Ij

This material is protected by U.S. copyright law. Unauthorized reproduction is prohibited.
To purchase reprints or request permission to reproduce, e-mail reprints@ons.org.
ONLINE EXCLUSIVE
Behavioral Adjustment of Children
and Adolescents With Cancer:
Teacher, Parent, and Self-Report
Ida M. (Ki) Moore, RN, DNS, FAAN, Julia Challinor, RN, PhD, Alice Pasvogel, RN, PhD,
Katherine Matthay, MD, John Hutter, MD, and Kris Kaemingk, PhD
Key Points . . .
Purpose/Objectives: To describe behavioral adjustment in children
and adolescents with acute lymphoblastic leukemia (ALL) and to deter-
mine whether behavioral adjustment is correlated with cognitive and
Children and adolescents with acute lymphoblastic leukemia
academic abilities.
experience specific behavioral adjustment problems, including
Design: Descriptive, cross-sectional design.
somatization; attention, adaptability, and learning problems;
Setting: Two pediatric oncology treatment centers.
and anxiety.
Sample: 47 children and adolescents who had been receiving ALL
therapy for at least one year or who were off therapy for no more than
Children and adolescents who experience central nervous sys-
three years and their parents and teachers. Wechsler Intelligence Scale
tem treatment-related cognitive and academic problems may
for Children­Revised (WISC-R) and Wide Range Achievement Test­
experience behavioral adjustment problems.
Revised (WRAT-R) data were available on a subset of 17 subjects.
Body image alterations during treatment may increase the risk
Methods: Parent, teacher, and self-report Behavioral Assessment
for behavioral adjustment problems.
System for Children (BASC) ratings were used to measure behavioral
adjustment. WISC-R measured cognitive abilities, and WRAT-R mea-
Ongoing assessment of specific areas of behavioral adjust-
sured academic abilities. Demographic, family, and treatment-related
ment is warranted, and interventions that target these at-risk
data also were collected.
areas are needed.
Main Research Variables: Behavioral adjustment and cognitive and
academic abilities.
Findings: At least 20% of teacher ratings for somatization, learning
problems, leadership, and study skills; parent ratings for somatization,
ne in every 333 children in the United States develops
O
adaptability, attention problems, withdrawal, anxiety, social skills, and
cancer before the age of 20, which translates into
depression; and self-report ratings for anxiety and attitude to school
12,400 new cases of cancer each year (Ries et al.,
were in the at-risk range (i.e., presence of significant problems that re-
1999). Acute lymphoblastic leukemia (ALL), the most com-
quire treatment). The majority of teacher BASC ratings were correlated
mon pediatric tumor, comprises 2,400 of these 12,400 new
significantly with WISC-R and WRAT-R scores. Self-report depression and
social stress ratings were correlated significantly with some WISC-R and
WRAT-R scores. Treatment-related experiences such as body image
Ida M. (Ki) Moore, RN, DNS, FAAN, is a professor and a director
alterations and mental and emotional problems were associated with
in the Division of Nursing Practice for the College of Nursing at the
problematic behaviors, including depression, somatization, withdrawal,
University of Arizona in Phoenix; Julia Challinor, RN, PhD, is an
and social stress.
assistant adjunct professor in the School of Nursing at the Univer-
Conclusions: Youth with ALL are at risk for some behavioral adjust-
sity of California, San Francisco (UCSF); Alice Pasvogel, RN, PhD,
ment problems, particularly anxiety, somatization, adaptability, atten-
is a senior research specialist in the College of Nursing at the Uni-
tion, and withdrawal. Cognitive and academic abilities are associated
versity of Arizona; Katherine Matthay, MD, is a professor in the
with some dimensions of behavioral adjustment.
School of Medicine at UCSF; John Hutter, MD, is a professor and
Implications for Nursing: Findings suggest the need for ongoing as-
Kris Kaemingk, PhD, is an assistant professor, both in the College
sessment of behavioral adjustment and cognitive and academic abilities
of Medicine at the University of Arizona. This study was supported
of children with ALL. Behavioral interventions that target at-risk manner-
by the National Institutes of Health (NR02557, NR04905, and
isms, such as somatization, depression, anxiety, and social stress, are
HD37816) and by the ONS Foundation and Ortho Biotech Products.
needed. Central nervous system treatment may contribute to behavioral
Additional support was provided by the Frank A. Campini Founda-
adjustment problems, as well as to cognitive and academic problems.
tion and the Bay Echo Foundation of San Francisco. (Submitted Sep-
Strategies to improve academic abilities also may have a positive effect
tember 2002. Accepted for publication March 7, 2003.)
on behavioral adjustment.
Digital Object Identifier: 10.1188/03.ONF.E84-E91
ONCOLOGY NURSING FORUM ­ VOL 30, NO 5, 2003
E84
cancer cases. Five-year disease-free survival rates for children
According to Merrell (2000), "Behavior-rating scales provide
with cancer are 77% overall and 85% for patients with ALL
summative judgments of general types of behavioral character-
(Jemal et al., 2003). Using current incidence and survival rates
istics that may have occurred in a variety of settings and over
among children with cancer in the United States alone, ap-
a long period of time" (p. 204). Behavioral-rating scales with
proximately 400,000 human years are being saved annually;
forms for multiple respondents allow researchers to identify
by 2010, 1 in 250 young adults will be a childhood cancer
problematic behavior under specific conditions. Significant
survivor (Bleyer, 1993, 1995). Aggressive therapies have led
behavior problems tend to be expressed consistently in differ-
to dramatic improvements in survival, but they also have
ent surroundings or situations and with different measurement
raised concerns about the impact of pediatric cancer and its
tools (Merrell). Self-report scales complement informant be-
treatment on quality of life and behavioral adjustment.
havior scales and typically measure children's and adolescents'
The primary aim of this study was to describe the behav-
emotional or behavioral adjustment in domains such as internal-
ioral adjustment of children and adolescents with ALL using
izing problems, externalizing problems, or school maladjust-
teacher, parent, and self-report measures. The intent was to
ment (Eckert, Dunn, Codding, & Guiney, 2000).
measure adaptive and problematic behaviors. All subjects
Methods
received some type of central nervous system (CNS) treat-
ment to decrease the risk of meningeal disease. The late ef-
Design and Sample
fects of CNS treatment on cognitive and academic abilities
A descriptive, cross-sectional design was used. To be eligible
have been reported since the 1980s (Kramer & Moore, 1989;
to participate, patients had to be receiving ALL therapy for at
Moore, Glasser, & Ablin, 1988; Moore, Kramer, & Ablin,
least one year or be off therapy for no more than three years,
1986). Survivors of ALL who experience CNS late effects
report no CNS or bone marrow relapse, and be able to speak
also may be at risk for behavioral adjustment problems. How-
English. Children and adolescents who had not completed their
ever, little is known about potential links between cognitive
first year of treatment were excluded to minimize the potential
or academic abilities and behavioral adjustment. Therefore, a
confounding effects of acute physical and emotional stress as-
secondary aim of the study was to investigate relationships be-
sociated with a new cancer diagnosis and the consolidation
tween behavioral adjustment scores and cognitive and aca-
phase of therapy. Individuals with CNS or bone marrow relapse
demic abilities in a subset of subjects for whom these data
were excluded because of more aggressive therapy or decreased
were available (Moore et al., 2000).
probability for long-term survival. All eligible children and
adolescents with ALL and their parents were recruited from two
Background and Literature Review
pediatric oncology treatment centers. The goal was to obtain a
Deficits in academic abilities following CNS treatment with
representative sample of children and adolescents with the same
whole-brain radiation have been well established (Appleton,
cancer diagnosis who had received relatively similar treatment.
Farrell, Zaide, & Rogers, 1990; Cousens, Ungerer, Crawford,
Institutional review committee approvals were obtained prior to
& Stevens, 1991; Silber et al., 1992). Because of the over-
subject recruitment and data collection.
whelming evidence that whole-brain radiation causes signifi-
Instruments
cant neurotoxicity, the majority of current CNS treatment
regimens for children with ALL include intrathecal chemo-
The Behavioral Assessment System for Children
therapy alone or in combination with intermediate- to high-
(BASC) (Reynolds & Kamphaus, 1992) was used to measure
dose systemic methotrexate. Findings suggest that chemo-
behavioral adjustment. BASC is a reliable and valid measure
therapy-based CNS treatment regimens also are associated
consisting of scales that provide a comprehensive behavioral
with academic problems (Armstrong, Blumberg, & Toledano,
assessment of children and adolescents (see Table 1). BASC
1999; Mulhern, Armstrong, & Thompson, 1998), especially
was selected because it provides an assessment of risk (inter-
mathematics (Brown et al., 1996; Copeland, Moore, Francis,
nalizing problems, externalizing problems, school problems,
Jaffe, & Culbert, 1996; Moore et al., 2000).
and behavioral symptoms) as well as positive factors (adap-
tive skills). According to Merrell (2000), "As a thorough and
Behavioral and emotional problems, including withdrawal,
depression, anxiety, and attention problems, have been re-
comprehensive system of behavior-rating scales, the BASC is
ported among children with ALL (Anderson, Smibert, Ekert,
representative of the best of what is currently available" (p.
& Godber, 1994; Fossen, Abrahamsen, & Storm-Mathisen,
215). BASC teacher, parent, and child and adolescent self-re-
1998; Sharan, Mehta, & Chaudhry, 1999). Several studies
port forms were used in this study. Content validity was estab-
determined that long-term survivors of childhood cancer ex-
lished by expert judgment. Construct validity was established
perience a greater number of problems with social compe-
by determining factor structure of the scales and by correla-
tence and more symptoms of depression compared to healthy
tions with other behavioral-rating scales, including the Child
children and siblings (Cavusoglu, 2001; Olson, Boyle, Evans,
Behavior Checklist, Personality Inventory for Children­Re-
& Zug, 1993; Pendley, Dahlquist, & Dreyer, 1997; Sharan et
vised, Conners' Parent-Rating Scales, and the Minnesota
al.).
Multiphasic Personality Inventory. Reliability was established
Behavioral-rating scales and systematic observation have
by tests of internal consistency, test-retest reliability, and
been used since the 1980s to assess behavioral adjustment;
interrater reliability (Reynolds & Kamphaus).
however, psychometric testing of instruments was limited
The BASC teacher- and parent-rating scales include items
(Shapiro & Kratochwill, 2000). More recently, comprehen-
regarding observations of negative and positive behavioral
sive psychometric testing of instruments has been performed,
performance. Separate forms target three age levels: pre-
and several excellent nationally standardized measures for
school (4­5 years), child (6­11 years), and adolescent (12­
assessing behavior in children and adolescents are available.
18 years). Therefore, BASC is developmentally sensitive and
ONCOLOGY NURSING FORUM ­ VOL 30, NO 5, 2003
E85
Table 1. Behavioral Assessment System for Children Composite and Scale Definitions
Score Indicating At-Risk Statusa
Scale
Definition
The ability to adapt readily to changes in the environment
< 40
Adaptability
Anxiety
Feelings of nervousness, worry, and fear; the tendency to be overwhelmed by problems
> 60
Attention problems
The tendency to be distracted easily and unable to concentrate more than momentarily
> 60
Feelings of alienation, hostility, and dissatisfaction with school
> 60
Attitude to school
Attitude to teachers
Feelings of resentment and dislike of teachers; beliefs that teachers are unfair, uncaring,
> 60
or overly demanding
Feelings of unhappiness, sadness, and stress that may result in an inability to carry out
Depression
> 60
everyday activities or bring thoughts of suicide; belief that nothing goes right
Leadership
The skills associated with accompanying academic, social, or communication goals, par-
< 40
ticularly the ability to work well with others
Learning problems
The presence of academic difficulties, particularly in understanding or completing school-
> 60
work
Feelings of self-esteem, self-respect, and self-acceptance
Self-esteem
< 40
Sense of inadequacy
Perceptions of being unsuccessful in school, unable to achieve goals, and generally in-
> 60
adequate
Social skills
The skills necessary for interacting successfully with peers and adults in home, school,
< 40
and community settings
Social stress
Feelings of stress and tension in personal relationships, a feeling of being excluded from
> 60
social activities
Somatization
The tendency to be overly sensitive to and complain about relatively minor physical prob-
> 60
lems and discomforts
Study skills
The skills that are conducive to strong academic performance, including organizational
< 40
skills and good study habits
Withdrawal
The tendency to evade others to avoid social contact
> 60
a
Score is more than one standard deviation, which indicates at-risk status.
istics; T scores greater than 60 (one or more standard devia-
has normative data for males and females in each of the three
tion) are in the at-risk range. At risk is used to indicate the
age levels. The parent and teacher forms provide composite
presence of significant problems that require treatment but
scores for adaptive skills, externalizing problems, internaliz-
may not be severe enough to warrant a formal diagnosis. A
ing problems, and behavioral symptoms. The teacher form
score in the at-risk range may signify potential or developing
also includes a school problems composite. Composite scores
problems that require careful monitoring (Reynolds & Kamp-
are based on relevant scales. The internalizing composite is
haus, 1992). Higher scores on scales that measure personal
comprised of anxiety, depression, and somatization scales; the
adjustment (parent and teacher forms) or adaptive skills (child
externalizing composite is comprised of aggression, hyperac-
or adolescent form) represent positive or desirable character-
tivity, and conduct problems scales. The behavioral symptoms
istics. For these scales, T scores less than 40 (one or less stan-
index is a measure of the overall level of problem behavior
dard deviation) are considered at risk.
and is based on scores from the aggression, hyperactivity,
Demographic (i.e., date of birth, gender, ethnicity, grade in
anxiety, depression, attention problems, and atypicality scales.
school) and treatment-related (i.e., diagnosis, diagnosis date,
All scale, composite, and index scores are T scores with a
CNS treatment) data were obtained for all subjects. Parents
mean of 50 and a standard deviation of 10.
were asked whether their family had experienced residence
The child and adolescent self-report of personality has
changes, family problems, the death of a friend or family
forms at two age levels: child (8­11 years) and adolescent
member, divorce, serious illness, or job loss during the past
(12­18 years). Items elicit information regarding children's or
year because these experiences could influence behavioral
adolescents' self-perception and emotional status. The self-
adjustment. Parents were asked about treatment-related
report form provides composite scores for personal adjust-
problems experienced by their child or adolescent. Treatment-
ment, clinical maladjustment, school maladjustment, and the
related problems included mental or emotional problems dur-
emotional symptoms index. Each composite is made up of
ing treatment, body image alterations during treatment, and
relevant scales. The school maladjustment composite includes
energy to stay in school.
attitude to school, attitude to teachers, and sensation-seeking
Data about general intellectual and academic abilities were
scales; the clinical maladjustment composite is based on anxi-
available for a subset of subjects (n = 17) who also partici-
ety, atypicality, locus of control, social stress, and somatiza-
pated in a study of the cognitive and academic effects of CNS
tion scales. The emotional symptoms index is comprised of
treatment for childhood ALL (Moore et al., 2000). Intellectual
scores from social stress, anxiety, interpersonal relations, self-
abilities were assessed using the Wechsler Intelligence Scale
esteem, depression, and sense of inadequacy scales. These
for Children­Revised (WISC-R) (Wechsler, 1974). Overall
scale, composite, and index scores are T scores with a mean
intellectual abilities were measured by the WISC-R Full Scale
of 50 and a standard deviation of 10.
Intelligence Quotient (FSIQ). Verbal intellectual skills were
Higher scale, composite, and index scores for negative be-
measured by the WISC-R Verbal IQ (VIQ), and nonverbal or
haviors (e.g., internalizing problems, externalizing problems,
performance-based intellectual abilities were measured by the
school problems) represent negative or undesirable character-
ONCOLOGY NURSING FORUM ­ VOL 30, NO 5, 2003
E86
WISC-R Performance IQ (PIQ). IQ scores have a mean of 100
Approximately 22% of children and adolescents received
and a standard deviation of 15. The Wide Range Achievement
teacher ratings on the internalizing problems composite in the
Test­Revised (WRAT-R) was used to assess academic
at-risk range. This finding was primarily a result of ratings on
--
achievement (Jastak & Wilkinson, 1984). Standard scores (X =
the somatization scale because 39% of children and adolescents
100, SD = 15) from the reading, spelling, and arithmetic
received at-risk scores. The adaptive skills composite summa-
subtests were used. The number-questions subscale of the
rizes prosocial, organizational study, and other adaptive skills
McCarthy Scales of Children's Abilities (McCarthy, 1972)
(Reynolds & Kamphaus, 1992). According to teacher ratings,
was used to measure emergent math skills in children younger
approximately 18% of patients with ALL received at-risk rat-
than the age of five at the initial (seven months after ALL di-
ings on the adaptive skills composite, particularly in the areas
agnosis) evaluation.
of leadership and study skills (21% and 24% at-risk scores,
respectively). Finally, teachers rated 25% of subjects in the at-
Data Analysis
risk range for learning problems. According to Reynolds and
BASC, demographic, treatment, treatment-related prob-
Kamphaus, a score of 60 or higher on the learning problems
lems, and family data were analyzed using descriptive statis-
scale indicates a need for careful investigation of academic
tics (mean, standard deviation). Percentages were used to
skills.
summarize teacher, parent, and self-report scores in the at-risk
According to parent ratings, 44% of their children or ado-
range. Pearson correlation was used to examine relationships
lescents were at risk for internalizing problems and 29% were
between behavioral adjustment and intellectual and academic
at risk for adaptive skills problems. Only 17% of children and
abilities measured 45 months after ALL diagnosis; Pearson
adolescents received high scores on the behavioral symptoms
correlation also was used to examine relationships between
composite, which is an indication of the overall level of prob-
behavioral adjustment and family or treatment-related prob-
lematic behavior. At least 20% of scores on 7 of the 12 par-
lems. Level of significance was set at p = 0.05.
ent-rating scales (somatization, adaptability, attention prob-
lems, withdrawal, anxiety, social skills, and depression) were
Results
in the at-risk range.
Fewer child and adolescent self-report scores were in the at-
Sample
risk range. Approximately 16% of self-report scores for the
The final sample was comprised of 47 children with ALL.
clinical maladjustment composite, emotional symptoms in-
Informed consent and assent were obtained from subjects and
dex, and school maladjustment composite were 60 or greater.
parents; written permission to contact teachers for data collec-
Only 4 (anxiety, somatization, attitude to school, and self-es-
tion also was obtained. The average age of subjects at the time
teem) of the 14 self-report scales had at least 20% of scores in
of data collection was 9 years, 11 months (range = 5.2­16
the at-risk range.
years), and the mean length of time since diagnosis was 40.4
Behavioral Adjustment and Cognitive Abilities
months (SD = 21.8). The majority of children were Caucasian
(60%); other ethnic groups were Hispanic (27%), Filipino
Mean FSIQ, PIQ, and VIQ scores obtained 7 and 45 months
(4%), African American (2%), Asian (2%), and other (5%).
after ALL diagnosis were within one standard deviation of the
mean (85­115) for children who received triple intrathecal
Fifty-six percent of the subjects were female. CNS treatment
consisted of triple intrathecal chemotherapy (methotrexate,
chemotherapy or intrathecal methotrexate (Moore et al.,
cytosine arabinoside, and hydrocortisone) (n = 31), intrathe-
2000). FSIQ, PIQ, and VIQ scores obtained from the child
cal methotrexate (n = 15), or intrathecal chemotherapy and
who received cranial radiation were much lower and had de-
whole-brain radiation (n = 1). The majority of parents had
clined dramatically from the initial evaluation completed prior
completed high school, and approximately 50% of mothers
to radiation treatment (126 to 71 for FSIQ, 109 to 69 for PIQ,
and 114 to 77 for VIQ). Table 3 summarizes the correlations
and fathers had undergraduate, graduate, or postgraduate col-
lege education.
between cognitive abilities and behavioral adjustment.
Data regarding intellectual abilities and academic achieve-
FSIQ, PIQ, and VIQ scores were correlated significantly
ment were obtained 7 and 45 months after ALL diagnosis on
with teacher ratings on learning problems, attention problems,
a subset of 17 children (10 females and 7 males). CNS treat-
leadership, and study skills scales. Correlations ranged from
ment for this subset included triple intrathecal chemotherapy
r = ­0.78 (for scales with maladaptive behaviors) to r = 0.87
(n = 9) or intrathecal methotrexate (n = 8); one child received
(for scales with adaptive behaviors). Based on these signifi-
24 Gy of whole-brain radiation in combination with intrathe-
cant correlations, the finding that the school problems com-
cal methotrexate. The mean age of this subset of participants
posite, which reflects academic problems including problems
was 10.4 years (range = 6.7­16 years), and the average length
of motivation, attention, learning, and cognition, was corre-
of time since diagnosis was 54.7 months (SD = 19.1). All of
lated strongly with FSIQ, PIQ, and VIQ scores was not sur-
the fathers and 71% of mothers had graduated from high
prising. The adaptive skills composite was correlated signifi-
school; 64% of fathers and 50% of mothers had undergradu-
cantly with FSIQ and PIQ scores but not with VIQ scores (see
ate, graduate, or postgraduate college education.
Table 3).
Parent ratings of attention problems were significantly cor-
Behavioral Adjustment
related with FSIQ, PIQ, and VIQ scores. FSIQ and PIQ scores
were correlated highly with the social skills scale and the
BASC teacher, parent, and child or adolescent composite
adaptive skills composite. Only the child and adolescent self-
and index scores are summarized (mean, standard deviation,
report depression and social stress scale scores were correlated
and range) in Table 2. Scale scores in which 20% or more of
strongly with FSIQ scores (r = ­0.62 and ­0.67, respectively)
T scores fell in the at-risk range (60 or more for maladaptive
and PIQ scores (r = ­0.74 and ­0.87, respectively); however,
and 40 or less for adaptive behaviors) also are included.
ONCOLOGY NURSING FORUM ­ VOL 30, NO 5, 2003
E87
Table 2. Teacher, Parent, and Self-Report Behavioral Assessment System for Children Composite and Scale Scores
--
Scale
% at Risk
SD
Range
X
Teacher-rating scale
Internalizing problems composite
22
54.00
19.96
39­791
Behavioral symptoms index
16
49.08
18.23
38­761
Adaptive skills composite
18
51.00
10.08
33­711
School problems composite
13
50.37
19.28
33­711
Externalizing problems composite
10
48.55
18.68
41­861
Somatization scale
39
58.05
15.06
42­111
Learning problems scale
26
51.45
19.65
35­681
Study skills scale
24
50.11
10.25
29­701
Leadership scale
21
49.47
19.26
34­691
Parent-rating scale
Internalizing problems composite
44
58.69
15.46
32­971
Adaptive skills composite
29
48.11
10.65
30­721
Behavioral symptoms index
17
50.81
11.38
31­861
Externalizing problems composite
13
48.56
10.59
32­871
Somatization scale
49
62.03
16.08
37­120
Withdrawal scale
34
55.15
11.90
34­851
Adaptability scale
31
45.35
12.74
24­701
Attention problems scale
28
52.26
10.68
33­761
Anxiety scale
28
53.77
12.50
33­901
Social skills scale
24
49.84
10.63
29­711
Depression scale
23
53.41
14.28
34­100
Child and adolescent self-report
Clinical maladjustment composite
17
48.37
11.07
34­731
Emotional symptoms index
17
48.90
18.68
36­671
School maladjustment composite
16
47.77
10.89
34­791
Personal adjustment composite
10
50.80
19.03
17­611
Anxiety scale
29
48.42
11.53
33­691
Somatization scale
23
51.92
13.22
39­821
Self-esteem scale
23
48.77
10.31
24­591
Attitude to school scale
22
49.09
10.81
36­761
spelling (r = ­0.61). Importantly, negative correlations were
not all correlations achieved statistical significance because of
sample size limitations.
found between academic abilities and child and adolescent
self-report scores on social stress, depression, and attitude to
Behavioral Adjustment and Academic Abilities
school. However, only correlations between social stress and
A significant decline in academic arithmetic was found
reading (r = ­0.81) and spelling (r = ­0.84) achieved statisti-
cal significance.
from the 7- to the 45-month evaluation for children who re-
--
ceived triple intrathecal chemotherapy (X decline = 10 points)
Family Problems and Treatment-Related Problems
--
and for those who received intrathecal methotrexate (X de-
Seventeen of the 47 respondents reported experiencing no
cline = 8.9 points). Mean reading and spelling scores at the
45-month evaluation were 92.3 (± 14.2) and 93.3 (± 18.4), re-
family problems during the prior year; one family did not re-
spond to these questions. Seventeen families experienced at
spectively, in the triple intrathecal chemotherapy group and
95.5 (± 16.7) and 94.1 (± 12.8), respectively, in the intrathe-
least one problem, including residence change, death of a fam-
ily member or friend, divorce, serious illness, or job loss. Eight
cal methotrexate group. Reading, spelling, and arithmetic
scores obtained at the 45-month evaluation were much lower
families experienced two of these problems, and three families
(55 in all areas) for the child who received intrathecal methotr-
reported experiencing three problems. According to parents,
exate and whole-brain radiation. Table 4 summarizes correla-
only two (9%) children did not have energy to stay in class.
tions between BASC scale or composite scores and WRAT-R
However, a greater percentage experienced physical problems
(n = 14, 30%), mental or emotional problems (n = 12, 26%),
standard scores.
WRAT-R reading, spelling, and math scores were corre-
and body image alterations (n = 26, 55%) during treatment. A
lated negatively with teacher ratings of learning problems and
"yes" response to mental and emotional problems during treat-
ment was correlated significantly with BASC teacher or parent
positively correlated with teacher ratings of leadership. Aca-
demic abilities (especially reading and spelling) were corre-
ratings of depression, learning problems, attention problems,
lated with teacher ratings of study skills, attention problems,
withdrawal, leadership, and study skills (r = 0.32­0.48, p =
school problems, adaptive skills, and behavioral symptoms. In
0.05). A "yes" response to body image alterations during treat-
general, a trend existed for correlations between parent ratings
ment was correlated significantly with child and adolescent
of behavioral adjustment and academic abilities. However, the
self-ratings on somatization, social stress, depression, interper-
correlations did not achieve statistical significance with the
sonal relations, clinical maladjustment composite, and emo-
exception of attention problems and reading (r = ­0.59) and
tional symptoms index (r = 0.44­0.59, p = 0.01).
ONCOLOGY NURSING FORUM ­ VOL 30, NO 5, 2003
E88
Table 3. Correlations Between Behavioral Assessment System for Children Scores and Wechsler Intelligence Scale for
Children­Revised Intelligence Quotient
Full Scale Intelligence Quotient
Performance Intelligence Quotient
Verbal Intelligence Quotient
Score
Score
Scale
p
Score
p
p
Teacher report
Learning problems scalea
< 0.01
­0.76
­0.68
< 0.02
­0.78
< 0.01
Attention problems scalea
­0.68
< 0.01
­0.57
< 0.01
­0.66
< 0.05
Leadership scale
< 0.01
­0.88
­0.74
­0.87
< 0.01
< 0.01
Study skills scale
< 0.01
­0.76
­0.66
< 0.01
­0.77
< 0.01
School problems compositea
­0.73
< 0.01
­0.63
< 0.01
­0.72
< 0.05
Adaptive skills composite
< 0.01
­0.73
­0.54
­0.68
ns
< 0.01
Parent report
Attention problems scalea
­0.79
< 0.01
­0.78
< 0.01
­0.79
< 0.01
Social skills scale
< 0.03
­0.60
­0.53
­0.62
ns
< 0.03
Behavioral symptoms indexa
­0.72
ns
­0.39
ns
­0.55
< 0.01
­0.64
­0.52
Adaptive skills composite
< 0.04
< 0.02
­0.58
ns
Child and adolescent report
Depression scalea
­0.74
< 0.05
­0.42
ns
­0.62
< 0.02
Social stress scalea
­0.87
­0.42
ns
< 0.01
­0.67
ns
a
Higher composite or scale scores indicate more problems.
ns--not significant
Discussion
ternalizing problems, especially depression, somatic com-
plaints, social withdrawal, and high anxiety. Researchers
Findings from this study suggest that children and adoles-
also have reported somatic complaints in long-term survi-
cents who are receiving or recently completed ALL treatment
vors of childhood cancer. Mulhern, Wasserman, Friedman,
may be at risk for some behavioral adjustment problems.
and Fairclough (1989) found that school problems and so-
These children and adolescents appear to be particularly vul-
matic complaints were increased fourfold relative to age-
nerable to internalizing problems, specifically somatization,
and gender-adjusted rates for peer groups in the general
depression, anxiety, and withdrawal. Boekaerts and Roer's
population. Somatic complaints could be related to other late
(1999) report of studies on stress, coping, and adjustment in-
effects of treatment and have been attributed to hypochon-
dicates that children with a chronic disease, including can-
driacal tendencies or functional impairments (Mulhern et al.,
cer, have more behavior problems compared to children
1989). The significant correlations between body image al-
without chronic disease, normative groups, and children
terations during treatment and self-report measures of be-
with acute non­life-threatening illnesses. The increased be-
havioral adjustment found in the current study corroborate
havior problems primarily are related to a higher level of in-
previous findings.
Table 4. Correlations Between Wide Range Achievement Test­Revised and Behavioral Assessment System for Children
Scores
Spelling
Math
Reading
Scale
r
p
r
p
r
p
Teacher report
Learning problems scalea
­0.91
< 0.01
­0.88
< 0.01
­0.61
< 0.03
Leadership scale
­0.85
< 0.01
­0.87
< 0.01
­0.74
< 0.01
Study skills scale
­0.85
< 0.01
< 0.01
ns
­0.83
­0.48
Attention problems scalea
­0.80
< 0.01
­0.81
< 0.01
­0.47
ns
School problems compositea
­0.86
< 0.01
­0.86
< 0.01
­0.55
ns
Adaptive skills composite
­0.67
< 0.02
­0.68
< 0.02
­0.44
ns
Behavioral symptoms index
­0.63
< 0.04
­0.67
< 0.02
­0.34
ns
Parent report
Attention problems scalea
­0.59
< 0.05
­0.61
< 0.01
­0.44
ns
Child and adolescent report
Social stress scalea
­0.81
< 0.01
­0.84
< 0.01
­0.24
ns
Depression scalea
­0.53
ns
­0.59
ns
­0.59
ns
Attitude to schoola
­0.64
ns
­0.62
ns
­0.47
ns
a
Higher composite or scale scores indicate more problems.
ns--not significant
ONCOLOGY NURSING FORUM ­ VOL 30, NO 5, 2003
E89
Previous studies of outcomes from CNS treatment for ALL
In the present study, 18% of child and adolescent self-report
have not investigated relationships between cognitive or aca-
scores were in the at-risk range for depression; high depression
demic abilities and specific areas of behavioral adjustment.
scores were correlated significantly with lower academic arith-
Thus, findings from this study are preliminary and warrant
metic abilities and with FSIQ and PIQ scores. Cavusoglu (2001)
replication. Future studies are needed to determine predictive
reported that children with cancer had significantly higher de-
relationships among these variables.
pression scores than a comparison group of healthy children
In summary, these findings suggest that, although the ma-
and that 22% of children with cancer had scores of 19 (the cut-
jority of children with cancer are not at risk for significant be-
off point for depression) or higher on the Children's Depression
havioral adjustment problems, specific areas of concern exist.
Inventory. These findings underscore the need to assess for
Scales in which at least 25% of scores from teacher, parent, or
depression in cancer survivors. The current study's findings
self-ratings were at risk were somatization, adaptability, atten-
also suggest a relationship between CNS treatment-related late
tion problems, withdrawal, anxiety, social skills, and learning
effects and depressive symptoms, as well as body image
problems. These areas are potential targets for behavioral and
changes during treatment and depression.
cognitive intervention strategies. Body image alterations may
Approximately 15% of teacher and parent behavioral
increase the risk for behavioral adjustment problems in chil-
symptoms index (overall level of problem behavior) ratings
dren and adolescents with ALL. CNS treatment-related de-
in the present study were in the at-risk range. This finding is
clines in cognitive and academic abilities were correlated with
in agreement with those of other researchers who reported
social stress and depression among cancer survivors. This is
that parents and teachers described an increased rate of clini-
a preliminary finding on a relatively small sample, but it also
cally significant behavioral problems in children who have
underscores the need for interventions designed to improve
survived cancer. One study reported that 14% of parent rat-
outcomes from CNS treatment.
ings and 7% of teacher ratings on the Child Behavior Check-
list met the criteria for poor adjustment (Newby, Brown,
Limitations
Pawletko, Gold, & Whitt, 2000). Butler, Rizzi, and Bandilla
(1999) assessed psychological functioning in 88 children on
These findings were based on data collected from two pe-
(n = 24) or off cancer therapy (n = 64) using the Personal-
diatric oncology treatment centers. Subjects were treated ac-
ity Inventory for Children and the Child Behavior Checklist.
cording to Children's Cancer Group or Pediatric Oncology
They found that, on average, 19% of participants were iden-
Group protocols, and treatment intensity varied across proto-
tified as having an adjustment problem on any one of the
cols. These limitations must be considered with respect to the
scales of the Personality Inventory for Children that mea-
generalizability of these findings. Oral glucocorticoids (e.g.,
sured self-control, social incompetence, internalization or
dexamethasone, prednisone) are used routinely as part of ALL
somatic symptoms, and cognitive development. The most
treatment. Data were not collected regarding type or dose of
common difficulties were in the areas of cognition, somatic
oral glucocorticoids; however, the researchers recognize that
complaints, anxiety, and family cohesiveness. Twelve per-
this also could affect behavioral adjustment and cause body
cent of the sample had adjustment difficulties according to
image alterations. Data about intrathecal hydrocortisone was
the Child Behavior Checklist (Butler et al.). Olson et al.
collected; the number of intrathecal hydrocortisone doses was
(1993) reported that cancer survivors who resided in a rural
not correlated with BASC, WISC-R, or WRAT-R variables.
setting were four times more likely than their age- and gen-
Future Directions
der-matched school peers to have social competence scores
(parent ratings) below the normal range. In the present study,
Based on an average age of 5 years at the time of cancer di-
24% of parent social skills ratings and 22% of sense of in-
agnosis and 77 years as the projected length of life, 72 years
adequacy self-ratings were in the at-risk range.
of every childhood cancer survivor's life are influenced by late
Butler et al. (1999) found that PIQ was a significant predic-
effects of cancer and its treatment (Bleyer, 1993). According to
tor of social competence, cognitive development, and with-
Bleyer (1990), the dramatic improvement in survival from
drawal. Teacher ratings of learning problems, school problems,
childhood cancer is worthwhile only if the quality of survival
leadership, study skills, and attention problems also were cor-
justifies the increased prolongation of life. Childhood cancer
related significantly with performance on a standardized mea-
survivors have not lived long enough in adequate numbers to
sure of academic abilities (Butler et al.). The current study's
accurately document the overall long-term impact of late effects
findings suggest a link between intellectual abilities (that may
on quality of life. However, this population continues to in-
be affected adversely by CNS treatment) and school problems,
crease and currently outnumbers other pediatric populations ex-
such as the ability to understand and complete schoolwork, or-
periencing chronic conditions related to hearing, visual, or or-
ganizational skills and study habits, the ability to concentrate,
thopedic impairments (Peckham, 1991; U.S. Department of
and the ability to successfully interact with peers and teachers.
Health and Human Services, 1997). Recognition is increasing
Body image changes during treatment, academic abilities,
that CNS treatment is associated with late effects that adversely
and FSIQ and PIQ scores were correlated negatively with
impact behavioral adjustment, as well as cognitive and aca-
BASC self-report of social stress and depression. Eighteen
demic abilities. Future studies that characterize patterns of defi-
percent of subjects had self-report scores in the at-risk range
cits and measure the efficacy of interventions designed to im-
for depression, and 16% of self-report scores were in the at-
prove behavioral, cognitive, and academic abilities among
risk range for social stress. To the researchers' knowledge, the
children receiving CNS treatment for cancer are needed.
current study is the first to examine relationships among be-
havioral adjustment (parent, teacher, and self-report ratings),
Author Contact: Ida M. (Ki) Moore, RN, DNS, FAAN, can be
body image alterations, and cognitive and academic abilities
reached at kmoore@nursing.arizona.edu, with copy to editor at
following CNS treatment for childhood ALL.
rose_mary@earthlink.net.
ONCOLOGY NURSING FORUM ­ VOL 30, NO 5, 2003
E90
References
Anderson, V., Smibert, E., Ekert, H., & Godber, T. (1994). Intellectual, edu-
Moore, I.M., Kramer, J., & Ablin, A.R. (1986). Late effects of central nervous
cational, and behavioural sequelae after cranial irradiation and chemo-
system prophylactic leukemia treatment on cognitive functioning. Oncol-
therapy. Archives of Disease in Childhood, 70, 476­483.
ogy Nursing Forum, 13, 45­51.
Appleton, R.E., Farrell, K., Zaide, J., & Rogers, P. (1990). Decline in head
Mulhern, R.K., Armstrong, F.D., & Thompson, S.J. (1998). Function-specific
growth and cognitive impairment in survivors of acute lymphoblastic leu-
neuropsychological assessment. Medical and Pediatric Oncology,
kaemia. Archives of Disease in Children, 65, 530­534.
1(Suppl.), 34­40.
Armstrong, F.D., Blumberg, M.J., & Toledano, S.R. (1999). Neurobehavioral
Mulhern, R.K., Wasserman, A.L., Friedman, A.G., & Fairclough, D. (1989).
issues in childhood cancer. School Psychology Review, 28, 194­203.
Social competence and behavioral adjustment of children who are long-
Bleyer, W.A. (1990). The impact of childhood cancer on the United States and
term survivors of cancer. Pediatrics, 83, 18­25.
the world. CA: A Cancer Journal for Clinicians, 40, 355­367.
Newby, W.L., Brown, R.T., Pawletko, T.M., Gold, S.H., & Whitt, K. (2000).
Bleyer, W.A. (1993). What can be learned about childhood cancer from "Can-
Social skills and psychological adjustment of child and adolescent cancer
cer Statistics Review 1973­1988." Cancer, 71(10 Suppl.), 3229­3236.
survivors. Psycho-Oncology, 9, 113­126.
Bleyer, W.A. (1995). The past and future of cancer in the young. Pediatric
Olson, A.L., Boyle, W.E., Evans, M.W., & Zug, L.A. (1993). Overall func-
Dentistry, 17, 285­290.
tion in rural childhood cancer survivors. Clinical Pediatrics, 32, 334­342.
Boekaerts, M., & Roer, I. (1999). Stress, coping, and adjustment in children
Peckham, V.C. (1991). Educational deficits in survivors of childhood cancer.
with a chronic disease: A review of the literature. Disability and Rehabili-
Pediatrician, 18, 25­31.
tation, 21, 311­337.
Pendley, J.S., Dahlquist, L.M., & Dreyer, D. (1997). Body image and psycho-
Brown, R.T., Sawyer, M.B., Antoniou, G., Toogood, I., Rice, M., Thompson,
social adjustment in adolescent cancer survivors. Journal of Pediatric Psy-
N., et al. (1996). A 3-year follow-up of the intellectual and academic func-
chology, 22, 29­43.
tioning of children receiving central nervous system prophylactic chemo-
Reynolds, C.R., & Kamphaus, R.W. (1992). Behavioral assessment system for
therapy for leukemia. Developmental and Behavioral Pediatrics, 17, 392­
children. Circles Pine, MN: American Guidance Service.
398.
Ries, L.A.G., Smith, M.A., Gurney, J.C., Linet, M., Tamra, T., Young, J.L.,
Butler, R.W., Rizzi, L.P., & Bandilla, E.B. (1999). The effects of childhood
et al. (1999). Cancer incidence and survival among children and adoles-
cancer and its treatment on two objective measures of psychological func-
cents: United States SEER Program 1975­1995 [NIH Pub No. 99-4649].
tioning. Children's Health Care, 28, 311­327.
Bethesda, MD: National Cancer Institute.
Cavusoglu, H. (2001). Depression in children with cancer. Journal of Pediatric
Shapiro, E.S., & Kratochwill, T.R. (2000). Introduction: Conducting a mul-
Nursing, 16, 380­385.
tidimensional behavioral assessment. In E.S. Shapiro & T.R. Kratochwill
Copeland, D.R., Moore, B.D., Francis, D.J., Jaffe, N., & Culbert, S.J. (1996).
(Eds.), Conducting school-based assessments of child and adolescent be-
Neuropsychologic effects of chemotherapy on children with cancer: A lon-
havior (pp. 1­20). New York: Guilford Press.
gitudinal study. Journal of Clinical Oncology, 14, 2826­2835.
Sharan, P., Mehta, M., & Chaudhry, V.P. (1999). Psychiatric morbidity in
Cousens, P., Ungerer, J.A., Crawford, J.A., & Stevens, M.M. (1991). Cogni-
children suffering from acute lymphoblastic leukemia. Pediatric Hematol-
tive effects of childhood leukemia therapy: A case for four specific deficits.
ogy and Oncology, 16, 49­54.
Journal of Pediatric Psychology, 16, 475­488.
Silber, J.H., Radcliffe, J., Peckham, V., Perilongo, G., Kishnani, P., Fridman,
Eckert, T.L., Dunn, E.K., Codding, R.S., & Guiney, K.M. (2000). Self-report:
M., et al. (1992). Whole-brain irradiation and decline in intelligence: The
Rating scale measures. In E.S. Shapiro & T.R. Kratochwill (Eds.), Conduct-
influence of dose and age on IQ score. Journal of Clinical Oncology, 10,
ing school-based assessments of child and adolescent behavior (pp. 150­
1390­1396.
169). New York: Guilford Press.
U.S. Department of Health and Human Services, Office of the Assistant Sec-
Fossen, A., Abrahamsen, T.G., & Storm-Mathisen, I. (1998). Psychological
retary for Planning and Evaluation. (1997). Trends in the well-being of
outcome in children treated for brain tumor. Pediatric Hematology and
America's children and youth. Retrieved June 24, 2003, from http://
Oncology, 15, 479­488.
aspe.hhs.gov/hsp/97trends/intro-web.htm
Jastak, S., & Wilkinson, G. (1984). The Wide Range Achievement Test (revised
Wechsler, D. (1974). Manual for the Wechsler Intelligence Scale for Children.
ed.). Wilmington, DE: Jastak Associates.
New York: Psychological Corporation.
Jemal, A., Murray, T., Samuels, A., Ghafoor, A., Ward, E., & Thun, M.
(2003). Cancer statistics, 2003. CA: A Cancer Journal for Clinicians, 51,
For more information . . .
5­26.
Kramer, J., & Moore, I.M. (1989). The late effects of cancer therapy on the
central nervous system. Seminars in Oncology Nursing, 5, 22­28.
Cancer Source: Follow-Up Care for Childhood Cancer Survi-
McCarthy, D. (1972). McCarthy Scales of Children's Abilities. New York:
vors
Psychological Corporation Harcourt Brace Jovanovich.
www.cancersourcekids.com
Merrell, K.W. (2000). Informant report: Rating scale measures. In E.S.
Shapiro & T.R. Kratochwill (Eds.), Conducting school-based assessments
Patient-Centered Guides: Childhood Cancer Survivors
of child and adolescent behavior (pp. 203­233). New York: Guilford Press.
www.patientcenters.com/survivors
Moore, I.M., Espy, K.A., Kaufmann, P., Kramer, J., Kaemingk, K., Miketova,
The Children's Cause
P., et al. (2000). Cognitive consequence and central nervous system injury
www.childrenscause.org
following treatment for childhood leukemia. Seminars in Oncology Nurs-
ing, 16, 279­290.
Links can be found using ONS Online at www.ons.org.
Moore, I.M., Glasser, M.E., & Ablin, A.R. (1988). The late psychosocial con-
sequences of childhood cancer. Journal of Pediatric Nursing, 3, 150­158.
ONCOLOGY NURSING FORUM ­ VOL 30, NO 5, 2003
E91