In 2014, 230 million children lived in countries and areas affected by armed conflicts [
1]. With increasing emigration and refugeeism, millions of children who have experienced war now also inhabit countries with no active conflicts. Research confirms high rates of mental health problems among war-affected children [
2], even higher than those of similarly affected adults [
3]. Studies demonstrate both general dose-effect relationships between exposure to war and trauma-related stress symptoms and specific types of war experiences that are particularly traumatizing [
2]. As many as 47 % of children exposed to war may suffer from posttraumatic stress disorder (PTSD) and 43 % from depression, while reviews and surveys suggest a prevalence of 10–30 % for PTSD among all refugee and asylum seeking children resettled into high-income countries [
4,
5].
Though most trauma survivors recover with time, for some, posttraumatic stress symptoms due to war experiences in childhood show persistence for years and even decades [
6,
7], and have been shown to be related to a number of psychological and physical health issues even in older age [
8]. PTSD in children also appears to be connected to lower verbal memory function and overall cognitive performance [
9,
10] and is linked to impairment in academic performance [
11,
12]. In addition to the decreased quality of life and increased suffering of trauma survivors, PTSD carries enormous economic costs associated with loss of personal income, inability to work, as well as increased utilization of treatment and support services. For example, in 2004, the social and welfare costs of claims for incapacitation and severe disablement in the UK from severe stress reactions and PTSD amounted to £104 million per annum [
13]. Posttraumatic stress symptoms in children due to organized violence are thus a pressing, global problem. Developing effective, evidence-based interventions to address this problem should be a public health priority globally.
For traumatized children in general, cognitive-behavioral therapy (CBT) has been repeatedly found to be effective in reducing PTSD and other symptoms [
16‐
18], and trauma-focused cognitive-behavioral therapy (TF-CBT) is recommended as the primary treatment for PTSD for both adults and children in several countries [
19,
20]. For children affected by war specifically, a variety of group and individual interventions have been implemented and studied all over the world. Reviews have reported evidence on the effectiveness of such psychosocial interventions in alleviating PTSD, depression and anxiety symptoms in children affected by armed conflict [
21‐
23]. However, reductions in symptoms have often been modest, and study designs have lacked rigor, with only a small minority representing RCTs.
The majority of intervention techniques among children traumatized by war are also based on CBT and its derivatives. CBT-based interventions for traumatic stress share similar creative, narrative, and cognitive elements, such as creative-expressive exercises (dream work and fantasy), cognitive restructuring, attention control, body-oriented methods, building a sense of safety, and providing psychoeducation. However, it is still largely unclear which specific treatment elements in these interventions might be most significant for recovery. Furthermore, there is a general lack of rationales based on clear theoretical frameworks as to which particular CBT tools are chosen for use in treatment [
22]. Therefore, there have been calls for research on the underlying mechanisms of change that contribute to the success or otherwise of interventions among war-affected children, i.e., particular processes mediating their effectiveness [
21].
NET as a CBT-based intervention
Narrative Exposure Therapy is a manualized, individual, short-term intervention program for the treatment of PTSD resulting from exposure to organized violence or other repeated traumatic events. NET is based on CBT principles, with its development influenced by exposure-based and testimonial therapies [
14]. The specific focus in NET is on habituation to and contextual anchoring of traumatic memories. This focus stems from the clinical model of repeated traumatization underlying NET, drawing on 1) dual representation theories of PTSD [
24,
25], see also [
26], and 2) Emotional Processing Theory and the idea of fear networks [
27].
According to dual representation theory, during a highly emotional event, two differing types of parallel memory representations are encoded. The sensory, cognitive-emotional and physiological features of the event are stored in long-term perceptual memory [
26], and have been called hot memories [
14,
24], Situationally Accessible Memories [
25] or sensation-near representations [
28], while the contextual, verbalizable elements of the situation are encoded into episodic memory [
26], and have been called cold memories [
14,
24], Verbally Accessible Memories [
25] or contextualized representations [
28].
Based on Emotional Processing Theory, hot memories are thought to be stored as sensory-perceptual representational networks, containing memories of stimuli in different modalities, together with cognitive and emotional states experienced during the event. Such representational networks may be created for any emotionally significant event. However, in the case of a traumatic event, the generated representation (called a fear network) would be unusually expansive and contain a great number of sensory, cognitive, emotional and physiological elements, most of which were previously considered safe and non-threatening. The large number of such elements and the strong connections formed between them mean that activation of just one element in the network may be enough to activate the entire structure. As the activation of the network (such as, in the form of a flashback) is a frightening event, this leads to trauma survivors avoiding any elements included in the network, as well as any possible cues reminding them of these elements. This understanding of the disconnected encoding of sensory-perceptual memories accounts for the avoidance, intrusion and numbing (avoiding even positive emotional experiences) symptoms seen in PTSD.
Cold memories, for their part, are seen to be selective representations of the contextual and factual elements of the event, consciously and verbally accessible to the trauma survivor. For a single event, a fear network usually remains mostly connected to its cold memory counterpart, and some autobiographical context is maintained. However, if a new traumatic experience becomes integrated into an already existing fear network of previous traumatic experiences, this connection may be partly or wholly lost. Thus, with each additional traumatic event, the fear network grows, and the hot memories become increasingly disconnected from the contextual referents of the cold memories (such as time and place). Hence, with repeated traumatization, the resulting fear/trauma network may end up containing sensory elements and disconnected perceptual or physiological memories from many traumatic events, all mixed together with little spatio-temporal context.
Based on this clinical model of PTSD, the central treatment element in NET is activating sensory-perceptual representations of traumatic events, especially the most intensely emotional and most autobiographically fragmented ones, and reconnecting them to the contextual episodic memories of the events in question. In other words, NET aims to provide a distinct time and place for the disorganized, highly emotional memories trauma survivors have of their experiences.
As most survivors of organized violence have experienced a number of traumatic events, it is often difficult or impossible for them to identify a single worst event to be processed by traditional pure exposure methods. To address this difficulty, NET treatment begins by the participant constructing a representation of his whole life from early childhood up to the present, placing important events, both positive and traumatic, on a lifeline. All traumatic events identified in this manner are then narrated in chronological order. The trauma survivor is thus exposed to repeated and detailed elaboration of what happened during the traumatic events and may become desensitized to trauma reminders. Subsequently, physical and psychological hyperarousal and the need to avoid painful reminders should decrease. At the same time, survivors are assisted in integrating their conscious, verbally accessible memories and thoughts with sensory-based traumatic memories into coherent, emotionally versatile and meaningful stories of these important moments in their lives. Creating this trauma narrative and processing the trauma emotionally and cognitively contributes to the integration of the trauma and its meaning into an optimal self-concept and life history [
14] and to the disconfirmation of maladaptive beliefs or appraisals that may have developed after the trauma.
Current trial
The general objective of this study is to contribute to the search for the most effective, evidence-based intervention methods to help children traumatized by war. The study aims to achieve this objective by comparing the effectiveness of NET in the treatment of war-affected children suffering from posttraumatic stress reactions with both a waiting list condition and a treatment as usual (TAU) control condition in a parallel-group randomized controlled trial. In addition, the potential mediating and moderating effects of a number of factors related to memory and other cognitive processes will be explored. The trial is pragmatic by nature, being carried out inside the Finnish healthcare system and emphasizing direct applicability of its results to that system and others similar to it.
This trial aims to extend the evidence from previous RCTs [
5,
30,
31] on the effects of NET on children’s psychopathology in at least six different ways.
First, in comparison to earlier studies carried out on NET with children, this trial benefits from a comparatively large and inclusive sample drawn from immigrant children settled or seeking to settle in Finland. Only one study has previously studied the use of NET with refugee or immigrant children in their new home country, with no active control group [
5].
Second, all previous studies on NET with children up to 2014 have involved one or more of the developers of NET in some position. This trial will be carried out by independent researchers with no affiliation to the institutions involved in the development of NET.
Third, whereas much of the earlier research on NET may be more accurately described as efficacy studies, this trial is one concerned with effectiveness in a current real-world clinical framework. Thus, it tends towards the pragmatic end of the pragmatic-explanatory continuum, in the sense defined by the extension of the CONSORT statement on pragmatic trials [
32]. It is to our knowledge the first such pragmatic effectiveness study on NET with children carried out in a high-income setting. The setting in question is the existing Finnish healthcare system, mostly in the Tampere region, including three outpatient clinics and two in-patient psychiatric wards. The clinicians carrying out the intervention and acting as assessors are healthcare professionals who would in any case treat these children, mostly psychologists, psychiatrists and psychiatric nurses, some with formal psychotherapy training. The pragmatic nature of the trial adds to its practical significance, as a new treatment method and targeted intervention will be rolled out in the context of the Finnish healthcare system.
Fourth, going beyond studying the simple effectiveness of NET, this trial aims at improving our understanding of the process of healing and recovery from posttraumatic stress symptoms due to war exposure. The trial does this by analysing some of the mechanisms of change that the theory underlying NET suggests would contribute to its success or otherwise.
Based on that theory, we expect that the effects NET has on posttraumatic stress symptoms would occur through recontextualizing disconnected sensory-perceptual memories, and linking them with their verbalizable episodic memory counterparts. This process would result in and be evinced by less fragmented memories of traumatic events with more spatial and temporal contextualization and coherence. Such memories would also be less vivid, biased, and intrusive and would include more verbal and fewer sensory elements.
In addition to changes in memories of traumatic events, this study considers improvements in dysfunctional, overly negative appraisals of the trauma and its sequelae [
33] as potential mechanisms of change. The theory underlying NET [
27] would suggest that symptom reduction in PTSD via exposure methods might also be achieved through reductions in such dysfunctional cognitions related to the trauma. Some evidence already exists of the involvement of dysfunctional trauma-related cognitions in the maintenance of PTSD symptoms in children [
34,
35], as well as for changes in trauma-related cognitions predicting PTSD symptom reduction in adults undergoing prolonged exposure [
36] and cognitive processing therapy [
37].
Fifth, by including a wide range of children without strict exclusion criteria and collecting a wide range of biographical information on these children together with information on peri-traumatic dissociation and cognitive performance, the trial also aims to explore factors possibly limiting the effectiveness of NET. At the same time, it may be possible to identify groups of traumatized children for whom NET is a particularly useful and effective treatment. Including analyses of potential moderators in RCTs has been recommended to reveal possible heterogeneity of effect sizes within samples and to avoid over-generalization of study results [
38].
The effects of significant dissociation on the effectiveness of PTSD treatments, especially with children, is still an open question. One trauma intervention among Palestinian children significantly reduced the proportion of clinical posttraumatic stress symptoms only among girls who had a low level of peritraumatic dissociation [
39]. On the other hand, for NET in particular, levels of derealization and depersonalization were not found to moderate treatment outcomes with adult refugees in Norway [
40]. As regards cognitive performance, Aupperle, Melrose, Stein, and Paulus have presented a model on how cognitive impairments may contribute to the clinical profile of PTSD and lead to the use of alternative coping styles such as avoidance [
41]. In light of this possibility, we will also study whether the effects of NET differ according to the children’s level of cognitive performance.
Finally, we study the effects of NET treatment, as compared to a waiting list and TAU, on a number of other outcome variables in addition to PTSD symptom levels. From a practical perspective, the effects of the intervention on depressive symptoms, overall psychological distress as well as resilience and cognitive performance are relevant indicators of clinical effectiveness and impact.
Taking into account the very high levels of comorbid depression in children with PTSD [
42], examining whether NET, though targeted at PTSD symptoms, might also reduce depressive symptoms is an important goal. Previous research on prolonged exposure therapy has found that successful treatment of PTSD symptoms lead to reductions in depressive symptoms as well, both in female adults [
43] and children [
44].
Very few studies have assessed the effects of PTSD treatment on cognitive performance, and there have been calls for more such research [
41]. For that reason we also explore possible deficits in attention, working memory, executive functioning and general cognitive performance, their relationship to levels of PTSD symptoms and traumatic exposure, as well as the effects of NET on such deficits, in a subsample of participants. Reviews suggest that in adults some cognitive impairment, at least impaired executive functioning and attention, is associated with PTSD symptoms, as separated from the effects of exposure to trauma per se [
45,
46]. However, twin studies of combat-exposed adults suggest this may be at least partly due to higher pre-trauma cognitive capacity acting as a protective factor promoting resilience in the face of traumatic experiences [
47].