A pilot evaluation of a training and supervision pathway for Augmented Depression Therapy (ADepT) in a Devon NHS Talking Therapies Service

Abstract

Abstract

Augmented Depression Therapy (ADepT) is a novel wellbeing and recovery-oriented psychological treatment for depression. A recent pilot trial run in a university clinic setting suggests ADepT has potential to be superior to cognitive behavioural therapy (CBT) at treating anhedonic depression in a NHS Talking Therapies for anxiety and depression (NHS–TTad) context. Before proceeding to definitive trial in pragmatic settings, it is important to establish if therapists in routine NHS-TTad settings can be trained to deliver ADepT effectively and to assess therapist views on the feasibility and acceptability of ADepT in this context. A bespoke training and supervision pathway was developed (2-day workshop, four 2–hour skills classes, and 6 months of weekly supervision) and piloted with 11 experienced therapists working in a single NHS–TT service in Devon. Nine out of 11 therapists completed the placement, treating 24 clients with a primary presenting problem of depression; 21/24 completed a minimum adequate dose of therapy (≥8 sessions), with 17/24 (71%) showing reliable improvement and 12/24 (50%) exhibiting reliable recovery. Eight out of nine therapists submitted a session for competency assessment, all of whom were rated as competent. Nine therapists submitted feedback on their experiences of training. Eight out of nine therapists felt the ADepT model would be effective in an NHS–TTad context; that training was interesting, useful, well presented and enhanced their own wellbeing; and that they felt sufficiently skilled in core ADepT competencies at the end of the placement. This suggests that NHS–TTad therapists can be trained to deliver ADepT competently and view the treatment as feasible and acceptable.

Key learning aims

1. (1) To become familiar with the Augmented Depression Therapy (ADepT) approach for enhancing wellbeing in depression.

2. (2) To evaluate the potential utility and feasibility of ADepT model in NHS Talking Therapies Services (NHS–TTad).

3. (3) To understand the pilot ADepT training and supervision pathway for CBT therapists in NHS–TTad services.

4. (4) To consider the opportunities and challenges of training therapists to deliver ADepT in NHS–TTad services.

Introduction

Depression is a widespread, disabling, chronically recurrent condition that is a significant contributor to global disability and results in extensive social and economic costs (Judd, 1997; Kessler et al., 2003; König et al., 2020). Change during current gold-standard psychotherapies like cognitive behavioural therapy (CBT; Beck et al., 1979) and behavioural activation (BA; Martell et al., 2010) is clinically significant but sub-optimal (Cuijpers et al., 2013). In the clinical trials literature, around half of those with depression remit, half of whom relapse within 2 years, which means sustained recovery rates are around 25% (Cuijpers et al., 2014; Vittengl et al., 2007). There is a pressing need to innovate novel therapies that can lead to enhanced treatment outcomes, that are cost-effective, and that can be implemented in routine care settings.

In a UK context, a majority of psychotherapy for depression is delivered in primary care settings via NHS Talking Therapy Services for Anxiety and Depression (NHS–TTad). Formerly known as the Improving Access to Psychological Therapies (IAPT) initiative, NHS–TTad services provide a stepped-care model. Milder cases of depression are treated using a short dose of low-intensity therapies delivered by Psychological Wellbeing Professionals (PWPs). More complex cases of depression (or those who have not responded to low-intensity care) are treated using a longer dose of evidence-based treatments like BA or CBT delivered by high-intensity therapists (HITs). Outcomes in NHS–TTad broadly mirror the depression clinical trials literature. National outcome data suggest around two-thirds of clients treated in NHS–TTad settings show reliable improvement, around 50% meet reliable recovery criteria (no longer showing clinically significant depression or anxiety post-treatment and exhibiting reliable improvement) and approximately 6% of clients show reliable deterioration. When focusing on the group of clients reporting a primary problem of depression, and receiving high-intensity CBT, rates of reliable recovery drop to around 40% according to 2021–2022 outcomes (NHS Digital, 2022).

Those who do not recover in NHS–TTad settings are typically not offered any further intervention, resulting in a significant ongoing disability burden for these clients. Moreover, many of those who do recover will experience a relapse or recurrence. While NHS–TTad does not routinely track relapse/recurrence rates of those receiving treatment, there is emerging evidence that a key driver of treatment return is relapse – with around 32% of treatment returners having recovered during previous NHS–TT treatment and subsequently relapsed (Lorimer et al., 2023).

One way to improve psychotherapy treatment outcomes in NHS–TTad settings and beyond is to target clinically important features of depression that are not effectively addressed by current psychotherapies like CBT and BA. The two core symptoms of depression are pervasive depressive mood and anhedonia (reduced interest/pleasure in previously enjoyable activities). These symptoms are believed to emerge from disruptions to two partly dissociable neurobiological systems (Insel, 2010). Over-activation of the negative valence system (NVS) drives withdrawal behaviour and increases levels of negative affect (NA), while under-activation of the positive valence system (PVS) inhibits approach behaviour and reduces positive affect (PA) and broader wellbeing (experience of pleasure, meaning and social connection). Existing depression psychotherapies like CBT and BA effectively repair depressed mood but not anhedonia and wellbeing, meaning many individuals experience residual PVS dysfunction after treatment (Alsayednasser et al., 2022; Dunn et al., 2020; Widnall et al., 2020).

Greater levels of anhedonia and reduced wellbeing have been shown to predict reduced likelihood of remission and greater risk of relapse/recurrence (Alsayednasser et al., 2022; Dunn et al., submitted), meaning a failure to repair PVS disturbances is contributing to sub-optimal symptom relief outcomes. Moreover, service-user definitions of recovery from depression prioritise reconnecting to positive affect and wellbeing to the same degree as repair of depressed mood (Cummergen et al., 2022; Fusar-Poli et al., 2023; Zimmerman et al., 2006), suggesting improving PVS function is of value in it its own right, irrespective of broader symptom relief outcomes.

Augmented Depression Therapy (ADepT) has been developed to target both depressed mood and wellbeing in clients presenting with anhedonic depression (Dunn et al., 2019a). ADepT is a solution-focused, cognitively augmented, behavioural activation approach. Reflecting that depression often follows a recurrent, relapsing course, ADepT conceptualises depression as a chronic vulnerability that individuals need to learn to live well ‘alongside’ (a recovery focus) rather than an acute condition to ‘cure’. ADepT is an integrative approach that builds on core CBT and BA techniques, integrating into them third wave approaches and aspects of positive psychology and solution-focused therapy. The delivery format is 15 acute individual sessions delivered weekly, with five flexibly scheduled booster sessions offered in the year after acute treatment has finished. Treatment involves identifying client values; behaviourally activating clients to work towards values consistent goals; and overcoming barriers to being resilient (managing challenges to reduce NA) and thriving (taking opportunities to maximize PA). Building wellbeing (capacity to experience pleasure, meaning and social connection in life) and functional recovery is the primary focus, with depression conceptualised as patterns of thinking, feeling and behaving that serve as barriers to achieving this goal. Therapists adopt a positive, future-focused, solution-focused and reinforcing style that helped clients develop the capacity to notice and utilise strengths. The end phase of therapy develops a wellbeing plan to continue progress over the coming months and booster sessions review and troubleshoot progress with this plan. To enhance implementation, ADepT has been designed so that existing CBT/BA therapists will be able to deliver it with minimal additional training (in a UK context high-intensity therapists working in NHS–TTad services). ADepT has also been designed to be of a similar dose to standard care (CBT/BA) and is therefore likely to be comparable in delivery cost.

Early evaluation of ADepT is encouraging. A multiple randomised baseline case series preliminarily examined feasibility, acceptability and clinical efficacy of ADepT in 11 individuals with depression primarily recruited from NHS–TTad high intensity waiting lists (Dunn et al., 2019a). ADepT was acceptable to both therapists and clients, resulted in large effect size and clinically meaningful improvements in depression, anhedonia and wellbeing, and gains were largely sustained over one year follow-up.

In a pilot randomised controlled trial, 82 depressed clients with anhedonic features were recruited predominantly from NHS–TTad high-intensity waiting lists and randomised to receive either 20 sessions of CBT (following a Beckian protocol) or 15 weekly sessions of ADepT plus up to five optional booster sessions (Dunn et al., 2019b; Dunn et al., 2023). The pilot trial met its feasibility aims, in that clients could be recruited, randomised, retained, engaged adequately with each treatment, and provided an adequate amount of follow-up data. Within-arm analyses showed ADepT led to statistically significant and large magnitude effect size improvements from intake to 6-month assessment in the co-primary wellbeing and depression outcomes and a range of secondary outcomes including anhedonia, anxiety and functioning. As this was a pilot trial, it was not powered to examine the statistical significance of between-group differences in clinical outcomes, nor to formally assess likely cost-effectiveness of ADepT relative to CBT. Nevertheless, examination of the 95% confidence intervals of the between-group clinical effect sizes suggested that ADepT has promise to be superior to CBT. On the co-primary depression and wellbeing outcomes, the lower bound of the confidence interval did not cross into the range where CBT could be superior to ADepT; and both the mean effect size estimate and the upper bound of the confidence interval fell into the range where ADepT could be superior to CBT at the primary outcome point. When considering anhedonia outcomes, the entire confidence interval fell into the range where ADepT was superior to CBT, which would meet formal frequentist criteria for superiority. Exploratory Bayesian analyses found that clients had a <1% chance of showing a minimum clinically important disadvantage and around a ≥50% chance of showing a minimum clinically important advantage if treated with ADepT rather than CBT for both depression and wellbeing outcomes. These gains largely held at longer-term (18–month) follow-up, with numerically lower rates of relapse/recurrence at 18–month follow-up in ADepT (of 80% of clients who remitted at the primary outcome point, 33% had relapsed) relative to CBT (of 56% of clients who remitted at the primary outcome point, 61% subsequently relapsed). Exploratory cost-effectiveness analysis showed ADepT cost a similar amount to deliver as CBT but resulted in greater quality of life (quality adjusted life years; QALYs) gains, leading to a ≥80% probability of likely cost-effectiveness when using a health-related QALY measure and a ≥96% probability of likely cost-effectiveness when using a wellbeing-related QALY measure. These results suggest ADepT in this trial was non-inferior to CBT (current best practice), establishing it as a viable treatment for anhedonic depression. Furthermore, ADepT shows potential to be superior to CBT, but this now requires testing in a definitive trial in NHS–TT setting with an adequate sample size to test this inferentially.

Before considering implementing ADepT more broadly in NHS–TTad settings, and proceeding to a pragmatic definitive trial to formally test if it is superior to CBT, caution is however warranted. ADepT in the initial cases series and pilot trial was delivered in a university research clinic setting. The therapists in the cases series were experienced research therapists, who had learnt and delivered other novel therapy protocols prior to being introduced to ADepT. The ADepT arm therapists in the pilot trial were a mixture of the therapists previously involved in the cases series and two high-intensity IAPT therapists learning ADepT for the first time. Due to the small number of therapists who needed to be trained in ADepT, it was possible to tailor and personalise the training these two new ADepT therapists received during the pilot trial. Any larger scale roll-out of ADepT in NHS–TTad settings, whether for the purposes of definitive trial or routine care, will require a more formal training and supervision pathway to be developed. It is critical that this pathway is feasible and acceptable to key stakeholders and can train therapists to deliver ADepT competently. It is also important to further assess the feasibility, acceptability, and potential efficacy of the ADepT intervention in more routine care settings (and to look for ways to refine the protocol to ensure optimal fit with the NHS–TTad context) before proceeding to a pragmatic definitive trial. Training therapists in routine practice to deliver ADepT, examining their clinical outcomes and canvassing their views on the model provides one way to achieve this aim. Considering implementation early on in the development phase is consistent with the MRC revised guidance for developing complex interventions (Skivington et al., 2021), which emphases the importance of understanding how the intervention interacts within the context within which it is implemented, and the MRC guidance on process evaluation of complex evaluations, which emphasises the importance of understanding the implementation process so that what is delivered has fidelity to the treatment protocol (Moore et al., 2015).

To explore these implementation issues, the Devon NHS–TTad service (TALKWORKS) that referred clients into the original ADepT pilot trial agreed to take part in a mixed-methods pilot evaluation of an ADepT training and supervision pathway in 2021. An audit of this pilot is reported here to examine the feasibility, acceptability and potential efficacy of ADepT (and a training and supervision pathway to support it) in a routine NHS–TTad service settings. The following feasibility and acceptability questions were explored:

• Would NHS–TTad therapists express interest to take part and complete the training and supervision pathway?

• Could NHS–TTad therapists be trained to reach competence in ADepT by the end of the training and supervision pathway?

• Could NHS–TTad therapists achieve similar engagement and clinical outcomes to those observed in the ADepT pilot trial after completing the training and supervision pathway (and that were broadly comparable to or superior to outcomes observed for other high-intensity depression treatments in NHS–TTad settings)?

• What are NHS–TTad therapists’ views on the feasibility, acceptability and effectiveness of the training and supervision pathway, and did they perceive themselves as competent upon completing it?

• What are NHS–TTad therapists views of the feasibility, acceptability and likely efficacy of implementing ADepT in routine NHS–TTad settings (and what adaptations to the existing protocol might this require)?

As well as informing the further development, evaluation and implementation of ADepT in NHS–TTad settings, findings may also be relevant to the opportunities and challenges around implementing any new care pathway within this clinical setting.

Method

A training and supervision pathway for ADepT was piloted in a single NHS–TTad service in Devon (TALKWORKS), with an audit conducted alongside the pilot to assess feasibility, acceptability and clinical outcomes. At the time of the audit, TALKWORKS employed around 60 high-intensity therapists across six teams. Ten training places were offered to TALKWORKS, with supervision and training delivered by experienced ADepT therapists from the Accessing Evidence Based Psychological Therapies (AccEPT) clinic at the Mood Disorders Centre, University of Exeter. Leads of each local TALKWORKS team (including a mixture of standard teams and long-term condition teams) were asked to identify high-intensity therapists to take part, on the basis of interest and capacity to complete the training and placement. The AccEPT supervisors were not involved in therapist selection. Eventually 11 therapists were recruited.

Therapists were trained to deliver ADepT following the protocol described in the ADepT pilot trial (Dunn et al., 2023), with the exception that due to service constraints clients were not offered the five optional booster sessions in the year after acute treatment finished. As this was an audit of a routine service improvement initiative, ethical approval for research was not applied for, and trainees were not required to give written informed consent. Therapists did all however give written permission for the audit to be published and quotes from their feedback questionnaires to be reported verbatim. As anonymised group-level data from routine outcome data collection were used to assess client clinical outcomes, individual clients were not asked for consent for publication. Therapists were given time in their routine workload to engage with the training, but there was no reimbursement for taking part in the training.

Overview of training and supervision pathway

The training and supervision pathway was designed to ensure by the end of the placement the therapists would have acquired the core style, structure and technique competences outlined in the ADepT therapist competency rating scale. Style is the interpersonal style therapists use with clients to engage in a positive, future-oriented and solution-focused dialogue (7 competencies); structure is managing session structure in a way that enables a balanced focus between positive and negative content and to ensure content is remembered (6 competencies); and technique is using ADepT-specific language and tools during sessions (8 competencies; cf. Dunn et al., 2023).

Initial training consisted of a 2-day introductory workshop. This outlined the training model and learning journey; presented the rationale for ADepT and outcomes to date; outlined the ADepT model in detail (including session by session structuring); and focused on the ADepT therapist style in detail (including changing relationship to depression and consolidating learning). Therapists then engaged in four 2–hour skills classes (one a week for a month after the initial workshop), including practice of solution-focused questioning; clarifying and consolidating values; setting and monitoring values consistent goals; formulating what helped and hindered clients reaching their goals using the mapping tool; engaging with simple everyday pleasures; utility-based cognitive change work in ADepT; keeping a positive journal; and setting a future wellbeing plan at the end of therapy.

The workshops and skills classes were informed by the declarative-procedural-reflective model of therapy acquisition (DPR; Bennett-Levy, 2006) and the COM-B model of behaviour change (Michie et al., 2011). Ideas and practices were introduced declaratively, before being modelled by the trainers. Trainees were then encouraged to engage in experiential practice (either role-play in session or self-practice between workshops to learn ‘ADepT from the inside out’) and reflect on this practice. There was an emphasis on ensuring trainees had the knowledge and skill, motivation and opportunity to implement the various elements of the ADepT model. Therapists were given a copy of the treatment protocol and all materials and handouts used in ADepT. They were also invited to keep a reflective journal throughout the training.

Following the formal training component of the training placement, therapists were invited to start treating two or more clients with ADepT, with weekly 90–minute supervision being provided in small group format (three supervision clusters of three or four trainees). Clients were selected by each individual therapist and invited to take part from the routine service high-intensity waiting list, who described depression as their primary presenting problem. No other inclusion or exclusion criteria were used beyond standard eligibility criteria for NHS–TTad settings (which had been assessed prior to clients being placed on the treatment waiting list). Therapists followed standard TALKWORKS service procedures to select clients to take on for treatment, apart from explaining to clients that they were learning a new experimental approach and checking that they were happy to engage with this. The AccEPT supervisors and trainers were not involved in client selection.

Supervision followed a pre-specified structure (solution-focused enquiry around a moment of personal resilience or thriving; a ‘traffic light’ review of all clients being treated, with greater time allocated to those showing risk or not engaging with therapy; before a detailed presentation of a single case). Therapists took it in turn to present a detailed case for discussion. When presenting they were asked to bring a specific supervision question to address and where possible to bring along a session clip to review. As required, supervisors introduced skills practice, role play and further conceptual learning into supervision during the detailed case presentation.

The supervisors aimed to model the ADepT style in workshops and supervision, with an emphasis on use of appreciative enquiry around trainee practice followed by some constructive suggestions for further refinement. There was an emphasis on building a culture of wellbeing for therapists into training and supervision and encouraging them to continue with ADepT self-practice for the duration of the placement.

As the pilot overlapped with the Covid pandemic, all training and supervision was delivered via video conferencing (and a vast majority of clients received ADepT via video conferencing sessions). The trainers and supervisors were clinical psychologists experienced in ADepT and who had delivered ADepT as part of the earlier pilot trial.

Evaluation of training and supervision pathway

The number of therapists engaging with each stage of the training and supervision pathway was audited. To assess therapist competency, trainees were invited to submit a video or audio recording of a session, which the supervisors then rated using the ADepT competency scale (cf. Dunn et al., 2023). This is a bespoke tool developed for the purposes of the ADepT pilot trial (Dunn et al., 2023), with preliminary evidence suggesting it has acceptable inter-rater reliability (see online Supplementary material; Dunn et al., 2023). The ADepT competency rating scale generates a competency rating for style, structure and technique, on a scale from 0 (‘incompetent’) to 6 (‘expert’). Overall competency is assessed by averaging these scores. The competency rating was completed by one of the two ADepT supervisors in the audit (half by one supervisor, half by the other supervisor), both of whom are experienced in delivering ADepT having been therapists in the earlier ADepT case series and pilot trial (Dunn et al., 2019a; Dunn et al., 2019b; Dunn et al., 2023). The sessions were single rated, so no measure of internal-rater reliability is reported.

Engagement outcomes for the clients each therapist treated with ADepT were captured, including the number of clients each treated, how many sessions each client attended, and how many clients completed a minimum adequate dose of therapy (≥8 sessions) and had a planned discharge.

Clinical outcomes for clients treated utilised the minimum dataset routinely collected in each session as part of NHS–TTad care. The Patient Health Questionnaire (PHQ-9; Kroenke et al., 2001) measures depression severity and consists of nine self-report items covering common symptoms of depression (e.g. ‘little interest or pleasure in doing things’). Participants rate frequency of experiencing symptoms in the past 2 weeks, on a scale ranging from 0 (not at all) to 3 (nearly every day). Item scores are summed, with scores ranging from 0 to 27, a cut-off of ≥10 indicating clinical levels of depression, and a change of ≥6 points used to indicate reliable change in NHS–TTad settings. The Generalized Anxiety Disorder Scale (GAD-7; Spitzer et al., 2006) measures anxiety severity and is made up of seven self-report items covering typical symptoms of anxiety (e.g. ‘feeling nervous, anxious or on edge’, each rated on the same scale as the PHQ-9 and then summed. Scores range from 0 to 21, with scores ≥8 indicating clinical levels of anxiety, and changes of ≥4 points indicating reliable change in NHS–TTad settings. The Work and Social Adjustment Scale (WSAS; Mundt et al., 2002) assesses functional impairment linked to poor mental health and consists of five self-report items covering functioning in work, home management, social leisure activities, private leisure activities, and close relationships, each measured on a scale from 0 (not at all impaired) to 8 (very severely impaired). Items are summed, with scores ranging from 0 to 40 and scores ≥10 being used to indicate moderate impairment and scores ≥20 to indicate severe impairment. Reliable change criteria have not been established for the WSAS. All three scales have good internal reliability (Cronbach’s α = .89, .92 and .94 for the PHQ-9 GAD-7 and WSAS, respectively) and adequate psychometric properties (Kroenke et al., 2001; Mundt et al., 2002; Spitzer et al., 2006).

First and last treatment session data were analysed using paired sample t-tests to examine if there were significant changes, with effect sizes interpreted according to Cohen’s rules of thumb for Cohen’s d (Cohen, 2013). All participants had complete first and last treatment session data, making this an intent-to-treat analysis. The number of clients meeting NHS Talking Therapy criteria at the end of treatment for reliable improvement, recovery, and reliable recovery were also indexed. Reliable improvement is reliable improvement in either the PHQ-9 or GAD-7; recovery is scoring below clinical caseness for both the GAD-7 and PHQ-9; and reliable recovery is meeting both criteria for reliable improvement and recovery. These were benchmarked against national IAPT services outcomes for clients with a depression problem descriptor receiving CBT in 2021 (54.5% reliable improvement, 44.1% recovery, 39.5% reliable recovery; NHS Digital, 2022).

To gain therapists’ views on the training placement and the ADepT model, at the end of the placement they were invited to complete a bespoke feedback questionnaire (for a copy, see online Supplementary material, Section 1). This asked therapists to rate their agreement with a series of statements regarding their experience of and views on the ADepT model (e.g. ‘I think ADepT is an effective treatment for depression’), on the training they had received (e.g. ‘the training was clinically useful’), and their views as to whether the training had built knowledge, skills and confidence in the key ADepT style, structure and technique competencies outlined in the ADepT competency scale (e.g. ‘using a positive interpersonal style’, ‘helping clients consolidate and remember session content’, ‘identifying and working with values’). Each statement was rated on a 5-point Likert scale, ranging from strongly disagree to strongly agree.

A series of open-ended qualitative questions were also asked about their views of the ADepT model (‘what elements of ADepT have worked well in your practice?’, ‘what elements of ADepT have you struggled with in your practice?’, ‘what impact has learning ADepT had on your own personal and professional wellbeing?’) and the training (e.g. ‘what did you like about the training?’, ‘what could be improved about the training?’). These were explored using thematic analysis, informed by the framework approach (Gale et al., 2013).

Results

Eleven therapists were recruited (nine female, two male), all with at least two years of experience as high-intensity therapists. Eight came from standard NHS–TTad teams and three came from long-term condition teams. Therapists had a range of prior professional backgrounds (four occupational therapists, three low-intensity PWPS, two mental health nurses, two other professional background) prior to high-intensity training. All except one were already CBT accredited therapists (one had over 10 years of experience of CBT and was working towards accreditation).

All 11 therapists completed the initial 2-day workshop, 10 completed the four skills workshops, and nine completed the supervised placement. One therapist withdrew at the skills workshop stage because of pressure of routine work. One therapist withdrew during the supervision phase (taking a leave of absence from all work due to personal circumstances).

The nine therapists who completed the supervised placement treated between them 24 clients [five therapists treating two clients each, three therapists treating three clients each, and one therapist treating five clients (two of whom were taken over from the therapist who took a leave of absence from work during the supervision phase)]. The sample had a mean age of 39.38 years (SD = 13.66), were predominantly female (17/24); and a majority identified as ‘white British’ (19/24; three clients identified as other white backgrounds; and the remaining two had not provided ethnicity data). A minimum adequate dose of therapy was completed by 21/24 (87%) of clients, with the mean number of sessions attended being 13.54 (SD = 3.84) out of 15 total sessions; 20/24 (83.33%) of clients had a planned discharge, with 4/24 dropping out (one of whom completed a minimum adequate dose and three of whom completed seven or fewer sessions).

Eight of the nine therapists who completed the supervised placement submitted a session recording for competency assessment (the remaining therapist said their two ADepT clients had not given permission for sessions to be recorded before both dropped out). All therapists who submitted recordings met overall competence criteria (mean score across style, structure and technique domains = 4.67, SD = 0.61; all scores ≥4; style mean = 4.75, SD = 0.65; structure mean = 4.56, SD = 0.68; technique mean = 4.69, SD = 0.59). Five therapists were rated overall as proficient (average scores across domains 4–5) and three were rated as excellent (average scores across domains ≥5).

At first session, clients presented with moderate to severe depression (PHQ-9 mean = 17.08, SD = 4.42; 24/24 scoring ≥10) and anxiety (GAD-7 mean = 12.88, SD = 4.78; 21/24 scoring ≥8); and moderately severe functional impairment (WSAS mean = 23.54, SD = 7.91; 18/24 scoring ≥10). At final session, PHQ-9 mean score was 11.25 (SD = 7.90; 11/24 scoring ≥10) and GAD-7 mean score was 8.08 (SD = 5.67; 10/24 scoring ≥8). There were significant, large magnitude improvements from first to last treatment session in PHQ-9 depression, mean reduction = 5.83, t = 3.96, p < .001, d = 0.808 (95% CI 0.34, 1.26) with 13/24 individuals showing reliable improvement (≥6 point reduction) and 1/24 individuals showing reliable deterioration (≥6 point worsening). There were significant, large magnitude improvements from first to last treatment session in GAD-7anxiety, mean reduction = 4.79, t = 4.39, p < .001, d = 0.90 (95% CI 0.41, 1.36), with 14/24 individuals showing reliable improvement (≥4 point improvement) and 1/24 individuals showing reliable deterioration (≥4 point worsening). There were significant medium to large magnitude improvements from first to last treatment session in WSAS functional impairment, mean reduction = 7.46, t = 3.70, p = .001, d = 0.76 (95% CI 0.29, 1.20).

Using IAPT outcome metrics, reliable improvement (on either of GAD-7 or PHQ-9) was shown by 17/24 clients (71%), recovery (scoring <10 on PHQ-9 and <8 on GAD-7) was shown by 12/24 clients (50%), and reliable recovery (meeting both reliable improvement and recovery criteria) was shown by 12/22 clients (50%). If repeating analyses on the eight therapists assessed as competent for the 20 clients they treated with a minimum adequate dose of therapy, rates of reliable improvement were 15/20 (75%) and rates of recovery and reliable recovery were 12/20 (60%).

All nine therapists who completed the supervised placement completed the feedback questionnaire. Table 1 summarises trainees’ views on the ADepT model and the training. Table 2 summarises trainees’ views on their competence at the end of the training.

Table 1. Trainees’ view on the ADepT model and the training/supervision pathway

* Only 8 trainees submitted a tape for feedback, so ratings are based on these 8 responses. No trainees disagreed or strongly disagreed with any statements.

Table 2. Trainees’ views on acquisition of ADepT competencies by the end of the training/supervision pathway

No trainees strongly disagreed with any statements.

Trainees tended to agree or strongly agree with statements, suggesting the ADepT model was effective; they had enjoyed practising it; that it could fit in the NHS Talking Therapies landscape; and that they would recommend its implementation in this setting (93% of responses). One of nine therapists neither agreed nor disagreed about enjoying ADepT, and two of nine neither agreed nor disagreed about recommending its implementation in NHS Talking Therapies.

Trainees also tended to agree or strongly agree with statements that the training and supervision pathway was interesting, useful, well presented, and fostered their own wellbeing (99% of responses). One therapist neither agreed nor disagreed overall that the training had been useful, well presented, and fostered their own wellbeing (the same therapist who also was uncertain regarding both enjoying ADepT and recommending the ADepT model overall). The element of the training that was rated most positively by trainees was feedback on a session recording using the competency rating scale (88% strongly agree responses). The initial training workshop and supervision group elements of the training were also rated positively (64% and 69% strongly agree responses). The skills workshop had slightly lower rates of positive endorsement (47% strongly agree responses).

Therapists tended to agree or strongly agree with statements that the training helped them gain core competencies (84% of all responses across all domains; 81% of all response in style domain; 75% of all responses in structure domain; and 91% of all responses in technique domain). Competencies where there were a relatively greater proportion of neutral or disagree ratings (three or more out of nine trainees) were: in the style domain, attending to the therapeutic alliance (3/9 trainees; 33%); in the structure domain, giving and receiving feedback (3/9 trainees; 33%) and pacing sessions (5/9 trainees; 56%); and in the technique domain, using cognitive change techniques (3/9 trainees; 33%), suggesting these aspects of the training and supervision may require refinement.

Thematic analysis (informed by the framework approach) of the qualitative data explored trainees’ views of the ADepT model and its fit in an NHS–TTad context. Overall, the model was seen as effective for working with depression; trainees commented particularly on the benefits of a positive focus and working on values; trainees felt that the model could be improved by being more explicit which intervention elements can be used to target each depressive mechanism; ADepT was seen as being helpful working with co–morbidities (although views on its capacity to support clients with complex trauma histories were mixed); the values work was felt to need to shift emphasis to exploration rather than clarification when working with young adults; and trainees generally felt ADepT would be a good fit in an NHS–TTad context (although there were mixed views if this was as a stand-alone model or via integration of ADepT ideas into standard CBT practice).

Thematic analyses also explored trainees’ views on the training pathway. Therapists reported having had enjoyed the training and found it useful; that it had been helpful for training to have a clear structure and rationale and to integrate experiential practice; that the learning journey had felt like ‘learning a new language’; that the training had enhanced their personal wellbeing; that a key determinant of engaging well with the training was having protected time in their workload; and that to continue practising ADepT with fidelity they would need ongoing supervision (see Supplementary material Section 1 for more detailed thematic analysis).

Discussion

The current study examined the feasibility, acceptability and effectiveness of implementing ADepT in a routine NHS–TTad service via a structured training and supervision pathway. The pathway was shown to be feasible and acceptable to high-intensity therapists who took part. Eleven therapists were selected to take part by their services (working in a mixture of standard and long-term condition teams), nine of whom completed the entire pathway. Two therapists did not complete the full training and supervision package due to broader pressures of work and personal reasons. Due to the selection method used, it was not possible to assess broader interest in the offering across all of the high-intensity therapists working in the service, but it is nevertheless encouraging that all available places were filled. The therapists who did complete the pathway treated 24 clients between them. NHS–TTad therapists could be trained to deliver ADepT competently, with all eight therapists who submitted a digitally recorded session for assessment meeting competence criteria. The remaining therapist could not submit a session recording as both clients they treated dropped out of therapy before giving permission for a session to be recorded and shared.

Therapists achieved promising engagement and clinical outcomes using ADepT; 21/24 of clients treated received a minimum adequate dose of therapy (≥8 sessions), the majority of whom completed the entire course of 15 acute sessions; 20/24 clients had a planned discharge at the end of therapy, with four dropping out (which compares with 35% drop-out in the depression trial literature and 31% in NHS–TTad settings; Furlong-Silva, 2020; Sharf and Primavera, 2009). At the group level, there were significant and large effect size improvements observed in PHQ-9 depression (d = 0.81) and in anxiety (d = .90); and medium to large effect sizes in functioning (d = 0.76). The depression effect size in the ADepT sample was equivalent to those observed in national NHS–TTad reports for clients with a depression problem descriptor receiving standard CBT (d = 0.8) and the anxiety and functioning effect sizes in the ADepT sample were numerically higher than those receiving CBT for depression in routine care (d = 0.7 and d = 0.5, respectively; NHS Digital, 2022). Rates of reliable improvement were 71 (compared with national average of 55% reliable improvement in CBT with a depression problem descriptor) and rates of reliable recovery were 50% (compared with 40% reliable recovery observed for clients with a depression problem descriptor receiving CBT in the national IAPT dataset).

While caution should be taken over-interpreting these figures due to a small sample size in the current study, the present results nevertheless suggest therapists can be trained to deliver ADepT in a 6-month placement and achieve broadly comparable outcomes to existing established treatments. The magnitude of improvement was however slightly smaller than that observed in the ADepT arm of previous pilot trial (80% reliable improvement, 55% reliable recovery; d of 1.61 for depression; d of 1.64 for anxiety; d of 0.95 for functioning; cf. Dunn et al., 2023), which recruited primarily form the same waiting lists that therapists had accessed for this training evaluation. One difference was the inclusion of long-term condition clients in the current audit, and co–morbid physical health conditions have been shown to result in slightly poorer IAPT outcomes (Seaton et al., 2022). Only one client (4%) showed reliable deterioration on both depression anxiety outcomes (<6% national average).

A majority of therapists provided positive feedback on the training and supervision pathway. They reported finding it interesting, clinically useful, well-presented, and as contributing to their own personal wellbeing. Given workforce pressures in NHS–TTad and the broader NHS (Owen et al., 2021; Westwood et al., 2017; Wheatley, 2023), the latter comments on staff wellbeing are encouraging. Qualitative comments suggested therapists had valued a clear structure to the training and a strong experiential focus (observing, practising and reflecting on use of ADepT ideas and tools). Therapists also reported feeling competent in a majority of ADepT style, structure and technique competencies at the end of training, which also aligned with supervisors’ session recording ratings, suggesting they had all acquired the capacity to deliver ADepT to a good standard. There were, however, suggestions from therapists that they would have benefited from more training in how to attend to the therapeutic alliance, how to give and receive feedback, how to pace sessions, and how to use cognitive change techniques within the ADepT framework. Therapists also commented that they would benefit from some additional workshops and skills sessions midway through the supervised placement, to consolidate learning of ADepT after they had started to practise it. A key contributor to receiving the training positively was therapists being given time and space by their line managers to engage with it. There was a recognition that if this was not available, the impact of the training would have been diminished.

Therapists generally also provided positive feedback on ADepT as a treatment. It was rated as an effective treatment of depression, that would fit well in an NHS–TTad context, and could manage a range of common co–morbidities often co-presenting with depression like long-term conditions and anxiety. Some adaptations were felt to be necessary as a function of client developmental stage, for example a focus on exploring rather than clarifying values with young adults. One therapist was less positive about the ADepT model (both of whose clients dropped out before completing treatment), raising questions about how well ADepT would meet the needs of clients with complex trauma alongside depression and anxiety (although other therapists felt ADepT might work as a stabilisation treatment for this group). Trainees felt that ADepT could either be delivered as a stand-alone approach or that aspects of ADepT could be incorporated into standard CBT practice and that either of these would likely be effective at improving depression outcomes.

A potential obstacle to implementing ADepT in NHS–TTad settings, however, is the use of flexible booster session in the year after completing therapy. The host service said this was not possible to integrate in this pilot (so therapists were only trained in the acute sessions of ADepT). A number of therapists commented how it would be very difficult to offer flexibility around boosters given a very busy therapeutic schedule and high case load. However, updates to the IAPT manual (National Collaborating Centre for Mental Health, 2023) now recommend offering brief booster sessions to detect and reduce relapse before symptoms become more problematic. As a result, the TALKWORKS service that implemented this pilot now adopts a more flexible stance around booster sessions, where up to four booster sessions are considered for clients requesting to re–engage within 6 months of responding well to high-intensity treatment with a planned discharge. There is clear evidence that offering maintenance/booster sessions reduces subsequent risk of relapse/recurrence (Bockting et al., 2015), suggesting this is an important element of ADepT to retain.

To scale up the capacity to implement ADepT, it will also be necessary to develop a cadre of supervisors and trainers. Future work will need to construct and evaluate a supervisor and trainer development pipeline.

A number of limitations need to be held in mind with this audit. Given the anhedonia focus of ADepT, it would have been useful to also have collected an anhedonia outcome. However, the current audit relied on the IAPT minimum data set (and we only had access to total scores on the PHQ-9 rather than individual item data), so it was not possible to analyse anhedonia outcomes. The audit is at risk of selection bias, both in terms of the therapists who opted into the training, and also the patients who were willing to be treated with the novel protocol. No additional strategies were deployed to mitigate this risk, other than ensuring that the AccEPT clinic supervisors and trainers were not involved in therapist or client selection. These therapists and patients may not be representative of the broader workforce and patient group treated by NHS–TT services. The therapists selected to take part were all experienced (having worked in high-intensity settings for at least 5 years), and it is unknown how less experienced therapists would acquire the model. The session recordings were rated for competence by the two supervisors/trainers, rather than by independent and blinded raters. The ADepT competency scale is still at an early stage of development (cf. Dunn et al., 2023) and has yet to be extensively psychometrically validated. Ratings in the current study were predominantly intended to provide developmental feedback to each therapist, so were only coded by a single trainer/supervisor, meaning inter-rater reliability of the present ratings was not established. Further work is needed to validate the competency rating scale, for example examining if it predicts therapist outcomes in a larger cohort. Therapists were aware that their feedback questionnaires were being analysed by the trainers/supervisors (after the placement had finished), which may have introduced social desirability biases. The training was done in a single NHS–TTad service in Devon, so caution should be taken generalising findings to other settings. The senior author of the study (BD) is the developer of ADepT and one of the supervisors/trainers, so there is a risk of allegiance bias.

Overall, the current findings further support ADepT as a promising treatment approach for depression in NHS–TTad (and potentially broader) settings). They demonstrate that it is possible to train therapists in routine settings to deliver ADepT competently and that this leads to good clinical outcomes. This supports continuation to a definitive pragmatic trial evaluating if ADepT is clinically superior (and cost effective) relative to CBT at managing anhedonic depression and further support findings from the pilot trial that ADepT is an effective treatment for this group.