Background
Patient and public involvement (PPI) increasingly features in shaping the design and conduct of mental health research, both in the UK and internationally. It is well established in the Global North, driven, in part, by major funder requirements, mandated reporting of PPI in prominent journals and good practice recommendations from mental health service users (Faulkner et al., Reference Faulkner, Crepaz-Keay, Kalathil, Griffiths, James, Perry and Yiannoullou2015; NIHR, 2020; Perot et al., Reference Perôt and Chevous2018; Wicks et al., Reference Wicks, Richards, Denegri and Godlee2018). Within the UK, the National Institute for Health Research (NIHR) established INVOLVE in 1996 (now the Centre for Engagement and Dissemination) to support active involvement in research, and many resources have since been published, supporting PPI and co-production (NIHR, 2019). PPI is increasingly being employed in lower and middle-income countries, often in collaboration with those in the Global North (Cook et al., Reference Cook, Siddiqi, Twiddy and Kenyon2019; Florence et al., Reference Florence, Bocalini, Cabrini, Tanzi, Funaro, Jordan and Yasui2023; Semrau et al., Reference Semrau, Lempp, Keynejad, Evans-Lacko, Mugisha, Raja and Hanlon2016). However, international differences in funders’ emphasis, research cultures, and ethics review processes can all influence how PPI is implemented and reported (Biddle et al., Reference Biddle, Gibson and Evans2021; Cook et al., Reference Cook, Siddiqi, Twiddy and Kenyon2019; Kaisler et al., Reference Kaisler, Kulnik, Klager, Kletecka-Pulker, Schaden and Stainer-Hochgatterer2021; Lang et al., Reference Lang, King, Jenkins, Boddy, Khan and Liabo2022; Semrau et al., Reference Semrau, Lempp, Keynejad, Evans-Lacko, Mugisha, Raja and Hanlon2016).
PPI can broadly be conceptualized into one of two paradigms (Beresford, Reference Beresford2002; Caron-Flinterman et al., Reference Caron-Flinterman, Broerse and Bunders2007; Madden & Speed, Reference Madden and Speed2017). Democratic approaches emphasize the right of those affected by research to be involved, particularly in decision-making. Consumerist approaches, which are more common within university settings, engage public insight to improve the relevance and quality of research. The latter are more consultative processes, with little consideration of the power distribution within the research process (Beresford, Reference Beresford2002). The conceptual shift toward co-production and survivor research reflects a more democratic rationale, with an emphasis on the examination of power dynamics and ownership of research (Beresford & Boxall, Reference Beresford, Boxall and Staddon2015; Hickey et al., Reference Hickey, Brearley, Coldham, Denegri, Green, Staniszewska and Turner2018; Sweeney, Reference Sweeney2016). Definitions vary greatly by time and place, with involvement ranging from consultation to partnership and full control. For the purposes of this review, we adopted the definition set by INVOLVE (2012): ‘research being carried out with or by members of the public rather than to, about or for them’.
The organizational and personal impacts of PPI within health and social care research are well documented, including improvements in research design, delivery, and relevance, as well as enjoyment and personal development (Brett et al., Reference Brett, Staniszewska, Mockford, Herron-Marx, Hughes, Tysall and Suleman2012, Reference Brett, Staniszewska, Mockford, Herron-Marx, Hughes, Tysall and Suleman2014; Cook et al., Reference Cook, Siddiqi, Twiddy and Kenyon2019; Crocker et al., Reference Crocker, Boylan, Bostock and Locock2017; Ennis & Wykes, Reference Ennis and Wykes2013; Modigh et al., Reference Modigh, Sampaio, Moberg and Fredriksson2021; Staley, Reference Staley2009). Despite this, the implementation of PPI faces significant challenges. Across the field, one of the largest barriers identified is insufficient time and resourcing (Agyei‐Manu et al., Reference Agyei‐Manu, Atkins, Lee, Rostron, Dozier, Smith and McQuillan2023; Higgs et al., Reference Higgs, Liao, Windsor and Ben-David2023; Jones et al., Reference Jones, Callejas, Brown, Colder Carras, Croft, Pagdon and Zisman-Ilani2023; Ocloo et al., Reference Ocloo, Garfield, Franklin and Dawson2021; Paul & Holt, Reference Paul and Holt2017; Rasmus et al., Reference Rasmus, Charles, John and Allen2019; Shen et al., Reference Shen, Doyle-Thomas, Beesley, Karmali, Williams, Tanel and McPherson2017). Universities, where much mental health research is conducted, are competitive environments, characterized by an emphasis on productivity, efficiency, and performance, and this can leave little space for PPI (Heney & Poleykett, Reference Heney and Poleykett2022; Papoulias & Callard, Reference Papoulias and Callard2022; Paylor & McKevitt, Reference Paylor and McKevitt2019; Sweeney, Reference Sweeney2016). Limited knowledge about research and PPI, a lack of role clarity, limited support, and training have also been identified as barriers (Bombard et al., Reference Bombard, Baker, Orlando, Fancott, Bhatia, Casalino and Pomey2018; Jones et al., Reference Jones, Callejas, Brown, Colder Carras, Croft, Pagdon and Zisman-Ilani2023; Ocloo et al., Reference Ocloo, Garfield, Franklin and Dawson2021; Paul & Holt, Reference Paul and Holt2017; Shen et al., Reference Shen, Doyle-Thomas, Beesley, Karmali, Williams, Tanel and McPherson2017; Vinnicombe et al., Reference Vinnicombe, Bianchim and Noyes2023). Difficulties with communication and collaboration, as well as the emotional impact on both researchers and service users have been recognized (Boylan et al., Reference Boylan, Locock, Thomson and Staniszewska2019; Carr, Reference Carr2019; Faulkner & Thompson, Reference Faulkner and Thompson2021; Ocloo et al., Reference Ocloo, Garfield, Franklin and Dawson2021; Todd et al., Reference Todd, Coupland and Randall2020), as well as concerns over the diversity of PPI contributors (Evans & Papoulias, Reference Evans and Papoulias2020; Jones et al., Reference Jones, Callejas, Brown, Colder Carras, Croft, Pagdon and Zisman-Ilani2023; Ocloo et al., Reference Ocloo, Garfield, Franklin and Dawson2021; Vinnicombe et al., Reference Vinnicombe, Bianchim and Noyes2023).
Recent systematic reviews of youth involvement in mental health research (McCabe et al., Reference McCabe, Amarbayan, Rabi, Mendoza, Naqvi, Thapa Bajgain and Santana2023; Totzeck et al., Reference Totzeck, van der Meer, Christiansen, Durlach, Sanchez and Schneider2024) generally echo these findings. However, discussions of epistemic injustice and issues of power within mental health research are largely absent. Significant power asymmetries between clinicians and patients exist within psychiatry, potentially based on coercive treatment and involuntary hospitalization (Rose & Kalathil, Reference Rose and Kalathil2019; Zisman-Ilani et al., Reference Zisman-Ilani, Roth and Mistler2021). Within the scientific research paradigm, the privileging of professional, academic, and clinical knowledge over experiential, lay knowledge means that service users’ views are considered less credible, and so their contributions can be marginalized (Beresford & Boxall, Reference Beresford, Boxall and Staddon2015; Green & Johns, Reference Green and Johns2019; Paul & Holt, Reference Paul and Holt2017). Some assert that mental health service users are at greater risk of this epistemic injustice than those in general medicine, due to negative assumptions around vulnerability, reliability and capacity (Crichton, Carel & Kidd, Reference Crichton, Carel and Kidd2017; Sanati & Kyratsous, Reference Sanati and Kyratsous2015). Within this context, genuine collaboration can be problematic and tokenistic (Caron-Flinterman et al., Reference Caron-Flinterman, Broerse and Bunders2007; Green & Johns, Reference Green and Johns2019; Ocloo et al., Reference Ocloo, Garfield, Franklin and Dawson2021; Patterson et al., Reference Patterson, Trite and Weaver2014; Paul & Holt, Reference Paul and Holt2017; Rose & Kalathil, Reference Rose and Kalathil2019).
Most reviews concern PPI within health and social care settings. While mental health research mirrors many of these findings, we speculate that the inherent power dynamics and differential value attached to knowledge may detrimentally influence genuine involvement. Most of the authors of this review are researchers with lived experience (LE) of mental ill health and so have first-hand experience and understanding of some of the inherent challenges associated with PPI practice within academic mental health research. Our longstanding association with the Maudsley Biomedical Research Centre and the Service User Research Enterprise puts us in a strong position to co-produce the first systematic review identifying barriers and facilitators to meaningful patient and public involvement in mental health research within academia. We believe this knowledge will aid understanding and help to provide evidence for best practice in the future.
Method
The systematic review protocol was pre-registered with PROSPERO (CRD42022309260) and followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (see Supplementary Table S1).
The involvement of experiential knowledge
People with lived experience of mental health difficulties were involved throughout, including study design, data collection, synthesis, and authorship. Authors JE, CDCL, and AS are researchers with lived experience. A Working Group of people with experience of mental health distress or caregiving and working within PPI contributed to the protocol, data synthesis, and are authors.
Search strategy
The search strategy was based on a PICO framework, combining Medical Subject Headings (MeSH) terms and free text. The search included a combination of the following concepts: population (patient, carer, service user, consumer, survivor, stakeholder, public, ally, lived experience, expert by experience), type of involvement (involvement, collaboration, participatory research, co-produced, user-led, etc.), mental health research setting (psychiatry, psychology, mental health), and facilitators and barriers (impact, change, facilitate, enable, improve, empower, barrier, challenge, decrease disempower, etc.).
Inclusion criteria were all peer-reviewed, primary research studies with a qualitative element, reporting barriers and facilitators of PPI in mental health research within academia. The publication dates were 2015–2024, and all ages and geographic locations were included. Exclusion criteria were neurodiversity, alcohol or substance use, dementia or Alzheimer’s, learning difficulties or disabilities, and studies in policy, planning, and education. We excluded autoethnographies as we wanted to understand team-based accounts.
We searched the following peer-reviewed databases: EMBASE, PsycINFO, MEDLINE, Cochrane, Web of Science, CINAHL (Cumulative Index to Nursing and Allied Health Literature), NICE Evidence, and NIHR Evidence. For context, we searched grey literature databases: Google Scholar, WorldCat Dissertation and Theses, and the National Grey Literature Collection and UK charities databases (McPin, MIND, the Mental Health Foundation, Rethink, the Co-Production Collective, and the Centre for Public Engagement). This was supplemented with backward and forward citation searching. The search was completed on 13th May 2024, the results were imported into EndNote and analyzed in Rayyan after deduplication. An example search strategy is shown in Supplementary Table S2.
Screening
All results were screened by title and abstract by two service user researchers independently. The full texts of the remaining articles were then screened independently. At each stage, the results were discussed and any discrepancies resolved.
Data extraction
Structured data extraction captured: authors, publication date, country, study aim, participant and research team demographics, data collection and analysis methods, and levels of PPI. Data were extracted independently from the whole papers, before reaching consensus.
Critical appraisal
The CASP Qualitative Studies Checklist (CASP, 2023) was modified to include an additional criterion on diversity (Sweeney et al., Reference Sweeney, Clement, Gribble, Jackson, Carr, Catty and Gillard2019). All articles were independently assessed, and as a reliability exercise, lead authors were provided with their CASP scores and asked for supporting information if they felt the scores were inaccurate. A sensitivity analysis was conducted to understand the variability of the data (Garside, Reference Garside2014).
Data synthesis
A thematic synthesis (Braun & Clarke, Reference Braun and Clarke2006; Thomas & Harden, Reference Thomas and Harden2008) took an inductive approach, starting with an open question and then generating a conceptual framework via the emerging codes. The synthesis took a three-stage approach. Following data familiarization, data were coded line-by-line to form initial codes. We looked for similarities and differences between the coded data before grouping them into distinct, descriptive themes within a hierarchical structure. Over-arching analytical themes were developed through comparison, identifying patterns and relationships contributing to the overall understanding of the review question. This analysis was conducted independently by two lived experience researchers using NVIVO, before the results were discussed and merged to reflect a shared understanding. To increase the credibility of the findings (Lincoln & Guba, Reference Lincoln and Guba1985), a peer debriefing workshop was held with the lived experience working group. Members discussed three included articles, selected at random, and the resulting themes were compared with the existing coding frame and used to inform the analysis.
Results
Search summary
As shown in Figure 1, the search strategy resulted in 14,105 articles. After deduplication, 11,780 articles were screened by title and abstract, and 276 were found relevant for full-text screening. The final review includes 51 articles. Most studies were conducted in the UK and Australia, followed by Europe and the USA. Most had a qualitative design and employed interviews, which were thematically analyzed. Thirty-five studies included lived experience researchers, ten did not, and six studies were reported as user-led. Demographic data were reported for participants but rarely for the lived experience component of research teams. For further details, see Supplementary Table S3.
PRISMA (2020) flow diagram.
The final CASP scores ranged from 8 to 22 out of a total of 22, with a mean of 17.94 (SD 2.53). Generally, the studies scored highly on the descriptions of their aim, design, data collection and analysis, findings, and research value. However, there was heterogeneity in the total scores, with seven studies scoring lower on research design and conduct (see Supplementary Table S4). Across all studies, areas with relatively lower scores were reflexivity on the relationship between researcher and participants (mean = 1.06, SD 0.9), intersectionality (mean 0.65, SD 0.74), and ethical considerations (mean 1.22, SD 0.73). After removing the lowest scoring studies from the data set, the sensitivity analysis maintained consistency across the themes. No papers were excluded based on CASP scores.
Thematic synthesis
The included publications provided many positive examples of involvement within mental health research, including setting research priorities, study design, the use of advisory groups, recruitment, data collection, analysis, authorship, and dissemination. However, over half of the studies also reported limited or late involvement, where service users were present but had limited influence over decision-making. Both researchers and service users reported feeling frustrated at what they perceived to be tokenistic involvement. Supplementary Table S5 shows how the themes are distributed across the included publications and how they correspond with the Working Group peer debriefing, and it is illustrated by quotations from the included studies and the Working Group. Overviews of the main themes and the barriers and facilitators are shown at Figure 2 and Table 1 respectively.
An overview of the main themes.
An overview of barriers and facilitators
How research is structured
Navigating governance procedures
Many studies reported feeling constrained by research governance processes, such as funding bodies and ethics committees. Limited funding and staff resourcing for PPI, competing research demands, and time constraints were the most cited barriers. Over a third of studies highlighted how fixed research protocols, grant deadlines, and concerns over methodological rigor led to rigid working practices which were not conducive to accommodating meaningful PPI. Including PPI in applications to ethics committees seemed to generate additional bureaucracy due to misunderstandings about PPI and concerns over safeguarding. A key facilitator was PPI inclusion during the proposal development. This ensured input into research design and helped to generate a sense of influence and ownership, thereby reducing tokenism. It was also crucial in ensuring that sufficient time, resources, and flexible working practices were determined before studies started.
Inclusion and influence over research processes
Power differentials in mental health research were highlighted by half the studies, evidenced by researchers’ ownership and control of research processes. Examples included researchers deciding on roles, service user exclusion from decision making, the marginalization of views, and the recruitment of people who would not challenge the status quo. Concerns were raised about a lack of diversity, particularly the over-representation of white, middle-class and university-educated people, with those from marginalized and racialized communities or with literacy difficulties often excluded. Researcher reticence to recognize these as issues or to share power was highlighted by several studies.
At an institutional level, embedding and monitoring PPI within academic governance and throughout research processes was a key facilitator for inclusion and redistribution of power. Successful practices included involvement in management committees, priority setting exercises, PPI facilitators, lived experience advisory groups, salaried service user researchers, and the recruitment of voluntary advisors in sufficient numbers to be effective, as well as regular monitoring and systematic measurement of PPI progress and impact. At study level, some researchers were able to navigate power dynamics by holding explicit team discussions around power differentials and making clear commitments to shared decision-making. Providing choice and flexibility regarding involvement was considered to promote autonomy and explicit, early discussion of PPI roles was seen to generate transparency and manage expectations around capacity to influence the research process. An active focus on promoting diversity was viewed as central to inclusion. This involved challenging exclusion, understanding and addressing institutional and individual barriers to involvement to ensure that research environments were more widely accessible, reaching out and collaborating with local communities and the use of targeted advertising.
Organizational culture
The cultural value attached to PPI and lived experience
The lack of value attached to PPI was seen as a major barrier, evidenced by insufficient funding, limited knowledge on implementation, and a lack of commitment. Researchers were less able to recognize the relevance of PPI within accepted scientific models of research. Researchers and service users also reported that the perspectives of people with lived experience, traditionally seen as patients or research subjects, held lower legitimacy within the academic structure. The review highlighted the importance of training for researchers to view PPI as a new way of working, based on collaboration and mutual learning, as well as emphasizing respect and value for lived experience alongside academic learning. This, alongside flexibility and a willingness to adapt research processes, would in turn lead to a growing sense of agency, inclusion, and equality for all those involved. Organizational commitment and the installation of feedback mechanisms were also seen as ways of demonstrating value and recognizing contribution.
Interactions and relationships
Strained relationships were apparent, often because of the clinician/patient hierarchy and reluctance to share power, and this led to service users’ reports of marginalization. Paternalistic attitudes and researcher insecurity about interaction with service users, particularly fear of causing offence or hearing critical feedback, inhibited communication, and active involvement. A third of studies reported difficulty integrating different perspectives and expectations within research teams, which could lead to reports of frustration and a sense of division.
Key facilitators were identified as relationship building, collaboration, and open, honest, and regular communication. Central was the creation of informal opportunities for teams to mix and share experiences, as well as an emphasis on the importance of partnership, collaboration, and compromise. Open and honest communication helped to maintain connection, share information, build relationships, and reconcile differences, as well as to sustain engagement and identify any support needs. This required accessibility of research team members and regular contact with PPI contributors.
Individual needs
Unfamiliarity with research
Both researchers and people with lived experience highlighted how service users were unable to contribute fully due to unfamiliar research processes and terminology, confusion over roles and expectations, and insufficient training. This was compounded by a lack of confidence derived from unfamiliarity with the academic environment. Training on research methods and development for service users, as well as the provision of clear information and the use of lay language, supported autonomy and the capacity to engage in research.
Feeling safe within the research environment
Negative experiences of psychiatric services and research participation, as well as the stigma and discrimination associated with mental health, could lead to anxiety and mistrust and hinder full involvement. Many studies highlighted how emotionally challenging involvement could be, leaving service users feeling overwhelmed. Involvement could also be affected by poor mental health, leading to periods of absence, and life factors such as a lack of qualifications, the need for sufficient income, employment, disability, travel needs, low literacy, and English as a second language. Creating informal, supportive research environments, where researchers were open to different perspectives and ensured healthy workloads, and where service users felt respected, listened to, and comfortable enough to voice their opinions, was crucial. Almost half of the studies stressed the importance of providing ‘safe spaces’. Service users reported the positive impact of this way of working, including the opportunity to create positive change, increases in self-worth and confidence, self-development, social connection, and the building of trust. Consideration of individual needs, for example, facilitating travel, accessible meetings, reimbursement, support with the benefits system, the provision of accessible documents or interpreters, and reasonable adjustments for disabilities, was also a key facilitator.
Working group: peer debriefing
The Working Group identified salient themes and discussed how they corresponded with their personal experiences of undertaking PPI (Supplementary Table S4). Areas that resonated were the existence of paternalism and risk aversion within ethics committees, and service users’ lack of understanding of research and roles within teams. Members agreed that relationship building was instrumental in breaking down researchers’ assumptions and increasing value for lived experience. They also reinforced the importance of providing safe, supportive working environments, clarity of roles, and choice around levels of involvement. Where their views differed were in relation to power dynamics. Members had witnessed power differentials, tokenism, and negative researcher attitudes, but felt that this had improved over time and was dependent on individual personalities.
Discussion
PPI is increasingly employed within mental health research; however, significant barriers endure. Authentic involvement requires time, flexibility, inclusion, and equity, and our findings suggest that this can be difficult to achieve within academia. Our results confirm previous research (Agyei‐Manu et al., Reference Agyei‐Manu, Atkins, Lee, Rostron, Dozier, Smith and McQuillan2023; Heney & Poleykett, Reference Heney and Poleykett2022; Paylor & McKevitt, Reference Paylor and McKevitt2019) showing that despite funders’ emphasis on PPI inclusion, fixed timescales and an emphasis on productivity, combined with a lack of understanding of or value for PPI, make it difficult for researchers to undertake PPI in a meaningful way. Our review points to the existence of significant power differentials and resistance to change, corroborating research by Hopkins et al.’s (Reference Hopkins, Verlander, Clarkson and Jacobsen2024). It also highlights the low status of experiential knowledge, as consistently raised by lived experience researchers (Beresford & Boxall, Reference Beresford, Boxall and Staddon2015; Rose & Kalathil, Reference Rose and Kalathil2019: da Cunha Lewin, Reference da Cunha Lewin2025). Both aspects combine to make genuine collaboration difficult and leave service users feeling unsupported, underutilized, and undervalued. As within the health and social care field (Ocloo et al., Reference Ocloo, Garfield, Franklin and Dawson2021; Vinnicombe et al., Reference Vinnicombe, Bianchim and Noyes2023), we found that reliance on the same convenient pool of contributors can restrict the diversity of perspectives and lead to systemic exclusions. Given these factors, it is perhaps unsurprising that over half of the review studies reported limited involvement and experiences of tokenism.
Quality appraisal
Overall, the quality of included studies was high, although the heterogeneity of scores indicates some variability in methodological rigor. Without transparent reporting, there can be uncertainty around the credibility of results. Rather than exclude studies based on their scores, a sensitivity analysis revealed consistency across themes, thereby increasing confidence in our findings (Carroll et al., Reference Carroll, Booth and Lloyd-Jones2012). Whilst most studies were collaborative, ten were not and only six were reported as user-led, which casts some doubt as to whose opinions are prioritized within the review. More lived experience/co-led research is required within academia to ensure that lived experience perspectives are given equal weighting.
The quality appraisal also showed relatively little attention paid to the researcher-participant/co-researcher relationship and ethical considerations. Reflexivity is an essential element within qualitative research, ensuring that researchers consider their influence over participants and the data (Newton et al., Reference Newton, Rothlingova, Gutteridge, LeMarchand and Raphael2012). Without it, it is difficult to assess whether bias has influenced the studies within this review. Studies employing PPI methodology do not necessarily require formal ethical approval. Nonetheless, given the complex and often sensitive nature of PPI in mental health research, the absence of discussion around ethical concerns such as confidentiality, safeguarding, and support was notable (Biddle et al., Reference Biddle, Gibson and Evans2021). Considering the significance of power dynamics and the need for safe and equitable working relationships highlighted within this review, it is imperative that researchers use reflexivity to ensure the well-being of co-researchers and participants, as well as consider their own influence over findings.
Study strengths and limitations
To our knowledge, this is the first systematic review and meta-synthesis of PPI in mental health research within academia. Lived experience perspectives were embedded throughout. To maximize reliability, screening, quality appraisal, and analysis were conducted independently. The Working Group then discussed and reflected on the major themes and related these to their own experiences, in line with Shimmin et al. (Reference Shimmin, Wittmeier, Lavoie, Wicklund and Sibley2017). We recognize that some of our screening criteria may have prioritized interpretations of PPI from the Global North. Whilst our specific interest was in PPI within academia, we acknowledge that using broader PPI definitions and sources of literature, particularly those from the Global South, as well as a multilingual approach, may have captured a more comprehensive picture of barriers and facilitators of PPI in mental health research. Our core research team is predominantly white, older, and cisgender, which inevitably influences our interpretation of the data. We were reflexive in our approach and ensured that the Working Group was diverse in terms of gender, ethnicity, and age, as well as service user/carer representation and working experience of different PPI groups.
Implications for future research and PPI
The redistribution of power
This review has highlighted the necessity for a more equitable distribution of power within mental health research. This is achievable, in part, by embedding PPI into governance and research procedures, ideally at the pre-funding stage before rigid research procedures are established. However, this does require a shift in understanding on the part of researchers, funders, and ethics committees. Rather than viewing PPI as an adjunct, it should be recognized and valued as a novel way of working, requiring proper funding, time, and flexibility (Callejas & Jones, Reference Callejas and Jones2022; Totzeck et al., Reference Totzeck, van der Meer, Christiansen, Durlach, Sanchez and Schneider2024). Power dynamics can also be negotiated at the study level, through collective reflexivity and relationship building. It requires sustained effort but can be successful in breaking down barriers, generating equity, and promoting genuine collaboration (Cox & Simpson, Reference Cox and Simpson2015; Roper et al., Reference Roper, Grey and Cadogan2018; Sangill et al., Reference Sangill, Buus, Hybholt and Berring2019; Shimmin et al., Reference Shimmin, Wittmeier, Lavoie, Wicklund and Sibley2017).
Building capacity
For power sharing to be effective, both researchers and lived experience contributors need the capacity to engage. We found a strong emphasis on providing service users with training on scientific research knowledge and skills, to enable participation. Whilst this is a positive step, it fails to account for the power imbalances and the systemic change required to facilitate meaningful involvement. The review focused less on capacity building for researchers. Expecting researchers to undertake PPI without understanding its relevance or implementation hinders success and makes tokenism more likely (Friesen et al., Reference Friesen, Lignou, Sheehan and Singh2021; Thompson et al., Reference Thompson, Barber, Ward, Boote, Cooper, Armitage and Jones2009). Training is critical in increasing knowledge of PPI and an understanding of power dynamics, as well as enabling attitudes to move away from traditional conceptions of the clinician/patient, toward the respect and value for lived experience, which is fundamental to agency and inclusion. We recommend that any training is co-produced and co-delivered to increase relevance and reduce the ‘us and them’ dynamic.
Safe working environments
Almost half the included studies emphasized the importance of feeling safe within the research environment, which suggests that involvement in mental health research feels unsafe for many and can have an emotional cost. Shimmin et al. (Reference Shimmin, Wittmeier, Lavoie, Wicklund and Sibley2017) have shown how trauma-informed, intersectional approaches can create settings and interactions that support safety for everyone involved. The onus is on academic institutions to reflect on their own practices, to become aware of barriers that impede full involvement, and to adapt working practices. Taking a holistic approach focused on communication, relationship building, and capacity development can make involvement more meaningful and inclusive for everyone (Roche et al., Reference Roche, Shimmin, Hickes, Khan, Sherzoi, Wicklund and Sibley2020).
Widening inclusion
The thematic analysis, quality appraisal, and data extraction all highlighted how consideration of diversity, as in the health and social care field (Ocloo et al., Reference Ocloo, Garfield, Franklin and Dawson2021; Ocloo & Matthews, Reference Ocloo and Matthews2016) is relatively scarce. This means we are unable to comment on the barriers and facilitators of PPI for marginalized groups and limits the generalizability of the findings. Addressing this requires proactive strategies. Ongoing efforts should be made to reach out to, build lasting relationships with, and address the barriers faced by traditionally excluded communities (Chauhan et al., Reference Chauhan, Leefe, Ní Shé and Harrison2021; Ekezie et al., Reference Ekezie, Routen, Denegri and Khunti2021; Golenya et al., Reference Golenya, Chloris, Panteli, Giannoudis and Howard2021; Kalathil, Reference Kalathil2008).
Conclusion
Our review highlights the existence of good PPI practice in mental health research within academia. It is evidently a valuable experience for many and has a positive influence on research processes. However, the wider research environment within universities and enduring power differentials, constrain PPI, leading to a consumerist model and experiences of tokenism. Insufficient value attached to PPI and lived experience, researcher resistance, and a lack of understanding of service users’ needs all present significant barriers. Our findings indicate that embedding PPI throughout the research process, ensuring diversity of representation, the provision of safe, supportive working environments, researcher training, and capacity building for service users are all key facilitators. Most importantly, by relinquishing some control over the research process and adopting more flexible, truly collaborative working practices, patient and public involvement in mental health research can strive to become an authentic and impactful process.
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1017/S0033291725101748.
Acknowledgements
We wish to thank the Lived Experience Working Group, some of whom have chosen to be named authors, for their invaluable contribution to this review.
Author contribution
JE, AS, TW, and the Working Group were involved in protocol development. JE conducted the data searches, and JE, CDCL, and HF undertook screening, data extraction, and quality appraisal. JE, CDCL, HF, and the Working Group were involved in data synthesis. JE, AS, and TW were involved in the writing.
Funding statement
This paper represents independent research funded by the NIHR Maudsley Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King’s College London. The views expressed are those of the authors and not necessarily those of the NIHR or the Department of Health and Social Care. Grant number NIHR203318.
Competing interests
The authors declare none
Open access
