. Inclusion, transparency and honest reporting matter deeply, but they cannot substitute for accountability, nor should they require communities to continually absorb the labour of compensating for institutional failures.
Your final question captures this tension very powerfully. It pushes the discussion beyond “how do we bring people in?” towards a more ethically demanding question of what conditions, responsibilities and forms of care must be in place before participation can be meaningful, safe or even desirable.
Thank you again for engaging with the post in such a careful and generous way.