Chapter 43. Posttraumatic Stress Disorder in Children and Adolescents Following a Single-Event Trauma

Single-event traumas experienced by children and adolescents include involvement in motor vehicle accidents, natural disasters, and other unexpected, paroxysmal events such as being assaulted or raped. In environments where the latter occur with unfortunate regularity, such as when a family experiences protracted domestic violence, ongoing sexual assault or lives in a war-zone, the individual is likely to be exposed to multiple traumatic experiences and the mental health sequelae are qualitatively different to the experience of a single-event trauma. Readers should consult the chapters on dissociative disorders, personality disorders, and references to sexual abuse to further understand the sequelae of exposure to multiple-event trauma. This chapter will focus on the mental health outcomes following exposure to a single-event trauma. Typical diagnoses include posttraumatic stress disorder (PTSD), other anxiety disorders including specific phobias, and depressive presentations. The Diagnostic and Statistical Manual, fourth edition, Text Revision, criteria for these conditions are detailed in Chapters 20, 41, and 38, respectively.

Two diagnostic considerations influence the chapter. The overarching frame is that of developmental psychopathology. Individuals function by integration of subsystems that interact vertically (e.g., gene to cell) and horizontally (person to person). Infants, children, and adolescents also need to organize across time. Time course introduces new developmental constructs: novelty of cause and effect of inputs at different developmental stages, differential rates of development across subsystems, critical periods, and negotiating developmental challenges such as the changing relationship between parent and child. The developmental science perspective subsumes other useful heuristics such as a bio-psycho-social and a systemic perspective. Inherent is the conceptualization of an individual's trajectory over the infant–child–youth developmental span and abnormality as deviance from a normal trajectory. The concept of a normal trajectory is very useful in understanding the impact of diverse traumatic events. Repetitive traumatic events and their sequelae are discussed elsewhere in this text (see Chapter 42 on Child Maltreatment). The cumulative adversity inherent to experiencing repetitive traumatic events leads to a picture of continuity of developmental abnormality. That is, the child has been symptomatic and functionally impaired for some time. The mental health sequelae of a significant isolated, single-event trauma is seen as a developmental discontinuity—a subjective deviation from the developmental norm. For a more comprehensive account of the developmental approach see Cairns and Angold (1996). The seminal paper of Lenore Terr advocated the useful nosology of type I trauma describing the developmental discontinuity, type II the developmental continuity (Terr, 1994).

An adverse mental health outcome following a single-event trauma is not limited to PTSD. Clearly PTSD is of central importance, however, researchers have long noted other postdisaster anxiety and depressive states. Current data consistent with greater disaster-related fear of death and event exposure being more likely to have a PTSD/anxiety outcome, while greater loss and grief predisposes to a depression outcome. However, PTSD-depression symptom concordance is great and comorbid presentations are common.

The post-single-event trauma burden of mental illness is not restricted to new mental health presentations. Many individuals with preexisting mental illness will experience a postevent symptoms relapse. Reasons may be pragmatic, such as difficulty obtaining medications or access to therapy in the postdisaster environment. Systemic issues such as an increased likelihood of parent distress and/or depression may be significant. Psychopathology may be related to event phase; immediately postevent individuals and service providers are appropriately focused on physical safety, water, food, and shelter needs. Mental health presentations and service provision generally begin in the weeks following the disaster. It is often the case that mental health issues come to dominate in the months and years postdisaster. An example is major depression some months following a disaster, yet with clear temporal links to disaster-related events.

Epidemiology

Estimates of the prevalence of post-single-event PTSD vary greatly. Uniform PTSD rates would be unexpected given every disaster or traumatic event is unique. PTSD estimate variability is also due to disparity in research design. Published research varies in the elapsed time since the disaster, developmental stage of participants, whether the child or parent was the primary informant, and instruments used to measure psychopathology; the last mentioned range from qualitative accounts, self-report questionnaires to structured diagnostic interviews. Some early studies included data obtained from mental health professionals working for either the plaintiff or defense in lawsuits after a disaster. Design factors include bias due to selective mortality, recall and interview bias and use of convenience samples. Significant “unpackaging” of the event, both of event-related variables and the personal meaning of the event is required for a more meaningful between-event comparison.

Examples of prevalence variability include PTSD rates between 90% and 100% of children who were kidnapped and endured harrowing hours in captivity and children trapped in a school playground by sniper fire, intermediate rates of 40–60% in children who experienced major transport disasters and lower rates of 5–15% of children screened after natural disasters. Occasional studies report no children meeting diagnostic criteria for PTSD. Even in these examples limited symptom PTSD and other anxiety disorder presentations are usually reported.

Single-event trauma research is often population-based initiatives following natural disasters. Significant research initiatives have followed the devastation of Hurricanes Hugo (1989), Andrew (1992), Iniki (1992), and Katrina (2005). Research findings include female gender, younger age, high-trait anxiety, and emotional reactivity during hurricanes independently predicted PTSD symptoms more strongly than self-report hurricane exposure. Australian wildfire disasters have been subjected to similar scrutiny and surprisingly similar rates of child and adolescent PTSD have been reported, as well as high-trait anxiety as a PTSD-predictive factor and high rates of comorbid depression. Natural disasters potentially involve large numbers of individuals and/or a large proportion of a given population. An epidemiological perspective is informative and a public health approach to service provision justified. For example, a sophisticated postdisaster-intervention strategy would include screening for PTSD, the latter consistent with the National Institute for Clinical Excellence (NICE) guidelines, followed by universal and targeted responses to identified children, adolescents, and families.

Etiology

A potentially confusing range of causal theories have been advocated for PTSD presentations in childhood. Such theories are often extrapolated from adult research and include psychodynamic, behavioral, cognitive, and biological explanatory models. Given variations in a potential victim's integration of biological, emotional, and behavioral systems and age-appropriate differences in individual–context, individual–peer, and individual–family interactions, stressful events may affect individuals in different ways across the life span. Etiology can be divided into research findings, often from epidemiological designs, that identify risk, vulnerability, and resilience factors. PTSD is also informed at a more fundamental level by cognitive theory and neurobiological research. The area is not without complexity; seemingly simple associations such as gender, age, and PTSD have proven difficult to clarify. Published findings include no effect of age on PTSD prevalence, higher rates in older children and adolescents, and higher rates in younger children. McDermott and Palmer (2002) studied 2379 grade 4–12 students after a wildfire disaster and reported a peak prevalence of PTSD in the middle school years.

Risk Factors

An overarching principle is whether the functioning of the individual with PTSD has diminished. If significant, this diminution will have rendered the individual's functioning below the normal trajectory for age. A second principle is that the impaired functioning is due to a combination of event-related, proximal factors and distal factors inherent to the individual and present before the trauma. The event-related factor with the greatest empirical evidence of association with subsequent PTSD symptoms is increased traumatic event exposure. The event-exposure–PTSD association has been demonstrated in children residing in cities subject to more earthquake damage, witnessing more hurricane damage, closer proximity to gunfire, and being closer to actual wildfire flames. Note that high exposure, while a useful summary variable, is itself a multifactorial construct that includes the perception of threat of death, concern about possible loss and bereavement, witnessing a range of disaster-related phenomena including death or serious injury, perceptual experiences such as smoke from fires, and witnessing the coping or lack of coping of parents, caregivers, and other adults. Dislocation in the immediate aftermath of a disaster has been argued as a positive factor by decreasing disaster exposure and a negative factor given the lack of continuity with family and community and removing the opportunity to constructively engage in postdisaster restorative activities. In a case-controlled study of Armenian child earthquake survivors, at 30 months postdisaster, there was no difference in PTSD and depression symptoms in a group of children removed from the disaster and those who remained, distal factors significantly associated with higher PTSD symptoms in children include high-trait anxiety, a past history of emotional problems, children with a subjective sense of low self-efficacy and low social support, children employing poor coping strategies and an internal causal attribution.

Cognitive Factors

Cognitive information-processing theory postulates that there is the potential of being overwhelmed by the everyday volume of information requiring processing. To counter this, information is encoded and recalled in specific ways. Cognitive schemata are developed as a primary organizing heuristic. The schema incoming information interaction influences subsequent memory encoding and retrieval. It is postulated that trauma-related information is also encoded with relation to existing schema. Dissonance with existing schema leads to conflict and the disruption of the usual memory encoding and comprehension and memory retrieval sequence. The greater the dissonance, the greater is the likelihood of trauma symptoms and impairment.

One postulate is that there are very specific schemata, which, if violated, are more likely to precipitate trauma symptoms. Fundamental beliefs such as “good things happen to good people” may be radically undermined after a traumatic event such as the experience of being raped. Possible psychological responses to information dissonant to existing schema include assimilation, altering information to fit preconceptions and accommodation, changing existing schemata to account for, and accept the new information. The last mentioned may facilitate recovery. Unfortunately accommodation may create a secondary impairment if, for instance, new schema are punitive, self-deprecatory, undermine adaptive beliefs about safety or intimacy, and make the tasks of daily living more difficult. Child and adolescent trauma complications can be more complex given the added possibility of altered parent schema about their child's safety. An example is the double jeopardy of the child who experienced a paroxysmal, unitary (type I) sexual trauma. Such an event may lead to PTSD psychopathology. If, for a protracted period thereafter, altered parent–child safety schema leads to the child not being allowed to visit friends, go on school outings, “sleep-overs,” and other events without direct parent monitoring, the normal process of child–adolescent individuation–separation may be retarded generating a secondary impairment. Further, accommodation in young children may manifest as self-blame for the event, heightened by a child's tendency to regress to a more egocentric worldview. Or by developing schema that predicts future disastrous events and ongoing extreme danger, either to themselves, friends, or family. Other changed schema may be that the event will reoccur, the event will become greater with a larger number of affected people and parents and other adults cannot protect children from traumatic events.

Biological Factors

Neurobiological research including neuroimaging studies has been less active with child and adolescent participants. Studies that have been conducted are more likely to be with maltreated children than children who experienced a single-event trauma. Extrapolation from research with adults and maltreated children suggests possible decreased intracranial and cerebral volume in children who experience a single-event trauma. However, it may be that multiple insults during critical periods are needed for altered brain morphology. Neuroendocrine investigations suggest that dysregulation of the hypothalamus–pituitary stress system is likely following traumatic events. More definitive conclusions await further research.

Detailed understanding of the genetic aspects of an individual's response to single-event trauma is not yet available. Consistent with many psychiatric presentations, the concordance for monozygotic twins, in this case those exposed to a traumatic event and who develop PTSD, is higher than for dizygotic twins. Monozygotic twins also share their intrauterine environment and so the monozygotic twin excess may also be due to nongenetic factors. In the review of Broekman and colleagues (2007) the authors make the assumption that genetic effects are likely to be polygenetic and note a range of target endophenotypes such as alterations in the HPA axis. They also report methodological difficulties such as small sample sizes that make current findings inconclusive. Theoretical problems include differentiating potential genetic vulnerability to PTSD versus a genetic vulnerability to being exposed to a traumatic event, and so different pathways for a genetic effect to be expressed are possible. One clear advance is the gene–environment studies of early life trauma and subsequent depression or antisocial behavior at age 26 years (Caspi et al., 2003). The relationship of functional polymorphisms in the serotonin promoter region and monoamine oxidase gene modifying the long-term outcome of early traumatic experience is an exciting area of research and likely to inform our understanding of individual resilience and vulnerability to traumatic events.

Assessment and History Taking

No two single-event traumas are identical either at individual or population level. The understanding that trauma exposure can be “lumped” into a nominal variable is simplistic. It is more typical that within the event setting exposures vary widely. For the clinician it is important to “unpack” the potentially traumatic event. One useful strategy is to obtain a chronological narrative, depending on the duration of the event, at 1-, 10-, or 60-minute intervals postevent. The clinician should consider potential traumatic stimuli in all sensory modalities, for example, what did the child see, hear, smell? What were the child's thoughts at different stages of the event? What is their understanding of what occurred and the meaning they gave to the event or the roles individual's played. What were the reactions of others, especially adults in the vicinity?

The type of traumatic event, for example, experiencing a motor vehicle accident, is usually identifiable in the PTSD-symptom content including the subject matter of flashbacks, dreams, and nightmares; symbolic reminders; and, if accessible, cognitions underlying avoidance behavior. Some individuals with intellectual impairment and the very young will not have a verbal narrative of the event. In these cases, the clinicians seeks to identify behavioral discontinuities, temporally related to event exposure, in various domains such as sleep, satiety, elimination, aggression, and impulsivity. Another useful exercise is to consider the event from a non-mental-health perspective. As is often the case following natural disasters, establish, if there was environmental degradation, significant postevent economic hardship (short and long term), disruption of services (electricity, water, sewerage), and disruptions to institutions such as schools? Failure to provide services in the postevent setting emphasizes personal and collective powerlessness. Postevent concerns in these domains may have little relationship to PTSD symptoms; nevertheless, be very significant in understanding overall impairment levels.

Signs & Symptoms

The clinical presentation of child and adolescent PTSD is dependant upon the child's age; given the age is a proxy for underlying developmental constructs such as cognitive, speech, and language development and sophistication of peer and societal relatedness.

Infant School Children

The lack of systematic research is more pronounced in preschool and infant school children, in part because of difficulties administering questionnaires and completing semistructured interviews with young children.

Most knowledge of PTSD in this age group is from expert opinion. Young, preverbal children communicate distress by behavioral change temporally linked to a given trauma. General signs include alterations in the ease of feeding, sleeping, or generally settling the child. Parents may report regressed behavior including the child's unwillingness to explore the environment, increased stranger danger, increased aggression, or clingy behavior.

Verbal preschool children may display or voice broad emotions such as anger, sadness, and excitement. Mixed or rapidly changing emotional states are frequent. Separation anxiety is common, so too are specific trauma-related fears. Posttraumatic play, in Terr's terminology “reenactments” given the absence of the fun element of play, is typical (Terr, 1994). Reenactments are compulsive, repetitive behavioral sequences that are unconsciously linked to the traumatic event. Nontraumatized children can be involved in a traumatized child's reenactment. This involvement can place children in danger, for example, setting or playing with fire following a bushfire disaster.

Primary and High School Children

With increasing age, symptomatology is more typical of adult PTSD. Age-related phenomena still exist such as aggressive or withdrawn behavior; however, behavior becomes more sophisticated. Fear of death, separation anxiety, or fear of the event recurring are common. Magical thinking and ascribing omen status to events before the trauma is common in younger children, so too are phenomena such as nightmares and sleep disturbance. Flashbacks are reported; however have a more daydream quality than the sudden, intrusive adult phenomena. A fear generalization gradient has been described such that stimuli approaching the traumatic event evoked increasing distress. For example, fear seeing any car, but greater fear seeing the same color car that was involved in a motor vehicle accident.

Denial and disavowal of the traumatic event is not often seen in childhood PTSD presentations that follow a single-event trauma. However, children do often withhold the extent of their distressing experiences from their parents, either because of perceived difficulties talking to their parents following trauma or children specifically not wanting to make their parents more anxious by the added burden of their feelings of distress. Another distinction from adult PTSD is that numbing and restriction of affect is not frequently reported by children, in part because of the difficulty in younger children understanding and discussing these concepts.

Psychological Testing

Given the reticence of many children to discuss their PTSD symptoms and the low parent identification posttraumatic symptoms, rating scales can play a valuable role in postevent screening and symptom identification. Some measures such as the PTSD-reaction index can be administered by a clinician as a semistructured interview or by child self-report. Care is required to assess the developmental appropriateness of item wording; most children younger than 8 years of age will have difficulty with multiple response answer fields but can complete Yes/No formats. Questions to adolescents should include important comorbidities such as postevent substance abuse. See Chapter 41 for more details on specific measures.

Laboratory Findings and Imaging

There are currently no practice recommendations advising the routine use of specific laboratory tests or neuroimaging studies in children or adolescents who experienced single-event trauma.

There is a danger in equating single-event trauma with PTSD. Care should be taken to elicit or exclude PTSD diagnostic criteria as well as excluding depression and other anxiety disorders especially phobias, panic disorder with and without agoraphobia and generalized anxiety disorder. For some children the primary presenting complaint will be a sleep disorder or a restrictive eating pattern. Increased physical symptoms such as musculoskeletal complaints and alcohol and substance use have been reported in adolescents following a single-event trauma.

Overarching treatment concerns are the need to be developmentally appropriate, especially the communication style of children at different developmental stages.

Psychopharmacological Treatment

The Cochrane PTSD review of Stein and colleagues (2006) identified 35 randomized controlled trials, found no trials with pediatric subjects, and concluded medication was effective in decreasing PTSD and depression symptoms and impairment in adult PTSD sufferers. Dedicated child and adolescent reviews concur with the absence of research in the pediatric domain and suggest, at the level of expert opinion, that SSRI medication may be useful for general PTSD symptoms, while alpha agonists may have a role with hyperarousal symptoms. The UK NICE guidelines advice against prescribing for children and adolescents with PTSD, with the caveat that medication may have a role in acute postevent sleep disturbance.

Psychotherapeutic Interventions

Current treatment guidelines emphasize that many individuals who experience a potentially traumatic event do not become symptomatic or have mild symptoms that spontaneously remit over the first month. There is currently no evidence that a single session debriefing intervention will lead to greater or more rapid symptom reduction.

Psychological First Aid

Psychological first aid (PFA) is recommended. PFA is an initiative of the National Child Traumatic Stress Network (NCTSN) and the National Centre for PTSD. The Field Operations Guide, second edition is available from the NCTSN website. This resource provides training on preparation to provide PFA and engagement issues. Core intervention areas are then covered: safety, mental status stabilization, gathering essential information, practical assistance, facilitating social connectiveness, psycho-education about emotional trauma and grief, and service linkage.

Trauma Focused Cognitive–Behavioral Therapy

If symptoms persist or cause impairment then treatment should be instigated. The current treatment of first choice is a Trauma focused cognitive–behavioral therapy (TF-CBT). While in children treatment will often include some element of play and relaxation therapy, it seems for effectiveness therapy must address the specific trauma memories and ameliorate the impact of such memories by CBT techniques such as exposure and habituation and/or cognitive restructuring. More indepth analysis is difficult given the current level of evidence, Taylor and Chemtob reported only 8 trials out of 102 identified included more than a single subject design or a control group. There is some evidence that early presentations may require brief five-session interventions. Later presentations or those that include comorbid traumatic bereavement, disability, and past history of other traumatic events may require longer intervention, that is, 8–12 sessions. TF-CBT delivery may be equally effective by group or individual therapy. Online training in TF-CBT can be obtained via the NCTSN website. One study demonstrated effectiveness of EMDR in children following a natural disaster. PTSD treatment should generally be initiated first. Comorbidities such as depressive symptoms may remit with PTSD therapy. If the severity of depressive or substance-abuse symptoms does not allow effective TF-CBT, these issues may need to be addressed first. Lastly, the usual child and adolescent strategies of parent or caregiver involvement and addressing systemic issues such as school-based initiatives to limit impairment are recommended.

Unnecessary treatment with medication or psychotherapy can occur following the good intentions of parents and clinicians to treat children who are exposed to a traumatic event. A competent assessment should prevent undue intervention. Similarly, all screening tests have false positive and negative rates. Negative screening results should not preclude subsequent referral for diagnostic clarification. In an ideal service provision scenario, an individual with a positive screening result on a self-report measure would then be individually assessed to determine the need for an intervention.

Time may diminish the emotional impact and symptoms of PTSD. Possible mechanisms include habituation to the reminders of a given event and a process of more normative forgetting. However, symptom reduction may be misleading. A perception of mastery may in reality be diminished reexperiencing symptoms as a reaction to greater emotional numbing and avoidance symptoms. Over time, mastery of trauma-related fears may occur with the reestablishment of non-trauma-dependant schema. However, symptom chronicity is well described. Persistent symptoms of PTSD have been reported by children and adolescents 6 months after shipping disasters, the 2004 Indian Ocean Tsunami and wildfires, 1 year after a sniper attack on a school, 18 months after an earthquake disaster, and 2 years after bushfires. Medium-term outcome studies, defined as 1–5 years postdisaster have demonstrated symptom chronicity. Few long-term outcome studies have been undertaken. Green and colleagues conducted a 17-year follow-up of survivors of the Buffalo Creek mudslide disaster. Generalizing from the Buffalo Creek findings is difficult given the original research included data obtained retrospectively from court reports prepared for victims. Original estimates of PTSD were 32% of children experienced “probable” PTSD. At 17-year follow-up, the PTSD rate had decreased to 7%. More remaining PTSD sufferers were women. The only symptoms that increased over time were substance abuse and suicidality. Clinicians should be alert for anniversary reactions, which may increase over time and often their occurrence is not immediately attributed to the traumatic event (Terr, 1994). Anniversary symptoms may include any symptom in the PTSD spectrum as well as altered beliefs such as the view that their lives would not be full, long, and their career or marriage successful.