My chosen question: Which treatment model provides the highest treatment effect for PTSD. The answer is not as simple as originally believed. Let's read...
Exploration on the efficacy of treatments for PTSD
This paper is intended to serve as an overview of treatment models for Post Traumatic Stress Disorder (PTSD). PTSD is seen with increasing prevalence in the United States, likely because of current military operations in Iraq and Afghanistan. Other potential causes of PTSD include childhood abuse, sexual abuse, rape, and exposure to violent crime.
Neurobiology and Neuropsychology of PTSD
The neurobiology of PTSD is an emergent topic in modern psychology. Many tools are used to investigate the underlying neural systems that are effected by, and expressed from, PTSD. While the label PTSD is just a name for a cluster of symptoms consistent with the DSM–IV–TR description, there are actual changes on the physical, neurological layer in the brain that tends to correspond with expression of PTSD symptoms.
The authors of Neurobiology of PTSD: A Review of Neuroimaging Findings, Garfinkel and Liberzon, 2009, discuss four areas that show changes in the neuronal activity in response to exposure to trauma, which trigger expression of PTSD symptoms. Further, they note five different tools for neural imaging: single–photon emission tomography (SPECT), positron emission tomography (PET), magnetic resonance spectroscopy (MRS), magnetic resonance imaging (MRI), and functional magnetic resonance imaging (fMRI).
First, the authors discuss the effects of trauma on the hippocampus, noting that “[t]he hippocampus is a known target for stress hormones” (Garfinkel & Liberzon, 2009, p. 371). The reason that the hippocampus is so important, the authors state, is the role of the hippocampus in all they types of memory – from episodic to declarative, to sensory short term memory (p. 371). The authors note that the volume of the hippocampus, following trauma, has been noted to reduce by “5% to 26%, and have tended to be found bilaterally across studies” (p. 371). This seems to indicate that the entire hippocampus is affected by trauma, not just portions of the hippocampus. The authors mention that the reductions are not seen in all cases, and mention the limitations of analyzing just the hippocampus for PTSD–effects. The authors note that it is currently unclear whether a reduced hippocampus is a PTSD–byproduct or a predisposing factor for PTSD symptoms. Twin studies, long the mainstay of determining the relative effects of environment and biology, have been inconclusive, largely because of small sample size. The authors note one study on Vietnamese twins, one with PTSD and one without – but both exposed to trauma, have shown both reduced size and impairment in hippocampus–related functional tests (p. 371). The authors indicate that this study seems to indicate that reduced hippocampus size is a predisposing factor (p. 371).
The anterior cingulated cortex (ACC) is another area that shows change in response to exposure to trauma. The authors state, “the ACC shares close neuroanatomical relationships with subcortical components of the ‘central fear system’” (Garfinkel & Liberzon, 2009, p. 371). The ACC has been demonstrated to be “smaller in patients with PTSD, relative to control patients” (p. 371). This reduction, the authors state, is in the grey matter, and that meta-studies suggest that the ACC is smaller in patients with PTSD, compared to trauma-exposed controls (pp. 371–2). Twin studies, the authors note, have shown that the ACC in combat–exposed twins is significantly smaller than non–combat exposed twins (p. 372). The authors discuss the functional structures of the ACC, indicating that the dorsal ACC is involved in cognitive functions, while the ventral ACC is in involved in emotional information (p. 372). The authors note that studies indicate that combat–exposed veterans have an increased regional cerebral blood flow (rCBF) in the ventral ACC, and decreased activity in the dorsal ACC, as well as other areas involved in mood (p. 376). The authors note that studies reveal that reduced ACC activation may also be characteristic of pediatric PTSD (p. 376). The authors note that most studies have focused on adults; however, studies with adolescent subjects corroborate the research with adults and show its applicability to children (p. 376). The authors state that neuroimaging has “repeatedly demonstrated ACC to be implicated in the neurocircuitry associated with PTSD (p. 377). The author notes several studies, using PET and fMRI, that agree in the lower activation of the ACC, and reduced rCBF in the ACC. “Taken together, these data provide compelling evidence that functional impairments of the ACC and mPFC [medial prefrontal cortex] are associated with the neurocircuitry implicated in PTSD, and this holds for both symptom provocation and cognitive activation studies” (p. 377).
The amygdala is another structure that is involved with emotional processing (Garfinkel & Liberzon, 2009, p. 377). The amygdala, the authors note, “is a region implicated in rapidly assessing the salience of emotional and especially threat–related stimuli”, in addition to its role in “generat[ing] and maintenance of emotional responses” (p. 377). Imaging studies shown limited reduction in size, but thus far not any statistically significant size differences (p. 372). However, the authors note a study by Rauch, et al., which shows the increased rCBF in the limbic structures (p. 377). Another study confirmed the activation of the amygdala in PTSD, subjecting combat–exposed veterans to combat noise or white noise; only the combat noise subjects experienced increase rCBF in the amygdala (p. 377). The authors note that the findings suggest that “PTSD is associated with increased amygdala responsivity to threat–related […] stimuli” (p. 377).
The insula is a large cortical structure which connects to the amygdala on the anterior, and the sensory cortex on the posterior (Garfinkel & Liberzon, 2009, pp. 377–8). The dorsal anterior insula, the authors state, is responsible for “developing and updating emotional states, autobiographical memory, cognitive control, affective processing, and pain” (p. 378). The authors note that “[a]bnormal functioning of the insula has been linked to anxiety”, and imaging studies have linked the insula with “anticipating the arrival of aversive stimuli” (p. 378). The insula is one of the “key regions involved in PTSD,” and especially the anterior portion of the insula (p. 378).
Another article that develops a neuropsychological perspective of PTSD is that of Perry, et al. . The article Childhood Trauma, the Neurobiology of adaptation, and “Use–dependent” Development of the Brain: How “States” Become “Traits”, describes PTSD symptoms, and their activity on the developing brain. In a child, responses occur on a continuum from “calm” to “terror,” and these responses correspond to the brain level at which the subject is operating on. (Perry, et al., 1995, p. 274). As stress increases, the level of processing decreases in a linear fashion, from top to bottom of the brain (p. 274). Calm is linked with the neocortex, and involves abstract cognition; arousal is linked with the subcortical system, and is involved with concrete thought; alarm is linked with the limbic system, and is involved with emotional thought; fear is linked with midbrain structures, and is involved in reactive thought; and terror is linked with the brainstem, and is involved in reflexes (p. 274). The authors state that when threatened, the individual moves down the continuum, and arrives at the highest level capable of reacting to a threat. The lower the level of processing, the more “primitive” the behavior and cognition becomes (p. 274). The authors state that state–dependent memory is built into the organizing framework of a child, and the trauma can become the organizing framework for new experiences (p. 275). The brain “develops in a sequential and hierarchical fashion” from back to front, and right to left, and the higher brain regions become fully functional at different times during childhood (p. 276). Because the layers of the brain are built one atop another, the developing brain is crucially dependent upon stability at both the “critical periods” and the “sensitive periods”, and disruption of these periods are implicated in deficits in neurocognitive functioning (p. 276). The authors state that “although experience may alter the behavior of an adult, experience literally provides the organizing framework for an infant and child”, and the enormous plasticity of a developing brain makes childhood critical for proper development (p. 276).
The article is primarily concerned with two types of childhood responses to trauma: arousal and dissociation. These responses are the familiar “fight or flight”, only expressed differently (Perry, et al., 1995, p. 277). Children exposed to trauma can express a hyperarousal response to trauma. This response plays a role in “vigilance, affect, behavioral irritability, locomotion, attention, […], sleep and the startle response” (p. 278). When a child is exposed to trauma, the hyperarousal state is stored as a state–dependent memory; upon a reminder of the traumatic event, these systems are “reactivated” at the heightened level of activation, and these reminders may generalize (p. 278). The authors note “everyday stressors that previously may not have elicited any response now elicit an exaggerated reactivity” (p. 278).
Another behavioral pattern is that of the dissociative response, which the authors call the “freeze or surrender responses (Perry, et al., 1995, p. 279). Children, especially younger children and females, use this response, and report going “to a ‘different place,’ or assuming persona of heroes or animals, a sense of ‘watching a movie that I was in’, or ‘just floating’" (p. 281). This response seems to be adaptive if the child lacks control of a situation, and there seems to be nothing they can do to avoid being victimized (p. 283).
An important characteristic of these responses is that repeated use of either style of threat response lowers the amount of perceived–threat needed to elicit the same response, a concept called use–dependent sensitization (Perry, et al., 1995, pp. 275–8). Neurons in the fear–system will need less and less input to put out a high response to threat, and thus these “states” can very well become “traits”, disrupting both life as a child and life as an adult (p. 275).
WIIFM – What’s In It For Me (or why I should care about neuropsychology?)
Neuropsychology is a rapidly developing field. It is also expensive and not widely available to clinicians. The question is continuously brought up: “This is all great stuff, but why do I need to know about it?” The answer lies in the practical applicability of the research.
Neuropsychology has provided clinicians with tools to assess brain functioning, such as the Wisconsin Card Sorting Task for frontal lobe functioning or the Stroop Test for assessing the ACC. Knowledge of the effects of impairment in the brain must inform the treatment. A child who externalizes their trauma can present symptoms that can look like a conduct disorder, but the interventions for conduct disorder and PTSD are very different. An adult with PTSD who experiences memory deficits will need both occupational therapy and other interventions that target working around the presenting memory deficits.
Treatment must be targeted to the level of brain functioning the client presents with. A highly anxious client may be receptive to anti–anxiety interventions, such as breathing or relaxation exercises. Other clients may present with anger or aggression, and treatment needs to be directed to reduce the incidence of the expression of the “fight” portion of the “fight–or–flight” response. A clinician who can adjust treatment to the level of functioning of the client can achieve a targeted and efficient response, tailoring treatment to the individual. Clinicians need to know what the differences in abilities of lower portions of the brain in order to decide which treatment intervention is necessary.
Another facet is the psychopharmaceutical approach, a client with high anxiety may need a short–term anxiolytic to get to a low level of stress before examining the source of anxiety, just as a depressed presenting client may need short–term anti–depressants to be able to meaningfully evaluate and interpret the cause of depression. These interventions need to occur before addressing the trauma.
Cognitive Behavioral Therapy
Perhaps the most empirically validated, and highly utilized, treatment model, is that of cognitive behavioral therapy (CBT). CBT is centered on a triad of thoughts, emotions, and behaviors. The literature suggests that an intervention of any one of these can influence the others. A key article for understanding the cognitive–behavioral model is Toward a Cognitive–Behavioral Model of PTSD in Children and Adolescents, 2002, by Meiser–Stedman. The article discusses two models of PTSD, and the applicability of those models to children.
Underlying CBT theoretical models of PTSD
The first model, by Brewin, Dalgleish, and Joseph, states that “memories of the traumatic event are stored in a manner different to that of normal memories,”, and that these memories are encoded in different neural structures than normal memories (Meiser–Stedman, 2002, p. 220). The Brewin model is informed by neuropsychology, and suggests that “’traumatic memories’ are laid down in a way that bypasses the hippocampus” (p. 220). The hippocampus is responsible for processing and consolidation of memories (p. 220). The traumatic memories are saved with the “sensory […], physiological, and motor aspects” that allow the “memory to be recreated” (p. 220). These “situationally accessible memories” bypass the hippocampus, and are elicited as a result of trauma–related cues (p. 220). The difference in pathways, compared to normal memories, “means that traumatic memories are not easily accessible by conscious means” (p. 220). Part of the consolidation of autobiographical memories is their conversion to verbal, explicit facts. These “secondary emotional reactions arising from subsequent conscious appraisal may interfere with the emotional processing of the traumatic experience” and impede the memory consolidation processes (p. 220). The author stresses that these secondary emotions must be dealt with prior to exposure treatment (p. 221). Completion of treatment is the integration of the memory into the sense of self, and is a key approach in Brewin’s model of PTSD (p. 221). When the memories are integrated, they are said to have been “worked through” (p. 221). When PTSD symptoms persist, the subject is undergoing “chronic emotional processing”, which may be the result of secondary emotions interfering with the integration of the trauma memory (p. 221).
The second model, proposed by Ehlers and Clark, is built upon the Brewin model. The Ehlers model focuses on the “pathological role of the trauma memory” and the “cognitions [and metacognitions [...] involved in the maintenance of PTSD” (Meiser–Stedman, 2002, p. 221). The PTSD symptoms serve to reinforce the sense of “current threat”, which continues the maintenance of heightened arousal (p. 221). Further, negative evaluations, such as a “dysfunctional meaning attached to symptoms […,] perceived negative responses from others […,] a sense of permanent change […,] and a change in global beliefs” add to the secondary emotions, and interfere with normal processing of the traumatic event (p. 221). The dysfunctional cognitions and metacognitions “produces a sense that the trauma continues to have damaging implications”, generating apprehension, and maintaining the sense of current threat (p. 221). Thought control, where a subject actively tries to avoid thoughts, paradoxically, increases the frequency of the intrusive thoughts (p. 221).
“The early onset of intrusive phenomena is considered a normal reaction to an extreme event [but the] maintenance of such symptoms is considered to result from a variety of maladaptive responses”, meaning that the subject is using the right reaction to the wrong situation (Meiser–Stedman, 2002, p. 222). The child’s response to PTSD is concordant with the adult models, but a child model must accommodate differences in neurodevelopment (p. 222).
More treatments applied to a subject do not mean better outcomes
Another article describing the use of CBT is Randomized Trial of Prolonged Exposure for Posttraumatic Stress Disorder With and Without Cognitive Restructuring: Outcome At Academic and Community Clinics, by Foa, et al. The article describes a randomized trial of two treatment groups: prolonged exposure, and prolonged exposure with cognitive restructuring. Of an initial intake group of 285, 179 were admitted for treatment in this study, with 96 completing the trials, and 26 waitlisted. Subjects were primarily rated on the PTSD Symptom Scale—Interview, but were also evaluated on the Beck Depression Inventory, the Social Adjustment Scale, and the PTSD Symptom Scale—Self–Report. Subjects were evaluated pretreatment and post-treatment at three, six, and 12–month intervals (Foa, et al., 2005, p. 956). Both groups were seen once a week for 90 to 120 minute sessions (p. 957).
The prolonged exposure group was made to close their eyes and recount the story of their trauma as many times as would fill 45–60 minutes. A tape recording was made for the subjects, and the subjects were encouraged to listen to the tape daily (Foa, et al., 2005, p. 957). In the first session, subjects made a list of activities they avoided because of feelings associated with the trauma, and the beliefs associated with the avoidances were evaluated in the final session (p. 957).
The prolonged exposure–cognitive restructuring group was “identical to the [prolonged exposure group] alone with two exceptions” (Foa, et al., 2005, p. 957). First, the clinician presented “the idea that post-trauma symptoms are maintained in part by trauma–related thoughts and beliefs”, and the subjects were provided information on the use of cognitive restructuring to break out of the cycle of maladaptive thoughts. Second, subjects were instructed to utilize a daily diary, recording their thoughts, assertions, and attitudes – and to challenge the maladaptive statements (p. 957).
Waitlisted subjects were informed that treatment would start in 9 weeks, and were called half way through the study. The waitlisted subjects were encouraged to call if they were having problems, or their symptoms worsened (Foa, et al., 2005, p. 957).
Of the 96 subjects who completed the trial, 40 terminated at between eight to nine sessions, 27 because they met termination standards, and 13 who declined additional sessions (Foa, et al., 2005, p. 958). The standard for CBT–based interventions for PTSD is 8 sessions (p. 952). The remaining 56 terminated at between 10 and 12 sessions (p. 958). Although the study criteria was met, “33 participants received at least three sessions of therapy during follow–up”, with 14 from the prolonged exposure group and 19 from the prolonged exposure–cognitive restructuring group (p. 960).
The outcome of the study seems to indicate that prolonged exposure alone is more effective than the prolonged exposure–cognitive restructuring, but the disparity is only slightly statistically significant. A surprise effect was also uncovered during this study – that treatment from masters–level clinicians performed better than CBT experts (Foa, et al., 2005, p. 962). The authors note that when combining the two treatments, subjects may not be receiving the full dose of cognitive restructuring. However, the success in the prolonged exposure group underlies the need for CBT practitioners to reevaluate what they believe to be critical components of CBT, and focus on what intervention is best for the client. If a component is not needed or is less effective, it should be left out and simplified. The authors note that “combining separately efficacious psychological treatments does not yield better outcome[s]”, and that alternate models of treatment are still needed to increase treatment effect sizes (p. 963). The authors conclude with a call for increased training in the use of CBT in community mental health because of its effectiveness (p. 958).
The difference between short and long term effects is all in the recipe
A third article, by McDonagh, et al., titled Randomized Trial of Cognitive–Behavioral Therapy for Chronic Posttraumatic Stress Disorder in Adult Female Survivors of Childhood Sexual Abuse, focuses on women with a history of childhood sexual trauma. This is the first article examining this issue – PTSD secondary to female childhood sexual abuse, the authors note (McDonagh, et al., 2005, p. 515). Subjects were evaluated using the Clinician Administered PTSD Scale, and were selected on the basis of “sexual contact with anyone or more years older when the study participant was under the age of 16 years”, and subjects experiences “intrusive and avoidance symptoms […] clearly related to [childhood sexual abuse] (p. 516).
Subjects were divided into two treatment groups: CBT and present–centered therapy, and utilized a waitlist group as a control. Both treatment conditions were allotted 14 individual sessions, the first seven of which were two hours long, and the remaining sessions were one and a half hours long (McDonagh, et al., 2005, p. 518). All of the clinicians were female, and all had specialized training in female sexual assault. (p. 518).
The CBT treatment group utilized the treatment protocols designed by Foa et al., utilizing prolonged exposure (McDonagh, et al., 2005, p. 518). At the fourth session, cognitive structuring was employed and taught to the subjects, and psychoeducation was provided as well (p. 518). Adherence to these protocols by the subjects was 87.8% (p. 518).
The present–centered therapy group was “designed to describe an active therapeutic intervention that non–CBT clinicians might use in the treatment of PTSD–CSA” (McDonagh, et al., 2005, p. 518). The technique was specifically designed to omit what the authors believe to be the magical (“hypothesized active”) ingredients in CBT: “breathing retraining, [prolonged exposure], in vivo exposure, and cognitive restructuring” (p. 518). Adherence to these protocols by the subjects was 88.5% (p. 518).
The authors noted a very high dropout rate in this study, 23% total: 41% in the CBT group and 13% in the present–centered therapy group (McDonagh, et al., 2005, p. 519). Further, 27.6% of subjects in the CBT group no longer met criteria for PTSD post-treatment, and 31.8% of the present–centered therapy group no longer met criteria for PTSD post-treatment (p. 519). However, the effect of present–centered therapy diminished at the three and six month follow–up, with “CBT [being] associated with greater reductions in [Clinician Administered PTSD Scale] scores”, and indicated a “moderate to large effect size” (p. 519). The authors note that the high dropout rate is a serious limitation the generalizability of this study, and may have been a confounding variable (p. 522).
More than one type of crazy at a time
A fourth article by Mueser, et al., is titled A Randomized Controlled Trial of Cognitive–Behavioral Treatment for Posttraumatic Stress Disorder in Severe Mental Illness. This article discusses CBT treatment of PTSD when PTSD is comorbid with schizophrenia, schizoaffective disorder, major depressive disorder, bipolar disorder, borderline personality disorder, and substance use/abuse (Mueser, et al., 2008, p. 266). Subjects were divided into two treatment groups, CBT and TAU, or treatment as usual.
The CBT treatment group consisted of a structured program, and was conducted at the subjects’ local community health center (Mueser, et al., 2008, p. 265). Each subject was randomly tested for their compliance with treatment protocols, and the clinicians ensured that subjects completed at least six sessions. The authors wanted to ensure that at least three sessions were devoted to cognitive restructuring, which is the assumed active variable (p. 263).
The TAU group was given no active therapy, and they continued receiving the treatments they had been outside the study, as the community mental health facilities offered neither cognitive restructuring nor exposure therapy (Mueser, et al., 2008, p. 263).
This study seems to indicate a medium effect size for CBT compared to the TAU group, but the TAU group still showed improvement. The authors note that this may be a result of supportive counseling that clients receive at the community mental health centers (Mueser, et al., 2008, p. 266). Key differences between this and the Foa, et al., article, which this trial attempted to replicate, is the use of a comorbid with severe Axis I and Axis II population, and the use of community mental health facilities that provide long–term care to clients with severe comorbid disorders (p. 265).
Traffic collisions cause PTSD? I guess if you and your car are one with the mangling
The fifth article, this one by Taylor, et al., is called Posttraumatic Stress Disorder Arising After Road Collisions: Patterns of Response to Cognitive–Behavior Therapy. This article aims to analyze subjects’ responses both pretreatment and post-treatment following motor vehicle accidents. Subjects were assessed on the Clinician Administered PTSD Scale. This study utilized three treatment groups: individual CBT, group CBT, and a waiting list group (Taylor, et al., 2001, p. 544).
CBT groups were provided psychoeducation in the first session, cognitive restructuring in the next three sessions, applied relaxation in the fifth through twelfth sessions, imaginal exposure for the fifth through eighth sessions, and in–vivo exposure in the eighth though twelfth sessions (Taylor, et al., 2001, p. 544). Session twelve serves to review the gains made during the sessions, and discusses “methods for maintaining and extending treatment gains” (p. 544). Inter–rater reliability was maintained by comparing Clinician Administered PTSD Scale with assessments by doctoral–level psychologists, who rated fifteen recorded sessions. The concordance rate between the Clinician Administered PTSD Scale and the doctoral–level psychologist was 87% (p. 544).
This study contained 8 dropouts, compared to 50 who completed the study. Both groups had markedly high and statistically significant improvement in Clinician Administered PTSD Scale scores, with a reduction of four or more points compared to the pretreatment assessments. Further, the data indicates that scores continued to decrease at the three month follow–up (Taylor et al., 2001, p. 547).
The authors discuss limitations of this research, indicating that the data set precludes spontaneous remission of the symptoms of PTSD, but recognizes the effects of time in easing the symptoms of PTSD (Taylor et al., 2001, p. 550). The authors further acknowledge that comorbidity with pain and depression may confound the response rate to CBT. The authors address this by hypothesizing that more treatments, especially individual sessions, may be required to address the partial responding group (p. 550).
Freud explored differences between “’normal’ grieving and ‘pathological melancholia in response to loss’” (Krupnick, 2002, p. 919). In the article, Brief Psychodynamic Treatment of PTSD, Krupnick describes the most empirically validated psychodynamic treatment of PTSD, named “brief trauma–focused psychotherapy”, developed by Horowitz (p. 919). The model is a twelve session treatment that involves elements from
general psychodynamic psychotherapy, for example bringing conflicts into conscious awareness, helping individuals gain insight into their difficulties through scrutiny of the therapist–client relationship, with particular attention paid to ways in which clients may react to their therapists on the basis of unresolved feelings related to significant figures of the past, and analysis of how they keep threatening thoughts, feelings, and impulses from awareness. (p. 920)
The author states that psychodynamic research suggests that patients respond in “characteristic” ways to traumatic events, with shock and disbelief gradually turning into intrusive thoughts (p. 920). Individuals seeking psychotherapeutic help “typically do so because of intrusive thoughts and heightened arousal [… and] feel that their emotions are out of control and that they cannot cope” (p. 920). The authors state that the aim of the practitioner is to evaluate what stage the client may be stuck at, what barriers exist that prevents processing of the event, and to help the client process and integrate the trauma.
The authors note that the client must have a pre–trauma history of emotional stability, stating that “dynamically oriented PTSD treatment, although offering structure and support, also can be painful and intense” (Krupnick, 2002, p. 921). The clinician needs to be able to “inquire about the nature and circumstances of the traumatic event, including the events leading up to the trauma and the circumstances and reactions following it, [as well as] a comprehensive development history” (p. 921). The author notes that brief dynamic therapy advocates, such as Milan, advocated this intervention only if the client has a history of at least one meaningful relationship (p. 921). In the assessment, the clinician is to note the “psychological strengths and weaknesses” of the client, which helps to target the best way to help the client cope with the trauma (p. 921).
Krupnick conceptualizes the brief psychodynamic therapy as existing with three phases, each with discrete tasks and goals: initial, middle, and final. The client and clinician should be able to arrive at a mutual focus, but clinician needs to use the trauma to organize and focus the treatment (Krupnick, 2002, p. 922). The initial phase consists of four sessions, emphasizing to the client that the session is a safe–space, reinforcing that the clinician is solution–oriented, and allowing the client to tell their story (p. 922). Rather than having the client use thought suppression to control their intrusive thoughts, the author suggests that clinicians encourage the repeated narrative. This serves to reinforce to the client that the clinician is able to tolerate the story and emotions, allowing the clinician to foster a sense of safety (p. 922). Pursuant to fostering feelings of safety, the clinician will tell their client that the therapy will go at the client’s pace and depth, again, reinforcing the feeling of safety and tranquility in the session (p. 923). Unconditional acceptance of the client, not judgment, the author states, will motivate the client to set aside the fear in sharing their experience (p. 923).
Krupnick describes the middle phase of treatment, as “the ‘working–through' phase of therapy” (Krupnick, 2002, p. 923). This phase centers on the beliefs and attitude which “have made the particular trauma so difficult to integrate” (p. 923). Clients are advised to be aware of their expectations for themselves, and to examine the effects of trauma on their relationships, and how their reaction to the trauma may be influenced by their past experiences (p. 923). In other words, it’s the right reaction to the wrong situation. A client who learns this lesson, the author states, is provided with a sense of “control and coherence to individuals who felt only chaos and disorganization” (p. 924).
The final sessions should focus on termination, Krupnick states. Krupnick seems to indicate that the feelings of loss can be transferred from the trauma to the session, and that the anticipated feelings of loss can allow the client to plan and mourn the loss of the sessions (Krupnick, 2002, p. 924). Clinicians are supposed to direct this time of loss, allowing the client to resolve their feelings of loss, which vicariously resolves their outstanding feelings of grief from the trauma (p. 924).
Another example of psychodynamic treatment of PTSD comes from Schottenbauer, Glass, Arnkoff, and Gray. Their article, Contributions of Psychodynamic Approaches to Treatment of PTSD and Trauma: A Review of the Empirical Treatment and Psychopathology Literature, analyzes the psychodynamic approach to PTSD. The authors assert that while CBT may have high rates for treatment efficacy, it is plagued by high drop–out rates, and thus cannot provide a true solution. The authors assert that psychodynamic treatment can be used as a model for alternative treatment (p. 13). The authors cite a randomized controlled study whose results seem to indicate that brief psychodynamic therapy has a similar treatment effect as CBT, but benefits from the psychodynamic models’ ability to address “various personality traits” (p. 14).
The authors make note of the two related, but different concepts of trauma: simple and complex (Schottenbauer, et al, 2008, p. 14). Simple PTSD consists of a single “discrete traumatic event”, while complex trauma consists of “prolonged, repeated trauma” (p. 14). As a reaction to the large non–response rate, as well as the high dropout rate, of CBT–approaches to PTSD, several short term psychodynamic therapies have been developed for PTSD (p. 15). The authors cite an article by Krupnick, where 87.5% reported a good outcome, compared to 12.5% who reported a fair outcome (p. 16). It should be noted, however, that this is an uncontrolled trial had a sample size of just eight, less than half of those who were offered treatment (p. 16). The article states that 18 were offered treatment, 11% declined, and 13% dropped out (p. 16). The authors also state that this intervention is only applicable to those who were relatively well functioning before the trauma (p. 16).
Complex PTSD, the authors note, has a high co–morbidity with Axis I and Axis II disorders (Schottenbauer, et al, 2008, p. 17). The authors believe that psychodynamic treatments are superior to CBT interventions for complex trauma because of the psychodynamic model is grounded on the person in relation to their relationships and the world around them, compared to CBT’s specific symptom–targeted approach (pp. 16–7).
Interpersonal problems are areas that psychodynamic therapy excels at, the authors note (Schottenbauer, et al, 2008, p. 17). The authors state that patients expressing PTSD symptoms tend to have more interpersonal problems, like divorce and disturbance in relationships, and that the rates are twice as high as the normal population (p. 17). The authors note that when patients have stable social relationships, their ability to cope with the stress of trauma increases (p. 17). When personality disorders are mixed in, the patient can experience even further complications when their social network is taxed and face social rejection (p. 17). This social rejection can increase the likelihood of severe disturbances from PTSD symptoms (p. 17).
Since “psychodynamic psychotherapy, unlike CBT […], focuses on the interpersonal relationships of the client” it is uniquely qualified to address the issues that arise from discrepancies related to psychosocial disturbances caused by PTSD symptoms (Schottenbauer, et al, 2008, p. 18). The authors note that through the relationship with the clinician, clients can reflect on normal and healthy interpersonal relationships (p. 19). Based on the Krupnick article, the authors determined that the interpersonal skills gained from psychodynamic therapy can induce PTSD symptoms into remission (p. 19).
Another psychodynamic approach is called Control Mastery Theory (CMT), which derives from “psychodynamic ego therapy” (Schottenbauer, et al, 2008, p. 19). This intervention focuses on “facilitating patients’ inherent unconscious capacity to solve problems” (p. 19). This approach centers on the client identifying evidence that both supports and refutes their beliefs, both positive and negative (p. 19). Clinicians are to remain neutral, but provide the client with tests of logic that elicit the supporting evidence provided by the client (p. 19). This allows the client to be aware of the effects of their behavior on their interpersonal relationships, allowing the client to identify maladaptive behaviors and make changes to their interpersonal behavior (p. 19). The authors also discuss the implications of attachment theory, and that the clinician must also take into account the style of attachment to increase the chances of successful treatment (p. 20).
The authors discuss developmental issues, and that these issues can impair a mature coping style in response to trauma (p. 20). The authors note that significant trauma before the age of 18 is highly correlated to greater impairment as an adult (p. 20). The reverse is also true, that a higher level of functioning serves as a protective factor in coping with trauma (p. 20).
Another issue that psychodynamic models are excellent at catering to is the presence of comorbid Axis I or Axis II disorders, which can be as high as 99% (p. 25). The authors note that PTSD is often comorbid with “major depressive disorder, substance use or dependence, panic disorder, agoraphobia, obsessive–compulsive disorder, generalized anxiety disorder, social phobia, specific phobias, and bipolar disorder” (Schottenbauer, et al, 2008, p. 25). Further, child abuse has been found to be related to “borderline, self–defeating, narcissistic, histrionic, sadistic, and schitzotypal traits” (p. 25). Whether the personality disorder preceded the trauma, or is a response to childhood trauma, psychoanalysis is well suited for dealing with both the trauma and the personality disturbances (p. 26). Research suggests that CBT has reduced treatment outcomes with comorbid Axis I or Axis II disorders, but psychodynamic interventions do work for these patients (p. 26).
So which one is it? Cake, or the eating of it?
Should clinicians practice cognitive–behavioral therapy or psychodynamic therapy? It seems like an easy question, considering the glut of research focused on CBT, but the answer is both, and neither. The question of which approach to use sets up a false dichotomy, as both have tremendous strengths – as well as weaknesses.
CBT is a highly empirically validated treatment; it is efficient, rapidly deployable, and symptom focused. A key goal of CBT is making sure that the clinician is interchangeable, a sure way to ensure that it is the treatment that is effective – and not the clinician. A very important aspect of CBT is its response to new research in neuropsychology. CBT realigns with factual findings in neuropsychology, assuring that CBT stays current as a model of treatment.
A client comes into the office complaining of intrusive thoughts, and we explore those thoughts in a critical light. “I’m a failure because I can’t deal with these feelings on my own” gets a response of, “No, you are not a failure. You have kids, a family that loves you, and the feelings you have are perfectly normal considering what has happened to you.” CBT is focused on the triad of thoughts (cognitions)–emotions–behavior, and CBT practitioners try to influence the thoughts – which is really all that can be dealt with in a 90–minute session.
CBT, however, has its limitations. It provides poor support for comorbidity with other Axis I and Axis II disorders. A review of the literature reveals few studies that even attempt to address a population that includes Axis I disorders, save for a few anxiety disorders. CBT is also lacking true standardization, leaving bits and pieces up to individual clinicians. This is undesirable because the term CBT may, in fact, refer to different mixes and ratios of ingredients. And this is seen in the published literature, to say nothing of practitioners who do not publish.
The psychodynamic approach is the treatment that won’t die. It has existed since the 1800’s, and has long since evolved from Freudian introspection to a client–centered model that is highly capable of incorporating comorbidity. A key strength of psychodynamic therapy is the close, intimate relationship between client and clinician. Engendering an environment of trust, and a nonjudgmental attitude, seems to be a key factor in the reduction of symptoms. Clients for whom CBT may not work, or clients who need closeness to be able to address their issues, will certainly find psychodynamic approaches a meaningful, worthwhile, and effective intervention. Further, psychodynamic models better allow the client to perceive relationships around them; psychodynamic models as the primary focus is not only symptom reduction, but normal functioning in all domains of life – including the psychosocial.
Despite the favorable press, psychodynamic models do have limitations. First, psychodynamic models tend to lack generalizability. This is because of the intense focus on the client–clinician relationship; these strong, intimate relationships are not easily forged, especially in patients who exhibit states of constant fear. Second, psychodynamic models really rely on the ability of the clinician. Because the relationship between client and clinician is so crucial, the psychodynamic clinician cannot allow the client to feel unsafe – even for an instant. Clients who come for psychological intervention may be expecting social rejection, and feelings of safety are crucial in successful psychodynamic approaches. Third, psychodynamic research tends to be limited to case studies, which inherently lack generalizability. Lastly, psychodynamic models work – only not necessarily why the clinicians or clients believe they do. Psychodynamic theories need to incorporate the realities of neuropsychology and cognitive neuroscience into their underlying model, as they are more reflective of the biological reality. The software cannot solely be explained in a vacuum, as the software necessarily interacts with the hardware it is running on.
So what is the right question?
Asking which treatment model is the best is the wrong question. The correct questions are: what parts of each treatment form the best and most effective treatments, and when should each treatment model be used. By asking these different questions, we get out of the false dichotomy, and we are left with a better understanding of treatment options that can help clients.
The Magic of CBT Isn’t Just Pixie Dust
CBT is particularly noted for its rapid deployability and large effect size. It is good for using when simple trauma presents, not comingled by the presence of active Axis I or Axis II disorders, which complicate treatment. Complex trauma can be approached with CBT, however, the effect size may be limited.
Parts of CBT that seem to involve a reduction in symptoms are: breathing retraining, prolonged exposure, in vivo exposure, and cognitive restructuring, as elucidated by the article by Foa, McDonagh, and others.
Breathing retraining is a controversial component of CBT. Studies have failed to show any long term treatment effects between breathing retraining and CBT without breathing retraining. However, the short–term effect of providing relaxation to the client probably maintains its status in most CBT models.
Prolonged exposure is a component of trauma–related CBT, in which clients are made to close their eyes and recount the story of their trauma as many times as would fill the session. In the Foa article, tape recordings were made for the subjects, and subjects were encouraged to listen to the tape daily (Foa, 2005, p. 957). The underlying mechanism of prolonged exposure seems to be narrative rehearsal, a low–level method for remembering. The hope is that since narrative rehearsal is low–level, it may precede the level of the trauma memory, aiding in the processing and integration. Further, clients may be able overcome dissociation because the memory is integrated into the explicit memory–and a clear ownership of the memory is established.
In-vivo exposure is the actual real-world use of the learned skills, bringing the skills from the clinician’s office to actual daily situations. A key measure of the effects of PTSD symptoms is the ability to integrate back into normal society, free from avoidance of situations which may have previously been too stressful or painful to be near. All of the different checklists, inventories, and measures cannot compare with an objective fact: the client can again do things they previously could not because of symptoms.
Cognitive restructuring sounds ominous, but is really simple. Cognitive restructuring focuses on the thoughts of the client and presents either confirmatory or refutatory evidence that corresponds to the truth value of the thought. “My life is ruined,” is an example thought. The clinician would have the client come up with evidence that the statement, now known as a “thought distortion”, is false, as well as having the client change the statement to a more positive, self-referential statement, such as “I have the ability to change my life, starting now, and it can only improve”. When thought distortions pop-into the client’s head, clients are encouraged to change their inner monologue to reflect positive assertions.
As a note, the Foa article specifically mentioned that adding more CBT techniques does not a better treatment make. Short-term effects of a control group, lacking any of the above key components of CBT, showed a larger effect size than a combination of cognitive restructuring and prolonged exposure, yet smaller than prolonged exposure alone.
Psychodynamic models required components
Psychodynamic treatments are noted for their ability to reintegrate the client back into normal functioning. The models work because they simultaneously target the primary, secondary, and tertiary facets of the trauma. Primary emotions are those of panic, stress, and loss, while secondary emotions are the beliefs that maintain these feelings. The tertiary facet is the relation of the client to the world around him. Psychodynamic models are very adept at addressing issues that CBT either misses, or is not designed to handle.
Both articles indicate that any successful psychodynamic treatment requires one major factor: an adept clinician. A good, well–educated clinician will be able to assess the client, determine their level of capacity, and tailor the intervention to the client’s needs.
Another key factor is the clinician’s ability to conceptualize the client’s relationships. Because the main treatment goal is it to integrate the client back into their normal social functioning, clinicians need to be able to offer advice, analysis, and reassurance to the client over the range of their interpersonal relationships.
Clinicians also need to be able to move the client toward termination in a tactful and emotionally meaningful way. Termination is the end goal of psychodynamic treatment, second only to the relief of symptoms. The transition out of treatment is intended to serve as a surrogate for the trauma and feelings of loss. The transference of feelings from the trauma to the session enables the client to deal with the feelings on a concrete basis, rather than the vague feelings of trauma.
Mutual symbiosis, or why we should all just get along
Each approach has many things to learn from the other. CBT should be considered the first-line therapy for PTSD, as it may be able to quickly and simply reduce the effect of PTSD on global functioning across domains. As a second line, the much more expensive and involved psychodynamic therapies should be utilized for clients that continue to face disturbances. Why should psychodynamic therapies be utilized second? The simple answer is because it requires more expertise, time, work, and money to be effective. Much education is required to properly administer psychodynamic therapy, and CBT’s effects can significantly lighten the caseload on psychodynamic clinicians. CBT has been empirically validated in both individual and group sessions, and the high effect sizes of CBT treatments means that psychodynamic clinicians can spend adequate time with the clients who need their expertise the most.
Foa, E., Hembree, E., Cahill, S., Rauch, S., Riggs, D., Feeny, N., et al. (2005). Randomized trial of prolonged exposure for posttraumatic stress disorder with and without cognitive restructuring: Outcome at academic and community clinics. Journal of Consulting and Clinical Psychology, 73(5), 953–964.
Garfinkel, S., & Liberzon, I. (2009). Neurobiology of PTSD: A review of neuroimaging findings. Psychiatric Annals, 39(6), 370.
Krupnick, J. (2002). Brief psychodynamic treatment of PTSD. Journal of Clinical Psychology, 58(8), 919–932.
McDonagh, A., Friedman, M., McHugo, G., Ford, J., Sengupta, A., Mueser, K., et al. (2005). Randomized Trial of Cognitive–Behavioral Therapy for Chronic Posttraumatic Stress Disorder in Adult Female Survivors of Childhood Sexual Abuse. Journal of Consulting and Clinical Psychology, 73(3), 515–524.
Meiser–Stedman, R. (2002). Towards a Cognitive–Behavioral Model of PTSD in Children and Adolescents. Clinical Child and Family Psychology Review, 5(4), 217–232.
Mueser, K., Rosenberg, S., Xie, H., Jankowski, M., Bolton, E., Lu, W., et al. (2008). A randomized controlled trial of cognitive–behavioral treatment for posttraumatic stress disorder in severe mental illness. Journal of Consulting and Clinical Psychology, 76(2), 259–271.
Perry, B. D., Pollard R. A., Blakley T. L., Baker, W. L., and Vigilante, D. (1995). Childhood trauma, the neurobiology of adaptation, and "use–dependent" development of the brain: how "states" become "traits". Infant Mental Health Journal, 16(4), 271–291.
Schottenbauer, M., Glass, C., Arnkoff, D., & Gray, S. (2008). Contributions of psychodynamic approaches to treatment of PTSD and trauma: A review of the empirical treatment and psychopathology literature. Psychiatry: Interpersonal and Biological Processes, 71(1), 13–34.
Taylor, S., Fedoroff, I., Koch, W., Thordarson, D., Fecteau, G., & Nicki, R. (2001). Posttraumatic stress disorder arising after road traffic collisions: Patterns of response to cognitive–behavior therapy. Journal of Consulting and Clinical Psychology, 69(3), 541–551.