Impact in Qualitative Research: Kylee Trevillion - A reflection on using co-production approaches in mental health research

Headshot of Kylee Trevillion

Kylee Trevillion is a senior lecturer at the IoPPN at King's. She is a mixed-methods researcher who specialises in women’s mental health, conducting research on practice and policy responses to violence and abuse in the context of mental illness, and perinatal mental health. Here, she reflects on the process of co-production in research, and how it can lead to richer, and higher impact, findings. 

Recently, I had the privilege of leading a co-produced qualitative research study that explored how community services can best support the needs of people who: (1) have been given a diagnosis of a ‘personality disorder’, (2) have used services for ‘personality disorder’, or (3) appear to have similar needs (you can read the full paper here). In this blog, I will outline and reflect on the process of doing this study and how the co-production approach resulted in a richer and more impactful study.

Co-production in research and how it was applied in this study

My colleagues at the NIHR Mental Health Policy Research Unit and I worked with a group of experts by experience and occupation to design and deliver a qualitative interview study which explored peoples’ expectations and experience of community service use. This was my first experience of co-producing a research study and I was grateful to have support and direction from lived experience academics (i.e. Sarah Carr, Vicky Nicholls) and a qualitative methodologist with extensive involvement experience (i.e. Nicola Morant).

Co-production can be defined as research that brings together experts by experience, experts by occupation and researchers, who work together and share power and responsibility to form equitable partnerships on a study, from start to finish. Co-production approaches take the position that those affected by the research have skills and knowledge of equal importance to researchers and are best placed to design and deliver it. This approach has been shown to result in a range of positive research impacts, including better formulation of research questions, increased recruitment, improved quality and validity of research, and the generation of data that is more relevant and appropriate to service users and clinicians (1-5).

In addition, co-production approaches foster two-way learning between researchers and experts by experience and occupation, and can increase experts’ sense of self-confidence, empowerment and evidence-based knowledge (5). This approach requires careful thought and attention at the planning stages, however, to avoid potential negative impacts on experts (e.g., feelings of frustration about limited opportunities to influence the research, confusion about clarity of role, time, and financial burdens of involvement) (5).

Co-production approaches foster two-way learning between researchers and experts by experience and occupation, and can increase experts’ sense of self-confidence, empowerment and evidence-based knowledge

Holding in mind the points described above, we involved experts by experience and occupation early on in the research planning stages of our work; we developed Terms of Reference which clearly outlined role expectations; we established regular expert working group meetings, which met throughout the study process, and was chaired by one of our experts and not the research team; we provided financial support to our experts for their involvement, and we were flexible in our expectations of experts’ time commitments to the study.

With the aim of increasing experts’ general knowledge and understandings of the topic area, we ensured that our expert working group meetings focused not just on the research activities at hand but also included educational components, via presentations from external speakers about wider research and policy developments in the field. In addition, myself and colleagues developed and delivered training to our experts by experience on qualitative research methodologies and practices, including qualitative research theory, interviewing skills and qualitative analysis training. Our experts by experience then took the lead on conducting the qualitative interviews with service users, and in analysing the data. In addition, our research team and experts by experience and occupation all worked together to synthesise the study findings and to write up the work collectively.

Examples of co-production activities applied in this study

One of the first co-produced exercises that our team undertook was to create a working definition to describe the population under study. This priority arose through team discussions about the considerable stigma attached to the label ‘personality disorder’ and the associated harms of the diagnosis, which are identified by service users and clinicians alike. Indeed, the diagnosis is often critiqued as being stigmatising and associated with an absence of hope and of progress in the delivery of healthcare. We acknowledged that although some service users find the diagnostic label helpful in explaining the nature of their needs, and it has had a role in ensuring consistency in research, many feel it is unhelpful and they do not identify with it.

We, therefore, co-produced an alternative working term - complex emotional needs – and we used this throughout the study, including in our study advertising materials. Through its application, we felt the revised terminology helped promote diversity in the sample - as it encouraged people who disliked the clinical label to come forward and share their treatment experiences. We recognise the term complex emotional needs has its limitations with respect to being over-generalised, and we acknowledge that further co-production work is required to develop new ways of describing this need. We felt, however, that it was important to establish an alternative working term to show that we recognised the critiques of the diagnosis.

We co-produced an alternative working term - complex emotional needs - and used this throughout the study. We felt this helped promote diversity in the sample - as it encouraged people who disliked the clinical label to come forward and share their treatment experiences.

Co-producing work with experts by occupation and experts by experience helped our research approaches be more appropriate and innovative in other ways too. For instance, our main recruitment strategies involved advertising through the networks of relevant third-sector organisations, via posts on social media platforms, and via online lived experience communities/networks. Our expert members guided us on the most appropriate third-sector organisations and online communities/networks to engage with and helped connect us with these groups. With their involvement, we ensured that we engaged groups that are often under-represented in research (e.g. ethnic minorities, traveller communities) and helped us achieve diversity in our sample with respect to ethnicity, age, sexuality, and type and extent of service use.

Through applying a co-production approach, we created a qualitative paper that generated richer and more impactful insights than could have been developed without the involvement of experts. A key component of this was having the experts by experience lead the qualitative interviews with service users and the qualitative analysis. Indeed, I believe we would not have produced such profound interview data if academic researchers had led the interviews.

Since its publication in January of this year, the paper has not only been well received by the academic community but has received important public attention by mental health charities (e.g. the Centre for Mental Health), mental health online forums (e.g. The Mental Elf) and policy makers alike. Some of our experts by experience and occupation also worked with us to deliver an online public event to share the findings of this work, and other work we have done in this area, which has helped to foster the public profile of our work (such as in this Mental Health Question Time event). I hope my reflections on the benefits of active expert involvement, and the application of co-production approaches, will inspire others to adopt a similar approach in their work.


1. Domecq J.P., Prutsky G., Elraiyah T., Wang Z., Nabhan M., Shippee N. et al. Patient engagement in research: a systematic review. BMC Health Serv Res. 2014.

2. Ennis L. & Wykes T. Impact of patient involvement in mental health research: longitudinal study. Br J Psychiatry. 2013;203(5):381–6

3. Staley K, Kabir T, Szmukler G. Service users as collaborators in mental health research: less stick, more carrot. Psychol Med. 2013;43(6):1121–5

4. Wykes T. Great expectations for participatory research: what have we achieved in the last ten years? World Psychiatry. 2014;13(1):24–7 10.1002/wps.20086

5. Russell, J., Fudge, N. & Greenhalgh, T. The impact of public involvement in health research: what are we measuring? Why are we measuring it? Should we stop measuring it? Commentary: Research Involvement and Engagement. 2020.