Cognitive remediation therapy for anorexia nervosa: Current evidence and future research directions



No abtsract.

The Need for New Ideas in the Treatment of Anorexia Nervosa

The effective treatment of anorexia nervosa (AN) remains a significant challenge. This had prompted new research into ways of engaging and keeping patients in treatment, and ultimately achieving better outcomes, not only on a symptomatic level but also in broader aspects of life.1 In an attempt to improve treatment outcomes for AN, there has been a move toward approaches that target the core maintaining factors of the disorder. Impaired cognition has been implicated in the maintenance of AN, contributing to individuals' difficulties in processing and in engaging with psychological therapy. An established body of existing research (including both systematic reviews and large sample studies) currently highlights two main areas of difficulty—cognitive flexibility (particularly set-shifting difficulties) and an extreme attention to detail (or weak central coherence).2 Both inflexibility and an overly detail-focused way of thinking are prevalent in adults with AN, and appear to be exaggerated in the acute stage of the illness.3

Translation of Neurocognitive Research into Treatment: Cognitive Remediation Therapy

Behavioral correlates of cognitive inflexibility can be seen clinically in patients' fixation with certain behaviors or routines (e.g., specific rules and ritualized eating behaviors at meal times). Furthermore, a rigid thinking style can have negative consequences on patients' interactions with the outside world and their ability to organize information. Unsurprisingly, this cognitive style poses challenges for the treatment of adults with AN. Extreme focus on detail and related behaviors (e.g., calorie counting, exercise, and obsessional rituals) and resistance to change each negatively impact upon engagement.

The way an individual processes information is crucially important to how they make sense of situations in their environment. Both an extreme focus upon detail and limited flexibility can impair social interactions. Therefore, before attempting to change strategies and behaviors, it is important to encourage patients to develop an awareness of their thinking style. Research findings relating to extreme attention to detail and cognitive inflexibility have led to the adaptation and development of cognitive remediation therapy (CRT) for AN.4 Initially developed for use with patients with brain lesions and acquired brain injury, cognitive remediation has since been used effectively in a range of disorders including psychosis and attention deficit hyperactivity disorder, with the format and focus adapted depending upon the nature of the illness and context. Thus, the development of a CRT approach for AN involved tailoring the treatment toward set-shifting, cognitive exercises for holistic thinking, and the development of behavioral exercises.

Cognitive Remediation Therapy for Anorexia Nervosa: A Snapshot

CRT for AN targets cognitive style broadly, rather than directly focusing on issues with eating, weight, or shape. The intervention consists of simple exercises designed to develop and encourage a more flexible and holistic thinking style. These include focusing on the “bigger picture” rather than details, prioritizing information, and estimating (which challenges maladaptive perfectionist tendencies). The simple cognitive exercises provide specific and non-threatening material (at the level of “doing” tasks) to encourage curiosity, reflectiveness and insightfulness about one's own thinking process (the metacognitive level). There is emphasis upon practicing such skills first in everyday behaviors (e.g., changing one's hairstyle or accessories) before targeting illness-specific rules and thinking.

CRT is a brief individual intervention (8–10 sessions), which provides a good introduction to further psychological treatment. It encourages patients to discover the ways in which they think and decide what they would like to change and how. CRT also encourages the use of intra-session experimentation with different ways of thinking and behaving, which can then be reflected upon during the next session. The style of delivery is motivational, and the therapist is able to model inefficiencies and imperfect performance. CRT can be delivered by therapists from across a range of disciplines (psychology, nursing, occupational therapy, and social work). Training and supervision are essential, as with any other form of psychological treatment.

The Evidence So Far: Published and Forthcoming Cognitive Remediation Therapy Studies

Following a small case series and production of a treatment manual, a further early case series was carried out.4 Existing studies demonstrate improvements in neuro-cognitive performance (medium to large effect sizes in flexibility and central coherence tasks) and low drop-out rates from treatment (around 10–15%). The approach has a high level of acceptability to both patients and therapists.5 Whilst a detailed discussion of study findings is beyond the scope of this article, readers are referred to Table 1 for a summary of the existing published studies, including both quantitative and qualitative analyses. We are also aware of three large randomized controlled trials of CRT that are in progress.

Table 1. Summary of published studies reporting CRT in AN
PublicationDescription of patients and study designMain findings
  1. M: mean; SD: standard deviation.

Tchanturia K, Davies H, Campbell IC. Cognitive remediation for patients with Anorexia Nervosa: preliminary findings. Ann Gen Psychiatry 2007;6:14. N = 4; Inpatient adults; age range = 21–42; illness duration range = 7–24 years.Improvements in neuropsychological task performance (set-shifting measured in different domains) post treatment (medium to large effect sizes).
Pretorius N, Tchanturia K. Anorexia nervosa how people think and how we address it in psychological treatment. Therapy 2007;4(4):423–433.Case study; inpatient adult (31 years); illness duration = 1 yearPatient's clinical improvements and feedback on the tasks reported.
Tchanturia K, Davies H, Lopez C, Schmidt U, Treasure J, Wykes T. Neuropsychological task performance before and after cognitive remediation in anorexia nervosa: A pilot case series. Psychol Med 2008;38(9):1371–1373. N = 23; inpatient adults; M age = 28.8 (SD = 9.6); illness duration M = 13.1 (9.6)Medium to large effect size improvements in set-shifting and global processing cognitive style.
Low drop-out rate (4/27)
Cwojdzińska A, Markowska-Regulska K, Rybakowski F. Cognitive remediation therapy in adolescent anorexia nervosa—Case report. Psychiatr Pol 2009;43(1);115–124. [Article in Polish].Case study, adolescentThe paper is in Polish; The abstract reports general improvements in clinical symptoms and set-shifting.
No information available on age or illness duration.
Genders R, Tchanturia K. Cognitive remediation therapy (CRT) for anorexia in group format: a pilot study. Eat Weight Disord, 2010;15(4),e234–239. N = 18; inpatients, group formatShort group format found to be acceptable for patients. Statistically significant changes on the 'ability to change' scale. No significant changes in self-reported cognitive flexibility.
Pitt S, Lewis R, Morgan S, Woodward, D. Cognitive remediation therapy in an outpatient setting: a case series. Eat Weight Disord 2010;15(4),e281–286. N = 7; outpatient adults; M age = 29; illness duration range = 3–22 yearsImprovements in self-reported flexibility of thinking after individual CRT. Mixed results for self-reported perfectionism.
Wood L, Al-Khairulla H, Lask B. Group cognitive remediation therapy for adolescents with anorexia nervosa. Clin Child Psychol Psychiatry 2011;16(2):225–231. N = 9; Inpatient adolescents; age range = 9–13.Positive observations from group members and clinicians were reported.
Abbate-Daga G, Buzzichelli S, Marzola E, Amianto F, Fassino S. Effectiveness of cognitive remediation therapy (CRT) in anorexia nervosa: A case series. J Clin Exp Neuropsychol 2012. N = 20; outpatient adults; M age = 22.5 (SD = 3.9); illness duration = 5.8 yearsMedium to large effect sizes reported on most neuropsychological tasks, e.g. TMT, WCST (Cohen's d = 0.6) and positive improvement in clinical characteristics.
Pretorius N, Dimmer M, Power E, Eisler I, Simic M, Tchanturia K. Evaluation of a Cognitive Remediation Therapy group for adolescents with Anorexia Nervosa: pilot study. Eur Eat Dis Rev 2012;20(4):321–325. N = 30 (7 groups); daypatient adolescents; M age = 15.6 (1.4) ; Illness duration M = 2 yearsSmall effect size for self-reported cognitive flexibility post group. Adolescents found the group interesting and useful; however, some wanted more support with application to real life.

In Table 2, we also report on recent findings from our own service in the form of previously unpublished data from 46 individual cases and from 25 patients who took part in a short group format of CRT in an adult inpatient ward. For those in individual treatment, Table 2 shows improvements of moderate size in flexibility of task performance, and small effect sizes in both global processing and self-reported flexibility. In the group format, four to five sessions of the intervention in an adult inpatient setting was associated with improvements in confidence to change (measured using a motivational ruler), and small-sized effects in self-reported cognitive flexibility. Further work is needed to test the effectiveness of group CRT and to define the optimum number of sessions. For example, feedback from both patients and clinicians suggests an increase in the number of group sessions.

Table 2. Updated evaluation of the case series (n = 46) presented: cognitive task performance before and after CRT in individual format
 Before CRT (n = 46)After CRT (n = 46)Significance (p-value)Effect size (Cohen's d)
  1. In Tchanturia et al. (2008), data for 23 cases was presented; this table contains an additional 23 cases.

Brixton (set shifting task)— Total (n = 46)13.3 (6.2)9.5 (6.1)< .0010.62
RCFT Style index (N = 45) Central coherence task1.14 (0.4)1.26 (0.5).010.25
Cognitive Flexibility Scale CFS (n = 36) Self-report measure43.05 (10.1)46.19 (10.0).030.32

Summary and Future Directions

In summary, the role of CRT in the treatment of AN is an idea worth researching, given its clear hypothesized links between brain function, psychological function, and treatment. Findings from our own published studies and current data, along with other replications and extensions demonstrate a relatively consistent picture—CRT is associated with cognitive improvements in this population. It is also associated with low drop-out rates and high levels of acceptability among both patients and therapists. Such findings will be of interest to clinicians given the difficulty of engaging patients with AN in therapy and the role of early engagement as a strong predictor of treatment outcome.

There are a number of promising avenues for future research. These include:

  • 1Developing our understanding of cognitive characteristics in the adolescent population by synthesizing the literature on young cases.
  • 2Direct comparison with the effects of other therapies, to determine whether the benefits reported here are specific to CRT or represent broader changes (e.g., refeeding).
  • 3Exploring how cognitive improvements may influence AN symptomatology and broader outcomes (e.g., work and social aspects).
  • 4Examining the length of CRT intervention, to clarify which patients benefit from short- and long forms of the intervention. Further research is also needed regarding whether individual or group formats of CRT are most beneficial for patients with AN.
  • 5Research has also highlighted the importance of involving family members and carers in the treatment of AN. Future research aims to focus on the development of a “user-friendly” module of CRT, which could be delivered by carers. Further work is needed to evaluate the effectiveness and acceptability of CRT delivered in this way.
  • 6Further work is needed to explore the brain mechanisms implicated in cognitive impairment in AN. Whilst there is strong evidence for such impairments, little is known regarding the specific neural mechanisms involved. Neuroimaging studies are needed in order to explore this, and would also enable investigation of the cognitive and functional brain changes associated with clinical improvements following CRT. This would both strengthen the support for neuropsychological inefficiencies in AN and the evidence base for CRT.

A further, more ambitious direction for future research is to investigate how socio-emotional focused interventions can be integrated with this line of treatment. There is support for such an approach, given strong evidence for socio-emotional impairments in AN and for the superior effectiveness of interventions that focus upon associations between cognitions, emotions, and behaviors in treating eating disorders. Initial research in this area has already started.6