CHAPTER 18 Internalizing Conditions
18A. Mood Disorders
There is a growing emphasis on psychosocial problems in pediatric care, and pediatricians are in a unique position to monitor children over time to prevent, identify, and address psychosocial concerns.1 Primary care visits account for an increasing portion of mental health visits, and pediatricians play an important role in the management of childhood mood disorders.2,3 Mood disorders in childhood and adolescence have received increased clinical and research attention. Mood disorders include those characterized by depressed or irritable moods (major depressive disorder (MDD), dysthymic disorder, and depressive disorder not otherwise specified) and those characterized by fluctuations between depressed and manic or hypomanic moods (bipolar I disorder, bipolar II disorder, cyclothymic disorder, and bipolar disorder not otherwise specified). This chapter summarizes how depression and bipolar disorder in childhood and adolescence are currently understood, focusing on research concerning the significance, causes, diagnosis, and treatment of these disorders. Current practice standards, as well as issues necessitating further research, are discussed.
SIGNIFICANCE
Prevalence
The incidence of depression in adolescents is similar to that found in adults, ranging from 0.4% to 8.3%. Rates are lower in preadolescents, ranging from 0.4% to 2.5%.4 Epidemiology studies rarely include preschool-aged children; however the available data suggest that depression occurs in approximately 1% of preschool children.5,6 Gender ratios in rates of depression also change with age. Among preadolescents, similar rates of depression are found in boys and girls. In adolescence, the rate of depression in girls increases dramatically, resulting in a gender ratio of 2 : 1, similar to that found among adults.4
Estimation of the prevalence of bipolar disorder in childhood and adolescence is complicated by the general lack of agreement concerning the core characteristics of the disorder, which are discussed further in the diagnosis section of this chapter.7 No epidemiological studies currently exist for children. In a study of 14- to 18-year-olds, Lewinsohn and colleagues8 found lifetime prevalence rates for a diagnosis of bipolar disorder to be approximately 1%; an additional 5.7% reported subthreshold symptoms. Adolescents with subsyndromal symptoms of bipolar disorder experienced functional impairments similar to those in adolescents with bipolar disorder, which continued into young adulthood. This study was based solely on interviews with adolescents, however, and more recent research has emphasized the importance of including parent report in diagnosing bipolar disorder.9
These incidence rates indicate that a significant proportion of children and adolescents suffer from mood disorders. In addition, the overall rates of depression in children and adolescents appear to be increasing.5
Course
Mood disorders often have a chronic or relapsing course, associated with continuing impairment. Episodes of MDD last 7 to 9 months on average, with relapse rates of 40% within 2 years and 70% within 5 years.4 Weissman and colleagues10 monitored adolescents with MDD into adulthood and found that 63% had additional depressive episodes within the following 10 to 15 years. These adolescents had a fivefold risk of suicide in comparison with control participants who had no psychiatric diagnosis in adolescence. Those with MDD also experienced continuing impairment in family, work, and social functioning. Dysthymic disorder lasts an average of 4 years and is associated with a high risk of developing MDD within 2 to 3 years, which results in double depression.4 Bipolar disorder has a particularly chronic course in childhood, as evidenced by the findings of Geller and colleagues11 in their 4-year follow-up of children with bipolar disorder, in which they reported that children met criteria for a mood episode, on average, for two thirds of the follow-up period.
Effect on Child and Family
Mood disorders can affect several aspects of a child’s development, including social and academic development. Children accomplish many developmental tasks through normal interactions with their environment, such as interpersonal relationships and academic tasks. Mood disorders interrupt these normal interactions. For example, social interaction may be disrupted when a child experiences social withdrawal or is rejected by peers because of unusual behaviors. Similarly, a child may miss considerable instructional time in the classroom because of impaired concentration or behavioral problems related to irritability or disruptive manic behaviors. Once a child has fallen behind in social or academic development, catching up can be extremely difficult. When children experience long mood episodes and frequent relapses, the effect on development is dramatic. As one example, U.S. Department of Education statistics indicate that only 29% of children with an “emotional disturbance” graduated with a standard high school diploma in 2001, whereas 65% dropped out, in comparison to an 86% high school completion rate for all students in the same year.12,13 Teachers rate children with depression as having more withdrawn and disruptive social behaviors than do nondepressed peers.14 Adolescents with MDD also experience impairments in social, family, and academic functioning.15 In comparison with children with attention-deficit/hyperactivity disorder (ADHD), children with bipolar disorder have been found to have impaired relationships with parents and peers, higher rates of placements in special education classes, and higher rates of hospitalization.16,17
A child’s mood disorder can affect the entire family. Symptoms of irritability and mood lability can increase conflict in the child’s family interactions. Parenting stress also increases because parents are faced with a child’s mood and behavior problems that do not respond to typical parenting strategies.18 Besides the stress associated with managing the child’s symptoms, families can also suffer from the financial burden associated with the cost of medication and other treatments and lost time from work for doctor appointments. With more severe cases of mood disorders, multiple hospitalizations and legal difficulties related to the child’s behavior further disrupt family life.
Effect on Society
Beyond the direct effect of these conditions on the child and family, there is also a significant effect on society in terms of both human and financial costs. These costs are difficult to estimate, because they include the amount of money invested in treatment and educational services, reduced productivity, lost employment, mortality, and juvenile justice services.19 Ringel and Sturm3 estimated a national annual expenditure of $11.68 billion on child mental health services (inpatient treatment, outpatient treatment, and psychotropic medications) in the United States, with average costs of $293 per adolescent, $163 per child, and $35 per preschooler. In addition, in 2001, more than 475,000 students aged 6 to 21 received educational services for emotional disturbance under the Individuals with Disabilities Education Act (IDEA); this population constituted 8.1% of all students served under IDEA.13 The risk of death or physical injury related to mood disorders is also substantial. Approximately 2000 adolescents in the United States die from suicide every year, many of whom suffer from mood disorders, and an additional 700,000 require medical attention after a suicide attempt.20 According to World Health Organization estimates, MDD is the first and bipolar disorder is the fifth leading cause of years of living with a disability among 15- to 44-year-olds worldwide.19
CAUSES
Genetics
There is considerable evidence for the heritability of mood disorders in adult populations, with bipolar disorder more strongly influenced by genetics than is unipolar depression. Meta-analyses of adult studies have attributed approximately 60% of the variance in bipolar disorder and 37% of the variance in MDD to genetic factors.21,22 More recent research, using family and twin studies, has focused on the genetic influences on child and adolescent mood disorders. There is some evidence that earlier onset of a mood disorder is associated with increased prevalence of mood disorders in family members in comparison with later onset, which suggests that earlier onset may signify a more substantial genetic basis.21
Studies of the offspring of depressed parents have clearly demonstrated a familial association in childhood and adolescent depression, which could be the result of genetic influences, parent-child interactions, or other environmental influences. Having a parent with depression is one of the strongest predictors of depression in childhood and adolescence.5 Several twin studies have been conducted to explore heritability; results have varied widely, depending on measurement strategy, informant, age, and gender. Heritability estimates of parent-rated depressive symptoms range from 30% to 80%.23 Twin studies have been based on questionnaire reports of depressive or more general internalizing symptoms, and further research is needed with clinical interviews to establish diagnosis.23
Studies of parents with bipolar disorder have indicated that their offspring are at increased risk for mood disorders in general and bipolar disorder specifically. Among children of parents with bipolar disorder in a meta-analysis, 52% developed some type of mental disorder (2.7 times the risk in comparison with parents without bipolar disorder), 26.5% developed a mood disorder (4 times the risk in comparison with parents without bipolar disorder), and 5.4% developed bipolar disorder (in comparison with none of the control group).24 Studies have consistently demonstrated that for children with bipolar disorder, the rates of bipolar disorder in family members are higher, and younger age at onset is related to stronger family statistical loading of bipolar disorder.7,21 Children with psychotic depression also tend to have a family history of bipolar disorder and have a higher chance of going on to develop bipolar disorder.7 Twin studies of childhood-onset bipolar disorder have yet to be conducted, but the evidence available from family studies suggests that early-onset bipolar disorder may have a particularly strong genetic basis, and young patients may be good candidates for molecular genetic studies.21,25 Investigators are beginning to explore the molecular genetics of childhood-onset bipolar disorder, but consistent findings have yet to emerge.7,21
Biological Factors
Research in adults has identified several neurobiological correlates of mood disorders, including abnormalities in basal cortisol, cortisol regulation, corticotropin-releasing hormone, thyroid hormones, growth hormone regulation, and electroencephalographic sleep measures. Research in children and adolescents has been relatively sparse and has inconsistently replicated adult patterns.26,27
A review by Kaufman and colleagues26 indicates that consistency among child, adolescent, and adult studies has been found only in response to the dexamethasone suppression test and to selective serotonin reuptake inhibitors (SSRIs). Across the lifespan, patients with depression demonstrate nonsuppression of cortisol after the dexamethasone suppression test, which is suggestive of dysregulation of the body’s stress response system. There is also evidence that children and adolescents with depression respond to some SSRI medications in similar ways as do adults, which is discussed in more detail later in this chapter. However, responses to serotonergic probes in children have generally opposed findings in the adult literature, which indicates that there may be developmental differences in the dysregulation of the serotonergic system.26
Other neurobiological studies have yielded inconsistent findings. Studies in children with depression indicate that they show blunted response to agents that trigger growth hormone release, which is similar to adults’ responses; however, results have not been as consistent in adolescents.26 Blunted responsiveness to growth hormone has also been found in nondepressed children who are at increased risk for depression because of family history; this finding indicates that this response may reflect a predisposition to depression.5 Sleep studies have shown that adolescents with depression may demonstrate some electroencephalographic sleep responses similar to those of adults with depression, including reduced rapid-eye-movement latency and increased rapid-eye-movement density; however, these patterns have typically not been found in children.26 In contrast to the adult literature, abnormalities in basal levels of thyroid hormones, basal cortisol levels, and corticotropin-releasing hormone have not been consistently observed in children and adolescents.26
Investigators have only begun to examine the brain anatomy and functioning of children and adolescents with mood disorders; therefore, many results are preliminary. Neuroimaging studies with adults can be confounded by long duration of illness and the effects of treatment. Studies with children hold particular promise for identifying brain regions associated with the pathogenesis of mood disorders.28
The prefrontal cortex is influential in mood regulation and has been the focus of much of the neuroimaging research in adult depressive disorders.5,29 A growing series of child and adolescent studies have also focused on this area. One study revealed patients with MDD who had no family history of mood disorder had larger prefrontal cortical volume than did control patients and patients with MDD who did have a positive family history of mood disorder.28 In another study, glutamatergic concentrations in the anterior cingulate cortex were shown to be decreased by approximately 19% in patients with MDD in comparison with matched controls.30,31 In addition, significant increases in choline compounds have been found in the left dorsolateral prefrontal cortex of child and adolescent patients with MDD.32,33 Together, these studies reveal anatomical and biochemical anomalies in the prefrontal cortex in childhood and adolescent MDD. Furthermore, within a group of patients with MDD, Ehilich and colleagues34 found that white matter hyperintensities were associated with history of suicide attempts. Replication and extension of these findings are necessary to establish a clearer understanding of the role of the prefrontal cortex in childhood mood disorders.
In at least 11 studies, children and adolescents with bipolar disorder have been studied with magnetic resonance imaging.35 As reviewed by Frazier and associates,35 these studies collectively show that early-onset bipolar disorder is associated with a variety of functional, anatomical, and biochemical abnormalities in brain regions associated with emotional regulation and processing, including the limbic-thalamic-prefrontal circuit and the limbic-striatal-pallidal-thalamic circuit.
Environmental Factors
The strong familial association in early-onset mood disorders, discussed earlier, probably reflects a combination of genetic influences and family environment influences. Parental mood disorders can affect parent-child interaction, as well as events in the home. In comparison with control families, parent-child interaction in families with depressed children is characterized by higher levels of criticism, less warmth, more conflict, and poorer communication.15,36–39 Research on expressed emotion has suggested that a low level of parental criticism is predictive of recovery from depressive symptoms, whereas a high level of criticism is associated with the persistence of the mood disorder.40 The depressed child’s behavior also plays a role in evoking more negative interactions from parents.41 Disruptions in the family environment, such as marital discord, abuse, and poor support, can also affect parent-child interaction and the child’s risk for depressive symptoms.4,42 Furthermore, life stressors in general have been found to precede and exacerbate depressive symptoms.5
The mechanisms by which environmental factors are associated with depressive symptoms have been the focus of more recent research. Diathesis-stress models suggest that depression results from an interaction between an internal predisposition and environmental stressors. This predisposition can take the form of a genetic or biological tendency or may be related to cognitive factors, such as poor coping skills or depressive cognitive style. According to the review by Hammen and Rudolph,5 several investigators have tested the diathesis-stress model as it pertains to cognitive vulnerabilities and have found significant interactions between cognitive style and stressful events.
The association between family or psychosocial factors and depressive disorders has been clearly demonstrated across the lifespan.4,5 In a few studies, Researchers have also begun to examine those factors in childhood bipolar disorder. In Geller and colleagues’ long-term follow-up of children with bipolar disorder, children experiencing poor maternal warmth were about four times more likely to suffer relapse after recovery than were children with high maternal warmth.11,43 Children and adolescents with bipolar disorder have also been found to experience more life stressors than children with ADHD and children with no psychiatric diagnosis.44 These stressors included those clearly caused by the child’s symptoms or behavior (e.g., hospitalization), those possibly related to the child’s symptoms or behavior (e.g., removal from the home), and those unrelated to the child’s symptoms or behavior (e.g., death of a parent).
DIAGNOSIS
Accurate diagnosis of mood disorders in children is important, because underdiagnosis and misdiagnosis can lead to delays in the delivery of appropriate treatment or to the selection of treatments that may be harmful.45 For example, some research findings suggest that the use of SSRIs or stimulants can induce mania in children and adolescents with bipolar disorder; although not all researchers have replicated these findings.7,46
Barriers to Identification
The rate of recognition and treatment for children with psychological disorders in general is quite low. Data suggest that approximately 20% to 50% of children with a psychological problem are identified and only a portion of those cases are referred for evaluation and treatment.1,47,48 Internalizing problems, such as mood disorders, are identified much less frequently than are externalizing problems and are more likely to be identified when they are accompanied by a comorbid externalizing condition.49 In view of the negative effects of mood disorders on the child’s school, social, and family functioning, as well as the suicidal behaviors that can accompany mood disorders, the low rate of identification and treatment is a significant concern. Because of low rates of identification and treatment, outcomes in community settings have not kept pace with advances in the development of effective treatments for mood disorders in children and adolescents.50
Primary care pediatricians are in an ideal position to identify mental health conditions in general and mood disorders in particular. Most children have at least one primary care visit per year, and children with psychological problems are likely to have more frequent visits.49 Many children, particularly those from families of low socioeconomic status, receive care from only a primary care pediatrician.1 Several factors can prevent accurate identification in a primary care setting, including pediatricians’ limited training in mental health issues, parents’ failure to report mental health concerns without direct questioning, limited time available for screening of nonsomatic concerns, limited referral resources, and limited availability and use of screening instruments.1 Pediatricians can play a very important role in improving identification of mood disorders but may require education and resources to support them in this role.1,49 The development of screening instruments and the availability of onsite support and treatment options are promising strategies for removing these barriers to identification and treatment.1,2,50
Symptom Manifestation
The same criteria used for adults are used to diagnose childhood mood disorders. There is consensus that childhood depressive disorders have the same clinical features as the adult form of the disorders, with a couple of differences as outlined in the Diagnostic and Statistical Manual of Mental Disorders, 4th edition, text revision (DSM-IV-TR)51 (i.e., irritability can take the place of depressed mood; considering failure to make expected weight gains; 1-rather than 2-year duration for dysthymia). The child’s cognitive and emotional development can affect symptom manifestation and profile over time, but the clinical features of depressive disorders remain fairly consistent over time.52
There is controversy, however, over the definition of bipolar disorder in children. The core symptoms necessary for diagnosis, the necessity of discrete episodes, and the definitions of cycling in children all continue to be points of debate in the literature, and definitions of bipolar disorder have varied across studies.7 DSM-IV-TR diagnostic criteria for a manic episode include “a distinct period of persistently elevated, expansive, or irritable mood.”51 However, because irritability is so pervasive across childhood disorders, some investigators have required hallmark criteria of expansive/elated mood or grandiosity to help distinguish mania in children. In addition, many children meet symptom criteria for mania, with the exception of the duration criterion. These children may have intense rapid mood swings and often receive a diagnosis of Bipolar Disorder Not Otherwise Specified.
To help clarify the various conceptualizations of childhood mania, Leibenluft and colleagues53 proposed definitions for narrow, intermediate, and broad phenotypes of mania. In the most narrow phenotype, children meet strict DSM-IV-TR criteria for mania or hypomania, with the hallmark symptoms of elevated mood and/or grandiosity, and meet full duration criteria. Two intermediate phenotypes were identified: mania not otherwise specified (hallmark symptoms present, but symptoms do not meet duration criteria) and irritable mania (irritability without hallmark symptoms; full duration criteria met). The broad phenotype includes symptoms of severe mood and behavioral dysregulation without the hallmark symptoms or episodic cycling. Further research is needed to determine how these various phenotypes are related and whether there are differences in terms of etiology, treatment, and prognosis among the types.
Diagnosis of mood disorders in children and adolescents can evolve and change over time. Seventy percent of children with dysthymic disorder eventually experience a major depressive episode, and 20% to 40% of children who initially present with MDD eventually experience a manic episode.4 It is important to monitor the progression of symptoms over time to ensure that appropriate treatment strategies are used. Children who have early-onset depression, depression with psychotic features or psychomotor retardation, a family history of bipolar disorder, or a very strong family history of any mood disorders are at increased risk for developing mania.4
The clinical symptoms of mood disorders in children are often different from those typically seen in adults. Depressed affect, low self-esteem, and somatic complaints are more common in children than in adults, whereas anhedonia, diurnal variation, hopelessness, psychomotor retardation, and delusions increase with age.54 Children with bipolar disorder are more likely than adults to display continuous cycling (<365 cycles per year), mixed episodes, irritable mood swings with an insidious and chronic course beginning in early childhood, and high rates of comorbidity.16,45,55
It is important to consider developmentally relevant symptom manifestations in diagnosis in children. Cognitive maturation influences the ways children experience and express emotion.56 Children may be less able to express symptoms such as hopelessness, which require abstract thought until around puberty, when children begin to develop more abstract cognitive abilities.56 As another example, during their early school years, children become cognitively capable of comparing and evaluating themselves with regard to others; thus, the symptom of low self-esteem is more relevant than in younger years.56 In diagnosing childhood mania, it is particularly important to consider how DSM-IV-TR symptoms such as grandiosity, “increase in goal-directed activity,” and “excessive involvement in pleasurable activities” may manifest at different ages and how they differ from typical childhood behaviors. Children have certain constraints on their behavior by virtue of being monitored by adults, being required to attend school and other activities regularly, and not having resources such as credit cards or independent transportation to engage in the types of behaviors that adults may display.
BIPOLAR DISORDER: CASE ILLUSTRATION
Pretend play is normal in childhood, but the child who describes elaborate scenarios and cannot readily identify the play as pretend may be experiencing grandiosity, particularly if the play becomes inappropriate for the situation and impairs functioning.57 In David’s case, grandiose thoughts that are causing interference can be seen in his elaborate plans to run a restaurant to make money. He also provides several examples of increased goal-directed activity. His intense focus on making plans for his restaurant and creation of string webs that fill his bedroom reflect his increased energy and goal-directed activity. Children are not able to display typical adult pleasurable activities, such as spending sprees, but may instead display excessive “daredevil” behaviors without considering the dangerous consequences, or they may become hypersexual; David exhibits both behaviors.
Normal Variations in Mood
All children undergo development in their understanding of and ability to regulate emotions. Children begin to understand and use basic emotional terms around ages 2 to 3; however, the understanding of more complex emotions and mixed emotions continues to develop through childhood. Toddlers and preschool-aged children tend to display tantrums in response to frustration but become better able to regulate their emotional expression by the time of school entry. Children begin to regulate their subjective feeling of negative emotions first through problem-focused coping strategies and later become able to use emotion-focused coping to tolerate situations that they cannot change.58 During puberty, hormone changes can lead to an increase in mood lability. There are also temperamental differences among children in general levels of reactivity. These normal developmental processes and individual differences should be considered in the evaluation of mood disorders.
Response to Bereavement
When a child or adolescent has experienced a loss, such as the death of a loved one or pet, a family move, or a broken relationship, symptoms of depression are common. Children’s reactions to the death of a loved one can vary and may include dysphoria, crying spells, clinging to familiar routines and caregivers, impairments in school functioning, behavior problems, bedwetting, loss of interest in activities, sleep problems, and psychosomatic symptoms.59 These symptoms are usually transient and can be differentiated from a mood disorder on the basis of the duration and associated impairment.60 According to DSM-IV-TR criteria, symptoms of bereavement should not be diagnosed as a mood disorder unless they last longer than 2 months or are associated with significant functional impairment, worthlessness, suicidal ideation, psychotic symptoms, or psychomotor retardation.
In children who have experienced parental death, grief or sadness lasting a year or more is common.61 In the largest prospective study to date of children after parental death, Cerel and colleagues62 found that bereaved children demonstrated more impairment over a 2-year period than did a control group but less impairment than did a comparison group of nonbereaved children with a diagnosis of a depressive disorder. Overall impairment and depressive symptoms improved significantly in the bereaved group over 2 years, and this improvement was more rapid than that seen in the depressed group. Level of impairment and coping skills should be carefully monitored after parental death, because up to 20% of children display symptoms serious enough to warrant specialized treatment.61 A child is more likely to display clinical levels of disturbance after parental death when he or she had a psychiatric disorder before the death, when the surviving parent displays high levels of depression before or after the death, or when the family has fewer socioeconomic resources.61,62 The presence of multiple stressors in the child’s life is associated with slower improvement in depressive symptoms and overall impairment.62 Parental death by suicide is also related to higher levels of overall psychopathology.63 When children witness elements of a traumatic death, such as parental murder or suicide, they are at risk for posttraumatic stress disorder.61
After the death of a parent, many children experience suicidal ideation; however they are less likely to attempt suicide than are children with depressive disorder.64 The suidical ideation expressed by bereaved children more often reflects a desire to be with the deceased parent rather than a wish to end their lives.61,64 As with all reports of suicidal ideation, bereaved children who report such thoughts should be carefully assessed for risk factors, development of a suicide plan, and access to means of harming themselves.
ASSESSMENT
As with all childhood mental health concerns, a thorough assessment is necessary to establish an accurate diagnosis. Mood problems can reflect underlying medical conditions or drug reactions; therefore, medical causes for symptoms should be explored and ruled out as part of the assessment process.45 Gathering data from multiple informants is important, because agreement between parent and child report is often low.65 Children are generally better reporters of internal mood states, whereas parents tend to be more accurate in reporting behavioral symptoms and symptom history, although exceptions to this generalization are readily found in clinical settings.66 Teachers also have a unique perspective, inasmuch as they see children in a structured setting where behavior can differ from home and have experience with a same-age comparison group. If possible, data from teachers, such as questionnaires, notes, and report cards, should also be integrated into the clinical assessment process.
Differential Diagnosis
Cross-sectionally, the symptoms of mood disorders can appear similar to other childhood disorders and may be misdiagnosed as ADHD, anxiety disorders, developmental disorders, or behavior disorders. Furthermore, differential diagnosis among the mood disorders can present a challenge when only current symptoms are considered. Examination of the evolution of symptoms over time can help establish the presence of episodic mood changes and whether other symptoms fluctuate with mood. For example, symptoms of social withdrawal and self-doubt may reflect depression or social anxiety, which can be difficult to distinguish with a cross-sectional assessment. A history of social anxiety preceding the development of other symptoms of depression would signify the presence of an anxiety disorder. If the symptoms developed at the same time or became significantly worse along with other mood symptoms, then they could be considered symptoms of depression.
Irritability is frequently a symptom of childhood mood disorders, but it can also be prominent in ADHD, behavior disorders, anxiety disorders, and pervasive developmental disorders, as well as in children without psychopathology who are hot, hungry, tired, or stressed.45 Manic irritability often can be distinguished by its episodic, intense, and prolonged nature.
Children with ADHD also tend to experience difficulties with emotional regulation related to general impairments in behavioral inhibition, which can lead to quick expressions of emotional reactions that change easily.67 Children with mood disorders also have difficulty with emotional regulation; however, they can be differentiated by their intensity, duration, associated symptoms, and environmental triggers. Children with ADHD typically experience emotional overarousal in response to environmental disorganization and overstimulation. Children with depressive disorders, on the other hand, experience depression or irritability as their predominant mood state, and their mood does not change as much in response to environmental triggers. Bipolar disorder can be distinguished from the emotional overarousal of ADHD by the episodic, intense, and prolonged nature of emotional reactions, which are accompanied by associated symptoms not typically seen in ADHD.
Because of the symptom overlap between mania and ADHD, differential diagnosis can be particularly difficult. Distractibility, rapid speech, and increased energy are symptoms of mania that overlap with those of ADHD. Geller and colleagues55 identified five symptoms of mania that provide the best discrimination between mania and ADHD: elated mood, grandiosity, racing thoughts, decreased need for sleep, and hypersexuality. In assessment for bipolar disorder, the clinician should pay careful attention to these distinguishing symptoms, as well as symptom fluctuation with mood changes over time. In children with comorbid ADHD, distractibility, rapid speech, and increased energy should be considered symptoms of mania only when they increase beyond the child’s unique baseline level as his or her mood changes.
Hypersexual symptoms, such as those demonstrated by the example of David previously, are common in childhood-onset bipolar disorder. In Geller and colleagues’ sample,55 43% of children displayed hypersexuality. Children who display hypersexual behaviors should be carefully assessed for evidence of sexual abuse or exposure to sexual content inappropriate for the child’s age.68 Sexual behavior with a pleasure-seeking quality that fluctuates with other mood symptoms may be a symptom of mania.69
Psychosis is common in both MDD and bipolar disorder in childhood and may be incorrectly diagnosed as a schizophrenia spectrum disorder.5,70 Psychotic symptoms that are congruent with mood and fluctuate with mood symptom severity are more likely to be an associated symptom of the mood disorder.70
Mood Symptoms
Specific information should be gathered during the clinical interview about severity of mood symptoms. Mood symptoms should be evaluated in the context of an understanding of normal variations in children’s mood. Frequency, intensity, number, and duration (FIND) guidelines can be used to assist in establishing the presence or absence of mood symptoms45: Individual symptoms should fluctuate with mood and occur most days of the week (frequency), at a level that causes impairment (intensity), several times per day (number), and should last a significant portion of the day (duration).45
Prospective mood charting can be helpful in making a diagnosis, as well as monitoring progress. Daily mood logs completed by parents or adolescents can provide valuable information about situational variables that trigger mood symptoms and response to treatment. Examples of mood logs can be found at www.bpkids.org/site/PageServer?pagename=lrn_mood or can be individually tailored to meet the needs of a particular child and family.
Clinician-rated mood scales can also be helpful in summarizing mood symptom severity and in tracking progress. The Children’s Depression Rating Scale-Revised has been shown to be a reliable, valid, and sensitive measure of depressive symptoms in both inpatient and outpatient samples.71 The Young Mania Rating Scale, which was developed for adult populations, has been shown to have acceptable reliability and validity in child samples.72,73 Although widely used, this scale has several limitations, including a lack of published developmentally appropriate anchor criteria for interview-based ratings.69 The Kiddie Schedule for Affective Disorders and Schizophrenia (KSADS) Mania Rating Scale and Depression Rating Scale are promising new instruments developed for child and adolescent populations and are based on DSM-IV-TR criteria.74 Preliminary studies have found that these instruments have good psychometric properties, and further validation with larger samples is under way.69,74
Family History
Information about family history can further help establish the probability of mood disorder.69 Although a family history of mood disorders is not diagnostic of mood disorders per se, it does add additional information about the child’s risk for the disorder.69 Furthermore, because data suggest that children with MDD who have a family history of bipolar disorder are at increased risk for developing a manic episode in the future, this information may guide treatment decisions and follow-up strategies.4
Structured Interviews
It is important to assess for behavior, anxiety, mood, and other symptoms as part of the clinical interview. Structured or semistructured interviews can be used to systematically gather information about various childhood problems. Several options have demonstrated sensitivity and specificity in identifying relevant conditions and require varying levels of training and time to administer. Examples include the Diagnostic Interview Schedule for Children,75 Children’s Interview for Psychiatric Symptoms (ChIPS),76,77 Diagnostic Interview for Children and Adolescents,78 and the Washington University at St. Louis KSADS (WASH-U-KSADS),79 which includes an expanded section on the diagnosis of manic symptoms. Variations of the KSADS, such as the WASH-U-KSADS, are most commonly used in research settings, but require extensive time and specialized training for administration, which makes this instrument impractical for clinical use.69 Symptoms of mania are often not thoroughly evaluated in developmentally appropriate terms in the briefer structured interviews; however the ChIPS shows the most promise for identifying manic symptoms in youth.69
Questionnaires
Diagnosis of mood disorders can never be made on the basis of questionnaires alone. However, questionnaires can be useful as screening instruments to guide a clinical interview or as another source of information to integrate with interview data. The Child Behavior Checklist (CBCL) is a norm-referenced and widely used instrument in clinical practice and research to assess a variety of behavior problems in children and adolescents.80 The CBCL behavior scales are not specific enough to differentiate depressive from anxiety disorders; however, high scores on the internalizing scale signal that additional information should be gathered about specific mood and anxiety disorders. The Children’s Depression Inventory81 is a self-report measure specifically designed to assess the severity of depressive symptoms. This inventory has been shown to differentiate between psychiatric patients and control subjects, but it does not differentiate well among psychiatric diagnoses.82
Low scores on the externalizing scales of the CBCL are useful in ruling out bipolar disorder, but high scores are not specific enough to draw conclusions about the presence of bipolar disorder.9 The General Behavior Inventory83 is a questionnaire that is used specifically to assess manic symptoms. Parent and youth versions of this inventory have demonstrated excellent psychometric properties; however, the complexity of many items may make it difficult for individuals with limited education or reading abilities.69 Data suggest that youth and teacher questionnaires do not add anything beyond parent questionnaire data in the prediction of bipolar disorder diagnosis.9
The Pediatric Symptom Checklist84 has been developed specifically to screen for a variety of mental health problems in primary care settings. It is brief, has empirically derived cutoff scores, and has been validated with racially diverse populations and populations of low socioeconomic status.1 In settings in which resources are available to score and interpret the CBCL, it may be administered before the clinician meets with the family and used to help guide the interview. When a mood disorder is suspected, the Children’s Depression Inventory or General Behavior Inventory may be useful in the decision of whether to refer for a more thorough evaluation.
ASSOCIATED CONDITIONS
Comorbidity
Comorbidity is common among childhood diagnoses. Approximately 40% to 70% of children and adolescents with depression have at least one other psychiatric condition.4 A meta-analysis by Angold and colleagues85 found that depression is most closely associated with anxiety disorders (odds ratio, 8.2), followed closely by conduct disorders (odds ratio, 6.6) and ADHD (odds ratio, 5.5). Substance abuse disorders are also commonly comorbid with depression and tend to begin an average of 4.5 years after the onset of the depressive disorder.4
High rates of comorbidity have also been found with early-onset bipolar disorder, particularly with ADHD, behavior disorders, and anxiety disorders. Rates of comorbidity range from 66% to 75% for ADHD, 46% to 75% for oppositional defiant disorder, 5.6% to 37% for conduct disorder, 12.5% to 56% for anxiety disorders, and 11% for pervasive developmental disorders.7,69 The rates of comorbid substance abuse disorders increase with age, with rates up to 40% in adolescents.7
Treatment of Comorbidity
In treating conditions comorbid with bipolar disorder, it is important to first stabilize the mood symptoms and then evaluate the need for psychosocial or pharmacological treatment of any comorbid conditions.45 There are no clear guidelines for the treatment of comorbid conditions with depression. If the mood disorder appears to be secondary to another condition, such as social anxiety or posttraumatic stress disorder, it may be useful to treat the primary condition first or concurrently with the treatment for depression.
Suidical Ideation
Children with mood disorders are at increased risk for suicidal ideation, attempt, and completion. Suicidal ideation has been reported in more than 60% of depressed children and adolescents, and MDD is the most common diagnosis among suicide victims.5,86 Children with bipolar disorder are also at high risk for suicide, particularly when depressed, during a mixed episode, or when psychotic.8,55,87 Geller and colleagues55 reported suicidal ideation in 25% of their 7- to 16-year old participants with bipolar disorder. Comorbidity between mood disorders and substance abuse or disruptive behavior disorders further increases the risk of suicide.86 These data highlight the importance of assessing suicidality in youths with mood disorders. Data from a randomized controlled trial indicate that assessing for suicidal ideation does not increase distress or suicidal ideation in adolescents.88
In assessing for suicide risk, several factors should be taken into account; these are outlined in the American Academy of Child and Adolescent Psychiatry’s 2001 Practice Parameters regarding the assessment and treatment of adolescent suicidal behavior.20 In addition to the presence of a mood and/or substance abuse disorder, individuals with previous suicide attempts, suicidal thoughts, plans for suicide, agitation, and psychosis are at greatest risk for suicide.20 Other risk factors include family history of suicide; history of physical or sexual abuse; school problems; poor communication with parents; recent suicide of a peer; and gay, lesbian, or bisexual orientation.86 Several questionnaires that have been developed to assess risk of suicide have high sensitivity but poor specificity because of the low base rates of suicide.86 These questionnaires can best be used as screening tools in community samples. In children and adolescents at high risk for suicide, such as those with mood disorders, assessment should include direct interview with the child and parent.20
Psychosis
Psychosis is more common in child and adolescent mood disorders than in adult mood disorders. Approximately 33% to 50% of preadolescents with MDD and up to 31% of adolescents with MDD experience hallucinations, most commonly auditory hallucinations.5 Estimates of the rates of psychosis in early-onset bipolar disorder range from 16% to 88%, depending on assessment strategy.70 The most common type of psychotic symptom reported in early-onset bipolar disorder is mood-congruent grandiose delusions.70
TREATMENT
As evidenced by the prevalence, chronicity, and impairment associated with mood disorders described previously, effective intervention strategies are needed to manage these conditions. Treatment outcome studies for both biological and psychosocial therapies have helped informed treatment decisions for children with mood disorders. Considerable research on treatments for MDD has accumulated.89,90 Much of the treatment research on childhood bipolar disorder, in contrast, is preliminary and is currently evolving.45
Depression
BIOLOGICAL INTERVENTIONS
Tricyclic antidepressants have not been found to be effective in the treatment of children and adolescents.26 Only one SSRI, fluoxetine, has received approval from the U.S. Food and Drug Administration (FDA) to be marketed for children and adolescents.
Cheung and associates91 reviewed the efficacy and safety of published and unpublished randomized controlled trials of antidepressants in children and adolescents. Throughout the studies reviewed, various outcome measures were used; however, a Clinical Global Impression Improvement (CGI-I) rating of 1 or 2 (very much improved or much improved) was the most frequent definition of response to treatment that produced significant results and these response rates are reported as follows: (1) Three large double-blind placebo-controlled trials of fluoxetine indicated significant differences in clinician-rated response and symptom level between fluoxetine and placebo across all three studies; response rates for fluoxetine ranged from 53% to 60%, in comparison with 33% to 37% rates for placebo.90,92,93 (2) Of three double-blind, placebo-controlled studies of paroxetine, the one published study demonstrated superiority of paroxetine over placebo (66% response to paroxetine, 48% to placebo), whereas the other two adequately powered but unpublished studies failed to demonstrate significant results.94 (3) Two studies of sertraline, combined a priori for analysis, were identified. Response rates were 69% for the sertraline recipients and 59% for the placebo recipients. The difference was statistically significant because of the large number of subjects; however, neither study produced significant results when analyzed separately.95 (4) Two studies of citalopram were reviewed, one published and one unpublished, with an unusual pattern of results. Neither study revealed differences in CGI-I response rates; however, one did reveal group differences in depression symptom severity, as rated by the Children’s Depression Rating Scale-Revised.96 The meaning of this finding without CGI-I response differences is unclear. (5) Two studies of nefazodone were identified; one revealed a significant effect of the drug (65% response rate in comparison with 46% response rate with placebo), whereas the other study revealed no significant effect.97 (6) In two studies of venlafaxine and two studies of mirtazapine, no differences were found between the drugs and placebo on any measure.98 Additional details about the methods and results of all these studies were described by Cheung and associates.91
In October 2004, the FDA issued a black box warning requiring that antidepressant medications be accompanied by information indicating that antidepressant use is associated with increased risk of suicidality in children and adolescents. This warning was based on a review of 26 studies that demonstrated that the average risk of suicide-related events was 4% with antidepressants, in comparison with 2% with placebo.99 No deaths by suicide were reported in any of the studies reviewed.99 Examining the risk of suicidal behavior associated with naturalistic antidepressant use in the United Kingdom, Jick and colleagues100 found that risk was increased in the first month after initiation of antidepressant therapy and was highest in the first 1 to 9 days. Concerns have also been raised concerning children’s and adolescents’ risk of becoming agitated or switching to mania with antidepressant medications, which was found to occur in very small numbers of patients participating in randomized controlled trials.91 The Society for Adolescent Medicine emphasizes the high risk of suicide associated with untreated depression, however, and supports the continued use of antidepressant medication in adolescents along with careful monitoring, particularly at the beginning of treatment and after dose changes.99
Herbal remedies are gaining popularity, and St. John’s wort has been shown to have antidepressant effects superior to those of placebo in mild to moderate adult depression.101 Open-label pilot studies in children and adolescents have indicated that St. John’s wort is well tolerated and may be beneficial in treating MDD in youth.101,102 Randomized clinical trials are needed to further evaluate the safety and efficacy of St. John’s wort in children and adolescents.
Among adults who experience seasonal variation in mood symptoms, exposure to bright light or to dawn simulation has been found beneficial. Research extending these findings to pediatric samples has shown that light therapy is effective and superior to placebo in children and adolescents.103
Electroconvulsive therapy has been shown to be a very effective treatment for severe depression in adults, with remission rates of 70% to 90% in clinical trials.104 Since 1990, several studies of the use of electroconvulsive therapy in adolescents with a variety of diagnoses have been published.105 Response rates range from 50% to 100%, with higher response rates reported for mood disorders. In addition, high rates of satisfaction with the treatment have been reported among adolescents who received electroconvulsive therapy.105 There are not enough data on the use of electroconvulsive therapy in preadolescents with which to draw conclusions about its efficacy. The American Academy of Child and Adolescent Psychiatry practice parameters advised that electroconvulsive therapy be considered for adolescents after previous interventions have been ineffective and if a second psychiatrist agrees to the appropriateness of the treatment.105 Overall, electroconvulsive therapy is rarely used in adolescents despite the efficacy data and American Academy of Child and Adolescent Psychiatry guidelines.106 Lack of both knowledge and experience with electroconvulsive therapy among child and adolescent psychiatrists and public controversy surrounding the treatment may contribute to the low rates of utilization.105
PSYCHOSOCIAL TREATMENT
The most extensively researched psychosocial treatment for depression in children and adolescents is cognitive-behavioral therapy (CBT). Compton and colleagues89 reviewed 12 randomized controlled trials of CBT for depression in children and adolescents. Overall these studies showed that CBT is superior to no treatment. One study found that CBT was superior to an attentional control group;107 however others have failed to find differences in comparison with nonfocused therapy, treatment as usual, or pill placebo.90,108,109 In comparison with other specific treatment modalities, CBT has been found to be superior to relaxation training110 and systemic behavior family therapy,107 but no differences were found with interpersonal therapy.111
The cognitive-behavioral model is based on the idea that depression is maintained by cognitions and behavioral patterns that decrease effective interaction with the world.89 Cognitive distortions are thought to bias the way an individual obtains and interprets information from the environment, leading to negative thoughts about one’s self, the world, and the future and to attribution of negative events to stable, internal, and global factors. Deficits in social skills and problem solving also prevent successful interactions with the environment. Further, decreased participation in potentially enjoyable activities decreases opportunities for pleasure. Specific components of CBT include psychoeducation, goal setting, problem solving, and tailored interventions based on the individual’s cognitive and behavioral patterns. As family conflict and disruption is quite common in child and adolescent depression, a parent or family component is often included in treatment. Two studies have specifically investigated the effect of adding a family component to individual therapy, and have failed to find group differences, however this may be due to insufficient power to detect incremental differences in treatment efficacy.112,113
Another psychosocial treatment that shows promise in the treatment of adolescents with depression is interpersonal psychotherapy (IPT). In randomized clinical trials, IPT has been shown to be more effective than control groups and has shown similar efficacy to CBT.111,114,115 IPT focuses treatment on patterns of interpersonal interaction and communication. Specific targets of treatment include the problem areas of grief, interpersonal disputes, role transitions, interpersonal deficits, and single-parent families.116
COMBINATION BIOLOGICAL AND PSYCHOSOCIAL TREATMENT
The Treatment of Adolescents with Depression (TADS) study90 is the first to systematically explore the relative efficacy of fluoxetine, CBT, and their combination in adolescent depression. Results suggest that the combination is most effective, followed by fluoxetine, which was superior to both CBT alone and placebo. The TADS study90 showed that combination pharmacotherapy and CBT are more effective in decreasing both depressive symptoms and suicidal behavior than either intervention alone in adolescents with depression, highlighting the importance of a combination of interventions in the treatment of childhood-onset mood disorders.
Bipolar Disorders
BIOLOGICAL INTERVENTIONS
Pharmacological interventions are an essential component of comprehensive treatment for bipolar disorder.116a The only medication approved by the FDA for adolescent bipolar disorder is lithium. Only five randomized placebo-controlled studies to date have evaluated medications for use in child and adolescent bipolar disorder.7 Lithium was shown to improve global functioning in adolescents with bipolar disorder and comorbid substance use disorders.117 The only other study of lithium monotherapy, however, did not find continuing effects of lithium in a randomized discontinuation study.118 Following mood stabilization with combination lithium and divalproex sodium, a comparison of the two medications indicated that they were equally effective in maintenance treatment.119 DelBello and colleagues120 found that the combination of quetiapine with divalproex sodium was more effective in treating adolescent mania than dovalproex sodium alone. In a study of children and adolescents with comorbid ADHD, the addition of Adderall following mood stabilization with divalproex sodium was found to be effective in treating the ADHD symptoms.121 Much of the data used in determining treatment options for bipolar disorder in children and adolescents come from open trials, adult studies, and clinical experience.45
Kowatch and colleagues45 developed treatment guidelines for children and adolescents with bipolar disorder based on the evidence currently available. For bipolar mania, with or without psychosis, various combinations of treatment with mood stabilizers and/or atypical antipsychotics were recommended. Following nonresponse to multiple medication trials, clozapine or electroconvulsive therapy was recommended. See Kowatch et al45 for detailed treatment algorithms. No algorithm was developed for the treatment of bipolar depression, as the currently available data are too limited to draw conclusions about the treatment of depressive symptoms.
Despite the limited efficacy data for child and adolescent bipolar disorder treatments, mood stabilizers and antipsychotics are commonly used for the treatment of early onset bipolar disorder in clinical settings. American Academy of Child and Adolescent Psychiatry practice parameters recommend considering the evidence from child and adult studies, symptom presentation, phase of illness, medication safety profile, child’s history of medication response, and family preference in choosing medication(s).116a The practice parameter also suggests that most youth with bipolar disorder will require ongoing medication treatment to prevent relapse.116a
PSYCHOSOCIAL TREATMENT
Medication is a critical component of treatment for bipolar disorder; however psychosocial interventions play an important role in promoting medication compliance and teaching skills to help decrease relapse.7,116a Outcome studies of medication treatment in children and adolescents are limited, as described above, however adult studies describe residual symptoms, poor outcomes for bipolar depression, and high medication nonadherence rates.122 Additionally, while the cause of bipolar disorder in children appears to be strongly influenced by biological factors, the course can be shaped by psychosocial factors.123 Adjunctive interventions are clearly needed to enhance treatment outcome and adult studies indicate family-based psychosocial interventions decrease relapse by 33%.124,125
The addition of psychotherapy to pharmacotherapy has been recommended as soon as the child’s mood is stable enough to learn new skills.45 Research to identify efficacious treatments for bipolar disorder is in the early stages of development and three research teams have developed and reported preliminary data on therapies for bipolar disorder. These therapies are all adjunctive, include family involvement, and are psychoeducationally focused.
Fristad and colleagues have conducted the only randomized controlled trials of a psychoeducational treatment for families of children with mood disorders to date. These studies included children age 8-11 with both depression and bipolar disorder. In a pilot study of 35 children, 46% of whom had bipolar disorder, Multi-Family Psychoeducation Groups (MFPG) were found to increase parental knowledge about mood disorders, increase positive family interactions, increase the parental support perceived by children, and increase utilization of appropriate services by families.126 MFPG involves parents and children meeting separately in a group format to receive education, support, and learn skills to cope with symptoms and improve the child’s functioning. Forming groups of children with both bipolar disorder and depressive disorder diagnoses had the benefits of making it easier and faster to recruit an adequate number of families to start a group. Due to the significant portion of children initially diagnosed with MDD who later develop bipolar disorder, it was also considered important to provide information on the symptoms and management of bipolar disorder to families of children with both diagnoses. No difficulties were found in conducting these combined groups, but it was considered beneficial to include at least two children with each type of diagnosis in a group. In the pilot study, families of children with bipolar disorder had worse mood symptoms, a history of more treatment experiences, and greater knowledge about mood disorders at the beginning of treatment than the families of children with depressive disorders; however families of children with bipolar disorder and depressive disorders both benefited from treatment.127 A larger randomized controlled trial of 165 children is currently underway to evaluate MFPG.
The content of MFPG has also been adapted for delivery in individual family sessions, which has been tested in a pilot study of 20 children with bipolar disorder. Results suggest treatment led to decreased severity of mood symptoms, with improvements continuing for 12 months following treatment, improved treatment utilization, more positive family climate, and high levels of satisfaction with treatment.128 Further research with a larger sample size will be needed to more clearly evaluate the efficacy of individual family psychoeducation.128
Family-Focused Therapy (FFT) in adults with bipolar disorder has been shown to delay relapse, decrease hospitalization, decrease symptom severity, and improve medication adherence.129–131 FFT involves patients and family members in education about bipolar disorder, communication training, and problem solving skills training. Miklowitz and colleagues132 have adapted FFT for an adolescent population (FFT-A), and in open trials the combination of FFT-A and pharmacotherapy was associated with improvements in symptoms of mania and depression and reductions in problem behaviors. Pavuluri and colleagues133 adapted the FFT model and combined it with cognitive behavioral principles to develop Child and Family-Focused Cognitive-Behavioral Therapy (CFF-CBT) for younger children. In open trials, CFF-CBT in addition to pharmacotherapy led to reductions in mood symptoms and improved global functioning.133
PREVENTION
Research has begun to address the prevention of depression in youth with subclinical symptoms. Group cognitive behavioral interventions for children and adolescents have shown promise in reducing symptoms and subsequent development of clinical depression; however data on the long term effects of these programs have been mixed.134–136 Gilham and colleagues134 found that benefits were sustained and the prevention effects grew over the course of two years, while Spence and colleagues136 found that initial benefits were lost by one year follow up. Intervention programs in primary care have also targeted adolescents at risk for depression, and findings suggest that prevention programs can be successfully implemented in such settings with the consultation of mental health professionals or use of internet-based programs.50,137
Given the often chronic and relapsing course of mood disorders, the prevention of relapse in children who have had a prior mood episode is an important consideration. One study suggests that the continuation of fluoxetine treatment for depression may help reduce and delay relapse in children and adolescents over eight months.138 Follow-up studies of CBT for depression have indicated that treatment gains are generally maintained or continue one to nine months following treatment, but over longer term follow-up (nine months to two years), lack of recovery and relapse are common.89 Further research is needed to determine whether booster sessions of CBT may be helpful in maintaining treatment effects.89 Research on pharmacological and psychosocial strategies for reducing and delaying relapse for children and adolescents with bipolar disorder is in the early stages and has not yet yielded any conclusive findings.7
CLINICAL IMPLICATIONS
Mood disorders in pediatric populations are associated with impairment at home, at school, and with peers. The experience of a mood disorder also increases risk for suicide and future mood problems. Identification of these disorders in children and adolescents is a crucial first step in reducing the associated impairment and risks. Developmental-behavioral pediatricians, as well as other health and educational professionals, play an important role in identifying mood disorders in this population and recommending appropriate treatments. A significant portion of mental health services are currently provided in primary care settings, with primary care visits accounting for nearly 40% of all mental health services among a small sample of privately insured children.3 The availability of child and adolescent psychiatry services nationwide is far short of the need.139 These data highlight the importance of all child and adolescent health care providers being educated about the symptoms and effective treatments for mood disorders. Further, only about half of children referred for services with a mental health provider show up for their first appointment with that provider.1 Families may be more likely to follow through with treatment when services are available in the primary care setting or there is significant collaboration among professionals.1,2,50
Models of collaboration among professionals to treat mental health problems in a primary care setting have been developed and show promise for providing cost-effective, beneficial services for children and adolescents. A stepped-care approach has been shown effective in treating adult mood disorders and has been adapted for the treatment of child mental health problems.2,50,140 Campo and colleagues2 described a program in which primary care physicians, advanced practice nurses, social workers, and pediatric psychiatrists work together to provide appropriate levels of care for children and adolescents. In their model, primary care physicians identify children with possible mental health issues and provide treatment for less complex cases. Advanced practice nurses with training in psychiatry complete onsite mental health assessments, make diagnoses, and provide patient education and support as needed. Social workers provide case management and onsite psychotherapy for cases of moderate to severe complexity. The pediatric psychiatrist manages more complex cases and provides consultation for the team. The team also met regularly to discuss ongoing cases. With this level of support, two-thirds of mental health cases were successfully managed by the primary care pediatrician and advanced practice nurse. Asarnow and colleagues50 reported on a similar program testing the benefits of having a care manager with mental health training available in the primary care setting to coordinate and support the care of adolescents with depression. Adolescents who received the collaborative care reported fewer depressive symptoms, greater utilization of mental health services, and greater satisfaction with their treatment than those who received usual care. These models highlight the ways that mental health screening, in-house treatment options, and consultation among professionals can improve outcomes for children and adolescents.
RESEARCH IMPLICATIONS
Further research in childhood mood disorders is needed to improve prevention, identification, diagnosis, and treatment efforts. The existing research base is considerably stronger for depressive disorders than for bipolar disorders. There is a growing base of knowledge about the risk factors involved in mood disorders, including genetics, neurobiology, and environmental influences. Future molecular genetics research will improve our understanding and early identification of at-risk children. It will also be important for future research to examine the ways genetics, neurobiological factors, and environmental influences interact, which may lead to more targeted intervention strategies.139 Research is necessary to establish a clear definition of bipolar disorder in children and determine its continuity with adult forms of the disorder. To accomplish this, developmentally appropriate criteria with high interrater reliability and validity will need to be developed.139
We now have a growing base of randomized controlled trials of treatments for depression to guide treatment decisions. Future research should expand on this knowledge by examining treatment options for treatment-resistant depression, identifying the active components of psychotherapy through dismantling studies, increasing attention to prevention strategies, and examining effectiveness in community-based studies.139 There is currently very little research base for the treatment of bipolar disorder and the identification of effective pharmacological and psychosocial treatments will be very important in improving outcomes for the children and families who are coping with this chronic and relapsing condition.139
Changes are also needed in the way mental health services are provided to increase identification and appropriate treatment. Programs to increase knowledge about mood disorders among professionals who work directly with children, such as teachers and primary care physicians, should be developed and evaluated. Recent research has begun evaluating ways of incorporating mental health treatment into primary care.2,50 Continued research and public policy changes to allow for more effective treatment will help improve access to appropriate services.139
1 Simonian SJ. Screening and identification in pediatric primary care. Behav Modif. 2006;30:114-131.
2 Campo JV, Shaver S, Strohm J, et al. Pediatric behavioral health in primary care: A collaborative approach. J Am Psychiatr Nurses Assoc. 2005;11:276-282.
3 Ringel JS, Sturm R. National estimates of mental health utilization and expenditures for children in 1998. J Behav Health Res. 2001;28:319-333.
4 Birmaher B, Ryan ND, Williamson DE, et al. Childhood and adolescent depression: A review of the past 10 years: Part 1. J Am Acad Child Adolesc Psychiatry. 1996;35:1427-1439.
5 Hammen C, Rudolph KD. Childhood mood disorders. In: Mash E, Barkley RA, editors. Child Psychopathology. 2nd ed. New York: Guilford; 2003:233-278.
6 Kashani JH, Holcomb WR, Orvashel H. Depression and depressive symptoms in preschool children from the general population. Am J Psychiatry. 1996;143:1138-1143.
7 Pavuluri MN, Birmaher B, Naylor MW. Pediatric bipolar disorder: A review of the past 10 years. J Am Acad Child Adolesc Psychiatry. 2005;44:846-871.
8 Lewinsohn PM, Seeley JR, Klein DN. Bipolar disorder in a community sample of older adolescents: Prevalence, phenomenology, comorbidity and course. J Am Acad Child Adolesc Psychiatry. 1995;34:454-463.
9 Youngstrom EA, Findling RL, Calabrese JR, et al. Comparing the diagnostic accuracy of six potential screening instruments for bipolar disorder in youths aged 5 to 17. J Am Acad Child Adolesc Psychiatry. 2004;43:847-858.
10 Weissman MM, Wolk S, Goldstein RB, et al. Depressed adolescents grown up. JAMA. 1999;281:1707-1713.
11 Geller B, Tillman R, Craney JL, et al. Four-year prospective outcome and natural history of mania in children with a prepubertal and early adolescent bipolar disorder phenotype. Arch Gen Psychiatry. 2004;61:459-467.
12 U.S. Department of Education. FY 2004 Performance and Accountability Report. Washington, DC: U.S. Department of Education, 2004.
13 Office of Special Education Programs: 25th Annual (2003) Report to Congress on the Implementation of the Individuals with Disabilities Education Act, vol 1. Washington, DC: U.S. Department of Education, 2005.
14 Rudolph KD, Clark AG. Cognitions of relationships in children with depressive and aggressive symptoms: Social-cognitive distortion or reality? J Abnorm Child Psychol. 2001;29:41-56.
