تغییرات در خاطرات مزاحم تصورات و پندارها در اختلال استرس پس از سانحه در ارتباط است

Abstract The study investigated changes in intrusive memories associated with imaginal reliving of traumatic events in posttraumatic stress disorder (PTSD). The study population comprised 44 patients treated with imaginal reliving in the context of cognitive therapy for PTSD [Behav. Res. Ther. 38 (2000) 319–345]. For most patients, imaginal reliving did not lead to exacerbations in intrusion frequency. The decrease in intrusion frequency after reliving was gradual, as was the decrease in their distress, vividness, and perceived “nowness.” Poorer outcome, i.e., a smaller reduction in residual gain scores for intrusion frequency with reliving, was associated with greater initial PTSD severity, greater anger, greater perceived “nowness” of intrusive memories, and more negative interpretations of PTSD symptoms. The patient's anxiety, depression, self-blame, and dissociation were not predictive of response to reliving.

1. Introduction Intrusive memories are a core symptom of posttraumatic stress disorder (PTSD). They usually consist of relatively brief sensory fragments of the traumatic experience that occur in a stereotyped, repetitive way (e.g., Ehlers, Hackmann, & Michael, 2004; van der Kolk & Fisler, 1995). Recent research has pointed to several features of intrusive memories that may be of clinical and theoretical interest. First, emotions that accompany intrusive memories in PTSD are the same as those experienced at the time of the trauma (Brewin, Dalgleish, & Joseph, 1996; Foa & Rothbaum, 1998), and the sensory components of intrusive memories appear to be reexperienced as if they were features of something happening right now, rather than being aspects of memories from the past (Hackmann, Ehlers, Speckens, & Clark, 2004). Second, as Ehlers and Clark (2000) highlighted, the intrusive memories appear to lack a context of other relevant information, so that the intrusive memory is reexperienced even if the person later acquired new information that contradicted their original impression. The relevance of the lack of time perspective and lack of context of intrusive memories in PTSD was supported by two studies by Michael, Ehlers, Halligan, and Clark (in press), who found that these intrusion characteristics distinguished between assault survivors with and without PTSD and predicted subsequent PTSD severity. Whether or not participants reported intrusive memories in the first few weeks after the assault only explained 9% of the variance of PTSD severity at 6 months after assault. Among survivors with intrusions, intrusion frequency only explained 8% of the variance of PTSD symptom severity at 6 months. Lack of time perspective (perceived “nowness” of the memories), distress and lack of context explained an additional 43% of the variance. These intrusion characteristics also predicted PTSD severity at 6 months over and above what could be predicted from PTSD diagnostic status at initial assessment. The present study investigated whether and how these intrusion characteristics change during treatment. 1.1. Change of intrusive memories with therapy Prolonged imaginal and in vivo exposure has been established as an effective treatment for PTSD (Foa, Dancu, et al., 1999; Keane & Kaloupek, 1982; Marks, Lovell, Norshirvani, Livanou, & Trasher, 1998; Tarrier et al., 1999). In imaginal exposure (imaginal “reliving”), patients are asked to relive the traumatic experience in the presence of the therapist, starting from the beginning and continuing to the point when they were safe again, while putting the experience into words (Foa & Rothbaum, 1998). In repeated relivings, the patients add more and more detail until all aspects of the traumatic memory are tolerated well. Sessions are audiotaped and patients are asked to listen to the tape for several times in between sessions. Given the predictive power of intrusion characteristics for the persistence of PTSD (Michael et al., in press), one would expect that they change with successful treatment. A preliminary study of Hackmann et al. (2004) in 22 patients with PTSD supported this hypothesis. The authors found that frequency, distress, vividness and perceived “nowness” changed with imaginal reliving (delivered in the context of a cognitive therapy program) and that the change was gradual rather than abrupt. In accordance with earlier findings by Foa, Zoellner, Feeny, Hembree, and Alvarez-Conrad (2002), reliving of the traumatic event in the initial treatment sessions increased the frequency of intrusive memories in only a minority of patients. 1.2. Predictors of treatment outcome Several studies have aimed to identify predictors of response to exposure-based PTSD treatments. Foa, Riggs, Massie, and Yarczower (1995) found that patients who showed anger rather than fear when reliving their traumatic event had a poorer treatment response. Van Minnen, Arntz, and Keijsers (2002) investigated predictors of treatment outcome in two groups of 59 and 63 PTSD patients with mixed traumas treated with prolonged imaginal exposure. The only stable predictor of outcome across the two groups, both post-treatment and at follow-up, was initial severity of PTSD symptoms. Use of benzodiazepines was also predictive of treatment outcome, but demographic variables, depression and general anxiety, personality, trauma characteristics, and feelings of anger, guilt, or shame were not. Blanchard et al. (2003) found that pretreatment severity of PTSD symptoms and functional impairment predicted outcome of treatment in 30 patients treated with a cognitive behavioral treatment for their PTSD. In a study of 50 people who received a cognitive behavioral treatment for PTSD following a road traffic accident, Taylor et al. (2001) also found that poor treatment outcome was related to lower levels of functioning, but also to greater numbing, anger, depression and pain. In 62 patients with PTSD treated with either cognitive therapy or prolonged exposure, Tarrier, Sommerfield, Pilgrim, and Faragher (2000) found that gender and suicide risk were associated with post-treatment outcome and comorbid generalized anxiety disorder and residential status with outcome at 6 months follow-up. 1.3. Other predictors of chronic PTSD The search for predictors of treatment response may be informed by recent theoretical and empirical work on cognitive factors that predict chronic PTSD. For example, Ehlers and Clark (2000) developed a cognitive model of PTSD suggesting that PTSD becomes persistent when individuals process the trauma in a way that leads to a sense of serious, current threat. The sense of threat arises as a consequence of (1) a disturbance of the autobiographical memory for the trauma characterized by poor elaboration and conceptualization, and (2) excessively negative appraisals of the trauma and/or its sequelae. Several studies supported the role of these memory and appraisal factors. For example, in a large prospective study of road traffic accident survivors, Ehlers, Mayou, and Bryant (1998) found that initial negative interpretation of intrusive memories, anger cognitions and dissociation during the accident predicted PTSD at 1 year. Similarly, Halligan, Michael, Clark, and Ehlers (2003) found in two studies of assault victims that negative interpretation of intrusive memories and dissociation were significantly correlated with both concurrent and subsequent PTSD severity. The present study investigated whether these cognitive factors are useful in predicting treatment response to imaginal reliving. 1.4. Aims of the study The present study had two aims. First, we aimed to replicate the preliminary findings of Hackmann et al. (2004) about the changes in intrusive memories with imaginal reliving in a separate and larger sample of patients with PTSD. Second, we aimed to identify possible predictors of changes in intrusion frequency in response to imaginal reliving. The hypotheses were as follows: (1) Decrease in frequency of intrusive memories after reliving will be gradual rather than sudden. (2) Imaginal reliving will not lead to an increase in intrusive memories. (3) The following factors will predict a smaller decrease in the frequency of intrusive memories after reliving: initial severity of PTSD symptoms, anxiety, depression, anger, dissociation, characteristics of intrusive memories and negative interpretation of PTSD symptoms.

Results 3.1. Study sample The sample consisted of 12 (27%) male and 32 (73%) female patients. The mean age was 39.4 (S.D. = 11.3) years. Twenty-five (57%) of the patients were married or co-habiting, 14 (32%) were single, 4 (9%) divorced and 1 (2%) widowed. The majority of the patients were employed (N = 37, 84%, including patients on sick-leave), 3 (6%) were students, 2 were homemakers (5%), and 2 (5%) were unemployed. Fourteen (32%) had professional jobs, 14 (32%) white collar, and 12 (27%) blue collar jobs. Twelve patients (27%) were using psychotropic medication, mostly tricyclic antidepressants (N = 7, 16%) or selective serotonin re-uptake inhibitors (N = 4, 9%). Patients had to be on a stable dose for 2 months before the first treatment session to be included in the trials. Medication was not related to outcome. The severity of symptoms at the beginning of treatment is shown in Table 1. Table 1. Severity of symptoms, intrusion characteristics, and other predictor variables at initial assessment; and their association with changes in intrusion frequency Mean (S.D.), N = 44 Correlation with residual gain scoresa Posttraumatic symptoms (PDS) 25.9 (8.0) .47** Anxiety (BAI) 20.0 (9.2) .11 Depression (BDI) 18.6 (6.5) .16 Main intrusive memory Frequency (week) 5.38 (5.69) N/A Vividness 76.8 (20.3) .20 Distress 67.9 (24.3) .14 Nowness 57.2 (26.1) .36* Self-blame (PTCI) 11.8 (7.1) .06 Negative interpretation of symptoms (PTCI) 15.5 (6.3) .31* Anger (PDS) 2.3 (.9) .34* Dissociation (TDQ) 50.0 (25.0) .03 PDS: Posttraumatic Diagnostic Scale, PTCI: Posttraumatic Cognitions Inventory, BAI: Beck Anxiety Inventory, BDI: Beck Depression Inventory, TDQ: Trait Dissociation Questionnaire, N/A: not applicable. a Greater scores reflect poorer outcome (smaller decreases in intrusion frequency). * P < .05, two-tailed. ** P < .001, two-tailed. Table options 3.2. Course of intrusive memories The frequency of the patients’ main intrusive memory from the assessment prior to treatment to the fourth session after reliving is shown in Fig. 1. The frequency did not change between initial assessment (mean = 5.5 per week, S.D. = 5.8) and the first reliving session (mean = 4.9, S.D. = 4.1, P > .57), or between the first reliving session and the following session (mean = 5.0, S.D. = 6.0, P > .86). By the second session after reliving, frequency declined significantly to a mean of 3.0 (S.D. = 3.1, P < .005) and remained significantly decreased for the following sessions (all P's < .001). Session-to-session changes in the mean frequency of the patients’ main intrusive ... Fig. 1. Session-to-session changes in the mean frequency of the patients’ main intrusive memory. Figure options The decline of vividness, distress and nowness is shown in Fig. 2. Only patients who still reported to have intrusive memories were included in this analysis, i.e., 37 by session 2, 32 by session 4 and 27 by session 5. Even though they still reported to have intrusive memories, the vividness, distress and nowness gradually declined over the course of treatment. Vividness, distress and perceived nowness of the patients’ main intrusive memory ... Fig. 2. Vividness, distress and perceived nowness of the patients’ main intrusive memory (for patients who still report intrusive memories at each time point). Figure options 3.3. Prediction of outcome Table 1 shows the correlations between the predictor variables before treatment and the patients’ response to reliving in terms of residual gain scores for intrusion frequency. Low initial PTSD symptom severity, low nowness of the intrusive memory, low negative interpretations of PTSD symptoms and low anger were predictive of greater decreases in intrusion frequency.