Clinical Outcomes Of Cognitive Analytic Therapy Delivered By Trainees

Darongkamas, J., Newell, A. Hewitt-Moran, T. & Jordan, S, 2017. Clinical Outcomes Of Cognitive Analytic Therapy Delivered By Trainees. Reformulation, Winter, pp.38-42.

Can Cognitive Analytic Therapy provide significant results for people with mental ill health when it is delivered by trainees in the approach?

There are many evidence-based, effective psychological therapies (Roth & Fonagy, 2005). The Improving Access to Psychological Therapies (IAPT) scheme continues (IAPT, 2015); particularly to provide a choice of therapies in a timely manner to a wider range of people and reach target recovery rates (We Need to Talk coalition, 2014; New Savoy Partnership, 2011).

Cognitive behavioural therapy (CBT) was one of the forerunners in IAPT and random controlled trials (RCT) – a method considered valid by bodies such as the National Institute for Health and Care Excellence (NICE) – had been conducted into it.

As IAPT developed, the need to train workers in approaches other than CBT was noted. “There will also be a proportionate increase in every region in training for therapy workers who deliver other NICE-approved modalities for treating depression and anxiety disorders.” (DH, 2011)

The target for ‘recovery rates’ (i.e. move from ‘caseness’ to ‘non-caseness’) for IAPT services have been set at 50%. Key performance indicators for IAPT (August 2013) state that “...of those completing treatment it is expected that at least 50% will recover (defined by recovery to no longer being a “case” on both the questionnaires, GAD-7, PHQ9). However, the more practice-based concept of reliable change is being considered (IAPT, 2014), “the IAPT year one audit (Gyani et al) reported reliable recovery rate at 40% and the reliable improvement at 64%” (IAPT, 2014). But this data will only be for IAPT services as opposed to secondary care mental health services for those at higher levels of the Stepped Care approach (NICE, 2011).

Introducing IAPT within this context led to training becoming more available but mostly in CBT. However, given that “not one size fits all” with regard to therapy (e.g. Faddis, 2009), some patients suffer the consequences of this lack of choice. For example, some patients do not respond to CBT treatment (Roth & Fonagy, 2005).

Even where CBT has shown to be effective and is recommended (e.g. CBT-E for eating disorders) up to 50% of patients do not respond positively (Fairburn et al, 2003). This shows that the question of “what works for whom” (Roth & Fonagy, 1996), when, and where is still highly relevant.

Cognitive analytic therapy (CAT) is often recommended for patients, especially where personality difficulties contribute to the symptoms (Chanen et al, 2008; Clarke et al, 2013). CAT is also in guidelines such as the NICE Guidelines for Eating Disorders (NICE, 2004). If such therapy is to be available for clients, it is necessary to have sufficient CAT trained therapists: currently there is a dearth.

In the West Midlands, prior to 2009, there were a tiny number of trained CAT therapists, partly because there were no CAT training courses in the region. Therefore, a bid was submitted to the then Workforce Deanery of the West Midlands Strategic Health Authority to establish a West Midlands CAT Training Course.

To initiate CAT Training, the Workforce Deanery funded a CAT Practitioner Training Course, which was hosted by South Staffordshire and Shropshire Healthcare NHS Trust and ran from 2009 to 2011.

Trainees on the Midlands CAT course were asked, as part of good practice, to collect “routine patient-reported outcome data” (Turpin & Fonagy, 2008; DH 2008; New Savoy Partnership, 2011). Trainees were asked to use the Core Outcome Measure (Core OM, previously known as CORE34) and the Inventory of Interpersonal Personality problems (IIP 32) questionnaires to assess the severity of symptoms and interpersonal difficulties of the patients at the start and end of therapy. At course completion, the outcome measure data was analysed to see if the therapy had benefited the patients.

The CAT model

CAT is an integrative model of psychotherapy, described in Ryle & Kerr, 2002. In summary, it is proposed that, in combination with inherited predispositions, early interpersonal experience with carers plays a central part in the development of the sense of self. The experiences of the various aspects of each player, in relationship to the child, are internalised. These patterns are called reciprocal roles – each incorporating beliefs, feelings, emotions, part identity, memories and so on. These roles are interpersonal and occur in relation to the other, or can also be how a person can relate to themselves. Attempts are made to attain desired roles, which may or may not be reached, and feared roles are avoided (or attempts for self-affirmation made) by ‘intentional patterns of behaviour’ known as ‘procedures,’ (or patterns). The latter is largely derived from cognitive theory (Kelly, 1955; Beck, 1995). CAT is also identified as one of several specific interventions for those with a diagnosis of personality disorder (Roth & Pilling, 2013).

Procedures that are no longer appropriate to the current situation but are not amenable to change are postulated to have a role major role in the maintenance of psychological problems (target problem procedures – TPP). Helping patients to change TTPs is an aim of therapy. Treatment is time-limited, lasting typically eight, 16 or 24 sessions, although other lengths of therapy can be recommended where appropriate.

Clinical psychologists’ training equips them to offer a breadth and range of psychological interventions. They often go on to CAT for further specialist post-qualification training. Other professionals are also selected for a place on a CAT training course.

This aspect of the course evaluation reported here aimed to assess the effectiveness of CAT delivered by trainees in their usual work setting, as demonstrated on outcome measures taken before the start and at the end of therapy.




The patients – 35% male and 65% female – were aged between 16 and 69 years of age, with a mean age of 41. They were being treated for a range of mental health difficulties in NHS mental health trusts across the West Midlands, primarily in secondary care community mental health teams. There is evidence that the patient sample is typical of those seen in mental health services; by comparing the data with a mental health patient population as assessed via Core Outcome Measure (Core OM) (Core System Group, 1998).

A Shapiro-Wilk (1965) test was used to test for normality, which suggests that the data is from a normally distributed sample.


There were 15 course members who were all employed in NHS trusts across the West Midlands. They were all qualified mental health professionals with training and experience in conducting psychological therapy. The range of professions included two community mental health nurses, nine clinical psychologists, and a counselling psychologist, social worker, psychiatrist and an art therapist.

CAT practitioner training

Prior to starting the course, trainees attended a two-day introductory workshop. During the course, trainees had 20 training days over two years, in addition to monthly reading seminars, personal therapy and weekly clinical supervision.

For this report, certain assumptions have been made that appear reasonable in the light of trainees’ prior level of clinical experience and the training model. For example, given that weekly clinical supervision was provided for each trainee concerning their work with every patient treated during the training, it can be expected there was a good standard of compliance to the model. While trainees may not have been fully proficient at the start of the course, it is reasonable to assume they would be delivering therapy in a manner that conformed sufficiently to the CAT model.

Supervision was conducted in groups of three. Furthermore, as all supervision was conducted by an accredited CAT supervisor, this should ensure that the supervision focused on delivering CAT according to the principals of the model.

Instruments used



The CORE OM was designed to measure psychological difficulties in a range of patient groups across a variety of care systems (Core System Group, 1998). It consists of a two-page self-administered questionnaire assessing four dimensions: wellbeing, problems/symptoms, life functioning and risk. The scores of these four subscales can be added together to produce a reliable measure of the severity of any mental ill health. The test has been reported to be reliable and valid. Data is provided on two main normative groups: a non-clinical population and a clinical population. The clinical data originates from a variety of sites, generally across the NHS and is based on 863 patients awaiting treatment for psychological problems in the NHS and thus is an appropriate measure to assessing the outcome of psychological therapy. Moreover, cut-off data is provided to determine if any individual meets the ‘cut off’ of caseness, or not, in a mental health service. The test has good psychometric properties including internal consistency and test retest reliability (Evans et al, 2002).

Inventory of Interpersonal Personality problems (IIP 32)

The IIP was developed by Horowitz et al (1988) to measure distress arising from interpersonal sources. It was also designed to assess psychological change associated with improvement as a consequence of psychological treatment. The measure has eight subscales: hard to be sociable, hard to be assertive, too aggressive, too open, too caring, hard to be supportive, hard to be involved and too dependent. All of these show high internal consistency and test-retest reliability (Barkham et al, 1994). A shorter version was developed by Barkham et al (1996) consisting of 32 questions which the patient rates using a five-point rating scale from zero to five. This version of IIP was used here.

Therapists were asked to give the CORE OM and the IIP-32 at the initial and final therapy sessions. Scores of anonymised patients’ outcome data were collated and analysed.


Forty-seven patients completed pre- and post-CORE 34 questionnaires and 39 completed pre- and post-IIP 32 questionnaires, from the potential 116 who completed treatment as training cases. The overall mean CORE 34 score (the mean of the four subscales) has been used in the analysis. The IIP 32 scores were calculated from the total individual scores. Patients showed significant improvement after CAT, with 74% of patients improved to a less severe banding, while 26% did not change bands. None deteriorated. Table 1 shows the means and standard deviations pre- and post-therapy. Table 2 shows the latter with clinical cut-offs for the two gender subgroups.


Core OM

A paired t-test was performed to determine the effectiveness of CAT delivered by course members. The mean improvement (M = 0.74, SD = 0.54, n = 47) was significantly greater than 0, t(46) = 9.36, one-tail p <0.01, providing evidence that CAT was effective in reducing patients global distress symptoms. A 95% confidence interval about mean outcome was calculated (0.58, 0.9).












IIP 32

A paired t-test was performed to determine the effectiveness of CAT delivered by course members. The mean improvement (M = 0.59, SD = 0.63, n = 39) was significantly greater than 0, t(38) = 5.88, one-tail p <0.01, providing evidence that CAT is effective in reducing patients interpersonal problem symptoms. A 95% confidence interval about mean outcome is (0.39, 0.79).


Shows statistically reliable change obtained on the CORE 34 scores pre-post therapy. Scores outside the tramlines represent change that could not be achieved by chance.


This provides practice-based evidence on some outcome measures that CAT psychotherapy delivered by trainees leads to improvement in psychological symptoms and distress. Firstly, there was a statistically significant change in the means of the CORE OM and IIP 32 following the CAT therapy: p<0.01 for both outcome measures. Both the CORE OM and IIP 32 outcome measures showed an overall mean improvement post-therapy.

The norms of the CORE OM include cut off scores for ‘caseness’. This means a person with scores in the caseness range has significant mental health problems. As Table 2 shows, prior to treatment the male and female means were above the cut off for ‘caseness’, whereas afterwards both were below the cut offs. In addition, the graph shows all mean CORE OM scores improved with the exception of one. Although this patient’s scores increased a little, it was at a level that could have occurred by chance. For those who showed improvement, many did so to a degree above that expected by chance – as shown by being outside the tram lines on figure 2.

Strengths and Limitations

Results from this evaluation should provide data to support evidence-based practice in psychological therapies through such a systematic clinical audit and demonstrating the utility of CAT on the outcome measures.

A further strength is that the patients were taken from waiting lists of people receiving care in a mental health setting and were not selected in anyway. That the therapy was delivered in a public health setting to a typical outpatient population suggests that the study has good external validity.

Due to using different measures, we have not been able to compare directly our patients’ reported ‘recovery rates’ with those published nationally by IAPT.

There were no specific exclusions, apart from patients with psychosis or organic disorders. The course supervisors decided for whom this type of therapy was appropriate.

A number of recommendations are typically utilised in RCTs to ensure therapists comply with standard procedures/techniques. One method is manualising treatment and systematic supervision (Roth & Fonagy, 2005). While CAT is difficult to manualise, the course used a standard text (Ryle & Kerr, 2002) around which training was based. Trainees had regular supervision linked to training, in which each trainee had a minimum of 15 minutes supervision per patient per week from experienced CAT providers. Therefore, it is likely that trainees delivered therapy that was consistent with current CAT models. However, the therapists were trainees in using CAT and therefore might not be expected to deliver it as efficiently as fully-trained therapists. There may be some fluctuation of fluency in delivering the CAT model, as it could be hypothesised that more experienced trainees will deliver it more proficiently. But the overall results exceeded the course leaders’ expectations.

The duration of therapy was consistent with the usual CAT format. The majority of patients had 16 sessions, with a few having eight or 24 sessions with no other duration.

The mean of the CORE OM sample pre and post therapy was 1.9 and 1.16 respectively, compared with a clinical cut-off from the clinical sample from the Core Systems Group (1998) of 1.19. The mean of the IIP 32 sample pre and post-therapy was 1.77 and 1.18 respectively, compared with a clinical cut-off of 1.59.

The study was not a RCT and therefore had no control group hence it is unclear how patients would have responded to treatment as usual using another psychological therapy, such as CBT.

The study’s sample size was relatively small. The data used was not from a complete sample due to the low return of outcome data. However, the trainees were effectively volunteers and there was no set requirement to complete the data collection as part of the training. Anecdotal data from some of the trainees suggests that there was no systematic bias that determined the returns rate of the patients’ questionnaires.

If the evaluation were replicated, it would be desirable to ensure that all trainees returned all patients’ outcome measures. The CAT competency measure (Bennett & Parry, 2004) could also be used to demonstrate the fidelity to the CAT model as used by psychological therapists training in it.


The study supports the existing evidence base that CAT produces meaningful and significant clinical changes in patients undergoing routine treatment in local mental health services across one NHS region.

Patients showed statistically significant post-treatment improvement on the CORE OM and IIP 32 outcome measures. There is evidence that this change was clinically meaningful (Table 2). The post-treatment means for both outcome measures was below the scores for clinical caseness (Tables 1 and 2). Further data confirming this relates to changes in the bands of symptom severity is shown in Figure 1.


Barkham M, Hardy G & Startup M (1994) The structure, validity and clinical relevance of the IIP. British Journal of Medical Psychology 67 171–85.

Barkham M, Hardy GE & Startup, M (1996) The IIP-32: A short version of the Inventory of Interpersonal Problems. British Journal of Clinical Psychology 35 21-35.

Barkham M, Mellor-Clark J, Connell J & Cahill J (2006) A core approach to practice-based evidence: A brief history of the origins and applications of the CORE-OM and CORE system. Counselling and Psychotherapy Research 6(1) 3-15.

Beck J (1995) Cognitive Therapy: Basics and Beyond. New York: The Guilford Press.

Bennett D & Parry G (2004) A measure of psychotherapeutic competence derived

from Cognitive Analytic Therapy. Psychotherapy Research 14 176-192.

British Psychological Society (2013) Using Practice Based Evidence to Improve Outcomes in Psychotherapy. Available at: (accessed August 2013).

BPS (2015) Clinical Psychology. Available at: (accessed July 2015)

Chanen AM, Jackson HJ, McCutcheon LK, Jovev M, Dudgeon P, Yuen HP, Germano D, Nistico H, McDougall E, Weinstein C, Clarkson V & McGorry PD (2008) Early intervention for adolescents with borderline personality disorder using cognitive analytic therapy: randomised controlled trial. British Journal of Psychiatry 193 477-484.

Clarke S, Thomas P & James K (2013) Cognitive analytic therapy for personality disorder: randomised controlled trial. British Journal of Psychiatry 202 129-134.

Core System Group (1998) CORE System (Information Management) Handbook. Leeds: Core System Group.

Department of Health (2008) Guidance on the routine collection of Patient Reported Outcome Measures (PROMs). Available at:

Department of Health (2008a) IAPT implementation plan: national guidelines for regional delivery. Available at:

Department of Health (2011) Talking therapies: A four-year plan of action. London: DH.

Evans C, Connell J, Barkham M, Margison F, Mc Grath, G, Mellor-Clark J & Audin K (2002) Towards a standardised brief outcome measure: Psychometric properties and utility of the CORE-OM. British Journal of Psychiatry 180 51 -60.

Faddis J (2009). Therapy is Not One Size Fits All. Available at: (accessed August 2013)

Fairburn CG, Cooper Z & Shafran R (2003) Cognitive behaviour therapy for eating disorders: a “transdiagnostic” theory and treatment. Behaviour Research and Therapy 41 509-528.

Horowitz LM, Rosenberg SE, Baer BA, Ureño G & Villaseñor VS (1988) Inventory of Interpersonal Problems: Psychometric properties and clinical applications. Journal of Consulting and Clinical Psychology 56(6) 885-892.

IAPT (2015) (accessed July 2015).

IAPT (2013) Key performance indicators. Available at: (accessed August 2013).

IAPT (2014) Measuring improvement and Recovery adult services version 2. Available at:

Kelly GA (1955) The Psychology of Personal Constructs. New York: Norton.

We Need to Talk Coalition (2010) We need to talk: Getting the right therapy at the right time. London: Mind.

National Institute for Clinical Excellence (2004) Eating disorders: Core interventions in the treatment and management of anorexia nervosa, bulimia nervosa and related eating disorders. London: NICE.

New Savoy Partnership, Consensus Statement on Evidence, 2011. Available at:

NICE, (2011) a-stepped-care-approach Available at: (accessed July 2015).

Roth A & Fonagy P (1996) What works for whom? A Critical Review of Psychotherapy Research. New York: Guilford Press.

Roth A & Fonagy P (2005) What Works for Whom: A Critical Review of Psychotherapy Research (2ND ed.) New York: Guildford Press.

Roth AD & Pilling S (2013) A competence framework for psychological interventions with people with personality disorder. UCL: London.

Ryle A & Kerr IB (2002) Introducing Cognitive Analytic Therapy. Chichester: Wiley.

Shapiro S & Wilk MB (1965) An analysis of variance test for normality (complete samples). Biometrika, 52, 591-611. Cited by Jarque CM & Bera AK (1987) In: A Test for Normality of Observations and Regression Residuals, International Statistical Review 55 (2) 163-172.

Turpin G & Fonagy P eds (2008) New Ways of Working for Psychological Therapists

Workstream one. The Evidence base for Psychological Therapies: Implications for Policy and Practice. Available at:

We Need to Talk coalition (2014) Commissioning talking therapies to meet need. Available at:


Financial support from the NHS (previously the West Midlands Deanery) for funding the first cohort of the West Midlands CAT practitioner course; Dr Tina Griffiths, course director; the Association for CAT; Dr Dawn Bennett and David Dobel-Ober for statistical analysis; Dr Kim Dent-Brown for his CORE 34 spreadsheet and Charlotte Blower with an earlier stage of data management.


Dr Jurai Darongkamas is a consultant clinical psychologist, registered practitioner psychologist (HCPC); director, West Midlands Cognitive Analytic Therapy Practitioner course; chair West. Midlands, Division of Clinical Psychology, British Psychological Society; lead for personality disorders in her NHS Trust and psychologist specialising in psychotherapy, senior practitioner status. Dr Adrian Newell is a CAT psychotherapist and trainer; consultant psychologist. Teresa Hewitt-Moran is the West Midlands Mental Health Institute LETC lead, NHS. with Sam Jordan, honorary assistant Psychologist, sandwich year, B.Sc. Psychology, Aston University.

Petition to NHS England - The Case for Funding Training in the NHS 2021 Alert!
ACAT's online payment system has been updated - click for more information

Full Reference

Darongkamas, J., Newell, A. Hewitt-Moran, T. & Jordan, S, 2017. Clinical Outcomes Of Cognitive Analytic Therapy Delivered By Trainees. Reformulation, Winter, pp.38-42.

Search the Bibliography

Type in your search terms. If you want to search for results that match ALL of your keywords you can list them with commas between them; e.g., "borderline,adolescent", which will bring back results that have BOTH keywords mentioned in the title or author data.

Related Articles

Measuring Outcomes in CAT
Freeman, E., Williams, L., 2008. Measuring Outcomes in CAT. Reformulation, Winter, p.14.

Generating Practice-Based Evidence for CAT
Marriott, M. and Kellett, S., 2007. Generating Practice-Based Evidence for CAT. Reformulation, Summer, pp.40-42.

A Randomised Controlled Trial of the Efficacy of CAT for the Treatment of Personality Disorder
Clarke, S., 2001. A Randomised Controlled Trial of the Efficacy of CAT for the Treatment of Personality Disorder. Reformulation, ACAT News Autumn, p.x.

Audit of Factors Predicting Drop Out from Cognitive Analytic Therapy Kerrie Channer and Alison
Channer, K., Jenaway, A., 2015. Audit of Factors Predicting Drop Out from Cognitive Analytic Therapy Kerrie Channer and Alison. Reformulation, Winter, pp.33-35.

The West Midlands CAT Practitioners Course 2009-2011
Darongkamas, J., Griffiths, Dr T., Williams, O., 2010. The West Midlands CAT Practitioners Course 2009-2011. Reformulation, Winter, pp.41-44.

Other Articles in the Same Issue

A case study of cognitive analytic therapy with a client diagnosed with psychotic depression
John Cartmell, 2017. A case study of cognitive analytic therapy with a client diagnosed with psychotic depression. Reformulation, Winter, pp.22-27.

An Alternative Psychotherapy File
Alison Jenaway, 2017. An Alternative Psychotherapy File. Reformulation, Winter, p.48.

Austerity, Power and CAT’s Opportunity for Change
John Cartmell, 2017. Austerity, Power and CAT’s Opportunity for Change. Reformulation, Winter, pp.5-7.

Clinical Outcomes Of Cognitive Analytic Therapy Delivered By Trainees
Darongkamas, J., Newell, A. Hewitt-Moran, T. & Jordan, S, 2017. Clinical Outcomes Of Cognitive Analytic Therapy Delivered By Trainees. Reformulation, Winter, pp.38-42.

Concepts of the Self, Social Inequality, Culture and Power in reflecting on therapeutic work with Asylum Seekers and Refugees: A Cognitive Analytic Approach
Dr Claire Wilson, 2017. Concepts of the Self, Social Inequality, Culture and Power in reflecting on therapeutic work with Asylum Seekers and Refugees: A Cognitive Analytic Approach. Reformulation, Winter, pp.44-47.

How CAT-Informed Practice Can Help To Improve Mental Health Services
Ben Bell, 2017. How CAT-Informed Practice Can Help To Improve Mental Health Services. Reformulation, Winter, pp.33-37.

On Learning From Our Patients: Reformulating CAT Training
Robert Marx, Joanne King, Kylie, Mary Wilson-Verrall, Karen Spencer and Annabel J McDonald, 2017. On Learning From Our Patients: Reformulating CAT Training. Reformulation, Winter, pp.13-18.

Perfectly Imperfect: Reflections of a Newly Qualified Clinical Psychologist in using CAT to Explore the Relational Contribution to Binge Eating
Dr Lyndsey Holt, 2017. Perfectly Imperfect: Reflections of a Newly Qualified Clinical Psychologist in using CAT to Explore the Relational Contribution to Binge Eating. Reformulation, Winter, pp.28-32.

The Clinician’s Guide To Forensic Music Therapy - Book Review
Wakeling, N, 2017. The Clinician’s Guide To Forensic Music Therapy - Book Review. Reformulation, Winter, pp.56-57.

Using The Parent-Adult-Child Model Alongside CAT
Louise Elwell, 2017. Using The Parent-Adult-Child Model Alongside CAT. Reformulation, Winter, p.43.

When Therapy Feels Like A Roller Coaster Ride: A CAT Perspective
Dr Diane Agoro, 2017. When Therapy Feels Like A Roller Coaster Ride: A CAT Perspective. Reformulation, Winter, pp.19-21.

Where Should We Keep Reformulation Letters?
Kate Freshwater, Tilait Tanweer & Wendy Walker, 2017. Where Should We Keep Reformulation Letters?. Reformulation, Winter, pp.8-12.


This site has recently been updated to be Mobile Friendly. We are working through the pages to check everything is working properly. If you spot a problem please email and we'll look into it. Thank you.