Vous êtes sur la page 1sur 13

Metacognitive Therapy for PTSD


treatment of PTSD in Viemam War veterans.Journal chology, 53, 917-923.

of Clinical Psy-

Weathers, E W., Litz, B. T., Herman, D. S., Huska, J. A., & Keane,

  • T. M. (1993,

October). The PTSD checklist(PCL): Reliability, validity,

and diagnostic utility. Paper presented at the 9th Annual Meeting of the International Society for Traumatic Stress Studies, San Antonio, TX. Yalom, I. (1995). The theory and practice of group psychotherapy (4th ed.). NewYork: Basic Books.

The author wishes to thank the female therapists involved in conducting treatment: Stephanie Fallon, Evelyn Sandeen, Ella Nye, Maureen McGlynn, Annette Brooks, Andrea Blmnenthal, and Sharon Sprague;

thanks to Celia Michael for encouraging this work; and thanks to Jan

Wallner for her support.

Address correspondence to Diane T.

CastiUo, Psychology Service

(l16B), New Mexico VA Health Care System, 1501 San Pedro S.E., Albuquerque, NM; e-maih diane.castillo@med.va.gov.

Metacognitive Therapy for PTSD: A Core Treatment Manual


Wells, University of Manchester



University of Liverpool

This article describes a new brief treatment for PTSD based on a metacognitive model (Wells, 2000). The treatment derived from this approach can be divided into core and supplementary treatment components. The core treatment manual is presented hoe. The core treatment does not require imaginal reliving of trauma or cognitive challenging of thoughts and beliefs about trauma. It enables pa- tients to develop a metacognitive perspective and disengage unhelpful thinking styles such as worry/rumination and attentional monitoring that block the natural propensity for cognitive-emotional adaptation following trauma. The content, techniques, and sequence of the basic program are described in detail to support practical application of the new treatment by therapists.


have consisted

of various methods and the study of treatment effi-

cacy is still at an early stage. Treatments comprised of ex- posure methods and cognitive therapy focused on modi- fying negative appraisals have been shown to be effective in controlled trials. In comparative studies, exposure therapy alone and cognitive restructuring without expo- sure have been shown to be equally effective (e.g. Rich- ards, Lovell & Marks, 1994; Tarrier et al., 1999). However, a proportion of patients do not respond or fail to com- plete treatment (Sherman, 1998; van Etten & Taylor, 1998). Prolonged imaginal reliving of trauma is distressing and is poorly tolerated by many clients (Scott & Stradling, 1997). Exposure and cognitive therapy approaches are relatively brief interventions involving about 10 sessions. hnplementation of these treatments requires a high de- gree of therapist skill. We believe that for treatment to be more widely accessible, interventions that are brief, less demanding, and less potentially distressing are needed. Ideally, such approaches should be grounded in empiri- cally testable theories of the psychological mechanisms un- derlying normal and abnormal posttraumatic processing.

Cognitive and Behavioral Practice 11,365-377,



Copyright © 2004 by Association for Advancement of Behavior Therapy. All rights of reproduction in any form reserved.

In this article we present a new cognitive therapy for PTSD. We describe the theoretical background, briefly review evidence for a central tenet of the model, and then provide a detailed guide to the treatment. The model on which the treatment is based was first advanced by Wells (2000) and is grounded in an information-processing theory of psychological disorder (Wells & Matthews, 1994,


A Cognitive Model


The starting point for the new model is the assump- tion that following trauma an internal goal of processing is the development of a blueprint or plan for guiding thinking and action in potential future encounters with threat.Just as individuals can be conceptualized as having a plan or script that guides thinking and behavior when eating in a restaurant, people have plans that guide cog- nitive and behavioral activities during encounters with threat. The goal of emotional processing, which normally proceeds unimpeded and spontaneously, is the strength- ening of such a plan. We have termed the process by which emotional processing takes place and plans are de- veloped the Reflexive Adaptation Process (RAP), a term intended to capture the idea that this is initiated auto- matically in response to intrusive thoughts. Whether or not adaptation occurs depends on the style of thinking and coping adopted by the person. Internal beliefs and


Wells & Sembi

environmental factors lead to a style of thinking and cop- ing behaviors that block the development of adaptive plans and prevent the return of cognition to a "normal" state of processing. Beliefs of a metacognitive nature (i,e., about thinking itself) are of central importance because they guide activities of the individual's cognitive system and can lead to styles of thinking that facilitate or impede emotional processing. Such metacognitive beliefs can be represented verbally but are linked to implicit plans that guide thinking. For example, much of the knowledge used to guide thinking and behavior is not verbally ex- pressible. In verbal form, examples of such beliefs in-

elude the following: I must worry in order to be prepared; I must try to remember all of the details of the trauma so that I can understand why me; I should be alert to all sources of danger; If I pay close attention to all possible threat I won't be taken by sur- prise; I must not think about what happened or I'll not cape.

These metaeognitions (and plans they represent) give rise to perseveration, which is persistent and recurrent thinking about trauma, threat, and one's reaction to it. Perseveration consists of worry/rumination, threat moni- toring, and maladaptive thought-control strategies. It is maladaptive because it strengthens and maintains per- ceptions of threat and it blocks the processes necessary

for the RAP. The model is novel

in its emphasis on styles

of thinking (e.g., worry, attentional monitoring for threat, mental control) rather than on content, and for its focus on metaeognitions, which are the beliefs and strategies used to appraise and regulate thinking itself. Like other models, the approach sees coping through avoidance, use of alcohol, and so on as further examples of unhelp- ful coping. In the present model alcohol is a problem be- cause it alters the conscious experience of intrusions so that mental simulation is disturbed, whilst avoidance pre- vents the person discovering that the environment is not still dangerous. In each case cognition cannot return to a "normal" state of functioning. A diagrammatic outline of the model is presented in Figure 1. In the model symptoms such as intrusive thoughts and hyperarousal are normal in-built responses following trauma. They act as biasing agents on cognition and lead to the selection and revision of metacognitions for guid- ing thinking and coping. Normally, symptoms such as intrusive recollections (flashbacks) coupled with atten- tional orienting responses provide an impetus for run- ning mental simulations of dealing with trauma. Such mental simulations are a rudimentary mechanism for lay- ing the foundations of a metacognitive plan for coping and subsequent action (the RAP). The flexible use of im- agery is a typical and potentially useful medium of pre- liminary plan compilation. It is useful because it repre- sents the dynamics of action and cognition over time and relates multiple modes of responding (thinking, behav- ing, feeling) in concert in a way that can be controlled by




1-'1 !

Adaptive situational


Low ruminative activity

Flexible attention control

Mental simulation (planning)

Acceptance of symptoms


cognitivere-tuning Exit


Maladaptive situational



Threat monitoring

Negative appraisal of coping/ symptoms

Avoidance/thought control



  • i !

  • i |

;" Psych0s0cial





Figure 1. A schematic representation of the metacognitive model of PTSD.

the individual. In PTSD, the resources needed for mental simulation are drained by, and ending the threat cycle in- compatible with, competing processing activities. In par- tieular, if the person believes that the way to cope is to en- gage in repeated conceptual (i.e., verbal) analysis of past events and/or to worry about the future, this can interfere with more adaptive imaginal simulations. Furthermore, the person may not be flexible enough in their use of thinking strategies such that there is an overuse of re- cyclic thinking (e.g., trying to remember all of the details, trying to work out blame) that constitutes chronic analyt- ical dwelling on trauma-related information. If the per- son believes that the best way to cope is to avoid situations, avoid thoughts of trauma, and/or execute hypervigilant threat-monitoring strategies, this interferes with mental simulation and maintains perceptions of danger so that the anxiety program persists. The anxiety program is also maintained by negative interpretation of traumatic symp- toms such as interpreting intrusive thoughts and hyper- arousal as a sign of mental breakdown. As a consequence

Metacognitive Therapy for PTSD


of these factors, the traumatized individual is unable to

exit the emotional processing (RAP) cycle because men- tal simulation is incomplete and/orcognitiondoes not re-

tune to the normal threat-free environment. As depicted in Figure 1, maladaptive and adaptive strategies arise from metacognitions as represented by the arrow leading from metacognitions to strategy. The strategies implemented in turn have an effect on meta- cognitions as depicted by the return arrow. For example, strategies such as trying not to think certain thoughts activate metacognitive thought-monitoring strategies in which the person monitors for the occurrence of target thoughts. This can have the effect of triggering the un- wanted thought itself. The resulting failure to avoid thoughts can strengthen negative metacognitive beliefs about the uncontrollability of thoughts about trauma. To take an- other example of the effect of strategies on metacogni- dons, the use of worry and hypervigilance strengthens metacognitive plans for these activities, so in essence the person becomes an increasingly skilled "threat detector." Moreover, the nonoccurrence of threat can be taken as evidence that worrying and hypervigilance are working. Such an effect strengthens positive beliefs about the use- fulness of these strategies. Adaptive strategies, on the other hand, allow reflexive adaptation to occur so the person spontaneously develops plans for coping in the future rather than being stuck in attempting to cope with non- existent threat in the present. Symptoms subside when a satisfactory plan for coping has been established and the person is able to exit the RAP. Exiting requires metacognitive monitoring and con- trol processes, and the RAP continues until discrepancies between the current status of the self (e.g., feeling vulner- able) and a desired or a normative internal goal state is eliminated. This process of checking for discrepancies consists of in-built metacognitive monitoring and control processes that are part of the RAP. However, discrepan- cies persist when metacognitive beliefs maintain the focus of processing on danger. We have seen how metacognitive beliefs about the value of worry/rumination underlie perseverative pro- cessing in the form of negative conceptual activity. Other factors also lead to a general repetition of processing. For example, an individual may be dissatisfied with the way in which she or he coped during the trauma. This will lead to repeated activation of the RAP leading to intrusions about the trauma so that the individual can plan more sat- isfactory responses. Such dissatisfaction may not occur immediately after the trauma but can be initiated by later negative social factors such as criticism or blame, which is one potential pathway to delayed-onset PTSD. Stresses that are difficult to bring under personal control also contribute to perseveration because the individual is in a state of persistent mental simulation and planning.

The Model in Action

A "walk through" of the model for a particular case will help to clarify how the model works. A case conceptu- alization based on the model is presented in Figure 2. In this example the RAP is repeatedly triggered by in- trusive images and reexperiencing of pain. The occur- rence of these and related symptoms leads to the activa- tion of metacognitions to guide information processing required for the RAP. In this case the metacognitions are maladaptive and lead the individual to activate worry/ ruminative strategies of self-regulation, hypervigilance for danger, bodily checking, avoidance, and thought con- trol. Metacognitive beliefs also lead the individual to fear symptoms themselves, thinking that this could lead to "losing it." As a result of these strategies, normal activities of the RAP are blocked and so intrusions automatically persist as a means of kick-starting the adaptive RAP process. The maladaptive strategies used also keep anxi- ety and a sense of threat going, thereby fueling anxious symptoms. Some strategies such as bodily checking in com- bination with anxiety contribute to feeling dead/unreal.




(aroUsal/threat level) I


.I n~r~~iii~uilain



Worrying eepame I

Adaptive situational






I Maladaptivesituational processing

Worryabout danger Look out for danger Avoid traffic

Avoid TVprogrammes/

reminders Try to controlthoughts




(' Psychosocial '~ ',,




Figure 2. A case conceptualization based on the metacognitive model.


Wells & Sembi

Thus, intrusive thoughts and anxious symptoms persist, culminating in a state of PTSD.

Predictions and Clinical Implications

The model assumes that posttraumatic s)maptomatol- ogy--that is, intrusions (memories and flashbacks), night- mares, heightened arousal, and hypervigilance--are nor- mal in the immediate aftermath of a traumatic event. They are indicators that the individual's cognitive system is commencing the RAP process. Difficulties arise, how- ever, when the individual uses inappropriate coping strat- egies and/or interprets the symptoms in a threatening way. These factors disturb the compilation of an appro- priate plan for dealing with threat and subsequent fading of the RAP. A basic implication is that treatment that re- moves the barriers to normal adaptive processing should be effective in relieving PTSD. The model makes several testable predictions. One of the most prominent is that specific internal aspects of processing interfere with the RAP underlying emotional processing. Worry and rumination drain resources neces- sary for processing images and running simulations. They also focus the individual on negative outcomes, thereby fueling the anxiety program and draining the resources needed for constructing a coping configuration (plan). A further prediction is that some coping strategies (e.g., avoidance and dissociation) will interfere with emotional processing. Excessive efforts to control or avoid thoughts of trauma will interfere with normal simulation pro- cesses. Finally, since the intrinsic goal of processing is to develop a plan for coping, factors that contribute to per- ceptions of the self as ineffective at coping with the threat will contribute to perseveration of the RAP and symp- toms of failed emotional processing. Candidates of inter- est are negative self-appraisals resulting from beliefs about the self or aspects of the trauma, and psychosocial factors such as negative social support. It is apparent from these predictions that responses to symptoms of stress, such as intrusive thoughts, startle re- sponses, and hypervigilance, should be managed in a par- ticular way that blocks worry/rumination, threat monitor- ing, and avoidant coping. Adjustment processes should be allowed to run their own course without inflexible or maladaptive upper-level involvement. In the following section we describe briefly how the model accounts for important PTSD-related phenomena.

Explanation of Imaginal Exposure Effects

While the model predicts that prolonged imaginal re- living of trauma is not necessary for effective treatment, the efficacy of the technique can be explained within the context of the model. The technique will be effective when it allows the individual to run mental simulations of experience that facilitate the RAP and lead to plan forma-

tion. Moreover, imaginal reliving will promote habitua- tion, which will weaken reflexive influences on process- ing and increase the flexible control of attention and coping so that adaptive strategies may be implemented. Imaginal exposure will also be beneficial in cases where patients fear anxious symptoms, and when the evocation of symptoms during exposure provides evidence that challenges symptom-related worries.

Dose-Response Phenomena

The model explains the dose-response relationship in PTSD in which repeated stressors or events of greater magnitude (or causing greater arousal) lead to more se- vere PTSD. This effect is attributed to the dynamic inter- play between lower- and upper-level processing. Repeated and intense events lead to the formation of strong associa- tive links between lower-level processors, producing sensiti- zation to threat and more readily activated RAP responses. As a result, threat-related material is more likely to in- trude into consciousness and diminish higher-level flexible control over processing, needed for the implementation of adaptive processing and subsequent fading of the RAP.

Delayed-Onset PTSD

Delayed-onset PTSD is explained by the model in the following way. Initially following trauma, individuals may fulfill the goals of the RAP. However, at some point there is a shift to maladaptive responding before complete con- solidation of the plan. This shift can be triggered by envi- ronmental and social events that lead to negative reap- praisal of the self, a reoccurrence of threat appraisals, and/or won T . Competing processing priorities can also temporarily block the RAP as the capture of attention by stress symptoms is avoided. For instance, dealing with on- going current personal issues may suppress stress symp- toms until competing self-regulatory demands are re- moved, at which point PTSD symptoms emerge.

PTSD Symptom Clusters

The model explains the three PTSD symptom clusters-- intrusions/reexperiencing, hyperarousal, and avoidance/ dissociation--as follows. Intrusions/reexperiencing and hyperarousal are seen as normal consequences of stress that are part of the RAP process. They indicate activity of reflexive processing that serves to bias conscious process- ing and retrieval of information from long-term memory by repeatedly introducing material into consciousness as the basis of forming a cognition-action plan. Symptoms of avoidance/dissociation are seen as coping strategies that are maladaptive if used in the long term because they disrupt the RAP.

Treatment Implications

An implication of the model as depicted in Figure 1 is that it can be translated into an individual case conceptu-

Metacognitive Therapy for PTSD


alization, which is used to socialize to treatment and as a framework supporting individual techniques. Patients can readily identify the maladaptive strategies that form a vicious cycle of PTSD symptoms and the adaptive responses that are to be developed in treatment. The goal of treat- ment is to free up the traumatized patient's natural ca- pacity for self-regulation and adaptation following trauma. This consists of enabling a shift to a metacoguitive mode of processing in which worry/rumination strategies and threat monitoring are discontinued and a strategy of "de- tached mindfulness" (cf. Wells & Matthews, 1994, 1996) is established in dealing with symptoms. This offers a means of improving flexible responding and attenuating excessive active involvement with intrusions, symptoms, and memories. The suspension of maladaptive strategies unlocks the barriers to in-built adaptive emotional pro- cessing. These techniques are the rudimentary basis or "core" of treatment, which is described in this manual. However, additional strategies are part of the broader metacognitive treatment approach, namely, running men- tal simulations of coping with the trauma and direct chal- lenging of metacognitive beliefs about symptoms. The core treatment is described later in this article. Before doing so, in the next section we review the empirical status of one of the central and novel predictions of the model-- that maladaptive strategies of worry/rumination and thought control contribute to the development of PTSD.

A Brief Review

of Research

on Worry

and Thought Control

A central prediction of the model is that worry/rumi- nation should be associated with the presence or devel- opment of PTSD symptoms. In particular, worry should block the RAP and therefore lead to a persistence or in- crease in intrusive images as the person's cognitive system attempts mental simulations and adaptation. Several studies have tested for a relationship between worry and trauma reactions, and two experimental studies have examined the effects of worry on intrusive images following stress. Butler, Wells, and Dewick (1995) asked nonpatient participants to watch a gruesome and stress- ful film about a workshop accident, and then to engage in one of the following postfilm mentation strategies for a period of 4 minutes: worry in verbal form about the film, image the negative aspects of the film, or settle down (control condition). The individuals who worried experi- enced the highest frequency of intrusive images about the film over the next 3 days. In a larger study, Wells and Papageorgiou (1995) used a similar film-stressor method- ology, only this study compared the effects of five differ- ent postfilm mentation conditions. These strategies were:

(a) worry about the film, (b) worry about usual concerns, (c) image the negative aspects of the film, (d) distraction

by speeded letter-cancellation, or (e) settle down. These conditions were selected because they were considered to differ in the extent to which they caused blocked emo- tional processing of images and "tagging." Tagging refers to the activating of memories of the stressor and engag- ing in processing that sets up a wide range of associations so that an increasing number of concepts become re- trieval cues for stress-related material. The results showed that the two worry conditions were associated with the highest frequency of intrusive images about the trauma over the 3 days following exposure to the film. Moreover, there was a linear increment in frequency of intrusions across different conditions that was consistent with pre- dictions concerning a cozjoint incubation mechanism in- volving blocked emotional processing and tagging. The results of these analogue studies of trauma-related stimu- lation support the hypothesis that worry following stress can lead to an increase in poststress symptoms of intru- sive images. The model also suggests that particular metacognitive coping strategies should be linked to PTSD and negative outcomes following stress. Several studies, both cross- sectional and longitudinal, have examined the relation- ship between thought control strategies and posttrauma stress symptoms. Individual differences in strategies used to control distressing, intrusive thoughts can be measured with the Thought Control Questionnaire (TCQ: Wells & Davies, 1994). The TCQ assesses five factorially derived domains of strategy: distraction; social control; worry; punishment; reappraisal. The scale appears to have a sim- ilar factor structure in patient and nonpatient samples (Reynolds & Wells, 1999). The use of worry and punish- ment to control thoughts is positively associated with stress vulnerability and appears to be elevated in some clinical syndromes. Holeva, Tarrier, and Wells (2001) conducted a longitudinal study of the predictors of PTSD following serious motor-vehicle accidents in which vic- tims required hospital treatment. Measures of thought control (TCQ) and social support administered within 4 weeks of the accident were used as predictors of PTSD 4 to 6 months after the accident. The presence of stress

symptoms (acute stress disorder) at Time 1 was controlled. The use of worry to control thoughts at Time 1, a change in perceived social support, and an interaction between perceived social support and the use of social thought control strategies significantly predicted subsequent PTSD. In cross-sectional analyses of symptoms, thought-control strategies were predictive of acute stress disorder (ASD)

at Time 1 and of PTSD at Time 2. Both

distraction and

social control TCQ subscales were negatively correlated with ASD and PTSD caseness, suggesting a possible posi- tive benefit of these metacognitive control strategies. How- ever, worry and punishment both emerged as positive pre- dictors of ASD and PTSD. These findings for worry and


Wells & Sembi

punishment control strategies are echoed in a study by Warda and Bryant (1998). They found that individuals with ASD used more worry and punishment thought- control strategies than non-ASD patients. Reynolds and Wells (1999) showed that particular TCQ strategies distin- guished recovered and nonrecovered patients with major depression and/or PTSD, and that change in TCQ strat- egies was associated with recovery. The recovered group was more likely to use distraction and reappraisal and less likely to use worry and punishment. Studies of rumination show that the tendency to rumi- nate, defined as recyclic negative thinking about the causes and symptoms of depression, is associated with negative outcomes following stressful life events. Rumination has been found to be very similar to worry in both form and function (Papageorgiou & Wells, 1999, 2003). In prospec- tive studies rumination has been shown to be positively as- sociated with depressive symptoms following significant negative life events involving an earthquake (Nolen- Hoeksema & Morrow, 1991) and following bereavement (Nolen-Hoeksema, Parker, & Larson, 1994). The results of these studies support a central supposi- tion of the present model that thinking stTles and meta- cognitive coping strategies can adversely affect outcomes following stress and that perseverative styles of thinking involving worry/rumination and choice of particular thought-control strategies are unhelpful for adaptation.

The Core Treatment Manual

In the remainder of this article we present the core metacognitive treatment based on the model. We have presented this in the format of eight treatment sessions, although longer is required in some cases, depending on the rate of patient progress.

Structure and Duration of Treatment

Treatment sessions are held on a weekly basis. The ini- tial sessions last 45 to 60 minutes. Once patients are en- gaged and able to effectively implement control over worry and detached mindfulness, the duration of sessions is reduced to 30 minutes. Effective mastery is indicated by the presence of each of the following: (a) the individual reports having successfully disengaged worry/rumination from the occurrence of intrusions, arousal, and orienting responses; (b) the individual accepts symptoms as a nor- mal part of adaptation that do not require active avoid- ance or suppression; (c) the individual reports allowing intrusions/memories/thoughts related to the trauma to occupy their own "mental space" while watching the spontaneous behavior of them as a passive observer. In our preliminary evaluation of the effectiveness of core treatment, 8 to 11 sessions were required in order to achieve PTSD-free outcomes. The rate of progress through

the different treatment phases influences the number of sessions required.

Session h Case Conceptualization and Socialization

The first step is elicitation and consolidation of knowl- edge concerning the nature and base rates of specific PTSD symptoms (intrusions, flashbacks, nightmares, and arousal symptoms). Following this, an individual case conceptualization is constructed based on the metacog- nitive model depicted in Figure 1. The task of conceptualization is simplified by direct- ing the course of questions to exploring a series of spe- cific and recent episodes in which the patient was trou- bled by symptoms associated with the trauma, or in which there was an exacerbation of anxious affect. The aim is to elicit examples of each of the elements in the model. Much of the discussion focuses on exploring the presence and extent of (a) worry/rumination about the tFauma, (b) attentional monitoring strategies, (c) strategies for coping with symptoms/distress (e.g., avoidance, thought control, distraction, alcohol), (d) beliefs about symptoms, worry, and attentional strategies. An effective sequence for ob- taining this material is to begin by asking about symptoms and then exploring the strategies used to manage or avoid symptoms. The therapist next asks directly about at- tentional monitoring for threat, and worry/rumination. Questions are then targeted at eliciting beliefs about symp- toms, worry/rumination, and threat monitoring. This se- quence is illustrated in the following dialogue which was the grounding for the conceptualization presented in Figure 2.



I'd like to begin by asking about the symptoms you


have been experiencing in the past month. Can you describe them to me? I feel as if I'm dead. Like I don't exist anymore.


Is it like being detached from things around you?

P: Yes, it's unreal.

I have

to check my pulse and



to make sure I'm still alive.

T: What








thoughts or memories? I keep seeing myself on


floor and




pain in my legs and the blood

flowing from


head. I can actually feel the pain again. T: What about feeling anxious or frightened?


I feel scared all the time when I go out now, I'm

constantly thinking something bad will happen.




Do you




try and




these symptoms? P: I avoid things. T: What are you avoiding?

Metacognitive Therapy for PTSD


P: Walking in streets where





of traffic,

holes in the street, crossing the road.

T: Do you avoid things that remind you of what happened?


I avoid watching hospital scenes on the television.

T: Do you avoid the scene of the accident? P: If possible. But I have to go there. I look away when driving past, but sometimes I force myself to look at the whole scene to see if it was my fault. I only look at the whole scene when I feel I can cope, but it's only for a few seconds. T: What do you do when you are looking? P: I try to look at all of the possibilities to analyze what happened. But when I see myself on the floor with the pain and blood, I look away, and try not to think about it. T: You mentioned something I'd like to ask you more about. You said you try not to think about it. Do you try to control your thoughts at other times too? P: Yes, I try not to think about what happened. If I get a thought I try to push it out of my mind or think about something else. T: Have you found that what you pay attention to has changed since the event? P: Yes, I pay more attention to things that are not safe. I'm constantly looking around for traffic and listen- ing for sounds of lorries. It can be anything, I've been watching a ceiling fan at work because it's wobbling.


T: You mentioned paying more attention to danger, and it sounds like you are spending time worrying and dwelling on what happened. P: Yes, everyday I'm worrying, and so I end up avoid- ing things.

T: How much of the day are you worrying about bad things that could happen? P: It's usually in the background and when I have to go out I get really worried.


How much of the day are you dwelling and rumi-

nating about what happened? P: I have periods when I think about it a lot, usually when

I feel depressed. But I try not to get into that state. T: What are you trying to achieve by worrying about things? P: I'm trying to be cautious and avoid accidents. T: What are you trying to achieve by repeatedly think- ing about what happened?


I'm trying to work out if it was my fault.





T: Do you think worrying is helpful in any way? P: It makes me more aware of the potential risks

T: How does that work?


It makes me think of what could happen

so that


act more cautiously. But it also means I don't do so many things now.

T: Is being cautious something you do? P: Yes, I make an effort to be cautious.


How do you do that?

P: I keep a lookout for danger. T: So it sounds as if you believe that worrying and keeping a lookout for danger keep you safe.




T: Do you worry about your symptoms? P: Yes, I think it's not normal to be like this and I'm concerned it means I'm losing it. T: What do you mean by losing it? P: That I can't cope anymore. Maybe there's some- thing wrong with my mind. T: What's the worst that could be wrong? P: Well, I'm afraid this problem means that I can't cope as well as other people. T: Do you ever believe that you are going crazy? P: Not crazy.Just that I'm mentally weak some way.


Do you do anything to stop yoursefffrom losing it?

P: I try to control my thoughts. T: Do you think anything bad could happen if you didn't do that?


I suppose I could lose it.


So it sounds as if you believe that if you don't con- trol your thoughts that could happen?

P: Yes.

Following case conceptualization the therapist moves onto socialization. This consists of presenting the formu- lation in which the therapist stresses that PTSD symptoms

are a normal part of adapting to traumatic experiences-- that, under normal circumstances, the symptoms subside over time as necessary information about the traumatic

event and

how to

deal with it is learned.

However, this

process of adaptation can be disrupted when individuals engage in specific types of thinking and behavior. Several factors can block adaptation, and these include:

* worrying or ruminating about the trauma or one's responses • paying too much attention to threat and danger after the event trying to avoid or excessively control thoughts about the trauma * negative beliefs about the meaning or consequences of symptoms


The nature and pervasiveness of worry/rumination is then highlighted by asking patients about the thoughts they have had during the day about their traumatic ex- perience or reaction to it. This typically results in the


Wells & Sembi

description of many negative thoughts and beliefs and the report of circular thinking based on: "what if


.... and why me?" type questions. It is not a



.... principle focus of treaunent to challenge the content of these ruminative thoughts, but to enable patients to dis- continue this verbal iterative style of negative thinking. Thus, although these "automatic thoughts" are elicited, this is done only to highlight the extent of the patient's rumination. The thoughts are not challenged/balanced in the traditional cognitive-behavioral sense. Patients are introduced to the idea that their intrusive thoughts, flashbacks, nightmares, startle responses, and arousal symptoms are normal and necessary following trauma. The symptoms are a sign that their cognitive sys- tem is attempting to process the trauma and recalibrate or adjust to the event that has taken place. However, their responses and coping strategies have the effect of pre- venting this processing from taking place. The therapist emphasizes that it is important not to avoid these symptoms because they are part of an automatic adaptation process. The next step is to provide an overview of the nature and goals of treatment. The case formulation provides a vehicle for doing this. Notice that in Figure 2 the "adap- tive processing" box is empty. The therapist describes to the patient how treatment will consist of emptying the "maladaptive" box in the formulation and putting new strategies that the patient will learn in treatment into the adaptive box in order to exit the PTSD cycle. The patient is given a copy of the conceptualization to take home and think about before the next session.


Sessions 2 to 3: Worry Postponement and Detached Mindfulness

The aim of the next session(s) is to reinforce aware- ness of the problematic nature of perseveration and to facilitate alternative responding to symptoms. There are three basic components to this: (1) the advantages/ disadvantages analysis, (2) practice of detached mindful- ness, (3) worry postponement.

Advantages~disadvantages analysis.

The first step is to

help clients see that engaging in worry/rumination serves no purpose and contributes to "locking" them into merely replaying negative aspects of the trauma or their dissatis- faction with their own coping responses. The therapist guides the patient through an advantages/disadvantages analysis of worry/rumination as a means of socialization and motivating clients to abandon preservative styles of thinking. The therapist inquires as to whether there are any advantages to rumination and a list of advantages is drawn up. This is followed by drawing up a list of disad- vantages. The disadvantages are prompted by questions

such as the following: What happens to your anxiety when you worry? Does worrying help you move on from the trauma? Is worrying realistic or just negative? Does worrying help you feel

better about yourself?. Does worrying create problems ? Does worry- ing help you see the situation more clearly ?

The next step is to weaken beliefs about the advan- tages of perseveration. Frequently, an advantage to rumi- nation that patients report is that it may help them to find answers. This belief can be weakened by asking the patient why this has not happened yet given that they ap- pear to have spent a considerable amount of time think- ing about what has happened. Patients quickly come to accept that perhaps there are no answers and this there- fore becomes a reason to abandon ruminative thinking. In some instances, as in the case illustrated in Figure 2, patients express the belief that worrying acts as a safety strategy by enhancing preparedness or cautiousness. A two-pronged approach is used here. The disadvantages of worry are contrasted with the advantages with the aim of showing how the disadvantages outweigh the advantages. The therapist then questions whether preparedness and cautiousness can be achieved without worrying (How can

you be cautious without worrying?). The aim here is to show

how worrying and cautiousness are not synonymous and therefore one can decide to reduce worry without sacri- ficing safety. An advantages/disadvantages analysis is also under- taken in examining the motivations for other unhelpful coping behaviors included in the conceptualization such as alcohol use, trying to suppress thoughts, and so on. When thought suppression is a feature of the formula- tion, a within-session suppression experiment is used to show how attempts to avoid and control thoughts can be disadvantageous. Here the therapist asks the patient to try not to think a target thought (e.g., "Try not to think about a blue tiger") for a period of two minutes. Typically patients report that they experience the thought and this is used as an illustration of how trying to suppress thoughts is not particularly effective. Detached mindfulness. Individuals with PTSD repeatedly engage with intrusive thoughts and symptoms in counter- productive ways involving worry/rumination, overcon- trol, attentional monitoring for threat, and negative ap- praisals. Some of these responses exaggerate the current sense of danger, and each of them can interfere with the processes involved in normal adaptation. A goal of treat- ment is to drop these unhelpful influences on adaptation so that normal adaptation processes may resume. An ini- tial step in achieving this consists of training in "detached mindfulness" (Wells & Matthews, 1994), which increases awareness of unhelpful thinking styles, disrupts them, and facilitates flexible control over responding. Detached mindfulness refers to taking a perspective on one's own thought processes in which they are ob- served in a detached way, without interpreting, analyzing, controlling, or reacting to them in any way. Patients are instructed to respond in a particular way when they

Metacognitive Therapy for PTSD


experience intrusive thoughts, flashbacks, and nightmares as follows:

"When you have an intrusive thought, flashback, or have had a nightmare, it is important that you do the following. Acknowledge to yourself that these symptoms are occurring, and remind yourself that engaging with these symptoms is unhelpful. Some people find it helpful to say to themselves: This is


a symptom,

I don't

need to do anything

with it. I


just going to leave it alone.



not going to try to avoid

it or equally ruminate on it. Remember

that engage-

ment with these symptoms includes questioning the meaning of the symptom, trying to work out what has happened to you, ruminating about why it should have happened, asking What if Why ....



Why me ....

or If only

. . .

type questions, worrying

about symptoms, trying to control or avoid thoughts or symptoms. It is important to let your thoughts or symptoms occupy their own space and time without engaging with them."

To facilitate comprehension, several analogies are used in sessions to demonstrate the way that intrusive symptoms should be treated.

Analogy 1. The Recalcitrant Child.

Patients are asked to

treat their intrusions in the same way that they might deal with an annoying child that they had to look after (i.e., they could not avoid). They need to acknowledge that the child was there but paying too

much attention to the child (engaging with it) would merely serve to reinforce its bad behavior, and attempting to punish the child (suppress it)

would upset the child even further. Thus, the best

thing to

do is to just leave the child alone and let it

settle of its own accord.

Analogy 2. Pushing Clouds. Intrusive




treated as if they were clouds in the sky. That is, they

are something that is passing by and something we can do nothing about. They are part of a natural self-regulating weather system and attempting to stop or push them away is not necessary and not possible. Even if we could, this would disturb the balance necessary for rainfall and nature. There- fore, the thing to do is let them occupy their own space and passively watch their behavior over time. A critical treatment component at this stage is ensuring that patients understand the difference between nonengagement in a detached mindful mode, and avoidance of thoughts. Avoidance--for instance, turning attention to other distracting activities--is a form of active engagement with thoughts in the sense that attempts are being made to exclude the thoughts from consciousness.

Practical illustrations. Use of the detached mindful mode and the consequences of engagement with symp-

toms is then illustrated in the therapy session. In one ex- periment patients are asked to first create a mental image of a green tiger, and then to engage with the image by try- ing to exclude all thoughts of tigers from consciousness. This is then contrasted with forming an image of a green tiger and observing the image without doing anything with it. Patients typically report difficulty excluding thoughts of tigers, but find that if they assume detached mindful observation, that thoughts of tigers take on their own life and become less salient. In another experiment,

patients can

be asked to sit quietly and observe in a de-

tached way a bodily sensation. For instance, patients are asked to passively observe the sensations in their mouth without moving their mouth-parts or swallowing for a pe-


of 3 to 4 minutes.

A useful strategy for facilitating detached mindfulness is for the therapist to ask the patient to sit quietly and let

his or her thoughts roam freely during a free-association exercise while observing these internal events. The in- structions for this task are as follows:

"One way to experience the sense of detached mindfulness that will allow you to apply it to your distressing thoughts is to practice first with more general thoughts and feelings. In a moment I will

say a series of words and what I would like you to do

is sit and passively watch

the movement of thoughts

in your mind and feelings in your body as I say different words. For example, I might say the word blue and your task is to watch what happens in your mind and body as a result. Do not try to deliber- ately form any thoughts or activate any feelings or memories--sometimes nothing related to the word may happen, that doesn't matter, you just need to watch your spontaneous thoughts and feelings with- out influencing them. [Pause.] Let's start: apple [pause for 10 secs], ocean [pause], tree [pause], chocolate

[pause], home [pause], birthday [pause], orange juice."

Worry postponement. Once the patient understands the idea of detached mindfulness and in-session practice has been completed, the therapist moves on to reducing worry/rumination. For this purpose the therapist intro- duces the worry-postponement strategy. The instruction is given that whenever intrusive symptoms occur, the pa- tient should acknowledge that the thought/flashback/ nightmare has occurred, and tell him- or herself not to worry or ruminate about the trauma or symptoms now,


let the

symptom fade


its own

time, and


think about it later. Patients are asked to allocate 15 minutes each evening as a designated worry or analysis time. The worry time should take place at least 2 hours before they go to bed,


Wells & Sembi

and they should review the whole day. If patients happen to remember what had been worrying them, they can de- cide to engage in as much worry/rumination as they feel they need to over the 15-minute period. However, it is emphasized that this is not compulsory and many patients decide not to worry. At the end of this period patients are asked to stop worrying and to deal with any further worry as they had in the day, by applying detached mindfulness, and carrying any thoughts over to the next day's worry period if necessary.

Application of detached mindfulness and worrypostponement.

Clients are instructed to apply detached mindfulness for homework in response to intrusive thoughts, flashbacks, and nightmares, and are asked to discontinue daily worry/rumination-based thinking by using the postpone- ment strategy. Careful therapist monitoring is required to ensure that patients are applying the method consistently to the full range of intrnsive thoughts and worry/rumina- tion experienced.





In the next two sessions the therapist monitors progress with detached mindfulness homework and worry post- ponement and facilitates continued practice and general- ization of the techniques. The first issue concerns whether both techniques are being used consistently and frequently. The use of de- tached mindfulness is assessed by asking patients to esti- mate the percentage of time that they have been able to apply detached mindfulness to intrusive thoughts. It is important that the therapist and patient do not confuse this as a rating of the amount of distress. The next ques- tion asked by the therapist assesses if there has been any decrease in usage of the technique over time, and if so,

what the cause of this is. In some cases this is due to a re- duction in distress associated with intrusions. The thera- pist should emphasize that the main aim of the technique

is not to change distress

processing," and therefore

but to "unlock barriers to natural

it is necessary to apply the tech-

nique to most instances of intrusions. The third question the therapist asks concerns the breadth of application of detached mindfulness. It is important that the patient ap- plies it to all types of distressing intrusions related to the trauma and its consequences. In particular, some patients report a specific recurring intrusion that predominates, and having applied detached mindfulness to this intru- sion they notice that other intrusions take precedence, but they do not apply detached mindfulness to these new events as they should. To assess consistency of usage of controlled worry peri- ods, the therapist asks about the amount of time spent worrying/rnminating per day, and how often the patient has succeeded in postponing worry/rumination. Any dropoff of usage of the technique should be explored.

Reduction in the frequency of postponement strategies are to be expected if there has been a reduced frequency of worry. Generalization. The therapist then proceeds to intro- duce the idea that worry/rumination postponement can be applied to all types of worry and persistent negative

thinking. At this stage it helps

to list a range of current

concerns that the patient has had in the past week in order to raise awareness of the pervasiveness of persever- ative thinking, All types of dwelling and worry are then targeted for subsequent homework practice of postponed worry periods. Further practice of detached mindfulness can be im- plemented in these sessions if necessary, The therapist then moves on to introducing the application of detached mindfulness to the after-effects of nightmares. Some pa- tients report that after trauma-related dreams/nightmares they are troubled by thoughts or feelings elicited by them. The therapist instructs patients to apply the tech- nique of detached mindfulness to such after-effects when they occur.

Eliminating other maladaptive strategies. At this point in

treatment, the therapist undertakes a review of the pa- tient's use of other coping strategies that are counterpro- ductive for adaptation. These strategies include use of al- cohol or other substances to avoid thoughts and feelings, thought suppression strategies, avoidance of stimuli such as television news, and so on. The therapist helps the pa- tient to see how these strategies are a problem. For ex- ample, many of these strategies can be seen as a form of avoidance of thoughts and memories of trauma, and this leads to a discussion of the problems caused by cognitive avoidance. Once the patient identifies the unhelpful con-, sequences of these strategies, the therapist asks the pa- tient to ban them for homework.


6 to 7: Attentional Modification

The attentional phase of core treatment is introduced when patients have in the past week: (a) mastered the use of detached mindfulness and reported success in using the strategy in response to at least 75% of intrusive symp- toms, and (b) successfully abandoned worry/rnmination and all forms of dwelling on past, present, and future events such that no episodes last longer than approximately 2 to 3 minutes. In this phase, treatment focuses on hypervigilance, an attentional coping strategy that maintains the perception of danger and anxiety. Two types of attentional monitor- ing strategies are problematic: attention to internal sources of threat (i.e., sensations and feelings) and external at- tention to threat in the form of scanning the environ- ment for danger. Systematic manipulations of attention are an impor- tant component of the core treatment as they shift

Metacognitive Therapy for PTSD


patients out of threat-modes of processing that repeatedly generate information concerning danger. Rather than

persisting in a loop of repeated processing of danger, pa- tients should be moving on to developing a plan for deal- ing with danger, and for controlling cognition that allows threat-related processing to decay. The search for threat is not synonymous with having a plan for dealing with threat once detected, perceiving the self as an effective agent of coping, and allowing cognition to re-rune to the normal



Stage I. Explanation and rationale. The following out-


is used



basis for

therapists to describe the role

that attention plays in the maintenance of PTSD:

'3(ou have seen how worry/rnmination and attempts to control symptoms can maintain your problem, and you have been successful in reducing those responses. We should now consider another impor- tant aspect of the problem that can keep your sense of danger and anxiety going. This is the role played by your focus of attention. Following trauma, it is quite natural for people to become overly aware of people or objects around them that are reminders of the trauma. This is one type of attention that can maintain a sense of danger and stop you from return- ing to a balanced view of the world. For some people, there is a tendency to focus too much on internal thoughts about the trauma or anxiety symp- toms. For instance, when in a situation similar to that in which the trauma occurred, the person focuses on a memory or picture of what happened. This is often an image fragment of a particular moment, which may be the worst moment. Focusing in this way increases the sense of threat and anxiety, and takes attention away from focusing on current events that could provide a better sense of safety and control."

The rationale is illustrated by asking questions concern-

ing the consequences of idiosyncratic threat-monitoring strategies. For instance, the therapist asks: Do you think there are any problems with constantly scanning the environment

for signs of threat? Is scanning for threat likely to increase or de- c,ease your anxiety ? Does paying attention to threat give you a balanced picture of how safe a situation is? Does paying atten- tion to threat mean you will cope better?This process is under- taken for external attentional monitoring for threat and also for internal monitoring. The therapist therefore moves toward a conceptualization of hypervigilance as being another form of unhelpful preoccupation similar to worry/rumination. Before the patient is willing to give up threat monitor- ing, it is often necessary to weaken the positive beliefs supporting its usage. The therapist does this by question- ing whether hypervigilance would have actually averted

the traumatic event, how the person would know exactly what to be hypervigilant for, and by examining counter- evidence concerning the potential unhelpful role of hy- pervigilance. The following transcript illustrates a typical line of questioning used to raise awareness of the role of attention and to weaken beliefs about its usefulness:


Have you noticed that what you pay attention



has changed since you were attacked? I'm not sure.

T: For instance,


you find

that you notice



things more than you did before?



I've noticed

how much







seems to be in the news. Do you think that is because

crime has suddenly

increased since your assault, or has something else


P: Well


obviously in

my mind,



way I'm


thinking about things.







observation. Something

has changed in what you pay attention to. Has your attention changed in any other way? For example, what do you pay attention to when you go out now? P: I'm on the lookout for groups of youths, and when I see them I walk the other way. T: Any other changes to what you look for?


I'm always looking to see if I can see anyone look- ing suspicious.



you think


are any problems with using

your attention in this way?



Well, it makes


feel safe,


if I'd



before maybe I would have been safe.



That sounds like an advantage. If you had been like this, would that have prevented the attack?

P: No, probably not, as they were not acting suspiciously.


So it may not have helped.



see any


advantages of doing this? For example, does it help

you to feel calm when you are out?







opposite, because I see danger




So the



is there

really danger


where or is your strategy keeping your anxiety and stress going? P: I'm keeping it going. T: So we need to take a look at doing something about your attention.

Stage 2: Awareness and abandonment. Once the problem with threat monitoring is understood, the therapist asks the patient to consciously acknowledge the direction of their attention the next time they feel anxious in a situa- tion and to stop threat monitoring. In order to apply this technique patients are encouraged to return to their nor- mal routine of daily life. In most cases this does mean


Wells & Sembi

returning to the situation in which the trauma occurred, or in similar situations. This is the only point in the core treatment where a degree of in-vivo exposure may take place. However, it is not habituation but the facilitation of awareness of and disruption of threat monitoring that is the goal of this procedure.

Session 8

The aims of the next session are to review progress with abandonment of threat monitoring and its applica- tion during the patient's normal daily routine. The first thing assessed by the therapist is the extent to which the patient has been returning to his or her normal pretrauma routine. At this stage, depending on the nature and sever- ity of threat, there should be some indication of a return to situations that were usually frequented. If avoidance of low-risk situations is an issue, then patients are encour- aged to go into these situations for homework while prac- ticing abandonment of threat monitoring. Attention refocusing. After abandonment of threat mon- itoring, the next step is active attention refocusing, con- sisting of asking patients to deliberately redirect attention away from themselves and away from threat, and onto nonthreatening aspects of the external environment when in situations that remind them of the trauma. (Note, however, that in our preliminary evaluation of the effects of the core treatment, we did not use this addi- tional strategy. We found that it was unnecessary as pa- tients responded well to the basic instruction to be aware of and abandon threat monitoring.) The therapist intro- duces the idea that, "in order to allow thinking to re-tune to the normal environment it is helpful to practice focus- ing attention on the environment in a benign way. This means looking for signs of safety instead of signs of im- probable threat." This is done by practicing different fo- cusing strategies during the treatment session. First, the therapist asks the patient to sit in the waiting room and focus on aspects of the environment that signal that it is a safe place. This is followed by walking in the street with the therapist and practicing focusing on safety signals. Fi- nally, a strategy is practiced involving focusing on neutral external stimuli (i.e., unrelated to concepts of danger or safety) such as focusing on the array of different colors that can be seen.

Relapse Prevention

During the final session of treatment the original for- mulation is discussed with personalized examples from the patient's experience of how tackling rumination via the use of detached mindfulness and controlled worry periods has placed the individuals' concerns in perspec- tive. Should patients find themselves disturbed by memo- ries of the trauma in the future, they are advised to look for signs of worry/rumination. On noticing worry or ru-

mination they should once again adopt the techniques they have learned until the symptoms subside.

Summary and Conclusions

In this article we have presented our core treatment protocol for metacognitive therapy of PTSD. This treat- ment is based on a model in which PTSD results from the failure to meet an intrinsic goal following trauma. This goal consists of developing a metacognitive plan that serves as a blueprint for guiding cognition and action in dealing with subsequent threats. It is assumed that plan compila- tion normally proceeds unhindered over a time course stimulated by the processing of intrusive symptoms. Such symptoms provide an imprint that has to be worked upon by upper-level processing to establish a plan for cognition and coping. The flexible processing of imagery provides one medium for running mental simulations of coping with trauma. Several factors interfere with these normal adaptation processes including: (a) worry/rumination that diverts resources away from running simulations and selectively focuses the individual on additional sources of threat; (b) threat monitoring that perpetuates percep- tions of threat and strengthens a danger-awareness plan rather than coping plan; (c) avoidant types of coping, in- cluding thought control, that interrupt the normal work of intrusions; (d) negative self-appraisals/beliefs about symptoms and coping. These factors block adaptation and prevent cognition from re-tuning to the normal threat-free environment. A role of worry/rumination and maladaptive metacog- nitive control strategies in the persistence of stress symp- toms is supported by data from experimental and corre- lational studies of patients and nonpatients. Moreover, a study of the longitudinal predictors of PTSD following road traffic accidents provides support for the idea that worrying as a means of coping with unwanted thoughts is causally related to the development of PTSD even when stress symptoms at Time 1 are controlled. A preliminary evaluation of the effectiveness of the metacognitive treatment has produced encouraging re- sults (Wells & Sembi, 2004). In this study, six consecutive patients (five females, one mate) referred for treatment following a variety of traumas (armed robbery, physical assault, rape) received the new treatment. The duration of PTSD ranged from 3 to 7 months across cases and each patient showed a stable baseline of symptoms for 4 weeks before the commencement of treatment. All participants met DSM-IV criteria for moderate-severe PTSD and all met criteria for major depressive disorder during base- line. All participants lived in or around a deprived inner- city area. A further two patients (the next consecutive re- ferrals, both male victims of physical assault) treated were subsequently included for effect-size analysis. All patients

Metacognitive Therapy for PTSD


showed marked improvement in PTSD symptoms and general symptoms of anxiety and depression. None of the patients met criteria for PTSD at posttreatment or follow- up on the Posttraumatic Stress Diagnostic Scale (Foa, 1995).

Pretreatment and posttreatment means on outcome mea- sures were as follows (pre first): Impact of Events Scale, 53.6-10.3; Davidson Trauma Scale, 92.5-15.5; Penn Inventory, 48.2-14.8; BAI, 31.9-3.5; BDI, 20.4-3.6. The treatment effects were statistically significant, and post- treatment effect sizes across measures ranged from 3.0 to 5.0. Treatment gains were maintained at follow-up (3 and

6 months,




41 months).

These preliminary findings support continued evalua- tion of the treatment, and further evaluations are in progress. Controlled studies and developments of this new treatment approach are required. It is our hope that this treatment manual will provide the basis for further

evaluations and implementations of treatment by other clinicians working in the area of PTSD.


Buffer, G., Wells, A., & Dewick, H. (1995). Differential effects of worry and imagery after exposure to a stressful stimulus: A pilot stud}~ Behavioural and Cognitive Psychotherapy, 23, 45-56. Foa, E. (1995). Posttrnamatic Stress Diagnostic Scale Manual. Minneapo- lis: National Computer Systems. Holeva, V., Tarrier, N., & Wells, A. (2001). Prevalence and predictors of acute stress disorder and PTSD following road traffic accidents:

Thought control strategies and social support. Behavior Therapy, 32, 65-84. Nolen-Hoeksema, S., & Morrow, J. (1991). A prospective study of depression and posttraumatic symptoms after a natural disaster:

The 1989 Loma Prieta earthquake.Journal of Personality and Social Psychology, 61, 115-121. Nolen-Hoeksema, S., Parker, L. E., & Larson, J. (1994). Ruminative coping with depressed mood following loss. Journal of Personality and Soeial Psychology, 67, 92-104.

Papageorgiou, C., & Wells, A. (1999). Process and meta-cognitive dimensions of depressive and anxious thoughts and relationships with emotional intensity. Clinical P~chology and Psychotherapy, 6,


Papageorgiou, C., & Wells, A. (2003). Rumination: Nature, theory and treatment of depressive thinking. Chichester, UK: Wiley. Reynolds, M. & Wells, A. (1999). The Thought Control Questionnaire:

Psychometric properties in a clinical sample, and relationships with PTSD and depression. Psychological Medicine, 29, 1089-1099. Richards, D. A., Lovell, F,2, & Marks, I. M. (1994). Post-traumatic stress disorder: Evaluation of a behavioural treatment program.Journal of Traumatic Stress, Z 669-680. Scott, M.J., & Stradling, S. G. (1997). Client compliance with exposure treatments for post-traumatic stress disorder Journal of Traumatic Stress, I0, 523-526. Sherman, J. J. (1998). Effects of psychotherapeutic treatments for PTSD: A meta-analysis of controlled trials. Journal of Traumatic Stress, 11, 413-436. Tarrier, N., Pilgrim, H., Sommerfield, C., Faragher, B., Reynolds, M., Graham, E., & Barrowclough, C. (1999) A randomised trial of cognitive therapy and imaginal exposure in the treatment of chronic posttraumatic stress disorder. Journal of Consulting and Clinical Psychology, 67, 13-18. van Etten, M. L., & Taylor, S. (1998). Comparative efficacy of treat- ments for post-traumatic stress disorder: A recta-analysis. Clinical Psychology and Psyehotherapy, 5, 126-144. Warda, G., & Bryant, R. A. (1998). Cognitive bias in acute stress dis- order. Behaviour Research and Therapy, 36, 1177-1183. Wells, A. (2000). Emotional disorders and metaeognition: Innovative cogni- tive therapy. Chichester, UK: Wiley Wells, A., & Davies, M. I. (1994). The Thought Control Questionnaire:

A measure of individual differences in the control of unwanted thoughts. Behaviour Research and Therapy, 32, 871-878. Wells, A., & Matthews, G. (1994). Attention and emotion. A clinicalperspec- tire. Hove, UK: Erlbaum. Wells, A., Xc Matthews, G. (1996). Modelling cognition in emotional dis- order: The S-REF Model. BehaviourResearch and Therapy, 32, 867-870. Wells, A., & Papageorgiou, C. (1995). Worry and the incubation of intrusive images following stress. Behaviour Research and Therapy, 33, 579-583. Wells, A., & Sembi, S. (2004). Metacognitive therapy for PTSD: A pre- liminary investigation of a new brief treatment. Journal of Behavior Therapy and Experimental Psychiatry, 35, 307-318.

Address correspondence to Adrian Wells, Ph.D., University of Man- chester, Division of Clinical Psycholog5 Rawnsley Building, Manchester Royal Infirmary, Manchester, M13 9WL UK; e-mail: adrian.wells@ man.ac.uk.