Developing, delivering and evaluating a university-led cognitive behavioural therapy service for students

Abstract

Demand for student mental health services is growing, as is the complexity of presentations to university student wellbeing services. There is a need for innovative service delivery models to prevent students falling in the gaps of existing provision, where outcomes from traditional talking therapies services have been shown to be poorer for students than non-student peers. In 2018, Newcastle University established a pilot in-house cognitive behavioural therapy (CBT) service to provide high-intensity CBT for students at the university, harnessing the expertise of qualified and training staff from the psychological professions. This subsequently expanded into the Psychological Therapies Training and Research Clinic, appointing additional clinical staff. Here we present the journey of the clinic, from inception to implementation and expansion. We also present a descriptive evaluation of the first three years of operation, reporting on clinical activity, clinical outcomes and client experiences of the service. Data are presented from 605 referrals. Over 70% of referrals were assessed and over 60% transitioned into treatment. The treatment completion rate was 50%, with an overall recovery rate of 47.3% [using the same definition of recovery as NHS Talking Therapies for Anxiety and Depression (NHS TTAD)]. Satisfaction, measured by the Patient Evaluation Questionnaire, was high. These outcomes are commensurate or better than seen in NHS TTAD services for students and young adults. Overall, the clinic has been a successful addition to the wellbeing offer of the university and has provided a number of positive further opportunities for both research and the clinical training programmes.

Key learning aims

1. (1) To understand the process followed to establish a university-run cognitive behavioural therapy service for students and enable other institutions to replicate this model.

2. (2) To identify whether universities can deliver safe, effective mental health services that are fully evaluated and result in commensurate clinical outcomes to other service contexts.

3. (3) To reflect on key learning, challenges and ethical considerations in establishing such services.

Introduction

Between 2011 and 2021 there has been a 450% increase in student mental health declarations to the body managing the application process for UK universities (Universities and Colleges Admissions Service, 2021). In response to this, and with increasing awareness about the mental health needs of students (Lewis and Bolton, 2023), the provision of mental health services within universities is increasing (Broglia et al., 2018). The past decade has seen a sharp increase in demand from students for University Counselling services (The Insight Network, 2019) and, nationally, young people aged 18–24 access NHS Talking Therapies services at a higher level than any other age group (Office for National Statistics, 2022). However, students seen in Improving Access to Psychological Therapy services have poorer outcomes than their non-student peers (Barnett et al., 2022). Taken together, these data suggest that young people have a high and growing need for mental health support services, and that some of the factors unique to students may affect the accessibility and effectiveness of usual care (Dodd, 2021).

Furthermore, qualitative research identified that providers of university counselling services report seeing students who are experiencing more – or more complex – problems with their mental health and it is becoming increasingly difficult to meet the needs of students presenting with complex or urgent mental health needs (IFF Research, 2023). There is a need for higher intensity interventions (i.e. those that are more specifically tailored to the individual and administered over a longer time period), but students often fall through the gaps of existing provision. Whilst the NHS provides mental health care, there are marked barriers to students accessing this – for example long waiting times, dual addresses, not being registered with a local GP, and difficulties travelling off-campus to services (Student Minds, 2017; Salaheddin and Mason, 2016). In addition, person-related barriers such as the type and severity of psychological distress, privacy concerns, attitudes, fears, and other personal beliefs about receiving mental health services also negatively affect the accessibility of mental health services (Broglia et al., 2021; Eisenberg et al., 2012; Penchansky and Thomas, 1981; Saunders et al., 2006).

The majority of students seeking support for mental health difficulties will attempt to access support from their university: according to the Student Minds (2017) report, 91% of the students referred to NHS services had also accessed support from a university counsellor. Universities are well-placed to host services providing care for students – having a central administration, specialism in the age group, being easy for students to access, and having an understanding of the demands of the academic environment, as well as being well placed to manage some of these demands and refer for appropriate institutional support. However, mental health care (rather than pastoral support) is not part of a University’s core brief and the extent of a University’s duty of care to their students is a current area of debate (Hansard, 2023).

The mental health-focused services available to students at their university vary greatly from one institution to another, meaning the student experience of support is far from uniform. A common wellbeing offer – time-limited 1:1 counselling – has been the stalwart of traditional student services provision and has been helpful for many students (Broglia et al., 2023). However, students have described the challenges of this provision (Priestley et al., 2021) and large-scale evaluation of the clinical effectiveness of counselling services is lacking (Broglia et al., 2023).

The increased complexity of student presentations – with many meeting diagnostic criteria for depression, anxiety, eating disorders, psychosis, or other problems – suggests that counselling (which is not a recommended treatment for any of these conditions under NICE Guidance) may not be sufficient to meet the needs of all students seeking support. Many universities have been remarkably creative in response to this increase in complexity, identifying ways to work more closely with NHS services to ensure streamlined access to services for students (Broglia et al., 2022). Furthermore, supporting joint working between the NHS and wider services in supporting the mental health needs of students has become a focus of the government’s agenda more recently (Department for Education, 2023; Lewis and Bolton, 2023; Office for Students, 2022) and has been a highlighted need for over a decade (Royal College of Psychiatrists, 2011). But is it feasible and effective for universities to run their own high-intensity mental health services?

Universities provide the majority of training courses for staff who ultimately make up the psychological professions workforce. Universities that run such courses often employ clinically qualified staff with either dual-roles (an NHS role and a University role) or with previous experience working in the NHS. At Newcastle University, for example, we deliver the Doctorate in Clinical Psychology programme, the Postgraduate Diploma in Cognitive Behavioural Therapy and the Certificate in Low Intensity Psychological Therapies. Seeing the need for more high-intensity clinical provision for students and opportunities to retain professional accreditation for our staff, we set out to establish a pilot Cognitive Behavioural Therapy service run by the University. In line with the NHS Long-Term plan to increase numbers of qualified clinicians (Health Education England, 2021), the Clinic also supports the increased capacity requirement by providing supervised training placements for our students.

Here, we outline our experience of developing and embedding this service in the University’s existing pastoral support, including the process followed to set up the pilot service and the service’s subsequent expansion. We also report a descriptive evaluation of the service since January 2020 (the year the service expanded), including brief demographic details of the clients referred, a summary of clinical activity (throughput from referral to treatment, waiting times, average number of clinical sessions, and attendance data), and outcomes data (recovery rate and main discharge reason). We also report client feedback from structured interviews and responses to the patient experience questionnaire.

Method

1. Set up and structure of the pilot service

Setting up the pilot service involved a period of stakeholder consultation and the design of an initial service. Following a period of operation, the service expanded. This part of the Method section describes the stakeholder consultation and the structural elements of the service when it operated as a pilot service and after it expanded. The quantitative and qualitative data collected during the pilot phase and after the expansion are described in sections 2 and 3, respectively.

Stakeholder consultation

We began with a period of stakeholder consultation, including informal discussions with the Student Health and Wellbeing Service (SHWS), senior academic staff in the School of Psychology, and round table discussions with students (n=12; recruited through public advert, student welfare reps and the ‘Mind the Gap’ mental health society).

Pilot service

A pilot service began in February 2018, which involved two members of qualified clinical staff each working 0.1WTE (whole time equivalent), alongside six students from the Postgraduate Diploma in Cognitive Behavioural Therapy, and a clinic manager – the latter was the only additional appointment. The manager was a therapist trained in cognitive behavioural therapy (CBT) and accredited with the British Association of Behavioural and Cognitive Psychotherapies (BABCP), and they were paid on a sessional consultancy basis for 0.1WTE. They were responsible for co-ordinating referrals and allocations, as well as providing supervision.

The pilot service initially ran from rooms in the SHWS in Newcastle University’s central campus and used their record-keeping system, before moving to a bespoke system more suited to clinical purposes and outcomes reporting. Clinical Governance arrangements were developed covering supervision, indemnity, risk assessment, service hosting, data security, and outcomes monitoring. The process of developing the governance arrangements subsequently formed the basis of the University Mental Health Advisers Network (2023) guidance for clinical governance within universities.

Given the limited capacity of the service, referrals were only open when a therapist had availability, in order to avoid very long waiting times. At this stage in the clinic’s operation, the clinic co-ordinator contacted the SHWS directly, letting them know how many new clients could be referred and once that number was reached, new referrals could not be made until a space was available. Clients were referred to the service by the SHWS team, with guidance to refer students self-presenting to the University Counselling Service with anxiety or depression and scoring either 10 or above on the PHQ-9 (Kroenke et al., 2001), or 8 or above on the GAD-7 (Spitzer et al., 2006). Referrals were assessed for suitability for CBT. Those assessed as not suitable for CBT were referred elsewhere as appropriate.

Expansion to a full-time service

Although the outcome of the pilot evaluation is provided in full below, in summary, the pilot demonstrated that a high-intensity CBT service could operate safely and effectively in a higher education context. When exploring opportunities to expand, the clinic team were successful in securing funding through the Office for Students Mental Health Challenge Competition (Office for Students, 2020). As part of the matched funding for this bid, the University built a dedicated suite of rooms – the Cedar Clinic – which gave the clinic its own space and identity.

The additional funding was used to appoint two full-time BABCP-accredited CBT therapists, a 0.5WTE clinic manager (also a BABCP-accredited CBT therapist) and a 0.6WTE administrator. From February 2020, it began operating as a full-time service, offering appointments every day of the week with some evening clinics available.

The service has a simple care pathway (Fig. 1). Referrals from the SHWS are assessed for suitability for CBT and those suitable are allocated to the waiting list for either a trainee or qualified practitioner (clinical judgement is used to determine whether the client is allocated to a trainee or qualified practitioner based on the chronicity and severity of the presenting problem, whether the client has more than one disorder, and the presence and severity of any risk). If unsuitable, appropriate alternative services are identified with referrals either made directly, or via the GP. In all instances (and with client consent), the client’s GP is informed at assessment and discharge that they are being seen/have been seen by our service.

Figure 1. Service care pathway.

Relationship to wider University processes and structures

The service is confidential and independent of the SHWS, using its own database and records keeping system (although there is regular communication between the teams about referrals and discharges). Academics are not informed that their students are using the service. Increase in risk to harm to self is managed in the same way as in other primary care settings, involving the GP and crisis team if necessary and appropriate.

The records of students referred to the service who are known to staff or are peers of trainees working in the service are locked down so only the administrator and allocated clinician can see the record, and, if necessary, alternative meeting space elsewhere in the university can be used. Whilst it is a theoretical possibility that a student could be referred to us who raises fitness to practise or fitness to study concerns, these have not arisen in practice, and this can happen in any context and the same process would be followed as if it happened outside of the university setting. The clinic provides evidence of attendance for treatment when a student requests it, and this might form part of a submission for a fitness to study or fitness to practise process, but these would not be processes that are triggered or started by staff in the clinic.

2. Quantitative evaluation data

Pilot service

At referral, brief demographic details are collected from students via a self-completed client registration form. Routine outcomes monitoring was embedded from the beginning and aligned to NHS Talking Therapies for Anxiety and Depression [NHS TTAD; formerly Improving Access to Psychological Therapies (IAPT)] services. Measures of depression (PHQ-9), anxiety (GAD-7), and social function [Work and Social Adjustment Scale (WSAS); Mundt et al., 2002] were taken at every session, and, where applicable, anxiety disorder-specific measures at the start and end of treatment and every six sessions in between.

Further metrics were derived from the questionnaire measures, including whether the client met clinical caseness at the start and end of treatment [a score ≥8 on the GAD-7 or ≥10 on the PHQ-9 or above the relevant threshold on a disorder-specific measure (National Collaborating Centre for Mental Health, 2021)], reliable improvement (a score improvement of 4 or more points on the GAD-7 or 6 or more points on the PHQ-9), reliable deterioration (a score worsening of 4 or more points on the GAD-7 or 6 or more points on the PHQ-9), recovery (a score in the final session of treatment <8 on the GAD-7 or <10 on the PHQ-9), and reliable recovery (both the criteria for reliable improvement and recovery are met).

Service use metrics were also collected, including the number of referrals, the number of referrals who were offered and accepted treatment, waiting time in days from referral to first treatment session, number of discharges and reason for discharge. We also recorded the number of clients who completed treatment [at that point, we defined treatment completers as ‘clients discharged at the point when the full course of required sessions had been delivered or the point at which a mutual agreement between therapist and client was made that further sessions were not necessary’; we note that this deviates from the definition used in NHS TTAD, but the NHS TTAD definition of any client that received two treatment sessions was used from January 2020 onwards (see below)].

Additionally, research was embedded within the day-to-day practice of the clinic. All clients are given an information sheet and asked for their consent to use routine clinical outcomes data for research purposes (approved by Newcastle University’s Faculty of Medical Sciences Research Ethics Committee, code 2004/4677/2020). At assessment, in addition to the standard mood, anxiety and functioning measures, students complete an additional battery of measures including a broad screen for mood and anxiety disorders, a measure of childhood trauma and of daily hassles.

Full-time service

In January 2020, the clinic began to use IAPTus (https://iaptus.co.uk/) as its record keeping system. IAPTus is a cloud-based electronic patient record which allows for safe, confidential data storage and ease of reporting clinical outcomes. At this point, the clinic began to use the same definitions as NHS TTAD for all reporting outcomes (National Collaborating Centre for Mental Health, 2021). The routine data collected and the frequency at which it was collected did not change, but the record-keeping system enabled more streamlined data-reporting.

At discharge, all students are sent the link to a Patient Experience Questionnaire (PEQ; Table 3), that they can complete anonymously to rate their experience of the service and provide feedback. This includes both quantitative and free-text response data.

3. Qualitative evaluation data

All clients discharged between September 2019 and March 2022 were offered the opportunity for a qualitative interview with an independent member of the team not involved in their care. This was not part of usual care and was specifically for research and evaluation purposes. Eight individuals consented to an interview and a further five opted to respond to the interview questions by email. For the interviews, open questions were created and developed through consultation of the relevant academic literature and discussions within the research team. The questions used in the interviews explored the overall experience for the clients, areas of improvement for the service, and positive aspects.

Interview transcripts and free text responses from the PEQ were analysed using framework analysis. Framework analysis fits under the broad umbrella of thematic analysis and was originally developed for policy research as it concentrates on finding solutions to practical problems (Ritchie et al., 2003). More recently, it has successfully been applied and used within healthcare research (Banck and Bernhardsson, 2020; Cree et al., 2015; Sim et al., 2023).

In the current study, analysis of the transcripts and PEQ free text responses followed the seven stages of the framework method: transcription, familiarisation with the interview, coding, developing an analytical framework, applying the analytical framework, charting data into the framework matrix, and interpreting the data (Gale et al., 2013). Based on the level of richness within the qualitative dataset and the scale of the project, it was deemed appropriate to present the findings as topic summaries, rather than in-depth, rich themes usually associated with thematic analysis (Braun and Clarke, 2022).

Firstly, each interview transcript and PEQ free text response underwent multiple readings to facilitate the researchers’ immersion in the data. Then four interview transcripts were inductively coded, and the codes generated at this stage were used to develop a first draft of the analytical framework. This framework was then applied to the remaining four interview transcripts. Any discrepancies or additions in coding were discussed within the research team, leading to adjustments in the analytical framework where needed. Once no new codes were required, the revised framework was then used to code all answers from email interviews and the PEQ free text responses. Next, similar codes were grouped into categories which were used to construct the framework matrix, allowing for charting of the data. Finally, topic summaries and sub-topic summaries were created and named to describe each category and were supported with data.

Results

1. Pilot service

Outcome of stakeholder consultation

Discussions with SWHS highlighted a number of practical issues (e.g. how suitable students would be identified and referred; GDPR, security and confidentiality; room usage; indemnity cover). In the round table discussions, student views about introducing a CBT service to the university were almost universally positive, especially at the prospect of a higher number of sessions and more streamlined access based within the footprint of the university. Concerns of sharing sensitive personal information with the university was raised by one student, who said they would prefer to access services unconnected with the university, but this concern was not widely held.

Clinical activity

In alignment with open science and open clinical practice, documentation relating to the clinic’s processes, outcome measures and outcomes were hosted on the Open Science Framework (https://osf.io/cwa7u/).

In September 2019, after 18 months of operating for 1 day per week, the pilot service had received 45 referrals, of which 33 had been offered and accepted treatment (https://osf.io/f8qwp/). Average waiting time from referral to the first treatment session was 70.7 days (note: once treatment begins, it continues in one continuous block of sessions – there are no additional waits). Twelve referrals were not offered treatment in that time frame – one was waiting for assessment, two were discharged due to non-attendance, four declined treatment, and the clinic was unable to meet the student’s needs for five individuals (for example due to high levels of risk that could not be safely managed in the setting, significant substance abuse, or presenting with a problem for which CBT is not an indicated intervention). These students were referred to appropriate alternative sources of help.

Clinical outcomes

Of those whose treatment ended between February 2018 and September 2019, 18 met our criteria for completing treatment. Of the clients meeting caseness at session 1, 70% met criteria for reliable recovery at discharge on the PHQ-9 and 70% on the GAD-7. Conversely, four clients were treatment non-completers, having stopped attending sessions without prior discussion with the therapist. For these four individuals, of those meeting caseness at session 1, only one met criteria for reliable recovery at the point treatment ended. There were no adverse events reported.

2. Full-time service

Data are reported for the 605 referrals received between 1 January 2020 and 31 August 2023 (Tables 1 and 2). Data are broken down by academic year (September to August) and also reported in total over the whole time period.

Table 1. Gender identity, disability status and ethnicity of student referrals between 1 January 2020 and 31 August 2023

Table 2. Clinical activity and outcomes

Table 3. Mean (standard deviation) of scores on the Patient Experience Questionnaire

The first seven items are rated on a 1 to 5 scale, where 1 is ‘never’ and 5 is ‘at all times’. ¹ n=77; ² n=39 (this question was introduced later).

Demographic characteristics

The majority of referrals identified as female (67%), with 30% identifying as male, and 1% defining their gender in another way. Twenty per cent of referrals identified as having a disability and 59% reporting no perceived disability. The remaining data relating to disability status was either unknown or missing. With regard to ethnicity, 64% of individuals referred were white, 6% were Asian or Asian British, 4% of mixed ethnic background, 4% from other ethnic groups, and 2% were black. However, a sizeable proportion of this data is missing (20%), typically because this question was not answered by the student, so the true rates of referral for students from different ethnic groups is not known.

Clinical activity

Of all referrals that began in the time period, 76% were assessed and 62% entered treatment. The 24% of referrals not assessed dropped out in the waiting period. The mean wait in days from referral to first treatment session was 92.6 days. The mean number of treatment sessions per referral that entered treatment was 13.6. In total, there were 6698 assessment and treatment sessions offered in the time period. The average non-attendance rate was 6% and the cancellation rate was 10%.

In terms of observed changes over time, there was a markedly higher number of referrals and clinical sessions in the 2022–23 academic year compared with previous years, which was also reflected in a lower percentage throughput to assessment and treatment, as well as a longer waiting time. The two academic years that were most-affected by the global COVID-19 pandemic were relatively stable with regard to referral numbers and other indices of clinical activity. The non-attendance rate and cancellation rate remained relatively stable over time.

Clinical outcomes

The two main groups of outcomes reported here are reason for discharge and recovery rate. These outcomes are reported for referrals that ended in the relevant time period.

Over the whole time period, the main reason for discharge was completing treatment (50%). One-third of referrals (33%) ended treatment before the care professional planned, and almost one-fifth (17%) were not suitable for the service and/or were referred elsewhere. The percentage of referrals completing treatment remained relatively stable over the three full academic years reported. However, there was a 9% increase in referrals that ended before the care professional planned from 2020–21 to 2022–23 and a commensurate decrease (11%) in unsuitable referrals or those that were referred elsewhere.

The overall recovery rate was 47.3%, with 45.3% meeting criteria for reliable recovery. The rate of reliable improvement was 71.1% and the rate of reliable deterioration was 6.2%. These rates have varied much more over the time period, with a marked drop in recovery (to 40%) and reliable recovery (to 38.9%) rates in the 2021–22 academic year.

3. Patient experience

Data from 78 students who completed the PEQ are summarised in Table 3. Satisfaction is high across the dimensions rated, with average ratings of 4.7 or above for the standard patient experience items (questions 1 to 5). In March 2022, two additional items were added to assess the extent to which access to treatment in this service has enabled students to engage better with their education and contributed positively to their educational experience. Ratings for these questions were also high (mean of 4.4 and 4.5, respectively).

Framework analysis from the interviews and PEQ free text responses generated three main topic summaries (‘Past experience and current needs’, ‘Experience of therapy’ and ‘Impact of therapy’), each with three or four sub-topic summaries. For the first topic summary, ‘Past experience and current needs’, many of the students discussed their experiences before accessing the clinic and how this impacted on their own expectations and hopes for their therapy sessions and outcomes. The second topic summary, ‘Experience of therapy’, looks at the therapy journey and the experiences and feedback from the students regarding the therapy they received. The sub-topic summaries in this section focus on the therapist relationship and competencies, the experiences within the sessions, the CBT approach, and the logistics of the sessions during this time. The final topic summary, ‘Impact of therapy’, is based on the impact of the therapy and the students’ experience and feedback once they had finished therapy. The feedback from the participants covered personal outcomes, dealing with future challenges and recommending the clinic to others.

Table 4 provides an overview of the topic summaries along with descriptions and supporting quotes from the participants.

Table 4. Topic summaries with supportive quotes from patient interviews and PEQ responses

Overall, the students were generally positive about their experiences but also raised some broader issues regarding waiting time and communication from the clinic that will need to be addressed by the service.

Key learning and reflections

This service began as an idea that we had that seemed to bring several ‘wins’ – for students, for staff on the clinical training programmes, and for clinicians in training. The service has grown organically from the initial question we had of whether this was possible and safe, seizing on opportunities that presented themselves along the way (e.g. the Office for Students Mental Health Challenge competition), rather than growing into a particular goal or service design that was present from the outset. As such, it was not designed to mirror directly other existing services (e.g. NHS TTAD, with all levels of stepped care available), but it has always been centred on a commitment to the provision of care that is supported by the best current evidence base and the embedding of routine outcomes and evaluation.

There has been a lot of concrete learning along the way, as well as opportunities to reflect on the issues raised of doing something out of the ordinary. The more concrete learning is included here and broader reflections in the Discussion.

Some of the concrete learning includes things like the challenges of employing clinical NHS staff on university pay grades. In order for the clinical positions to be attractive to appropriately qualified individuals, they needed to be offered at competitive salaries. However, the nature of the posts – when subject to university grading procedures, which emphasise teaching and research responsibilities – meant they were graded at too low a level to be a competitive offer. This took substantial time and negotiation to resolve.

Establishing adequate clinical governance arrangements in the absence of sector guidelines or recommended practice also took significant time. Wellbeing services have developed out of pastoral care provided by universities for their students and have traditionally had greater emphasis on support and the educational experience rather than active treatment of mental health disorders. However, there is a growing need for formalising how Wellbeing Services in universities are governed, as students present to Wellbeing Services with increasingly complex difficulties and there are greater pressures presented by the challenges of calibrating expectations for support services versus those providing treatment. There has been a clarion call by organisations such as Student Minds, the University Mental Health Charter, and the University Mental Health Advisors Network (UMHAN) for universities to develop clearer governance for their wellbeing and mental health services. As mentioned previously, using our own experience in this area, we developed foundation guidance that was used as the basis of a UMHAN consultation exercise to develop recommendations for clinical governance arrangements in universities (University Mental Health Advisers Network, 2023).

There is marked seasonal demand in our service with the majority of referrals coming in October to March. Although our service offers students online appointments if they return home during the holidays, a large proportion of students do not take up this offer. This makes service planning challenging, as the staff hours in the service remain constant throughout the calendar year. Also, students tend to have lower attendance around exam time and frequently request to defer treatment or assessment until, for example, a work placement or international placement has finished. We have found there is a need for flexibility in accommodating students’ needs around starting their treatment, as these hurdles are not of their own creation.

There are also challenges presented by students ending their course – either through finishing their period of study or choosing to terminate their studies early. Our indemnity insurance only covers a short period of time after the student is no longer registered with the university. This presents barriers to starting treatment with students who are near the end of their period of study.

Discussion

Establishing our service has demonstrated that University-led high-intensity CBT services can be safe and effective, as well as offering relatively short wait times to access this level of treatment. Between 1 January 2020 and 31 August 2023, the service received 605 referrals. The majority of referrals transitioned through to treatment and 50% of referrals completed treatment. The overall waiting time from referral to first treatment session was approximately 3 months and the overall recovery rate was 47.3%. Student experience of the service indicated high satisfaction and qualitative findings from interviews and the free text boxes of the Patient Experience Questionnaire indicated that students felt they had good rapport with their therapists, had trust in their therapeutic skills, and felt supported and in control of their sessions. They found the on-campus location easy to access, but not all found the online provision of sessions necessitated by the pandemic met their needs. The service also hosted clinicians in training on the Diploma in Cognitive Behavioural Therapy and the Doctorate in Clinical Psychology, which increasingly plays an important role in allowing the workforce expansion as outlined by Health Education England’s Psychological Professions plan (Health Education England, 2021), and also has supported the development of specialist training pathways for those wanting to attain BABCP accreditation during their doctorate in clinical psychology training.

The clinical outcomes we report compare favourably with outcomes for students and young people in other contexts. For the 2021–22 financial year (the most recent data available), the national recovery rate in England for 18- to 25-year-olds in NHS TTAD was 46.4% (Baker, 2023), which is very similar to the average rate we found over the whole time period reported (47.3%). Furthermore, outcomes for students have been shown to be poorer than for comparable non-students in NHS TTAD. A large retrospective cohort study covering a 12-year period contrasted outcomes for students versus non-student peers attending Improving Access to Psychological Therapies services in North Central and East London (Barnett et al., 2022). They reported a reliable recovery rate for students of 41.9%, whereas for non-students it was 47.8%. Our reliable recovery rate of 45.3% sits between these two figures and suggests, at the most conservative interpretation, that our service delivered commensurate clinical outcomes to NHS TTAD services for students.

The lower recovery rate reported in the 2021–22 academic year was one of the main years affected by the global pandemic, which is known to have disproportionately affected young people’s mental health, as well as other groups that students often fall into, such as those living in rented accommodation and those with low incomes (Office for National Statistics, 2021). The sharp rebound in the 2022–23 academic year to recovery rates more commensurate with the rates seen prior to the pandemic would lend support to the likelihood that this unusual drop was related to the pandemic in some way. However, without a longer time series and, as a relatively small service where larger fluctuations in statistics year on year may be more likely, it is impossible to say with certainty that this was the reason.

There is increased awareness of the mental health needs of certain groups that are traditionally marginalised, do not have equitable access to services, or who do not report benefit from traditional psychological therapies. This includes individuals from racially or ethnically minoritised communities (Stoll et al., 2022), neurodivergent individuals, women and those from the LGBTQ+ community (Lewis and Bolton, 2023). The ways in which students from these groups have access to this service is being explored in ongoing research.

Universities are approaching the challenges of linking students up with appropriate mental health care differently and it is interesting to see the variety of models being explored, as well as consider the effectiveness, impact, and scalability of each different approach. The 2019–22 Office for Students Mental Health Challenge Competition was focused specifically on creating a ‘stepped change’ in mental health outcomes for students (Office for Students, 2020). Whilst a variety of projects were funded under this call, of those focusing directly on intervention, some (such as ours) used in-house provision of services, whereas others brought NHS services closer to students, either with NHS-based student-specific liaison services or seconding NHS staff into the University environment. Unfortunately, there was no one common set of outcomes to enable straightforward comparison of the impact of different approaches.

Some of the advantages of our in-house service are shared by other approaches that make services more accessible to students (i.e. those that bring services onto campus, making them easier for students to access). Other advantages were unique to our service setting, for example we were able to see clients irrespective of the location of their GP, provide the option for online therapy to continue if the student returned home during the holidays or temporarily relocated for a placement year, and link students with additional/alternative support networks in university. From a staffing perspective, some of our staff attained additional academic roles alongside their therapy work, thereby diversifying their workloads and providing progression opportunities. Furthermore, from a research point of view, being able to incorporate research and evaluation routinely within the operation of the clinic has opened up opportunities for funded and unfunded research (projects linked to the clinic have attracted almost £1 million in external funding since its inception).

The success of our trial of offering in-house CBT for students has seen an additional portfolio added to the service, namely a Neurodevelopmental Assessment Service, whereby students can access assessments for autism spectrum disorder and attention deficit hyperactivity disorder. This service is also staffed by qualified and trainee clinicians.

However, there are question as to how straightforward our model is to replicate. Employing additional staff or adding to the university estate is costly. The necessary expertise and appropriate governance is required to operate services safely and is not present in all Higher Education Institutions; offering mental health services in universities risks blurring the boundaries between education and health care; and university-delivered services divert some of the already small pool of qualified practitioners away from the NHS to serve the needs of a relatively small group that typically have better overall outcomes across a range of indices over the life span (Britton et al., 2020; Davies et al., 2018; Raghupathi and Raghupathi, 2020).

Universities with existing clinical training programmes are in the strongest position to replicate this model and a variety of possibilities exist that circumvent some of these potential objections – e.g. fully trainee-led services, or hub-and-spoke models that serve more than one University. With further expansion of the psychological professions required by the NHS, we believe that such provision could form an important part of universities’ capacity planning.

Initiating a service like this prompts broader questions about the appropriateness of using university money for healthcare and also sustainability questions given the rising demand for mental health services – where should the line be drawn on the scale of provision? These are questions that are difficult to answer directly with data, as they draw more on ideas about what is right, proportional, appropriate and necessary, not simply what is possible.

With regard to sustainability, referrals have stayed relatively constant over the past couple of years and the clinic has been of wider benefit to the university through its ability to enhance existing programmes (e.g. providing placements for students, which facilitated the accommodation of a 100% increase in trainee numbers on the Doctorate in Clinical Psychology programme), to enable clinically qualified staff to remain clinically active in a way that dovetails more easily with their other role(s) and even to increase their skills portfolio by achieving BABCP accreditation through working into the clinic, and through profile-raising and achieving external funding awards. Longer-term follow-up of students who have been seen would be informative to understand the service’s impact and wider consideration of the effect the service has had on educational attainment and experience rather than just symptom resolution would be interesting. Whilst we did secure ethical approval to follow students up for 5 years post-discharge, we have been unable to implement this due to technical and staffing challenges.

It would also be interesting to map out the downstream effects on other services, such as NHS Talking Therapies, to identify whether this service has filled a gap and offered treatment to students who otherwise would not have gone elsewhere, or whether it has transitioned students who would typically be seen in an NHS context into the University context instead. A full economic evaluation would be a useful exercise, although it is noteworthy that neither existing NHS services nor Student Wellbeing Services are commonly subject to a full economic evaluation, so there would be limited comparators to contextualise any findings.

Overall, establishing this service has been an exercise in learning and overcoming barriers and has required proactive problem-solving and filling gaps where we have been able to. We believe the service is unique and wanted to document our journey for the benefit of those who may have envisaged something similar, but have not had a roadmap to follow. We are also keen to add to the evidence base for student mental health provision, an area where evaluation is sorely needed. The service we have developed has successfully integrated alongside existing provision, has met a need for students, staff and clinicians in training, and for our clinical training programmes. We have demonstrated that it is possible for universities to run their own services for their students, where existing enablers are present, and they can deliver positive clinical outcomes and high service satisfaction.