A critical review of mechanisms of adaptation to trauma: Implications for early interventions for posttraumatic stress disorder

A critical review of mechanisms of adaptation to trauma: Implications for early interventions for posttraumatic stress disorder. Richard A. Bryant. Clinical Psychology Review, January 24 2021, 101981. https://doi.org/10.1016/j.cpr.2021.101981

Highlights

• Current early intervention strategies to limit PTSD after trauma exposure have achieved limited success.

• Longitudinal studies indicate that trauma survivors follow distinct trajectories of stress response.

• There are diverse acute predictors of PTSD, encompassing a range of biological and cognitive factors.

• The different posttrauma trajectories indicates that early intervention strategies need a more flexible approach.

• The diversity of acute risk factors for PTSD indicates that early intervention strategies need a more flexible approach that accommodate the different mechanisms.

Abstract: Although many attempts have been made to limit development of posttraumatic stress disorder (PTSD) by early intervention after trauma exposure, these attempts have achieved only modest success. This review critiques the biological and cognitive strategies used for early intervention and outlines the extent to which they have prevented PTSD. The major predictors of PTSD are reviewed, with an emphasis on potential mechanisms that may underpin the transition from acute stress reaction to development of PTSD. This review highlights that there is a wide range of biological and cognitive factors that have been shown to predict PTSD. Despite this, the major attempts at early intervention have focused on strategies that attempt to augment extinction processes or alter appraisals in the acute period. The documented predictors of PTSD indicate that a broader range of potential strategies could be explored to limit PTSD. The evidence that people follow different trajectories of stress response following trauma and there is a wide array of acute predictors of PTSD indicates that a flexible and tailored approach needs to be investigated to evaluate more effective early intervention strategies.

Keywords: Posttraumatic stress disorderEarly interventionAcute stressTreatment

13. Implications for early intervention

Although acute symptoms are statistically related to subsequent PTSD, the relationships are modest at best. Sufficient prediction requires balancing

sensitivity (the probability that someone who eventually develops PTSD satisfied the acute predictor), specificity (the probability that someone who does not develop PTSD did not satisfy the acute predictor), as well as positive and negative predictive power (the probability that a person who does/does not satisfies the acute predictor subsequently does/does not develop PTSD). Our modest ability to predict subsequent PTSD points to several important implications in early intervention for PTSD. First, the modest predictive power of any acute markers suggests that identifying people who will require early intervention has a long way to go before we can confidently rely on predictive tools to identify those at risk of PTSD. Second, the conclusion that not all people follow the same trajectory of posttraumatic stress indicates that no single early intervention strategy may be efficacious for all people. Third, the array of acute markers that have been identified as being predictive of later PTSD points to the need for a broad approach to encompass potential intervention targets that may address these different mechanisms. On the basis of these conclusions, and the review conducted above, the following considerations are offered regarding future directions for developing a better science and practice for early intervention to limit PTSD.

13.1. Timing of early intervention

Although some models of fear learning and extinction may suggest that intervening very early in the period after trauma exposure may reduce consolidation of trauma memories (Norrholm et al., 2008; Rothbaum et al., 2012), there is also considerable evidence that the later one attempts to classify people who are at risk of subsequent PTSD the more accurate the prediction may be. For example, acute symptoms that persist in the weeks after trauma are more predictive than those assessed in the very initial period (Briere, Scott, & Weathers, 2005; Panasetis & Bryant, 2003; Solomon et al., 1988). This suggests that attempts to provide early intervention to those identified as being at risk of later PTSD are more likely to be unnecessary because the early identification of risk will yield more false positive identifications. Perusal of trajectory studies of PTSD point to the same conclusion because they suggest that a proportion of people will develop PTSD over time as their condition worsens, and these people may not be identified if the initial assessment of risk is conducted too early (Galatzer-Levy et al., 2018). It appears that there is a balance between offering intervention early because some evidence suggests this may yield a more potent effect (Carpenter et al., 2018) and the risk of providing the intervention to people who may naturally remit. This issue is underscored by evidence from one trial that found that early intervention with TF-CBT facilitated recovery but it did not necessarily lead to greater longer-term outcomes than people's natural recovery trajectories (Shalev et al., 2016). At a practical level, many early intervention studies have attempted to identify risk very soon after trauma exposure, often within the first week, whereas delaying this identification process several weeks after trauma exposure may yield more accurate classification of people at risk of later PTSD and at the same time not comprise the benefits of treatment.

13.2. The benefit of identifying resilient individuals

Trajectory studies of the course of posttraumatic stress have robustly indicated that approximately three-quarters of trauma survivors do not experience PTSD over repeated assessments (Galatzer-Levy et al., 2018). This strongly suggests that these individuals are not in need of mental health assistance for PTSD. Traditionally attempts to focus on trauma survivors who may benefit from early intervention have focused on those who are deemed at risk of PTSD development. Another approach would be to identify the resilient people and focus limited mental health resources on those trauma survivors who are not resilient. This strategy would reduce the problem of identifying people in the acute phase who may not be very distressed but who may worsen over time. Most trajectory studies indicate that approximately 75% of trauma survivors can be classified as resilient, and so excluding these people from consideration would allow mental health resources to be allocated to those in need. Such an approach could involve early intervention and/or ongoing monitoring of the person's condition and intervention could be offered when the condition deteriorates. This strategy would be consistent with the evidence that PTSD is a fluctuating condition that often surfaces following recent stressors (Bonanno, Kennedy, et al., 2012; Bryant et al., 2017; Pietrzak, Van Ness, et al., 2013), which could be identified with regular monitoring of this group of trauma survivors.

13.3. Focus on novel mechanisms related to predictors

This review has highlighted that the early predictors of PTSD encompass a range of factors that are present in the acute period after trauma that appear to represent mechanisms that contribute to PTSD development. Despite the breadth of predictive factors, most early interventions have focused on either extinction-based mechanisms or processes involving adjustment of maladaptive appraisals. These mechanisms form the basis of most TF-CBT programs, as well as novel pharmacological and direct stimulation techniques. However, there are other options that could be used that would address some of the factors identified as early predictors of PTSD. For example, we have noted that overgeneral autobiographical memory retrieval in the acute phase is predictive of later PTSD (Harvey et al., 1998)(Kleim & Ehlers, 2008). In recent years this deficit has been the focus of memory specificity training, which coaches people on retrieving specific memories and has been shown to have a small to moderate effect in reducing depression (Hitchcock, Werner-Seidler, Blackwell, & Dalgleish, 2017). One pilot study has found that this training can also reduce PTSD symptoms (Moradi et al., 2014). This approach could be trialled in the acute posttrauma period to determine its efficacy in limiting PTSD.

We also noted that rumination is a strong predictor of later PTSD. Several treatment options have promise for managing rumination, including metacognitive therapy (Wells & King, 2006), rumination-focused cognitive behaviour therapy (Watkins et al., 2011), and mindfulness-based therapy (Kuyken et al., 2008). There is initial evidence that these interventions can be applied to PTSD (Polusny et al., 2015; Wells & Colbear, 2012). Considering the strong predictive role of rumination in PTSD development, potentially adapting metacognitive and mindfulness-based approaches in the early period may be adaptive for recently traumatized people who present with marked rumination.

Elevated distress and deficient emotion regulation strategies in the acute period have been linked to later PTSD (Bardeen et al., 2013). There are numerous strategies available to improve emotion regulation that have proven capacity in managing PTSD symptoms. Programs that teach skills in identification of emotional states, strategies to reduce distress, addressing emotional avoidance, and tolerating bodily sensations of emotions can effectively limit PTSD or associated anxiety symptoms (Barlow et al., 2017; Bryant et al., 2013; Bryant, McGrath, & Felmingham, 2013; Cloitre et al., 2010). Teaching acutely traumatized people who present with emotion dysregulation skills to identify and manage emotional reactions may also be useful in limiting subsequent PTSD.

The predictive role of depression and negative affect in the acute phase points to the possibility of targeting this initial symptom as a potential early intervention candidate. There has been renewed interest in treatments of anhedonia and negative affect by training people to develop greater awareness of and exposure to positive experiences (Craske, Meuret, Ritz, Treanor, & Dour, 2016). This approach emphasizes enhancement of reward processes, including broadening exposure to positive stimuli, rehearsing appreciation of the pleasurable experiences associated with positive events, and deep learning techniques to enhance the sense of pleasure. This strategy is indicated because of evidence that positive affect can impede fear reacquisition after extinction (Zbozinek & Craske, 2017) and that positive affect can serve as a buffer against the adverse mental health effects of chronic stress (Sewart et al., 2019). Importantly, one controlled trial found that training patients with severe depression or anxiety reported greater reductions in negative affect and improvements in positive affect than those who treatment targeted negative affect (Craske et al., 2019). Considering the role of negative affect in the acute period, this intervention offers an opportunity target this risk factor in the acute posttrauma period.

There is also potential is adapting some of the agents used to augment exposure therapy for chronic PTSD to determine if this can facilitate early interventions. Most attempts at augmenting exposure therapy for PTSD have relied on targeting extinction mechanisms by modulating neural processes implicated in the associative learning process, including pharmaocological agents such as d-cycloserine (de Kleine, Hendriks, Kusters, Broekman, & van Minnen, 2012; Difede et al., 2014; Rothbaum et al., 2014), yohimbine (Tuerk et al., 2018), and MDMA (3,4-methylenedioxymethamphetamine (Barone, Beck, Mitsunaga-Whitten, & Perl, 2019; Mithoefer et al., 2018; Ot'alora et al., 2018). Although these attempts have only been marginally beneficial and are yet to be definitively shown to enhance exposure therapy (Lebois, Seligowski, Wolff, Hill, & Ressler, 2019; Weisman & Rodebaugh, 2018), they are yet to be properly tested in early intervention frameworks. There is also potential in extending extinction procedures beyond those traditionally studied. For example, there is evidence that PTSD is characterized by greater generalization of fear, reflected in both the common symptoms of fear of stimuli that are reminiscent of the trauma and hypervigilance to a range of potential threats and also experimental evidence that people with PTSD demonstrate greater fear overgeneralized fear conditioning (Kaczkurkin et al., 2017). This raises the possibility that exposure therapy in the acute phase may be augmented by targeting exposure to situations or stimuli that approximate the traumatic event as well as trauma-specific stimuli. Based on studies that the cholinergic system appears to contribute to poor discrimination between danger and related stimuli (Thiel, Bentley, & Dolan, 2002; Weinberger, 2007), anticholinergic agencies given at the time of exposure therapy may augment early intervention treatment with those displaying early symptoms of acute posttraumatic stress.

In noting these potential mechanisms, however, there is a need to be cautious about the strength of evidence these factors are significant in fact the mechanisms of change underpinning successful adaptation to trauma. Kazdin (2007) outlined several key criteria for how mechanisms of change should be defined; these include (a) the strength of the associations between an intervention, the purported mediator, and the outcome; (b) specificity of the change mechanism such that the change is not occurring as a result of many other, possibly related, constructs; (c) reliability of the change mechanism across samples and studies; (d) causal proof as shown by experimental manipulation of the proposed mechanism; (e) proof of the temporal sequence of the causes and mediators on the target outcome; (f) a dosage effect such that there is an association between the extent to which the mechanism has been activated and the strength of the outcome; and (g) theoretical plausibility given broader body of knowledge. It should be recognized that our current evidence base for the potential mechanisms that influence successful adaptation to trauma does not currently meet most of the criteria, and so we should be very tentative about claims of how these mechanisms can be utilised in the clinical context until further evidence is obtained.

13.4. Developing more sophisticated prediction models

The capacity for more precise prediction of who will develop PTSD may open new opportunities for targeted early intervention strategies. Although many studies have identified a range of acute symptom, cognitive, and biological factors that can predict later PTSD, this evidence base is traditionally limited by understanding the individual predictive role of these factors rather than considering them in association with each other. There are two advantages in understanding how predictive factors are associated with each other. First, by considering the relationships between predictive factors there is the potential to identify more specifically the mechanisms that should be targeted in early intervention because we can isolate factors that are more causally related to later PTSD by determining their influence over and above related factors. An example of this approach is a recent study that has focused on the role of cognitive factors that we have reviewed above have been shown many times to predict later PTSD. In one multifactorial analysis, this study found that as hypothesized by prevailing cognitive models (Ehlers & Clark, 2000), negative appraisals of the trauma and its sequelae predicted later PTSD both directly and also via maladaptive behavioural and cognitive coping strategies, and that processing of the trauma at the time of the event impacts later PTSD via its effects on later appraisals about oneself, one's environment, and the fragmentation of one's trauma memories; interestingly, including acute symptoms in this model did not improve prediction of later PTSD (Beierl, Bollinghaus, Clark, Glucksman, & Ehlers, 2020). This finding does suggest that early intervention strategies that address negative appraisals, cohesion of the trauma memory, improving coping strategies may be important for facilitating better posttraumatic adjustment.

The second benefit of considering the interplay between acute posttrauma factors that can predict later PTSD is the capacity for more sophisticated statistical modelling to use all available information to derive predictive algorithms that can be used in common acute trauma settings. One of the most commonly studied contexts for acute traumatic stress is emergency rooms, which also collect much routine data on biological indicators that can also be used as markers of acute stress. One recent study used a machine learning approach and tested a predictive model encompassing a range of psychological and biological measures collected in the emergency room, then validated this in a separate large sample, and found that the resulting algorithm achieved 87% accuracy in identifying patients who developed PTSD 12 months after presentation to the emergency room (Schultebraucks et al., 2020). This study found that prediction was best found with a combination of 20 variables that included acute psychological stress reports, combined with biomarkers such as white blood cell count and lymphocytes. Although this approach requires further validation in different settings, it holds the promise of being able to identify people who may benefit from early intervention or at least from ongoing monitoring to determine the trajectory of subsequent psychological health.

13.5. The need for a precision medicine approach

The evidence that there are multiple pathways to developing PTSD highlights that not one form of early intervention will be optimal for all trauma survivors. To date all attempts to limit PTSD by early intervention have focused on a uniform protocol administered to all participants identified as being at risk of PTSD development. There has been considerable attention in recent years on precision psychiatry as a more sophisticated approach to targeting treatments to patients with specific clinical presentations or profiles. Precision medicine is defined as “an emerging approach for treatment and prevention that takes into account each person's variability in genes, environment, and lifestyle” (National Research Council Committee on a Framework for Developing a New Taxonomy of Disease, 2011). Although not a new notion of understanding diseases in terms of the variability between individuals, it has only received focused attention in more recent times. The speed at which precision approaches have been adopted has been much faster in other areas of medicine, such as oncology, and psychiatry is only now embracing a precision approach (Fernandes et al., 2017). In this regard, most work has been done in psychosis and mood disorders (Salazar de Pablo et al., 2020), with most attention focusing on genetic, molecular, and neuroimaging techniques to aid diagnosis and treatment prediction (Williams, 2016).

The precision psychiatry approach can be equally applied to PTSD. This work is in its infancy, with the initial work focusing on genetic (gene PTSD) and neuroimaging (me) approaches to identify people with stress conditions and subtypes of PTSD. The evidence reviewed above indicates that people follow distinct trajectories, and this suggests that a precision psychiatry approach is appropriate for more nuanced screening and early intervention strategies for people shortly after trauma exposure. The study of early intervention for posttraumatic stress has not adequately addressed precision psychiatry approaches and continues to focus on one-size-fits-all programs for screening and interventions. This approach is contrary to the evidence. If early interventions after trauma exposure are to be advanced in a significant way, there is a need for research paradigms to understand the different phenotypes that are present in the acute and chronic phases (e.g. dissocative, hyperaroused, dysphoric), as well as the genetic, neurobiological, cognitive, behavioural, social, and emotional mechanisms that underpin development of the genesis of PTSD following the acute phase. This approach represents an ambitious research agenda but other domains of medicine and psychiatry have shown that it can yield very promising results.

13.6. The role of process therapy

Another approach is to adopt a process therapy approach which does not apply a uniform protocol to all patients but rather tailors the intervention to specific clinical needs (Hofmann & Hayes, 2019). This strategy would not identify trauma survivors simply in terms of the ASD or early PTSD symptoms categorization but would rather identify key acute symptoms or early markers of risk and target them with evidence-based strategies. For example, trauma survivors with re-experiencing symptoms may be provided with emotional processing strategies (e.g. imaginal exposure), avoidance with in vivo exposure, depressive symptoms with behavioural activation, rumination with mindfulness, overgeneral memory with memory specificity training, and anhedonia with positive affect training. The development of precision treatments in mental health have not kept pace with many areas of medicine, and there is much we have to learn about how to match specific treatments with distinct clinical presentations. Nonetheless, matching trauma survivors with primary presenting difficulties with the strategies that optimally meet their needs may provide a new opportunity for greater success in early intervention.