November 2012  << Back  

  Can't open the PDF? Click here for help.



Characteristics of children with juvenile bipolar disorder or disruptive behavior disorders and negative mood: Can they be distinguished in the clinical setting?

Daniel F. Connor, MD

Professor of Psychiatry, Division Chief of Child & Adolescent Psychiatry, University of Connecticut School of Medicine, Farmington, CT, USA

Leonard A. Doerfler, PhD

Professor of Psychology, Assumption College, Associate Professor of Psychiatry, University of Massachusetts Medical School, Worcester, MA, USA

BACKGROUND: Because of continuing controversy over distinguishing juvenile bipolar disorder (JBD) from disruptive behavior disorders (DBDs) in the clinical setting, we investigated whether referred children with a DBD and a negative mood component could be differentiated from those diagnosed with JBD. The distinction is important because treatments differ.

METHODS: In this single-site sample, 96 children with non-attention-deficit/hyperactivity DBD and depression were compared with 27 JBD children and 187 psychiatric comparison children on measures assessing behavior, functional impairment, symptom severity, psychopathology, and comorbid psychiatric diagnosis

RESULTS: Few differences were found between children with DBD and depression and those with JBD on measures of conduct problems, oppositionality, aggression, hostility, and psychopathology. More functional impairment was found in the JBD group who also had higher rates of comorbid posttraumatic stress disorder (PTSD), substance use disorders, and suicidality than the other groups.

CONCLUSIONS: These results do not support the specificity of aggression as a defining criterion for JBD and clinicians assessing such patients also should consider complex DBDs with an associated depressive component in the differential diagnosis. Children with JBD must be specifically assessed for comorbid developmental trauma, substance abuse, and suicidality. The association between JBD and PTSD needs further investigation in clinical research

KEYWORDS: juvenile bipolar disorder, childhood depression, disruptive behavior disorders



There has been considerable discussion among researchers and clinicians as to the characteristics of bipolar disorder (BD) in children.1-4 One source of disagreement is whether the diagnostic criteria for mania should be broadened to include children who exhibit severe irritability, “affective storms,” and prolonged aggressive, sometimes violent, temper outbursts.1,5 Proponents of a “broad phenotype” de-emphasize distinct episodes of euphoria, arguing instead that manic children may present with severe irritability, along with disruptive, unmanageable, and explosive behaviors.1,3

There is clear evidence that many children and adolescents exhibit chronic and severe irritability, emotional lability, and disruptive and oppositional behavior. The diagnostic dilemma for researchers and clinicians is whether these children should be diagnosed with BD. The use of adult criteria for children is challenged by reports that mania in children varies from the classic descriptions of episodic mood disturbance in adults. However, the evidence is not sufficient to conclude that a “broad phenotype,” such as that described by Biederman and colleagues,1 evolves into the well-established disorder observed with adults. Hence, recently formulated practice guidelines for diagnosing and treating juvenile BD (JBD) state that clinicians should use the DSM-IV-TR diagnostic criteria.6 Another important diagnostic issue is the relationship between JBD and externalizing disorders.7 JBD often co-occurs with externalizing disorders such as attention-deficit/hyperactivity disorder (ADHD), oppositional defiant disorder (ODD), and conduct disorder (CD).8-12 These externalizing disorders also have high rates of irritability, hostility, and aggression. The overlap in symptoms of disruptive behavior disorders (DBDs) with JBD has fueled a debate about the relationship between these disorders.7

Although a relationship between JBD and DBDs is suggested by several findings, the nature of these relationships remains unclear. First, many studies report high rates of JBD comorbidity with ADHD and DBDs, concluding that even though there is frequent overlap across symptoms of irritability, hostility, and aggression, CD or ODD and JBD are distinct yet often comorbid disorders.13,14 Other studies did not find such comorbidity.15 This issue is further confused because clinical studies report an asymmetrical relationship between ADHD and JBD—high rates of ADHD have been found in children diagnosed with JBD, but lower rates of JBD have been reported in children with ADHD. For example, in 1 study, 74% of 27 JBD patients were diagnosed with co-occurring ADHD, yet only 7.4% of 249 ADHD patients were diagnosed with comorbid JBD.16 Second, JBD is associated with irritability, hostility, and severe aggression.8 However, aggression also has been described in clinically referred children with ADHD without JBD,17 as well as in many other types of non-JBD psychiatric disorders in clinically referred children.18,19 Finally, both ADHD and DBD children share a vulnerability to mood and affective dysregulation. In ADHD and DBD this may take the form of emotional impulsiveness and rapid mood swings,20 and in children with JBD this may take the form of chronic irritability and severe mood dysregulation (SMD).2,21,22 Indeed, Leibenluft and colleagues2,21 have suggested criteria for a range of narrow to broad phenotypes of JBD, incorporating irritability, aggression, and negative affective valence with symptoms of overarousal into a broad clinical phenotype of JBD identified as “children with SMD.” Taken as a whole, these studies suggest that there is a relationship between JBD, externalizing behavior disorders, irritable aggressive behavior, and negative affective valence in clinically referred children but that the exact nature of the relationship remains unclear.

Fewer studies have examined relationships between ODD/CD and JBD than between ADHD and JBD in referred youngsters.10 In a comparison of 23 boys age 6 to 10 with DSM-III-R manic symptoms, 23 “next-seen” controls, and 23 comorbidity controls with ADHD, ODD, or CD, few differences were found between boys with manic symptoms and boys with DBDs after controlling for disruptive behavior symptoms. These findings suggest that these disorders may be difficult to distinguish from one another in clinically referred children.23 In contrast, a study examining 3 groups of children—1 group of 110 youths with DSM-III-R mania, 1 group of 116 children with CD, and 1 group of 79 youths with both mania and CD—found evidence to support a clear distinction between diagnostic groups.24 Thus, the extant literature remains unclear as to whether JBD and DBDs can reliably be distinguished from one another in the clinical setting.

Moreover, research that examines the relationship between BD and DBDs often ignores the evidence that children and adolescents with conduct problems frequently exhibit co-occurring depression. There is strong evidence of a high degree of comorbidity between depression and DBDs in children and adolescents, especially among youths who are referred to mental health professionals.25 These findings raise the possibility that depressed or dysphoric mood, rather than mania, may account for the association between DBDs, aggression, and JBD.

Given a suggested relationship between DBDs with a negative mood component and the spectrum of JBD, it is important to directly compare these groups in clinically referred youths. Yet few studies have directly compared children with non-ADHD DBDs and a negative mood component with children with JBD in referred clinical samples. To better understand these relationships, we investigated characteristics of children with JBD and compared them with a group of children diagnosed with non-ADHD DBD plus a comorbid depressive or dysthymic disorder, and a third group of psychiatric comparison children in a clinically referred sample. We specifically sought to determine if the 3 groups were distinguishable, based on a comparison of clinical characteristics that included aggression, hostility, oppositional defiant and conduct problems, mood, functional impairment, symptom severity, hyperactive-impulsive symptoms, general psychopathology, and psychiatric diagnosis.


The sample included 310 children and adolescents referred to the Pediatric Psychopharmacology Clinic at the University of Massachusetts Medical School who were consecutively ascertained and evaluated independently of any specific psychiatric diagnosis or a priori patient characteristic. Assessment procedures included information from parents, teachers, and the child. Procedures have been reported previously.16 Briefly, diagnoses were based on the parent-report, child psychiatrist–administered Kiddie Schedule for Affective Disorders and Schizophrenia–Epidemiologic Version (K-SADS-E) for lifetime disorders and on clinical assessment interviews of the child, supplemented by child self-report, parent-report, and teacher-report rating scales of psychopathology. Using all available information, final DSM-IV-TR diagnoses were assigned after an assessment team review of all available clinical information using a best estimate procedure.26 Clinicians assigned final impairment and symptom severity ratings. Parents and legal guardians provided clinical consent for assessment procedures. The study was approved by the University of Massachusetts institutional review board and also approved under a waiver by the University of Connecticut School of Medicine institutional review board. Exclusion criteria included children with schizophrenia, autism, mental retardation (IQ ≤70), or unstable medical or neurological illness.

Three groups were created for comparison: Group 1: 27 children with juvenile bipolar disorders; Group 2: 96 children with either CD or ODD and either depression or dysthymia (DBD + depression); and Group 3: a comparison group of 187 referred children who did not fit into Group 1 or Group 2, including depressed children without DBDs, children with ODD or CD but without depression or dysthymia, children with other psychiatric diagnoses and problems, including anxiety disorders, phobias, obsessive-compulsive disorder (OCD), ADHD, posttraumatic stress disorder (PTSD), cigarette smoking, substance abuse, and suicidality (lifetime gesture and/or attempt).

Demographic information for the 3 groups is presented in TABLE 1. The 3 groups differed significantly on only 1 demographic variable: age (F[2,307] = 4.96; P < .01). The JBD group (mean age, 12.0) and the DBD + depression group (mean age, 11.1) were both significantly older than the comparison group (mean age, 10.1). The 3 groups did not differ significantly in terms of sex, self-reported family income, self-reported highest parental educational level achieved, or race/ethnicity. For the entire sample of 310 children, 92% were white and 70% were male, with a mean age of 10.65 ± 3.4 years (range: age 3 to 19). The mother’s average educational level was 14.5 ± 2.5 years and fathers had an average of 14.5 ± 2.95 years of education. Family yearly income was stratified as: ≤$39,000 = 26%; $40,000 to $49,000 = 14%; $50,000 to $75,000 = 23%; and >$75,000 = 37%.

Table 1

Demographics of study participants

(N = 27)
DBD + depression
(N = 96)
Comparison group
(N = 187)
Age, mean (± SD), years 12.0 (3.46) 11.1 (2.98) 10.1 (3.56)
Male, n (%) 16 (59%) 66 (69%) 135 (72%)
Family income ≥$50,000, % 48% 59% 64%
Mother’s education, mean (SD), years 14.4 (2.58) 14.4 (2.39) 14.6 (2.60)
Father’s education, mean (SD), years 14.6 (3.03) 14.5 (3.09) 14.5 (2.88)
Race/ethnicity, white 89% 93% 93%
DBD: disruptive behavior disorder; JBD: juvenile bipolar disorder; SD: standard deviation.

K-SADS-E. A board-certified child psychiatrist individually administered the K-SADS-E27 to the parent or guardian of the referred child. A total of 5 child and adolescent psychiatrists conducted the interviews. Interrater reliability for diagnosis was good to excellent and has been previously reported.16 In addition to the DSM-IV diagnosis, the child psychiatrist rated impairment using the Children’s Global Assessment Scale (CGAS), which is a reliable measure of daily functioning and impairment.28,29 Agreement on this scale was good, as reported previously.16 Lower scores on the CGAS indicate more functional impairment. Severity of illness was assessed using the psychiatrist-rated Clinical Global Impression–Severity of Illness (CGI-S) scale.30 This scale was used to assess severity of psychopathology regardless of diagnosis. The CGI is a 7-point rating scale that ranges from 1 (no signs of illness) to 4 (moderately ill) to 7 (extremely ill). The scale has been shown to have acceptable concurrent validity in children with severe emotional disturbance.31

Aggression and hostility. The Modified Overt Aggression Scale (MOAS)32,33 is a 20-item scale that assesses the frequency and severity of overt aggression during the previous month. Parents completed this scale, which commonly is used to rate aggression in clinical settings. The MOAS assesses 4 categories of aggression, including verbal (threats to harm others), object (impulsive property destruction), self (self-injurious behavior), and other (physical assault). Proactive and reactive aggression were assessed using the Proactive/Reactive Aggression Rating Scale, which was completed by parents.34,35 This scale consists of 3 questions assessing reactive aggression and 3 questions assessing proactive aggression. Reactive aggression is in response to frustration or threat and carries a high affective valence and autonomic nervous system arousal. Proactive aggression usually is premeditated, planned, and driven by contingency reinforcement, with low affective valence and arousal patterns.36 The Feelings Scale37 based on the Buss-Durkee Hostility Inventory37,38 was used to assess the child’s self-report of hostility. This scale yields an Expressed Hostility subscale and a Perceived Hostility subscale, as well as a Total Hostility score. The Behavior Scale assesses the 15 DSM-IV-TR criteria of CD and the 8 items of ODD.39 Completed by parents to assess symptoms over the past week, the scale is scored from 0 (not at all/rarely) to 3 (always).

Attention-deficit/hyperactivity symptoms. Classroom teachers assessed the child’s hyperactive and impulsive behaviors using the 10-item Abbreviated Conners’ Teacher Rating Scale (ACTRS).40 The ACTRS is a 10-item measure that is scored on a 0-to-3 scale. Parents completed the ADHD Rating Scale–IV.41 This measure uses a 0-to-3 scale to rate the frequency of 18 ADHD items listed in the DSM-IV.39 Internal consistency and test-retest reliability is well established for both measures.

Children’s Depression Inventory (CDI). The CDI42 is a 27-item child self-reported measure of depressive symptoms. The CDI has sound psychometric properties, including high internal consistency, test-retest reliability, and discriminant validity.

Child Behavior Checklist (CBCL). Parents completed the CBCL,43 a 118-item parent-reported measure of a child’s behavioral and emotional problems. The CBCL contains 8 narrow-band subscales (Anxious/Depressed; Withdrawn; Somatic Complaints; Social Problems; Delinquent Behavior; Thought Problems; Attention Problems; and Aggressive Behavior) and 2 broad-band subscales (Internalizing problems and Externalizing problems). A total score can be calculated. For this study, T-scores are reported.

Data analysis. Descriptive statistics were calculated for the entire sample, and separately for the JBD, DBD + depression, and psychiatric comparison groups. A 1-way analysis of variance series was conducted to determine whether there were significant between-group differences for the measures of anger/hostility, aggression, ADHD/impulsivity, depression, and overall psychopathology. A 95% CI for the means and partial eta-squared (a measure of effect size) also was calculated. Partial eta squared provides an estimate of the proportion of the variance explained by the variable.44 Chi-square tests were conducted to determine whether there were significant between-group differences for categorical variables, such as DSM-IV-TR diagnoses. Cramer’s V was calculated to determine the magnitude of the association between the group variable and DSM diagnoses.44


Hostility, aggression, ODD, and CD symptoms

TABLE 2 presents results for hostility, aggression, ODD, and CD symptoms across the 3 groups. With respect to hostility, there were significant group differences for expressed hostility (F[2,245] = 3.27; P < .05). CIs for the mean scores for the 3 groups overlapped, and partial eta-squared was .03. There were no significant group differences for perceived hostility.

There were significant group differences for both reactive aggression (F[2,246] = 12.69; P < .01) and proactive aggression (F[2,246] = 8.12; P < .001). For reactive aggression, inspection of the CIs revealed that mean scores for the JBD and DBD + depression groups were significantly higher than the comparison group. The CIs for the JBD and DBD + depression groups overlapped. Partial eta-squared for reactive aggression was .09. For proactive aggression, inspection of the CIs revealed that mean scores for the JBD and DBD + depression groups were significantly higher than the comparison group. The CIs for the JBD and DBD + depression groups overlapped. Partial eta-squared for proactive aggression was .06.

There were significant group differences for the MOAS Verbal Aggression (F[2,247] = 10.92; P < .001) and MOAS Physical Aggression (F[2,272] = 4.43; P < .05) scales. For MOAS Verbal Aggression, inspection of the CIs revealed that mean scores for the JBD and DBD + depression groups were significantly higher than the comparison group. The CIs for the JBD and DBD + depression groups overlapped. Partial eta squared for MOAS Verbal Aggression was .08. For MOAS Physical Aggression, inspection of the CI suggested that mean scores for the JBD and DBD + depression groups were higher than the comparison group. Partial eta-squared for MOAS Physical Aggression was .03. There were no significant group differences for MOAS Self-Aggression or MOAS Other Aggression scales.

There were no significant group differences for the Behavior Scale.


Means and CIs for hostility, aggression, ODD, and CD behaviorsa

  JBD DBD + depression Comparison group
Expressed Hostilityb 6.5 (5.63)
(3.9 to 9.1)
7.0 (6.71)
(5.5 to 8.5)
5.0 (5.27)
(4.1 to 5.8)
Perceived Hostilityb 6.0 (9.48)
(1.7 to 10.3)
6.0 (7.62)
(4.3 to 7.7)
4.2 (5.70)
(3.3 to 5.1)
Reactive Aggressionc 3.6 (1.33)
(3.0 to 4.2)
3.4 (1.16)
(3.2 to 3.7)
2.7 (1.20)
(2.5 to 2.9)
Proactive Aggressionc 2.1 (1.36)
(1.5 to 2.7)
1.7 (0.87)
(1.6 to 1.9)
1.4 (0.73)
(1.3 to 1.5)
Verbal Aggressiond 14.3 (10.33)
(9.8 to 18.7)
12.4 (9.54)
(10.3 to 14.6)
7.7 (8.11)
(6.4 to 9.0)
Object Aggressiond 13.5 (13.43)
(8.1 to 18.9)
9.4 (8.79)
(7.5 to 11.2)
7.4 (10.30)
(5.8 to 9.0)
Self-Aggressiond 6.8 (9.00)
(3.1 to 10.4)
6.9 (7.85)
(5.3 to 8.6)
4.9 (8.95)
(3.5 to 6.3)
Other Aggressiond 8.5 (9.39)
(4.7 to 12.2)
7.2 (9.27)
(5.2 to 9.2)
5.5 (8.30)
(4.2 to 6.8)
Behavior Scale 2.1 (1.36)
(–2.1 to 2.9)
0.81 (5.90)
(–0.46 to 2.1)
0.36 (4.65)
(–0.3 to 1.1)
aMeans and SDs are presented with CIs below.
bBuss-Durkee Hostility Inventory.
cProactive/Reactive Aggression Rating Scale.
dModified Overt Aggression Scale.
CD: conduct disorder; DBD: disruptive behavior disorder; JBD: juvenile bipolar disorder; ODD: oppositional defiant disorder; SD: standard deviation.
Depression and ADHD symptoms

TABLE 3 presents data on child self-reported depression symptom severity and parent- and teacher-rated inattention and hyperactive-impulsive symptom severity. There were significant group differences on the CDI (F[2,257] = 18.28; P < .001). Inspection of the CIs revealed that the mean scores for the JBD and DBD + depression groups were significantly higher than the comparison group. The CIs for the JBD and DBD + depression groups overlapped. Partial eta-squared for the CDI was .13.

There were significant group differences on the ADHD Rating Scale (F[2,269] = 5.70; P < .005). Inspection of the CI revealed that the DBD + depression group had higher scores than the comparison group. The CI for the JBD group overlapped with the other 2 groups. The partial eta-squared for the ADHD Rating Scale was .05. There were no significant group differences for the ACTRS.

Impairment and symptom severity

TABLE 3 also reports data on clinician-rated functional impairment (CGAS) and overall symptom severity (CGI). There were significant group differences on the CGAS (F[2,301] = 14.59; P < .001). Inspection of the CIs revealed that the JBD group was rated as more impaired than the other 2 groups. Partial eta-squared for the CGAS was .09.

There were significant group differences on the CGI-Severity scale (F[2,299] = 4.30; P < .05). Inspection of the CIs revealed that clinicians rated the JBD group as having higher global symptom severity than the comparison group. The CI for the DBD + depression group overlapped with the CIs for the other 2 groups. Partial eta-squared for CGI-Severity was .03.


Means and CIs for mood, ADHD symptoms, impairment, and symptom severitya

  JBD DBD + depression Comparison group
CDI 15.2 (9.88)
(10.9 to 19.5)
14.7 (8.65)
(12.9 to 16.6)
9.2 (6.10)
(8.2 to 10.2)
ADHD Rating Scale–IV 29.7 (20.76)
(21.2 to 38.3)
38.2 (25.44)
(32.8 to 43.6)
30.0 (13.54)
(27.9 to 32.1)
ACTRS 16.5 (12.18)
(11.3 to 21.8)
16.5 (12.33)
(13.8 to 19.3)
14.2 (10.41)
(12.5 to 15.9)
CGAS 43.5 (6.02)
(41.1 to 45.9)
48.0 (5.10)
(46.9 to 49.0)
49.7 (5.94)
(48.9 to 50.6)
CGI-S 4.9 (1.18)
(4.4 to 5.4)
4.4 (1.19)
(4.1 to 4.6)
4.2 (0.95)
(4.1 to 4.4)
aMeans and SDs are presented with CIs below.
ACTRS: Abbreviated Conners’ Teacher Rating Scale; ADHD: attention-deficit/hyperactivity disorder; CDI: Children’s Depression Inventory; CGAS: Children’s Global Assessment Scale; CGI-S: Clinical Global Impression–Severity of Illness scale; DBD: disruptive behavior disorder; JBD: juvenile bipolar disorder; SD: standard deviation.
CBCL scores

TABLE 4 presents parent-reported CBCL data across the 3 groups. There were significant group differences for the Aggressive (F[2,300] = 18.30; P < .001), Anxious/Depressed (F[2,300] = 18.90; P < .001), Delinquent (F[2,297] = 15.47; P < .001), and Somatic (F[2,301] = 8.69, P < .001) subscales. For all 4 subscales, inspection of the CIs revealed that the JBD and DBD + depression groups had higher scores than the comparison group. CIs for the JBD and DBD + depression groups overlapped for all 4 subscales. Partial eta-squared was .11, .11, .11, and .06 for the Aggressive, Anxious/Depressed, Delinquent, and Somatic subscales, respectively.

There were significant group differences for the Social subscale (F[2,297] = 5.95; P < .005). Inspection of the CIs revealed that the DBD + depression group had higher scores than the comparison group. The CI for the JBD group overlapped with the other groups. Partial eta-squared for the Social subscale was .04. There were significant group differences for the Thought subscale (F[2,296] = 3.36; P < .05).


Means and CIs for parent-reported Child Behavior Checklist T-scoresa

  JBD DBD + depression Comparison group
Aggressive 72.8 (12.74)
(67.7 to 77.8)
70.3 (9.87)
(68.3 to 72.3)
63.3 (10.88)
(61.7 to 64.9)
Anxious/Depressed 70.8 (16.57)
(64.2 to 77.3)
70.2 (8.64)
(68.5 to 72.0)
62.8 (10.18)
(61.3 to 64.3)
Delinquent 68.3 (7.97)
(65.1 to 71.4)
64.9 (9.02)
(68.3 to 72.3)
60.1 (8.97)
(58.8 to 61.5)
Somatic 67.3 (9.32)
(63.6 to 71.0)
64.7 (9.59)
(62.8 to 66.7)
61.0 (8.98)
(59.7 to 62.3)
Social 68.7 (11.7)
(64.3 to 73.1)
69.2 (10.23)
(67.1 to 71.3)
64.8 (10.90)
(63.1 to 66.4)
Thought 69.9 (8.39)
(66.6 to 73.2)
66.2 (9.51)
(64.2 to 68.1)
65.1 (8.90)
(63.8 to 66.4)
Attention 73.5 (9.75)
(69.7 to 77.4)
73.1 (8.65)
(73.1 to 74.9)
71.0 (9.55)
(69.6 to 72.4)
Withdrawn 55.3 (23.73)
(45.9 to 64.7)
58.2 (19.38)
(54.2 to 62.2)
58.3 (15.96)
(55.9 to 60.6)
Internalizing subscale 60.2 (25.52)
(50.1 to 70.3)
60.3 (23.25)
(55.5 to 65.0)
58.6 (18.44)
(59.9 to 65.3)
Externalizing subscale 61.6 (23.87)
(52.2 to 71.1)
59.7 (22.50)
(55.0 to 64.3)
57.5 (17.83)
(54.9 to 60.1)
Total score 63.7 (25.95)
(53.4 to 73.9)
62.9 (23.9)
(58.0 to 67.8)
62.6 (18.33)
(59.9 to 65.3)
aMeans and SDs are presented with CIs below.
DBD: disruptive behavior disorder; JBD: juvenile bipolar disorder; SD: standard deviation.
Psychiatric diagnosis

TABLE 5 presents data on psychiatric diagnoses across the 3 groups. Considering our data set as a whole, consecutively referred children to our clinic with DBDs + depression (N = 96; 32%) were much more common than referred children with JBD (N = 27; 8.6%) as assessed by our methodology. All 3 groups had high rates of comorbid psychiatric diagnoses, including anxiety disorders and ADHD (see TABLE 5). Children in the DBD + depression and JBD groups had significantly more diagnoses of separation anxiety (χ2[2] = 18.27; P < .001) and generalized anxiety (χ2[2] = 14.12; P < .001) than the comparison group. Cramer’s V was .24 for separation anxiety and .21 for generalized anxiety disorder (GAD). Children in the JBD group had significantly higher rates of PTSD (χ2[2] = 22.65; P < .001), substance use disorders (SUDs) (χ2[2] = 12.34; P < .001), and suicidality (lifetime history of suicidal gestures and/or suicide attempts) (χ2[2] = 30.18; P < .001) than children in the other 2 groups. Cramer’s V was .27, .20, and .31 for PTSD, substance abuse, and suicidality, respectively. All associations between group classification and diagnoses were moderately strong. No differences were found across the 3 groups for combined ADHD, the inattentive subtype of ADHD, cigarette smoking, OCD, or phobias.


Comorbid psychiatric diagnosis

(N = 27)
DBD + depression
(N = 96)
Comparison group
(N = 187)
Separation anxiety 56% (n = 15) 48% (n = 46) 26% (n = 49)
Social phobia 19% (n = 5) 27% (n = 26) 16% (n = 29)
Specific phobia 26% (n = 7) 33% (n = 32) 27% (n = 51)
GAD 67% (n = 18) 63% (n = 60) 42% (n = 78)
OCD 11% (n = 3) 8% (n = 8) 9% (n = 17)
PTSD 30% (n = 8) 9% (n = 9) 4% (n = 7)
Combined ADHD 48% (n = 13) 57% (n = 55) 43% (n = 80)
Primarily inattentive subtype 11% (n = 3) 21% (n = 20) 23% (n = 42)
SUDs 26% (n = 7) 8% (n = 8) 6% (n = 11)
Cigarette smoking 11% (n = 3) 5% (n = 5) 4% (n = 7)
Suicidality 22% (n = 6) 2% (n = 2) 2% (n = 3)
ADHD: attention-deficit/hyperactivity disorder; DBD: disruptive behavior disorder; GAD: generalized anxiety disorder; JBD: juvenile bipolar disorder; OCD: obsessive-compulsive disorder; PTSD: posttraumatic stress disorder; SUDs: substance use disorders.


The purpose of our descriptive study was to ascertain if clinically referred children with DSM-IV-TR mania and hypomania (JBD group) could be distinguished from a group of children with non-ADHD DBDs and a diagnosis of depression or dysthymia (DBD + depression group), and a psychiatric comparison group. This distinction is clinically important, as many behaviors and symptoms such as conduct problems, depression, irritability, and aggression may be common both to children with non-ADHD DBDs19 and to children with JBD.5 However, both psychotherapeutic and psychopharmacologic treatments differ across the groups, making the distinction between them clinically important and meaningful.

Significant between-group differences were found, indicating youths in the JBD and DBD + depression groups exhibited higher levels of proactive and reactive aggression as well as verbal and physical aggression than the comparison group. However, youths with JBD did not differ from youths with DBD + depression on any aggression measure. Between-group differences accounted for ≤10% of the variability in aggression, indicating that the differences obtained for this sample were modest. From a diagnostic point of view, this study indicates that children with JBD exhibit aggressive behavior, but that aggressive behavior is not specific to BD. These findings suggest that clinicians should consider diagnoses of co-occurring externalizing and depressive disorders when evaluating youths who present with aggressive behavior because these disorders are more common than JBD.

High lifetime rates of diagnostic comorbidity were found in our sample. The association between group classification and psychiatric diagnosis was moderately strong. Significantly more children in the JBD group met DSM-IV-TR criteria for lifetime PTSD than did children in the other 2 groups. Although the overall numbers were small, JBD children were greater than 3 times more likely to meet PTSD criteria as DBD + depression or comparison group children. These results support previous research reporting that childhood traumatic stress is prevalent in bipolar disorder patients,45 is associated with a more adverse course of bipolar illness,46 and may be a risk factor in the development of PTSD in juvenile bipolar illness.47 It is suggested that symptoms of childhood BD and childhood posttraumatic stress problems as reported on the CBCL may reflect a measure of a single syndrome of affective and behavioral dysregulation.48 We extend these findings to report a higher cross-sectional prevalence of DSM-IV-TR PTSD in clinically referred JBD children compared with children with DBDs and children with non-JBD affective and ADHD disorders. Our finding of a significantly higher rate of PTSD diagnosis in JBD children supports an association between the 2 disorders, albeit in clinically referred children, that requires further replication and investigation. In addition to an association between the 2 conditions, the question of traumatized youths presenting with JBD-like symptoms is important for the field to address. Overlap exists between the overarousal symptoms of PTSD and the symptoms of JBD, including irritability, aggression, and sleep problems, which may cause the clinician to overlook the possibility of PTSD as a primary diagnosis in children presenting with JBD-like symptoms.49,50

Similar to previous findings in studies of JBD, 43% to 57% of referred children in our sample had comorbid ADHD. The JBD group had a significantly higher rate of lifetime suicide gestures/attempts than the other 2 comparison groups in our sample. Our findings are consistent with an extant literature documenting that suicidality is a risk for children with pediatric BD.6 Children with DBDs + depression and JBD had higher rates of separation anxiety and GAD than the comparison group. Our findings are consistent with reports that find high rates of comorbid anxiety disorders in children with DBDs51 and in those with JBD.13,52 The JBD group had a significantly higher rate of substance abuse compared with our 2 comparison groups. Epidemiologic and clinical studies report that juvenile-onset BD confers an even greater risk of SUDs than adult-onset BD—a risk that markedly escalates among bipolar adolescents53 and is associated with legal and academic difficulties, pregnancy, and suicidality.54 Because JBD typically precedes SUDs in development, there may exist a window of opportunity for substance abuse prevention in adolescence if JBD can correctly be identified and treated in childhood.53

Results should be considered within the context of some study limitations. Our study design is cross-sectional and results are correlational. Thus, causality cannot be determined. Our study sample was overwhelmingly white and referred to a tertiary medical school child psychopharmacology clinic. Most referrals to our clinic have complex, comorbid, and severe psychopathology and may not have responded to previous interventions. As such, our results might not generalize to other populations.

Another possible limitation is our evaluation procedure to establish a diagnosis of JBD. Although our evaluation procedures adhered to recent practice guidelines requiring the presence of a clearly defined manic or hypomanic episode,6 variations in diagnostic and evaluation tools and procedures may yield differing results. Another possible limitation is that we only interviewed caregivers using the K-SADS-E and did not use a structured interview of the child in addition. However, previous research has established that caregiver report of childhood psychopathology shows excellent accuracy, specificity, reliability, and validity for JBD55 and that parent-completed assessment measures for JBD provide better information about child manic symptoms than child- or teacher-reported measures.56


Our results do not support the diagnostic specificity of aggression in children and adolescents, and clinicians assessing such youngsters also should consider complex DBDs with an associated depressive component as well as JBD in the differential diagnosis. Assessing children with JBD comorbidity with substance abuse, suicidality, and PTSD requires careful evaluation. The association between JBD and PTSD needs further investigation in clinical research.

DISCLOSURES: Dr. Connor receives grant or research support from Shire Pharmaceuticals; is a consultant to Shire Pharmaceuticals, Supernus Pharmaceuticals, and Neos Therapeutics; and is a speaker for Shire Pharmaceuticals. Dr. Doerfler reports no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.


  1. Biederman J, Mick E, Faraone SV, et al. Pediatric mania: a developmental subtype of bipolar disorder? Biol Psychiatry. 2000;48:458–466.
  2. Leibenluft E, Charney DS, Towbin KE, et al. Defining clinical phenotypes of juvenile mania. Am J Psychiatry. 2003;160:430–437.
  3. Leibenluft E, Rich BA. Pediatric bipolar disorder. Annu Rev Clin Psychol. 2008;4:163–187.
  4. Youngstrom EA, Birmaher B, Findling RL. Pediatric bipolar disorder: validity phenomenology, and recommendations for diagnosis. Bipolar Disord. 2008;10:194–214.
  5. Biederman J, Faraone SV, Wozniak J, et al. Parsing the association between bipolar, conduct, and substance use disorders: a familial risk analysis. Biol Psychiatry. 2000;48:1037–1044.
  6. McClellan J, Kowatch R, Findling RL, et al. Practice parameter for the assessment and treatment of children and adolescents with bipolar disorder. J Am Acad Child Adolesc Psychiatry. 2007;46:107–125.
  7. Kim EY, Miklowitz DJ. Childhood mania attention deficit hyperactivity disorder and conduct disorder: a critical review of diagnostic dilemmas. Bipolar Disord. 2002;4:215–225.
  8. Biederman J, Faraone S, Mick E, et al. Attention-deficit hyperactivity disorder and juvenile mania: an overlooked comorbidity? J Am Acad Child Adolesc Psychiatry. 1996;35:997–1008.
  9. Carlson GA. Mania and ADHD: comorbidity or confusion. J Affect Disord. 1998;51:177–187.
  10. Galanter CA, Leibenluft E. Frontiers between attention deficit hyperactivity disorder and bipolar disorder. Child Adolesc Psychiatr Clin N Am. 2008;17:325–346 viii-ix.
  11. Kent L, Craddock N. Is there a relationship between attention deficit hyperactivity disorder and bipolar disorder? J Affect Disord. 2003;73:211–221.
  12. Luckenbaugh DA, Findling RL, Leverich GS, et al. Earliest symptoms discriminating juvenile-onset bipolar illness from ADHD. Bipolar Disord. 2009;11:441–451.
  13. Spencer TJ, Biederman J, Wozniak J, et al. Parsing pediatric bipolar disorder from its associated comorbidity with the disruptive behavior disorders. Biol Psychiatry. 2001;49:1062–1070.
  14. Wozniak J, Biederman J, Kiely K, et al. Mania-like symptoms suggestive of childhood-onset bipolar disorder in clinically referred children. J Am Acad Child Adolesc Psychiatry. 1995;34:867–876.
  15. Jaideep T, Reddy YC, Srinath S. Comorbidity of attention deficit hyperactivity disorder in juvenile bipolar disorder. Bipolar Disord. 2006;8:182–187.
  16. Doerfler LA, Connor DF, Toscano PF Jr. Aggression ADHD symptoms, and dysphoria in children and adolescents diagnosed with bipolar disorder and ADHD. J Affect Disord. 2011;131:312–319.
  17. Connor DF, Chartier KG, Preen EC, et al. Impulsive aggression in attention-deficit/hyperactivity disorder: symptom severity, co-morbidity, and attention-deficit/hyperactivity disorder subtype. J Child Adolesc Psychopharmacol. 2010;20:119–126.
  18. Connor DF, McLaughlin TJ. Aggression and diagnosis in psychiatrically referred children. Child Psychiatry Hum Dev. 2006;37:1–14.
  19. Connor DF, Carlson GA, Chang KD, et al. Juvenile maladaptive aggression: a review of prevention, treatment, and service configuration and a proposed research agenda. J Clin Psychiatry. 2006;67:808–820.
  20. Barkley RA, Fischer M. The unique contribution of emotional impulsiveness to impairment in major life activities in hyperactive children as adults. J Am Acad Child Adolesc Psychiatry. 2010;49:503–513.
  21. Leibenluft E. Severe mood dysregulation irritability, and the diagnostic boundaries of bipolar disorder in youths. Am J Psychiatry. 2011;168:129–142.
  22. Leibenluft E, Blair RJ, Charney DS, et al. Irritability in pediatric mania and other childhood psychopathology. Ann N Y Acad Sci. 2003;1008:201–218.
  23. Carlson GA, Loney J, Salisbury H, et al. Young referred boys with DICA-P manic symptoms vs. two comparison groups. J Affect Disord. 1998;51:113–121.
  24. Biederman J, Faraone SV, Chu MP, et al. Further evidence of a bidirectional overlap between juvenile mania and conduct disorder in children. J Am Acad Child Adolesc Psychiatry. 1999;38:468–476.
  25. Wolff JC, Ollendick TH. The comorbidity of conduct problems and depression in childhood and adolescence. Clin Child Fam Psychol Rev. 2006;9:201–220.
  26. Leckman JF, Sholomskas D, Thompson WD, et al. Best estimate of lifetime psychiatric diagnosis: a methodological study. Arch Gen Psychiatry. 1982;39:879–883.
  27. Orvaschel H. Schedule for Affective Disorder and Schizophrenia for School-Age Children Epidemiologic Version 5. Ft. Lauderdale FL: Nova Southeastern University, Center for Psychological Studies; 1994.
  28. Green B, Shirk S, Hanze D, et al. The Children’s Global Assessment Scale in clinical practice: an empirical evaluation. J Am Acad Child Adolesc Psychiatry. 1994;33:1158–1164.
  29. Shaffer D, Gould MS, Brasic J, et al. A children’s global assessment scale (CGAS). Arch Gen Psychiatry. 1983;40:1228–1231.
  30. Guy W. ECDEU assessment manual for psychopharmacology. Rockville MD: U.S. Department of Health, Education, and Welfare, Public Health Service, Alcohol, Drug Abuse, and Mental Health Administration, National Institute of Mental Health, Psychopharmacology Research Branch, Division of Extramural Research Programs; 1976.
  31. Mattison RE, Bagnato SJ, Mayes SD, et al. Reliability and validity of teacher diagnostic ratings for children with behavioral and emotional disorders. J Psychoeduc Assess. 1990;8:509–517.
  32. Sorgi P, Ratey J, Knoedler DW, et al. Rating aggression in the clinical setting. A retrospective adaptation of the Overt Aggression Scale: preliminary results. J Neuropsychiatry Clin Neurosci. 1991;3:S52–S56.
  33. Yudofsky SC, Silver JM, Jackson W, et al. The Overt Aggression Scale for the objective rating of verbal and physical aggression. Am J Psychiatry. 1986;143:35–39.
  34. Dodge KA. The structure and function of reactive and proactive aggression. In: Pepler DJ Rubin KH, eds. The development and treatment of childhood aggression. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.; 1991:201–218.
  35. Dodge KA, Coie JD. Social-information-processing factors in reactive and proactive aggression in children’s peer groups. J Pers Soc Psychol. 1987;53:1146–1158.
  36. Raine A, Dodge K, Loeber R, et al. The Reactive-Proactive Aggression Questionnaire: differential correlates of reactive and proactive aggression in adolescent boys. Aggress Behav. 2006;32:159–171.
  37. Treiber FA, Musante L, Riley W, et al. The relationship between hostility and blood pressure in children. Behav Med. 1989;15:173–178.
  38. Buss AH, Durkee A, Baer MB. The measurement of hostility in clinical situations. J Abnorm Psychol. 1956;52:84–86.
  39.  Diagnostic and statistical manual of mental disorders, 4th ed, text rev. Washington, DC: American Psychiatric Association; 2000.
  40. Conners CK. Manual for Conners’ Rating Scales. North Tonawanda NY: Multi-Health Systems; 1997.
  41. DuPaul GJ, Power TJ, Anastopoulos AD, et al. ADHD rating scale IV: checklists, norms, and clinical interpretation. New York, NY: Guilford Press; 1998.
  42. Kovacs M. Child Depression Inventory technical manual. North Tonawanda NY: Multi-Health Systems; 2003.
  43. Achenbach TM. Child Behavior Checklist–Cross-Informant Version. Burlington VT: University of Vermont, Department of Psychiatry; 1991.
  44. Trusty J, Thompson B, Petrocelli JV. Practical guide for reporting effect size in quantitative research in the Journal of Counseling & Development. J Couns Dev. 2004;82:107–110.
  45. Alvarez MJ, Roura P, Osés A, et al. Prevalence and clinical impact of childhood trauma in patients with severe mental disorders. J Nerv Ment Dis. 2011;199:156–161.
  46. Leverich GS, McElroy SL, Suppes T, et al. Early physical and sexual abuse associated with an adverse course of bipolar illness. Biol Psychiatry. 2002;51:288–297.
  47. Strawn JR, Adler CM, Fleck DE, et al. Post-traumatic stress symptoms and trauma exposure in youth with first episode bipolar disorder. Early Interv Psychiatry. 2010;4:169–173.
  48. Ayer L, Althoff R, Ivanova M, et al. Child Behavior Checklist Juvenile Bipolar Disorder (CBCL-JBD) and CBCL Posttraumatic Stress Problems (CBCL-PTSP) scales are measures of a single dysregulatory syndrome. J Child Psychol Psychiatry. 2009;50:1291–1300.
  49. Chessen CE, Comtois KA, Landes SJ. Untreated posttraumatic stress among persons with severe mental illness despite marked trauma and symptomatology. Psychiatr Serv. 2011;62:1201–1206.
  50. Quarantini LC, Miranda-Scippa A, Nery-Fernandes F, et al. The impact of comorbid posttraumatic stress disorder on bipolar disorder patients. J Affect Disord. 2010;123:71–76.
  51. Polier GG, Herpertz-Dahlmann B, Matthias K, et al. Associations between trait anxiety and psychopathological characteristics of children at high risk for severe antisocial development. Atten Defic Hyperact Disord. 2010;2:185–193.
  52. Geller B, Tillman R, Bolhofner K, et al. Child bipolar I disorder: prospective continuity with adult bipolar I disorder; characteristics of second and third episodes; predictors of 8-year outcome. Arch Gen Psychiatry. 2008;65:1125–1133.
  53. Goldstein BI, Bukstein OG. Comorbid substance use disorders among youth with bipolar disorder: opportunities for early identification and prevention. J Clin Psychiatry. 2010;71:348–358.
  54. Wilens TE, Biederman J, Millstein RB, et al. Risk for substance use disorders in youths with child- and adolescent-onset bipolar disorder. J Am Acad Child Adolesc Psychiatry. 1999;38:680–685.
  55. Faraone SV, Biederman J, Milberger S. How reliable are maternal reports of their children’s psychopathology? One-year recall of psychiatric diagnoses of ADHD children. J Am Acad Child Adolesc Psychiatry. 1995;34:1001–1008.
  56. Youngstrom E, Meyers O, Youngstrom JK, et al. Diagnostic and measurement issues in the assessment of pediatric bipolar disorder: implications for understanding mood disorder across the life cycle. Dev Psychopathol. 2006;18:989–1021.

CORRESPONDENCE: Daniel F. Connor, MD, Department of Psychiatry/MC 1410, University of Connecticut School of Medicine, 263 Farmington Avenue, Farmington, CT 06030 USA E-MAIL: