Bipolar Diagnoses in Community Mental Health:
Achenbach Child Behavior Checklist Profiles and
Patterns of Comorbidity
Eric Youngstrom, Jennifer Kogos Youngstrom, and Maryjean Starr
Background: There are converging findings about pediatric bipolar disorder (PBD) in terms of associated comorbidity and behavior
problem profiles on the Achenbach Child Behavior Checklist (CBCL). However, no study has examined clinical or demographic
characteristics of youths clinically diagnosed with bipolar disorder in a low-income, diverse community clinical sample.
Methods: Archival data (N = 3086 cases) from six urban community mental health centers (CMHC) were reviewed to determine the
base rate of bipolar disorder and the demographic and clinical characteristics (comorbidity and CBCL profiles) associated with the
diagnosis.
Results: Roughly 6% of the sample received clinical diagnoses of PBD. Patterns of comorbidity and CBCL profiles were highly similar
to published samples. However, elevated CBCL scores were not specific to bipolar disorder, since they were also frequently high for
nonbipolar cases.
Conclusions: There appears to be substantial convergence between the demographic and clinical characteristics of cases clinically
diagnosed with PBD versus those diagnosed with semistructured research interviews, strengthening the validity of both sets of
diagnoses. At the same time, the CBCL appears to do poorly discriminating clinical diagnoses of PBD, due to the pervasive externalizing
behavior problems in CMHC samples and the variable presentation of PBD cases.
Key Words: Pediatric bipolar disorder, clinical diagnosis, validity,
comorbidity, demographics
A
lmost no research has been done on pediatric bipolar
disorder (PBD) in low-income or demographically di-
verse samples. This is a significant gap, given the demo-
graphic changes in the United States leading to increased diver-
sity throughout the country, as well as in the families seeking
clinical care (Hernandez 1997). It also is unclear whether recent
gains in knowledge about the phenomenology, assessment, and
treatment of pediatric bipolar disorder generalize to low-income
and non-European American families.
Pediatric bipolar disorder is a controversial diagnosis, espe-
cially in prepubertal youths (Klein et al 1998; McClellan 1998).
Some epidemiological studies have documented few (Lewinsohn
et al 1995) or no cases of bipolar disorder (Costello et al 1996). At
the same time, the rate of clinical diagnosis of bipolar disorder
has increased rapidly (Hellander 2002; Naylor et al 2002; Young-
strom et al, in press). It is unclear whether clinicians are hewing
close to adult-oriented definitions versus using more broad
categories, such as Bipolar Disorder Not Otherwise Specified
(NOS) (Leibenluft et al 2003), or modifying their application of
criteria to be more developmentally sensitive (Geller et al 2002).
Thus, the extent of overlap between the sets of children identi-
fied as having bipolar disorders by researchers versus practicing
clinicians is uncertain.
There also may be demographic differences in the rate at
which bipolar disorder is identified. Research samples typically
have more male than female participants identified with bipolar
disorder, and the gender discrepancy is larger in prepubertal
than adolescent samples (Birmaher et al, unpublished data;
Findling et al 2001; Geller et al 2000; Wozniak et al 1995). In
constrast, epidemiological studies typically have not identified
gender differences in the rate of bipolar disorders (American
Psychiatric Association 2001; Kessler 1994), with the possible
exception of bipolar II appearing more common in adult females
(Goodwin and Jamison 1990). Furthermore, bipolar disorder is
diagnosed less frequently in African Americans in adult (Stra-
kowski et al 1996) and adolescent patients (DelBello et al 2000).
It would be helpful to assess whether gender and racial differ-
ences in identification are evident in community mental health,
where diverse families are heavily represented.
Characterization of youths diagnosed clinically with pediatric
bipolar disorder also has the potential to contribute to the
validation of the construct of pediatric bipolar disorder. Clinical
diagnoses, especially discharge diagnoses, have often been used
as the criterion against which assessment procedures are vali-
dated (Kendler and Roy 1995; Trull et al 1995), including
structured and semistructured diagnostic interviews (Spitzer et al
1992). As Table 1 delineates, clinical diagnosis and research
interviews have complementary strengths and weaknesses.
Thus, convergence between clinical diagnoses (especially taking
treatment or discharge diagnoses into account) and semistruc-
tured diagnoses in terms of the clinical characteristics of identi-
fied cases would enhance confidence in the validity of the
construct of pediatric bipolar disorder in general, because dis-
similar methods would be generating similar findings (Meyer
2002). Studying clinical diagnoses also provides important infor-
mation about secular trends in diagnostic practice (Stoll et al
1993).
The present study sought to address these issues by analyzing
archival data from a consortium of community mental health
centers (CMHC) serving a large Midwestern metropolitan region.
The consortium tracked intake, treatment, and discharge diag-
noses. In addition, the standard intake protocol used at all sites
required the primary caregiver to complete the Achenbach Child
Behavior Checklist (CBCL) (Achenbach 1991). The combination
of these data made it possible to examine both demographic
factors and aspects of clinical phenomenology associated with
From the Department of Psychology (EY, MS), Case Western Reserve Univer-
sity, Cleveland; and Applewood Centers, Incorporated (JKY, MS), Cleve-
land, Ohio.
Address reprint requests to Eric Youngstrom, Department of Psychology,
Case Western Reserve University, 10900 Euclid Avenue, Cleveland, OH
44106-7123; E-mail: Eric.Youngstrom@case.edu.
Received August 6, 2004; revised February 3, 2005; accepted April 5, 2005.
BIOL PSYCHIATRY 2005;58:569 –575 0006-3223/05/$30.00
doi:10.1016/j.biopsych.2005.04.004 © 2005 Society of Biological Psychiatry