CONTENTS

Keynote Addresses
Social Relationships and Schizophrenia. Christine Barrowclough
On the Psychology of Psychosis. Richard Bentall
Early Intervention in Psychosis, Max Birchwood
Was Epictetus Right? Do Negative Thoughts Cause Depression? David Burns
Clinically Effective CBT for Insomnia - and How it Might Work, Colin Espie
Cognitive Behavioural Treatment for Eating Disorders, Christopher G. Fairburn
Cognitive Therapy: A Treatment for the New Millenium, Melanie Fennell
Intrusions, Obsessions, Rumination and Worry, Mark H. Freeston
Active Ingredients in Child and Adolescent Therapy, Philip C. Kendall
Depressive Decision-Making, Robert Leahy
A Regulatory Systems Approach to Anger and Violence, Raymond W. Novaco
The Development and Prevention of Anxiety Disorders, Ron Rapee,
Cognitive Models of Bipolar Disorder: Theory and Therapy, Jan Scott

Symposia
1. Prognostic Factors in Cognitive Behaviour Therapy for Schizophrenia
2. Advances in the Understanding & Treatment of Body Image Disturbance
3. CBT Self-help Treatments
4. The Understanding of Cognition in Children and Adolescents.
5. Issues of Coping and Vulnerability
6. Psychological Processes and Psychotic Symptoms
7. Family Care in Psychoses
8. Sexual Dysfunction
9. Dissemination of Effective Interventions
10. Cognition And Emotion In Children And Adolescents
11. Intrusions, Rumination, Control and Suppression Across Disorders.
12. Trauma and Mood Disturbance Issues
13. Cognition and Emotion Across Disorders
14. Early Intervention in Psychosis
15. Psychological Approaches and Sleep, Part 1
16. New Developments in Anger Treatment
17. Therapeutic Strategies Across Disorders
18. Disgust - The Forgotten Emotion?
19. Is There a Role for Cognitive Behaviour Therapy in Bipolar Disorder?
20. Psychological Approaches and Sleep, Part 2: Recent Empirical Findings
21. Clinically Effective and Efficient Approaches for RoutineClinical Work.
22. Domestic Violence
23. Investigating Imagery
24. Do the Effects of CBT Endure? Follow-up of GAD and Panic Disorder
25. Evaluating The Effectiveness Of CBT Training
26. Basic and Applied Issues in Eating Disorders
27. Issues of Cognition and Emotion

Roundtables
1. Analogies, Stories, Metaphors: Tools for Clinicians
2. Children and ADHD, Alternatives to Amphetamines
3. Supervision in Cognitive-behaviour Therapy: Who, What and How?
4. Voluntary Self-regulation or Statutory Registration: What Does it Mean for You?

Open papers

Poster Sessions

KEYNOTE ADDRESSES

Social Relationships and Schizophrenia
Christine Barrowclough, University of Manchester
There is a dynamic association between social relationships and emotional well being and this is illustrated very well in schizophrenic illness. Schizophrenia is characterised by impairments and difficulties in interpersonal functioning, and at the same time interpersonal problems may exacerbate the symptoms. Understanding this dynamic association has considerable clinical implications.
Social relationships and schizophrenia have been studied most intensively in the context of families where attitudes are measured using expressed emotion (EE) dimensions which have been shown to have an influential role on patient outcome. Recent studies have demonstrated that such a problematic response is not something unique to the family nor to schizophrenia: negative staff attitudes are associated with worse patient outcomes; and the relationship between high EE and worse outcome holds for many psychiatric and medical conditions.
Given the importance of social relationships in schizophrenia, a key question pertinent to both familial and staff -patient interactions is why do some people develop high EE? Whilst it is widely accepted that cognitive processes mediate people's adaptations to their own health threats, carers perception and evaluations of the illness experience have had very limited study. However, there is now some evidence that cognitive appraisals of illness symptoms underlie reactions of family members to schizophrenic illness. Well replicated work on attributions has demonstrated that carers make many and varied spontaneous attributions about the patient's symptoms which are related to EE dimensions. Recent studies also suggest that carers' models of mental illness may have a similar structure to those held by people about physical illness and that these models carry behavioural and coping outcomes. What are the mechanisms by which high EE exacerbates symptoms? Whilst it is acknowledged that the EE/ heightened arousal hypothesis for symptom exacerbation is very important, a new study raises the possibility that critical attitudes from significant others may impact on negative self concept which in turn maintains or exacerbates positive symptoms.

On the Psychology of Psychosis: Towards a Unifying Framework for Understanding Depression, Paranoia and Mania.
Richard Bentall, University of Liverpool
In the last decade, significant advances in understanding psychopathology has been achieved by using psychological models to explain specific type of behaviours and experiences ('symptoms') rather than broad diagnostic categories such a 'schizophrenia'. However, it is unlikely that entirely different cognitive systems are affected in each type of abnormal behaviour. Instead, different abnormal behaviours may reflect different types of abnormalities in the same cognitive systems. Moreover, cognitive abnormalities are for the most part not trait-like, but vary with time. By taking into account these factors it may be possible to gain insights into why psychopathological phenomena sometimes covary, and why their course is often unpredictable.
These ideas will be examined by exploring the mechanisms involved indepression, paranoia and mania. Starting with what might be called a 'generic cognitive model', in which appraisals of events effect beliefs about the self, thereby bringing about changes in mood, it will be shown how developments of this model can explain persecutory delusions and manic symptomatology. This exploration will lead to the idea of the 'attribution - self-representation cycle', a model of the nonlinear relationships between the appraisals and self-representations (beliefs about the self, standards of self-evaluation, autobiographical memories) that attempts to account
for the instability of these symptoms over time. Implications of this model for psychiatric classification, for biological research into psychopathology, and for clinical interventions will be discussed.

Early Intervention in Psychosis
Max Birchwood, University of Birmingham
The UK Government has recently announced its firm intention to fund 50 new early intervention services for young people experiencing psychosis for the first time.
In this paper, I will critically review the conceptual basis for the early intervention approach and discuss the service gaps which currently exist in our services for first episode psychosis. I will argue that CBT interventions should be bought to bear on the key transition points in first episode psychosis, including transitions to the first psychotic episode, the management of treatment resistance, but most importantly, the prevention of depression, suicidal thinking and traumatic

Was Epictetus Right? Do Negative Thoughts Cause Depression?
David Burns, Stanford University School of Medicine, Stanford, USA
Cognitive behavioral therapy (CBT) proposes that negative cognitions cause depression, anxiety, and anger. They also propose that self-defeating beliefs (SDBs) trigger episodes of depression and that changes in SDBs mediate recovery. Typical SDBs include the belief that one must be perfect, or the belief that one must be loved to be worthwhile.
It is difficult to test these theories because of the problem of circular causality. Negative cognitions may lead to negative emotions, but negative emotions may have reciprocal causal effects on cognitions. In addition, it is possible that negative cognitions and emotions are not causally linked. The correlation between these variables could result from an unknown third variable that activates them simultaneously.
Elucidating the causal linkages is crucial. If negative cognitions do not have a causal effect on emotions, then the theoretical basis of CBT would be invalid. Dr. Burns will present two new studies that examine the causal links between cognitions and emotions. The results suggest a powerful, high speed reciprocal causal link between negative cognitions and emotions, as proposed by Epictetus nearly 2000 years ago.
However, there no causal linkages between SDBs and depression or anxiety were detected. Furthermore, recovery during CBT did not appear to be mediated by changes in SDBs.

Clinically Effective CBT for Insomnia - and How it Might Work
Colin Espie, University of Glasgow
A least 10% of adults and 20% of older adults experience persistent difficulty in getting to sleep or remaining asleep. Sleeplessness and its daytime consequences, therefore, represent a considerable public health problem, and one that is increasingly recognised in terms of health costs. Pharmacotherapy is of limited benefit in chronic insomnia, but cognitive-behavioural treatments have been widely evaluated and may now be the treatment of first choice. This paper briefly reviews the literature on the efficacy of CBT in this population, but concentrates primarily on addressing three commonly asked questions. First, will CBT really work with my patients? The evidence on clinical effectiveness will be considered, that is, whether or not CBT is useful in ordinary clinical settings. Second, can CBT be delivered cost-effectively in practice? A validated model using manualised, small group intervention format will be described. Third, if CBT works, why does it work? A recently revised, integrated model of normal sleep and sleep disturbance will be presented. Critical pathways in the development and maintenance of insomnia will be discussed and consideration given to possible critical ingredients in effective therapy. In relation to each of these questions, avenues for future research will be highlighted.

A New "Transdiagnostic" Cognitive Behavioural Treatment for Eating Disorders
Christopher G. Fairburn, Department of Psychiatry, University of Oxford
The research on the treatment of eating disorders has focused largely on the treatment of bulimia nervosa. The most effective treatment is a specific form of cognitive behaviour therapy (CBT-BN). A mounting body of evidence provides support for the cognitive behavioural theory that underpins CBT-BN but the treatment only results in about half the patients making a full recovery. This suggests that either the treatment procedures need to be improved (or implemented more effectively), or the theory needs to be modified and the treatment adapted accordingly, or both.
In Oxford we have developed an enhanced and extended form of CBT-BN. It has been enhanced by the use of improved treatment procedures and it has been extended to address additional maintaining mechanisms. A novel feature of the resulting "modular" treatment is that it is "transdiagnostic"; that is, it is designed to be suitable for all forms of clinical eating disorder. The new treatment, and the theory upon which it rests, are shortly to be the subject of a large randomised controlled trial.

Cognitive Therapy: A Treatment for the New Millenium
Melanie Fennell, University of Oxford
The evidence base for cognitive therapy has expanded dramatically in the last 10-15 years, encompassing not only common mental health problems but also severe mental illness. At the same time, the resources to provide effective, high quality psychological treatments within the National Health Service are increasingly constrained. The presentation argues that cognitive therapy is ideally placed to meet the standards and objectives of contemporary mental health care - except that access to competent practitioners is limited. Rather than trying to solve this problem by quick and dirty means, we should think carefully about how knowledge and skills in cognitive therapy can best be taught, so as to provide NHS patients with the high quality treatment they need and deserve

Intrusions, Obsessions, Rumination and Worry: What Do We Know and Where Are We Going?
Mark H. Freeston, University of Newcastle. Director of Research and Training, Newcastle Cognitive and Behaviour Therapies Centre
There have been significant advances in the understanding and treatment of obsessions and worry over the last decade with the development of increasingly specific models and specific treatment strategies. Worry and obsessions are particularly interesting because their respective disorders, Generalised Anxiety Disorder and Obsessive-Compulsive Disorders, are in fact defined by the very presence of these types of thinking which are naturally targets for treatment. The knowledge acquired in the study of these phenomena has also influenced developments in other fields where intrusive and ruminative thinking is experienced such as insomnia, acute stress, chronic pain, health anxiety, depression, and psychosis. Although the distinction between obsessions and worry is relatively simple at a prototypical level, the clinical reality suggests that there is great deal of overlap. Thus, they may also be better conceptualised as different points on a continuum and may share some common processes but also have some distinct features. This presentation will present some key findings that have contributed to current conceptualisations of obsessions and worry, which in turn have resulted in the development of effective treatments. It will highlight some common and distinct features of obsessions and worry and how they may have implications for treatment. Despite the significant advances in our understanding of these phenomena, there remain a number of paradoxes that we have yet to account for in a satisfactory way. Two of these apparent paradoxes will be addressed in more detail where there are currently gaps in both the theory and in our capacity to investigate them due to limitations in methodology. Finally, some additional areas where our understanding of these processes may be applied in conceptualising and treating other intrusive or ruminative thinking will be suggested.

Active Ingredients in Child and Adolescent Therapy
Philip C. Kendall, Temple University, Philadelphia, USA
The keynote address will provide an overview of several components of psychosocial interventions for children/youth that are considered to be influential in the achievement of positive outcomes. Specifically, therapist, child, and type of treatment factors will be presented and discussed, and examples from clinical and research practice will be provided.

Depressive Decision-Making
Robert Leahy Cornell University Medical College & American Institute for Cognitive Therapy, New York, USA
A major goal of cognitive therapy is to assist the patient in initiating change. Rational models of decision making are offered in therapy, based on "expected utilities", such as costs and benefits of future action. However, individuals often ignore expected utility and base their decisions on other considerations. I shall present two decision models that guide depressive resistance: pessimistic risk aversion based on modern portfolio theory and commitment to sunk costs.
According to modern portfolio theory individuals utilise different assumptions and goals in considering how they will allocate their resources and how much risk they will tolerate. A portfolio theory is the individual's perception of his or her resources, ability to produce future resources, diversification, emphasis on maximisation or minimisation of rewards or costs, potential for regret, hedonic utility for gains and losses, and risk-tolerance. For example, optimistic "rational individuals", who view themselves as having abundant current resources and potential for future earnings with a long duration, would be likely to tolerate greater risk in their investments than would individuals lacking current and future resources or who would view themselves as having a shorter duration of investment. Presumably, the individual who has abundant current and future resources can "absorb" a loss, should it occur, especially if she believes that there is a long duration in which this recovery could occur. Furthermore, greater risk tolerance would be assumed if the individual believed that she had many potential replications of behaviour, for example, many potential "hands" to play in order to win
Modern portfolio theory proposes that individuals will tolerate greater risk if they are more highly diversified, since a loss in one investment may be offset by the performance in other investments. Thus, diversification may be viewed as a way in which costs may be contained or compartmentalised, avoiding the risk of over-generalising a failure. Given the negative cognitive schemata of depressed individuals, we would anticipate that these individuals would view losses as having a spreading activation effect, such that they would trigger other losses, thus adding to risk aversion. Empirical data are reviewed that indicate that depressed individuals resist change based on their pessimistic portfolio assumptions and strategies.
A second model---commitment to sunk costs-is reviewed that proposes that individuals utilise prior losses and commitments as a rationale for future escalation of commitment. Rather than extinguishing their behaviour, these prior losses provide a "rationale" for further action aimed at redeeming bad decisions, avoiding finality of regret, and preventing "waste".

A Regulatory Systems Approach to Anger and Violence
Raymond W. Novaco, University of California, Irvine, USA
Clinicians, researchers, and patients tend to view anger as attributable to immediate circumstances, current thoughts, and sustained beliefs. In contrast, systems-oriented thinking approaches anger as a contextual and dynamic phenomenon in which personal dispositional systems of anger (cognitive, physiological, and behavioural) are embedded in an interdependent network of interpersonal and environmental systems. Recurrent anger and violence are maintained by their functionality, as well as by entrenched schemas, and by the relative absence of inhibitory controls and neutralizing influences that counteract antagonism. The more functional and embedded anger is within a system, the greater is its inertia or resistance to change. For persons having serious anger difficulties, it is engaged by threat perception with considerable automaticity that challenges its regulation.
The treatment of anger, like anger itself, should be understood contextually. Cognitive-behavioural anger treatment augments anger-regulatory mechanisms and seeks to do so through both person-centered and environmental intervention. The social distancing effect of chronic anger and the negative expectations that others form of angry people, including clinical staff in treatment settings, interfere with therapeutic change efforts and can reverse treatment effects. Recent treatment outcome studies with patients having severe anger problems, compounded by other clinical conditions (PTSD, psychoses, and developmental disabilities) have found significant gains associated with anger treatment in both individual and group intervention. New results will be presented, and key ingredients of successful intervention in institutional settings will be discussed. As significant treatment gains have been achieved with highly distressed clients having very complex needs, there is much hope for many other people who struggle to maintain their anger and seek clinical assistance.

Building Resilience: The Development and Prevention of Anxiety Disorders
Ron Rapee, Macquarie University, Sydney, Australia
Anxiety disorders are responsible for a huge cost to society as well as a great deal of personal suffering. Past research has focussed on the formulation of detailed models of maintenance and concomitant treatment programs. Little research has so far been directed at the more difficult question of causation. But understanding cause may help to elucidate ways of preventing the development of anxiety disorders from a young age. Over the past few years, we have been focussing on several factors of possible importance in the development of anxiety. These include a combination of a withdrawn temperament, overprotective parenting, and modelling of anxious behaviour from parents. This talk will describe some of our data and integrate them with data from other laboratories to describe a model of the development of anxiety disorders that may be amenable to change in prevention programs. I will then describe a program that we have been working on for the past few years to prevent the development of anxiety disorders. The program began with 3-4 year old preschool children who are at risk for anxiety disorders through their high scores on behaviourally inhibited temperament. The primary focus of the program is to provide education for their parents to help reduce inhibition and withdrawal. In the long term, we hope that this will also alter their risk for anxiety disorders. In this talk, I will present data from our 12-month follow-up that provides the first evidence that temperament can be altered with a relatively brief parent education intervention.

Cognitive Models of Bipolar Disorder: Theory and Therapy
Jan Scott, University of Glasgow, Scotland
Cognitive therapy is widely established as an alternative to medication for individuals with acute major depression. However, arguments prevail about the cost versus benefit of this approach as compared to treatment with antidepressants. This view is less valid when considering the problem of 'treatment-resistant' depression. Previous studies indicate that clients with persistent symptoms of depression following antidepressant treatment are common, experience considerable morbidity and have very high rates of relapse. There is evidence that cognitive therapy may reduce relapse rates in depression, but limited evidence about which clients with chronic depression will most benefit from cognitive therapy.
To explore these issues, we undertook a controlled of 158 subjects with recent major depression, partially remitted with antidepressant treatment (mean doses approximately equivalent to 185 mg amitriptyline or 33 mg fluoxetine), but with residual symptoms of 2 - 18 months duration. Subjects were randomised to receive clinical management alone, or clinical management plus cognitive therapy for 16 sessions over 20 weeks, with two subsequent booster sessions. Subjects were assessed regularly throughout the 20 weeks treatment and for a further year. They remained on continuation and maintenance antidepressants at the same dose throughout.
Recovery rates in those treated with additional cognitive therapy were more than twice that of individuals receiving clinical management. Cognitive therapy also reduced relapse rates for acute major depression and persistent severe residual symptoms, in both intention to treat and per protocol treated samples. The cumulative relapse rate at 68 weeks was reduced significantly from 47% in the clinical management control group to 29% with CT (hazard ratio 0.54, CI 0.32-0.93 intention to treat analysis). Cognitive therapy also reduced specific psychological and social symptoms such as guilt, pessimism and interpersonal dependency. We found evidence that CT may mediate its relapse prevention effects through changes in thinking style rather than thinking content.
The paper also showed that CT for residual depression is more clinically effective (using evidence based medicine calculations of numbers needed to treat) than treatments for physical disorders with equivalent levels of disability. Using cost incremental analysis we calculated that adding CT significantly reduces use of other services, but that CT is not a substitute for all components of treatment as usual. We calculated that it costs about 4000 pounds to avoid a relapse with this approach, or about 12 pounds per depression free day. The issue now is whether decision makers will view this as value for money.
In this difficult to treat group of clients with medication-resistant, residual depression, cognitive therapy produced worthwhile relapse reduction. This paper will review these results in light of other recent studies and will also address the health economic issues offering an estimate of the cost of relapse avoidance.

SYMPOSIA ABSTRACTS

SYMPOSIUM 1

Understanding and Targeting Prognostic Factors in Cognitive Behaviour Therapy for Schizophrenia
Andrew Gumley, University of Stirling

Cognitive Approaches to the Treatment of Individuals at High Risk of Developing Schizophrenia
Paul French, Psychology Services, Mental Health Services of Salford, Manchester.
Primary prevention of psychosis has previously been viewed as unattainable. Currently interventions are aimed towards secondary prevention strategies through minimising further episodes and residual symptoms. However, current research indicates the possibility of identifying initial prodromal symptoms and, therefore, the possibility of primary prevention. A number of primary prevention teams have been established around the world with interventions geared towards medication. This has problems in that there are still false positives who will be treated with neuroleptic medication and be exposed to the side effects of these medications. The answer is to employ psychological interventions with no side effects that target the distress the individual is experiencing. During the symposium early detection strategies will be discussed and a psychological model describing the onset of psychosis will be presented.

Cognitive Behavioural Psychotherapy for Psychosis: From Formulation to Treatment
David Fowler, University of East Anglia
Recent randomised controlled trials carried out in the UK suggest that there is now strong evidence for the efficacy of cognitive behavioural therapy in the management of individuals with psychosis. At the core of this approach is the focus on the subjective experience of individuals with psychosis, in particular, the experience of voices, paranoia, delusions and depression. The starting point for cognitive therapy is the formulation and conceptualisation of this experience in order to make sense of the individual's experience of psychosis. This paper will describe an approach to cognitive formulation, which integrates recent research findings on cognition and emotional processing, and trauma experience of individuals with psychosis. In particular the paper will highlight the potential role of vulnerability to psychological disorder and distress in influencing the course and experience of psychosis.

Cognitive Behavioural Therapy Targeted During Early Relapse in Schizophrenia: The Results of a Randomised Controlled Trial
Andrew Gumley and the West of Scotland Early Intervention Trial Research Group, University of Stirling & Ayrshire and Arran Primary Care Trust
Relapse in schizophrenia is associated with increased probability of future relapse, increased residual symptoms and deteriorating social functioning. There is evidence that targeting increases in anti-psychotic medication during early relapse is effective in preventing relapse and re-hospitalisation. However, this approach to prevention has a number of disadvantages. Continuing questions concerning the sensitivity and specificity of early signs to relapse mean that increases in medication may be unnecessary, and indeed this treatment approach may not be acceptable to individuals due to increased side-effects. Therefore targeting cognitive therapy on the prevention of relapse provide a meaningful alternative treatment approach to relapse prevention. The West of Scotland Early Intervention Trial recruited a total of 144 individuals with a diagnosis of a schizophrenia spectrum disorder who were considered by the assessing clinician as relapse prone. Participants were randomised to either Treatment as Usual (n = 72) alone or in combination with Cognitive Therapy (n = 72). Participants were assessed at entry, 12 weeks, 26 weeks, and 52 weeks. Cognitive therapy was delivered as a two-stage intervention; a five-session engagement/ formulation phase between entry and 12 weeks, and a targeted cognitive therapy phase delivered on the appearance of early signs indicative of relapse. This targeted phase was available to participants throughout the follow-up period. Outcome was assessed according to (1) relapse rate, (2) remission at 52 weeks and 18 months, and (3) social functioning. Predictors of outcome were examined using Logistical Regression Analysis. The paper will present outcome results for the trial, alongside findings on predictors of outcome. Consideration will be given to future research into the delineation and targeting of psychological factors involved in the evolution of relapse in schizophrenia.

Improving Insight in Schizophrenia
Douglas Turkington, Jeremy Pelton and the Insight into Schizophrenia Research Group, University of Newcastle
353 patients with schizophrenia according to ICD 10 research criteria were randomised on a 2:1 basis to receive the Insight programme or treatment as usual. The Insight Programme consisted of 6 sessions of CBT for the patient along with psycho-educational material targeted on insight improvement. The sessions focused on engaging, developing explanations, case formulation, symptom management, adherence work, belief change and relapse prevention. The main carer received 3 sessions focused on stress management, formulation, helping with homework and relapse prevention. Insight was successfully improved by end of therapy in the Insight Programme group but there was evidence that improving insight does need the support of a therapeutic relationship and much in the way of explanation or depression and increased suicidal ideation can occur. These results are fully explored and their pertinence to working with psychotic patients explained.

Characteristics of Good and Poor Responders in Tayside-Fife Trial of CBT for Chronic Psychotic Symptoms
Rob Durham & R. Victor Morton, Department of Psychiatry, University of Dundee
The Tayside / Fife trial of cognitive-behavioural therapy for medication-resistant psychotic symptoms was completed in December 2000. Of 274 patients referred to the trial 65 were suitable and consented to participate. These patients were randomly allocated to one of three treatment conditions of 9 months duration: CBT delivered by clinical nurse specialists as an adjunct to routine psychiatric care (n=22), supportive psychotherapy delivered by volunteer members of the clinical team (mainly nurses) as an adjunct to routine psychiatric care (n=23), and routine psychiatric care alone (n=21). Four aspects of the trial methodology promoted a rigorous evaluation of the clinical effectiveness of CBT in routine clinical practice: (1) a three month pre-treatment baseline to assess stability of presenting symptomatology, (2) outcome evaluation by independent assessors blind to treatment allocation at post-treatment and 3 month follow-up, (3) supportive psychotherapy delivered by non-CBT trained therapists supervised by a consultant psychotherapist, and (4) a treatment as usual control group. Patients were recruited from Dundee, Perth and rural areas of Fife and Angus. Main outcome measures were the Positive and Negative Syndrome Scale (PANSS), the Psychotic Symptom Rating Scales (PSYRATS) and the Brief Symptom Inventory. In addition, the quality of the therapeutic relationship was assessed using the Penn Helping Alliance Scales. This paper explores the diversity of responses to psychological intervention by comparing the characteristics of the ten patients with the best and worst outcomes in terms of demographics, insight, clinical state, social functioning, and quality of therapeutic alliance and symptomatic change.

SYMPOSIUM 2

Advances in the Understanding and Treatment of Body Image Disturbance
Roz Shafran, University of Oxford

Predictors of Eating Disorder Scores in Children
Rick M. Gardner, University of Colorado, Denver, USA
A longitudinal study which identified variables that predict higher eating disorder scores in a non-clinical sample of boys and girls aged six through fourteen will be described. A TV-video procedure was used in conjunction with advanced psychophysical techniques to measure the perceptual and affective components of body image. Predictor variables examined include body sizejudgments as well as demographic, familial, sociocultural, esteem, and clinical variables.

A New Method of Assessing Body Image Disturbance
Roz Shafran, Christopher G. Fairburn, Zafra Cooper , Oxford University Department of Psychiatry
Body image disturbance has been proposed as a key mechanism that contributes to the persistence of dietary restriction in patients with anorexia nervosa. Despite the abundance of research in this area, body image disturbance remains ill-defined and poorly understood; consequently, we suggest that the existing methods of measuring body image disturbance may be flawed and limited.
In order to test the hypotheses relating to body image disturbance that derive from our new theoretical model, it became necessary to devise a new method of measurement. We suggest that this new method represents a significant advance over existing methods for at least two reasons. First, it uses an ecologically valid stimulus for the assessment of body image i.e., a mirror. Second, it separates the perceptual component of body image disturbance from the memory of body image by asking the participant to estimate her body size whilst simultaneously looking at her reflection in the mirror.
Results from an initial study on body size estimation with women who have an eating disorder and non-clinical comparison women will be presented. These results show that despite looking at their reflections in the mirror whilst performing this task, women with an eating disorder overestimate their body size compared to normal controls. The clinical and research implications of the results of this study and the new method of assessing body image disturbance will be discussed.

Development of the Body Checking Questionnaire: A Self-Report Measure of Body Checking Behaviors
Deborah L. Reas, Brooke L Whisenhunt, Rick Netemeyer & Donald A. Williamson, Department of Psychiatry and Behavioral Sciences, University of Texas Medical School at Houston,
The purpose of this study was to develop a brief self-report inventory that could be used for the assessment of body checking behaviors that are common in eating disorder patients. Using exploratory and confirmatory factor analyses, a 23-item measure called the Body Checking Questionnaire (BCQ) was developed. A variety of body checking behaviors are assessed by the BCQ, including subfactors that measure checking related to overall appearance, checking of specific body parts, and idiosyncratic checking rituals. The BCQ was found to have good test-retest reliability (.94) and the subfactors had good internal consistency (.88, .92, and .83). The measure correlated highly with other measures of negative body image and eating disorders, demonstrating evidence for the convergent validity of the BCQ. Additionally, the BCQ was found to differentiate normal controls and eating disorder patients, as well as non-clinical participants scoring high and low on the Body Shape Questionnaire. Since frequent body checking may reinforce body dissatisfaction in eating disorder patients by directing excessive attention to body shape and size, the BCQ may prove to be a useful clinical tool for the assessment and treatment eating disorder patients.

Aesthetic Sensitivity in Body Dysmorphic Disorder
David Veale. Royal Free Hospital & University College Medical School, London & The Priory Hospital North London
Individuals with Body Dysmorphic Disorder (BDD) have an extremely distorted body image. BDD is also associated with idealised values about the importance of appearance to the self and increased aesthetic sensitivity. Evidence will be presented from two studies: (1) The occupation and higher education or training was extracted from the casenotes of 100 consecutive patients with BDD and compared to 100 patients with a Major Depressive Episode, 100 patients with Obsessive Compulsive Disorder (OCD) and 100 patients with Post-Traumatic Stress Disorder (PTSD). 20% of the BDD patients had an occupation or education in art or design compared to 4% in the depressed group, 3% in the OCD group and 0% in the PTSD group, which was highly significant. (2) We have tested self-discrepancy theory and found that BDD patients have a high degree of self-discrepancy between how they see their appearance and how they would like to be in an ideal world. There was no discrepancy between how they saw their appearance and how they thought others saw them or would like them to look. This suggests that BDD patients are not like patients with bulimia or social phobia and are mainly driven by an internal aesthetic standard, which they fail to achieve. These findings will be discussed with a cognitive behavioural model of BDD and implications for treatment.

SYMPOSIUM 3

CBT Self-help Treatments: The Current Status of Computer-delivered Self-help Materials
Chris Williams, University of Glasgow

Self-help approaches and computer-based treatments are very topical. Several UK-developed treatment packages are available and are currently being evaluated. In spite of claims for high acceptability and the possibility that computer-based treatments will allow ready access to an effective psychosocial intervention, few clinical units currently offer such treatments. This symposium will review current computer packages, the evidence for their effectiveness, and discuss ways that such packages may be seen as part of a wider service delivery of CBT.

Evaluating a computer-based CBT treatment package for bulimia nervosa
U. Schmidt*, Williams, C., Barra-Carril, N., Reid, Y., Harkin, P., Cottrell, D., Treasure, J., Kovacs, D. Palmer, R
*: Senior Lecturer in Psychiatry, Institute of Psychiatry, De Crespigny Park, London.
Background: Bulimia nervosa (BN) is a psychologically and physically disabling condition. The disorder is not self-limiting, and without early treatment the risk of a chronic relapsing course is high. Cognitive behavioural treatment (CBT) is the treatment of choice for bulimia nervosa. Self-help formats of CBT using books have been used with good effect, but many patients find it difficult to motivate themselves to actively engage with working through a book.
Aims: The present study describes the ongoing pilot evaluation of a CD-ROM based CBT package (Williams et al., 1998) in the treatment of outpatients with bulimia nervosa. The interactive nature of this programme and the more individually tailored delivery of information is likely to make it acceptable to a much broader range of sufferers than traditional self-help. Individuals with bulimia nervosa are usually young women, a group with high computer literacy, and are therefore an ideal target group for this type of treatment. Moreover, the shame and secretiveness surrounding bulimic disorders may make computer-based treatment particularly appealing to sufferers as a first line treatment.
Subjects and Method: New out-patients with bulimia nervosa referred to two eating disorder clinics (London, Leicester) were entered into the study. Outcome was measured in terms of bulimic symptomatology, and indicators of psychological and social functioning. Acceptability of the programme was assessed by questionnaire and qualitative interviews.
Results: To date 35 patients have enrolled in the CD-ROM programme. The paper will describe take-up and drop-out rates, and preliminary symptomatic outcomes will also be presented.
Discussion: The clinical and research implications of using this novel approach to treatment will be discussed.

Beating the Blues' Computer CBT Program for Anxiety and Depression: Applications and Outcomes.
Judy Proudfoot*, Institute of Psychiatry, London; Jim Willis*, Invicta Trust, Maidstone:Research Team: David Goldberg, Isaac Marks, Anthony Mann, Jeffrey Gray, David Shapiro, Sharon Swain
'Beating the Blues'TM is a stand-alone computer-controlled, interactive multimedia package providing CBT for anxiety and depression (for details, visit www.Ultrasis.com). Developed at the Institute of Psychiatry with two private-sector companies, the program can be delivered on a personal computer located in the GP surgery or specialist care facility. Clinical supervision and responsibility continue to rest with the GP or other appropriately qualified personnel (nurse or clinical psychologist), to whom reports (including warnings of suicide or other risk) are automatically delivered by the computer program. The program is readily usable by patients with no previous computer experience. Like other versions of CBT, Beating the Blues can be given alone or in combination with pharmacotherapy.
Excerpts of the programme will be demonstrated and preliminary data presented from the various studies in which Beating the Blues is being evaluated.

If the Evidence is So Good, Why Doesn't Anyone Use Them? - Current Uses of Computer-Based Packages.
Graeme Whitfield., Leeds Community and Mental Health Trust. And Chris Williams, University of Glasgow.
Despite evidence that computerised CBT can be as effective as face to face therapy less than 7 per cent of practitioners accredited by the BABCP (British Association of Behavioural and Cognitive Psychotherapists) use them (Keeley and Williams 2001). This low uptake may reflect patient or therapist attitudes. Alternatively it may be the result of NHS commissioning mechanisms that have been noted to be relatively unreceptive to computerised CBT (Shapiro et al 2001). Preliminary results from a recent survey of a randomised sample of 500 BABCP members will be presented which explores practitioner attitudes to and experience of using computerised CBT. It also addresses therapist-identified factors that would need to change before the therapists not currently using computerised CBT would begin to do so. Factors such as the practitioners' current computer literacy, the availability of a secure workplace in which to house computer facilities as well as the availability of IT (Information Technology) support all affect uptake. Specific therapist concerns about computerised CBT such as patient confidentiality and the Data Protection Act are highlighted.
References:
Keeley, H, Williams C, Shapiro, DA (2001) A National Survey of BABCP Accredited Therapists' Attitudes towards and Use of Structured Self-Help Materials, Paper submitted to Behavioural and Cognitive Psychotherapy.
Shapiro, DA, Proudfoot J, Gray J (2001) Computer-Based Cognitive-Behaviour Therapy of Anxiety and Depression. Paper presented at the 'Psychological Therapies in the NHS' conference, Brighton, UK.

Using Computer-based Self-help: A Clinical Perspective.
Chris Williams, University of Glasgow.
Cognitive Behaviour Therapy (CBT) is a proven treatment for anxiety and depression, yet access to specialist therapists is often limited. Computerised self-help packages offer one way of delivering CBT within a clinical service. CBT packages offer essentially an educational form of psychotherapy and this to some extent overlaps with the goals of practitioners working in the arena of health promotion.
A significant literature exists within the field of health promotion concerning how to maximise change in attitudes and behaviour. When applied to CBT, creators of self-help materials need to be aware of the potential pit-falls in creating such materials. Average reading ages within the UK vary according to a roughly normal distribution and have a mean of between ages 9-11 in many studies. A poor reader will quickly become discouraged by texts that are difficult for them to read fluently. This occurs when text is poorly printed, contains complex sentences, long words or too much material containing entirely new ideas (www.hfht.org). This has significant implications for the development of computerised self-help materials.
Many existing CBT computer packages have tended to "put a book" onto computer. This is unhelpful because it takes 28% longer to read than when the user reads the same materials printed out on paper1. Materials should instead be produced with computer delivery in mind, and be accessible, delivered for a specific target audience, be clearly presented, legible, and readable. The Overcoming Depression course brings together key components for change: a clear structure, a focus on current problems and a jargon free content that has been produced in liaison with users. The course is available in different formats - as a book, via the web and shortly as an interactive CD Rom (www.calipso.co.uk). The development of the course and an overview of its content will be described.
References:
Muter, P., Latremouille, S.A., Treurniet, W.C., & Beam, P. (1982). Extended reading of continuous text on television screens. Human Factors, 24, 501-508.

A Systematic Literature Review of Computer-based Outcome Studies.
Martin Neal*, Chris Williams, Ian Cameron, David Cottrell, Helen Clarke, Karina Lovell.
*Leeds Community and Mental Health Services Trust.
This presentation will focus upon the Cochrane systematic review that is being undertaken considering those studies that specifically focus upon computer based self-help materials for the treatment of anxiety and depression. This presentation will identify the work that has been carried out by a multi institutional team focusing specifically upon depression.
The presentation will not only focus upon RCTs and CCTs, but attempt to make sense of the broader literature regarding the use of computer materials in the treatment of depression.
Additionally the presentation will seek to set the context for how computer assisted materials are being considered at both a national and international level.
Consideration of how the review was constructed and the key finding to date will also be discussed.

SYMPOSIUM 4

New Developments in the Understanding of Cognition in Children and Adolescents.
Jonquil Drinkwater, University of Oxford

Cognition in Younger Children: Some Theoretical Observations and Implications.
Derek Bolton, Psychology Department, Institute of Psychiatry, Kings College London
Piagetian and Vygotskian theories of cognitive development include points of direct relevance to cognitive therapy, including that cognition increasingly regulates behaviour (including affect), that in pre-rational mentality symbol can influence reality, and that mature rationality from adolescence on involves various competencies including meta-representation and systematic theory. Notwithstanding paradigm shifts from general stage theories to information-processing models of modular domains, the early developmental theories retain relevance and implications for cognitive therapy with children. The theories imply that basic cognitive therapy methods including cognitive restructuring may apply indefinitely far down the age-range once the child is using language, though these techniques are more likely to engage with what is regulating the child's affect and behaviour if they focus on the child's own spontaneously produced verbally encoded meanings. It is unclear how far regulation of affect and behaviour by language can be promoted or accelerated by telling the child what they are probably thinking, or what it would be helpful to think instead. Another implication is that those cognitive therapy techniques which focus on explicit systematic theory, core beliefs etc., are probably less applicable prior to mid-adolescence, but this belongs with the fact that childhood presenting problems typically do not involve explicit theory of this kind. Meta-representation can be found in children, such as feeling bad about having such-and-such thoughts or inclinations. Pre-teenagers can also have beliefs about the effects of cognition on reality, as for example in magical thinking. Again the implication is that cognitive therapy methods are likely to be applicable to children, providing they engage with the child's own cognitive processing.

Emotional Disclosures in School Children
Martina Reynolds, Department of Addictive Behaviour & Psychological Medicine, St. George's Hospital Medical School, London
Recent research with adults by Pennebaker and his colleagues has found that emotional disclosure through writing about stressful events appears to have significant benefits in terms of psychological and physical health outcomes. This report describes a controlled trial of emotional disclosure, adapted for schoolchildren, with the major hypothesis that the repeated description of negative events will have beneficial effects on measures of mental health, attendance, and school performance. The sample consisted of children aged 8-13 years from 4 schools, a primary and a secondary school from both a suburban and an inner-city area. Children were randomly assigned to 1 of 3 conditions, writing about negative events, writing about non-emotional events, and a
non-writing control group. Children in all groups were seen 4 times during a single week and were then followed up after 2 months with measures of health and school performance. The intervention was well received by both schools and children, and the scripts written by the emotional and non-emotional writing groups differed in content in the predicted ways. Contrary to expectation, here was little evidence of a specific effect of emotional disclosure, and several possible reasons for this are discussed. Nevertheless, there was a general reduction in symptom measures, indicating that children may have benefited from their involvement in the study. Although there are several possible explanations for our findings, they indicate that it is both feasible and potentially valuable to give children opportunities to engage in discussion about sources of stress and their reactions to them.

Beliefs and Cognitive Avoidance Associated with Worry in Teenagers:
Mark Freeston, Newcastle Cognitive and Behavioural Therapies Centre and
University of Newcastle
Current models propose that different beliefs about worry and cognitive avoidance contribute to the maintenance of worry. This study examined the relationship between worry intensity, beliefs, and cognitive avoidance among a large sample of high-school students aged 12-17. A wide range of worry intensity was reported, with surprisingly large numbers reporting levels that approach clinical levels. There were also age and gender related differences in worry themes. As predicted there was an association between the degree of worry and a range of beliefs. High worriers more strongly endorsed the notion that worry was useful than did moderate worriers. In particular there was significant interaction between two groups of beliefs. Although both high and moderate worriers believed that worry helped solve problems, high worriers reported a relatively greater belief that worry could help prevent the worst from happening. Likewise, a wide range of cognitive avoidance strategies was endorsed with worriers reporting greater avoidance. Once again an interaction
was observed for the perceived efficacy of the strategies. High worriers reported greater efficacy in the short term and lower efficacy in the long term than did the moderate worriers. The results support current models of worry, in particular the strategic and rule-governed nature of the worry process. In particular, this study extends findings to an adolescent age group, suggesting the processes are already in place at an early age. Implications for prevention and treatment are briefly discussed.

The Relationship Between Automatic Thoughts and Negative Emotions in Children and Adolescents
Ronald M. Rapee & Carolyn A. Schniering, Macquarie University, Sydney, Australia.
Cognitive theories of emotional disorders point to links between specific categories of beliefs and associated symptoms. For example, threat-related thoughts are said to be associated with anxious symptomatology while loss-related thoughts are said to be related to depressive symptoms. There has been some support for this argument in adults. However, in children, support for the argument has come from small studies examining individual types of thoughts or symptoms. The two main aims of the present studies were to examine the factor structure of a wide range of negative beliefs in children and adolescents and to examine the links between these empirically derived factors and symptom clusters. In study 1, 978 young people aged 7 to 16 years completed a child-generated measure of their experience of negative thoughts. Structural equation modeling provided the strongest support for a model in which four distinct factors were all related to a single higher-order factor. The four lower-order factors related to thoughts of social threat, physical threat, personal failure, and hostility. In study 2, 790 young people completed the same measure of negative thoughts as well as a measure of experienced symptoms. The four cognitive factors showed relatively specific associations with expected symptom clusters - i.e. social threat with social anxiety symptoms, physical threat with physical anxiety symptoms, personal failure with depressive symptoms, and hostility with oppositional symptoms. Similarly, the four cognitive factors distinguished relatively well between young people's principal diagnoses. These results provide support for cognitive specificity models of emotional problems in young people.

SYMPOSIUM 5

Issues of Coping and Vulnerability
BABCP Scientific Committee

Is Dissociation the Ultimate Form of Avoidant Coping? Dissociation, Self-efficacy, and Coping in Borderline Personality Disorder (BPD)
Dale Huey, Newcastle Cognitive & Behavioural Therapies Centre and University of Newcastle; Peter Whewell and Jonathan Espie, Regional Department of Psychotherapy, Newcastle; Matthew Philpott, University of Newcastle, England.
Dissociative phenomena co-occur with numerous forms of psychopathology and are particularly prevalent in BPD (Zweigfrank et al., 1994). The cluster of symptoms that we call BPD has been conceptualized as being the result of a cluster of beliefs about oneself and others, e.g. seeing oneself as vulnerable and inherently unacceptable and seeing others as dangerous (Arntz, 1994). As Janet originally formulated the phenomenon of dissociation as a response to overwhelming trauma (Gershuny & Thayer, 1999) in BPD dissociating, or disengaging from one's surroundings, as a response to threat is predictable given the double disadvantage of perceiving oneself to be powerless in a perceived-to-be malignant world. We have recently tested this conceptualization by observing the relationships between tendency to dissociate, perceived self-efficacy, and avoidant coping. Data will be reported from a sample of 105 participants who satisfy criteria for BPD. The role of self-reported childhood trauma and current level of psychological distress will also be discussed. Conceptualizing dissociation as a response to stressors which an individual implicitly perceives to be beyond their ability to manage demystifies and translates an involuntary response into an active, albeit currently unhelpful, strategy, i.e. it provides a plausible and benign conceptualization which has good potential treatment utility.
Arntz, A. (1994) Treatment of BPD: A challenge for CBT. Behaviour Research & Therapy, 32(4), 419-430.
Gershuny, B.S. & Thayer, J.F. (1999) Relations among psychological trauma, dissociative phenomena, and trauma-related distress. Clinical Psychological Review, 19(5), 631-657.
Zweigfrank et al., (1994) Dissociation in … patients with borderline and non-borderline personality disorders. Journal of Personality Disorders, 8(3), 203-218.

Defensive Coping and Underlying Self-esteem in Chronic Fatigue Syndrome
Cathy Creswell* & Trudie Chalder**
*Sub-department of Clinical Health Psychology, University College London
** Department of Psychological Medicine, Kings College Hospital
The cognitive behavioural model of Chronic Fatigue Syndrome (CFS) (Sharpe, 1997) proposes that low self-esteem is prevalent amongst people with CFS yet is protected by rigid coping mechanisms. This is the first study to investigate the prevalence of the Defensive High Anxious coping style amongst people with CFS and the potential impact of this mechanism on self-esteem.
The study comprised 68 participants (24 CFS; 24 Healthy volunteers; 20 chronic illness volunteers). Participants completed the Bendig short form of the Taylor Manifest Anxiety Scale and the Marlowe-Crowne Social Desirability Scale, a Self-Statements Questionnaire and an Emotional Stroop Test. A greater number of participants in the CFS group were classified as Defensive High Anxious compared to the two comparison groups (c²(2)=8.84, p=0.012). Participants with CFS reported lower self-esteem than the two comparison groups on overt (c²(2)=13.44, p<.0001) and covert measures (Emotional Stroop) (F(2,64)=8.75, p<0.001). When overt levels of self-esteem and self-reported depression were controlled for the group differences found for covert levels of self-esteem continued to approach a significant level (F(2,62)=2.97, p=0.059).
This study supports the existence of defensive coping mechanisms amongst participants with CFS. In contrast to previous studies, participants with CFS overtly reported lower levels of self-esteem than two comparison groups. Overt reports of low self-esteem did not, however, fully reflect their underlying levels of self-esteem. This discrepancy might result from the application of rigidly held defence mechanisms. These findings highlight the need to address low self esteem and defensive coping in cognitive behavioural therapy.

An Investigation into Cognitive, Emotional and Psychosocial Factors Influencing Vulnerability and Recovery in Bipolar Disorders
Matthias Schwannauer, Charlotte Brodie, Mick Power., Bipolar Disorder Service; Department of Psychiatry, University of Edinburgh.
The current study aims to investigate the efficacy and effectiveness of a psychosocial intervention for people with a diagnosis of bipolar disorder. This particular psychosocial intervention has been developed for the purpose of this trial and includes elements of cognitive therapy and interpersonal therapy in both a group and individual format. This study further aimed to develop a multi-factorial model of aetiology and treatment of bipolar disorders that takes account of mood specific changes in the perception and function of various psychological factors.
It is our intention in this study to move away from a structural and solely epidemiological understanding of psychosocial, cognitive and emotional risk factors, towards a more process-oriented model of perception and cognitive processing of these factors during varying stages of the disorder, and to investigate the connection of these processes with the development and course of the illness.
In this study we applied a partially randomised design in which patients were randomised into waiting list control or treatment group. In the treatment group, patients had the choice of either group or individual treatment. All patients were assessed at intake, mid-treatment, end-treatment and at 5 month follow-up. Psychosocial measures included quality of life, life events, social support and expressed emotion; psychological measures included self esteem, basic emotions, beliefs about illness, and meta-cognitions and relevant symptom measures included measures of depression, mania, and mixed episodes. In addition to these self-report measures, all subjects were assessed utilising observer rated measures of severity and chronicity. Individuals' perception of the therapeutic process was assessed at follow-up.

Circadian Rhythms, Multilevel Models of Emotion and Bipolar Disorder: An Initial Step Towards Integration?
Steven.H. Jones, Honorary Senior Research Fellow, Department of Psychology, University of Manchester and Consultant Clinical Psychologist, Birch Hill Hospital, Rochdale
Possible links between disruption of circadian rhythms in bipolar disorder and the affective symptoms which are experienced in this disorder are described. Evidence is drawn from Healy and Williams' (1989) review of circadian function in manic depression, along with later reports, which indicate a role for disrupted circadian rhythms in both depressed and manic phases of manic depression (bipolar disorder). This is integrated within a version of the multilevel model of emotion proposed by Power and Dalgleish (1997,1999). The aim of this process is to propose a possible psychological mechanism by which the disruption of circadian rhythms might result in the observed clinical symptoms of bipolar disorder. The integration of these approaches leads to a number of specific testable hypotheses which are relevant to future research into the psychological treatment and understanding of bipolar disorder.

The Relationship Between Schemas, Depression and Psychosis: An Analysis
Neshika Samarasekera, Steve Moorhead Douglas Turkington, Department of Psychiatry, University of Newcastle-Upon-Tyne.Newcastle and London Wellcome Research Groups, Project no. 039243.
Background: There is considerable evidence of a relationship between depressive and psychotic symptoms at all stages of schizophrenia, but its nature remains unclear. Cognitive therapy theory states that the affective and behavioural response to an event is determined by its appraisal. This, in turn, is mediated by underlying beliefs (schemas). Controlled trials have demonstrated the efficacy of cognitive-behavioural therapy (CBT) in schizophrenia, but there has been little systematic research into the types of beliefs present or their association with depressive or specific psychotic symptoms. We wished to survey schematic vulnerabilities in schizophrenia and determine their role, if any, in the relationship between psychotic and depressive symptoms.
Method: Schemas and symptoms were assessed in twenty-six patients with a diagnosis of schizophrenia. These were a subgroup of patients being followed up from a randomised controlled trial of adjunctive CBT.
Schemas were measured using the Burns Dysfunctional Attitudes Scale (DAS). This DAS comprises 35 statements of beliefs within six schemas that are hypothetically important in depression. These include love (need for love), approval (need for approval) and achievement (need for achievement). Overall symptoms were measured using the Comprehensive Psychopathological Rating Scale (CPRS) by a rater who was blind to the use of the data and the DAS ratings. Two subscales derived from the CPRS: the Schizophrenia Change Scale and the Montgomery-Asberg Depression Rating Scale were used to determine the extent of psychotic and depressive symptoms respectively.
Statistical analysis was made with Pearson's product-moment correlation coefficient.
Results: The majority of patients (88%) scored in the dysfunctional range of at least one schema. As shown in other diagnostic groups, higher DAS scores (indicating greater schematic vulnerability) correlated with a greater degree of depressive symptoms (p<0.001). Moreover, total DAS scores were significantly correlated with psychotic symptoms (p<0.001).
Partial correlation demonstrated a pivotal role for DAS scores in the relationship between psychotic and depressive symptoms in this group. Certain schemas were associated with specific psychotic experiences; for example, scores on the Love schema were correlated with scores on thoughts of persecution (p<0.01) and commenting voices (p=0.02). Various models relating psychotic phenomena and schemas to depression were tested. Control psychotic experiences (p=0.02) and an approval schema (p<0.01) made independent contributions to depression in a regression analysis. The results will be presented in further detail.
Conclusions: The connections between schemas and psychotic experiences shown in this preliminary study support the psychological meaningfulness of psychotic and depressive symptoms in schizophrenia. Although it is a small study, these findings are consistent with earlier literature which emphasises that psychotic beliefs are on a continuum with non-pathological beliefs. It may be possible to predict particular schemas on the basis of psychotic symptoms and consider targeting interventions for brief work. Limitations and implications of the study will be discussed.

SYMPOSIUM 6

Psychological Processes and Psychotic Symptoms
Craig Steel, Institute of Psychiatry, London.

The last decade has witnessed major theoretical developments within the understanding of psychological processes associated with the development and maintenance of psychotic symptoms. These advances have highlighted the roles of attention, perception, reasoning, attribution biases, theory of mind, metacognition and 'safety behaviours'. Recently there have been increased efforts at incorporating the increased understanding of these processes within integrated models. Several models will be presented within the symposia which have individual foci on (i) inhibitory mechanisms underlying schizophrenic symptomatology (ii) persecutory delusions (iii) a cognitive model of positive symptoms emphasising the role of maintenance factors such as safety behaviours and (iv) a cognitive model outlining processes argued to underlie symptoms of depression, mania and positive symptoms.

Inhibitory Processes Underlying Schizophrenic Symptomatology
Craig Steel, Institute of Psychiatry, London
Recent influential theoretical accounts of schizophrenic symptomatology have referred to a core dysfunction within inhibitory mechanisms or 'cognitive inhibition'. This account states that a failure to integrate previously occurring information with its current context underlie the heterogeneous symptoms associated with a diagnosis of schizophrenia. It is argued that the original use of the term 'cognitive inhibition' lacks a clear theoretical basis. Recent studies will be discussed which contribute to a clarification of the role of specific types of inhibitory processing in relation to specific symptomatology. A multi-dimensional model is proposed in which reduced levels of associative learning underlie reality-disordered symptoms (i.e. hallucinations and delusions), whilst a failure to inhibit the spread of activation of information is associated with disorganisation symptoms.

A Cognitive Model of Persecutory Delusions
Daniel Freeman, Institute of Psychiatry, London
A new multi-factorial model of the formation and maintenance of persecutory delusions is presented. Persecutory delusions are conceptualised as threat beliefs. The beliefs are hypothesised to arise from a search for meaning for unusual experiences; the explanations formed reflect an interaction between psychotic processes, pre-existing beliefs and personality (particularly emotion), and the environment. It is proposed that the delusion is maintained by processes that lead to the receipt of confirmatory evidence and processes that prevent the full processing of disconfirmatory evidence. Novel features of the model include the direct roles given to emotion in delusion formation, the consideration of the content and form of delusions, and the hypotheses concerning the associated emotional distress. A number of relevant research studies that have been carried out by the presenter are discussed, and the clinical implications of the model outlined.

A Cognitive Approach to Auditory Hallucinations: From Theory to Therapy
Tony Morrison, University of Manchester
In this paper, a cognitive approach to the understanding of psychotic symptoms that focuses on the interpretation of intrusions into awareness is outlined. It is argued that many positive psychotic symptoms (such as hallucinations and delusions) can be conceptualised as intrusions into awareness or culturally unacceptable interpretations of such intrusions, and that it is the interpretation of these intrusions that causes the associated distress and disability. It is also argued that the nature of these interpretations is affected by faulty self and social knowledge (including metacognition) and that both the intrusions and their interpretations are maintained by mood, physiology, and cognitive and behavioural responses (including selective attention, safety behaviours, and counterproductive thought control strategies). Experimental evidence from several studies of cognitive processes in psychotic patients that test specific predictions of this cognitive approach will be summarised. The clinical implications of this approach will be discussed.

SYMPOSIUM 7

Family Care in Psychoses
Liz Kuipers, Institute of Psychiatry, London

Caring for Carers of People with Psychosis
Liz Kuipers, Institute of Psychiatry, London

Carers of those with psychosis are most typically older age mothers, although other groups such as partners, siblings and children can also be involved
We know that such carers have needs of their own. They are more likely to be distressed, anxious or depressed than the general population, to ask for information, respite, practical and emotional support. The impact of care begins at first episode and may be lifelong.
Standard 6 of the National Service Framework requires that carers needs are assessed. The more difficult problem is to design interventions that begin to meet such needs.
Research projects that have been offering interventions in early episodes and beyond will be presented. The ongoing difficulties of designing and implementing services that reduce the impact of care in psychosis will be discussed.

A Randomised Controlled Trial of a Carers' Support Programme
G. Szmukler, E. Kuipers, J Joyce, T. Harris, M. Leese, W. Maphosa, E. Staples, and M. CunninghamHealth Services Research Department, Institute of Psychiatry, Kings College London
Background Despite an acknowledgement of the impact of serious mental disorders on informal caregivers, we still know little about how to best help them.Aims To evaluate the effectiveness of a carers' intervention of 'intermediate' intensity, that is, one lying between brief educational programmes and long-term family psychoeducational treatments.
Method A pragmatic randomised controlled trial comparing the experimental support programme with 'standard' care. All carers of patients with a psychotic disorder from a defined population were approached. Outcome measures were based on a 'stress-appraisal-coping' model of caregiving.
Results Despite concerted attempts to engage carers, only 40% participated in the study. The carers' programme did not offer any significant advantage on any of the primary outcome measures: psychological morbidity, negative appraisal, coping or social support. The severity of caregiving difficulties decreased over the study period for the group as a whole.
Conclusions There is still uncertainty about the most effective interventions for carers. Meeting 'needs' may not improve caregiver distress.

Family Work: Perspectives of Consumers and Providers
Frances Gere,
No abstract available

Family Work in Early Onset Psychosis: Perspectives from Specialist and Generic Services
Frank Holloway, South London and Maudsley NHS Trust and Health Services Research Department, Institute of Psychiatry, Bethlem Royal Hospital
Traditionally psychiatric services have been preoccupied with the needs of people with established and severe disabilities. A controlled trial of intensive treatment for people with "early" psychosis, defined as within five years of initial presentation, is underway within the Croydon Mental Health Services. This paper compares and contrasts the experiences of practitioners working in the specialist treatment service for "early" psychosis, the COAST Team, and staff working in one of the Community Mental Health Teams (CMHT) serving Croydon. Both teams consist of a full range of professionals.
COAST team members work with a clearly identified patient group, have low case-loads (currently eight clients per worker) and have dedicated weekly family supervision from an experienced therapist. Within the comparison CMHT staff work with the full range of people with mental health problems, have high case-loads (currently fifty clients per Community Psychiatric Nurse) and have minimal access to supervision about family work.
The majority of staff in both COAST and the comparison CMHT lack formal training in family management of psychosis and experience uncertainties in working with families. Objectively COAST staff and CMHT members differ in their capacity to undertake family work. In general staff find working with families difficult. Practitioners identify particular problems in working with "early onset" families because of the uncertainties associated with diagnosis and prognosis.

Family Intervention for Schizophrenia and Comorbid Substance Misuse
Christine Barrowclough, University of Manchester

SYMPOSIUM 8

Sexual Dysfunction
Padmal de Silva, Institute of Psychiatry, London

The Challenge of Physical Treatments in the Management of Male Erectile
Disorder
Kevan Wylie, University of Sheffield
Sexual problems can occur as a consequence of disruption of any of the four phases of the sexual response cycle.The therapist needs to obtain a detailed description of the current sexual status in an attempt to identify and understand psychological issues which may be either causing or contributing to the sexual disorder. These include: family-of-origin, behavioural , cognitive and/or systemic issues.In addition, around two-thirds of patients also have some physiological factors leading to the impairment of sexual function.Assessment of these factors is important in trying to offer a prognosis for response to specific interventions. Despite this, patient awareness and expectations of medical -including pharmacological - solutions to the problem can affect any resolution of the dysfunction.A review of currently available physical treatments, including Viagra, is given. This is followed by a consideration of potential psychological manoeuvres that may enhance overall response to clinical interventions.

Sexual Dysfunction in Patients with Eating Disorders
Gill Todd, Gerald Russell Eating Disorders Unit, Bethlem Royal Hospital, Beckenham, Kent
This paper discusses sexual difficulties in patients with eating disorders, especially anorexia nervosa. The focus is on female patients. The theoretical views on sexuality in anorexia nervosa is briefly reviewed. The sexual difficulties that these patients present with are then considered, with the help of case examples. Issues in the assessment and treatment of these difficulties are then discussed. The need for a sensitive and individually tailored approach in the treatment of these problems is highlighted. Timing of therapy, partner involvement, dealing with past memories, and body image issues are some of the problems considered.

The Role of Paraphilias in Sexual Dysfunction
Padmal de Silva, Institute of Psychiatry, King's College, London
This paper considers the role of paraphilias in sexual dysfunction . While paraphilias are usually considered as a separate category of clinical problme from sexual dysfunctions, there is overlap, and one domain can influence the other. With the help of case examples, the way in which paraphilias (e.g. fetishism, transvestism, masochism) can have an impact on sexual functioning, leading in some cases tp frank sexual dysfunction, is discussed. Issues in assessment and treatment are also considered. One aspect of treatment discussed is partner involvement.

Training Issues in Sex Therapy
Mary Griffin, Maudsley Hospital, London; and Padmal de Silva, Institute of Psychiatry, London
There are several programmes in the UK at present for the training of sex therapists. There is an established accreditation procedure, and regular monitoring of courses. This paper considers some of the key issues in sex therapy training.Some of the issues have been controversial, and there has been some debate in the literature on these matters. The topics considered include: the feasibility and advisability of offering training in sex therapy to those who are not in the mental health/selping professions; the need to be knowledgeable about medication and other physical treatments; theoretical orientations; need for training in research; evaluation of training; ethical and professional issues.Current thinking on these issues are reviewed, and suggestions made for discussion.

SYMPOSIUM 9

Dissemination of Effective Interventions
Christine Barrowclough, University of Manchester

Towards the Dissemination of CBT for Bulimia Nervosa
Christopher G. Fairburn, Department of Psychiatry, University of Oxford
A cognitive behavioural theory of the maintenance of bulimia nervosa was proposed in the early 1980's together with a cognitive behavioural treatment (CBT) derived from it. Since then, the theory has been supported by a variety of lines of evidence and the treatment has been shown to be the most effective treatment for the disorder.
This has stimulated interest in how to disseminate this specialised psychological treatment. The findings of various small-scale studies suggest that it can be simplified and abbreviated without great loss of potency. Importantly, it has also been shown that it can be converted from a "therapist-led" treatment into a "programme-led" format suitable for delivery in non-specialist settings or, indeed, as a form of pure self-help. This has led to "effectiveness" research (Carter and Fairburn, 1998) and the proposal that a stepped care approach to management be adopted (see Wilson et al, 2000). Such an approach is currently the subject of a large multicentre trial.

Psychosocial Interventions for Psychosis: Promoting Clinical Change in the Real World
Jo Smith, Worcestershire Community and Mental Health NHS Trust
Considerable research has demonstrated the effectiveness of psychosocial interventions in improving a number of key outcomes for individuals with psychosis and their families. These interventions include psychoeducation, family intervention, early signs monitoring and individual cognitive behavioural intervention. However, implementation in routine service settings has been more problematic: the number of trained personnel remains limited, many of these interventions are not routinely available beyond these research settings and little work has considered the development and maintenance of psychosocial interventions in ordinary service settings.
While it is appreciated that research findings do not automatically influence routine clinical practice, the task of promoting effective evidence based practice is highly complex. An understanding of the professional, educational, political and economic factors in the real world environment may be crucial if efforts to promote changes in services for individuals with psychosis are to be more than marginally successful. This paper will consider a framework which highlights key influencing factors and identifies potential change strategies for successful implementation. The paper will draw on the limited research evidence so far available and the author's own experience in trying to promote successful clinical change in relation to implementing psychosocial interventions for in Psychosis in 'the real world'.

The Politics of Training
Jan Scott, University Department of Psychological Medicine and Glasgow Institute of Psychosocial Interventions
There is a wealth of research evidence that brief psychological therapies such as cognitive therapy can improve the short term and long term outcome of a wide variety of common and severe mental disorders. However, there are major problems in ensuring that these effective interventions are available in day to day clinical practice. The last 15 years has seen the rise of the post qualification multi-disciplinary training course. However, this paper suggests that the cost versus benefits of these courses are unproven. Too often, individuals leaving these courses fail to continue to use their new skills on return to their previous clinical setting. This may be due to lack of skill (training in using the approach with one group of clients and then returning to work with another), lack of time (no-one adjusts case loads to allow for the additional time required for the early part of therapy), or lack of support (failure to provide ongoing supervision).
This paper will present an alternative model for trying to increase the availability of effective therapies in day to day clinical practice. The Glasgow Institute of Psychosocial Interventions is a training research and clinical multi-disciplinary group with the primary goal of increasing staff skills in delivering brief therapies and in changing organisational systems to allow people to use those skills. The project has been funded through the Scottish Executive Waiting List Initiative which has chosen waiting times of psychological therapies as a primary target for change. Four Trusts in Scotland were invited to participate in this pilot scheme. We also recruited the support of an organisational development group who helped us in gathering baseline data and setting up steering groups to look at gaps in training for staff within the Trust and how to develop effective delivery systems.
The paper concludes that if we are to truly increase the availability of psychological therapies, we need to use our skills to change the thinking within organisations so that training is not an expensive and frustrating waste of time!

Cognitive Behavioural Processes in the Dissemination of Evidence-based Guidelines for Dental Practice
Marie Johnston, Department of Psychology, University of St Andrews
Background: Guidelines are published to facilitate dissemination of evidence-based practice and procedures to promote their implementation have been developed, derived from medical education approaches. Although not developed on cognitive-behavioural principles, it is possible to examine a) which cognitive variables predict evidence-based dental practice and b) whether implementation interventions affect these cognitions and the behaviour.
Procedure: Following postal distribution of guidelines on the management of third molars, 51 dentists were randomly allocated to one of four different methods of disseminating guidelines of the management of third molars i.e. they had one, both or neither of two additional implementation methods: 'audit and feedback' and 'computer-aided learning with decision support'. Before and after these procedures, dentists completed a questionnaire based on social cognition models addressing two specific behaviours pertinent to the new guidelines: 1) the extraction of third molars; and 2) the use of guidelines. Data on actual dental behaviour was obtained from patient records.
Results and Conclusions: Results will be reported on the prediction of behaviour from the social cognitive variables. These results allow comparisons of the importance of beliefs about guidelines versus beliefs about the behaviour per se. The main study will examine how the normally used implementation methods affect critical cognitions and behaviour related to evidence-based practice. From that it should be possible to comment on the adequacy of current procedures and the potential value of cognitive-behavioural procedures in disseminating effective interventions.

The Dissemination of Anxiety Disorder Treatment: What are the Questions?
Paul Salkovskis, Department of Clinical Psychology, Institute of Psychiatry, London

No abstract available

SYMPOSIUM 10

Cognition And Emotion In Children And Adolescents
Rebecca Park, Department of Psychiatry, University of Cambridge, U.K.

Specificity of Autobiographical Memory and Mood Disturbance in Adolescents.
Swales, M, & Brennan, A.: School of Psychology, University of Wales, Bangor ; Bangor & Conwy & Denbighshire NHS Trust
Difficulty in retrieving specific autobiographical memories to cue-words has been found to be associated with a number of psychiatric disorders; depression, PTSD, acute stress disorder, borderline personality disorder and certain forms of behaviour notably parasuicide. In pilot study, comparing performance on the autobiographical memory test (AMT) of adolescents admitted to an in-patient psychiatric unit with a non-clinical school sample, the clinical group, who were more depressed and hopeless than the non-clinical group, were less specific in their response to cue-words on the AMT. Within the clinical group, greater recall of specific memories to negative cues was strongly associated with higher levels of hopelessness in both males and females (Swales, Williams & Wood, in press). However, this study had a number of methodological difficulties, not least that the clinical group were already engaged in treatment, which may have affected their responses on the autobiographical memory test. This paper will present data from a replication of this preliminary study comparing adolescents who have been referred for but have not yet undergone treatment with a non-clinical school sample. In addition this study will also examine data on other variables relevant to autobiographical memory e.g. the intrusiveness and avoidance of traumatic memories (as measured by the Impact of Events Scale) and dissociation.
Reference. Swales, M., Williams, J.M.G. & Wood, P. (in press). Specificity of autobiographical memory and mood disturbance in adolescents. Cognition and Emotion.

Rumination and Overgeneral Autobiographical Memory in Adolescents with Major Depressive Disorder
Rebecca J. Park*, Ian M. Goodyer*, John Teasdale**,
*Section of Developmental Psychiatry, University of Cambridge, U.K.
**MRC Cognition and Brain Sciences Unit Cambridge,U.K.
Both rumination and overgeneral (categoric) memory retrieval have been implicated in the maintenance of adult major depression (MDD), because they relate to poor prognosis. Recent evidence supports the idea that rumination may be involved in overgeneral memory retrieval. These phenomena have not been explored in adolescence and the processes underpinning them remain to be clarified. The current study investigated 1) whether overgeneral memory is a feature of MDD in adolescents, and 2) the impact of rumination on mood and overgeneral memory. In this study, 96 clinically referred adolescents (aged 12-17 years) with MDD and 26 non-depressed psychiatric controls were recruited from child psychiatry services in Cambridge, UK and 33 community controls were recruited form local state schools. All subjects were assessed using the Kiddie-Schedule for Schizophrenia and Affective Disorders, and completed experimental procedures involving manipulation of mood and autobiographical memory using induced rumination and distraction. Overgeneral memories were found to be more common, but not specific to MDD in adolescence (p<.005). In adolescents with MDD, induced rumination as compared to distraction increased negative mood (p<.0001) and increased overgeneral memory (p<.05) , yet had no such influence in nondepressed psychiatric patients. These findings suggest that overgeneral memory may be a modifiable feature of adolescent MDD, and that rumination in adolescents with MDD may impact deleteriously on mood and memory retrieval processes. Implications are discussed, with particular reference to psychological interventions for adolescent MDD.Biases in second-order mindreading in middle childhood

Biases in Second-order Mindreading in Middle Childhood
Carla Sharp, Ian M Goodyer, Section of Developmental Psychiatry, University of Cambridge
This paper discusses the link between deficits in theory of mind and emotional behaviour disorders of childhood. To this effect a new child-centred measure of social cognition, the Social Stories Test, was developed. The test is proposed to measure the quality of second-order theory of mind. Quality in this sense relates to biases in the emotional valence of children's thoughts about their peers thoughts about themselves. Children were asked to respond to stories that contain potentially hurtful social scenarios such that interpersonal processes governing automatic thinking (second-order mindreading) were accessed. Second-order mindreading was found to be positively biased in a group of 41 psychiatrically referred children compared to 65 children from a community sample who showed neutral biases. Biases in mindreading predicted change in subjective an objective depression and behaviour scores over an 8-month follow-up period.

Vulnerability and the Development of Depressogenic Schematic Models of the Self
Philip J. Barnard*, Sophie K. Scott* & Lynne Murray**
*MRC Cognition and Brain Sciences Unit Cambridge.
**Winnicott Research Unit University of Reading
The children of mothers who experience a period of post-partum depression have a high risk of developing depression in later life. It would therefore be expected that these children should acquire schematic models of the self with depressogenic properties. In order to assess children's schematic models of self and other family members, a doll's house play technique was used to elicit discourse about the experience of daily family life in five year old children. A case grammar analysis was then used to characterise properties of their schematic models of self and other family members. When referring to themselves, the children in a high vulnerability group, whose mothers had experienced depression, expressed less agency and more syntactic negation than did a low vulnerability group matched for age, sex and linguistic ability, but where vulnerablity factors were minimised. In contrast, when referring to parents both high and low vulnerability groups showed similar profiles of case role utilization and lower levels of syntactic negation. The data suggest that the two groups are, from an early age, developing distinctive schematic models of the self whose properties vary on dimensions that link nicely with those assumed for adult depressogenic self-models, and that the effects cannot simply be attributed to any differential use of case roles in wider discourse.

SYMPOSIUM 11

When Thinking Too Much is the Problem: Intrusions, Rumination, Control and Suppression Across Disorders.
Mark H. Freeston, Newcastle Cognitive and Behavioural Therapies Centre and University of Newcastle.

This symposium includes five papers that present recent work on distressing thoughts and their control. The presentations draw on previous work on intrusions, rumination and control in novel applications to a range of clinical problems that span chronic pain, depression, insomnia, substance abuse, and eating disorders.

Can the Thought of Suffering be Painful?
Paul Salkovskis & Dorothea Felten, Institute of Psychiatry, Department of Psychology, Kings College London.
Cognitive conceptualisations of pain related to health anxiety suggest that ruminations can be triggered by catastrophising interpretations of episodes of pain. Such catastrophising and ruminations can both result in an elevation of anxiety and/or depression, which in turn can increase catastrophising. These mechanisms are hypothesised to increase the experience of pain. Two investigations are reported addressing 1) the phenomenology and reported impact of rumination, and 2) the affect of rumination induction. The implications for the cognitive-behavioural theory and treatment of some forms of chronic pain are discussed.

Ruminative Intrusions Following Failure
Ed Watkins, Institute of Psychiatry, Kings College London
Recurrent and perseverative thinking about the self, about mood, and about problems is an important factor in the maintenance and relapse of depression (Nolen-Hoeksema, 1996; Teasdale & Barnard, 1993; Pyszczynski and Greenberg, 1987). However, transient increases in intrusive thoughts about problems are a common and often adaptive response to failure or goal frustration (Martin & Tesser, 1989, 1996; Carver & Scheier, 1990). How then do these intrusive thoughts become persistent rumination in people prone to depression? The consensus view is that persistent rumination occurs when people cannot resolve the problem triggering the intrusions (Martin & Tesser, 1989, 1996) and when people cannot give up on unfulfilled goals because they are too personally important (Pyszczynski and Greenberg, 1987). However, most experimental research has not looked at rumination over clinically relevant time frames (e.g. hours, days) nor examined rumination in clinically relevant or vulnerable groups. To rectify these limitations, a series of studies has been started to examine the persistence of dysphoric mood and intrusive thoughts over 2 days following failure feedback on an intelligence test in more vulnerable groups (e.g. people with low self-esteem). In particular, these studies will investigate recent suggestions (Watkins & Teasdale, in press; McFarland & Buehler, 1998) that there are different thinking styles within focus on self and problems, with different implications for the maintenance of rumination. Preliminary findings will be reported.

An Investigation of Strategies of Thought Control in Insomnia.
Allison G. Harvey & Suzanna Payne, Department of Experimental Psychiatry, University of Oxford.
A particularly robust finding in the insomnia literature is that people with sleep-onset insomnia experience unpleasant, intrusive thoughts and worries whilst trying to get to sleep. This paper will present the results of two studies relating to the control of intrusive thoughts and worries in insomnia. Study 1 involved a manipulation of one thought control strategy; thought suppression. Specifically, insomniacs and good sleepers were asked to either suppress their thinking during the pre-sleep period or not to suppress. The morning following the experimental manipulation, participants estimated sleep onset latency and rated their attempted suppression, sleep quality, feeling on waking, frequency of target thought, and discomfort associated with the target thought. Participants instructed to suppress their thoughts estimated their sleep onset latency to be longer, and their sleep quality to be worse, than participants given non-suppression instructions. This effect was observed across diagnostic status. Study 2 investigated whether brief training in identifying and elaborating an interesting and engaging imagery task for use during the pre-sleep period can reduce unwanted pre-sleep cognitive activity and sleep onset latency. Forty one people with insomnia were given one of three instructional sets to follow on the experimental night; instructions to distract using imagery, general instructions to distract, or no instructions. Based on previous findings reported by Salkovskis and Campbell (1994) and ironic control theory (Wegner, 1994), it was predicted that (1) 'imagery distraction' would be associated with shorter sleep onset latency and less frequent and distressing pre-sleep cognitive activity compared to the 'no instruction' group and that (2) 'general distraction' would be associated with longer sleep onset latency and more frequent and distressing pre-sleep cognitive activity compared to the 'no instruction' group. Support was found for the first but not the second prediction. The success of the 'imagery distraction' task is attributed to it occupying sufficient 'cognitive space' to keep the individual from re-engaging with thoughts, worries, and concerns during the pre-sleep period.

Intrusive Thoughts in Substance Dependent Patients
Martina Reynolds, Department of Psychiatry of Addictive Behaviour and Psychological Medicine, St. George's Hospital Medical School.
Thought processes have been hypothesised to play a role in addiction and relapse. Salkovskis and Reynolds (1994) showed that suppression was associated with an increase in smoking related intrusive thoughts, whilst a distracting task (relaxation) reduced intrusion frequency. This is a report of a similar study with a substance dependent sample undergoing detoxification. Subjects were randomly allocated to one of three groups (mention control, relaxation and suppress) and respectively were asked to monitor, suppress and do relaxation exercises, and suppress substance related intrusive thoughts in period 1. In period 2 they were told that they could think about anything. Subjects recorded all substance related intrusions in both periods. Participants were an inpatient substance dependent sample undergoing detoxification. Results showed that deliberate suppression of substance related intrusive thoughts did not result in an increase in frequency of the same for the suppression group compared to the mention control group. Relaxation facilitated thought suppression in the first period, but this effect was not carried over to the second period. In conclusion, in the absence of the task which acted as an effective structured distracter, feeling relaxed may act as a trigger for drug related thoughts. This may have some implications for the use of relaxation as a form of distraction in treatment for substance misusers.

Treating Obsessional Preoccupation with Body Image and Weight in Anorexia: A Pilot Study
Mark H. Freeston, Newcastle Cognitive and Behavioural Therapies Centre and University of Newcastle, United Kingdom; Gilles Gaudette, Yves Careau, Nicole Mainguy, Hôpital Louis-H. Lafontaine, Montréal, Canada
'Given the efficacy of CBT in treating OCD, it is tempting to wonder whether therapeutic leads might not evolve from the evolving analyses of obsessional states (Wilson, 1999; p. 85)'. Based on this proposition, concern with weight and body image was conceptualised as a form of ruminative thinking and an intervention was designed drawn directly from CBT for obsessional thoughts. An initial test of the intervention was conducted within an intensive single-case design to target an aspect of anorexia that may remain once initial weight gain has been achieved. The technique proved to be successful in reducing preoccupation for this participant who was not in an acute phase of the disorder.

SYMPOSIUM 12

Trauma and Mood Disturbance Issues
Nick Grey, Institute of Psychiatry, London

Did it Really Happen? The Vagaries of Spontaneous Imagery in Posttraumatic Stress Disorder from a Clinical Perspective.
NickGrey, Centre for Anxiety Disorders and Trauma, Institute of Psychiatry
A distinguishing characteristic of PTSD is the presence of re-experiencing symptoms in the form of flashbacks, nightmares and intrusive memories. Most commonly these take the form of visual imagery, but also include other somatosensory information. Whilst classically this imagery is a replica reliving of the traumatic event it is often noted that the images reported by patients do not exactly match their experience. This paper makes a distinction between veridical and non-veridical images. Non-veridical images include out-of-body experiences (dissociation), composite images, worst-case scenarios (Merckelbach et al, 1998), and reconstructed images (Bryant & Harvey, 1998). A second distinction is made between imagery that can be understood as arising from during the traumatic event itself (i.e. peritraumatically) and that which is associated with later appraisals. A variety of clinical case examples are presented to illustrate these differences. These are conceptualised using recent cognitive models of PTSD (Brewin et al, 1996; Ehlers & Clark, 2000) together with specific work on image control in PTSD (e.g. Laor et al, 1999). Possible clinical approaches to working with the various types of imagery are highlighted. It is argued that clinically it does not necessarily matter whether the images are an accurate representation of the traumatic event as working with the associated meanings for the individual is of greater importance (see Hackmann, 1998).

Developmental Risk Factors for PTSD: The Role of Parental PTSD and Childhood Trauma
Rachel Yehuda, Ph.D., Sarah Halligan, Ph.D. (presenter), Robert Grossman, M.D., Affiliated to the Mount Sinai School of Medicine, New York
Studies on the impact of trauma have identified a family history of psychiatric disorder as a risk factor for the development of PTSD (e.g. McFarlane, 1988). Preliminary research in Holocaust survivors has further suggested the parental PTSD may be a relatively specific risk factor for PTSD in offspring (Yehuda, Schmeidler, Giller, Siever, & Binder-Brynes, 1998). The current research represents a comprehensive examination of the role of parental experiences in mediating vulnerability to psychopathology in a sample of adult offspring of Holocaust survivors.
First, the contributions of parental trauma exposure versus parental PTSD to the development of PTSD and other psychiatric diagnoses in the offspring are assessed.
Second, the Childhood Trauma Questionnaire (Bernstein et al., 1994) is used to examine early trauma exposure in the offspring. The role of negative childhood experiences in mediating the relationship between parental symptomatology and vulnerability in the offspring is investigated. Results and implications will be discussed.
References
Bernstein,D.P., Fink,L., Handelsman,L., Foote,J., Lovejoy,M., Wenzel,K., Sapareto,E., & Ruggiero,J. (1994). Initial reliability and validity of a new retrospective measure of child abuse and neglect [see comments]. American Journal of Psychiatry, 151, 1132-1136.
McFarlane,A.C. (1988). The aetiology of post-traumatic stress disorders following a natural disaster. British Journal of Psychiatry, 152, 116-121.
Yehuda,R., Schmeidler,J., Giller,E.L., Siever,L.J., & Binder-Brynes,K. (1998). Relationship between posttraumatic stress disorder characteristics of Holocaust survivors and their adult offspring [see comments]. American Journal of Psychiatry, 155, 841-843.

Poor Memory as a Predictor of Poor Treatment Response in Adults Diagnosed with Posttraumatic Stress Disorder
Jennifer Wild, Institute of Psychiatry, London, UK, Dr Sallie Baxendale, Institute of Neurology, London, UK, Dr Peter Scragg, University College London, UK
Introduction: This study highlights the importance of memory functioning in cognitive-behavioural therapy (CBT) for Posttraumatic Stress Disorder (PTSD). Most research investigating PTSD has been divided, either drawing on biological or cognitive models. Biological models have focused on alterations in brain morphology (e.g. decreased hippocampal size), and altered levels of neurotransmitters. Cognitive models have focused on poor consolidation of the trauma memory and the role of appraisals in the onset and maintenance of PTSD. This study draws on both biological and cognitive models to explain the role of memory functioning in PTSD. Method: This study investigated the memory, attention and learning profiles of 27 adults diagnosed with PTSD prior to commencing CBT at a specialist treatment centre. Memory was assessed both subjectively and objectively. Subjective appraisal of memory was assessed by a self-report questionnaire in which participants rated their perception of memory difficulties. Objective assessment of memory was assessed by a neuropsychological test battery. This included: baseline cognitive functioning, memory, learning, attention, and executive function. In addition, measures of PTSD, anxiety, depression, and past and current history of alcohol/substance use were also obtained prior to memory assessment.
Results:Twenty-three adults were followed up at session eight of treatment, and their PTSD diagnosis was re-evaluated. The results were grouped into two categories: (1) treatment outcome, and (2) memory appraisal. Treatment outcome: Clients who did not improve with treatment had significantly poorer performance on intake measures of verbal memory. In particular, a measure of encoding meaningful verbal material was found to independently predict outcome. Differences were not accounted for by performance on tasks of attention and executive function. Further, severity of PTSD symptomatology, severity of anxiety and depression, length of time since trauma, and alcohol and substance use were not related to memory functioning. Memory appraisal: Perception of memory difficulties was significantly correlated with memory functioning. Those participants with a more negative appraisal of memory functioning had poorer neuropsychological memory scores.
Conclusion: Cognitive models emphasise the nature of the trauma memory and its meaning in PTSD. Biological models focus on areas of the brain implicated in memory functioning in general. This study found that objective memory performance (not merely for the trauma memory) is a predictor of treatment outcome, thereby bringing together the biological and cognitive perspectives on memory.
Although the sample size in this study was small and further research is desirable, the findings suggest that a self-report questionnaire for memory problems could be an important adjunct to initial psychological assessment and treatment planning for PTSD. The self-report questionnaire could indicate the potential utility of more detailed neuropsychological testing. Neuropsychological testing could then inform the modification of individual CBT so as to best meet the client's need. For example, progressing at a slower pace, use of more repetition, increasing homework regarding listening to tapes, use of visual stimuli and notes. However, testing is rarely indicated when the client rates his/her memory as good on the self-report questionnaire.

The Role of Negative Interpretation Thought Suppression and Rumination in the Maintenance of Post-Traumatic Stress Symptoms: A Prospective Longinitial Study
Ben Smith, Traumatic Stress Clinic, ,London
Objective: Recent cognitive conceptualisations of posttraumatic stress disorder (PTSD) emphasise the role of negative interpretation of initial symptoms and avoidant cognitive coping in symptom maintenance. It was investigated whether these factors function to maintain posttraumatic stress symptoms following a minor road traffic accident (RTA), and the extent of their influence compared to other variables.
Design: A longitudinal study with measurement within one month of trauma. Follow up conducted after 4 months. Dependent variables included the IES-R and SRS-PTSD(DSM-IV), Carlier et al (1998).
Setting: Subjects were recruited from A&E at a London Hospital and assessed in their homes.
Subjects: 50 subjects were assessed at baseline and 39 successfully followed up.
Results: A maintenance factors variable consisting of negative interpretation, thought suppression and rumination was the only variable to exhibit independent and significant prediction of the IES-R and SRS-PTSD at follow up. This variable was also found to strongly mediate the relationship between baseline and follow up scores on the IES-R and SRS-PTSD.
What is already known on this subject and what does this study add?
The independent and significant role of negative interpretation, thought suppression and rumination in the maintenance of PTSD has not been shown before in a sample exclusively of minor trauma victims. This adds to empirical evidence and theoretical conceptualisations of PTSD implicating these variables in the maintenance of symptoms in a wider range of traumas. The strong mediating role of the maintenance factors variable has important implications for the identification and Cognitive-Behavioural Treatment of potentially chronic PTSD in minor trauma victims

What We Think About Ourselves, or How We Think About Things? The Relationship Between Everyday Reasoning and Mood-related Reasoning
Dale Huey, Newcastle Cognitive & Behavioural Therapies Centre & University of Newcastle; Rosemary Stevenson, Jayne Alderdice, & Sharon Godfrey University of Durham, England.
We are interested in the way people think about non-self-referent events and how this may relate to thinking about self-referent events. We are particularly interested in how balanced, or flexible, people are in their thinking. We aim to determine how such everyday reasoning may relate to the kind of cognitive-processes, and -content, typically associated with psychopathology. Our theory presupposes individual differences in domain-independent reasoning style and predicts that a more balanced approach will be more adaptive; as, if there are generalized reasoning styles, a more balanced approach should provide a buffer against the extremes of self-denigratory cognition. On this view, there should be an observable relationship between reasoning style across domains. We propose that a biased, and inflexible, general reasoning style may provide a better predictor of vulnerability to mental health problems than more mood-related thinking. The first stage of the project is described and provisional data from clinical (n=40) and non-clinical samples (n=70) are reported.

SYMPOSIUM 13

Cognition and Emotion Across Disorders: Advances in Theory, Research and Practice
Warren Mansell, Department of Psychology, Institute of Psychiatry, London

Neurosis and Psychosis: The Influence of Emotion on Delusions and Hallucinations
Daniel Freeman, Department of Psychology, Institute of Psychiatry, Kings College, London
A core classificatory divide exists between neurosis and psychosis, leading to their separate study and treatment. The basis for the separation of the disorders will be outlined and reassessed. It will then be highlighted that emotional disorder frequently occurs prior to and accompanying psychosis, indicating that neurosis has the potential to contribute to the development of the positive symptoms of psychosis. Therefore, psychological theories and experimental evidence concerning the influence of emotion on the content and form of delusions and hallucinations will be reviewed. It will be argued that in many cases delusions are a direct representation of emotional concerns, and that emotion contributes to delusion formation and maintenance. The content of hallucinations less often directly expresses the emotional concerns of the individual, but emotion can trigger and contribute to the maintenance of hallucinatory phenomena, although how this occurs is not well understood. It will be concluded that study needs to be made of the interaction between psychotic and neurotic processes in the development of delusions and hallucinations, and that neurotic and psychotic disorders may have a number of common maintaining factors

The Truth About Mental sSmulation
Gary Brown, Salomons, Christchurch University College, Kent
The central intuition of the cognitive model is that emotional problems arise when people react on the basis of their misrepresentations of reality. It is therefore surprising that we know little about how individuals come to judge these misrepresentations as being true. This paper is a theoretical and empirical overview of mental simulation, which is potentially the central mechanism that people use to judge what is true or potentially true. Thus, for example, people who are better able to simulate some future catastrophe are more likely to be worried and anxious about such a catastrophe occurring. That simulation has been relatively neglected is likely due to (1) a preference in the field for convenient, static measures of cognition (e.g., questionnaires) that are not capable of capturing dynamic processes and (2) a neglect of the distinction between the representational function of cognition versus its heuristic function. The various theoretical strands relevant to mental simulation will be drawn together and the potential of this line of research for clinical psychology will be outlined. A particular focus will be on the implications of a new integrated framework recently proposed by Sanna (2000). Attention will be given to conditions in which simulation likely plays a central role (generalised anxiety disorder and worry, obsessive compulsive disorder, postraumatic stress) as well as its likely secondary function in other conditions. Methodological challenges will be discussed in the context of previous and ongoing research.

Perfectionism Across the Disorders
Roz Shafran, Warren Mansell, Zafra Cooper & Christopher G. Fairburn , Oxford University Department of Psychiatry, Oxford
Perfectionism is an important clinical problem that can impede the successful treatment of a range of disorders. This presentation will specify the cognitive characteristics of perfectionism as described by clinicians. It will be argued that the existing multidimensional constructs of perfectionism unjustifiably broaden the construct and that there is need for a return to a unidimensional construct. The core cognitive psychopathology of this construct is suggested to be the overdependence of self-worth on the pursuit and attainment of personally demanding standards, in at least one salient domain, despite adverse consequences. The existing literature suggests that at least two of the subscales of the multidimensional measures of perfectionism ("socially prescribed perfectionism" and "concern over mistakes") are associated with a range of psychopathology and are elevated in people with anxiety disorders and eating disorders. However, a re-examination of the existing literature in terms of the core psychopathology of perfectionism indicates that there is a specific relationship between perfectionism and eating disorders. It is suggested that the elevation in "socially prescribed perfectionism" and "concern over mistakes" in people with anxiety disorders reflects their beliefs about other people's evaluations and expectations of them, as opposed to "perfectionism". The treatment implications of this cognitive re-analysis of perfectionism will be discussed.

Cognitive Biases in Emotional Disorders: Current Controversies in Theory and Research
Karin Mogg & Brendan Bradley, Department of Psychology, University of Southampton
Recent cognitive theories of emotional disorders propose that cognitive biases for negative or threatening information play a critical role in causing and maintaining anxiety and depressive disorders, including generalised anxiety states, phobias and major depression (e.g. Williams, Watts, MacLeod & Mathews, 1997; Mogg & Bradley, 1998; Clark, 1999). However, there is considerable dispute about the nature of cognitive biases in emotional disorders. For example, some theories predict that anxiety disorders are characterised by vigilance for threat (e.g. Williams et al. 1997), whereas, according to other models, different anxiety disorders are associated with different patterns of attentional bias, e.g. vigilance for threat in simple phobia and generalised anxiety, but avoidance of threat in social phobia (Clark, 1999). Further controversy surrounds the nature of attentional biases in depressive disorders. These issues will be evaluated in the light of recent research evidence. The implications of this research for the identification of cognitive vulnerability factors in emotional disorders, and for the development of more effective cognitive treatments will also be considered.

If Our Minds Work Like Evolution, What Should We Do About It?
Warren Mansell, Department of Psychology, Institute of Psychiatry, Kings College, London
What have cognitive therapy, mass hysteria and evolution got in common? Several theorists (Blackmore, 1999; Campbell, 1974; Cziko, 1995; Dawkins, 1976; Dennett, 1995) have suggested that the principles of natural selection apply not only to evolution, but to human learning and the spread of beliefs and behaviour through society. Natural selection relies on (a) multiple variations of an entity, (b) the ability of the entity to copy itself, and (c) an environment in which certain entities will be more likely to survive and/or make copies of themselves. The case will be made that potentially dysfunctional beliefs, cognitions and behaviours can develop through this process during their competition for survival both within an individual's belief system, and between individuals in a society. This approach may have two beneficial influences on our understanding of psychological disorders. First, it may help to clarify how and why a variety of diverse techniques within therapy (e.g. decentering, cognitive-reappraisal, problem-solving and operant learning) can lead to effective change. Second, by conceptualising disorders as the end result of problems with adaptation at multiple levels (genes, behaviour, cognition and culture), it may be possible to provide a framework that can promote integration between different disciplines of clinical psychology and psychiatry.

SYMPOSIUM 14

Early Intervention in Psychosis
Max Birchwood, University of Birmingham

Abstracts not available

SYMPOSIUM 15

Psychological Approaches and Sleep, Part 1: Recent Developments in Treatment
Allison G. Harvey, University of Oxford

Insomniacs' Reported Use of CBT Components and Relationship to Long-term Clinical Outcome.
Colin A. Espie, Linda Harvey and Stephanie J. Inglis
Dept. of Psychological Medicine, University of Glasgow
Although there is considerable evidence for the efficacy of non-pharmacological treatment of insomnia, many of the larger trials have delivered CBT in multicomponent format. This makes it impossible to identify critical ingredients responsible for improvement. Furthermore, compliance with home implementation is difficult to ascertain in psychological therapies, and even more so when trying to differentiate across a range of elements. In the present report, 90 patients who had completed 12 month follow-up after participation in a clinical effectiveness study of CBT in general medical practice, responded to a questionnaire asking them about their use of the ten components of the programme. Reports of home use were then entered as predictors of clinical response to treatment. Results indicated that reported home use of stimulus control/ sleep restriction was the best predictor of clinical improvement in sleep latency and nighttime wakefulness. Cognitive restructuring also contributed significantly to reduction in wakefulness. In spite of being the most highly endorsed component (by 79% of respondents) use of relaxation did not predict improvement on any variable. Similarly, sleep hygiene was unrelated to sleep pattern change and use of imagery training was modestly predictive of poor response in terms of sleep latency. There are methodological limitations to this type of post hoc analysis, nevertheless, these results being derived from a large patient outcome series raise important issues both for research and clinical practice.
This research was supported by grants from the Chief Scientist Office, Scottish Office Department of Health and Ayrshire and Arran Health Board.

Psychological Treatment in the Management of Chronic Insomnia in Primary Care*
Kevin Morgan1, Simon Dixon2, Nigel Mathers3, Joanne Thompson3, and Maureen Tomeny4
1Gerontology Research Unit, Department of Human Sciences, Loughborough University,
2Institute of General Practice and Primary Care, University of Sheffield
3Sheffield Health Economics Group, University of Sheffield
4Department of Clinical Psychology, Central Nottinghamshire Healthcare Trust
Background. While the clinical trials evidence shows that chronic insomnia can be effectively treated with appropriate psychological therapies, such treatments are rarely deployed in primary care settings - where most complaints of insomnia are made.
Objective. To evaluate the practicality, cost and clinical effectiveness of a primary care 'sleep clinic' providing short-term cognitive/behavioural treatment in the management of chronic insomnia in primary care settings.
Methods. Patients presenting, or recalled, for repeat hypnotic prescriptions in 23 Sheffield general practices were invited to participate in either the clinic (50%) or control (50%) conditions of the study. Clinic patients were offered a package of psychological therapies for insomnia over 6 sessions (including health education, sleep hygiene, stimulus control procedures, relaxation and cognitive approaches). Treatment was delivered by primary care counsellors supervised by a clinical psychologist. Controls were assessed, but received no additional treatment. Outcomes included sleep quality, drug use, and patient ratings of service satisfaction. Economic analyses of trial costs and NHS insomnia care were conducted alongside the trial.
Results. A total of 101 and 109 patients participated in the control and clinic groups respectively (total = 210). The sleep clinic was associated with improved sleep quality, reduced hypnotic use, positive ratings of patient satisfaction, and high take-up by general practices. Economic analyses indicated substantial hidden costs associated with hypnotic drug usage.
Conclusions. The primary care sleep clinic model described here can offer a high quality, low risk response to chronic insomnia in general practice settings.
*This project was funded by the NHS Health Technology Assessment Programme (project 95/30/02)

The Efficacy of a Pennebaker-like Writing Intervention for Poor Sleepers
Clare Farrell1 and Allison G. Harvey2
1Department of Psychiatry, University of Oxford
2Department of Experimental Psychology, University of Oxford
Insomniacs commonly complain that they are unable to get to sleep because of unwanted thoughts and worries. This excess cognitive activity has been attributed to incomplete processing of daytime stressors and hassles. Previous research has demonstrated the benefits of writing about emotional experiences as a method to facilitate emotional processing. The present pilot study tested the hypothesis that writing about worries and concerns, with an emphasis on the expression and processing of emotion, will reduce sleep onset latency among poor sleepers. Forty two poor sleepers were randomly allocated to one of three groups for three nights; the instructions for the 'problems' writing group emphasised the expression and processing of worries and concerns, the instructions for the 'hobbies' writing group emphasised distraction from worries and concerns by writing about hobbies and interests, the 'no writing' group were not given a writing task. The 'problems' writing group reported shorter sleep onset latency compared to the 'no writing' group. It is concluded that a writing intervention, designed to enhance emotional processing, may assist poor sleepers to fall asleep faster and has potential to be a useful treatment component.

Correcting Inaccurate Perception of Sleep Onset and Total Sleep Time in Insomnia: A Novel Treatment Component?
Nicole K.Y. Tang and Allison G. Harvey, Department of Experimental Psychology, University of Oxford
Insomniacs tend to overestimate their sleep onset latency and underestimate the total number of hours they slept. These findings raise the possibility that distorted perception of sleep may contribute to the maintenance of insomnia. Aaron T. Beck (1976) has highlighted the importance of 'distortions of reality' in the maintenance of emotional disorders. Since then clinical researchers have been interested in identifying distortions and developing interventions to correct distorted perceptions across a wide range of disorders. This paper will present the results of a study that explored the role of distorted perception in the maintenance of insomnia and piloted the utility of a novel intervention to correct distorted perception. Insomniacs were asked to wear an actiwatch and keep a sleep diary for 3 nights. On the following day, half of the participants were shown the discrepancy between the actiwatch recording and their sleep diary (discrepancy demonstrated group), while the other half were not shown the discrepancy (no demonstration group). Participants were then asked to wear the actiwatch and keep a sleep diary for 3 further nights. On the following day, the discrepancy demonstrated group reported reduced sleep-related anxiety and were more accurate in their estimation of the amount of sleep they obtained. Clinically, these findings highlight the potential benefits of targeting inaccurate sleep perception when treating insomnia and provide a possible method to achieve this. Theoretically, the results provide evidence that inaccurate perception of the amount of sleep obtained may be involved in the maintenance of insomnia.

Psychological and Clinical Predictors of Compliance with Continuous Positive Airway Pressure Therapy for Obstructive Sleep Apnoea: A Prospective Study.
Matt Wild1, Colin A. Espie1, Neil Douglas2, and Heather Engleman2
1Dept. of Psychological Medicine, University of Glasgow
2 Scottish National Sleep Laboratory, Universiy of Edinburgh
Objectives: Obstructive sleep apnoea (OSA) is a disorder which is associated with excessive daytime sleepiness, cardiovascular disease and road traffic accidents. Continuous positive airway pressure therapy (CPAP) is the current treatment of choice and is highly effective. However, compliance with this treatment is low. As yet, compliance rates have not been consistently explained by physical or clinical variables. Wallston's modified social learning theory was applied to a population of OSA sufferers to establish whether an amount of variance could be explained on the basis of psychological variables.
Design: A prospective, naturalistic study was conducted. Psychological clinical and physical variables were considered in order to establish the best model of compliance.
Methods: One hundred and twenty consecutive OSA sufferers were recruited. Measures of heath locus of control, self-efficacy and health value were administered prior to initiation of CPAP therapy. Clinical and physical data were also collected at this time. Participants were followed up at three months and objective compliance rates were collected from their CPAP apparatus.
Results: Initial between group analysis found that Wallston's model did not usefully differentiate between compliers and non-compliers. However, logistic regression analysis revealed that a model comprising health value, body mass index and Epworth scores (i.e., a subjective measure of daytime sleepiness) explained a small, but clinically significant amount of variance.
Conclusions. These results support previous research findings that compliance with CPAP may be most productively researched by considering psychological as well as physical and clinical variables. This bio-psychosocial approach is to be encourages in other areas of medical research.

SYMPOSIUM 16

New Developments in Anger Treatment
Raymond W. Novaco, University of California, Irvine

Central Issues in Anger Treatment: Theoretical, Methodological, and Practical
Raymond W. Novaco, University of California, Irvine, USA
Anger is an important clinical subject for mental health professionals who work in a wide range of community and institutional settings, not only because it is a significant activator of violent behaviour, but also because it is evoked by many stress-related conditions. Recurrent anger is often a product of troubled life histories and readily becomes part of a dysfunctional style of dealing with life's challenges. Intrinsic to the predicament of people with serious anger problems is treatment resistance, and clinicians are often less than keen on seeking to engage them in the therapeutic enterprise. Major issues arising in the recruitment, referral, assessment, and preparation of angry patients will be outlined. Significant treatment gains that have been achieved with patients having serious anger problems will be overviewed. Key elements of successful treatment interventions, grounded in outcome evaluation research, will be presented. As significant gains have been achieved with highly distressed clients having very complex clinical needs, there is much hope for many other people who struggle to manage their anger and seek clinical assistance.

Anger Treatment Outcomes for Offenders with Learning Disabilities
John L. Taylor, University of Northumbria at Newcastle / Northgate & Prudhoe NHS Trust
Rates of aggression are high amongst people with intellectual disabilities living in institutional settings. Anger is a significant activator of aggressive behaviour. There is some limited evidence for the value of cognitive-behavioural treatments for anger problems with people with intellectual disabilities. No controlled studies of anger treatment involving intellectually disabled offenders living in secure settings have been conducted to date.
In a pilot study, detained patients with intellectual disabilities and histories of offending were allocated to specially modified cognitive-behavioural anger treatment or to routine care waiting-list control conditions. Eighteen sessions of individual treatment were delivered over a period of 12 weeks. Patients' self-report of anger intensity to an inventory of provocations was significantly lower following intervention in the treatment condition, compared to the wait-list condition. Limited evidence for the effectiveness of treatment was provided by staff ratings of patient behaviour post-treatment.
Detained offenders with intellectual disabilities can benefit from intensive individual cognitive-behavioural anger treatment. Further research is required to examine the mechanisms for change and their sustainability.

Anger Management with Offender Patients
Yvonne Shell and Estelle Moore, Department of Psychology, Broadmoor Hospital, Berkshire
This paper will draw on the experience of working with a group of young male offender-patients with complex clinical presentations detained in a high security hospital. The group participants undertook a 12-month programme of anger treatment utilising a cognitive-behavioural approach which incorporated the need for a 'preparatory phase' as identified by Renwick, Black, Ramm, and Novaco (1997). Using case material, we will consider: individual differences in response to the treatment; the progress of the group as rated by independent observers; and clinical outcome, as evidenced in their subsequent pathways through treatment in the hospital. Within this we will give voice to the patients feedback on the groupwork experience and discuss how this might inform our understanding of the offender-patients' use of strategies for managing anger. Particular attention will be given to the role of shame and its links with earlier trauma and the expression of anger, as this evolved from work within the group on offending behaviour. The impact of a high intensity of emotion on the group process, and on the therapists involved, will be considered. We will conclude with a brief discussion of the implications for the design and delivery of interventions for anger in the context of a period of detention for young men with offending histories.

Group Anger Management with Mentally Disordered Offenders
Mark Ramm, The Orchard Clinic, Edinburgh
Mentally ill patients who are detained because of their "dangerous and criminal propensities" often have substantial difficulties associated with anger. An integrated programme was therefore developed to address these needs at the State Hospital for Scotland and Northern Ireland. It involves comprehensive assessment and the provision of clinical intervention at several therapeutic levels. The presentation focuses on "anger management" groupwork, during which participants are helped to understand their anger-mediated difficulties and to develop effective self-control skills. The approach is based on the cognitive-behavioural approach developed by Novaco, but the groupwork intervention incorporates a number of innovative developments, which will be described. A brief account will be given of the group sessions and the weekly individual support sessions. Preliminary results of a comprehensive evaluation study will be reported and some observations made.

SYMPOSIUM 17

Therapeutic Strategies Across Disorders
BABCP Scientific Committee

A Multi-dimentional Approach to Working with Methadone Maintained Families: Preliminary Results from a Randomised Controlled Trial
Sharon Dawe1, Paul Harnett2, Petra Staiger3 & Vanessa Rendalls1
1School of Applied Psychology, Griffith University, Brisbane, Qld 4111, 2 University of Queensland, Brisbane, Qld, 3Psychology Department, Deakin University, Melbourne.
Children raised in substance abusing families show high rates of behavioural and emotional problems, in particular oppositional, defiant behaviours. Problems at the level of the individual parent, family relationships, and life-style factors are related to poor child outcomes and adult substance abuse. In order to decrease child behaviour problems, improve family relationships and decrease parental substance use, we have developed a multi-dimensional program, Parents Under Pressure, that address problems at each ecological level (individual, family, and community) in a structured but non-sequential format (SNS) for families who are currently on methadone maintenance. In PUP family problems are conceptualised more broadly than as a parenting skills deficit. Specifically, low confidence in parenting ability, guilt associated with poor parenting practices, emotional, and social problems of the parents, and high levels of stress (and the effect that each of these has on ongoing, nonprescribed drug use) are seen as factors that interfere with the provision of good enough parenting. Interventions are structured insofar as each identified problem area is addressed by a manualised intervention. However, the program is non-sequential in that the problems targeted vary from family to family, with the order of presentation determined by the salient presenting problem for the family in each session. Problem solving strategies for dealing with pressing issues provide parents with some control and direction for dealing with their immediate stress. Critically, the problem solving strategy allows parents to attend to other aspects of the intervention.
The current study aims to determine the relative effectiveness of this intensive, multidimensional approach compared to a brief behavioural parenting intervention and standard care in families in which either or both parents are currently enrolled in a methadone maintenance program. The intensive program, PUP, is delivered in families' homes and consists of up to 12 sessions focusing on child behaviour, the parent-child relationship, beliefs about parental adequacy and parental functioning (including drug abuse). The brief behavioural parenting intervention consists of 2 sessions and is delivered in the clinic. Standard care consists of the current treatment program of once monthly contact with a caseworker. Using measures sensitive to child behaviour (including behavioural observation of parent child interactions), parental functioning and parental substance abuse, we present outcome data on the first 50 families who have taken part in the trial.

Stress Control' Large Group CBT for GAD: Eight Year Follow-up
Jim White, Clydebank Health Centre, Scotland
This paper will look at an eight year follow-up study. The controlled trial involved cognitive therapy, behaviour therapy, cognitive-behavioural therapy, placebo ('Subconscious Retraining') and waiting list. 109 (DSM-III-R) GAD patients took part. The approach involved 6, 2 hour 'evening class' sessions. The sessions involved didactic teaching, slide and video presentations, workshops and homework. Patients in each treatment condition were given a booklet specific to that approach. This paper will describe outcome in terms of questionnaire data (pre, post, 6 month, 24 month and 8 year), further psychological and pharmacological treatment and patients experiences of the approach.

Goal Planning: A Retrospective Audit of Rehabilitation Process and Outcome
Jane Duff, Matthew Evans, Paul Kennedy, National Spinal Injuries Centre, Stoke Mandeville Hospital NHS Trust, Aylesbury, UK.
Background: Goal Planning is a well recognised approach used in a number of clinical settings. The approach is based on behavioural change principles (Houts and Scott, 1975) which emphasise the identification of specific goals and targets in relation to the strengths and needs of the individual. Within spinal cord injury rehabilitation, this multi-disciplinary approach aims to enhance client involvement and adaptation to injury. Active in-patient involvement in rehabilitation has been found to the best predictor of long-term medical and behavioural outcome (Norris-Baker et al, 1981). However, inspite of the general acceptance and wide usage of the approach, little research has been conducted on its application (Wade, 1998). The Goal Planning system used in this research integrates behavioural management strategies with a clinical outcome measure, the Needs Assessment Checklist (Kennedy and Hamilton, 1999), formed from 216 behavioural indicators.
Objectives: To evaluate the outcome, utility and effectiveness of the Needs Assessment and Goal Planning approach within spinal cord injury rehabilitation and audit the application of the approach against previously set standards for the service.
Method and Design: A retrospective audit of sixty-four patients who participated in the Needs Assessment and Goal Planning approach between 1997 - 2001 within an in-patient hospital based rehabilitation centre. The sample included 48 men and 16 women, with a mean age of 41 years (range 15-71 years). The sample were representative of a spinal cord injury population in relation to level and neurological completeness of injury.
Results: The total number of goal planning meetings for the sample was 381 (range of 1 to 15), with an average of 6 meetings per patient being held. A total of one thousand, two hundred and eight goals were set across the sample at the first goal planning meeting, with 75% of these goals being achieved by the second meeting. The total achievement rate across the meetings was 71%, with an average of 13.6 goals being set per meeting. Staff attendance at meetings was very high with 83.5 % of meetings being attended by physiotherapists, 77.9 % by Occupational Therapists and 68.2 % by Nursing Staff. Qualitative information about the process of rehabilitation and functional outcome were also obtained. In addition, information will be presented comparing the process of goal planning with the Needs Assessment Checklist, a tool used to assess outcome in this population.
Conclusions: This study provides evidence that Goal Planning is an effective tool for a patient-centred, multi-disciplinary team approach to rehabilitation management. The benefit of systematic audit of a service in providing information about the rehabilitation process is also highlighted.

Increasing the Odds for Change: Utilising Motivational Strategies with Problem Gamblers
Alison Salmon and David Ryder, Edith Cowan University, Perth, Western Australia
The 1990's witnessed an unprecedented proliferation of gambling venues in Australia in the form of casinos and electronic gaming machines, which have been patronised enthusiastically by Australians. In 1998 over 80 per cent of Australians gambled at least once and 40 per cent gambled on a regular basis. According to a 1999 report by the Australian Productivity Commission 15 per cent of regular (non-lottery) gamblers are problem gamblers. It is estimated that problem gamblers represent 2.1 per cent of the Australian adult population and account for expenditure of A$ 3.5 billion annually. Problem gamblers lose around A$12,000 per head each year compared with under A$650 for other gamblers. In addition to financial costs, there are significant emotional costs associated with problem gambling; for example one in ten problem gamblers said they have contemplated suicide due to gambling. Clearly in an environment where gambling continues to be increasingly accessible there is a need to provide effective treatment for problem gamblers.
In 1999 a study was carried out by Edith Cowan University in Western Australia to identify best practice strategies for the treatment of problem gambling. The study reviewed 46 research papers and interviewed 17 key informants. It was concluded that at this stage there is a dearth of methodologically sound research to support the efficacy of specific interventions although there is some support for the use of cognitive-behavioural strategies such as imaginal desensitisation, cue exposure/response prevention and the addressing of erroneous beliefs regarding gambling. The need to attend to motivational issues was consistently raised in the literature and emphasised by key informants. The study also identified a need to provide services for those who are unwilling or unable to attend specialist gambling services. These may be people who are ambivalent about changing their gambling behaviour or those who feel uncomfortable attending specialist gambling counselling services. The latter may include women or members of ethnic minority groups. Despite the paucity of rigorous research in the area of problem gambling, it is suggested that as in the treatment of other addictive behaviours, non-specialist services and practitioners are well placed to respond to problem gamblers providing that staff have appropriate training. Such training necessarily includes motivational strategies. It is also suggested that other strategies that have been successful in the alcohol and other drug area such as brief interventions and self-help manuals may be useful in assisting problem gamblers. This paper will explore ways in which existing knowledge and practice in the alcohol and other drug field can be incorporated into interventions with problem gamblers. In particular the author will discuss how non-specialists can incorporate motivational strategies for problem gamblers into their work. Drawing upon the Australian experience the paper will also suggest that whilst maintaining a focus upon the individual problem gambler, it is important to pay attention to environmental factors which can encourage or restrain gambling behaviour.

Coping Effectiveness Training: A Group Controlled Intervention for People With Traumatic Spinal Cord Injuries
Paul Kennedy, Jane Duff, Matthew Evans, National Spinal Injuries Centre, Stoke Mandeville Hospital NHS Trust, Aylesbury, UK.
Objectives: To evaluate the efficacy of a group based intervention programme to improve psychological adjustment and reduce anxiety and depression following traumatic spinal cord injury. The Coping Effectiveness Training (CET) programme is grounded on the cognitive theory of stress and coping and this study develops the results of a previous pilot investigation. Design: A trial comparing patients receiving the CET intervention with matched controls on measures of adjustment and coping. Methods: Forty-two intervention group participants and forty matched controls were selected from in-patients at a hospital based rehabilitation centre. Outcome measures of depression (BDI), anxiety (SAI) and coping (COPE) were collected before, immediately following and at six weeks post intervention. Results: Intervention group participants showed a significant reduction in depression (p<0.01) and anxiety (p<0.05) compared to matched controls immediately after the intervention and at six weeks follow-up. The pattern of coping strategies did not alter. Conclusions: These results confirm the results of the pilot study, that the CET intervention reduced mood disturbance following injury. Low depression is associated with a reduction in secondary complications, such as pressure sores. The issues of altering negative appraisals and post traumatic growth are considered.

SYMPOSIUM 18

Disgust - The Forgotten Emotion?
David Veale, The Priory Hospital North London

Fear and loathing in OCD: Does Disgust Play a Role in This Anxiety Disorder?
Paul Salkovskis & K. Wahl, Institute of Psychiatry, London

Abstract not available

"Only a Wafer Thin Mint, Sir?" A Survey of Vomit Phobics.
David Veale. Royal Free Hospital and University College Medical School, London
Vomit is a prototypal stimulus for the emotion of disgust. Little is known about the psychopathology or treatment of vomit phobia with only a few isolated case reports. Vomit phobics are generally regarded as difficult to treat by exposure because of the complexity of obtaining credible stimuli, which can be repeated and prolonged. The aim of this study was to learn more about the psychopathology of vomit phobia, the degree of handicap and their experience of treatment. Patients with vomit phobia were compared to patients with agoraphobia and panic and non-patient volunteers on a number of measures to compare beliefs about nausea and vomiting, avoidance behaviours, safety behaviours, and symptoms of anxiety.

Disgust in Eating Disorders: Sensitivity and Emotional Responses
Nicholas Troop1, Tara Murphy2 and Professor Janet Treasure3
1 Department of Psychology, London Guildhall University, London
2 Department of Psychology, Institute of Psychiatry, London
3 Department of Psychiatry, Institute of Psychiatry, London
Disgust is an emotion of avoidance where the focus of core disgust is considered to be food. Despite this, however, very little attention has been paid to disgust in eating disorders, disorders in which issues concerning food and its avoidance are clearly of importance. This talk will present the results of a series of studies on the relationship between disgust sensitivity, disgust emotions and eating disorders.
Two studies are presented on disgust sensitivity and categories of disgust. Study 1 uses patients from an eating disorder clinic and Study 2 uses women with a history of eating disorders from a Research Volunteer Register. Despite the use of different measures and very different samples, the results suggest that women with a history of eating disorders do report higher levels of disgust sensitivity but that this is restricted to categories of disgust related to food and the body.
Two more studies are presented on reported disgust and fear responses to food, body and emotion stimuli. In Study 3, women with abnormal eating attitudes reported higher levels of both fear and disgust in response to high calorie foods and overweight body shapes than did women scoring low on abnormal eating attitudes but groups did not differ on emotional responses to drinks or slim body shapes. Study 4 replicated and extended these results in a group of patients with anorexia nervosa.
Overall these results suggest that disgust sensitivity may play a role in eating disorders. In particular, while eating disorders have traditionally been conceptualised in terms of fear (e.g. fear of weight gain, morbid dread of fatness, weight phobia), disgust may be an equally salient emotional response to "dangerous" foods and "undesirable" body shapes. It is proposed that this disgust response may facilitate the avoidance of such stimuli.

The Role of Disgust Sensitivity in Body Dysmorphic Disorder
Kate Cavanagh & Ann Stevenson, University of Glasgow
Body Dysmorphic Disorder (BDD) is characterised by preoccupation and torment elicited by an imagined or exaggerated defect in appearance (e.g. a large nose, thinning hair, facial scarring). Whilst classified in DSM-IV as a somatoform disorder, considerable evidence suggests BDD is better understood as an affective disorder within the obsessive-compulsive spectrum (K. Phillips, McElroy, Hudson & Pope, 1995). Whilst, there is a growing body of research into the nature and treatment of BDD, the causes of BDD remain largely unexplained and its vulnerability factors largely unexamined. One candidate vulnerability factor for BDD is sensitivity to disgust.
Recent research has linked the pathological experience of disgust to a variety of nominally anxiety and affective disorders, including animal and medical phobias, obsessive-compulsive sympomatology and the eating disorders (e.g. Davey, Tantlow & Dallos, 1998; Matchett & Davey, 1991; Muris, Merckelbach, Nederkoorn, Rassin, Candel & Horseleberg, 2000; Tolin, Lohr, Sawchuk & Lee, 1997). The idea that BDD (dysmorphophobia) may be linked to the emotion of disgust is not new (M.Phillips, Senior, Fahy & David, 1998), however this relationship has not been tested empirically. This paper presents the findings from two studies investigating the link between disgust and BDD in non-clinical populations. The first study 1 measured disgust sensitivity, anxiety and depression in a group of undergraduates who had completed a screening measure for BDD. In a second study the relationship between BDD symptom severity and disgust sensitivity, anxiety and depression was investigated in a second undergraduate sample.
Whether disgust plays a significant role in Body Dysmorphic Disorder is discussed within the context of the research findings.

Disgust, the Self and the Brain.
Mary L. Phillips and Maike Heining, Institute of Psychiatry, London..
The basic emotion disgust (literally, "bad taste") has been defined as "...something offensive to the taste". Objects of disgust include waste products of the human/animal body, violation of body borders at non-mouth points, animal-origin disgust, interpersonal contamination, and moral or socio-cultural disgust. Human lesion and functional neuroimaging studies, employing as stimuli standardised facial expressions, have demonstrated that the amygdala is critical to the perception of fear, and the insula and putamen are important for disgust. The insula is also important for perception of disgust depicted either as a facial expression, or as an emotionally-salient odour or flavour, and for pain.
Inappropriate disgust, and the complex emotions derived from disgust (e.g. shame and guilt), may underlie disorders of self and body image. Self-perception includes several cognitive processes, in particular, recognition of physical attributes as self (e.g. one's own face, and information presented in other sensory modalities), and the experience of specific emotions, both positive and negative (e.g. disgust) in response to these. Studies of split-brain patients, and those employing psychophyical, psychophsiological and neuroimaging techniques have indicated that self-recognition is associated with right and left prefrontal cortex, and limbic regions, including the insula. These findings indicate that similar areas are important for perception both of emotionally-salient information and the self. Abnormalities of self-perception, e.g. body image disorders, may therefore be associated with dysfunctional regulation by the prefrontal cortex of activity in brain regions important for emotion, and disgust, perception.

SYMPOSIUM 19

Is There a Role for Cognitive Behaviour Therapy in Bipolar Disorder?
Jan Scott, University of Glasgow

Social Cognition in Bipolar Disorder
Peter Kinderman, Reader in Clinical Psychology, University of Liverpool
Studies of social cognition in bipolar disorder and paranoid ideation have revealed both similarities and differences between the processes operating in manic and paranoid states. Both conditions appear related to defensive or self-protective responses to personal threat. Both conditions, for example, appear to be associated with abnormalities in self-concept and causal attributions. However there appear subtle differences in the manner in which these defensive processes manifest themselves. Further research, in apparently unrelated conditions, suggests that schematic processes, rather than psychological deficit states, may explain these patterns. A tentative model of these schematic processes will be presented. It is hypothesised that appraisals of potentially threatening events precipitate specific circular feed-back loops that are self-maintaining. This model will be supported and illustrated with qualitative case material from CBT and with a year-long series of daily records of 'positive and negative activation'. These data will illustrate schematic thought, specific appraisal processes, as well as the behavioural and affective consequences. The possible research and clinical consequences of models such as this will be discussed.

Teaching Bipolar Disorder Patients to Identify and Manage Early Symptoms of Relapse
Richard Morriss, Alison Perry, Nicholas Tarrier, Eilis McCarthy, Kate Limb
We conducted a single blind randomised controlled trial of teaching patients to recognise early symptoms of manic and depressive relapses and then seek early conventional treatment from psychiatric services. 69 patients with DSM-IIIR bipolar disorder were recruited with a relapse in the previous 12 months. The intervention was conducted in 7-12 sessions by a psychology assistant and involved in recognising early and late early warning symptoms of manic or depressive relapse. A treatment-seeking plan was devised involving three points of contact with psychiatric services. The intervention was introduced in addition to standard psychiatric care. The 25th centile time to first manic relapse was 65 weeks as opposed to 17 weeks (p=0.008). there was a 30% decrease in manic relapses, improved social functioning and employment. In contrast there was no effect on depression and higher doses of antidepressants were used in the intervention group. Problems in implementing such an intervention into routine practice will also be discussed.

Cognitive Theory and Therapy of Bipolar Disorders
Jan Scott, University of Glasgow
This paper will give a brief overview of 3 pieces of research on cognitive therapy for bipolar disorders. First, it will describe cognitive style in people at risk of episodes of mania and depression. Second, it will highlight similarities and differences in cognitive style of individuals with unipolar and bipolar disorders. Third, it will look at a pilot study of the use of cognitive therapy for individuals with bipolar disorders. In the pilot study, 42 individuals were randomly allocated to usual treatment or cognitive therapy plus the usual treatment. The group receiving cognitive therapy showed significant reductions in symptoms, improvements in quality of life and lower relapse rates over the one year after cognitive therapy as opposed to the year prior to cognitive therapy. The paper will also briefly describe ongoing work in this area.

Cognitive Therapy for Bipolar Affective Disorder: A Randomised Controlled Study: Preliminary Findings
Lam, D., Watkins, E., Hayward, P., Bright, J., Wright, K. & Sham, P., Institute of Psychiatry, London, UK

One hundred and three patients suffering from bipolar 1 affective disorder were recruited in a randomised controlled trial of cognitive therapy (CT) specifically designed for bipolar affective disorder. The study targeted bipolar patients who are vulnerable to relapses. All subjects had to be taking a mood stabiliser on recruitment. The control group received minimal psychiatric input, i.e. mood stabilisers and outpatients follow-up. The therapy group received up to twenty sessions of CT plus minimal psychiatric input. With the exception of age of onset of the illness, there were no significant differences between the two groups in terms of demographics or the number of previous bipolar episodes. At the end of therapy, intention to treat analysis revealed that the therapy group had significantly fewer bipolar episodes, fewer days in a bipolar episode, higher social functioning and better medication compliance. Moreover, subjects in the therapy group had fewer episodes of bipolar depression and number of days hospitalised. The therapy group also had significantly less fluctuation according to the Activation subscale of the Internal State Scale that subjects returned monthly. The therapy group had significant reduction in BDI scores over the six months. When the subjects who received inadequate treatment (fewer than six sessions) were excluded, the therapy group also had significantly fewer hospital admissions and fewer manic and hypomanic episodes. This study replicated our earlier pilot study.

SYMPOSIUM 20

Psychological Approaches and Sleep, Part 2: Recent Empirical Findings
Colin A. Espie, University of Glasgow

Attentional Processes in Insomnia: The Role of Monitoring the Environment and the Body for Sleep Related Cues.
Christina Neitzert Semler and Allison G. Harvey, Department of Experimental Psychology, University of Oxford
Introduction: Previous research has implicated heightened attention to and monitoring of body sensations in the maintenance of a range of psychological disorders. Monitoring processes have also been highlighted as of potential importance in the maintenance of insomnia. The present study was undertaken to investigate the following six types of monitoring among insomniacs and good sleepers: (1) monitoring of physical state/body sensations during the pre-sleep period for bodily signs consistent with falling asleep (e.g., slowed heart rate, loss of muscle tone, physical signs of 'drifting off'), (2) monitoring of the environment during the pre-sleep period for signs of not falling asleep (e.g., being able to hear a dog barking and for noises outside and inside the house), (3) monitoring of the clock during the pre-sleep period to see how long it is taking to fall asleep, (4) monitoring of physical state/body sensations on waking for signs of poor sleep (e.g., sore head, achy and heavy eyes), (5) monitoring of physical state/body sensations during the day following a bad night of sleep for signs of fatigue (e.g., heavy legs, sore shoulders), and (6) monitoring of the clock on waking to determine how many hours of sleep were obtained.
Methods: Eighty participants between 18 and 35 years of age (40 good sleepers and 40 insomniacs) completed an extensive semi-structured interview assessing for the presence, nature and consequence of each of the six types of monitoring. Measures included an assessment of the frequency of each type of monitoring, along with associated thoughts, emotions, and safety behaviors.
Results: The results indicated that insomniacs attended to time and noise in their environment more frequently than good sleepers during pre-sleep and reported more frequent attention to their bodily sensations on waking compared to good sleepers.
Insomniacs reported more negative thoughts, more negative emotion and greater use of safety behaviors as a consequence of monitoring. A path analysis indicated that frequency of attending to cues related to sleep was positively related to negative thoughts, and that negative thoughts in turn were positively related to both negative emotion and safety behaviors.
Conclusions: These findings suggest that heightened attention to and monitoring of body state and the environment for sleep-related cues during the pre-sleep period, on waking and during the day following a poor night of sleep may perpetuate insomnia.

Clock Monitoring in the Maintenance of Insomnia
Allison G. Harvey and D. Anne Schmidt, Department of Experimental Psychology, University of Oxford
Introduction. Attentional bias toward threat-related material has been implicated as a trigger to threat perception and excessive negatively toned cognitive activity across a range of psychological disorders. In the context of insomnia, clinical observation has implicated 'clock watching' during the pre-sleep period to be a trigger for excessive negatively toned pre-sleep cognitive activity. The present study was designed to index the effect of 'clock watching' during the pre-sleep period on sleep-onset latency, anxiety and worry about not getting enough sleep. There were three predictions: (1) that participants instructed to monitor the clock would take longer to get to sleep than participants instructed not to monitor the clock, (2) on the basis that one mechanism by which clock monitoring may be detrimental to sleep onset is that it triggers worry, we expected that participants instructed to monitor the clock would report that worrying about how long it was taking them to fall asleep interfered with getting to sleep more than participants instructed not to monitor the clock and (3) on the basis of previous findings showing that a simple cognitive manipulation can induce a 'state' of insomnia in good sleepers we reasoned that hypotheses 1 and 2 would hold for both good sleepers and insomniacs.
Methods. Sixty adults participated; 30 good sleepers and 30 insomniac. Good and insomniacs were randomly allocated to one of two experimental conditions. In the first, participants were instructed to monitor the clock during the pre-sleep period and in the second participants were instructed not to monitor the clock during the pre-sleep period. Sleep onset latency was measured by self-report and actigraphy. Worry was measured by self-rating.
Results. The results indicated that participants instructed to monitor the clock experienced longer sleep onset latency and more worry about falling asleep compared to participants instructed not to monitor the clock. These findings were observed across diagnostic status. Interestingly, instructions to monitor the clock lead participants to overestimate sleep onset latency relative to instructions to not monitor the clock. This finding may be explicable with reference to two factors. First, previous research indicates that time seems longer as the number of units of information processed per unit of time is increased. Monitoring the clock involves processing more units of information than not monitoring the clock. Second, detailed analysis of the actigraphic data revealed that in the first 60 minutes after sleep onset those instructed to monitor had more awakenings than those instructed not to monitor (p < 0.001). It is possible that these awakenings may have been perceived as continuous wakefulness.
Conclusions. Consistent with the predictions, the present study demonstrated that monitoring the clock lead to longer sleep onset latency and more worry about not getting to sleep compared to not monitoring the clock. These findings are consistent with previous clinical observations and support the hypothesis that clock checking may be involved in the maintenance of insomnia.

The Evolution of Sleep-onset Latency Problems: An Experimental Investigation of Pre-sleep Cognition and Attribution in People with Cancer
Lynne Taylor, Colin A. Espie and Craig White, Dept. of Psychological Medicine, University of Glasgow
This study examined the evolution of insomnia by investigating pre-sleep cognition and attribution, through the administration of the emotional Stroop task, to two groups of cancer patients who had developed sleep-onset latency (SOL) problems since diagnosis: 15 individuals with cancer 0-3 months from diagnosis (early group); and 18 individuals with cancer 12-18 months from diagnosis (late group). Consistent with the hypothesis, both groups demonstrated attentional bias for cancer-related words but only the late group demonstrated attentional bias for sleep-related words. High levels of pre-sleep cognitive arousal were evident in both groups despite lower levels of psychological distress in the late group. Findings are discussed within the context of the current literature and implications for future research are proposed.

Pre-school Children with Attention Deficit Hyperactivity Disorder: A Naturalistic Assessment of Activity and Sleep
Stephanie J. Inglis and Colin A. Espie , Dept. of Psychological Medicine, University of Glasgow
Researchers and clinicians in the field of Attention Deficit Hyperactivity Disorder have recognised the importance of sleep problems in children with ADHD. However, little is known about the circadian rest-activity pattern of the preschool child with ADHD. This study investigated the nature of sleep and activity in a sample of 12 preschool children fulfilling diagnostic criteria for ADHD relative to age and gender matched controls using both objective and subjective measurements. Results indicated that parents of ADHD children reported significant daytime over-activity and sleep disruption in their children. Objective evaluation using actigraphy failed to detect differences in daytime activity between the ADHD group and their matched controls. ADHD children were found to be more active during sleep than their comparisons; however, this was only apparent in a sub-set of ADHD children. No significant difference was found in sleep pattern or sleep quality parameter between the groups. There was high variability both between and within groups in activity and sleep measures and objective evaluation of activity and sleep failed to consistently differentiate children diagnosed with ADHD from controls. Various explanations are provided to account for these findings and implications for future research are discussed.

The Role of Cognitive Coping Styles and Sleep Hygiene in Acute and Chronic
Insomnia
Jason Ellis and Mark Cropley, Department of Psychology, University of Surrey
To date, little attention has been paid to the factors which may explain why some acute insomniacs progress to chronic whilst others do not. Recent advances suggest that the role of cognitive coping styles may be important. This research, although in its infancy, provides a comprehensive framework from which to examine the cycle of chronic insomnia. Chronic insomnia is considered to be a progressively heterogeneous disorder, which includes sleep incompatible practices and negative conditioning. This conceptualisation however questions the utility of a cognitive approach for the creation of a preventative intervention, as the link between cognitive and behavioural factors can not be delineated.
The present, cross-sectional study examined the relationship between different cognitive coping styles and sleep hygiene factors in self-identified acute and chronic insomniacs, as well as normal sleepers. The Thought Control Questionnaire, a measure of cognitive coping styles, and the Sleep Disturbance Questionnaire (SDQ) were distributed to 1104 participants from the general population, along with questions on demographic and sleep hygiene practices. The definition of insomnia, based on the DSM-IV recommendations, identified 162 participants with chronic insomnia, 146 with acute insomnia, and 304 normal sleepers.
A series of stepwise multiple regression analyses identified the role of punishment in acute and chronic insomnia for each insomnia subtype, whereas sleep hygiene was only significantly related to chronic insomniacs who scored highly on the Mental Anxiety dimension of the SDQ. Differences in the cognitive coping styles used by acute and chronic insomniacs, and normal sleepers, who scored highly on each dimension of the SDQ are also highlighted. Distraction, as a cognitive coping strategy was found to be a 'buffer' against chronic insomnia but not acute insomnia, further highlighting differences between these groups.
These results are discussed in relation to tailored preventative strategies to stop an acute period of sleep disturbance from becoming a chronic condition.

SYMPOSIUM 21

Developing Clinically Effective and Efficient Approaches for Routine Clinical Work.
Jim White, Clydebank Health Centre

Three Session CBT for Common Mental Health Problems: How Far Can We Get?
Dale Huey, Pauline Callcott, Gary Robinson, & Mark Freeston, Newcastle Cognitive & Behavioural Therapies Centre.
Trying to meet an increased demand for psychological therapy with limited resources is an international problem. Brief interventions, e.g. less than six sessions, may provide a partial solution for this problem. For example, the '2+1' model appears to be a potentially effective option for sizeable proportion of 'sub-syndromal neuroses' (Barkham et al., 1999). As part of a waiting-list initiative we offered clients on our treatment waiting-list the opportunity to participate in a brief intervention project. The main aims of the project were: to establish whether we were able to provide effective brief interventions for our 'syndromal' clients; to test-out a three-session protocol; to provide an accessible interim service for people waiting for therapy; and where appropriate to reduce the waiting-list. Twenty-nine clients completed three sessions of CBT. The Clinical Outcomes in Routine Evaluation questionnaire was completed at two points prior to intervention, prior to each session and post-session three. The Beck Depression Inventory was completed prior to intervention and post-session three. Participants also completed a satisfaction questionnaire asking for their views on aspects of the therapeutic relationship and changes as a result of the sessions. The BDI and CORE measures were completed again prior to a naturalistic follow-up appointment (around three-months later). The outcome data from the project will be reported (self-reported symptomatic change, client satisfaction, and proportion discharged), the three-session protocol described, and our reflections on suitability issues summarised.
Barkham, M., Shapiro, D.A., Hardy, G.E., & Rees, A. (1999) Psychotherapy in two-plus-one sessions. Journal of Consulting & Clinical Psychology, 67(2), 201-211.

Developing a Primary Care Service Around the Routine Evaluation of Clinical Outcome
Tony Turvey , Tayside Area Clinical Psychology Dept
An adult primary care service is described that has been developed from the findings of an eight year service audit (n=8,500), a one year follow-up survey (n=288) and a brief therapy project (n=30). The current service uses the HAD (Hospital Anxiety & Depression Scale) routinely and its correlation with to the SCL90r is reported. A pilot measure for predicting allocation of referrals to brief, routine or longer term therapy is described. Some strengths & weaknesses in the current service are discussed. Finally a method for quantifying the generalisation from research studies to routine service provision is presented that can help clinicians determine if their service to a more heterogeneous client group is as effective as might be expected based on research using more carefully selected client groups.

Offering Choice in the Community: CD-ROM vs Written Self-help vs GP Treatment
Jim White, Clydebank Health Centre, Ray Jones, Dept of Public Health, University of Glasgow
Given the significant imbalance between need and resources in the community, there is a need for mental health therapists to innovate existing services to provide readily accessible and easily understood approaches for common disorders such as anxiety and depression. Following a successful pilot study with a chronic, severe and highly comorbid heterogeneous anxiety disorder population*, this paper will describe a recently completed controlled trial. We compared a three session interactive multi-media touch screen CD-ROM treatment based in public libraries, a self-help written version (directed by practice nurse or research assistant) and GP treatment as usual. The computer and written versions were based on a well validated CBT self-help package - 'Stresspac' - written by JW. Clients meeting DSM-IV criteria were recruited from (mainly) socially deprived districts across Glasgow. A 'one plus two' format was used - one information and self-assessment session followed by two self-treatment sessions. Seven options were available - Controlling your - (1) thoughts, (2) actions, (3) body, (4) panic attacks, (5) sleep problems, (6) depression and (7) future. Personalisation was achieved by computer-patient interview and on-screen completion of the Hospital Anxiety and Depression Scale. Results to six month follow-up will be presented. The discussion will look at how to further develop these approaches and how they can help offer real choice to service users.
*White, J., Jones, R. and McGarry, E. (2000). Cognitive behavioural computer therapy for the anxiety disorders: A pilot study. Journal of Mental Health, 9, 505-516.

Short versus Long CBT for Clients with 'Severe and Chronic Difficulties': A Pilot Study
David Westbrook and Gillian Butler, Warneford Hospital , Oxford
The Oxfordshire Adult Psychology Department generally offers short courses of cognitive behavioural treatment to patients with a wide variety of psychological problems. Most patients have 10 treatment sessions, with an overall mean of around 13 sessions including assessment. In an evaluation of this service, there was some indication that patients whom an assessor thought had more severe and chronic difficulties were less likely to benefit from such brief treatment. There is also some evidence from the literature, although not consistent, that patients with personality disorders may benefit less from ordinary CBT. We therefore set up a small pilot study to examine (a) whether patients classified at assessment as having chronic difficulties by our criteria also met criteria for personality disorder, and (b) whether they would benefit more from longer treatment. Thirty such patients were randomly allocated to receive either 10 sessions or 30 sessions of CBT from experienced therapists, and they were assessed on a variety of measures through the course of treatment and up to a one year follow-up. The results of this pilot study will be presented, and their implications discussed.

Large-scale Self-confidence Workshops: A Pilot Study
June Brown, Institute of Psychiatry
In line with the NSF goal of improving the health of the population, large-scale (for up to 30 people per workshop) day-long Self-Confidence workshops were run and members of the general public given the opportunity to self-refer.
This study had two aims. One was to assess if 'Self-Confidence' workshops could be effective in reducing symptoms of anxiety and depression. The other was to establish if these workshops could reach those with hidden problems of depression. Previous 'depression workshops' had failed to reach this group This study therefore attempted to see if a different label - 'self-confidence' - could attract those with depressive problems and who may not have consulted their GPs.
The results were promising. 231 people enquired and 149 attended Introductory talks preceding the workshops. 113 went on to attend the workshops, with 59 in the experimental group and 54 in the control group.
Those who attended the workshops reported significant decreases in their anxiety and depression scores as well as increases in their self-esteem scores as compared to a waiting list control group at the 3 month follow-up stage.
These workshops were also successful in attracting those with depression but who had not consulted their GPs about these problems. Most attenders had mild and moderately severe problems of anxiety and depression.
It is therefore concluded that the results from this pilot study show that this line of enquiry is worth pursuing further.

Fast Access for Brief Interventions for Anxiety and Depression in Primary Care : A Pilot Study
Karina Lovell, David Richards, Peter Bower, Dianne Oliver, University of Manchester
A fast access self-help clinic for anxiety and depression has undergone a 1 year pilot phase. The clinic offers fast access for patients in primary care with anxiety and depression and offers brief evidence based interventions (usually via self-help books and manuals). An evening telephone clinic is also available for those patients not able to attend during the day. Patients are seen for a 30 minute assessment and offered 15 minute follow-up appointments usually on a 2-4 week basis. Such a system allows for between 12-17 contacts each session. In the course of the past 10 months 130 patients have been referred with an average of 3-4 new referrals weekly. Patients are usually seen within a week of referral and no waiting list has accrued. Clinical outcomes measures include the CORE, GHQ and satisfaction, and are administered at baseline, 3 and 6 month follow up. Qualitative interviews have been conducted with both GP's and users of the service. Preliminary results have found significant reduction in the outcome measures and initial qualitative analysis is demonstrating satisfaction with the interventions and fast access to the clinic from both users and GP's.

SYMPOSIUM 22

Domestic Violence
Raymond W. Novaco, University of California, Irvine

Working Systemically with Family Violence: Risk, Responsibility and Collaboration
Arlene Vetere, The Tavistock Centre and University of East London, and Co-director of "Reading Safer Families" and Jan Cooper, Co-director of "Reading Safer Families"
A risk assessment and risk management approach to working with violence in family relationships and the associated ethical problems will be outlined. Our aim is to ensure prevention and continued protection from violence for family members. Such violence often occurs in the context of intergenerational patterns of repeating violence, with children as both
victims and witnesses. In our therapeutic work with couples and families we identify three recurrent themes: collaboration, responsibility and risk. We try to develop a collaborative relationship with our clients using systemic processes of reflection, reflexivity and consultation. We establish responsibility for the violence and develop a no-violence contract with the perpetrators. Therapists, families, and referrers all share the risk of the therapeutic work, with the referral agency acting as a stable "third" in the therapeutic triangle, enabling risks to be
monitored in the rehabilitation of families in the aftermath of violence. An example from practice will be illustrated with a violence-focused genogram to show how we work systemically with violence in family relationships.

A Therapeutic Jurisprudence Approach in Legal Interventions for Domestic Violence: How do the Children Fare?
Kathleen A. Ham-Rowbottom and Raymond W. Novaco, University of California, Irvine, USA
Legal interventions can have a therapeutic purpose but may instead have an anti-therapeutic impact. Therapeutic jurisprudence suggests that the consequence of the law's application be compared against legal/judicial objectives so that the practice of the law, like that of medicine, strives first to do no harm, and ideally acts to improve the well-being of those who come in contact with it. Children involved in the legal system as a result of domestic violence tend to be a neglected constituency. Legal interventions, including police action, criminal court involvement, and civil action, and the very study of the effects of these interventions, focus on adult victims and abusers. Children exposed to domestic violence are largely served by courts through the application of a "best interest of the child" standard in custody and visitation decisions. Just as witnessing violence in the home can be detrimental to children, judicial inaction or misguided legal action can short-term and long-term adverse effects. Understanding the family context is imperative, especially to the extent that the consequences of intervention operate through the mother's well being. The therapeutic efficacy of legal interventions should be assessed for all parties involved. Illustrative data from a family court intervention project will be presented to highlight dilemmas bearing on decisions regarding child visitation.

Development of Networks for a Comprehensive Service to Support Adolescents Affected by Violence Within Families
Matthias Schwannauer, University of Edinburgh, Royal Edinburgh Hospital
The key dynamic that this research describes is how different ways of understanding domestic violence lead to different responses by service providers. These different responses clash with (or bypass) each other, and this leads to a fragmentation of service provision. Lack of dialogue and conflict between service providers mirrors the lack of dialogue, blame, and demonisation that exists in situations of domestic violence. Service providers appear to find it hard to collaborate and seek advice and support from other agencies, without turning the whole case over to to the other agency. A lack of clarity concerning what each agency is able to do, an absence of procedures whereby different service providers can take joint responsibility for a case, a lack of trust resulting from the lack of a shared understanding, means that dialogue is very difficult. Rather than working in isolation, or passing individuals between service providers (who often know little of each other or are in conflict with each other), the need for much better networking. This approach could be taken further by setting up a specialist neutral agency (e.g. A centre for family relationships) that could address the needs of the individual or family by drawing upon and further developing expertise of the different service providers. As it is, the current difficulty of achieving a healthy dialogue among service providers, mirrors the difficulties, blame, and lack of dialogue prevalent in situations of domestic violence.

Domestic Violence Exposure and Children's Behavioural Adjustment: Impact of Maternal Depression Before and After Shelter Residence
Kelly L. Jarvis, Erin E. Gordon, and Raymond W. Novaco, University of California, Irvine, USA
Mothers and children from domestically violent homes who have sought assistance from a battered women's shelter in Orange County, California, were studied in two contexts, as provided by two samples: (1) women and children currently residing in an emergency shelter, and (2) women and children who had previously resided in the shelter but were now living in the community. Mothers participated in structured interviews and completed psychometric assessments of their depression and trauma symptoms and of the child's recent behavioural functioning. Maternal emotional health was found to be associated with maladaptive behaviours in children, particularly with respect to older children. Maternal depression was associated with child internalising behaviours, particularly being withdrawn and feeling anxious or depressed; maternal trauma was associated with more child internalising and externalising behaviours, such as aggression. Our results suggest that efforts to intervene therapeutically with children who have been exposed to domestic violence ought to also address maternal emotional health. Major reductions in the subsequent violence exposure of the clients receiving the community-based intervention service will also be discussed.

Exploring the Nature of Motivation to Change in Two Samples of Domestic Violence Offenders Attending CBT programmes
Erica Bowen, School of Psychology, University of Birmingham
Patient motivation for treatment has been considered to be vital to psychotherapy progress and outcome for some time. The most frequently stated hypothesis is that those people who are self motivated to change ie who volunteer for treatment are the most likely to successfully alter their behaviour in comparison to those who attend prescribed intervention. This assumption has been fed by the development of the transtheoretical model of change (Prochaska & DiClimente, 1985) which has identified several independent stages of motivation to change - pre-contemplation, contemplation, action and maintenance. The goal of treatment is to move the individual from the position of acknowledging the problem behaviour, to actively seeking to change the behaviour and maintaining the altered behaviour. Tools developed to aid the identification of levels of motivation to change include the URICA - a non-specific motivation to change tool based on the identification of stages commensurate to those identified by the trans theoretical model of change. Although the majority of research in this area has focused on clinical populations, there is a rapidly expanding literature questioning the efficacy of pro feminist cognitive-behavioural domestic violence interventions in which arguments regarding the effect of motivation to change upon intervention outcome are rife. Despite this, to date there is no evidence that explores precisely differences in motivation to change between those offenders who are court mandated, and those who volunteer for treatment. This paper presents the preliminary findings of a study comparing the motivation to change levels as measured by the newly developed domestic violence specific URICA-DV (Levesque et al 2000) of domestic violence offenders before court mandated and voluntary treatment. The URICA- DV has been validated on a sample of court mandated batterers, but differences between mandated and voluntary samples have yet to be explored. Results are discussed in relation to the potential of using motivational interviewing as a pre treatment intervention for these two groups

SYMPOSIUM 23

Investigating Imagery
Emily Holmes, University College London & Ann Hackman, University of Oxford

Recurrent Images: Would Mirrors Do Well to Reflect Again?
Ann Hackmann, Department of Psychiatry, University of Oxford, Warneford Hospital
Recurrent, distressing images have been noted in a number of disorders. Clinical observation suggests that whilst these images are often present or future orientated they generally appear to have very similar content to memories of past upsetting experiences. Like intrusive memories in PTSD they frequently have situational triggers, and may involve not only the visual modality, but also other sensory modalities. They are often accompanied by emotion, and may also involve some response components. The person experiencing them is sometimes unaware initially of any connection between the images and past experience. Sometimes the images are so fleeting that although they affect emotions and behaviour they may not immediately be noticed by the individual. However, once brought into awareness distressing images do have face validity: to the person experiencing them they appear to reflect reality.
In this paper consideration is given to the idea that the imagery system provides information from memory to guide behaviour in the present. Acting without reflection in response to such images may be adaptive to some extent, but can also result in a lack of updating and contextualisation of the meanings originally given to events. Other autobiographical memories and knowledge may not be brought to bear once images are triggered. Implications of these observations for cognitive therapy will be explored. Accessing and reflecting on images and their meanings and origins may be a beneficial prelude to ushering in a more realistic view, by comparing image and reality.

What Parts of a Trauma Become Images? -Examining Peri-traumatic 'Hotspots' in Memory
Emily Holmes, Sub-Department of Clinical Health Psychology, University College London
Why do patients with Post-traumatic Stress Disorder (PTSD) have intrusive imagery of some moments of a trauma and not others? In exposure/reliving therapy patients describe their trauma in detail. Within this they note distinct 'worst moments' of intense emotional distress and reliving, known as 'hotspots'. Examining hotspots provides information about trauma memory and what happened at the time (peri-traumatically) at points that give rise to images. Dual Representation Theory (Brewin, 2000) predicts in general why intrusive images arise. That is, where conscious, verbally accessible memory for trauma is most disrupted, sensory and emotional memory can be triggered as images. Ehlers et al (in preparation) propose intrusions indicate moments with greatest emotional impact. Points of verbally accessible memory disruption could occur at the time of trauma due to dissociation (Holmes and Brewin, 2000), and intense emotion . In PTSD such emotions are commonly assumed to be fear, helplessness or horror. Hotspots analysis in 8 patients illustrates a wider range of emotions such as intense disgust, sadness and shame also occur (Grey, Holmes and Brewin, in press). Further data will be discussed indicating the types of cognitions and meanings associated with hotspots. These illustrate shattering positive assumptions, and confirming prior negative beliefs about the self (Ehlers and Clark, 2000). Reported peri-traumatic cognitions and emotions in hotspots indicate specifically why points have high emotional impact, and illustrate their diversity. However, it appears not all reported hotspots in a trauma memory are linked with current intrusive images in patients with PTSD.

The Course of Intrusive Memories During Cognitive Behavioural Treatment for PTSD
Anne E.M. Speckens, Department of Psychology, Institute of Psychiatry, London
Ann Hackmann, Oxford Cognitive Therapy Centre, Warneford Hospital, Oxford
Anke Ehlers, Department of Psychology, Institute of Psychiatry, London
One of the most characteristic features of post traumatic stress disorder (PTSD) are intrusive memories. We investigated intrusive memories in 55 patients included in two different trials investigating the effectiveness of cognitive behavioural therapy for PTSD. Patients had suffered from different types of trauma, such as road traffic accidents, physical or sexual assaults. They received 12 regular and 3 booster sessions of cognitive behavioural therapy. Frequency, distress, vividness and nowness of the intrusive memories were measured at baseline and at every treatment session.
At baseline, the mean frequency of intrusive memories was 5.6 (SD 6.1) a week. Frequency of intrusive memories slightly increased after the first reliving session, but steadily decreased over the number of treatment sessions after that. According to the frequency of their intrusive memories after reliving, patients were classified as dramatic responders, responders and non-responders. The only variable that was related to response to reliving was negative interpretation of symptoms. Patients who endorsed beliefs like: "If I cannot control my thoughts and feelings about the event I will go crazy" or "Something terrible will happen if I do not try to control my thoughts about the event" improved significantly more after reliving than the other patients. Response to reliving did not seem to be related to depression, anxiety, self-blame or anger.

Compulsive Images in Post-Traumatic Reactions
Padmal de Silva, Institute of Psychiatry, London
This paper discusses the phenomenon of imagery occurring in post-traumatic reactions, with special reference to compulsive images. Much is known about intrusive images in post-traumatic reactions; little has been written about compulsive images in those with a history of trauma. Data are presented from clinical cases of post-traumatic disorder and other traumatic reactions on the presence, and prominence, of compulsive imagery. In some, the compulsive imagery is the predominant symptom in the domain of imagery. Comparisons are made with compulsive images in obsessive-compulsive disorder. The significance of this phenomenon is commented on. Some comments are also made on treatment issues.

SYMPOSIUM 24

Do the Effects of CBT Endure? Long-term Follow-up (5-14 years) of Clinical Trials for Generalised Anxiety and Panic Disorder
Rob Durham, University of Dundee

Aims and Methodology of Long-term Follow-up of CBT Clinical Trials in Central Scotland
Julie Chambers, Department of Psychology, University of Stirling, & Department of Psychiatry, University of Dundee
A 5 year project to follow-up 10 clinical trials of CBT for anxiety disorders and psychosis in central Scotland is now at the half-way stage. The project has been funded by the NHS Executive Health and Technology Assessment Programme and is a collaborative venture between the Universities of Dundee and Stirling and several NHS Trusts (Tayside Primary Care, Fife Healthcare, Forth Valley Healthcare and Ayrshire and Arran Health Board). Extended follow-up is needed to investigate the scope and limitations of psychological treatment in changing the overall trajectory of mental disorders and to clarify the relative costs and benefits of CBT in comparison with medication and other forms of therapy. In this paper an overview of the methodology used in the project is presented and illustrated with reference to follow-up data from the first three trials (two on generalised anxiety disorder and one on panic disorder with or without agoraphobia).
In summary, attempts have been made to trace all 414 patients (including drop-outs) entered into these three trials and a total of 52% of patients available for contact have taken part in the follow-up study. A comprehensive picture of overall functioning at follow-up has been collected using a combination of standardised questionnaires (including SF36 II, BSI), structured interview (Diagnostic status, social adjustment and attitude to original treatment) and case note review. A number of methodological issues resulting from the lengthy period of follow-up (7-14 years in the case of the first three trials) have arisen of which the most important concerns the degree to which the follow-up sample is representative of the original cohort. Extensive analysis of the characteristics of participants and non-participants in the follow-up study has, in fact, revealed few differences although there has been a tendency for those who had completed initial treatment to be more likely to respond. Ethical issues concerned with obtaining access to medical case note data which might shed light on the status of non-participants at long-term follow-up have been actively debated by the research team and these issues are presented for open discussion.

Long-term Follow-up of Two Clinical Trials of CBT for Generalised Anxiety Disorder
Kevin Power, Department of Psychology, University of Stirling, & Tayside Primary Care NHS Trust
Generalised anxiety disorder (GAD) is a complex and variable condition which typically runs a fluctuating and often chronic course associated with heavy demands on healthcare resources. There is evidence that CBT can be an effective treatment in the short to medium term with 50-60% of patients achieving recovery. However, investigations of the long-term effects of CBT in changing the course of GAD are of limited scope and duration. Current follow-up studies consist mainly of postal questionnaire of symptom severity over relatively short periods (6-12 months). In this paper results are reported from an extended follow-up (8-14 years) of two clinical trials of CBT for GAD in which self-report data were supplemented by a structured interview with an assessor blind to initial treatment condition. Study 1 was conducted in primary care settings in the Forth Valley in a mixed rural and urban setting and compared Diazepam, Placebo, CBT alone, CBT plus Diazepam and CBT plus Placebo (Power et al, 1990). Study 2 was conducted in a psychiatric outpatient clinic in the City of Dundee and compared cognitive therapy, analytical psychotherapy and anxiety management training (Durham et al, 1994).
Data was collected on 55% of original Study 2 participants and 30% of original Study 2 participants. The relatively lower figure in Study 2 was almost certainly due to the longer period of time since the original trial and the absence of a central medical record for tracing patients. Despite the rather low overall contact rate the follow-up samples were broadly representative of the original cohorts. The results were broadly consistent across measures with marked differences in outcome between the two studies. About two thirds of Study 1 patients were substantially recovered whereas only about one third of Study 2 patients could be categorised in this way. Study 1 patients treated in primary care settings had less severe and less chronic problems at the time of the initial clinical trial and came from a more affluent area.. In contrast, Study 2 patients had more chronic and severe problems and came from a more economically deprived urban setting. Despite these differences the majority of patients (60-70%) in both studies reported varying degrees of improvement over the years with about 50% of the sample markedly improved and about 30% free of symptoms. For a minority of patients (30-40%) overall outcome was poor with evidence of continued disability, complex clinical presentations and moderate or considerable dependence on medication and/or psychological treatment over the follow-up period. In comparison with medication and analytic psychotherapy, treatment with CBT was found to confer an advantage in terms of significantly lowering the overall severity of symptomatology. This advantage, however, did not affect diagnostic status or the likelihood of recovery in terms of Jacobson criteria for clinically significant change. It is concluded that CBT does not presently have the power to alter the long-term course of GAD but it does have the power to make the disorder more manageable.

Long-term Follow-up of Two Clinical Trials of CBT for Panic Disorder, With or Without Agoraphobia, in Primary Care.
Donald Sharp, Institute of Rehabilitation, School of Medicine, University of Hull
Kevin Power, Tayside Clinical Psychology Department, & Anxiety and Stress Centre, University of Stirling
Vivien Swanson, Anxiety and Stress Centre, University of Stirling
Panic disorder, with or without agoraphobia, is a prevalent condition which presents most commonly in the primary care setting. The disorder places high demands on primary care treatment resources and effective treatments are therefore of considerable clinical interest. Previous research has shown efficacy for both psychological (usually CBT) and pharmacological treatments. Most treatment outcome research has however shown efficacy in the short term only with most studies reporting treatment outcome data at 6 months to one year. The efficacy over the longer term of CBT and pharmacological treatments for panic disorder is not well researched. Long-term efficacy is of considerable relevance to the applicability of such treatments in wider clinical practice. Results will be presented for the long-term follow-up of two trials of CBT in the treatment of panic disorder and agoraphobia in primary care. The first study comprised a controlled comparison of the antidepressant fluvoxamine and CBT, used alone and in combination, in the treatment of panic disorder and agoraphobia. The second study was a controlled comparison of CBT for panic disorder and agoraphobia delivered with standard therapist contact, minimum therapist contact, and as a bibliotherapy. Patients were followed up between 5 and 8 years post treatment. The overall efficacy of CBT in long term follow-up will be discussed as will the comparative efficacy of the CBT and drug treatments. Clinical implications of the findings will be discussed.

Health Economic Analysis of the Long-term Efficacy of CBT for Anxiety Disorders
Kirsten Major, Health Economist, Ayrshire and Arran Health Board
Economic evaluation is the comparative analysis of the costs and consequences of particular actions. As part of the long-term follow up of CBT clinical trials in central Scotland a cost effectiveness analysis is being conducted with a view to examining if CBT is an efficient treatment option across a range of diagnoses when compared to alternative regimes. The measure of effectiveness being utilised for this element of the study is the generic health status measure, the SF36. This paper will explore the resource implications associated with the first three trials analysed - two in GAD and one in panic disorder - as well as draw tentative conclusions on the cost effectiveness of CBT, based on initial findings. The method employed has been to collect data on levels of resource use across all health care sectors from general practitioner case notes. A deliberate decision was made to include non mental-health services given the likelihood of chronic use of wider health services associated with mental health morbidity. Data have been collected for 2 years prior to initial treatment and 2 years prior to the date of long-term follow-up. One hypothesis would be that successful treatment of mental disorder should be associated with a reduced need for all types of healthcare resources.
A number of factors have been raised by the analysis thus far. The first of these is the highly positively skewed nature of the data, such that a few individuals consume very high levels of resources, whilst the bulk consume fairly low levels. This is problematic for economic analysis, where there is a requirement to use means to summarise data to allow the calculation of budgets. Novel approaches to non-parametric analysis of means have therefore been pursued in an attempt describe the mean levels of resource use and the uncertainty which surround these estimates. A further complexity associated with the use of case note based measurement and valuation of resource use is the number of imputed values due to difficulties in accurate interpretation of items recorded in case notes. Finally, the measurement of costs, which include the costs of original treatment, have been compared for those receiving and not receiving CBT. The total cost has then been compared with the SF36 as a generic measure of outcome to assess if CBT treatments are more or less cost effective than alternatives. The results of this interim analysis from the study will be presented.

What Influences the Long-term Outcome of Generalised Anxiety Disorder? The Development of Prognostic Indices Based on Complexity of Presenting Problems and Response to Original Treatment
Dr. Rob Durham, Department of Psychiatry, University of Dundee
Dr. Mike Dow, Fife Healthcare NHS Trust and University of St Andrews
Andrew Gumley, Ayrshire & Arran Primary Care NHS Trust and University of Stirling
Although extended follow-up studies provide some evidence of the positive long-term influence of CBT on the severity of generalised anxiety disorder, it is clear that the overall course of anxiety disorders are influenced by a number of different factors. In this paper we first describe a general framework for conceptualising the influences on long-term outcome and then present results from an analysis of prognostic factors within this framework based on the cohort of patients from the central Scotland long-term follow-up study. The conceptual framework assumes that accurate prediction of long-term outcome requires knowledge of two sets of variables: (1) general prognostic indicators of the overall likelihood of change irrespective of treatment offered and (2) specific treatment response indicators of the individual's ability to engage with, and respond to, the demands of a particular therapy. General prognostic factors reflect the overall complexity and severity of presenting problems. Thus, there is persuasive evidence from both psychotherapy outcome research and epidemiological psychiatry that greater symptom severity and comorbidity, economic adversity (chronic indebtedness, low socio-economic status, unemployment) and social disadvantage (absence of intimate confiding relationships, chronic social difficulties and low support) are all associated with a poorer long-term outcome irrespective of the effectiveness of treatment over the short-term. Treatment response indicators, on the other hand, concern the quality of the therapeutic alliance, the power of the treatment offered and the degree of improvement over the course of therapy. The model assumes that general prognostic factors will be of most influence in determining long-term outcome and that if general prognosis is poor at the start of original treatment (i.e. complexity and severity is high), the long-term outcome will be either moderate (i.e. a partial recovery) if the original treatment was positive, or poor if the original treatment was ineffective. Conversely, if general prognostic factors are favourable then either a moderate or good outcome is predicted depending on the effectiveness of the original treatment.
Although general in character this model does have the merit of making testable predictions. It would be undermined, for example, by evidence that patients who were found to have recovered at long-term follow-up had a combination of poor prognostic factors and negative treatment response indicators at the original treatment trial. In the second part of the paper we present evidence on the validity of the model using logistic regression analysis with overall outcome defined in terms of diagnostic status and Jacobson methodology for clinically significant change. Finally, we discuss the clinical and ethical challenges to our current treatment technology of identifying at an early stage those referrals who are likely to do relatively poorly. In particular, we suggest the potential value of independent audit and clinical supervision becoming an essential part of routine clinical work with poor prognosis patients.

SYMPOSIUM 25

The Road Less Travelled: Evaluating The Effectiveness Of CBT Training
Anne Garland, Glasgow Primary Care NHS Trust.

Psychological Skills Training of Primary Care Practitioners: Report of a pilot Project Developed by Calderdale and Kirklees Health Authority.
Project Facilitator, Frances Cole; Trainers Jo Hardy, Vicki Dutchburn
The aim of training primary care practitioners in psychological skills was to improve the ability of primary health care team members to enable patients to manage their own psychological problems by using specific cognitive techniques.
Primary care provides the initial assessment and further management of 90% of patients with mental health care problems. This training project was developed around the concept of self help materials being used to supplement therapeutic encounters by mental health workers or in consultation in primary care. A meta-analysis of 40 self-help studies form 1974-1990 indicated that depression and anxiety are amenable to change using this approach. Cognitive behaviour therapy is an effective treatment depression and anxiety but is time and therapist intensive, a current problem locally with a lack of adequately trained therapists. Cognitive behavioural approaches using structured self help materials facilitated by trained primary care practitioners offers an alternative approach. This approach was evaluated in the pilot project by the Health Authority to ascertain its overall efficacy in changing practitioner skills.
The initial evaluation results provide evidence that the project is addressing real perceived needs by the practitioners. They have provided constructive feedback believing firmly that this type of training should continue because it;
is very relevant to their practice
is structured compared to previous woolly practice
is a safe training environment which allows increase in skills
is valuable for personal development
gives skills useful in other areas of work
is helpful in working with patients with multiple problems
The evaluation assessed the skills change by practitioners before and after training using videotape simulated consultations. There were skill improvements in use of problem based interviewing skills, suicidal risk assessment, negotiation skills and setting agendas. Other skill improvements were in collaboration with the patient resulting improved relationship building and ennoblement. Practitioners were taught to implement a cognitive based five areas model and results indicate its successful implementation in the simulated consultations.
Further qualitative evaluation is planned to assess the impact of skills training on the practitioner and the outcome of use of self help materials by patients.

Three Years on the Road: But Are We Going in the Right Direction?
*Mark Latham, Department of Health Studies, University of York and Chris Atha., Wakefield & Pontefract Community Mental Health Trust
Data from three courses will be presented, showing an evaluation of the effectiveness of a one year part-time introductory CBT course. Measures were taken (pre and post) of both clinical skill and knowledge levels, and rated by two independent blind assessors. Skills were measured on the Cognitive Therapy Scale and on the scale designed by the authors, via a videotaped role-play of an initial assessment interview. Knowledge was rated by means of a written questionnaire that was related to course content. Results showed statistically significant change in both aspects, thus indicating the effectiveness of the training.
The authors will raise several key issues in relation to their own work and to the wider field of CBT training and clinical practice including
How can the "active ingredients" of CBT training be identified?
Which competencies should be measured in evaluating CBT training?
How can tools be developed for evaluating effectiveness of CBT training?
How can evaluation be standardised nationwide across courses?
To what extent does CBT training influence clinical practice through improving outcomes?
Proposals will be put forward as an initial attempt to answer these questions. An invitation will be made to others in the field to join an open discussion of the important issues.

A Short Course in Cognitive Behaviour Therapy Techniques for Palliative Care Practitioners.
*Kathryn Mannix, Consultant in Palliative Medicine, Newcastle Marie Curie Centre
Anne Garland, Nurse Director, Glasgow Primary Care NHS Trust, Ivy Blackburn, Consultant Psychologist, Newcastle upon Tyne, Jennifer Gracie, Assistant Psychologist, Newcastle Marie Curie Centre, Stirling Moorey, Consultant Psychiatrist, Maudsley Hospital and Jan Scott, Professor of Psychiatry, University of Glasgow
Psychological distress is common amongst patients with advanced illness, with problems ranging from adjustment reactions to severe anxiety, panic and depression. Palliative care professionals are therefore often called upon to support patients with such psychological problems, and yet few have received training in techniques for psychological intervention beyond counselling training. Our study was designed to assess whether a brief course in CBT techniques would allow palliative care professionals to acquire skills, which they could apply to help their client group (patients and families).
25 palliative care professionals were recruited for the study. The majority were nurses working in the hospices or as clinical nurse specialists in palliative care teams. The group also included 2 social workers and 2 occupational therapists. All were naïve to CBT training.
The training team comprised 2 experienced cognitive therapists with experience of delivering CBT training (IMB & AG), a palliative care consultant who is a trained cognitive therapist (KAM) and a consultant in CBT with experience of delivering CBT to cancer patients (SM)
Core elements of Cognitive Theory and key techniques were identified by the training team.
9 days equivalent training was delivered as a 3 day block followed by 12 half days at weekly intervals, with homework assignments (skills practice) between sessions.
Trainees received a minimum of 3 months supervision in small groups following their training. Thereafter they were randomised to continue or discontinue supervision, in order to assess the effect of supervision on maintenance of skills.
Acquisition of skills was assessed using audiotapes of interviews provided by the trainees prior to training, at the end of 12 weeks didactic training, after 3 months with regular supervision, and 3 monthly for a further 6 months. Tapes were assessed blind by raters using a modified Cognitive Therapy Rating Scale.
Outcomes of training will be presented and the implications for developing structured, brief training in CBT techniques for non-mental health practitioners will be discussed.

Practical Psychological Skills for Community Teams: Accessible CBT Skills-Based Training Using the Overcoming Depression Format.
*Chris Williams, University of Glasgow and Anne Garland, Greater Glasgow Primary Care NHS Trust
Specialist training in CBT often involves the practitioner completing an expensive (in terms of time and cost) specialist postgraduate course. After completion of such courses, 71% of practitioners alter their job and many migrate into specialist CBT posts. The consequence is that trained practitioners often become separated clinically from general clinical services thus reducing the formal and informal dissemination of skills within these settings. A second difficulty that may prevent ready dissemination of skills to non-CBT specialists is that the traditional language of CBT is highly technical and inaccessible to those who have not received a similar training. This can cause problems when trying to work with patients/clients clinically where concepts such as dysfunctional assumptions and selective abstraction have to be "translated" as specific knowledge is required to understand the concepts. The language used not only affects our clinical work with patients, but also can affect the take-up of CBT skills across health service settings within both primary and secondary care.
We described the development and evaluation of training courses offering multidisciplinary team-based training in generic CBT skills for primary and secondary based community teams. The training course of six 3-hour sessions encourages attendees to apply what they learn in their own clinical settings. Generalisation of the training - to ensure that training does not become compartmentalised only within the teaching sessions - is emphasised as is clinical supervision embedded within the training sessions.
References.
Ashworth P, Wiliams C J, Blackburn I-M. What becomes of Cognitive Therapy trainees? A survey of trainees' opinions and current clinical practice after postgraduate Cognitive Therapy training.
Behavioural and Cognitive Psychotherapy, 1999, 27,3,267-277

Experiences of Palliative Care Professionals Receiving Cognitive Therapy Training
*Jennifer Gracie, Assistant Psychologist, Newcastle Marie Curie Centre, Kathryn Mannix, Consultant in Palliative Medicine, Newcastle Marie Curie Centre, Anne Garland, Nurse Director Glasgow Primary Care NHS Trust, Ivy Blackburn, Consultant Psychologist, Newcastle upon Tyne, Stirling Moorey, Consultant Psychiatrist, Maudsley Hospital and Jan Scott, Professor of Psychiatry, University of Glasgow
Twenty palliative care professionals participated in teaching and supervision in a study that investigated the efficacy of brief training in Cognitive Therapy interventions. The experiences the participants had during the training were illuminating, giving and insight into palliative care practitioners' achievements and problems when applying newly learned cognitive therapy techniques to their unique patient group. Outcome measures were administered at three month intervals and the participants were interviewed regarding their participation concurrently. Among the questionnaires administered were psychometric measures of occupational stress, job satisfaction and psychological well being. Subjective measures of job stress and satisfaction were taken, along with qualitative information regarding the participants' experiences of using their new skills. Their views on the teaching, supervision and participating in the study were also reported.

SYMPOSIUM 26

Basic and Applied Issues in Eating Disorders
Glenn Waller, St georges Hospital Medical School, University of London

Bulimics' Responses to Food Cravings: Is Binge-eating a Product of Hunger or Emotional State?
Anne Waters, Doctoral Programme in Clinical Psychology, Department of Psychology, University of Southampton

Introduction: This study examined the roles of hunger, food craving and mood of 15 women with bulimia nervosa. The participants used food intake diaries and Craving Records to self-monitor their nutritional behaviour, hunger levels and affective state.
Results: Cravings leading to a binge were associated with higher tension, lower mood and lower hunger than those cravings not leading to a binge. Levels of tension and hunger were the critical discriminating variables.
Conclusions: The findings of the study support empirical evidence and models of emotional blocking in binge-eating behaviour, and challenge the current cognitive starve-binge models of bulimia. The role of food cravings in the emotional blocking model is discussed in terms of a classically conditioned motivational state. Implications for treatment are addressed.

Social Anxiety and Coping Strategies in the Eating Disorders
Hendrik Hinrichsen. Eating Disorders Service, South West London and St George's Mental Health NHS Trust, Fiona Wright, Department of Psychology, University of Hertfordshire, Glenn Waller
Department of Psychiatry, St. George's Hospital Medical School, University of London, Caroline Meyer, Department of Psychology, University of Warwick
Background: A range of affective states are linked to eating pathology. There is some evidence that women with eating disorders experience high levels of social anxiety. However, it is not clear whether this form of anxiety pertains equally to all eating disorders, and how patients with different eating disorders cope with social anxiety.
Method: Women with eating disorders (N = 114) and a non-clinical group (N = 50) completed a standardized measure of social anxiety (Fear of Negative Evaluation scale; FNE), along with measures of bulimic and dissociative pathology.
Results: The clinical group had higher levels of social anxiety than the non-clinical group, although the binge-purge anorexia nervosa patients scored higher than those with either restrictive anorexia nervosa or bulimia nervosa. However, there were also differences across the groups in the use of emotional regulation mechanisms. Among the non-clinical and the bulimia nervosa groups, social anxiety was positively correlated with levels of bulimic behaviours. In contrast, the restrictive anorexics showed a correlation of social anxiety with dissociation.
Conclusions: Social anxiety is very prominent in the eating disordersin the binge-eating episodes of bulimic patients, and identified the critical factors involved in the processes surrounding binge-eating episodes that follow cravings.
Method: This was a prospective study of the binge-eating behaviour, and merits clinical attention in such cases. However, it will be important to understand how women with different eating disorders use different strategies to help them cope with that affective state.

Anger and Core Beliefs in the Eating Disorders
Rhonda-Jane Milligan, Department of Psychology, Royal Holloway, University of London
Michelle Babbs, Eating Disorders Unit, South West London and St. George's Mental Health NHS Trust , Caroline Meyer, Department of Psychology, University of Warwick. Glenn Waller, Department of General Psychiatry, St. George's Hospital Medical School, University of London
Background: Clinical and empirical research suggests a role for anger in the eating disorders, but this relationship is not well understood. This paper examines levels of different facets of anger in the eating disorders, and investigates whether those aspects of anger are related to the individual's core beliefs.
Method: A case series of 140 eating-disordered women and 50 non-clinical women completed the State-Trait Anger Expression Inventory (STAXI) and the short version of the Young Schema Questionnaire (YSQ-S).
Results: There were different levels of two specific forms of anger across the groups. The women with bulimic disorders had more pathological levels of state anger, but anger suppression was linked to both restrictive and bulimic disorders. Among the clinical group, there were also links of specific eating-disordered behaviours with particular aspects of anger. While women with eating disorders had generally higher levels of unhealthy core beliefs, those beliefs were related to a limited set of anger constructs. Unhealthy core beliefs were strongly associated with anger suppression in the clinical women, but not in the non-clinical group.
Conclusions: When working with the eating disorders and their constituent behaviours, it is important to treat anger as a complex construct. In particular, it is necessary to understand the role of state anger and anger suppression. Treatment of anger suppression in eating-disordered women may depend on modifying cognitions at the schema level.

Preconscious Processing: Impact on Body Percept and Concept
Glenn Waller, Department of Psychiatry, St. George's Hospital Medical School, University of London
Background: There is considerable evidence that body image is an elastic construct, which can be influenced by environmental and internal factors. However, knowledge of the factors that influence this elasticity is confounded by the demand characteristics inherent in the research paradigms that have been used to date. To overcome this problem, the present study used a visual subliminal processing paradigm, with the aim of determining the impact of preconscious processing of body-related cues upon body image (percept and concept).
Method: Forty non-clinical women were divided into those with more or less healthy eating attitudes. They completed measures of body percept (body size estimation) and concept (body satisfaction) before and after being exposed to very rapid presentations of fatness and thinness cues (the word 'fat' or 'thin', presented ten times at 4ms each).
Results: The women with relatively unhealthy eating attitudes were influenced by the fatness stimulus, with a worsening of their body percept and concept. In contrast, the women with healthier eating attitudes showed an improvement in their body percept in response to the thinness stimulus, although their body concept was not affected.
Conclusions: The findings support the centrality of body image schemata in eating psychopathology, although there is a need for replication and extension in clinical groups.

Family Work in the Treatment of Patients with Anorexia Nervosa
Gill Todd, Kay Gavan and Tara Murphy, Bethlem Royal Hospital, Beckenham, Kent
This paper describes a programme of family work undertaken as part of the treatment given to patients with anorexia nervosa in a specialist eating disorders service. The family work is a specific element in a multi-faceted in-patient treatment programme. The rationale for this work is the recognition that family attitudes and conflicts play an important part in the maintenance of anorexia nervosa, and that family members need advice, guidance and support. We describe a trial where some of the families are seen on their own, and other families are seen in groups of two or three families. The same therapeutic principles are used for both groups. The allocation is random. Four therapists are involved in the treatment. Therapy time is held equal across the families. Outcome measures include the patient's clinical state at the end of therapy and at follow-up, and changes in family attitudes and reactions.
The paper describes the rationale and the implementation of this programme; it also supports preliminary results. Issues arising in such family interventions are also discussed.

SYMPOSIUM 27

Issues of Cognition and Emotion
Gary Brown, Salomons Centre, Tunbridge Wells, Kent

Distinguishing Depression from Dementia in Later Life: Performance on the Emotional Stroop task.
Robert Dudley1 John O'Brien2 Nichola Barnett2 Liz McGuckin2 Peter Britton3
1 Newcastle Cognitive and Behavioural Therapies Centre, Newcastle.
2 Wolfson Unit, Centre for the Health of the Elderly, Newcastle
5. Department of Clinical Psychology, University of Newcastle.
Objectives: In later life, cognitive impairment is common in depression often making it difficult to distinguish a dementing illness from depression. We examined whether people with depression could be differentiated from those with dementia on their performance on a task that examines attentional bias to depression related material. Methods: Twelve older adults who fulfilled DSM-IV criteria for major depression were compared with 12 people with Alzheimer's Disease and 12 age matched controls on a test of cognitive biases: the emotional Stroop task. In this task participants were presented with words written in different coloured inks, and they had to name the colour the word was written in. Four types of material were presented Neutral, Positive, and Negative emotion words as well as a condition of meaningless symbols. Results: People with depression and those with Alzhiemer's disease were both slower than the controls on the task generally. However, the depressed group alone showed a statistically significant and specific increase in response time when colour naming the negative emotion words. The other two groups did not demonstrate such a pattern and colour named neutral, positive and negative words equally quickly. Conclusions: The biased processing of depression related material may have a valuable role in distinguishing depression from dementia in later life. Although the emotional stroop in its present form is not sufficient for such a purpose. Furthermore, the demonstration that older adults with depression exhibit such biases helps provide a theoretical basis for the application of cognitive behavioural treatments with older adults.

Stimulus Generalisation in Social Phobics and Healthy Controls
Antje Horsch, University of Oxford, Department of Psychiatry, Warneford Hospital
Christiane Hermann, Department of Neuropsychology, Central Institute of Mental Health, Mannheim, Germany
Despite the significance of conditioning models that aim to explain the etiology of phobias, there has been a great deal of criticism regarding their inability to explain why and how conditioned fear responses are generalised from the original traumatic situation to other similar situations. The aim of the present study was to test the generalisation of a conditioned fear response by comparing generalisation gradients of 20 social phobics (generalised subtype) and 20 healthy controls when confronted with fear-relevant or fear-irrelevant stimuli. The aversive classical conditioning experiment consisted of two phases. During the first phase (differential conditioning), the participants learned to discriminate between the conditioned excitatory stimulus (CS+) and the conditioned inhibitory stimulus (CS-): CS+ was followed by the presence of a mildly aversive electric shock (UCS), while CS- served as a safety signal through the absence of such a shock (no aversive UCS). During the second phase (generalisation test) individual generalisation gradients were obtained. Seven similar pictures, which differed from the original conditioned stimulus in only one specific feature, were presented to each subject along with the CS+ and CS-, and subjects were again asked to predict the outcome (presence or absence of mild electric shock). Our findings provide evidence that the learning of a fear-relevant conditioned inhibitory stimulus as a safety signal is delayed in anxiety patients. Moreover, they suggest that social phobics show a greater tendency to generalise the conditioned fear response towards stimuli that share similar features with the original traumatic stimulus, thus contributing to the development and maintenance of social phobia.

'It Has to Be Perfect': Anxiety Attitude and Belief Scale Factor structure, Psychometric Properties and Validity in an Anxious Population
Nick Hawkes, Department of Psychology, Birkbeck College, University of London, UK, Gary Brown, Salomons: Cantebury Christchurch University College, UK and Anne Richards, Department of Psychology, Birkbeck College, University of London, UK
The Anxiety Attitude and Belief Scale (AABS; Brown et al, 2000) aims to index a stable, long-term, cognitive vulnerability to anxiety and anxiety disorders. Its factor structure and psychometric properties were examined in a sample of anxious people (N=108). An exploratory factor analysis identified three theoretically interesting factors: external catastrophising, vigilance-avoidance and internal catastrophising. The internal reliability of the scale and its factors were high. The AABS predicted scores on measures of psychiatric distress, anxiety and insomnia (GHQ28; Goldberg and Hillier, 1979) and negative affect (PANAS; Watson et al, 1988), adding incrementally to information from a measure of Anxiety Sensitivity (ASI-R; Taylor and Cox, 1998). The AABS internal catastrophising factor was closely related to the ASI-R. Qualitative data from the sample, based on first hand experience of significant anxiety, was used to suggest future improvements to the AABS.

"I Have to Wash Until it Feels Right.' An Experimental Study Investigating the Use of Inappropriate StoppingCcriteria in Obsessional Washers.
Karina Wahl, University of London; Paul Salkovskis, PhD, University of London; Imogen Cotter, D.Clin. Psych, University of Oxford
Recent modifications of CBT suggest that obsessional clients might rely on inappropriate stopping criteria in order to decide when to stop a compulsive action. These criteria are believed to be at least partly responsible for the repetitiveness and prolongation of a compulsion. A previous study from our group had found that obsessional washers were more likely to use subjective criteria in the decision to stop a hand wash compared to a group of non-washing obsessionals and non-clinical controls who used more external criteria. The obsessional washers also considered a greater number of criteria before they reached a decision and the decision making process required more conscious control compared to the control groups. The current study investigates whether the use of inappropriate stopping criteria can be observed during an experimental hand wash and whether the criteria are linked to the length of the hand wash. 40 obsessional hand washers, 40 non-washing obsessionals and 40 non-clinical controls were asked to wash their hands in one of two conditions: after covering their hands in floor wax that allegedly comes from a) an old tin that has been used in a hospital or b) a brand new tin. The use of inappropriate criteria was expected to be more prominent in the 'hospital condition' with the greatest effect for the obsessional washers. Results are presented and discussed.

A Cognitive-Emotional Model of Desire
David J. Kavanagh, Department of Psychiatry, University of Queensland, Australia.
Jackie Andrade and Jon May, Department of Psychology , University of Sheffield
Recent mode1s of craving still do not account for the full range of empirical data. A general model of motivation and desire is described, in which two types of appetitive cognitions are distinguished: intrusive thoughts and elaborations. Intrusive thoughts are fleeting and require few cognitive resources, and represent the usual gateway to desire cognitions. Elaborative processing typically involves the construction of appetitive: images. Both types of thought increase the probability of subsequent consumption, but they operate within a causal model with additionalfactors including other incentives, coping skills, self-efficacy and target availability . Both intrusivethoughts and elaboration are initially pleasurable, but delay in consumption makes them aversive. Attempted suppression of desire-related thoughts increases their intrusive quality. The paper briefly illustrates how the model accounts for existing research and offers new opportunities for intervention.

ROUNDTABLE DEBATES

ROUNDTABLE 1

Analogies, Stories, Metaphors: Tools for Clinicians
Margret Hovanec, Private Practice, Toronto Canada

All of us use unique metaphors, stories, images and analogies in psychotherapy. In this Roundtable, experienced clinicians will share the interesting and amusing tools they have found helpful in clinical work. Clinicians in the audience will be encourage to offer their own experiences with analogies, stories and metaphors.

Metaphor and Meaning.
Gillian Butler, Department of Psychology, Warneford Hospital, Oxford.
Metaphors can provide an extremely efficient way of encapsulating meanings. The way that they work enables therapists to use them in many different ways. Working with people with long term, complex difficulties, most of whom have suffered abusive experiences in childhood, I have found that metaphors can be helpful at many different stages of therapy: when helping people to talk about painful, avoided, memories; when putting together a formulation so as to apply the theory to the case; when developing an understanding of processes that maintain problems or reactions to problems, and in the many other aspects of therapy that focus more specifically on bringing about change. Illustrations of the metaphors used with this client group will be provided with the aim of helping us to think more clearly about how we use metaphor, and to raise questions about how it works, or does not work.

Metaphor, anecdote and analogy in the treatment of obsessive-compulsive disorder
Norma Morrison, Oxford Cognitive Therapy Centre.
Central to the cognitive-behavioural approach to the treatment of obsessive-compulsive disorder is the need to target the individual's appraisal of the intrusive thoughts rather than the thoughts themselves. These appraisals are based on the patient's particular interpretation of what the occurrence and content of these thoughts mean to them and on their background assumptions about responsibility, need for control and overestimation of danger. As these assumptions and appraisals are often longstanding and firmly held, it is useful to have some extra tools to help the patient see them from a new perspective. Metaphors, anecdotes and the use of humour can provide these tools which facilitate a new perspective, make therapy more understandable and increase the likelihood of the patient remembering important information. Ways of using these to demonstrate how the strategies used to cope with the problem actually maintain it and to help modify appraisals and assumptions will be discussed.

Using Stories, Metaphors, Pictures and Music to work with Core Beliefs and Assumptions in Cognitive Therapy.
Ann Hackmann, University Department of Psychiatry, Oxford
Verbal discussion and behavioural experiments can be helpful when working with core beliefs and assumptions. However, where these are strongly entrenched it may be difficult to bring about cognitive change accompanied by a real shift in affect and meaning. Experiential tools such as stories and metaphors can tap into meaning at the implicational level, and suggest new perspectives to the patient. This can help develop a meta-perspective from which individuals can reflect on their problems. Examples of the use of such material in cognitive therapy for people with long standing interpersonal issues will be provided in this presentation

Listening to Rumours and Gossip: Information Processing in Anxiety Disorders
Margret Hovanec Private practice, Toronto, Canada
Many people have a vicarious attachment to listening to rumours and gossip. Whether they gather bits and pieces of information from friends or tabloids they combine fragments into plausible but frequently sensational and false scenarios.
This presentation poses the possibility that patients with anxiety disorders are analogous to people engaged in gossip gathering. Specifically with regard to health anxiety, patients are inclined to listen to their bodies and symptoms and reach the same conclusions that those who listen for the sensational in unrelated pieces of information. If the cardinal feature of anxiety is believing in mis-information, then people suffering from anxiety disorders can be seen to listening to their bodies looking for ominous, reinforcing signs of sensational diseases and conditions.
Differences in rumours and gossip will be explored.

Spontaneous Therapeutic Analogies
Lorna Tener, Private practice, Ottawa, Canada and Margret Hovanec, Private Practice, Toronto, Canada
Spontaneous analogies are a creative and useful way of deepening understanding and improving alliance at many points in therapy. For example, patients frequently bring to sessions information about themselves or their predicaments that they don't fully understand. The therapist can improve the richness of the summary or feedback to such patients by creating analogies, drawn from traditional or contemporary sources, from the patient's areas of expertise, or from the therapist's imagination. This presentation will illustrate the use of analogies to facilitate case conceptualization, goal setting, accessing automatic thoughts, challenging dysfunctional thinking, constructing & evaluating behavioural experiments, and teaching factual knowledge. The use of analogies may be a helpful skill to teach patients as well as supervisees. The examples described here are intended to "prime the pump" and lead to contributions from the audience.

ROUNDTABLE 2

Children and ADHD, Alternatives to Amphetamines

In Memory of Professor Stephen Baldwin

ADHD is the most frequently diagnosed disorder of childhood (Kazdin 1999), Despite 100 years of clinical investigation and research there are few outright solutions to the problems of hyperactivity and impulsivity in children. Practitioners in the field need a clear direction with the provision of proven effective treatment. This roundtable aims to provide practitioners with at least two effective alternatives to amphetamine therapy for children/teenagers with ADHD, Roundtable participants are encouraged to plan and provide more ideal solutions to ADHD in their own locality,
The roundtable will include reviews of effectiveness of chemical treatments for ADHD; the MTA study, Patterson's social learning approach (1971) and Stein's Caregiver Skills Program (1999).

ROUNDTABLE 3

Supervision in Cognitive-behaviour Therapy: Who, What and How?
Mark Freeston, Director of Training and Research, Newcastle Cognitive and Behavioural Therapies Centre and University of Newcastle

The current popularity of cognitive-behaviour therapy among clinicians and government initiatives that increase the demand for CBT among management has created an unprecedented demand for supervision. With the number of courses, both short and long, springing up all over the UK, and growing demand for CBT input to various training initiatives such as PSI, there is an increasing pressure on the capability to deliver supervision. Compared to the ever-increasing literature on CBT for a wide variety of disorders, there is very little written on supervision in CBT. This roundtable addresses the question of supervision from a number of perspectives. The first part of the round table will allow speakers to present briefly from each of the perspectives. The second half will invite discussion involving panel members and the audience.

Current Supervision Practice Among BABCP Accredited Therapists.
Michael Townend, Assistant Programme Director, Community Mental Health, University of Birmingham
This paper will report on a national survey of supervision practice among BABCP accredited therapists. The survey addresses who currently receives supervision, from whom they receive it, the nature of the supervision they receive, and whether dual relationships interfere with supervision.

Supervisor Training.
Anni Telford, Head of Counselling & Psychotherapy, University of Derby
This paper will address formal supervisor training, an endeavour that provides as many challenges as training therapists themselves. The training is based on the idea that supervisors must work from a supervisory framework rather than simply adopting strategies from the CB model. As a group of people who perceive themselves as expert practitioners, supervisors must understand organisational influences on both themselves and trainees, negotiate different relationship skills from those employed in therapy, and identify and respond to different levels of supervisee competence and confidence.

Organising Internal Supervisor and Accreditation.
Joe Findlay, North Tyneside, and Northumberland NHS Mental Health Trust
This presentation will report on experiences in meeting the challenges of rapidly expanding CBT services within the former Northumberland MHT. The Trust greatly increased the number of CBT-trained staff both at foundation and advanced level over the last 8 years and plans are for further expansion. Supervision during and following training has been a priority. The results of planning on organisational issues and internal accreditation of supervisors will be discussed.

Who Supervises the Supervisors?
Peter Armstrong. Clinical Nurse Specialist, Newcastle Cognitive and Behavioural Therapies Centre
This paper will address the particular challenges of supervision that may arise for experienced CBT therapists, most of whom are in fact themselves supervisors. Although supervision models are clearer when there is a substantial experience gap between the supervisor and the supervisee, it is less clear what are the appropriate models when there is little gap. Based on a series of in house discussions at the NCBTC, common problems and potential solutions will be discussed.

Supervision in CBT: The Whys and Wherefores?
Melanie Fennell, University of Oxford
This presentation will consider some fundamental questions about supervision in cognitive behaviour therapy. What is the purpose of supervision? What should cognitive therapy supervisors be aiming to achieve? How can supervision enhance learning by echoing the empirical collaboration that is at the heart of cognitive therapy? What makes a good cognitive therapy supervisor?

ROUNDTABLE 4

Voluntary Self-regulation or Statutory Registration:
What Does it Mean for You?

This session will include two short presentations by Chris Williams (President-Elect of BABCP and BABCP's representative on the Alderdice Advisory committee) and Francis Lillie (Chair of the UKCP sub-committee and on the Governing Board of UKCP). The session will provide an update on definite moves by the Government towards the regulation of Psychotherapy and Counselling, and offer you the opportunity to discuss the issues involved. In addition, Amanda Cole, Chair of the BABCP Accreditation and Registration sub-committee will be available to discuss current accreditation arrangement, and moves towards the voluntary registration of clinical supervisors.

OPEN PAPERS

Cognitive Therapy and the Elderly. What is the State of the Evidence and What Awaits Investigation?
Paul L Merrick, Massey University at Albany, Auckland, New Zealand
As with other age cohorts, Cognitive Therapy has begun to look promising as an efficacious psychological treatment for a number of disorders in older people. Investigations in the 80's provided a platform for a series of controlled studies, that promised positive outcomes for mild to moderately, depressed elderly. A smaller body of evidence also suggests that, as with younger people, Cognitive Therapy may be helpful in the management of some anxiety disorders.
This paper will briefly canvass the evidence in support of Cognitive Therapy as a useful therapeutic approach in the management of mood and anxiety disorders in the elderly. Reasons for the relative paucity of findings in the past 10 years with this age group despite the projected large growth in client numbers will be explored and directions for future research will be raised. Finally the importance of incorporating modifications in treatment delivery and research for this age group will be highlighted.

A Group Rehabilitation Approach to Chronic Fatigue Syndrome
Vanessa Russell, Christine Atkinson, Bob Lewin, Andrea Gaston, Peter Campion, Hull and East Riding Community Health NHS Trust, Department of Psychology, Victoria House, Park Street, Hull, HU2 8TD.
Chronic Fatigue Syndrome (CFS) is characterised by prolonged disabling fatigue of at least 6 months duration for which no medical explanation can be given. This fatigue is exacerbated by once tolerated physical activity and affects both physical and mental function. It is usually associated with a range of other difficulties, which may include sleep disturbance, painful limbs and joints, sore throat and headaches. Research also suggests considerable overlap between CFS and anxiety and depression (Chalder et al 1996). A number of research studies have examined the efficacy of cognitive behavioural approaches in the management of CFS, with mixed results. Three randomised controlled trails have found beneficial effects of using the cognitive behavioural approach (e.g. Sharpe et al 1996; Deale et al 1997; Fulcher and White 1997) whilst two have shown no benefit (Lloyd et al 1993; Friedberg and Krupp 1994). Chalder et al (1997) have demonstrated the efficacy of a self-help approach to the management of CFS. They found that the provision of a self-help booklet with a cognitive behavioural emphasis and specific advice from a research nurse was more effective than no treatment in improving fatigue and psychological distress. There are limited references in the literature concerning the evaluation of group approaches in the management of CFS. Pemberton et al (1994) described a group rehabilitation approach, which was not based on a particular theory but focused on helping patients overcome dysfunctional behaviours and the personal consequences of CFS. They cite the power of group facilitation as a positive aspect for people with CFS. This paper reports the initial findings of a multi-disciplinary group rehabilitation approach to the management of CFS. The group entails the use of cognitive-behavioural approaches, including graded resumption of physical and mental tasks and the challenging of attitudes that may reduce the likelihood of recovery. Measures used include the Fatigue Scale (Chalder et al 1992), Handicap Scale (Harwood et al 1994), Automatic Thoughts Questionnaires (Hollon and Kendall 1980, Ingram and Wisnicki 1988) and the HAD (Zigmond and Snaith 1988). A cross-over design is employed to allow comparisons between groups who are on the waiting list with those already involved in the rehabilitation group. One year follow up data is also reported.

Cognitive Style of Depression
Viktoras Keturakis , Department of Clinical and Social Psychology,Vilnius University, Lithuania.
Cognitive style concept yielded big amount of research and publications. Clinical view of depression appeared to be non-homogeneous cognitively Cognitive style may be applied differentiating depressive client cognitively employing concept of field dependency and checking the hypothesis against recognized notion of attribution style. 114 subjects participated at the research, 59 from control group and 55 experimental(depression) group. There were used Raven Progressive Matrices test to control eductive ability variable, Group Embedded Figures Test, House-Tree-Person drawing test plus cognitive style scoring system, Attribution Style Questionnaire, Sociotropy-Autonomy-Scale. Preliminary analysis conform the hypothesis of heterogeneity of depression along the dimension of field dependency. Negative attribution style relates more prominently and significantly to depression than Positive attribution style. Field dependency is not related with eductive ability when comparing among depression and control groups. Cognitive style being heterogenic concept cannot be related linearly to occurrence of depression. So we can distinguish field dependent and field mindependent styles present in the depression.

Effect of Cognitive-behavioral Therapy on the Delusion in Patients with Schizophrenia
Young-Chul Chung, M.D.,* Jae-Hyun Kim, M.D.,** Hong-Bae Eun, M.D.,* Ik-Keun Hwang, M.D.*
*Department of Psychiatry, School of Medicine, Chonbuk National University, Chonju, Korea
**Department of Neuropsychiatry, School of Medicine, Wonkwang National University, Iksan, Korea
Objective: The effect of cognitive-behavioral therapy on the delusion in patients with schizophrenia was evaluated. Method: The patients admitted to psychiatric ward from September 1999 to June 2000 and diagnosed as schizophrenia, schizophreniform disorder, and schizoaffective disorder by DSM-IV were randomly assigned to cognitive-behavioral therapy group(n=9) and supportive psychotherapy group(n=8). During the 10 weeks' therapeutic period, conviction, preoccupation, and anxiety of delusion, explanatory mode about symptom, and recovery style were regularly measured and compared between the study groups. Results: 1) As for conviction and anxiety of delusion, both groups showed gradual reduction over time but there was no significant difference between the two groups. 2) As for preoccupation of delusion, both groups showed gradual reduction over time and CBT group had a significant reduction compared to ST group. 3) As for explanatory mode and recovery style, CBT group had more marked positive changes compared to ST group. Conclusion: Cognitive-behavioral therapy is effective on preoccupation of delusion, explanatory mode about symptom, and recovery style compared to supportive psychotherapy.

Clients' Perspectives of Change Processes in Cognitive Therapy
Helen Clarke, Anne Rees, & Gillian Hardy, University of Leeds and Leeds Mental Health and Community NHS Trust
There is substantial evidence demonstrating the usefulness of cognitive therapy for depression. However there has been little attention given to clients' experiences of therapy, and how clients' might help understand the processes through which therapeutic change is achieved. The present study reports on the analyses of five end-of-therapy evaluation interviews with clients who had received between 12 and 20 sessions of cognitive therapy for depression. One client was selected from each of the therapists working at a jointly managed health service and university research clinic. All clients achieved reliable clinical change on the Beck Depression Inventory at the end of treatment.
Clients were interviewed using the Client Change Interview approximately two weeks after the
end of therapy. The Client Change Interview is a semi-structured interview and assesses three aspects of the clients' experience of therapy: a) changes perceived by client s that have occurred over the course of therapy; b) clients' understanding of the sources of these changes, including helpful aspects of therapy; c) hindering or difficult aspects of therapy. The interviews were transcribed and then analysed using Grounded Theory analysis.
Three categories emerged concerning changes clients attributed to therapy. These were
Observable Changes and were often changes that other people had commented on. Example
statements include 'Everyone says how much better I look' and 'My mother sees a massive difference in my behaviour'. The second category comprised Cognitive Changes noted by clients: 'I've altered my attitude towards myself, and '1 internalise everything less'. The final category! " comprised statements such as '1 feel easier with myself as though I don't have so much to carry' and is summarised in the statement 'Letting Go of the Illusion of Control'.
Clients descriptions of how they thought therapy had helped formed two core categories, and
included difficult aspects as well as positive aspects of therapy. The first comprised the Big Idea and included categories of CT techniques ( 'The one I'm still using is the thought diary'), ('..that one little model stayed in my head throughout'), testing things out ('Work that I did
between sessions was just as productive'), and patterns or core beliefs(' A realisation within
myself.of my old beliefs').
The second core category was labelled General Therapeutic Principles. These principles included Resistance ('1 was a bit dubious at first because I didn't trust the methods'), Engagement (Being sort of excited, and I got very into it'), and ~ ('1 never felt that she switched off once, I always felt she was listening').
This study highlights clients awareness of the importance of both specific and non-specific therapeutic factors that promote client change. One case is presented in detail as an example of a client's view of the change process. Interestingly the changes the client describes map onto the levels of change described in the assimilation model.

POSTER SESSIONS

POSTER SESSION 1

A Social Skills Group for Children: The Importance of Liaison with Parents, Teachers and Professionals
Estelle Macdonald, Uttom Chowdhury, Jackie Dabney, Miranda Wolpert and Samuel Stein Bedfordshire and Luton Community NHS Trust
Due to a recent increase in referrals of children with numerous social communication difficulties the trust decided to set up groups, using a cognitive behavioural framework, to teach children and adolescents various forms of social skills. We describe one particular group and highlight the importance of liaison with professionals and parents to ensure that adequate transfer of skills takes place.
Seven children (5 boys and 2 girls) were referred to the group. Their parents were invited to a parallel parents group. Each group lasted for 90 minutes and ran weekly for six weeks. Teaching of social skills was based on cognitive behavioural techniques. Extensive liaison with parents and teachers took part before, during and after the group in the form of meetings, telephone calls, letters and homework given to the children. Pre and post group questionnaires were sent to parents and teachers for evaluation purposes. Qualitative analysis of the results indicated that parents valued the on-going dialogue with professionals and often felt understood and supported by therapeutic professionals for the first time. Therapists' contact with teachers ensured a complete picture of the child's behaviour was obtained and that some of the skills learnt were transferred to the school setting. We, thus, argue that the use of cognitive behavioural techniques with children undergoing group work requires an intensive level of liaison in order to facilitate generalisation of skills learnt.

The Prediction of Delusional Ideation In College Students
Yoshihiko Tanno, Sachiko Morimoto and Shinji Sakamoto 1) Department of Life Sciences (Psychology), Graduate School of Arts and Sciences, The University of Tokyo, JAPAN. 2) Department of Human Relations, Otsuma Women's University.
The purpose of the present study is to test the diathesis-stress model of delusional ideation in college students and to predict the development of delusional ideation. The diathesis-stress model is a popular model in the study of the development of psychopathologies, such as schizophrenia and depression. People with a strong diathesis are more likely to show symptoms of psychopathology than those with a weak diathesis, when they encounter stressors.
Questionnaires were administered to college students three times, at fortnightly intervals, in a longitudinal study design. At time 1, diatheses of persecution ideation were assessed using 18 subscales. The subscales were selected from the Self-As-Target Bias Questionnaire (Fenigstein, 1984), Paranoia/ Suspiciousness Questionnaire (PSQ; Rawlings & Freeman, 1996), Auditory Hallucination Experiences Scale (AHES; Tanno, Ishigaki & Morimoto, 1998) and Evaluative Beliefs Scale (Chadwick, Birchwood & Trower, 1996). At time 2, persecution ideation was assessed. At time 3, persecution ideation and stressors experienced between time 2 and time 3 were assessed. Persecution ideation was measured using the Persecution Ideation Scale (Tanno, Ishigaki & Sugiura, 2000) in which an example item was "Thoughts that someone may intend to entrap me." Stressors were measured using the College Life Experience Scale (Hisada & Niwa, 1987).
The data from 117 college students were analysed. A set-wise hierarchical multiple regression analysis (Cohen & Cohen, 1983) was used which could predict the change in persecution ideation between time 2 and time 3. This prospective research design allows us to analyse the causality of persecution ideation and diathesis.
A set-wise hierarchical multiple regression analysis showed significant interactions between stressors and four diathesis subscales: a) subscale of Anger/Impulsiveness in the PSQ, b) subscale of Resentment/Perceived Hardship in the PSQ, c) subscale of Auditory Illusion in the AHES, and d) subscale of Hearing Negative Voices in the AHES. These interactions could significantly predict the increase of persecution ideation from time 2 to time 3. The direction and form of the interactions were in keeping with the diathesis-stress model.
The present results predict that students who have high scores for these four diatheses will show increased persecution ideation when they encounter stressors. Where the development of delusional ideation is predicted, we may be able to prevent the high-score students' increase in persecution ideation by using psycho-education methods.

Use of Millon's MCMI-III to Assess Frequency of Personality Disorder in Referrals to a Primary Care Clinical Psychology Service: Impact on Treatment Process and Outcome
Steven Jones,University of Manchester, Academic Division of Clinical Psychology, Graham Tate, Barry Fowler and Amy Tattersall, Laurence Burns Unit, Birch Hill Hospital, Rochdale
149 people routinely referred to a primary care clinical psychology service for cognitive behaviour therapy were asked to complete Millon's MCMI-III (1997). Although none of the GP referrals cited personality disorder as a primary problem, 57% met Millon's criteria for at least one personality disorder and 31% met criteria for two or more. In terms of the three personality clusters used by DSM-IV (American Psychiatric Association, 1994), 34.5% were 'anxious-fearful', 18% 'dramatic-erratic' and 11.5% 'odd-eccentric'. These figures are at least suggestive that significant levels of personality disorder exist in routine caseloads of clinical psychologists working in primary care. In addition, following Millon's classification, 16% of the sample met criteria for one of the three 'severe personality disorder scales' (schizotypal, borderline or paranoid).
A clinician's questionnaire was also developed which assessed a number of clinical process and outcome variables as well as clinician's opinion as to whether each patient had a personality disorder. Results indicated a significant association between MCMI-III and clinician detection of personality disorder, although this association was weaker when individual personality disorder categories were considered. 11 out of the 14 personality styles assessed by MCMI-III were associated with significant difficulties with therapy process and outcome.

The Treatment of Comorbid Anxiety Disorders in a Patient with a Diagnosis of Schizophrenia and the Effect of this Treatment on Psychotic Symptoms: A Case Study
John Good, Severn NHS Trust, Gloucester.
It is well known that anxiety disorders frequently exist comorbidly with psychotic disorders but are rarely diagnosed or treated. Stress vulnerability models are one possible explanation for the link between these disorders, suggesting that stress can push a person beyond their threshold for the experience of psychotic symptoms. If this is the case, it is suggested that the presence of comorbid anxiety disorders could provide the necessary stressors. Treatment of these stressors should then lead to a reduction in psychotic symptoms. The case study presented here concerns the treatment of agoraphobia and social phobia in a patient with residual psychotic symptoms. These symptoms consisted of hearing voices referring to his illness and experiencing his thoughts being read and reading the thoughts of others. These symptoms were not dealt with directly during therapy. Treatment consisted of CBT and an exposure programme for the agoraphobia and CBT for social phobia. This resulted in a reduction in both of these phobias as measured by the Fear Questionnaire and an elimination of the psychotic symptoms as measured by the KGV Symptom Scale.
More research is needed in this area but this case study would seem to uphold the stress vulnerability models and suggest that treatment of comorbid anxiety disorders may have a profound effect on psychotic symptoms in at least some individuals.

The Role of Case Formulation in Therapeutic Change
Penny Stevens, Lakeside Counselling and Psychotherapy Service, Lakeside Surgery, Corby, Northants.
Aspects of the process of case formulation are considered with reference to the literature relating to the research theme of how the process of case formulation may facilitate therapeutic change. A comparison between case formulation and diagnosis suggests limitations of diagnosis. Aspects of the role of case formulation in the mechanisms of cognitive change, and in the establishment and maintenance of the therapeutic relationship are explored. Case formulation is argued to offer advantages with regard to positive therapy outcome by providing opportunities for understanding the problem, for motivation towards engagement in therapy, collaboration, empowerment of the client, providing a rationale for therapeutic strategy, and for constructive ending of therapy with guidelines for relapse prevention. Evidence from the literature is presented which points to a number of ways in which the process of case formulation may operate to facilitate therapeutic change. These appear to comprise three main interrelated areas: (1) the individuality of the case formulation (2) the function of the case formulation and (3) the process of the development and updating of the case formulation.
The importance of the appropriate management of the balance of power within the therapeutic relationship during the process of the development of the case formulation is considered with regard to its potential to motivate the engagement of the client in therapy and to promote empowerment of the client, and thereby initiate therapeutic change.
The implications for clinical practice, the training of therapists, and for further research are discussed.

Sudden Gains in Cognitive Therapy
Jane Cahill, Gillian Hardy, Caroline Massey, David Shapiro, Anne Rees, Michael Barkham, Norman MacGaskill, University of Leeds, Psychological Therapies Research Centre
Sudden gains in therapy is an important phenomenon that may enable us to progress our understanding of client change processes and enhance therapy outcome. Tang and DeRubeis ( 1999), using data from two efficacy studies, found that 29 out of 61 clients experienced sudden gains between single sessions of cognitive-behavioural therapy (CBT). These clients enjoyed superior outcomes post-treatment and at 18 month follow up to non-sudden gainers. The authors set out to establish if this phenomenon exists more generally. The object of the study therefore was to attempt to replicate Tang and DeRubeis' findings using data from a study carried out in an NHS service within a University/NHS collaborative environment, using usual referral routes for clients and involving NHS therapists with regular caseloads.
The authors followed the procedure and terminology of Tang and DeRubeis ( 1999) to identify between sessions sudden gains. The criteria were as follows: 1) the gain had to be at least 7 BDI points between two consecutive BDIs; 2) the gain represented at least 25% of the pregain session BDI score; 3) the mean BDI score of the three therapy sessions before the gain was significantly higher than the mean BDI score of the three therapy sessions following the gain. In addition, the
authors followed Tang and DeRubeis in identifying "reversals" which indicate a significant increase in depressive symptoms and significantly impact on therapy outcome. Reversals were defined as whenever a client gave up 50% of their symptom improvement resulting from the "sudden gain". This sample comprised 62 clients who were offered at least 12 sessions of cognitive therapy. Twenty-two of the 62 clients experienced at least one sudden gain during therapy. The median 50% of sudden gains occurred between session 4 and 10, with the 4th session being the mode and the 5th session being the median pregain session. Of those 22 clients who experienced sudden gains during therapy, 9 clients experienced a reversal before the end of therapy, more than the 4 reported in Tang & DeRubeis (1999).
On average sudden gainers dropped 20.82 BDI points (SD = 8.50) during therapy. The mean total decrease in BDI points across therapy for the whole group was 13.56 (SD = 11.69). To compare the outcomes of sudden gainers and non-sudden gainers the BDI before the last therapy session was used. Sudden gainers presented with a significantly lower BDI pre final session than non-sudden gainers (p < .05). On the follow-up data currently available, there is a non significant trend for sudden gainers to be doing better at 4 month follow up than non-sudden gainers. When non-completers were removed from the analysis, there were no significant differences in outcomes between sudden gainers and non-sudden gainers. However, there was still a non- significant trend (p = .094) for sudden gainers to be doing better pre-final session, suggesting that the relationship between sudden gains and improved therapy outcome is not spurious.

Therapist Competence as a Predictor of Outcome in the Cognitive Therapy of Depression
Chris Trepka, Anne Rees, David Shapiro, & Gillian Hardy, Bradford Community Health NHS Trust, Leeds University
Therapy process measures can show how well therapy is being performed, as well as providing a means of monitoring the progress of therapists learning new or more advanced techniques. Unfortunately there is a lack of well-developed process measures specific to cognitive therapy.
The Cognitive Therapy Scale (CTS) is the only established measure of how competently cognitive therapy is carried out by therapists. Although the CTS has been used widely in cognitive therapy training centres there is little published data on its validity . The only substantial study of the relationship of cognitive therapy competence and outcome found only limited
support for such an association. It seems inconceivable that therapist competence should not beclearly related to clinical outcome, but the relationship may be easily obscured, for instance by other factors affecting outcome or if there are only small differences in competence between therapists.
In this study an external expert rater used the CTS to rate audiotapes of one randomly selected therapy session from each of 30 clients. Clients were taken from a cognitive therapy clinic for the treatment of depression. They completed the Beck Depression Inventory (Bill) at the beginning and end of therapy and before each session. Following each session clients also completed a measure of therapeutic alliance. Therapist scores on the "cognitive therapy skill" subscale of the CTS were associated with outcome for the whole sample of clients. The relationship between competence and outcome was
appreciably stronger for clients who persisted in therapy; other competence measures, including the CTS total score, also predicted outcome when dropouts were excluded from the analysis. While therapists obtaining lower CTS scores tended to achieve substantial reductions in depressive symptoms with therapy completers, complete resolution of symptoms occurred more often for therapists who scored higher on the CTS. Regression analyses showed that therapeutic
alliance was a stronger predictor of outcome than competence, although both measures independently contributed to outcome.
These results suggest that the CTS is a valid measure of cognitive therapy competence. Weaknesses are apparent in the scale, however, and development of an improved version could provide an even more valuable instrument in quality control and clinical governance, as well as the training and supervision of therapists.

A Comparative Study on Personality Characteristics of Learning Disabled and Non Learning Disabled Among Primary School Children
Gitanjali Sharma,Consultant Psychologist, Chennai, India
A modest attempt was made in the present study to examining the personality characteristics of 180 subjects, both boys & girls of ages 8,9 &10yrs old with specific learning disabilities (LD) From 3rd,4th&5th Grades of Urban & Rural primary schools & also an equal number of normal children(180 Subjects both Boys & Girls of non-LD)with the same age group & grades examined for a comparative study. An adopted version Cattell's Children Personality Questionnaire (CPQ) was Administered to the subject with LD & Non-LD. The results find that there is a significant difference on personality characteristics of LD & Non-LD Children, the study also points out the older LD children tend to show more maladaptive behavioural disposition than the younger LD & also a Significant gender effect among LD children.

Structured Group Behavioural Psychotherapy for Depressive Disorder: A Prospective, Randomised, Controlled Trial of Clinical Efficacy and Cost-utility
John Swan Department of Psychiatry, University of Dundee
It is proposed to test the clinical efficacy, utility and cost-effectiveness of a structured group behavioural psychotherapy in the management of outpatients meeting ICD-10 criteria for a moderate to severe depressive episode (F32.1, 32.2, 33.1 or 33.2) referred to the Tayside Affective Disorders Service. Structured group psychotherapy will be compared with optimised antidepressant drug treatment, placebo-drug treatment and individual cognitive behavioural psychotherapy. Trained therapists will deliver both psychological treatments in 12 sessions over a period of 10 weeks with additional 'booster' sessions after 14, 18 and 22 weeks. Drug and placebo-drug treated groups will be managed according to protocolised assessment and treatment procedures that guide practice in the Tayside Affective Disorders Clinic. Clinical assessments will be conducted by raters blinded to treatment conditions at 0, 4, 8, 12, 26 and 52 weeks. Structured group behavioural psychotherapy may represent a potent, cost-effective non-stigmatising treatment for depression that can be delivered by a broad range of mental health care professionals

POSTER SESSION 2

'Assisted Self-Help' for Chronic Pain: A Pilot Study Using Cognitive and Behavioral Interventions
Helen Macdonald, Pain Management Unit, Doncaster and Bassetlaw Hospitals NHS Trust, S.Yorkshire.
Cognitive-behavioural psychotherapy is effective for many sufferers of chronic pain. (Morely, Eccleston and Williams, 1999) It is usually delivered in in-patient or out-patient groups, or as a one-to-one intervention. There is increasing evidence that an 'assisted self-help' format can be both clinically and cost-effective as a mode of CBT treatment delivery, (Lovell and Richards, 2001). Manual-assisted therapy with brief therapist contact has been shown to be useful in treatment for depression and several anxiety disorders, for example. It is also argued that pain sufferers benefit from perceived control over interventions used.
A small pilot study was carried out in a hospital outpatient Pain Management Unit, to investigate the effectiveness of an assisted self-help format for cognitive-behavioural interventions for chronic pain.
An 'assisted self-help' manual was developed, using existing evidence on effective interventions for chronic pain. 13 people took part in the pilot study, sampled from usual referrals for CBP for chronic pain, by invitation to participate.
For the treatment group, mean scores significantly improved on measures of mood. Quality of life, self-reported pain intensity and evaluative ratings all showed changes in the desired direction, but did not reach significance. Those waiting for treatment tended to become more depressed, but there were no significant trends in the waiting list group.
The outcome of this study indicates that changes tended to be in the desired direction, therefore a larger-scale study is planned, in which this approach will be compared directly with Cognitive-Behavioural 'Treatment as usual' for chronic pain.
Lovell, K and Richards, D (2001) Multiple Access Points and Level of Entry, (MAPLE): Ensuring Choice, Accessibility and Equity for CBT Services Behavioural and Cognitive Psychotherapy 28 379-391
Morley, S., Eccleston, C., and Williams, A. C. (1999), Systematic review and Meta analysis of randomised controlled trials of cognitive behaviour therapy and pain in adults, excluding headache. Pain 80, 1-13

Developing a "First Class" CBT Service: The Newcastle Approach.
Vivien Twaddle, Newcastle Cognitive and Behavioural Therapies Centre, Newcastle City Health NHS Trust & Derek Milne, Department of Clinical Psychology, University of Newcastle upon Tyne.
Quality is a fundamental and increasingly important requirement of the "New NHS", as epitomised by "A First Class Service" (Department of Health, 1998). In response, the Newcastle Cognitive and Behavioural Therapies Centre (NCBTC) has attempted to (i) evaluate the quality of the cognitive therapy service it provides, using a new measure, The NCBTC Patients' Needs Questionnaire and (ii) use the data to improve the quality of the service via a stakeholder-collaborative evaluation. By means of a longitudinal, quasi-experimental design, the study involved 183 stakeholders, (patients, referrers, managers, purchasers and staff) who completed the Questionnaire. On the basis of the questionnaire data, 28 stakeholders rated the quality of different aspects of the service, prioritised the areas for improvement and provided suggestions for change. Furthermore, on the basis of these extrapolated opinions, NCBTC staff, in turn, produced service objectives, clear action plans and explicit methods of evaluation which were ratified by the Trust Board and resulted in an improved service delivery model, "cascading cognitive therapy skills", designed for staff across the mental health service. The results showed that that, overall, the service was perceived as "good" with pockets of "excellence". Clear priorities for improvement emerged, including the need to disseminate therapeutic skills to staff outside the NCBTC via a systematic model of supervision and training. The study illustrates the viability and impact of a stakeholder-collaborative approach; one, it is argued, that is inherently cognitive-behavioural in philosophy and principles. It has also led to the development of a staff CBT training needs assessment tool and has generated ideas about a cascade model of training.

Panic Disorder in Primary Care. Research into a Novel Approach Using Group Work and Exercise
Pauline Callcott, Paul Cromarty, Gary Robinson, Mark Freeston, Newcastle Cognitive and Behaviour Therapies Centre
An initial service development by the Newcastle Cognitive Behavioural Therapies Centre and health and fitness staff in a healthy living centre combines Clark's panic model (1986, 1996) in a group format with a formal exercise programme. It is reported in 'Physical Activities and Health' a national consensus statement, that there is a clear beneficial effect of exercise on anxiety, however it quotes that 'certain people will not become less anxious as a result of physical activity alone (e.g. people with panic disorder and agoraphobia.)' The evaluation of the pilot groups has been encouraging with evidence of clinically significant improvement in participants post therapy scores on standardised assessments such as the Clinical Outcomes in Routine Evaluation, Beck Anxiety and Depression Inventories, as well as specific measures for panic disorder and agoraphobia and avoidance. However it is less clear why some participants are benefiting and the NCBTC is researching characteristics that might predict a better response for group participants. It is hypothesised that a combined CBT exercise package is superior to exercise alone in eliminating panic, reducing symptoms and increasing well being and functioning. The poster will outline clinical findings as well as discuss how a service like this can be cascaded to other settings. The pilot project work fits with National Service Framework (1999) guidelines, in that it targets gaps in service provision, involves users in their own care and promotes interagency working. The project also tackles social exclusion in that it provides the sort of service normally found in tertiary centres and delivers it in an effective manner in a non-stigmatising local community setting such as the West End Health Resource Centre.
Clark D.M.(1986) A Cognitive Approach to Panic. Behaviour Research and Therapy 24. 461-470
Department of Health (1999). National Service Framework For Mental Health: Modern Standards and Service Models. London. HMSO.
Grant T. (ED.) (2000). Physical Activity and Mental Health: National Consensus Statements and Guidelines for Practice. Health Education Authority.
Department of Health (1999) National Service Framework For Mental Health: Modern Standards and Service Models London. HMSO.

The Impact of Postgraduate Clinical Training on Clinical Performance.
Analise O'Donovan, M. Dyck, and J. Bain, Griffith University, Australia
This paper addresses the ongoing issue of the effectiveness of clinical training, a topic that has led to many debates over a number of decades. The necessity of higher degrees is based on the belief that additional years of training produce a more effective practitioner. However, the evidence for this argument has been consistently equivocal. There have been a number of explanations suggested in the literature for the limited evidence of training effectiveness, with lack of methodological rigour (e.g. unreliable measures, lack of control groups), frequently suggested as a reason for the failure to observe training effects. To a large extent, ethical and practical constraints on the random assignment of participants to training or no-training conditions, contribute to these problems. However, in Australia, graduates of 4-year psychology courses can obtain professional registration by one of two routes: postgraduate training or supervised practice. These alternative pathways to professional development afforded the opportunity to assess the effectiveness of postgraduate training courses by comparing the clinical performance of individuals who are selected for postgraduate training, with the clinical performance of individuals who do not enter postgraduate training.
A quasi-experimental design was used to evaluate the effectiveness of clinical training programs. As the scientist-practitioner model has been adopted as the basis for training in Australia, training outcome was assessed using measures representative of both clinical knowledge and clinical practice ability. 31 clinical psychology trainees sampled from 7 Australian universities were evaluated at the beginning of their postgraduate training courses and one year later; 30 psychology graduates from 5 Australian universities who did not pursue postgraduate clinical training served as controls.
Results showed that following training, trainees perform significantly better than non-trainees on a measure of assessment, treatment, and evaluation ability, a measure of diagnostic ability, and a measure of the ability to set goals and effective tasks with a standardised client. No differences were observed in their ability to develop an accurate case conceptualisation of a client that had just been interviewed, to develop an effective bond with that client, or to communicate empathic understanding.
As suggested by the scientist-practitioner model, results indicated that so-called clinical knowledge and practice ability do represent distinguishable knowledge structures. But our results also indicate that both clinical knowledge and practice ability contribute to a higher-order knowledge structure that we call global clinical ability. These results may be important because they illustrate something about the "ingredients' of clinical practice that, although obvious, is frequently overlooked. Clinicians can only use or apply their clinical knowledge in the context of the relationships that they have with their clients; the relationships that clinicians develop with their clients always develop in tandem with the clinicians' application of their clinical knowledge to helping clients achieve their aims.
Although the data to date is encouraging as indicated by the moderate to large effect sizes on some measures of trainees' knowledge and practice abilities, the way forward is to also consider some other variables which are likely to be influencing effectiveness of practitioners to understand the role of training within a wider context of relevant outcome. factors.

Factor Analysis of CBT Supervision Activity
O'Carroll, P.J. Chester College of Further Education,
The current literature on cognitive behavioural therapy (CBT) supervision is primarily about trainees with only a small part concerned with post-accredited CBT practitioners. This situation is seriously out of step with current NHS reforms where Clinical Governance requires health service providers to ensure quality and effectiveness in the delivery of health care in the UK, including psychological therapies. Effective supervision between accredited practitioners is a necessary component for ensuring maintenance and effectiveness of psychotherapy practice. Supervision is necessary both during training and during post-accredited practice. On this basis, there is a pressing need to increase research and development that should underpin post-accredited supervision practice.
One issue for examination in developing post-accredited CBT supervision practice concerns the perceived functions of supervision: whether supervision should be essentially personal or whether educative, that is, whether one examines the supervisees' experience or whether one examines what they do. Often this duality has led to polarisation and or compromise amongst other psychotherapy approaches, a situation that has also been observed within CBT. CBT has tended to be characterised as more pragmatic, educative and task focussed compared to other psychotherapy approaches. With these issues in mind, this study aims to explore and analyse the content of supervision activities of post-accredited CBT practitioners in the UK.
A postal survey of 240 UKCP registered CBT practitioners was carried out looking at the nature and extent of specific supervision activities. One hundred and twenty four (54%) questionnaires were returned. The questionnaire included 16 items about the content of supervision activities. The content areas were derived from a number of sources, including, competency checklists and literature on the structure, content and process of CBT supervision (e.g. Padesky, 1996). The items addressed several areas of activity including: assessment, formulation, intervention, evaluation, inter- and intra-personal processes, therapy alliance and ethics. Respondents were asked to rate the relative level of these activities over the previous 12 months.
The findings highlighted differences in the level of supervision activities with practical task activities reported more frequently than inter- and intra-personal process issues. A principle component factor analysis yielded 4 components labelled (1) tasks, (2) bond, (3) supervision and (4) critical decisions. These components equate closely with the dual personal and educative functions identified in other psychotherapy orientations. However, it is suggested that contemporary models of CBT supervision (Padesky, 1996) provide a more effective and theoretically consistent resolution of the multiple functions of supervision suggested in this study. The work of Padesky (1996) not only provides a basis for research but also a framework for CBT supervision practice and training.

The Development of Two New Instruments for the Evaluation of Foundation Level Training in Cognitive Behaviour Therapy
Pamela Myles-Kelly, Department of Psychological Therapies & Research, Northumberland Mental Health NHS Trust, Northumberland
Evaluation of training programmes is essential and requires the development of appropriate measures. Two new instruments were developed to assist in an evaluation of a new foundation level training programme in Cognitive Behaviour Therapy (CBT). The measures are described and their reliability and validity presented. Ninety mental health professionals were trained in foundation level CBT over a 12-week period, receiving a total of 48 hours of shared learning, covering core theory and techniques for depression and anxiety within an experiential workshop format. The research took the form of a waiting list, control group design. In this longitudinal design, participants served as their own controls. Multiple measures were administered during a double baseline assessment, a post-training re-assessment and a three-month follow-up assessment.
Instrument one, The Foundation Cognitive Behaviour Therapy Multiple Choice Questionnaire (FCBT-MCQ), was developed to assess knowledge of CBT. The original MCQ had 25 items and, following an item analysis, was reduced to thirteen items.
Instrument two was a Video Assessment Task (VAT), developed to measure CBT-relevant skill. An actor described a typical Panic Disorder, which included three symptoms from each of the following domains: cognitive, behavioural, physical, and affective. Participants were shown the video clip twice and given a total of 10 minutes to answer three questions related to what they had seen. Participants were asked to identify symptoms, name the problem and consider up to six suitable cognitive-behavioural techniques.
Good reliability and validity data were obtained for these two instruments, which are available as sound instruments to support training in evidence-based practice.

Group Cognitive Therapy for Low Self-Esteem: A Preliminary Evaluation
Craig Simpson, Manchester Mental Health Partnership
Low self esteem is commonly encountered in clinical practice. In many cases, self confidence can be restored by treating the individual's presenting problem. However, for some individuals, low self esteem is more pervasive, and may act as an underlying vulnerability factor that precedes the onset of their problems, and that predisposes them to multiple, recurrent or persistent psychological difficulties. In these cases, low self esteem in itself may warrant clinical attention. A recent cognitive model of low self esteem and associated treatment programme has been proposed (Fennell, 1997). The current study is a preliminary evaluation of a group cognitive therapy programme for low self esteem which is based on this conceptualisation, and which employs a modified version of the individual treatment protocol. The contents of the group therapy programme will be outlined and the theoretical and practical considerations of treating low self esteem in a group setting discussed. Finally, pilot outcome data will be reported.

Modified Dry-Bed Training for Treating Childhood Nocturnal Enuresis: A Replacement for the Bell-and-pad Method?
Shazia Nawaz, Peter Griffiths, Department of Psychology, University of Stirling & David Tappin, Department of Child Health, University of Glasgow.
We compared the efficacy of modified Dry-Bed and urine alarm training for treating monosymptomatic nocturnal enuresis in children aged 7-12 years attending health centres? in Glasgow. We simplified the original 1974 Dry-Bed method and described it in an illustrated instruction manual. Parents and children followed the procedure at home, aided by a videotape. We also wrote instructions forn conventional urine-alarm ('bell-and-pad') training as a self-help manual with an accompanying videotape. Dry-Bed training is based on an operant learning model and emphasises the complex sequence of skills the child needs to acquire to stay dry at night, such as wakening to the sensation of a full bladder and moving rapidly from bed to toilet. Urine-alarm treatment focusses more narrowly on promoting arousal in response to bladder fullness. The methods also differ in that, in Dry-Bed Training, the bedwetting alarm is used to alert the parent/trainer as opposed to the child directly, as happens in urine-alarm conditioning. Between November 1999 and July 2000, 36 children with primary enuresis were assigned to Dry-Bed, urine-alarm and waiting list control groups, each group being of equal size and matched for age, sex , social class (deprivation category) and baseline wetting frequency. In the two treated groups, initial interview was followed by two review appointments, otherwisefamilies carried out the training independently at home with fortnightly telephone support for16 weeks or until initial success was achieved (14 consecutive dry nights). Of the 12 children treated by the Dry-Bed method,8 achieved initial success compared with only 3 of the 12 treated by the conventional urine-alarm method. One waiting-list control child remitted spontaneously. Analysis of variance showed highly significant differences in wet nights per week immediately after intervention for both treatment and time factors (p<0.001) and their interaction (p<0.01). The Dry-Bed group averaged a mere 0.8 nights per week wet at short-term outcome compared with 3.25 for the urine-alarm group and 5.00 for the controls. Moreover, only the Dry-Bed group were superior to both the urine-alarm and control groups, the urine-alarm group's wet-night average being not significantly different
from that of the controls. At follow-up, three months after treatment, no child had relapsed or spontaneously remitted in any group. Our results point strongly to modified Dry-Bed Training being more efficacious than orthodox urine-alarm conditioning over comparable invervention periods. It may be that modified Dry-Bed Training, delivered as a self-help package, will become the treatment of choice for primary nocturnal enuressis in childhood. We are currently investigating the method with a larger sample.

Therapist Competence as a Predictor of Outcome in Cognitive Therapy for Depression
Chris Trepka, Anne Rees, David Shapiro & Gillian Hardy
Bradford Community Health NHS Trust (CT), Leeds University (co-authors)
The Cognitive Therapy Scale (CTS) has been widely used in cognitive therapy training centres to assess therapist competence but competence has not previously been shown to predict cognitive therapy outcome clearly. We compared competence with alliance as process variables that might moderate change in therapy. A randomly selected therapy session from each of 30 courses of cognitive therapy for depression was rated using the CTS. Regression analysis showed that therapeutic alliance was more strongly related to outcome than therapist competence, although both measures independently contributed to outcome. Prediction of outcome was better for clients who persisted longer in therapy. All six therapists in the study demonstrated competence, but three averaged higher CTS scores and achieved symptom reduction to low levels on the Beck Depression Inventory more often.
These results indicate that both modality-specific cognitive therapy ability and non-specific facilitation of alliance are moderators of change, and thus determinants of outcome, independently of each other. The effects upon outcome appear to be mediated by the extent to which subjects persist in therapy. Weak relationships between CTS scores and outcome in previous studies might be attributable to less variation in therapist competence levels. The validity of the CTS as a measure of cognitive therapy competence is supported by this study, although weaknesses are apparent in the scale.