Tchanturia et al.

BMC Psychiatry (2015) 15:53

DOI 10.1186/s12888-015-0434-9

RESEARCH ARTICLE

Open Access

Cognitive Remediation and Emotion Skills

Training (CREST) for anorexia nervosa in
individual format: self-reported outcomes
Kate Tchanturia1,2,3*, Eli Doris1,2, Vicki Mountford2 and Caroline Fleming2

Abstract
Background: To evaluate self-reported outcomes after a brief course of skills-based individual therapy for inpatients
with anorexia nervosa (AN).
Methods: In this case series study 37 adults with AN participated in cognitive remediation and emotion skills
training (CREST) sessions, and completed social anhedonia, alexithymia and motivational measures before and after
the intervention.
Results: The CREST primary outcome measures were total scores on the Revised Social Anhedonia Scale (RSAS),
which decreased significantly (p = 0.03) with an effect size of 0.31, and the Toronto Alexithymia Scale (TAS), which also
decreased significantly (p = 0.05) with an effect size of 0.35. The secondary outcome measures focused on motivation:
perceived importance to change and ability to change; the second of which increased significantly (p < 0.001) with a
medium effect size (d = 0.71).
Conclusions: The individual format of CREST led to a decrease in patients self-reported social anhedonia, an improvement
in the ability to label their emotions, and increased confidence in their ability to change. Considering the limited number of
individual sessions, this is a promising preliminary finding which warrants further research.
Keywords: Anorexia, Flexibility, Social interaction, Social cognition, Emotion skills, Anhedonia, Alexithymia

Background
Social and emotional difficulties in the anorexia nervosa
(AN) adult population are well recognised [1,2]. Additionally, the poor work and social adjustment which accompany chronic AN make recovery difficult [3]. An
increasing amount of research is being carried out in
order to explore the ways in which people with AN can
be helped to return to a normal level of functioning and
adjustment. A new wave of psychological interventions
in eating disorders which target improving patients social life should therefore benefit their overall quality of
life, and might also facilitate the process of recovery [4].
Cognitive Remediation and Emotion Skills Training
(CREST) was developed with the purpose of addressing
* Correspondence: [email protected]
1
Kings College London, Division of Psychological Medicine, Institute of
Psychiatry, SE5 8AF London, UK
2
Eating Disorders Unit, South London and Maudsley NHS Foundation Trust,
London, UK
Full list of author information is available at the end of the article

emotion processing, to be delivered over 10 sessions in

an individual format [4-6]. In a study by Davies and colleagues [5], patients from different inpatient programmes receiving either treatment as usual or CREST
were compared in terms of their performance on
emotion-processing behavioural (experimental) tasks. It
was found that the patients who had received CREST
demonstrated a larger magnitude of change in neuropsychological task performance, although the results
were not statistically significant. In previous research [6],
qualitative analysis showed that CREST was perceived
positively by patients, who reported that education regarding the function, management and expression of
emotions was useful [6]. A small study on the group format of the CREST intervention [4] found positive small
changes in social anhedonia and motivation; furthermore, when patients were asked what they found helpful
about the group, they commented on learning new strategies to deal with emotions and being with other people.

2015 Tchanturia et al.; licensee BioMed Central. This is an Open Access article distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and
reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain
Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article,
unless otherwise stated.

Tchanturia et al. BMC Psychiatry (2015) 15:53

Preliminary research therefore suggests that CREST might

be a valuable initial intervention enabling patients with AN
to process their emotions, which may in turn help them to
engage in other therapies [6]. Further investigation is required to explore the benefits of CREST in this patient
group and to identify the number of sessions sufficient to
yield improvements in emotion processing.
Our aim throughout the development of CREST has
been to translate updated experimental findings into a
manualised 10 session individual treatment package. The
aforementioned studies helped us to evaluate the first version of the manual; and feedback from clinicians and patients as well as more recent research findings (e.g. in
positive psychology [7], the expression of emotions [8-10],
social communication difficulties [11], social anhedonia
[12,3], and emotional intelligence [13]) enabled us to revise the manual. We then offered the revised version to inpatients within the adult service and asked them to
evaluate its benefits using self-report questionnaires.
The aim of this study was to explore the self-reported
outcomes from inpatients with AN who received an intervention which targets emotion skills. This was not done in
the previous quantitative and qualitative studies evaluating
the CREST manual. Specifically, we were interested in exploring whether social anhedonia and alexithymia (the
ability to recognise, describe and express emotions), both
of which are targeted by CREST, showed improvements
after 10 individual sessions of therapy. Social anhedonia
and alexithymia were chosen as targets due to the difficulties in these areas experienced by patients with AN, which
has been demonstrated extensively by research [3,12].
CREST provides psycho-education on the research findings from the studies mentioned above, as well as a variety
of skills based, concrete exercises which are tailored to suit
the specific needs of the patients. Some of these exercises
involve the following: helping patients to name their emotions and feelings (e.g. by giving them a list of emotion
words and encouraging them to think about how they feel
at that particular moment), encouraging patients to share
and express their emotions in a safe way, supporting them
to find good strategies around being with other people
and communicating their emotional states (e.g. by stating
what they think or how they feel), increasing their awareness of pleasurable hedonic experiences (e.g. by generating
lists and sharing readily available hand-outs of lists of
pleasurable things which do not involve social eating), and
helping them to develop positive biases (e.g. by looking at
things from different angles, training the brain to notice 3
positive things every day, and trying to identify what is
good about the treatment programme or the people with
whom we communicate on the ward).
Overall, the aim of CREST is to normalise emotions via
the provision of research evidence and psycho-education
as to the nature and function of emotion processing.

Page 2 of 6

Through the experiential exercises in sessions and homework tasks, it is hoped that patients will develop a vocabulary for emotional experience which can enable them to
become more assertive in meeting their needs.
The format of the intervention is simple, brief, and designed for nutritionally compromised patients in an individual format. The intervention used in the current study
overlaps with cognitive remediation therapy (CRT) as it includes specific and very concrete examples and activities to
facilitate discussion, with materials based on art, metaphors, and scenarios (for details on CRT exercises and
the clinical manual please see the following website:
http://www.national.slam.nhs.uk/about-us/our-experts/
dr-tchanturia/).
The aims of this study were to investigate whether the
individual format of CREST can produce changes on
self-reported social anhedonia, alexithymia, and motivation to change measures.

Methods
Participants

The participants receiving CREST were inpatients from

the South London and Maudsley National Adult Eating
Disorders service. All patients who had been referred consecutively to the inpatient ward were offered individual
CREST. Ethical approval was gained from the National
Health Service (NHS) ethics committee (ref: 08/H0606/58).
All participants gave written consent, prior to participation in the study. 37 patients were assessed before the
CREST intervention started. Of this number, 33 patients
(89%) attended all 10 sessions and provided both pre- and
post-intervention primary outcome measures. There were
some patients in the inpatient programme who received
CREST, but did not complete the questionnaires (N = 15)
(see Figure 1).
CREST

Cognitive Remediation and Emotion Skills Training

(CREST) is a low-intensity individual manualised treatment for inpatients with severe AN. It aims to target rigid
and detail-focused thinking styles (in 1 session), but places
greater emphasis on the development of emotion recognition skills (in ourselves and others), and the management
and expression of emotion in AN. In CREST, therapists
provide updated psycho-educative material (from KT who
supervises the therapy and updates the materials with new
experimental findings) and facilitate simple, collaborative
cognitive tasks and game-like activities that encourage reflection on emotion processing skills (for example, by presenting balls with emotion expressions and discussing what
it is like to be sad, happy, or curious) as we have found this
is more conducive to discussion than is the approach of
asking open ended questions. Therapists also encourage patients to practice implementing small behavioural changes

Tchanturia et al. BMC Psychiatry (2015) 15:53

Page 3 of 6

Figure 1 Consort diagram detailing the collection of self-reported outcome measures for CREST.

(e.g. noticing and saying what went well and letting their
key nurses know that I felt supported today or I enjoyed
group walk). Different modules within CREST are designed to help individuals learn about a) the function of
emotions, b) how to label and identify emotions in oneself
and others, c) the positive intentions of emotions and the
needs emotions communicate to the self and others, and
d) tolerating and expressing emotions. The intervention is
based on the cognitive interpersonal model, and has been
adapted to address additional emotional processes following empirical reviews of the existing literature which
highlighted what the main areas of emotional difficulty appear to be for people with AN. It was also informed by
service user, carer, and clinician input regarding which aspects of emotion processing were perceived to be the most
difficult and necessary areas for intervention [14].

Assessments

The first primary outcome measure of this study, the

Revised Social Anhedonia Scale (RSAS), assesses the reduced ability to experience social pleasure [15]. Higher
scores on the RSAS indicate greater discomfort in being
with other people. The scale is comprised of 40 items
measured by true or false statements. Examples of
items from the RSAS include: I prefer watching television to going out with other people and I attach very
little importance to having close friends. This measure

has been used in the eating disorder patient population

several times [3,12]. In one study [12], researchers found
a mean score of 16.2 among their group of 105 AN patients, while the corresponding value among 136 noneating disordered controls was 6.1. Similarly, another
study [3] reported a mean score of 16.4 among 72 AN
patients, whereas their non-eating disordered group had
a mean score of 5.5.
The Toronto Alexithymia Scale (TAS) [16] is the second primary outcome measure of this study. It consists
of 20 items and three subscales assessing difficulties in
identifying feelings (e.g. I am often confused about what
emotion I am feeling), difficulties in describing feelings
(e.g. It is difficult for me to find the right words for my
feelings), and degree of externally-oriented thinking
resulting in a preoccupation with the details of external
events (e.g. I prefer talking to people about their daily
activities rather than their feelings). This is the most
widely used measure of alexithymia. A score of less than
51 indicates non-alexithymia, equal to or greater than 61
indicates alexithymia, and 5260 indicates possible alexithymia. The TAS-20 has been found to have good internal consistency for the total score (a = .81), acceptable
internal consistency for the subscale scores, and good
test-retest reliability (p < 0.01) [16].
A motivational ruler was also administered and includes two questions which explore beliefs about the
importance to change and perceived ability to change.

Tchanturia et al. BMC Psychiatry (2015) 15:53

Page 4 of 6

Outcome measures

Figure 2 Effect sizes of the primary self-report outcome measures

(RSAS and TAS total score) and the secondary outcome measures
(Importance to change, Ability to change, and BMI), post-therapy.

The questions are answered on a Likert 010 scale.

Higher scores indicate more positive beliefs about ones
importance/ability to change.

Statistical analysis

SPSS version 21 was used to analyse the data.

Pairwise t-tests were conducted to explore differences
in pre- and post-intervention levels of social anhedonia,
alexithymia, motivation and BMI. Cohens d (mean1mean2/pooled standard deviation) was calculated to provide effect sizes for normally distributed data, with an effect size of <0.2 defined as small, 0.5 defined as
medium and 0.8 defined as large. The results are presented in Figure 2.

Results
Group characteristics

The mean age of the participants was 24.5 years (SD = 8.2),
and the range was 1854 years. The average duration of
illness (measured at the start of CREST) was 8 years
(SD = 7.2). The mean BMI at the start of therapy was
15.1 (SD = 1.95).

RSAS and TAS total score, motivation (both perceived

importance to change and ability to change), and BMI
were measured at pre- and post-intervention (the time between assessment before and after CREST was 10 weeks
on average). These values along with those from the ttests conducted are displayed in Table 1.
RSAS and TAS total scores were found to have decreased significantly following the intervention, with
small effect sizes. With regard to motivation, perceptions
of importance to change did not differ significantly between pre- and post-intervention, but ability to change
increased significantly after CREST, with a medium effect size. BMI also increased significantly between the
beginning and the end of therapy, with a medium effect
size (see Figure 1).

Discussion
The aim of this exploratory study was to evaluate the efficacy and impact of a novel and brief skills-based therapy
for inpatients with anorexia. Recent research findings have
highlighted that abstract and flexible thinking styles seem
to be less present in individuals with AN [17,3], despite
other cognitive strengths such as high IQ [18] and good
working memory [19] remaining intact. Eating disorders
research has also made progress in characterising low
emotional intelligence [13], poor positive emotional expression [8-10], difficulties with social communication,
difficulties in friendships [11], emotionally-driven thinking
as hot cognition [20,21], high social anhedonia [3,12], difficulties with social interactions [22,11], and the patterns
of both positive [7] and negative [23] emotion processing.
Qualitative assessments of patients emotional needs [14],
along with these recent research developments, have been
translated to inform treatment interventions such as cognitive and emotional remedial therapies; for example,
CRT [24,25] and CREST [5].
This study is the first of its kind in that we have explored the feasibility of delivering CREST in a short (10
sessions), individual format using self-report measures of
social anhedonia and alexithymia. We found a clinically
significant decrease in social anhedonia and alexithymia
among the participants, and these aspects were directly

Table 1 Primary and secondary outcome measures at the beginning and end of therapy
Outcome measures

Pre-intervention

Post-intervention

8.13

0.03

0.31

59.8

9.79

0.05

0.35

7.87

2.69

1.00

0.00

30

6.29

2.16

0.00

0.71

37

16.4

2.06

0.00

0.66

Mean

SD

Mean

SD

RSAS

34

15.3

7.49

34

12.9

TAS

33

63.1

9.18

33

Importance to change

30

7.87

2.71

30

Ability to change

30

4.66

2.47

BMI

37

15.1

1.95

Tchanturia et al. BMC Psychiatry (2015) 15:53

targeted by CREST. BMI and motivation were not directly addressed in CREST, but were used here as secondary measures to assess improvements in other important
domains. The results show that the main outcomes, such
as improved use of emotional vocabulary and development of the ability to be with other people, changed in
the positive direction. Small effect sizes (0.31-0.35) and
statistical significance together with patients comments
on how they benefited from CREST suggest that CREST
is a useful tool for this severe group of patients receiving
inpatient treatment.
We are aware of several limitations within the current
study and will try to improve upon these shortcomings
in future research and evaluation. For example, we
would like to explore in more detail exactly who benefits
the most from this intervention and audit more precisely
the clinical characteristics of patients who either do not
respond to or do not choose to take part in this therapy.
It would be desirable to explore the presence of autistic
characteristics within this group of patients, since the
current research findings suggest that many inpatients
with AN have elevated levels of autistic traits [26]. In
summary, this study provides some evidence that our
brief CREST intervention can influence social anhedonia, the ability to recognise ones own emotions, and
confidence in ones ability to change. These findings are
promising because from previous research we know that
social anhedonia is highly correlated with chronicity of
illness and the current data has been collected from
adult patients with a long duration of illness (the mean
duration of illness in this study is 8 years). However, further research is needed in order to corroborate these
findings; to be precise, a randomised controlled design
study with a larger sample size.
Strengths

This is our attempt to contribute towards intervention

developments targeting emotion skills in severely nutritionally compromised patients with AN. The intervention has been positively received by patients and
therapists, and the manual has been informed by the latest research updates. Well-tested outcome measures
which seem to tap into problematic areas for eating disorder patient groups have been selected, which also capture changes after CREST.
Limitations

The absence of accurate information explaining the reasons for drop out cases (e.g. early discharge, not wanting to complete questionnaires) is a clear limitation of
this study.
Further improvements of this study would include addressing the limitation mentioned above and refining our
assessment battery to include tests measuring emotion

Page 5 of 6

expression, an area into which CREST taps and which has

been shown to be sub-optimal in people with AN [8,9].

Conclusion
Promising results from the self-report measures combined
with meaningful feedback from the patients participating,
leads us to the conclusion that it would be worthwhile to
further develop this line of clinical research.
Competing interests
The authors declare that they have no competing interests.
Authors contributions
KT devised the therapy manual, selected the outcome measures and drafted
the manuscript. ED collected the data, and contributed to the writing of the
manuscript. VM delivered the therapy and commented on the draft of the
manuscript. CF delivered the therapy and commented on the draft of the
manuscript. All authors read and approved the final manuscript.
Acknowledgements
KT would like to acknowledge funding and support from the Swiss Anorexia
Foundation, the Maudsley Charity Health in Mind, the NIHR Biomedical
Research Centre for Mental Health at South London and Maudsley NHS
Foundation Trust, and the Institute of Psychiatry, Psychology and Neuroscience
at Kings College London. We would like to thank our patients for their
informative feedback and active participation in the development of this
therapeutic package. Many thanks also to our researchers and clinicians who
contributed to the development of this work: Helen Davies, David Hambrook,
Clare Money, Naima Lounes and Emma Smith.
Author details
1
Kings College London, Division of Psychological Medicine, Institute of
Psychiatry, SE5 8AF London, UK. 2Eating Disorders Unit, South London and
Maudsley NHS Foundation Trust, London, UK. 3Illia State University, Tbilisi,
Georgia.
Received: 4 December 2014 Accepted: 9 March 2015

References
1. NICE. Eating Disorders: Core Interventions in the Treatment and
Management of Anorexia Nervosa, Bulimia Nervosa and Related Eating
Disorders. London: National Institute for Clinical Excellence; 2004.
2. Treasure J, Claudino AM, Zucker N. Eating disorders. Lancet.
2010;375:58393.
3. Tchanturia K, Davies H, Harrison A, Fox J, Treasure J, Schmidt U. Altered
social hedonic processing in eating disorders. Int J Eating Dis.
2012;45:9629.
4. Tchanturia K, Doris E, Fleming C. Effectiveness of cognitive remediation and
emotion skills training (CREST) for anorexia nervosa in group format: A
naturalistic pilot study. Eur Eat Disord Rev. 2014;22:2005.
5. Davies H, Fox J, Naumann U, Treasure J, Schmidt U, Tchanturia K. Cognitive
remediation and emotion skills training (CREST) for anorexia nervosa: an
observational study using neuropsychological outcomes. Eur Eat Disord Rev.
2012;20:2117. 10.1002/erv.2170.
6. Money C, Genders R, Treasure J, Schmidt U, Tchanturia K. A brief emotion
focused intervention for inpatients with anorexia nervosa: A qualitative
study. J Health Psychol. 2011;16:94758.
7. Tchanturia K, Marin Dapelo M, Hambrook D, Harrison A. Why study positive
emotions in the context of eating disorders? Curr Psychiatry Rep.
2015;17:537. 10.1007/s11920-014-0537.
8. Davies H, Schmidt U, Stahl D, Tchanturia K. Evoked facial emotional
expression and emotional experience in people with anorexia nervosa.
Int J Eating Dis. 2010;44:5319.
9. Davies H, Swan N, Schmidt U, Tchanturia K. An experimental investigation
of verbal expression of emotion in anorexia and bulimia nervosa. Eur Eat
Disord Rev. 2012;20:47683.
10. Rhind S, Mandy W, Treasure J, Tchanturia K. An exploratory study of evoked
facial affect in adolescents with anorexia nervosa. Psychiatry Res. 2014;5:56.

Tchanturia et al. BMC Psychiatry (2015) 15:53

Page 6 of 6

11. Doris E, Westwood H, Mandy W, Tchanturia K. Patients with anorexia

nervosa show similar friendships difficulties to people with autism spectrum
disorders: a qualitative study. Psyc Special Issue Autism. 2014;5:133849.
12. Harrison A, Mountford V, Tchanturia K. Social anhedonia and work and
social functioning in the acute and recovered phases of eating disorders.
Psychiatry Res. 2014;218:18794.
13. Hambrook D, Brown G, Tchanturia K. Emotional intelligence in anorexia
nervosa: is anxiety a missing piece of the puzzle? Psychiatry Res.
2012;200:129.
14. Kyriacou O, Easter A, Tchanturia K. Comparing views of patients, parents and
clinicians on emotions in anorexia: a qualitative study. J Health Psychol.
2009;14:84354.
15. Chapman LJ, Chapman JP, Roulin ML. Scales for physical and social
anhedonia. J Abnorm Psychol. 1976;85:37482.
16. Bagby RM, Parker JDA, Taylor GJ. The twenty item Toronto Alexithymia Scale
I Item selection and cross-validation of the factor structure. J Psychosom Res.
1994;38:2332.
17. Lang K, Lopez C, Stahl D, Tchanturia K, Treasure J. Central coherence in
eating disorders: an updated systematic review and meta-analysis.
World J Biol Psyc. 2014;1:114.
18. Lopez C, Stahl D, Tchanturia K. Estimated IQ in anorexia: a systematic review
of the literature. Annals General Psyc. 2010;23:940.
19. Lao-Kaim NP, Giampietro V, Williams SCR, Simmons A, Tchanturia K.
Functional MRI investigation of verbal working memory in adults with
Anorexia Nervosa. Eur Psychiatry. 2014;29:2118.
20. Russell T, Schmidt U, Tchanturia K. Aspects of social cognition in anorexia
nervosa: affective and cognitive theory of mind. Psychiatry Res.
2009;15:1815.
21. Oldershaw A, Hambrook D, Stahl D, Tchanturia K, Treasure J, Schmidt U. The
socio-emotional processing stream in Anorexia Nervosa. Neurosci Biobehav
Rev. 2011;35:97088.
22. Harrison A, Genders R, Davies H, Tchanturia K. Experimental measurement
of the regulation of anger and aggression in women with Anorexia
Nervosa. Clin Psyc Psychother. 2011;18:44552.
23. Harrison A, Sullivan S, Tchanturia K, Treasure J. Emotional functioning in
eating disorders: attentional bias, emotion recognition and emotion
regulation. Psychol Med. 2010;40:188797.
24. Tchanturia K, Lloyd S, Lang K. Cognitive Remediation in eating disorders. Int
J Eating Dis Special Issue. 2013;46:4926.
25. Tchanturia K, Lounes N, Holtum S. Cognitive remediation in anorexia
nervosa and related conditions: a systematic review. Eur Eat Disord Rev.
2014;6:45462.
26. Tchanturia K, Smith E, Weineck F, Fidanboylu E, Kern N, Treasure J, et al.
Exploring autistic traits in anorexia: a clinical study. Mol Autism. 2013;4:44.

Submit your next manuscript to BioMed Central

and take full advantage of:
Convenient online submission
Thorough peer review
No space constraints or color gure charges
Immediate publication on acceptance
Inclusion in PubMed, CAS, Scopus and Google Scholar
Research which is freely available for redistribution
Submit your manuscript at
www.biomedcentral.com/submit

BioMed Central publishes under the Creative Commons Attribution License (CCAL). Under
the CCAL, authors retain copyright to the article but users are allowed to download, reprint,
distribute and /or copy articles in BioMed Central journals, as long as the original work is
properly cited.