Improving physical health-care equity for people with severe mental illness: evaluating a co-produced community initiative in Bexley, UK

Health inequalities experienced by people living with severe mental illness (SMI) represent a persistent and well-documented challenge for health systems globally. In England, the National Health Service (NHS) has identified this disparity as a policy priority, setting targets for annual physical health checks (PHCs) to address a substantial mortality gap driven largely by preventable physical conditions (NHS England, 2023). Despite these efforts, people with SMI continue to face disproportionately poor physical health outcomes, often due to fragmented care, low engagement and barriers within primary care. This aligns with recent evidence highlighting that primary care systems frequently struggle to equitably engage this population (Sheridan et al., 2025).

Local variation in PHC uptake persists, shaped by complex, intersecting barriers at patient, practitioner and system levels (Curry and Ham, 2010; Fuller, 2022; Lennox-Chhugani et al., 2021; Public Health England, 2018). In response, this paper critically evaluates an integrated, co-produced initiative in the London Borough of Bexley aimed at improving PHC uptake among individuals on the SMI register.

The initiative was developed by a multi-agency collaboration, hereafter referred to as the Bexley Partnership, comprising Mind in Bexley, Primary Care Networks (PCNs), General Practitioner (GP) leads, the Bexley Health Neighbourhood Care GP Federation Community Interest Company (CIC) and Oxleas NHS Foundation Trust. Together, this cross-sector partnership implemented a series of coordinated strategies, including lived-experience-informed practitioner education, collaborative data-sharing, targeted outreach and the delivery of care in trusted, community-based settings.

By drawing on qualitative feedback from both health-care professionals and residents, this evaluation explores the mechanisms through which relational, resident-focused, neighbourhood-level approaches can drive system change. It contributes to the growing evidence base for integrated, sustainable and culturally competent care models designed to reduce physical health inequalities for people with SMI, in line with the ambitions of the NHS Long Term Plan and Core20PLUS5 framework.

People living with SMI experience profound health inequalities in the UK, with life expectancy estimated to be 15–20 years lower than that of the general population (Public Health England, 2018; NHS England, 2023; Walker et al., 2015). This mortality gap is largely attributable to preventable physical health conditions, such as cardiovascular disease, diabetes and respiratory illness, rather than psychiatric symptoms. Approximately two-thirds of premature deaths among people with SMI are considered avoidable through timely and appropriate interventions (NHS England, 2023; Public Health England, 2018). Addressing this inequity is a central health policy priority. The NHS Long Term Plan aims to reduce the mortality gap by improving physical health care for people with SMI, including setting targets for annual PHCs (NHS England, 2019). These checks monitor six core risk factors: alcohol use, smoking, body mass index (BMI), blood pressure, glucose/HbA1c and lipid profile. When systematically delivered, PHCs can reduce emergency admissions and improve life expectancy.

Despite national commitments, coverage of annual PHCs for people with SMI remains inconsistent and below policy targets. The Core20PLUS5 framework identifies people with SMI as a key target group, with an aspirational goal that 60% receive a full PHC annually. However, fewer than half of eligible patients receive all six components, with disparities based on region, ethnicity, socioeconomic status and service engagement (Centre for Mental Health, 2024). Barriers to uptake operate at patient, practitioner and system levels. Medications for SMI, including antipsychotics and mood stabilisers, can lead to significant physical side effects such as weight gain, diabetes and cardiovascular disease. Polypharmacy compounds these risks, making the balance between managing psychiatric symptoms and preventing physical deterioration particularly challenging (Public Health England, 2018).

Lifestyle factors further contribute. Smoking rates among people with SMI remain disproportionately high, with approximately 40%–50% of adults affected, significantly exceeding the national average. This makes smoking the single most modifiable contributor to their reduced life expectancy (Royal College of Physicians and Action on Smoking and Health, 2019). Poor diet, sleep disruption, low physical activity and substance use are also common, often linked to coping mechanisms or negative symptoms such as apathy and avolition (Public Health England, 2018; Sheridan et al., 2025). Side effects from psychotropic medication, such as increased appetite and fatigue, can further exacerbate these behaviours. Structural inequalities, including low income, food insecurity and limited health literacy, add to these challenges by reducing access to preventative care. A recent systematic review highlights the compounding impact of socioeconomic disadvantage and behavioural risk factors, which together drive high rates of chronic multimorbidity (Álvarez Gálvez et al., 2023). For people with SMI, these issues intersect with illness-related barriers to further increase vulnerability.

Low health literacy and difficulties navigating complex health-care systems impede risk management and engagement with physical health interventions. While UK guidelines recommend brief interventions and motivational interviewing, the evidence base for their effectiveness in SMI populations remains limited, pointing to the need for more tailored approaches. Additional barriers include stigma, the misattribution of physical symptoms to mental illness and broader experiences of marginalisation (Sheridan et al., 2025). Psychiatric symptoms such as cognitive impairment and low motivation often hinder appointment attendance. Practitioners may inadvertently downplay physical symptoms by viewing them primarily through a mental health lens, resulting in delayed or missed opportunities for appropriate care. System-level challenges, including fragmented data systems, workforce shortages and poor coordination between primary and secondary care, further undermine service delivery (Thomson and Chatterjee, 2023; The King’s Fund, 2022). Recent literature consistently identifies stigma, systemic inflexibility and fragmented care as key obstacles to effective physical health support for people with SMI (Sheridan et al., 2025).

Integrated care has been proposed as a solution. NHS England’s Community Mental Health Framework recommends embedding mental health professionals within PCNs to promote holistic, joined-up care (NHS England, 2019). However, implementation has been uneven due to IT incompatibilities, information governance constraints, variable local leadership and staff instability (Thomson and Chatterjee, 2023). The South East London PHCs for SMI programme illustrates both progress and persistent challenges. A 2025 evaluation by the Health Innovation Network South London reviewed over 2,600 patient records across 31 general practices. While average PHC completion increased from 0.5 to 1.2 elements per patient, key gaps remained, particularly in blood testing. Barriers included delays in information-sharing, high staff turnover and the redirection of nursing roles from outreach to data validation.

In the London Borough of Bexley, uptake of PHCs among individuals on the SMI register stood at just 43% as of December 2024, significantly below the national target of 60% (Mind in Bexley, 2025). Barriers existed at multiple levels, including patient disengagement, practitioner capacity and systemic fragmentation. In response, and as part of a broader South London programme to improve physical health outcomes for people with SMI, the Health Innovation Network South London supported a series of co-produced initiatives, including one in Bexley delivered in partnership with Mind in Bexley, Oxleas NHS Foundation Trust and GP’s in Bexley. Central to the Bexley initiative was a commitment to co-production, with individuals with lived experience helping to shape the design, delivery and adaptation of interventions to improve accessibility, trust and engagement. The Clinical Care Professional Lead (CCPL) for Mental Health helped coordinate efforts with the CCPL for Cancer and Healthy Behaviours, the Chief Executive of Bexley Health Neighbourhood Care GP Federation CIC, PCN and GP’s leads and senior Oxleas leadership. Activities included structured engagement, data quality improvements and proactive outreach to patients who had disengaged from services.

Interventions were designed to address identified barriers and included interactive webinars for primary care staff, as well as a co-produced video featuring narratives from both staff and service users. This resource contextualised health inequality through lived experience, reduced stigma and promoted relational, person-centred care. The programme also tackled system-level issues such as data fragmentation, workforce turnover and underdeveloped primary care infrastructure. By fostering voluntary-sector partnerships and providing strategic leadership, the Bexley initiative aimed to support scalable, sustainable improvements in PHC delivery for people with SMI.

By March 2025, PHC coverage in Bexley rose to 63%, exceeding the national target and demonstrating the value of collaborative, locally tailored approaches (Mind in Bexley, 2025). However, qualitative data revealed ongoing barriers, including data integration issues, limited practitioner time and service-user challenges linked to psychiatric symptoms, stigma and medication side effects (Health Innovation Network South London, 2025).

This paper situates the Bexley initiative within wider regional and national efforts to improve PHC coverage for people with SMI, illustrating how co-production, education and lived experience can inform more effective, equitable care. While practitioner training remains essential, long-term progress will depend on concurrent organisational and system-level reforms to deliver high-quality physical health care for this underserved population.

This study was undertaken as a service evaluation of a co-produced initiative to increase uptake of annual PHCs for people with SMI in the London Borough of Bexley. The initiative was delivered by a multi-agency collaboration, referred to throughout as the Bexley Partnership, which brought together Mind in Bexley, PCNs, GP leads, the Bexley Health Neighbourhood Care GP Federation CIC and Oxleas NHS Foundation Trust. Funding and strategic support were provided by the Health Innovation Network South London.

The evaluation was coordinated by the CCPL for Mental Health in Bexley, in collaboration with wider system partners. The design reflects principles of participatory and formative evaluation, focusing on mechanisms of change within real-world, practice-based systems.

Primary qualitative and indicative quantitative data were collected between January and March 2025 from the following sources:

• Practitioner feedback was gathered from eight co-produced webinars and a GP roundtable event. Participants (n = 90) included GPs, practice nurses, PCN leads and administrative staff. Open-text responses were collected through post-session feedback forms, live chat comments and reflective polling. Sessions were co-designed with service users and included video content featuring lived-experience narratives from local residents with SMI.

• Service-user feedback was collected through structured telephone outreach to 80 individuals on the SMI register who had not recently attended a PHC. Peer support workers and trained staff from Mind in Bexley used a semi-structured script to explore barriers to attendance, preferred models of delivery and suggestions for improving access. Notes from these conversations were anonymised and summarised for thematic analysis.

• Pilot outreach feedback was collected informally from 18 residents invited to a community-based PHC session held at Mind in Bexley, of whom eight received full checks. Reflections from attendees were recorded by staff and peer supporters.

In addition, polling data from webinar participants (n = 90) provided pre- and post-training self-assessment scores of confidence and awareness using a five-point Likert scale. These were used to supplement qualitative findings.

All qualitative data were analysed thematically using an inductive, iterative approach. An initial coding framework was developed by the lead evaluator based on patterns in responses across practitioner and service-user groups. Codes were refined through multiple readings and grouped into themes reflecting barriers, enablers and system-level dynamics affecting PHC uptake. This analytic process followed Braun and Clarke’s widely recognised six-phase framework for thematic analysis, which is particularly suited to applied health research and flexible enough for service evaluation contexts (Braun and Clarke, 2006). Analysis was reviewed by Mind in Bexley colleagues to ensure contextual validity. Findings were organised across three analytical levels, patient, practitioner and system in line with existing public mental health frameworks. Representative quotes and practitioner reflections were included to enhance interpretive depth.

As a service evaluation, this project did not require NHS Research Ethics Committee approval under Health Research Authority guidance. All participants were informed of the purpose of the evaluation, their right to opt out and the anonymised use of their feedback. Verbal consent was obtained. Safeguarding procedures were co-designed with Mind in Bexley to support the emotional well-being of service users with lived experience of mental illness.

This evaluation has several limitations. Qualitative data were based on self-reported reflections and field notes rather than in-depth interviews or audio-recorded transcripts. The pilot outreach session engaged a relatively small number of residents, and no longitudinal follow-up was undertaken. Furthermore, the initiative occurred within a single locality with established voluntary-sector infrastructure and pre-existing collaborative relationships, which may limit generalisability to areas with less partnership capacity.

Nonetheless, this form of embedded, co-produced service evaluation is increasingly recognised as best practice in applied public mental health. Global guidance highlights the value of participatory approaches for evaluating complex interventions in real-world systems, particularly among underserved populations (Greenhalgh and Papoutsi, 2018). This evaluation contributes to that agenda by integrating practitioner and resident voices to inform equitable, community-based care models for people with SMI.

Thematic analysis of the qualitative data revealed interconnected themes highlighting both the intervention’s impact on practitioners and persistent, multi-level barriers to equitable uptake of PHCs among people with SMI in Bexley.

Practitioners consistently reported a shift in awareness of the health inequalities faced by people with SMI. The co-produced training combined epidemiological evidence, local data and lived-experience narratives, reframing PHCs from an administrative task into what many described as an ethical and clinical imperative. This aligns with literature on disruptive learning, where emotionally resonant, narrative-rich content fosters reflexivity (Wear and Zarconi, 2008). Participants described the training as “eye-opening”, “shocking in a good way” and admitted to underestimating the mortality gap. Poll data reflected this, with mean self-rated understanding rising from 3.45 to 4.53.

Many emphasised the power of hearing personal stories to humanise health inequalities and disrupt stereotypes: “Hearing their stories really brought it home”, and “We need that shock to make us act”. The co-produced video, featuring residents with lived experience alongside NHS and GP staff, was praised for its authenticity and emotional impact: “Essential for making the issue real” and “Very powerful…it really changed how I see it”. These reflections highlight the value of locally relevant, co-produced resources in promoting empathy and culturally competent, relational practice (Wear and Zarconi, 2008).

Beyond awareness, the training aimed to build communication skills and practitioner confidence, areas where many previously felt apprehensive. Concerns about “appearing insensitive”, “triggering disengagement” or “causing anxiety” were frequently cited. Post-training polling showed a marked increase in self-rated confidence (mean 2.91–4.87). One practitioner said: “I am confident I can support people living with SMI to attend a physical health check”. Another observed: “It gave me a clear way to talk about it without making it sound scary or blaming”. However, participants recognised the limitations of training alone. Many stressed that one-off sessions would not deliver lasting change without broader system-level integration. Comments emphasised engaging full practice teams to ensure consistent approaches: “All practice nurses should be encouraged to attend”. Others called for cross-borough coordination: “Please liaise with other (boroughs)… We want to raise this awareness with all clinicians”. Calls for ongoing resources were emphatic: “We need resources ongoing to keep this in mind”. These reflections mirror evidence that professional development requires repeated engagement and system-wide reinforcement (Kirkpatrick and Kirkpatrick, 2006).

Service-user feedback from 80 residents on the SMI register offered further insight into persistent barriers. While anxiety about receiving serious health news was common, it was rarely isolated. Fear interlinked with disrupted routines, low motivation, cognitive challenges and medication side effects. These accounts revealed the emotional and structural complexity of disengagement, shaped by disadvantage, stigma and exclusion. Relational barriers such as discomfort in clinical settings and mistrust from past experiences were noted. Participants expressed preferences for care in familiar, less formal environments. One explained: “If it was in the community centre I know, later in the day, I’d feel less alone and safer”. Another shared: “I panic if I think it’s going to be horrible so I just don’t go… it would help to have someone come with me”.

In response, the initiative piloted a community-based outreach session at Mind in Bexley. Over 80 residents were contacted; 18 agreed to attend, and eight received a health check in a non-clinical setting. Though small-scale but indicative, the pilot demonstrated the feasibility of engaging those likely to avoid traditional GP appointments. Participants valued the setting and support: “I wasn’t rushed… the staff supported me”, and “I haven’t been out of the house for two years… this has been a life-line”. These accounts illustrate the potential of relational, trust-building approaches to reduce fear and deliver person-centred care aligned with dignity and respect (Davidson et al., 2006).

Importantly, the outreach pilot was part of a broader collaborative framework. The Bexley Partnership worked together to share data, validate registers, coordinate contact and co-design delivery in a trusted community setting. This model demonstrated how cross-sector collaboration can operationalise integrated care in tangible, resident-focused ways. Nonetheless, persistent system-level challenges emerged. Persistent structural issues, including disjointed data flows, information-sharing limitations and overstretched primary care teams, remained key barriers. One noted: “We can’t fix this long term if we don’t know who’s who”, emphasising the need for interoperable systems to support outreach and planning. These barriers reflect well-documented challenges in integrated care, where sustainable change depends on shared vision, strong relationships and enabling infrastructure (Thomson and Chatterjee, 2023).

Taken together, these results show that while education can catalyse practitioner awareness and confidence, lasting improvements in PHC uptake for people with SMI require system-level change, cross-sector partnerships and community-based delivery models grounded in trust, dignity and relational care. Table 1 summarises the key multi-level barriers and enablers identified during the evaluation, spanning patient, practitioner and system levels.

This evaluation provides an in-depth, qualitative assessment of a co-produced initiative in the London Borough of Bexley aimed at improving uptake of annual PHCs for people with SMI. The findings demonstrate that practitioner-focused education integrating local data with lived-experience narratives can increase awareness of health inequalities, challenge diagnostic biases and build confidence in addressing physical health needs. However, these improvements cannot be sustained without system-level change.

The Bexley initiative illustrates how meaningful integration requires genuine collaboration across organisational boundaries. The Bexley Partnership formed formal and informal partnerships, shared resources and data and jointly designed, delivered and evaluated interventions. This approach moved beyond ad hoc training to create sustained relationships, shared problem-solving and outreach rooted in local context. Such partnership models reflect NHS England’s (2019) Community Mental Health Framework vision and echo academic calls for stronger cross-sectoral collaboration to tackle inequalities (Naylor et al., 2016; Thomson and Chatterjee, 2023; The King’s Fund, 2022). This aligns with findings by Sheridan et al. (2025), who argue that despite national guidance, effective care for people with SMI remains hampered by fragmented primary care processes and insufficient adaptation to their specific needs.

A major strength of this evaluation is its co-produced design, embedding service-user perspectives into training and evaluation. Triangulating practitioner feedback, polling data and resident accounts provides a nuanced view of multi-level barriers and enablers. It also documents practical examples of integration, including collaborative data cleansing, co-developed educational materials and coordinated community outreach, showing that integrated care can be realised through local relationship-building and shared resourcing.

These findings reinforce the value of relational, trust-based integration. Research shows that effective partnerships require time, leadership and shared goals (Valentijn et al., 2013; Rycroft-Malone et al., 2016). The Bexley model exemplifies this through collaborative leadership within the Bexley Partnership, with NHS and primary care partners coordinating delivery and Mind in Bexley providing voluntary-sector insight and strong community ties. This approach aligns with Core20PLUS5 priorities (NHS England, 2023), which call for local partnerships with the voluntary and community sector to address inequalities in marginalised populations.

Nonetheless, limitations remain. Practitioner outcomes rely on self-reported confidence, which may not translate to long-term behavioural change. Polling data lacked statistical power, and while the outreach pilot reached highly disengaged residents, overall attendance remained modest. Additionally, Bexley’s established voluntary-sector infrastructure and history of collaboration may limit replicability in areas with weaker partnerships or fragmented commissioning. Future research should adopt longitudinal, mixed-method designs to assess sustained behaviour change, health outcomes and scalability. While this evaluation was conducted within a single borough, its participatory, co-produced approach is aligned with global best practice in community mental health service design. The World Health Organisation (WHO) (2021) explicitly endorses community-based, person-centred models that integrate lived experience and local delivery to tackle structural inequalities. Findings from this evaluation demonstrate how such principles can be operationalised at a neighbourhood level through trust-based collaboration and outreach.

The use of thematic analysis (Braun and Clarke, 2006) within a service evaluation framework also reflects established guidance for studying complex interventions in practice settings (Greenhalgh and Papoutsi, 2018). Although not a formal research study, the rigour of the approach lies in the triangulation of perspectives, co-production methods and actionable insights for system change. This highlights the value of embedded qualitative evaluation for informing public mental health policy, particularly in the context of Core20PLUS5 and integrated neighbourhood models.

The barriers identified in this evaluation reflect long-standing concerns in the literature regarding disparities in physical health-care access for individuals with SMI. The tendency to misattribute physical symptoms to underlying psychiatric conditions continues to result in delayed or overlooked diagnoses within both primary and secondary care settings (Jones et al., 2008; Shefer et al., 2014). Although practitioner training can raise awareness of these patterns, evidence indicates that such efforts alone are insufficient to produce sustained behavioural change in the absence of wider structural and cultural reform (De Hert et al., 2011; Naylor et al., 2016).

The Bexley initiative contributes to this literature by showing how co-produced, lived-experience-informed training can provoke emotional engagement and foster empathy, key components in challenging entrenched assumptions (Jones et al., 2008). Participants described the sessions as “thought-provoking and eye-opening”, highlighting the value of personal narratives in humanising health inequalities. This echoes the concept of “disruptive learning” in health professions education, which stresses the role of emotionally resonant content in shifting practitioner mindsets (Wear and Zarconi, 2008).

Beyond training, this evaluation adds to the growing evidence base on integrated, neighbourhood-based models. The NHS Long Term Plan and Core20PLUS5 strategy call for holistic, locally driven approaches that bridge mental and physical health services through PCNs and voluntary-sector partnerships (NHS England, 2019, 2023). However, implementation is hindered by fragmented IT systems, variable leadership and data governance constraints (Thomson and Chatterjee, 2023; The King’s Fund, 2022). The Bexley case shows how these can be partially overcome through relationship-building, shared ownership and embedded outreach within trusted community settings.

It also supports theories of “relational integration” in health services research, which emphasise that successful collaboration is built not only on structures and incentives but also on trust, mutual understanding and continuous communication (Valentijn et al., 2013). Practical steps such as joint webinars, GP roundtables, shared video resources and co-produced data work in Bexley illustrate how relational practice can translate integrated care policy into meaningful local action.

These findings have implications for practice, design and policy. For practitioners, the results support mandatory, ongoing co-produced training to ensure emotional engagement and relevance to local realities. Continuing professional development should include all primary care staff, GPs, nurses and admin teams, to create whole-practice approaches to equity.

Service design must move beyond clinic-centred models to better meet the needs of people with SMI. Flexible appointments, rescheduling support, reminders and peer support can help address barriers such as anxiety, fear and disorganisation. Community-based delivery in familiar, non-clinical settings should be expanded, with voluntary-sector partners contributing cultural competence and trust. Systemically, reducing health inequalities for people with SMI requires integrated commissioning that enables joint planning across sectors. Sustainable funding must recognise the voluntary sector’s essential role in delivering person-centred care. Investment in interoperable data systems is also crucial for accurate SMI registers, proactive outreach and equity monitoring.

Future research should evaluate the long-term effects of integrated models on PHC uptake and outcomes, using mixed-method approaches. Cost-effectiveness studies and adaptation strategies will be key to scaling. By embedding dignity, trust and relational care in neighbourhood models, health services can make tangible progress towards reducing the mortality gap, and meet the NHS’s commitments to equity.

This evaluation highlights the critical role of locally tailored, co-produced strategies in addressing longstanding disparities in physical health care for people with SMI. While national frameworks and policy initiatives have sought to tackle the mortality gap, this study reveals that sustained progress depends on the capacity of systems to engage meaningfully with the lived realities of both service users and practitioners. The Bexley initiative illustrates how change becomes possible when policy ambition is translated into relational, place-based action.

Rather than viewing low uptake of PHCs through an individualised or deficit-based lens, this project emphasised the structural and cultural barriers that shape engagement. These include not only systemic issues, such as fragmented data, rigid appointment systems and stretched clinical resources, but also deeper relational dynamics, including distrust and stigma. By responding to these dynamics with co-designed, emotionally intelligent interventions, the initiative reframed health checks as opportunities for care and connection, rather than system obligations.

Practitioner-facing training that incorporated local data and first-person narratives created a sense of immediacy and moral urgency. Participants described shifts in perception, greater awareness of their role in reducing preventable harm and renewed commitment to resident person-centred care. However, these outcomes were not produced by training alone. They were made possible by the alignment of training with wider system reform: system leadership, improved data quality, shared ownership between primary care and the voluntary sector and a willingness to build relationships that extended beyond professional silos.

Importantly, the findings highlight the limits of short-term or fragmented approaches to tackling inequalities. While the co-produced interventions yielded meaningful changes in attitudes, awareness and practice, their impact will only be sustained through structural commitments, ongoing investment, integrated commissioning, data interoperability and the permanent inclusion of voluntary-sector partners and lived experience in service planning and delivery. The initiative suggests that when given the opportunity and support, practitioners are willing to adapt, collaborate and innovate. It also affirms the need for communities to play an active role in shaping how preventative care is offered, delivered and understood.

The Bexley experience offers a pragmatic and hopeful model of integrated neighbourhood care in action. It does not present a universal solution, but it provides concrete, transferable lessons for other localities seeking to embed relational care and improve equity for people with SMI. As health systems move towards more community-based, personalised models of delivery, this evaluation contributes to a growing body of practice-based evidence that positions trust, co-production and shared leadership as essential components of meaningful system change.

This study was conducted as a service evaluation of local initiatives to improve uptake of annual physical health checks among people with severe mental illness (SMI) in the London Borough of Bexley. As a service evaluation, it did not require NHS Research Ethics Committee approval under Health Research Authority guidelines. All service-user participants were approached voluntarily. Data were anonymised at source and analysed without identifiable information to protect participant confidentiality. Participants were offered clear information about the purpose of the evaluation and available sources of support, including referral back to Mind in Bexley for follow-up if any distress arose. The evaluation was designed in partnership with Mind in Bexley to ensure appropriate safeguarding procedures and to prioritise the dignity, autonomy and well-being of all participants.