Trauma- and Stressor-Related Disorders Among Children and Adolescents

Experiencing adverse environments early in life is associated with a range of negative outcomes for children, including trauma- and stressor-related disorders. Unlike most psychiatric disorders, trauma- and stressor-related disorders include a specified etiology as part of the diagnostic criteria. For posttraumatic stress disorder (PTSD) and related disorders, the etiology is exposure to a frightening, traumatic event. For attachment disorders, the etiology is severe social neglect. Trauma and neglect are the two most important etiological contributors to this group of disorders.

Before reviewing the most common of the trauma- and stressor-related disorders, we consider a framework for understanding how these two types of adversity may lead to disruptions of brain and behavioral development. Although excessive unwanted input and insufficient necessary input are often considered together, as when we use “maltreatment” to refer to abuse and neglect, there is growing evidence that these two types of adversity have different effects on the developing brain and compromise different circuitry (1, 2). For this reason, we find that an initial consideration for trauma- and stressor-related disorders should come from our understanding of brain development.

Contemporary studies of brain development of young children have noted that, to develop normally, the human brain anticipates specific kinds of input at select times. Periods of time when the brain is especially sensitive to specific kinds of input are known as sensitive periods. Input that occurs before a sensitive period opens or after it closes will not have the same effect on circuitry as that which occurs when input is expected.

Although much of the early research on sensitive periods was limited to animal studies and studies of perceptual systems, recent research with humans raised initially in large, impersonal institutions and later adopted or placed into family foster care has led to similar findings regarding sensitive periods in human brain development. These include findings compatible with sensitive periods in social and emotional domains. The age at which children are removed from depriving institutional care and placed in families seems to be related to their recovery from the effects of deprivation. For example, if children who were abandoned at birth and raised initially in institutions were placed in families before 6 months of age, they were less likely to show inattention/overactivity subsequently compared with those placed after 6 months (3). Similarly, after institutional placement, children placed in foster families before 15 months of age were more likely to exhibit normal expressive and receptive language (4), and those placed before 24 months were more likely to show higher IQs than those placed after 24 months (5). Earlier environmental enhancement, before the sensitive period closes, reduces the likelihood that lasting maladaptive outcomes will be observed.

What this means is that an absence of expected input during certain times may be especially harmful. However, what about excessive, unwanted input? Infants in the first year of life may become dysregulated by extremely frightening or threatening experiences, but they seem relatively protected from PTSD until their representational capacities are better developed toward the middle to latter part of the second year of life (6). Similarly, neglect that occurs after a young child forms an initial attachment relationship to a caregiving adult does not appear to lead to attachment disorders. Indiscriminate behavior, for example, appears only when severe neglect compromises the initial attachment relationship, implying that the formation of attachment closes a sensitive period.

An additional part of the framework for understanding trauma- and stressor-related disorders concerns the adaptive/maladaptive context of behavior. Children may be exposed to one or more types of adverse experiences and, as a result, they may manifest signs of one or more disorders that represent attempts to adapt to their species-atypical experiences. Generally, it is the persistence of behaviors even after the adversity is no longer present that defines what we describe as psychopathology. For example, hyperarousal, avoidance, and preoccupation with signs of danger (re-experiencing) that occur in PTSD may be adaptive in the face of ongoing threats, but their persistence after the threat is gone leads us to conceptualize them as disordered or maladaptive. Similarly, actively seeking social engagement in conditions of limited availability of human contact may have adaptive advantages for young children living in institutions, but the persistence of indiscriminate behavior after adoption into loving families is defined as pathological. Behaviors that represent adaptive responses to adverse environmental circumstances become maladaptive when the environmental adversity is no longer present but the behaviors persist.

With this framework in mind, we consider details of several prominent trauma- and stressor-related disorders that arise in response to either excessive, unwanted input or to inadequate, necessary input.

PTSD

Definition

PTSD describes a specific cluster of symptoms that follows exposure to an overwhelming and often frightening event or experience. The child may experience the event directly (e.g., a motor vehicle accident), witness it occurring to others (e.g., witnessing partner violence), or learn about it occurring to loved ones (e.g., learning a parent has been murdered). Although children’s emotional responses at the time of the trauma may vary, in the weeks and months after, children with PTSD exhibit various manifestations of re-experiencing the trauma, avoiding reminders, negative mood and thoughts triggered by the event, and manifestations of central nervous system hyperarousal.

Children as young as the second year of life who experience trauma may meet criteria for PTSD. The same symptom clusters are present as in older children and adolescents, but they manifest differently (Table 1). In addition, the algorithm for young children is adjusted so that either cluster C or D (rather than cluster C and D in older children) is necessary to meet the diagnostic threshold. This reflects in part the reduced prevalence of avoidance in young children. Preschool PTSD is defined in DSM-5 (7) to account for developmental differences in how the disorder manifests in young children (8).

Epidemiology and Natural History

Children are commonly exposed to stressful and traumatic events. In the Great Smoky Mountain Study, 68% of a community sample of 1,420 children had experienced at least one potentially traumatic event and 37.0% had experienced more than one (9). Conversely, only 13% of children developed posttraumatic symptoms and only 0.5% met the criteria for PTSD.

Interestingly, Costello et al. (10) found that “low magnitude” events (including deaths or losses) were more common and more likely to lead to PTSD in a community sample of children than “high magnitude” events such as abuse. Of course, prevalence in clinical samples is far greater than in community samples because more symptomatic children and families are more likely to seek help. There appears to be a larger variability in prevalence among studies of children compared with adults, which is consistent with DSM criteria being relatively insensitive to how PTSD presents among children.

Few studies have examined the natural history of childhood PTSD, although most indicate similarities to the general trend in adult studies that PTSD rates decline gradually over time without treatment (11–15). However, some children experience chronic PTSD over the course of many years. The available evidence indicates that PTSD symptoms in preschool children do not quickly remit (16–21).

Biopsychosocial Underpinnings

The biological basis of PTSD involves alterations in the regulation of neurotransmitters and stress response hormones, abnormalities in brain development and cognition, and recently investigated genetic and epigenetic influences. Cluster B re-experiencing symptoms can be conceptualized as a classically conditioned response mediated by the serotonin system; a trauma trigger (conditioned stimulus) activates intrusive memories of the trauma (unconditioned stimulus). Other cluster B symptoms such as nightmares may involve the dysregulation of multiple neurotransmitter systems including norepinephrine, dopamine, and GABA (22). Cluster C avoidant symptoms, including anhedonia and numbing of responses, are likely associated with dopamine deficiency and overactivation of the opioid system (23).

In response to stress, corticotropin-releasing hormone (CRH) from the hypothalamus stimulates the pituitary release of ACTH, which in turn stimulates adrenal release of cortisol. Ordinarily, elevated cortisol then feeds back to the hypothalamus, leading to reductions in levels of CRH and return to normal functioning. If the stresses are chronic or if traumas are severe, however, the system may become dysregulated and CRH levels may remain elevated. Chronically elevated CRH causes overstimulation of the sympathetic nervous system, leading to hyperarousal symptoms seen in PTSD (24). Pediatric PTSD studies measuring cortisol levels report altered levels (25–30), although they have been increased and decreased in different studies, perhaps depending on which mediating and moderating factors come into play (24).

Stress, and altered levels of stress hormones and neurotransmitters, may lead to adverse brain development through decreases in brain growth factors (31), delays in myelination (32), abnormalities in developmentally appropriate pruning (33, 34), inhibition of neurogenesis (35–37), and apoptosis (38–40). Structural imaging studies have shown that maltreated youth have smaller cerebral and cerebellar volumes compared with nonmaltreated youth (41–45).

An available attachment figure can buffer the child's response to stress, whereas an unavailable or frightening attachment figure can exacerbate the child’s fears. Furthermore, children may associate threat to the attachment figure with threat to the self (46). Scheeringa and Zeanah (6) found that children exposed to a severe trauma before 48 months of age had more PTSD symptoms when they witnessed a threat to the mother compared with children exposed to other traumas. The authors replicated this finding in a sample of children ranging in age from birth up to 18 years who were admitted to an inpatient unit in a level 1 trauma center for physical injuries; they found that witnessing a threat to the caregiver was the single best predictor of number of PTSD symptoms (47).

A meta-analysis of risk factors by Trickey et al. (48) included 64 studies of PTSD among 32,238 children and adolescents (6–18 years old). Broadly, and in line with previous meta-analyses of risk factors of adults, peritraumatic and posttraumatic factors showed larger effect sizes on persistent PTSD than pretrauma demographic factors and the severity of exposure. Peritraumatic factors such as fear and the perception of life threat during the event showed large effects, although these subjective factors are difficult to assess for young children. Posttrauma factors with large to medium effect sizes included cognitive factors (thought suppression, blaming others, distraction), psychological factors (comorbid psychological problems), and social and family factors (poor social support, social withdrawal, poor family functioning).

Childhood exposure to trauma has been associated with later problematic social interactions, additional peer victimization, and problem behaviors. Scheeringa et al. (49) found that preschool children with PTSD had significantly higher rates and more symptoms of separation anxiety disorder and oppositional defiant disorder and higher scores on the Child Behavior Checklist internalizing and total subscales compared with both preschool children who experienced trauma but did not have PTSD and healthy control groups (50).

Assessment and Differential Diagnosis

Clinicians should routinely ask children and their caregivers about exposure to commonly experienced traumatic events even if trauma is not the reason for a clinical referral. If a trauma history is endorsed, the child should be screened for the presence of PTSD symptoms using developmentally appropriate language. Obtaining information about PTSD symptoms from multiple informants is particularly important for prepubertal children because the addition of caregiver report significantly improves diagnostic accuracy (51).

Children ages ≥7 or 8 years often can report about their own trauma exposure and symptoms. Self-report measurements for PTSD, such as the University of California Los Angeles Posttraumatic Stress Disorder Reaction Index (52) or the Child PTSD Symptom Scale (53), can assist with screening and monitoring response to treatment. Instruments should be administered to caregivers when screening and assessing children of all ages.

If screening indicates significant PTSD symptoms, the clinician should conduct a formal evaluation with the child and caregivers to determine the severity of those symptoms, the degree of functional impairment, and whether PTSD is present. The most important initial determination is whether the child’s current living environment is safe and stable. If children are in danger, no meaningful treatment can occur.

Probes about symptoms of trauma must be tailored to the particular experiences of each child (e.g., “When you went past the house where the event occurred, did you get upset?”). Visual tools such as depictions of fearful or happy faces are often useful in assessing younger children. Given the frequency of comorbid conditions with PTSD (19), a thorough review of psychiatric symptoms is also indicated. Beyond symptoms, there are other crucial components in the assessment of traumatized children, including critical contextual details about the child and the traumatic event or circumstance (Box 1).

Evaluation should consider other psychiatric and physical disorders that may present, such as PTSD. PTSD symptoms such as restless, disorganized, and/or agitated activity or play, sleep disturbance, or poor concentration may be confused with attention deficit hyperactivity disorder (ADHD) symptoms. Timing of the onset or worsening of symptoms in relation to trauma exposure helps to distinguish these conditions. When predominant PTSD symptoms are angry outbursts and irritability, the condition may appear to be more characteristic of oppositional defiant disorder. These symptoms may be exacerbated if the child is being exposed to ongoing trauma triggers (e.g., the presence of the perpetrator). If the child has striking physiological distress on exposure to trauma reminders, PTSD may be confused with panic disorder. Misdiagnosis with another disorder, including social anxiety disorder or phobia may occur, particularly if the child has prominent avoidance symptoms that may preclude him or her from talking about the trauma.

Avoidant coping with trauma triggers may manifest as self-injurious behaviors, which may erroneously point to a depressive disorder. PTSD symptoms of social withdrawal, affective numbing, and/or sleep difficulties also may contribute to confusion with depressive disorder symptoms. PTSD may also mimic bipolar disorder due to hyperarousal and/or aggressive or hypersexual reenactment masquerading as hypomania. Primary substance use disorder may be suspected in adolescents because drugs and/or alcohol may be used to numb or avoid trauma reminders.

Severe PTSD symptoms of hypervigilance, flashbacks, sleep disturbance, numbing, and/or social withdrawal may mimic a psychotic disorder. In the presence of impaired sensorium and fluctuating levels of consciousness, differential diagnoses should include delirium. Physical conditions that may present with PTSD-like symptoms include hyperthyroidism, migraine, asthma, seizure disorder, and catecholamine- or serotonin-secreting tumors. PTSD among children is often associated with somatic complaints such as headaches and abdominal pain. Therefore, psychiatric assessment should be considered early in the medical evaluation of children with somatic complaints, especially those with a known history of trauma exposure.

Treatment for PTSD

Psychological interventions.

Trauma-focused psychotherapies are considered first-line treatment for PTSD in preschoolers, school-age children, and adolescents (54). Meta-analysis has found that therapies that are specifically focused on the child’s trauma are more effective than nondirective or nonspecific therapies (55). Of the trauma-related therapies, trauma-focused cognitive-behavioral therapy (TF-CBT) has the most evidence for efficacy in the treatment of childhood PTSD (54, 56). TF-CBT is typically administered as an individual therapy with caregiver involvement, although there is evidence that a school-based group format (the Cognitive-Behavioral Intervention for Trauma in Schools) is also effective (57, 58). The TF-CBT includes several components, starting with psychoeducation about PTSD, parental skills, relaxation skills, feeling identification, and cognitive processing. This culminates with the child sharing their trauma narrative, followed by graduated exposure to in vivo and imaginary trauma triggers during which children learn to use skills to reduce subjective distress experienced at reminders. This treatment also incorporates safety planning to help prevent the possibility of future trauma (8, 56). The TF-CBT has been studied and found to be efficacious compared with care as usual or wait-list control subjects in a variety of populations, including children exposed to the September 11, 2001, attacks on the World Trade Center and Pentagon (59), natural disasters (60), and sexual abuse (61). In addition, it has been demonstrated that TF-CBT can be adapted for and effective in the treatment of PTSD among children as young as 3 years (60).

Eye movement desensitization has evidence to support its use in the treatment of adult PTSD; however, studies with children are less conclusive. One randomized controlled trial (RCT) did demonstrate improvement in children’s PTSD symptoms compared with wait-list control subjects when a modified eye movement desensitization protocol was implemented (62).

Pharmacological interventions.

Although selective serotonin reuptake inhibitors (SSRIs) have been demonstrated to decrease PTSD symptoms in adults (63), evidence for their efficacy for children and adolescents is less convincing. An RCT comparing sertraline with placebo did not find that children receiving the SSRI had improved outcomes compared with the placebo group (64). Another RCT comparing TF-CBT plus sertraline with TF-CBT alone found that, although all children in the study had significant improvement in symptoms, there was no difference between children who received sertraline and those who received therapy alone (65). Reviews of the literature do not support the efficacy of SSRI use in the treatment of PTSD among children (55, 66).

In addition to SSRIs, other medication classes, such as alpha- and beta-adrenergic blockers, mood stabilizers, and second-generation antipsychotics, have been used in the treatment of adults with PTSD, but there is very limited evidence to support their use with children (66). Medications are justified primarily to treat comorbid symptoms among children and adolescents.

Reactive Attachment Disorder

Definition

Attachment disorders are a subgroup of trauma-related disorders that describe a number of abnormal and aberrant attachment behaviors that, by definition, result from social neglect and deprivation. There are two clinical patterns more frequently encountered in disturbed attachment: an emotionally withdrawn/inhibited presentation and an indiscriminately social/disinhibited presentation. In DSM-IV-TR, these two patterns were considered as subtypes under the diagnosis of reactive attachment disorder (RAD). However, given the distinct phenotypic differences between the two subtypes, as well as differences in comorbidities and prognosis, the disorders were separated in DSM-5 (7) into two distinct diagnoses: RAD and disinhibited social engagement disorder (DSED; see Table 2 for a comparison of the two disorders).

DSM-5 criteria for RAD focus on the absence or severely underdeveloped attachment between a child and caregiver. These children are described as showing inhibited, emotionally withdrawn behavior toward caregivers; failing to seek comfort from caregivers when distressed; not responding consistently to comfort when it is provided. The diagnosis also requires persistent social and emotional disturbance demonstrated by minimal social responsiveness to others, limited positive affect, or episodes of unexplained irritability, sadness, or fearfulness. Criteria for the diagnosis also includes the requirement that the child has experienced patterns of extremes of insufficient care, such as social neglect and deprivation, numerous changes in caretakers, or experience in an unusual caretaking environment (such as institutional care). This latter criterion is required because social neglect seems to be the key experience contributing to the development of symptoms; there have been no known cases of RAD without strong suspicion for severe neglect, nor have there been cases in which abuse was present without neglect (67). DSM-5 also requires that the child have reached a cognitive age of at least 9 months because before that, the child would be too young to form selective attachments. The disturbance must be evident before age 5 years.

Epidemiology and Natural Course

Most information regarding the prevalence of RAD comes from studies of children adopted out of institutional care or foster care. Nevertheless, even within these high-risk populations, the rate of attachment disorders is less than 10%, with rates of RAD being lower than those of DSED (68). Symptoms arise when the child experiences neglect within their initial caregiving relationships. Without intervention, symptoms may persist for several years. Children in the Bucharest Early Intervention Project (random assignment to foster care versus remaining in institutional care) who remained in continual institutional care showed moderate stability of RAD symptoms at each time point assessed up to age 54 months, with increased stability of symptoms among children who remained in institutional care through 54 months (69). Similarly, signs of RAD were associated with difficulty in interpersonal relationships and functional impairment at age 54 months (68). More longitudinal assessment is needed to determine the potential psychiatric, cognitive, and behavioral sequelae that may emerge later in childhood.

Biopsychosocial Underpinnings

Genetics and neurobiology.

Insights about the neurobiology of RAD are limited at present. Currently, there are no reports identifying genetic risks for RAD. However, it has been postulated that differences in genetic polymorphisms may contribute to an individual child’s vulnerability to the environmental factors implicated in the development of RAD. This is an area that has received limited attention to date.

Neuroimaging studies have reported reductions in gray and white matter volumes in studies of institutionalized children (70, 71), but it has not been demonstrated that these are associated specifically with RAD.

Social influences.

By definition, the diagnosis of RAD requires that the child has experienced serious neglect and adverse caregiving environments. Nevertheless, neglect appears to be necessary but not sufficient for a diagnosis of RAD. In addition to neurobiological contributions, it appears that differences in the social and emotional environment also mitigate risk for the development of disordered attachment and the severity of such disorders. One study showed that signs of RAD increased in conjunction with increase in adversity of the caregiving environment. In this study, investigators compared two groups of institutionalized children. One group was staying on a unit that restricted the number of caregivers that each child encountered. The other group comprised children on whose unit there was no attempt to limit the number of caregivers for a given child. They found that children who had a limited number of caregivers involved showed fewer signs of attachment disorders compared with children with a higher number of caregivers (72).

Quality of care.

The relation between the classification of a child’s attachment to a preferred caregiver (i.e., secure, insecure, disorganized, or unclassifiable) and attachment disorders is an area of interest. In the Bucharest Early Intervention Project, investigators examined this relationship in institutionalized children between the ages of 11 and 31 months. These children were assessed in the Strange Situation Protocol (SSP) with their favorite caregiver, if one existed, or one of their regular caregivers if no favorite could be identified. Signs of RAD were measured through caregiver report (73). All of the children rated as unclassifiable in the SSP had elevated signs of RAD. Because this classification was used to designate children who lacked attachment behaviors, these findings support the idea that the clinical signs of RAD represent a failure to develop a selective attachment relationship.

Assessment and Differential Diagnosis

Assessment.

Given the relational nature of attachment disorders, the assessment of RAD requires assessment of the interaction between the child and their primary caregiver(s), as well as collecting extensive history regarding early caregiving environment from caregivers and collateral sources. Signs of RAD are most apparent when observing the child’s relationship with the primary caregiver. The caregiver-child relationship may be assessed through a structured observation that includes a combination of various higher- and lower-stress activities (e.g., play, teaching/problem-solving, and separation/reunion), as well as comparing the child’s behavior with an unknown adult to that with their attachment figure. Specific aspects of the child’s behavior with their attachment figure should be areas of focus, including the presence or absence of proximity seeking in response to distress, affection toward the caregiver, looking to the caregiver for help, and cooperating with the caregiver.

For caregiver report of signs of RAD, the Disturbances of Attachment Interview has proven useful and is convergent with similar interview measures and validated against behavioral observations (73, 74).

Differential diagnosis.

Recent reviews suggest that autism spectrum disorder (ASD), intellectual disabilities, and depressive disorders are the most important considerations in the differential diagnosis of RAD (68, 74).

Aberrant social behaviors, including language delay, impaired social reciprocity, and limited positive affect, are present in both children with RAD and ASD. However, although experience with neglect is necessary for the diagnosis of RAD, children with ASD will rarely have a history of neglect. In addition, children with ASD will generally show attachment behavior with caregivers at a developmentally appropriate level, whereas children with RAD rarely or never display such behaviors. Children with ASD can also be distinguished by restricted interests and repetitive behaviors, which are not typical of RAD, except that stereotypies may occur in association with either disorder. Children with ASD also demonstrate specific communication deficits, particularly in intentional, social interactions.

Intellectual disabilities can coexist among children with RAD, likely as a result of severe neglect; however, this must be distinguished from the diagnosis of an intellectual disability in a nonattachment-disordered child. Children with developmental delays do not show reduction in positive affect, and their emotional skills are generally comparable to their cognitive skills. In addition, developmentally delayed children who have attained a cognitive age of 7–9 months should show selective attachments, unlike children with RAD, who are distinguished by a lack of preferred attachment by the time they have reached a developmental age of 9 months.

Although children with RAD and those with depression demonstrate a reduction in positive affect and social withdrawal compared with healthy peers, children with depression generally do not demonstrate aberrations in attachment relationships with caregivers. Although signs of depression and RAD are correlated moderately among children with histories of severe deprivation, the two disorders are distinct, and most children with depressive disorders do not have co-occurring RAD (68).

Treatment and Outcomes

Studies of children adopted from institutions have demonstrated that RAD symptoms seen earlier in childhood resolve after placement in families (69). These findings strongly support the hypothesis that signs of RAD can be eliminated with placement into adequate caregiving environments

Formal psychotherapies have not been formally studied in RAD, but several have been shown to promote secure attachments in high-risk samples; these include child-parent psychotherapy, a manualized, dyadic intervention used for traumatized young children and their families. This therapy focuses on emotional communication between the child and caregiver, as well as the parents’ early childhood experiences and current life stressors that may be contributing to destructive perceptions of and behaviors toward their children. Through development of appropriate communication with their child, caregivers are better able to recognize and respond effectively to their child’s expression of distress. Four RCTs support its efficacy (75).

The Circle of Security is another clinical intervention designed to enhance the attachment relationship between young children and their caregivers (76). This is a manualized, group psychotherapy intervention that uses videotape review and visual aids to emphasize the balance between a child’s need to be comforted by the parent and his need to explore his environment. In addition, the parent learns about his or her own conflicts that may lead to misreading the child’s needs. Although an RCT is under way (77), results are not yet available. A within-subject, pre-post study, however, supports its efficacy (78).

Attachment and Biobehavioral Catch-Up therapy is a 10-session intervention targeting the development of secure attachment between parents and young children who have experienced adversity. The intervention targets four potential barriers for the development of attachment in the dyad: the child’s rejection of care offered, the caregiver’s own history and its effect on caregiving, children’s special needs with regard to self-regulation, and the potential for young, maltreated children to be sensitive to frightening caregiver behaviors. Videotaped sessions of the caregiver and child are used to discuss treatment themes of parental nurturance, following the child’s lead, overriding negative reactions to the child, and avoidance of frightening behaviors. Two RCTs of the Attachment and Biobehavioral Catch-Up protocol with young, maltreated children showed that, compared with control groups (who received psychoeducational intervention), children in the Attachment and Biobehavioral Catch-Up group had normalization of their diurnal cortisol (suggesting normalization in stress response). A subset also showed a significant reduction in avoidance behaviors compared with control subjects (79, 80).

DSED

Definition

DSM-5 (7) diagnostic criteria for DSED comprise core social and behavioral features including venturing off without checking back with caregivers, lack of wariness of strangers, inappropriate approaches to unfamiliar adults, and a willingness to depart with them without being accompanied by a caregiver. Children with DSED also demonstrate a lack of appropriate physical and social boundaries, such as interacting with adult strangers in overly close proximity and actively seeking close physical contact or comfort. By the preschool years, verbal boundaries may be violated as the child asks overly intrusive and overly familiar questions of unfamiliar adults. These behaviors comprise a set of objectively defined signs of disorder (81) and have been reported in numerous studies (73, 82–86).

The criteria for DSED in DSM-5 also distinguish socially disinhibited behavior in the disorder from the impulsivity displayed by individuals with ADHD. This is included in the DSM description because several lines of evidence suggest that some signs of ADHD and of DSED overlap. Nevertheless, it is clear that children may have ADHD without socially indiscriminate behavior or socially indiscriminate behavior without ADHD, but there are often moderate correlations between the two symptom profiles (68).

According to Zeanah and Gleason (67), tying the phenotype to grossly inadequate caregiving was retained in DSM-5 for the important reason that children with a known biological abnormality do not necessarily qualify for the diagnosis of DSED. For example, children who have Williams syndrome (a chromosome 7 deletion syndrome) have been reported to demonstrate behavior phenotypically similar to those with DSED, even though the children have received adequate care.

Some have suggested that indiscriminate behavior represents an adaptive quality in the context of environmental deprivation. Seeking social contact and interaction could signify an effort to compensate for lack of expectable responsiveness. Chisholm (83), in her longitudinal study of children adopted into Canada from Romanian institutions, reported that adoptive parents did not view their children’s indiscriminate behavior as problematic. Conversely, Gleason et al. (68) showed that indiscriminate behavior was associated with functional impairment concurrently and predictively.

Epidemiology and Natural History

The epidemiology of DSED has not been well described. Insufficient caregiving appears to be necessary but not sufficient to cause DSED. Children who have experienced seriously adverse, neglectful caregiving environments have clearly demonstrated increased risk for indiscriminate social behavior compared with children who have not been exposed to adverse conditions (67). These environments can exist in child welfare institutions, as well as in neglectful conditions leading to intervention by child protective services.

Children who display manifestations of DSED can have persisting symptoms even after the provision of adequate caregiving and the development of a selective attachment relationship (67). In the Bucharest Early Intervention Project, signs of DSED remained moderately stable between the ages of 30 and 54 months among children with a history of institutional rearing (68). Stability was even higher among the continuously institutionalized group. Another study of previously institutionalized children showed continued display of overfriendly and attention-seeking behavior from age 4 to 8 years, even among those who were adopted (84). It was also found that children adopted out of Romanian institutions into Canada showed persistently elevated indiscriminate sociability 2–4 years after adoption (83). Furthermore, in the English and Romanian Adoptees Study, children who were adopted out of institutions before the age of 42 months showed persistent signs of indiscriminate behavior from 6 to 11 years of age (85).

Two longitudinal studies have tracked the course of DSED behaviors into adolescence. An older study found that children who in early and middle childhood had displayed high levels of indiscriminate behavior with caregivers showed indiscriminate behavior with peers at age 16 years (e.g., naming a new acquaintance as a best friend) (86). More recently, Rutter et al. (87), studying children adopted out of institutions, found moderate stability in indiscriminate behavior from early childhood through 15 years of age.

Studies have shown that children with DSED develop socioemotional (68), peer relational (86), and educational (85) impairments. For example, Lyons-Ruth et al. (88) found that socially indiscriminate behavior at 18 months of age predicted hostile behavior and hyperactivity at 5 years of age. Rutter et al. (85) reported that children with histories of institutional rearing who showed indiscriminate behavior used mental health services and special education more than children without such history.

Biopsychosocial Underpinnings

Many factors have been suggested as potential risks for DSED, including genetic factors, biological factors, subnutrition, cognitive factors, attachment status, and degree and length of deprivation (67).

Some have proposed models of gene by environment contributions to indiscriminate social behavior. For example, Soares et al. (89) proposed that children will be most likely to display indiscriminate behavior when they carry specific polymorphisms within the Williams syndrome critical region (the portion of the seventh chromosome that the deletion includes) and when they have experienced the low quality of care typical of many institutions.

Drury et al. (90) studied the effect of the interaction between group status in the Bucharest Early Intervention Project and specific genetic variations on levels of indiscriminate behavior over time. They concluded that individual differences in responsiveness to changes in the caregiving environment are influenced by functional polymorphisms in the serotonin transporter gene and in brain-derived neurotrophic factor. These findings may at least partly explain why only some children exposed to serious adversity develop indiscriminate behavior and why only some of these children benefit from an improvement in caregiving quality over time.

In a study of institutionalized young children in Portugal, Oliveira et al. (91) found that maternal prenatal risk predicted indiscriminate behavior. However, Gleason et al. (82) found that institutionalized children’s prenatal environment was not related to indiscriminate behavior. Results are also mixed regarding the relationship between growth delays and indiscriminate behavior. Johnson et al. (92) found that only postinstitutionalized children with stunted growth at adoption were at risk for indiscriminate behavior. In the English and Romanian Adoptees Study, however, there was no clear association between small head circumference or subnutrition and disinhibited attachment, but it was noted that head growth partially mediated the association between institutional care and indiscriminate behaviors (93, 94).

Mixed results have also been reported with respect to IQ and its relation to indiscriminate social behavior. Some studies have found no relationships (92), whereas other studies found that lower IQs were associated with more signs of DSED (78, 95).

Stunted growth, cognitive impairment, and indiscriminate social behavior are all outcomes of deprived environments, but this does not mean that they are causally related to signs of DSED. It is difficult to assess whether stunted growth and intellectual disability predispose to indiscriminate behavior when they result from the same exposure to deprivation.

Soares et al. (96) found that the absence of a preferred caregiver among institutionalized children predicted greater presence of indiscriminate social behaviors even after taking into consideration preinstitutionalization conditions, including prenatal and maternal risk. Other studies have found that even institutionalized children with preferred caregivers displayed significantly higher levels of indiscriminate behaviors compared with family-reared children (72, 73, 97). Smyke et al. (72) demonstrated that signs of indiscriminate social behavior were associated in stepwise fashion with increasingly adverse caregiving environments, at least as indexed by caregiver/child ratios. These results indicate that limited opportunities to form attachments in caregiving environments are associated with increasing risks for DSED.

Lyons-Ruth et al. (88) found indiscriminate behavior in high-risk, family-reared 18-month-old infants only if they had been maltreated or if their mothers had a history of psychiatric hospitalizations. The researchers observed that these mothers displayed uncomfortable and awkward patterns of interacting with their infants and they likened this to caregiver interaction in understaffed institutions. This suggests that indiscriminate social behavior may be related to the depth of emotional engagement of caregivers rather than to physical neglect or rotating caregivers, per se.

Pears et al. (98) found that signs of DSED among children in foster care were related to the number of placement disruptions rather than severity of maltreatment. Foster children who had greater numbers of foster caregivers demonstrated poorer inhibitory control, which was related to indiscriminate social behavior. The authors concluded that indiscriminate behavior is part of a larger pattern of dysregulation, which is associated with inconsistency in caregiving. This finding supports the inclusion in DSM-5 of repeated changes in caregivers as a type of insufficient care required for the diagnosis of DSED.

Some have suggested that indiscriminate behavior represents an adaptive strategy in the context of environmental deprivation. Seeking social contact and interaction could signify an effort to compensate for lack of expectable responsiveness. In a study of HIV-infected young children raised in institutions and families, Dobrova-Krol et al. (97) found higher levels of positive caregiving among institution-reared children with indiscriminate social behavior, an association in the opposite direction of that seen in home-reared controls. Conversely, Gleason et al. (68) showed that indiscriminate behavior was associated with concurrent and subsequent functional impairment among children with histories of institutional rearing.

Assessment and Differential Diagnosis

Parent report measures of DSED and RAD as defined by DSM-5 have shown acceptable to strong interrater and test-retest reliability with young children who have experienced adverse caregiving (67). Studies also have shown convergence between observational measures and caregiver reports of indiscriminate behavior (68, 91).

Symptoms of ADHD and disruptive behavior disorders were associated with signs of DSED among children aged 54 months involved in the Bucharest Early Intervention Project (72). There have been similar findings in other studies of Romanian adoptees. Rutter et al. (87) found that 35% of children with disinhibited attachment at age 6 or 11 years had challenges with inattention and overactivity at age 11. Despite these correlations, the existing research shows that ADHD and DSED are distinct clinical entities. For example, in the Bucharest Early Intervention Project, only four of the 20 children who met criteria for ADHD also met criteria for DSED and only four of the 16 children who met criteria for DSED also met criteria for ADHD (68).

Treatment and Outcomes

The existing research describes the effects of the intervention of family rearing (foster care or adoption) for children displaying symptoms of DSED who had previously been in institutions (83–85). The results regarding the effectiveness of adoption for ameliorating symptoms of DSED have been mixed. Tizard’s longitudinal study demonstrated both the persistence of indiscriminate behavior and its reduction after adoption (84, 96, 99). A longitudinal study of young children adopted into Canada from Romanian institutions revealed significant increases in parent reports of attachment during the first several years after adoption, but there were no comparable decreases in indiscriminate behavior over time (83).

Researchers have also attempted to explore which aspects of enhanced caregiving (e.g., sensitivity) are crucial for reducing signs of DSED. Among children adopted into U.K. families from Romanian institutions, investigators found no relationship between indiscriminate behavior and quality of care in adoptive homes (87). This suggests that more than enhanced caregiving is necessary to remediate DSED for some children. In the Bucharest Early Intervention Project, a modest but statistically significant decline in signs of DSED was found with the foster care intervention, and placement before 24 months of age predicted the lowest level of DSED (69). Rates of DSED behaviors were significantly lower in the never-institutionalized group.

Given the current knowledge base, it seems reasonable to recommend sensitive and responsive caregiving for children with DSED because this is known to lead to secure attachment formation. Soares et al. (96) recommended that institutional caregivers be trained to promote warm, sensitive, and responsive interactions with children within a context of routines and activities that promote social-emotional development. Garvin et al. (100) noted fewer formerly institutionalized children with disinhibited symptoms when adoptive parents were provided training regarding the types of behaviors that their children may exhibit. Therefore, specific training on minimizing and addressing DSED behaviors could be provided for foster caregivers, adoptive parents, and institutional caregivers alike. Zeanah and Gleason (67) suggested that evidence-based interventions for which effectiveness has been demonstrated with maltreated children should be attempted with children with DSED (79, 101), as well as interventions that target social cognitive abnormalities. In addition, preventive efforts that specifically target the causes of the disorder need to be developed and systematically studied.

Future Directions in the Study of Attachment Disorders

Although there has been a significant increase in research on attachment disorders over the last decade, there are still a number of remaining questions that warrant further investigation. DSM-5 separated the attachment disorders into RAD and DSED based on the significant phenotypic differences between children with these disorders, as well as differences in the course, correlates, and responses to intervention. The rationale for the changes was that other than arising from similar conditions of social neglect, the two disorders differ in most other important ways, including phenotypic characteristics, correlates course, and response to caregiving or intervention (67, 102).

However, there may be differential risk factors that may make a child more vulnerable to developing one disorder over the other (or none at all). Additional study on the contribution of temperament and neurobiological factors could improve understanding of the development of RAD and DSED as well as inform more effective interventions. In addition, most studies of RAD have focused on younger children, so there is limited information on the course of this disorder in later childhood, adolescence, and adulthood. Data on whether children who recover from RAD are at higher risk for other psychiatric, emotional, or social disorders later in life would be valuable from the perspective of intervention and prevention. Given the unique and changing attachment needs as children progress through childhood, understanding the manifestations of attachment disorders in older children could inform novel treatments for these children and their families.