Cognitive Distortions and Psychiatric Diagnosis in Dually

Diagnosed Adolescents
TRACY KEMPTON, PH.D., VINCENT B. VAN HASSELT, PH.D., OSCAR G. BUKSTEIN, M.D.,
AND JANE A. NULL, B.S.

ABSTRACT
Objective: The purpose of this study was to examine the characteristics and patterns of cognitive distortions among
psychiatrically hospitalized adolescents. Method: Measures of cognitive distortions, depression, and hopelessness
were administered to 135 adolescents on two psychiatric inpatient units. Subjects were grouped according to their Axis
I diagnoses: depression only, conduct disorder only, depression and substance abuse, conduct disorder and substance
abuse, all three diagnoses, and none of the three diagnoses. Results: Multivariate analyses of covariance indicated
that differently diagnosed adolescents exhibited varying levels of cognitive distorting as measured by the Children's
Negative Cognitive Errors Questionnaire (CNCEQ). In particular, adolescents with multiple Axis I diagnoses tended to
score highest. On all but one of four CNCEQ subscales, the depression only group evidenced as much cognitive
distortion as did the group with multiple diagnoses. However, each diagnostic grouping demonstrated its own somewhat
distinct distortions based on CNCEQ subscales. Conclusions: Findings are discussed in terms of the utility of differenti-
ating cognitive styles for subsequent treatment. It is suggested that disparate cognitive interventions could be matched
with adolescents displaying particular problems. J. Am. Acad. Child Ado/esc. Psychiatry, 1994, 33, 2:217-222. Key
Words: cognitive distortions, psychiatric diagnosis, inpatient adolescents.

Over the past three decades, considerable clinical and
investigative attention has focused on the relationship
between forms of cognitive distortions (e.g., catastroph-
izing, selective abstraction, overgeneralizing, personaliz-
ing) and psychopathology (cf Abramson et al., 1978;
Beck, 1967, 1976; Ellis, 1962). In particular, a bur-
geoning body of research has documented the associa-
tion between faulty or irrational cognitions and
depressed mood states (e.g., Krantz and Hammen,
1979; Nelson and Craighead, 1977). Although the
theoretical basis and nature of this association has
been questioned (Beidel and Turner, 1986; Coyne and
Accepted june 2, 1993.
Tracy Kempton is Associate Director, Child Inpatient Unit, and Assistant
Professor of Psychiatry, Kobacker Center, Medical College of Ohio; Dr. Van
Hasselt is Professor of Psychology, and Jane Null is Clinic and Research
Coordinator, Nova University, Ft. Lauderdale, FL; and Dr. Bukstein is
Assistant Professor ofPsychiatry, University ofPittsburgh School ofMedicine.
This study was supported in part by the Center for Education and Drug
Abuse Research (CEDAR), ftnded by the National Institute on Drug Abuse
(DA05605).
Reprint requests to Dr. Van Hasselt, Centerfor Psychological Studies, Nova
University, 3301 College Avenue, Fort Lauderdale, FL 33314.
0890-8567/94/3302-0217$03.0010©1994 by the American Academy
of Child and Adolescent Psychiatry.
Gottlib, 1983), such findings have been the impetus
for initial applications of cognitive therapy with a
wide range of adult disorders including depression,
personality disorders, and substance abuse (Beck and
Freeman, 1990; Beck et al., 1979; Foreyt and Rath-
jen, 1978).
Although investigative efforts regarding the role of
cognitive distortions in depression and other disorders
in younger groups are at the nascent stage, findings
of preliminary studies appear to support this associa-
tion. For example, in an investigation by Thurber
et al. (1990), the tendency for psychiatrically hospital-
ized 12- to 17-year-old adolescents to emit cognitive
errors, as evaluated via the Children's Negative Cogni-
tive Errors Questionnaire (CNCEQ), was significantly
correlated with scores on the Beck Depression Inven-
tory (Beck et al., 1961) and the Hopelessness Scale
for Children (Kazdin et al., 1986). However, in this
study, the possibility of common method variance must
be considered, as the measures completed by parents
concerning their children did not correspond with
the CNCEQ.
Haley et al. (1985) employed the Cognitive Bias

J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 33:2 FEBRUARY 1994 217

KEMPTON ET AL.
Questionnaire for Children (CBQC), a child and ado-
lescent version of the Cognitive Bias Questionnaire
(Krantz and Hammen, 1979), with 39 child and adoles-
cent inpatients ranging in age from 8 to 16 years (X
= 12.8). Seventeen of the subjects were diagnosed
with an affective disorder and 22 with a different or
with no Axis I disorder. Results indicated that scores
on the depressed distorted (DO) scale of the CBQC
were positively correlated with (1) psychiatric ratings
of dysthymic symptoms, (2) child-reported symptoms
on the Child Depression Inventory (Kovacs, 1981),
and (3) the child-completed Children's Depression
Scale (Lang and Tisher, 1978). In addition, the affective
group had significantly higher scores on the DO scale,
after adjusting for age (which differed between the
two groups).
The relationship between cognitive distortions and
substance abuse also has been examined. In an investiga-
tion of78 adolescents (age not provided) in a residential
drug treatment program, Denoff (1987) used regression
analyses to predict frequency of drug use from subscales
of the Rational Behavior Inventory (RBI) (Shorkey
and Whiteman, 1977). Specific beliefs, such as catas-
trophizing ("it is a catastrophe when things are not
the way one wants them"), approval ("it is absolutely
necessary to be approved of by everyone for everything
one does"), and blame and punishment ("certain people
are bad and should be punished") emerged as the
strongest predictors of substance abuse. A second at-
tempt to predict substance abuse in the same sample
employed the Perceived Parenting Questionnaire (a
modified version of the Cornell Parental Behavior
Scales) (Devereux et aI., 1969) and demographic infor-
mation, in addition to the RBI (Denoff, 1988). Again,
several RBI scales significantly predicted frequency of
drug use, with catastrophizing emerging as the strongest
predictor. Although preliminary, these findings suggest
a possible role of irrational beliefs in substance use
and related problems in adolescents.
The aforementioned studies indicate that some psy-
chiatrically impaired adolescents may present with
cognitive distortions. However, although many investi-
gators and clinicians have initiated cognitive treatment
strategies with adolescent groups, assessment research
has yet to adequately define the nature of the association
between cognitive distortions and various types of
psychiatric disorders. Moreover, there is only a modi-
cum of evidence showing that cognitive distortions are
218
present among psychiatrically disturbed adolescents.
Furthermore, although cognitive techniques are being
applied toward a number of problem areas, such as
depression (Lewinsohn et al., 1990; Reynolds and
Coats, 1986) and conduct disorder (Kendall
et al., 1990), investigative endeavors have yet to deter-
mine whether distortions differ among disparate psychi-
atric diagnostic groups. It is possible that adolescents
with distinct diagnoses will evince different types of
cognitive distortions. If so, forms of cognitive interven-
tions should be tailored to specific populations and
disorders. In light of this hypothesis, an instrument
was chosen that would delineate types of cognitive
errors in addition to substantiating their existence in
this population.
The primary purpose of this study was to provide
some of the first empirical data concerning cognitive
distortions in dually diagnosed adolescents. Further-
more, a variety of diagnostic categories were examined
to ascertain whether there are specific forms or charac-
teristics (type, amount) ofcognitive distortions endemic
to each.
METHOD
Subjects
The present study was part of a comprehensive evaluation of
social and emotional adjustment in dually diagnosed psychiatrically
hospitalized adolescents. The sample included 135 adolescents (78
females and 57 males). The age range of the subjects was 12 to
18 years (X = 15.3). Socioeconomic status (SES), as determined
by the Hollingshead Socioeconomic Status Index (Hollingshead,
1975) ranged from 1 to 5 (X = 1.91), reflecting that most subjects
were from lower socioeconomic levels. Ninety-seven (72%) of the
subjects were white, 34 (25%) were African American, and four
were other. Eighty-five of the participants were admitted to the
Adolescent Drug Abuse and Psychiatric Treatment (ADAPT) pro-
gram, and 50 were admitted to the Services for Teens at Risk
(STAR) Program at the University of Pittsburgh School of Medi-
cine. ADAPT is a psychiatric inpatient unit for adolescents with
a history of substance abuse and a comorbid psychiatric disorder.
STAR is a psychiatric inpatient unit for treatment of depressed or
suicidal adolescents. Length of stay on both units ranges from
several days to several months, with an average of between 4 and
6 weeks. Clinical interviews of the adolescents served as the source
of diagnoses. These interviews were completed by a psychiatrist
experienced in the use of semistructured diagnostic interviews.
Additional diagnostic information from an ongoing NIAAA re-
search project supplemented the clinical interviews in reaching
final diagnoses for each adolescent in the program.
A total of 158 admissions were considered for participation in this
study. Twenty-three subjects were eliminated because of inability to
match diagnostic configurations with current study groups or
noncompliance with assessment procedures. Chi square and analyses
j. AM. ACAD. CHILD ADOLESC. PSYCHIATRY. 33:2, FEBRUARY 1994

of variance were conducted to determine whether these nonpartici-
pants differed in any important way from the participants. No
significant differences emerged on race, gender, age, or SES.
Measures
Children's Negative Cognitive Errors Questionnaire (CNCEQ)
(Leitenberg et al., 1986). The CNCEQ is a 24-item self-report
questionnaire that presents vignettes illustrating four different types
of cognitive distortions (selective abstraction, overgeneralization,
catastrophizing, personalization) in three areas (athletic, social,
academic). The respondent ranks a thought following each vignette
on a five-point Likert scale describing how similar the thought is
to the way they would think in the situation. The Total Subscale,
as well as the subscales of Overgeneralization, Personalization,
Selective Abstraction, and Catastrophizing were used. Test-retest
reliability has been shown to be .69; internal consistency for the
total score is .89 (Leitenberg et al., 1986).
Beck Depression Inventory (BDI) (Beck et al., 1961). The BDl
has been widely used in psychotherapy studies of depression.
This measure is a 2l-item self-report inventory consisting of four
alternative statements scored on a 0 to 3 basis reflecting the range
of symptom severity. Split-half reliability for the BDl is .93 (Beck
er al., 1961). Moreover, the BDI has proved to show high test-
retest reliability as well as high internal consistency in various
studies with adolescents (Strober et al., 1981a,b). The BDI has
been shown to correlate well with ratings of depression made by
independent clinicians in a number of investigations (Beck et al.,
1961; Nussbaum et al., 1963). In addition, the BDI discriminates
well between anxiery and depression (Beck, 1978).
Beck Hopelessness Scale (BHS) (Beck et al., 1974). This 20-item
true/false scale taps the specific cognitive distortion of pessimism
concerning the future. Each item is extracted from affective, motiva-
tional, and cognitive aspects of hopelessness. Split-half reliabiliry
for the BHS has been shown to be .93 (Beck et al., 1974).
The BHS is predictive of repeated suicide attempts and suicide
completion. A score of nine or higher in adult suicidal idearors
has been found to be predictive of suicide.
Procedure
All subjects completed the above-mentioned measures within
the first week of admission (usually on day four) as part of
their comprehensive psychological assessment. Instructions for each
measure were read by a research assistant who was available to
answer any questions subjects may have had. For subjects with
low reading levels, questions were read aloud by the research
assistant, and responses were marked on each form. Completion
of the entire battery required between 2 Y2 and 3 hours. As an
incentive for measure completion, subjects advanced from the first
(evaluation) phase of a mulririer unitwide level (token economic)
system to the second level.
RESULTS
Subjects were divided into groups on the basis of
DSM-III-R (American Psychiatric Association, 1987)
Axis I psychiatric diagnoses. Diagnoses and group
composition were as follows: depression (major depres-
sion, dysthymia, or bipolar disorder) only (DE) (N =
22), conduct disorder only (CD) (N = 18), depression
J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 33:2, FEBRUARY 1994
DISTORTIONS IN DUALLY DIAGNOSED ADOLESCENTS
and substance abuse (DE + SA) (N = 22), conduct
disorder and substance abuse (CD + SA) (N = 37),
all three Axis I diagnoses (CD + SA + DE) (N = 15),
and none of these three Axis I diagnoses (ND) (N =
21). Diagnoses for subjects in the ND group consisted
primarily of adjustment disorder, attention-deficit hy-
peractivity disorder, and no Axis I disorder.
To determine whether groups differed with regard
to relevant demographic characteristics, one-way analy-
ses of variance were conducted. No significant differ-
ences between groups were found for SES, race, or
gender. However, significant differences were found
between groups for age, F(5,137) = 4.30, P < .01,
indicating that the DE + SA and the DE only groups,
were significantly older than the SA, CD, and ND
groups, with the SA + DE group also being significantly
older than the CD + SA + DE group. Means ranged
from 14.73 to 16.26. To ensure that age did not
influence group differences, age was used as a covariate
in subsequent analyses.
Next, because of the correlation shown to exist
between cognitive errors and depression and hope-
lessness, analyses were conducted to ascertain whether
there were group differences on the BDI and the BHS.
No differences emerged for either scale.
Because the Total subscale of the CNCEQ is the
sum of the cognitive errors subscales, this subscale
initially was tested using analysis of covariance; signifi-
cant findings were obtained, F(5,128) = 2.79 P < .05.
Consequently, the individual subscales were analyzed
using a multivariate analysis of covariance; a significant
multivariate analysis of covariance (MANCOVA)
emerged, F(20,512) = 1.64, P < .05. Univariate analy-
ses of variance indicated that three of the four subscales
were significant: Catastrophizing, F(5,128) = 2.89, P
< .05, Overgeneralizing, F(5,128) = 2.55, P < .05, and
Personalizing, F(5,128) = 2.54, P < .05. Differences
between groups approached significance on the Selec-
tive Abstraction subscale, F(5,128) = 2.06, P < .10.
Planned comparisons were conducted on each
CNCEQ subscale to determine which particular diag-
nostic groups differed from each other. For the Caras-
trophizing and Personalizing subscales, the CD + SA
+ DE group significantly differed from each of the
other groups, with the exception of the DE group.
For the Overgeneralizing subscale, the CD + SA +
DE group differed significantly from the CD group
as well as from the ND group. Similarly, the DE group
219
KEMPTON ET AL .
differed significantly from the CD and ND groups.
All differences were significant at the .05 level. For
the Selective Abstraction subscale, the CD + SA + DE
group differed significantly from every other group ,
despite the overall nonsignificance of this subscale in
the MANCOVA (see Table 1 for adjusted means of
each measure).
DISCUSSION
The present investigation provided some of the
first empirical data concerning the nature of cognitive
distortions in psychiatrically impaired adolescents.
Findings indicated that adolescents who suffer from
disparate disorders may actually think in differently
distorted ways. Additionally, adolescents with multiple
problems are correspondingly more likely to commit
cognitive errors. This was true although the subjects
did not report different levels of depression and hope-
lessness on self-report measures. One would expect
that adolescents with different diagnoses, particularly
diagnoses as distinct as conduct disorder and depres-
sion, would be differentiated by self-reported levels of
depression and hopelessness. Furthermore, one may
question the validity of the diagnostic groupings when
no such disparities occur. However, group differences
did emerge when the construct of cognitive errors
was used as the independent variable. Thus, types of
cognitive errors made by diagnostic groups appear
to be more sensitive measures of differences between
adolescents with varying psychiatric problems than
are self-reported levels of depression and hopelessness.
Although a few studies have implicated cognitive distor-
tions as potentially important variables in adolescent
depression (Thurber et al., 1990), the present investiga-
tion provides more conclusive evidence as to the nature
TABLE 1
Adjusted Means for Each Area of Children 's Negative Cognitive Errors Que stionna ire for Each Diagnostic Group
Carasrrophizing
Depr ession (N = 22) 15.36-·b
Conduct disorder (N = 18) 12.75b
Depression and substance abuse (N = 22) 13.n b
Conduct disorderlsbu stance abuse (N = 37) 12.78b
All three diagnoses (N = 15) 18.10-
None of the diagnoses (N = 21) 11.79b
p< .05
Note: Means with different superscripts differ significantly from each other at the .05 level.
220
of their relationship to a variety of psychiatric disorders.
In particular, this effort suggests their relevance beyond
the most commonly assessed symptoms of psychiat-
ric disorders.
The most salient finding in this study concerns the
group of adolescents with multiple diagnoses (conduct
disorder, substance abuse, and depression). For two of
the three significant CNCEQ subscales (Catastrophiz-
ing and Personalizing), this group reported the most
cognitive errors. Subjects in this condition also scored
significantly higher on cognitive errors than did every
other group on the Selective Abstraction subscale and
higher than every other group except the depressed-
only subjects on the Catastrophizing and Personalizing
subscales. Thus, adolescents with problems in multiple
areas have a greater tendency to cognitively distort
than do adolescents who have only single or dual
diagnoses. The exception to this finding , however,
relates to the depressed-only group. Specifically, on
three of the four subscales, the adolescent group with
all three diagnoses did not differ from the depression-
only condition. This finding suggests that the adoles-
cents with the single diagnosis of depression were most
similar to the subjects who were conduct disordered
and abusing substances in addition to their depression.
It appears that depression is the crucial diagnosis for
increasing the likelihood of making cognitive errors.
These data are consonant with previous research docu-
menting the association between cognitive distortions
and depression (e.g., Krantz and Hammen, 1979).
For one particular subscale, Overgeneralizing, a
slightly different pattern was evident. Here, the multiple
diagnosis group differed significantly only from the
groups with either none of the diagnoses or conduct
disorder only. Furthermore, the depression-only group
was also significantly different from the groups with
Overgeneralization Personalizing Selective Abstraction
17.22- 15.87-·b 13.15b
11.78b 13.51b 13.47 b
14.61 -·b 13.37 b 13.84b
13.6yb 13.27 b 13.64b
16.94- 17.62- 17.82-
12.67b 11 .50b 12.21 b
.05 .05 .10
J. AM . AC AD . CHILD ADOLESC. PSYCHI ATRY . 33:2. FEBRUARY 1994

none or conduct disorder only diagnoses. Again, the
depression-only group most closely resembled the mul-
tiple diagnoses group. However, the dually diagnosed
adolescents fell between the none and conduct disorder
groups on the one hand, and the depression and
multiple diagnosis groups on the other, and differed
significantly from neither.
For Overgeneralizing and all other subscales, the
dually diagnosed adolescents did not report higher
levels of cognitive errors than did the group with none
or conduct disorder only. Whereas one may not expect
an increase in cognitive errors among youth with
conduct disorder and substance abuse, the similar find-
ing with respect to the depression and substance abuse
group warrants attention. That is, adolescents who
have concurrent problems with depression and sub-
stance abuse do not appear to be as prone to cognitive
errors as do their peers who are only depressed. Perhaps
the depression in the dually diagnosed group is best
characterized as secondary depression. Several investiga-
tors (e.g. Schuckit, 1983; Weissman et al., 1977) have
examined the phenomenology of depression when it
is secondary to substance abuse in adults with equivocal
findings. Schuckit (1983) reported few minor differ-
ences between primary and secondary depressives,
whereas Weissman et al. (1977) showed that secondary
depressives were generally less depressed and less symp-
tomatic on other dimensions. The present findings
suggest that the diagnosis of depression in the dually
diagnosed adolescents in this study may be secondary
to the substance abuse and clinically distinct from
primary depression. Unfortunately, this study did not
distinguish primary versus secondary depression.
The existence of cognitive distortions and their evo-
lution over the course of normal and abnormal develop-
ment is an area that has not been adequately examined.
Although the adolescents in this study showed varying
degrees and types of distorting, it would be useful to be
able to compare their scores with those of a normative
sample. The closest comparison group available consists
of the 8th graders in a sample investigated by Leitenberg
et al. (1986), where subscale score averages ranged
from 13.12 to 15.13. Thus, it is likely that only the
DE and the DE + SA + CD groups in the present
study were distorting at a significantly higher rate than
their normative peers. Clearly, the absence of a normal
control group in this study represents a limitation;
more information in this area is needed.
j. AM. ACAD. CHILD ADOLESC. PSYCHIATRY. 33:2. FEBRUARY 1994
DISTORTIONS IN DUALLY DIAGNOSED ADOLESCENTS
The present investigation also is limited by the use
of a single, inhospital sampling of cognitive errors. If
mood and cognitions are indeed related, it would be
beneficial in future research to gather cognitive data
over multiple sessions. In addition, various types and
degrees of substance abuse were combined in the cur-
rent sample. It is possible that differences in thinking
styles exist between, for example, alcohol, cocaine,
and heroin abusers. Nevertheless, the present findings
provide preliminary evidence of the potentially im-
portant role of cognitive styles in psychiatrically im-
paired adolescents.
Clinical Implications
As research has identified the presence of cognitive
errors and distortions in adolescents, there is a potential
for the effectiveness of cognitive-based therapies for
adolescents (Kendall, 1991). Although cognitive distor-
tions have been identified in specific diagnostic groups
of adolescents (i.e., those with depression or conduct
disorder) and research has been conducted on the
efficacy of cognitive therapy applied to these specific
diagnostic groups, little consideration has been given
to cognitive errors in adolescents with comorbid or
multiple, coexisting diagnoses who comprise a sizable
proportion of youth in treatment (Bukstein and Van
Hasselt, 1993; in press). The clinical implications of
the findings of this study include not only the necessity
of considering assessment of cognitive distortions as
part of a comprehensive evaluation of an adolescent
with one or more psychiatric disorders but also assess-
ment of the type of distortion within the context
of the psychiatric diagnosis. Examination of possible
differences in cognitive distortions offers the potential
of individualizing cognitive approaches. Perhaps the
types or pattern of cognitive distortions displayed by
the adolescent may prove to be a better guide to
treatment selection than are categorical diagnoses.
The importance of depression as a constituent diag-
nosis of the multiple diagnosed group or alone in the
depression only group is manifested by these groups
having the most cognitive errors. Therefore, the pres-
ence of depression, whether alone or comorbid with
nonaffective diagnoses should direct the clinician to
carefully assess cognitive errors or distortions and con-
sider cognitively based therapies as part of the treatment
plan. On the other hand, dually diagnosed adolescents
221
KEM PT ON ET AL .
with depression and substance abuse may be clinically
distinct from depressed adolescents with depression
only or those with depression, substance abuse, and
conduct disorder. In assessing cognitive errors, the
identification of significant cognitive errors in the pres-
ence of depression, regardless of the coexisting diagno-
sis, may serve to alert the clinician that this depression
may be similar to a pure or primary depression and
perhaps responsive to similar cognitive interventions.
Thus, all adolescents presenting with nonaffective psy-
chopathology should not only be assessed for the pres-
ence of depression bur also for cognitive errors in
attempting to identify salient (homogenous) subgroups
for treatment planning.
REFERENCES
Abramson LY, Seligman MEP, Teasdale ] (1978), Learned helplessness in
hum ans: cririque and reformularion. J Abnorm PsychoI 87:49-74
American Psychiatric Association (1987), Diagnostic and Statistical Man ual,
3rd Edition- Revised(DSM-III-R). Washington, DC: American Psychi-
atric Association
Beck AT (1967), Depression: Clinical Experimental and Theoretical Aspects.
New York: Hoeber Med ical Division, Harper & Row
Beck AT (1976), Cognitive Therapy and the Emotional Disorders. New York:
International Universities Press
Beck AT (1978), Depression Inventory. Philadelphia: Center for Cogni-
tive T herapy
Beck AT, Freeman AM (1990), Cognitive Therapy of Personality Disorders.
New York: Gu ilford Press.
Beck AT, Rush A] , Shaw BF, Emery G (1979), Cognitive Therapy of
Depression, New York: Th e Guilford Press
Beck AT , Ward CH , Mendelsohn M, Mock ] E, Erbaugh ] (1961) . An
inventory for measuring depression. Arch Gen Psychiatry, 4:56 1-57 1
Beck AT, Weissman A, Lester D, Texler L (1974), Th e measurement of
pessimism: the Hopelessness Scale. J Consult Clin PsychoI42:861- 865
Beidel D, Turner S (1986), A critique of the theoretical bases of cognitive-
behavioral theories and therapy. Clin Psychol Rev 6:177-1 97
Bukstein OG , Van Hasselt VB (1993), Alcohol and drug abuse. In:
Handbook of Behavior Therapy in the Psychiatric Setting, eds, Bellack
AS, Hersen M . New York: Plenum Press
Buksrein OG , Van Hasselt VB (in press). Substance use disorders. In:
Handbook of Adolescent Psychopathology: A Guide to Diagnosis and
Treatment, eds, Van Hasselr VB, Hersen M. Boston, MA: Lexington
Books
Carlson GA, Cantwell DP (1980), A survey of depressive symptoms in a
child and adolescent psychiatric populat ion : interview data. JA m Acad
Child Psychiatry 18:587-599
Carlson GA, Garber] (1986), Developmental issues in the classification
of depression in children. In: Depression in Young People: Developmental
and Clinical Perspectives, eds, Rutt er M, Izard CE, Read PB. New York:
Gu ilford Press, pp. 399-434
Chambers W), Puig-Antich ] , Hirsch M, er al. (1985), Th e assessment of
affective disorders in children and adolescents by sernistructural inter-
view: test-retest reliability of the Schedule for Affective Disorders and
Schizophrenia for school-age children, present episode version. Arch
Gen Psychiatry 42:696-702
Coyne ]C, Gotlib IH (1983), T he role of cognition in depression: a critical
appraisal. Psychol Bull 94:472-505
222 j . AM . AC AD . CHILD ADOL ESC . PSYCH IATR Y, 33:2 . FEBRUARY 19 94
Denoff MS (1987), Irrational beliefs as predictors of adolescent drug abuse
and runn ing away. J Clin PsychoI 43:412-423
Denoff MS (1988), An integrated analysis of the contribution made by
irrational beliefs and parental interaction to adolescent drug abuse. Int
j Addict 23:655- 669
DevereuxE, Bronfenbrenner U, Rodgers R (I 969), Child- rearing in England
and the United States: a cross-national comparison. J Marr Family
3 1:257- 270
Ellis A (1962), Reason and Emotion in Psychotherapy. New York: L. Stuart
Foreyt ]P , Rathjen DP (1978), Cognitive Behavior Therapy: Research and
Application. New York: Plenum Press
Haley GMT, Fine SV, Marriage K, Moretti MM , Freeman R] (1985),
Cognitive bias and depression in psychiatrically disturbed children and
adolescents. j Consult Clin PsychoI 53:535-53 7
Hollingshead AF (1975), Four-Factor Index of Social Status. New Haven,
CT: Yale University Sociology Department
Kandel D (1975), Sequence and stages in patterns of adolescent drug use.
Arch Gen Psychiatry 32:923- 932
Kazdin AF, Rogers A, Colbus D (1986), The Hopelessness Scale for
Children : psychometric characteristics and concurrent validity. j Consult
Clin PsychoI 54:241-245
Kendall PC, ed (1991), Child and Adolescent Therapy: Cognitive-Behavioral
Procedures. New York: Guilford Press
Kendall PC, Reber M, McLeer S, Epps ] , Ronan K (1990), Cogn itive-
behavioral treatment of conduct-disordered children. Cognitive Therapy
and Research 14:279-297
Kovacs M (1981), Rating scales to assess depression in school-aged children.
Acta Paedopsychiatr 46:305- 3 15
Krantz S, Hamm en C (1979), Assessment of cognitive bias in depression.
j Abnorm PsychoI8 8:6 11-6 19
Lang M, Tisher M (1978), Children's Depression Scale. Victoria, Australia:
Australian Council for Educational Research
Leitenberg H, Yooc L, Carroll-Wilson M (1986), Negative cognitive errors in
children: questionnaire development, normative data, and comparisons
between children with and without self-reported symptoms of depres-
sion, low self-esteem, and evaluarion anxiety. j Consult Clin Psychol
54:528- 536
Lewisohn PM, Clarke GN , Hops H , Andrews] (1990), Cogn itive-behavioral
treatment for depressed adolescents. Behau Ther 21:385-4 01
Nelson RE, Craighead WE (1977), Selecrive recall of positive and negative
feedback, self-control behaviors, and depression. J Abnorm Psycho!
86:379-388
Nussbaum K, Wittig BA, Hanlon TE, Kurland AA (1963), Intravenous
nialamide in the treatment of depressed female patients. ComprPsychia-
try 4:105-1 16
Reynolds W , Coats K (1986), A comparison of cognitive-behavioral therapy
and relaxation training for the treatment of depression in adolescents.
j Consult Clin Psychology 54:653-6 60
Ryan N, Puig-Antich ] , Ambrosini P er aI. (1987), The clinical picture of
major depression in children and adolescents. Arch Gen Psy chiatry
44:854-860
Schuckit M (1983), Alcoholic patients with secondary depression Am J
Psychiatry 140:711-714
Shorkey C, Whit eman V (1977), Development of the Rational Behavior
Inventory: initial validity and reliability. Educational and Psychological
Measurement 37:527- 533
Strober M, Green ] , Carlson G (l981a), Utility of the Beck Depression
Inventory in hospitalized adolescents. J Consult PsychoI 49:482-483
Strober M, Green], Carlson G (1981b), Reliability of psychiatric diagnosis
in hospitalized adolescents. Arch Gen Psychiatry38:14 1- 145
Thurber S, Crow LA, Thurber ]A, Woffington LM (1990), Cognitive
distortions and depression in psychiatrically disturbed adolescent inpa-
tients. J Clin PsychoI46:57-60
Weissman M, Potten ger M, Kleber H, Ruben H , Williams D, Thompson
WD (1977), Symptom pattern s in primary and secondary deptession.
Arch Gen Psychiatry 34:854- 862