Caroline L. Donovan and Sonja March
This chapter describes background knowledge that a therapist should possess when working with youth who experience internalizing disorders. The chapter has four major foci. First it will discuss knowledge about the internalization of psychopathology, its presentation in youth, and the impact of comorbidity on treatment. The influence of cognitive, social, and emotional developmental issues on therapy will be presented next. Then the discussion will center around the impact of individual differences among young people on therapy, with a particular focus on demographic variables, health conditions, and learning disorders. Finally, the chapter will examine the impact on therapy of environmental factors (parental psychopathology, parenting style, and conflict) and life events (stressful life events, trauma, and bullying).
Young people, like adults, may be afflicted with any of the anxiety and mood disorders outlined in the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV; APA 1994) as well as with separation anxiety disorder and selective mutism, which are placed in the section “Disorders Usually First Diagnosed in Infancy, Childhood, or Adolescence.” Males and females are equally likely to have mood disorders in pre-adolescence, but by adolescence females are twice as likely to experience depression (Avenevoli, Knight, Kessler, and Merikangas 2008). With respect to anxiety disorders, girls are more likely to have an anxiety disorder than boys, although the differences are not large (Costello, Egger, and Angold 2005).
The presentation of mood and anxiety disorders may be somewhat different in youth from what it is in adults. For instance, for youth with mood disorders, DSM-IV adjusts time and symptom requirements. For dysthymia, symptoms need only be present for one year rather than two, while for depression youth may show irritability rather than depressed mood and failure to make expected weight gain rather than weight loss during depressive episodes. It has also been found that depressed children are more likely to present with somatic complaints than adolescents and adults, that girls tend to display more hypersomnia and reduced appetite during adolescence, and that suicide risk peaks during middle adolescence for girls and during late adolescence for boys (Avenevoli et al. 2008). Finally, for preschool children, anhedonia may manifest itself as a difficulty to enjoy play activities, and preoccupation with negative thoughts may manifest themselves as preoccupation with play themes (Luby 2010).
DSM-IV also suggests differences in the presentation of anxiety disorders in youth by comparison with adults. In specific phobia (SP) crying, tantrums, freezing, or clinging may be evident; the child may not realize that his or her fear is excessive or unreasonable; and the duration of symptoms must be at least six months. The “other” category of phobias may include avoidance of loud sounds or of costumed characters. For social phobia (SoP) crying, tantrums, freezing, or shrinking away from social situations with unfamiliar people may occur; there must be evidence that the child can develop and maintain social relationships with familiar others; the anxiety must be triggered by same-age peers as well as by adults; the child may not realize that his or her anxiety is unreasonable or excessive; and the symptoms must be present for at least 6 months. In obsessive compulsive disorder (OCD) the child may not recognize that the obsessions and compulsions are excessive or unreasonable, while in generalized anxiety disorder (GAD) only one of the six core symptoms must be present for diagnosis. For post-traumatic stress disorder (PTSD) the response may be expressed through disorganized or agitated behavior, and the re-experiencing of the event may be evidenced by repetitive play involving the trauma theme, frightening dreams where the child fails to recall or recognize the content, or episodes of re-enactment of the trauma. Thus it is evident that there are differences between the presentation of mood and anxiety disorders in youth and the presentation of the same disorders in adults.
When youth show symptoms of anxiety or depression, they frequently show symptoms of other disorders as well. for anxiety disorders, comorbidity with other anxiety disorders is most common (Ollendick, Jarrett, Grills-Taquechel, Hovey, and Wolff 2008), and our own research suggests that 89 percent of the youth involved in treatment for anxiety disorders hold more than one clinical-level anxiety diagnosis. Comorbid mood disorders are also common: 10 to 15 percent of anxious youth display comorbid depression (Garber and Weersing 2010). When comorbid depression is present in anxious youth, the young person tends to be somewhat older, the anxiety is more severe, and family dysfunction is higher (O’Neil, Podell, Benjamin, and Kendall 2010). Finally, externalizing disorders such as attention deficit hyperactivity disorder (ADHD), oppositional defiant disorder (ODD) and conduct disorder (CD) are commonly comorbid with youth anxiety (Ollendick et al. 2008). Angold, Costello, and Erkanli (1999) suggest that around 10 percent of youth with an anxiety disorder have comorbid CD or ODD and that anxiety and ADHD co-occur at a rate three times higher than what would be predicted by chance.
Comorbidity is also very common in youth with mood disorders. Comorbidity with anxiety disorders is most commonplace, with 25–75 percent of depressed youth also being diagnosed with an anxiety disorder (Angold et al. 1999, Garber and Weersing 2010). Rates of comorbid externalizing disorders seem to vary considerably across studies: ranges between 0 and 79 percent are found for CD and ODD, and rates of 0 and 5 percent are found for ADHD (Angold et al. 1999). Of particular concern is the high rate of suicidal ideation and attempts evidenced in depressed teenagers (Kovacs, Goldston, and Gatsonis 1993). Indeed, when youth with depression have comorbid disorders, the onset of depression tends to be earlier and suicidal behavior tends to be more common (Lewinsohn, Rohde, and Seeley 1998). Furthermore, when depressed youth also demonstrate suicidality, both depression severity and impairment are greater (Barbe, Bridge, Birmaher, Kolko, and Brent 2004).
From the above discussion it would seem that comorbidity is associated with greater severity and impairment, or at least with more complex diagnostic profiles. But how does this impact on treatment? For anxiety disorders, the news is relatively good. The results of two large reviews suggest that, for the vast majority of studies, comorbidity is not predictive of poorer treatment outcome for youth with anxiety disorders (Nilsen, Eisemann, and Kvernmo 2012; Ollendick et al. 2008). Furthermore, it has been found that focusing treatment on the primary anxiety disorder also results in the reduction of comorbid anxiety disorders and of other comorbid disorders such as depression, ADHD, and ODD (see, e.g., Kendall, Brady, and Verduin 2001). It may therefore not be necessary (or efficient) for clinicians to attempt to incorporate treatment components for comorbid conditions; they should rather focus on treating the primary anxiety disorder and determine later whether additional treatment is required.
For youth suffering from mood disorders, the presence of comorbidity seems less clear in terms of its effect on treatment outcome. Some reviews and large-scale studies have concluded that comorbidity, and particularly comorbidity with anxiety, is associated with poorer outcome (e.g., Emslie, Kennard, and Mayes 2011; Ollendick et al. 2008), while others have found little effect (Nilsen et al. 2012). What is clear is that suicidality predicts poorer treatment outcome and is associated with dropout, and that treating depression does not necessarily lead to a reduction in suicidality or vice-versa (Barbe et al. 2004). Furthermore, it has been found that focusing on adolescent depression has the effect of also reducing anxiety symptoms, but not externalizing problems (e.g., Weisz, McCarty, and Valeri 2006). Thus it may not be necessary for clinicians to target comorbid anxiety in depression treatment (at least not initially), but treatment should include strategies aimed at reducing comorbid externalizing problems and should allow for specific considerations related to the importance of suicidality.
Children demonstrate numerous developmental changes as they mature. With respect to youth and development and its impact on therapy, questions that are often asked include: “At what age can you use CBT with children?” and “Do older children benefit more from CBT than younger children?” Overall, for youth with anxiety and depression, age does not seem to have a large effect on treatment outcome or dropout (Berman, Weems, Silverman, and Kurtines 2000; Emslie et al. 2011; Gonzalez, Weersing, Warnick, Scahill, and Woolston 2011; Nilsen et al. 2012), perhaps because it is a proxy for developmental level (Bolton and Graham 2005).
There is some consensus that children as young as 7 years – an age that roughly corresponds with Piaget’s concrete operational stage – can benefit from CBT (Stallard 2004). During this stage children learn to classify objects and put them in series, their mathematical abilities develop, they can play games that have rules, and they can use logic to solve problems. Children at this level begin to understand reversibility and generalizability and their previous egocentrism is reduced, so that they are able to view things from the perspective of another. Also around the age of 7–10 years, the child begins to regulate his or her emotions autonomously, is able to employ distancing strategies to manage them, has a better ability to use emotional expression to regulate his or her relationships, and becomes aware of being able to feel multiple emotions about the same person. Furthermore, the child becomes able to use information he or she has acquired about his or her own emotions and those of others across contexts, for the purpose of developing and maintaining friendships (Carr 2006). Hence many important cognitive and emotional changes occur around 7 years of age.
The various changes that occur around the age of 7 may allow for a more successful implementation of particular CBT strategies. For instance, the ability to self-regulate or regulate his/her emotion autonomously may allow the child to engage in relaxation strategies when required. The emerging ability to move from the specific to the more general means that examples used during session might be more easily generalized to life outside the session. The child’s ability to take the perspective of others and to use the information learned about his/her own and others’ emotions may assist him/her in the successful learning and implementation of interpersonal skills, problem-solving strategies, and cognitive restructuring procedures.
Although 7 years is most commonly agreed upon as the age around which CBT may become useful, Stallard (2004) has argued that children as young as 5 years may also benefit. He suggests that, although CBT can be quite complex, it often is not when applied to children. Children of 5–6 years can articulate their thoughts and are able to talk to themselves. Furthermore, around the age of 5–7 children are beginning to regulate their emotions themselves and to use social skills to deal with their own emotions and those of others (Carr 2006). Thus Stallard (2004) suggests that children of 5 years can be taught less sophisticated, specific, and concrete cognitive techniques such as positive self-talk. Furthermore, given that children of this age have not yet mastered the ability to generalize, Stallard suggests providing them with information that helps them reach conclusions about specific problems they are experiencing outside the session. Thus clinicians should not necessarily reject out of hand the idea of simple cognitive techniques when working with 5–6-year-old children.
Rather than speak of age, or even of developmental level, it is perhaps more important for therapists to develop a sound conceptualization of the individual child case, as there is enormous variation in ability and developmental level among children. Indeed, it has been suggested by Bolton and Graham (2005: 17) that asking “What cognitive developmental level is needed for CBT?” is less useful than asking “What cognition is involved in the production and maintenance of the problem in the particular case?” For instance, if a child is not yet of a developmental level where he or she can engage in metacognition, then metacognitive processes will not be maintaining his or her anxiety or depression, and hence strategies targeting metacognitve processes will be unnecessary. In other words, “don’t fix it if it ain’t broke.” This points to the importance of both good assessment and strong case conceptualization. A thorough assessment of symptoms and of causal and maintaining factors is imperative, so that a sound case conceptualization can be made. For good CBT therapists, treatment is always based on conceptualization, and thus only cognitive and behavioral factors that are found to be contributing to the child’s issues should be targeted. Chapter 9 gives more details concerning case formulation and treatment planning.
Regardless of developmental level, children display a myriad of individual differences that the therapist must take into account. In particular, the literature highlights the importance of familial culture and learning disabilities as potential determinants of treatment outcome.
It is important for therapists to be aware of the ways in which a child’s ethnicity may impact upon the treatment of his or her internalizing disorder. There is some evidence to suggest that particular ethnic groups have higher rates of internalizing disorders (e.g., Twenge and Nolen-Hoeksema 2002) and are more likely to drop out of treatment (Gonzalez et al. 2011). Furthermore, although research is limited, it would seem that ethnicity is not highly associated with poorer treatment outcome for anxiety, but may be so for depression (Nilsen et al. 2012).
There are no exact guidelines as to how clinicians should work with clients who are from a culture that is different from their own. However, Harmon and colleagues (2006) discuss a number of issues that therapists should be aware of. Symptom expression may vary acoss cultures, and the family’s belief about the origins of a disorder may affect both treatment compliance and treatment outcome. In addition, the anxiety or the depressive disorder may be the result of prejudice or discrimination, and a history of prejudice or discrimination may lead to suspicion and lack of trust in the therapist. Therapists should also be aware of the family’s cultural practices in terms of organization and decision-making and should have knowledge of the child’s religion and acculturation. Indeed particular opinions, emotions, and behaviors may be prohibited or encouraged in various cultures, and attempts on the part of the therapist to increase or minimize these factors may result in a poor therapeutic alliance and in problems with treatment compliance and outcome. Thus therapists should educate themselves in the culture from which their clients come and should be aware of the various ways in which cultural beliefs and practices may impact on clients and on their progression through treatment.
Youth may suffer from a variety of learning disorders (LDs), dyslexia being the most common and the most researched one. It has been suggested that around 20–25 percent of all dyslexic young people across the span from grade 1 to university have internalizing problems (Mugnaini, Lassi, La Malfa, and Albertini 2009); and the results of meta-analyses support the notion that, compared to youth without learning disorders, youth with dyslexia demonstrate higher rates of anxious and depressive symptoms (Maag and Reid 2006; Nelson and Harwood 2011). It should be noted, however, that some researchers have failed to find an association between LDs and internalizing problems in children (see, e.g., Miller, Hynd, and Miller 2005) and that there is more evidence for internalizing symptoms than for clinical-level internalizing disorders (Maag and Reid 2006; Nelson and Harwood 2011).
Mugnaini and colleagues (2009) suggest that children with dyslexia have poor self-efficacy, are not motivated to do homework, demonstrate poor social integration, and have a sense of learned helplessness. They also note that youth with dyslexia are more likely to have insecure attachment styles, a lower sense of coherence, greater loneliness, to experience more bullying, and to be dissatisfied with the level of support they receive from parents, teachers and peers. The authors also discuss risk factors for the development of internalizing issues in the LD child, suggesting that greater LD severity, the presence of more than one LD, comorbid ADHD, borderline intellectual capacity, non-inclusion in mainstream classes, female gender, and inappropriate management of the learning disorder all place a LD child at greater risk of internalizing problems.
In addition to the psychosocial reasons for the link described above between LDs and internalizing problems, the clinician should also be aware of evidence that common brain etiology may account for at least some of the comorbidity between LDs and internalizing problems. Furthermore, the presence of internalizing issues in LD children can serve to worsen their academic achievement, as these youth are already more likely to have problems with particular cognitive functions, such as working memory and metacognition (Nelson and Harwood 2011). For example, when a young person becomes anxious, his or her attention is diverted and anxious cognitions can take up important space in working memory, which will in turn lead to even less efficient information processing (Eysenck, Derakshan, Santos, and Calvo 2007). Thus anxiety in a child with a LD may have a “double whammy” effect on his or her abilities. Clinicians should therefore look for both psychosocial and cognitive processes that might contribute to, or exacerbate, internalizing disorders in the LD child, or indeed they should check whether the internalizing symptoms are the result of unidentified learning difficulties.
In addition to the many and varied diagnostic and individual characteristics that may impact on therapy, specific environmental factors and life events have been associated with anxiety and depression in children and adolescents.
There are many stressors inherent to lower socioeconomic status (SES): lack of material and social resources, frequent chronic and uncontrollable stressful life events, health problems, dangerous neighborhoods, crowded and noisier homes, family conflict and instability, crime and exposure to violence, and frequent moves and transitions (Evans 2004; Lorant et al. 2003; Santiago, Wadsworth, and Stump 2011). The relationship between SES and youth anxiety or depression is not quite as straightforward as that demonstrated in the adult literature, and it has received significantly less empirical attention. Some studies have reported no association between SES and the internalizing disorders (e.g., Twenge and Nolen-Hoeksema 2002); some have shown that lower SES is negatively associated with internalizing disorders (e.g., Cronk, Slutske, Madden, Bucholz, and Heath 2004); some have shown mixed results within one and the same study, depending on the measure of SES (e.g., Ozer, Fernald, and Roberts 2008); and some have shown that higher SES is associated with greater anxiety (e.g., Merikangas et al. 2010).
The reasons for differences in findings pertaining to the association of SES with internalizing disorders are likely to be complex. For example, Vine et al. (2012) found different results depending on type of anxiety symptom, distinction between family and neighborhood incomes, and gender. Furthermore, although research is scant, there is some evidence to suggest that SES may impact in various ways upon the treatment of child anxiety and depressive disorders. For example, it has been found that higher SES may be related to better treatment outcome, at least for adolescent depression (Emslie et al. 2011); that disadvantage may be significantly related to dropout (Kazdin, Mazurick, and Bass 1993); and that lower income and lower social class may be more strongly related to dropout for younger than for older children (Pekarik and Stephenson 1988). Thus the relationship between SES and internalizing disorders in youth is complex, and clinicians should not only be aware of the child’s SES, but also assess and take into account the broader implications and bidirectional impact associated with it.
Living in a single-parent household has been inconsistently associated with higher dropout from treatment (e.g., Gonzalez et al. 2011) but has not been found relevant to treatment outcome. There is evidence to suggest, however, that family conflict, whether between parent and child, mother and father, or child and sibling, is associated with anxiety and depression (Drake and Ginsburg 2012; Kane and Garber 2004; Sheeber, Davis, Leve, Hops, and Tildesley 2007). Interparental conflict in particular has received considerable empirical attention, being consistently associated with youth anxiety and depression (Drake and Ginsburg 2012; Restifo and Bögels 2009). Indeed, it has been suggested that interparental conflict, rather than marital discord, marital status or divorce per se, affects child internalizing disorders (Drake and Ginsburg 2012; Rapee 2012). Furthermore, the effects of parental conflict have been found to be worse when there is aggression or the threat of aggression, when a parent is angry and withdrawn, and when the parent uses avoidant conflict tactics (Restifo and Bögels 2009).
The scant evidence conducted to date suggests that family conflict and dysfunction are associated with poorer treatment outcome for depression (Emslie et al. 2011) and anxiety (Crawford and Manassis 2001) in youth. Various mechanisms through which interparental conflict has its effect on internalizing disorders have been suggested: impairment of parenting practices; disruption of the youth–parent bond; parental modeling of anxiety, withdrawal, poor coping, and maladaptive conflict resolution; child appraisals of self-blame; and threat associated with the conflict (Drake and Ginsburg 2012). It is thus important for clinicians to look for family conflict (child–parent, interparent, and child–sibling conflict) and to incorporate it into the conceptualization and treatment plans for youth with internalizing issues.
Stressful life events have been found to be associated with internalizing and externalizing symptoms in the vast majority of studies (Grant, Compas, Thurm, McMahon, and Gipson 2004), even those investigating preschool-age children (Luby 2010). This section discusses particular types of life events, trauma, chronic health issues, and bullying, with respect to their relationship with youth anxiety and depression.
Unfortunately, young people may be exposed to numerous types of trauma, and consequently go on to develop anxiety and depression in response to them. One of the most studied forms of trauma is childhood sexual abuse (CSA). With respect to the effect of CSA on childhood internalizing disorders, it would seem that mood rather than anxiety disorders are a more common result. In his review, Rapee (2012) concluded that CSA increases the risk of anxiety disorders in children, but that the link is not as strong as it is with other disorders. The effects of CSA on the development of depression appear somewhat stronger: there are higher rates of depression in abused compared to non-abused children and the severity of the resulting depression is higher when there has been sexual abuse with contact, when the child knows the perpetrator, and when the child is male (Putnam 2003).
With respect to trauma more generally, the clinician should be aware that, when adolescents experience it, trauma tends to result in specific mood and anxiety disorders, whereas in younger children the effects on mood and anxiety are more diffuse and can take on a variety of internalizing difficulties (De Young, Kenardy, Cobham, and Kimble 2012; Yule 2001). Furthermore, it has been suggested that adverse effects following trauma are more likely to develop in children when the events are ongoing, when the events lead to longer term disruptions in the child’s life, when the child has had a high level of exposure, when the child is female, and when the child has less social support, or when the child’s social support networks are disrupted due to the trauma (Pine and Cohen 2002). Clinicians should therefore look for trauma and the risk factors associated with it, so that a thorough case conceptualization and treatment plan can be made.
Unfortunately many children have to contend with a chronic illness, and chronic illness has been associated with internalizing disorders. Youth with chronic illnesses such as epilepsy, asthma, diabetes, or migraine demonstrate higher levels of depression (Pinquart and Shen 2011). It would seem that this link is evident even in very young children, as chronic illness is also associated with the early onset of depression in preschool children (Curtis and Luby 2008).
When one considers the difficulties faced by youth beset by chronic illness, it is not surprising that they frequently develop internalizing disorders. Although a full review of health conditions is beyond the scope of this small section, it may be illustrative to take the example of the child suffering from type 1 diabetes. Diabetes can have numerous serious and potentially lethal medical complications if one does not adhere to the regime very closely, and the burden of treatment is heavy on youth afflicted by this disease. A number of areas that young people with diabetes tend to find stressful have been identified (Rubin and Peyrot 2001). First, dietary restrictions can be worrisome and upsetting for youth. Birthday parties and other social events frequently revolve around foods such as sweets, which diabetic children are unable to consume. As a result, these children may feel different from others, deprived, and anxious that they may appear to be “weird” in the eyes of their peers. Second, there can be stress associated with monitoring blood glucose levels, the unpredictability of blood sugar levels being a strong cause of stress and frustration. The finger prick itself, required to monitor blood glucose levels, can be painful and become associated with behavioral problems and anxiety. Third, insulin delivery can be very stressful, both in terms of needle phobia on behalf of the child and in terms of concern around inducing hyperglycemia by administering too much insulin. Finally, there can be much conflict around diabetic management, as the developing need for autonomy in the adolescent, the fear of injections, the deprivation of particular foods, and the concern about being “different” from peers leads to resistance to the diabetic regime and distress around it.
This example highlights a number of areas that the clinician should be aware of when working with a child with chronic health concerns and comorbid anxiety or depression. First, the clinician should be aware that depression and anxiety are common in chronic illness and should be assessed for. Second, given that chronic health conditions are often associated with physical pain, the clinician should ascertain whether the anxiety or the depression is associated (even in part) with pain in some way. Third, the clinician should assess for, and be aware of, any links (uni- or bidirectional) between the required medical regime and the child’s experience of anxiety or depression.
Unfortunately bullying among children is highly prevalent and is associated with a number of deleterious consequences, including anxiety and depression (O’Brennan, Bradshaw, and Sawyer 2009). It has been suggested that childhood is an important time for the development of cognitive vulnerabilities for depression and that bullying might be one pathway in which these depressive cognitions develop (Gibb, Stone, and Crossett 2012; Sinclaire 2011).
Within the bullying area there are distinctions made between bullies, victims, and those who are referred to as bully-victims (those who have been both a bully and victim). Unsurprisingly perhaps, victims of bullying are at high risk of developing internalizing problems, while bullies are not (Juvonen, Graham, and Schuster 2003; Nansel, Craig, Overpeck, Saluja, and Ruan 2004). Interestingly, however, it has also been found that bully-victims are the most likely to display internalizing symptoms and disorders (O’Brennan et al. 2009). Clinicians should be aware that internalizing problems associated with bullying seem to increase as youth get older, that the association is stronger for boys, and that social support has been found to be an important buffer between bullying and the development of internalizing disorders (Sinclaire 2011).
A distinction between direct and indirect bullying is also made in the literature. Direct bullying involves physical contact or threat thereof, while indirect (or relational) bullying involves covert social manipulation, such as exclusion of the victim and rumor spreading. Interestingly, youth seem to be more at risk of developing internalizing issues if they are the victim of indirect rather than direct bullying (Marini, Dane, Bosacki, and Cura 2006), indirect bullying being more directly related to changes in depressive cognitions than direct bullying (Sinclaire 2011). Thus clinicians should ensure that they assess for bullying, both indirect and direct, when working with youth with internalizing disorders, so that appropriate strategies can be included within treatment.
It is hoped that this chapter has provided the clinician with some important background knowledge regarding particular child characteristics that may impact on therapy. The take-home message we want to impart is that a thorough assessment of psychopathology, developmental issues, individual child characteristics, and environmental factors is essential. Through knowledge and subsequent assessment of the factors that may impact on youth internalizing disorders and treatment, the clinician can develop a more accurate case conceptualization, which will ultimately lead to greater therapeutic success.