Article review
One page review per article (2pages in total separate for each article)
tie in a reference of a topic from the article
show relevance and and accuracy
no plagiarism
Research in Developmental Disabilities 37 (2015) 143151
Contents lists available at ScienceDirect
Research in Developmental Disabilities
Developmental level and psychopathology: Comparing (R) CrossMark
children with developmental delays to chronological
and mental age matched controls
Barbara Caplan a,*, Cameron L. Neece b, Bruce L. Baker a
a University of California, Los Angeles, United States
b Loma Linda University, United States
A R T I C L E I N F O
Article history:
Received 16 July 2014
Received in revised form 27 October 2014
Accepted 29 October 2014
Available online 9 December 2014
Keywords:
Child psychopathology
Developmental delays
Dual diagnosis
Mental age
A B S T R A C T
Children with developmental delays (DD) are at heightened risk for developing clinically
significant behavioral and emotional difficulties as compared to children with typical
development (TD). However, nearly all studies comparing psychopathology in youth with
DD employ TD control groups of the same chronological age (CA). It is unclear, then,
whether the heightened symptomology found in age-matched children with DD is beyond
what would be expected given their developmental level. The present study assessed rates
of behavior problems and mental disorder in 35 children with DD at age 9 years. These
were compared with rates from 35 children with TD matched for CA at age 9 and also
earlier rates for these same children at age 6, when matched for mental age (MA). Children
with DD had significantly more behavior problems in 7 of the 17 scales of the CBCL when
compared to TD children matched for CA, and 6 of 17 scales when compared to the MA-
matched group. Rates of meeting DSM-IV criteria for a psychiatric disorder were
significantly higher in the DD group than both the CA- and MA-matched TD groups for
three and four, respectively, of the seven diagnoses examined. Descriptively, the mean
ratings for all variables assessed were higher for the DD group than both TD comparison
groups, with the exception of the Anxious/Depressed scale of the CBCL. These findings
validate the heightened risk for clinically significant behavior problems and mental
disorders in youth with DD above and beyond their developmental functioning.
2014 Elsevier Ltd. All rights reserved.
1. Introduction
Children with developmental delays (DD) are at substantially higher risk for developing clinically significant behavioral
and emotional difficulties than children with typical cognitive development. This comorbidity of DD and a mental disorder is
often referred to as dual diagnosis. Epidemiological studies indicate that 3050% of youth with DD meet the clinical cutoff for
behavioral and emotional problems and/or diagnosable mental disorder (see Einfeld, Ellis, & Emerson, 2011 for review).
Studies that include comparison samples of children with typical development (TD) highlight the considerable difference in
* Corresponding author at: Department of Psychology, UCLA, 405 Hilgard Avenue, Los Angeles, CA 90095, United States. Tel.: +1 3108259592.
E-mail address: [emailprotected] (B. Caplan).
http://dx.doi.org/10.1016/j.ridd.2014.10.045
0891-4222/ 2014 Elsevier Ltd. All rights reserved.
http://crossmark.crossref.org/dialog/?doi=10.1016/j.ridd.2014.10.045&domain=pdf
http://crossmark.crossref.org/dialog/?doi=10.1016/j.ridd.2014.10.045&domain=pdf
http://dx.doi.org/10.1016/j.ridd.2014.10.045
mailto:[emailprotected]
http://www.sciencedirect.com/science/journal/08914222
http://dx.doi.org/10.1016/j.ridd.2014.10.045
B. Caplan et al. / Research in Developmental Disabilities 37 (2015) 143151144
risk for psychopathology, with the relative risk for youth with DD (to youth with TD) ranging from 2.84.1 to 1 (Baker, Neece,
Fenning, Crnic, & Blacher, 2010; Dekker, Koot, van der Ende, & Verhulst, 2002; Emerson & Hatton, 2007).
For the purposes of the current study, the term developmental delay refers specifically to children with cognitive delay,
one of the six areas of developmental delay recognized by the Individuals with Disabilities Act (IDEA, 2004). It is estimated that
13.4% of children in the United States meet criteria for cognitive developmental delay (Rosenberg, Zhang & Robinson, 2008).
This developmental (cognitive) delay is associated with diverse impairments in multiple domains, including social
competence (Guralnick, 1999), adaptive behavior (e.g., Green, Caplan, & Baker, 2013) and the focus of the current study,
clinically significant behavioral and emotional problems (i.e. dual diagnosis; Baker et al., 2010).
Dual diagnosis is a concern for families, service providers, and researchers alike, as the presence of mental disorder in
individuals with DD contributes substantially to the burden of disability. Dual diagnosis has been found to have deleterious
effects on families, including increased parenting stress (Neece, Green, Baker, 2012), increased burden on the family
(Irazabal et al., 2012) and decreased parental well-being (Tonge & Einfeld, 2003) beyond the presence of DD alone.
Additionally, psychopathology in youth with DD often limits participation in the community, as severity of psychopathology
relates to unsuccessful adaption to independent living (Fotheringham, 1999), reduced occupational opportunities
(Anderson, Lakin, Hill, & Chen, 1992) and restrictions in recreational and educational programs (Parmenter, Einfeld, Tonge, &
Dempster, 1998).
The substantially increased risk for psychopathology in youth with DD, coupled with the implications of dual diagnosis
for the affected individual and family, highlight the need for early assessment and treatment of behavior problems and
mental disorders. However, there has been some debate regarding the validity of diagnosable mental disorders in youth
with DD. The Diagnostic and Statistical Manual for Mental Disorders (DSM) has consistently referenced developmental level
in defining many childhood mental disorders (e.g. APA, 2000), making it difficult for clinicians and researchers to determine
whether the presence of symptomatology in youth with DD is attributable to their developmental level, or to a distinct,
comorbid condition. This practice continues in the recently released DSM-5 (APA, 2013), which states that to meet criteria
for certain childhood mental disorders, an individual must present with symptoms: inconsistent with developmental
level (Attention Deficit Hyperactivity Disorder; ADHD), outside the range that is normative for the individuals
developmental level (Oppositional Defiant Disorder; ODD) or that are developmentally inappropriate (Separation
Anxiety).
Despite these developmental criteria, nearly all studies comparing rates of psychopathology in youth with or without DD
implement comparison groups of TD children matched for chronological age (CA; e.g., Baker et al., 2010; Emerson & Hatton,
2007), thereby comparing groups that by definition operate at different developmental levels. A more nuanced approach
rooted in developmental theory is to match targeted groups on an indicator of developmental functioning, such as mental
age (MA; Flanagan, Russo, Flores, & Burack, 2008). By comparing the behaviors and symptoms of children with DD to MA-
matched controls, one can determine whether children with DD are truly at heightened risk for developing psychopathology
beyond what would be expected given each childs cognitive level.
The goal of comparison group matching is to rule out non-central, or extraneous, explanations of group differences
(Jarrold & Brock, 2004). Using control groups matched by MA is a longstanding method for controlling for developmental
level in comparing the functioning of children with and without developmental disabilities. This methodology largely stems
from the developmental (e.g., Zigler, 1969) and difference (e.g., Ellis & Cavalier, 1982) approaches to intellectual disability,
through which researchers sought to illuminate the similarities or fundamental differences, respectively, in developmental
processes exhibited by children with or without intellectual disability when controlling for developmental level. More
recently, the approach has been used to demonstrate differences in many behavioral and cognitive processes, such as
adaptive behavior (Fidler, Hepburn, & Rogers, 2006), social communication (Strid, Heimann, Gillberg, Smith, & Tjus, 2013)
and joint attention (Wong & Kasari, 2012) between individuals with or without developmental disabilities.
The few studies that have implemented MA-matching to study behavior problems or psychopathology in children with or
without DD have primarily focused on children with specific disorders, such as Down syndrome. Evans and Gray (2000)
examined compulsive behaviors in children with Down syndrome and MA-matched TD controls, finding that children with
Down syndrome were rated to have the same number of different compulsive behaviors, but higher frequency and more
intensity of these behaviors. Evans, Canavera, Kleinpeter, Maccubbin, and Taga (2005) assessed differences in the fears,
phobias, and anxieties of children with autism spectrum disorder (ASD), Down syndrome (matched for MA), or MA- and CA-
matched children with TD. Children with ASD were reported to have more situational and medical fears than any of the other
groups, regardless of CA or MA matching. More recently, van Gameren-Oosterom et al. (2011) examined eight year-old
children with Down syndrome and MA-matched controls, finding that children with Down syndrome exhibited more
emotional and behavioral problems. Additionally, Evans, Kleinpeter, Slane and Boomer (2014) found that repetitive behavior
predicted poor adaptive functioning for children with Down syndrome, but not MA-matched controls. Together these
findings substantiate the increased risk for psychopathology in children with DD beyond developmental level. However,
there is a need to replicate such findings across heterogeneous samples of children with DD, rather than only those with
specific disorders such as Down syndrome or autism.
The present study expanded on prior work by comparing a broad range of behavior problems in children with DD to these
same problems in CA- and MA-matched TD comparisons. Additionally, we sought to determine whether children with DD
met DSM-IV criteria for the most prevalent childhood mental disorders, including ADHD, ODD, anxiety disorders and mood
disorders (Merikangas et al., 2010), at a higher rate than CA- and MA-matched TD controls.
B. Caplan et al. / Research in Developmental Disabilities 37 (2015) 143151 145
We addressed the following questions: (1) How do rates of behavior problems and mental disorder compare between
children with DD and children with TD matched for chronological age (CA)? Consistent with previous studies cited, we
expected that children with DD would exhibit higher rates of behavior problems and psychopathology than children with TD
matched for CA. (2) How do rates of behavior problems and mental disorders compare between children with DD and these
same children with TD matched for MA? We anticipated that even when controlling for developmental level, children with
DD would demonstrate elevated rates of behavior problems and psychopathology as compared to children with TD. Such
findings would support the claim that variables related to the developmental disability beyond developmental level may be
contributing to the development of behavior problems and psychopathology.
2. Materials and methods
2.1. Participants
Participants were drawn from an ongoing longitudinal study of children with or without DD and their families (the
Collaborative Family Study). Of the 194 families who participated in the 9-year assessment, 70 were selected for the current
study based on the criteria for our matching procedures (described below). The larger project was a collaborative effort between
UCLA, UC Riverside and Pennsylvania State University. Families of children with DD had been recruited at child age 3 years
primarily through agencies that provide and purchase diagnostic and intervention services for persons with intellectual and
developmental disabilities, while families of children with TD were recruited through daycare facilities and preschools.
DD group (n = 35) participants were selected if they had complete data at the 9-year assessment and IQ scores within the
range of 6084. This IQ range was chosen to allow for the closest MA-matching to TD children. Children within the DD group
exhibited cognitive developmental delays (IDEA, 2004) as evidenced by IQ scores greater than one standard deviation below
the population mean (IQ < 85) using appropriate diagnostic instruments. Fifteen participants in the DD group (42.8%)
demonstrated IQ scores that fell in the range of intellectual disability (IQ 70; APA, 2013), though, it should be noted that a
diagnosis for intellectual disability could not be confirmed as measures of adaptive behavior were not considered. Within the
DD group, the majority of children had undifferentiated developmental delay (i.e., did not have a specific syndrome), while a
smaller subset of the sample were reported to have an autism spectrum disorder (n = 5), Cerebral Palsy (n = 6), Down
syndrome (n = 1), or another syndrome or multiple syndromes (n = 5). TD group participants (n = 35) were selected if they
had complete data at the 6-year and 9-year assessments, had IQ scores ranging from 90 to 125 and matched a selected DD
participant for MA at age 6. Children in the TD group were not born prematurely and did not have any disability diagnosis.
Table 1 shows DD and TD group demographic characteristics at age 9. The combined sample was 57.4% male and 59%
White, non-Hispanic. Other race/ethnicities included Hispanic (13.4%), African American (7.5%), Asian American (2.2%), and
other (17.9%). Most participating mothers were married (79.3%). The majority of families (66.7%) had an annual household
income above $50,000, and mothers average years of education was 15.6 years. While mothers in the TD group completed
significantly more years of education and were more likely married relative to the DD group, these variables were not related
to the dependent variables and thus they were not covaried in analyses.
2.2. Procedures
Research procedures were reviewed and approved by Institutional Review Boards at the three participating universities
and informed consent was obtained from all participating families. At both the age 5 and age 9 assessments, families visited
the research center for child cognitive testing. To assess for the presence of cognitive developmental delay, trained graduate
students and research staff administered the Stanford Binet IV (SB-IV; Thorndike, Hagen, & Sattler, 1986) at child age 5 and
the Weschler Intelligence Scales for Children, Fourth Edition (WISC-IV; Wechsler, 2003) at child age 9. Within the broader
sample from this longitudinal study, the correlation between IQ at age 5 (SB-IV) and IQ at age 9 (WISC-IV) was r = .90. Given
Table 1
Demographic information.
Developmentally delayed (age 9) Typically developing (age 9) X2 or t
Child
Sex (% male) 68.6 57.1 X2 = .98
Race (% caucasian) 54.3 65.7 X2 = .95
IQa [mean (SD)] 72.1 (7.9) 111.2 (10.3) t = 17.6***
Mental agea 6.5 (.7) 10.0 (.9) t = 17.3***
Parent/family
Marital status (% married) 71.4 91.4 X2 = 4.63*
Mothers race (% caucasian) 57.1 68.6 X2 = .98
Mothers education (highest grade completed) 14.3 (2.3) 16.6 (2.4) t = 4.19***
Family income (% above $50,000) 60.0 85.3 X2 = 5.53*
* p < .05.
*** p < .001.
a IQ scores used to determine group classification and mental age for the age 6 typically developing group were calculated at child age 5.
B. Caplan et al. / Research in Developmental Disabilities 37 (2015) 143151146
this stability of cognitive performance, IQ assessed at age 5 was used as a proxy for IQ at age 6 to calculate MA. At assessment
ages 6 and 9, mothers completed a battery of questionnaires about child and family functioning, and were administered an
interview regarding family demographic and child diagnostic information, as well as a structured diagnostic interview about
child mental health.
2.3. Measures
2.3.1. Assessment of IQ
2.3.1.1. Stanford-Binet IV (SB-IV). The SB-IV (Thorndike et al., 1986), a widely used assessment instrument of cognitive
ability, yields an IQ score with a normative mean of 100 and a standard deviation of 15; it has sound psychometric properties
and is particularly suited to the evaluation of children with delays, given that the examiner adapts starting points to
developmental level. This test has strong test-retest stability (a = .91 for 5 year-olds) and high internal consistency for the
composite score (a = .95.99 across ages; Thorndike et al., 1986).
2.3.1.2. The Weschler Intelligence Scales for Children. Fourth Edition (WISC-IV). Child cognitive level was reassessed at age 9,
using the WISC-IV (Wechsler, 2003), a widely used assessment of cognitive functioning for children ages 616. The scores
from the Vocabulary, Matrix Reasoning, and Arithmetic WISC-IV subtests were used to derive an estimated full-scale IQ and
to confirm status group classification (TD or DD). The selection of the three subtests was based on the high correlation
(r = .91) of their composite score to the full-scale IQ from the full administration of the WISC-IV (Sattler, 2001).
2.3.2. Assessment of child behavior problems/mental disorders
2.3.2.1. Child Behavior Checklist for ages 618 (CBCL). The CBCL (Achenbach & Rescorla, 2001), completed by mothers at child
ages 6 and 9, contains 113 items that describe behavioral and emotional problems. For each item, the respondent indicates
whether the behavior is (0) not true, (1) somewhat or sometimes true, or (2) very true or often true of her child. The CBCL yields a
total problem score, broadband externalizing and internalizing scores, eight narrowband syndrome scores and six DSM-
oriented scores.
2.3.2.2. Diagnostic Interview Schedule for Children (DISC). The DISC (Costello, Edelbrock, & Costello, 1985), administered to
mothers at child ages 6 and 9 years, is a structured diagnostic interview targeting criteria for child psychiatric disorders as
defined in the Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev. [DSM-IV-TR]; APA, 2000). We employed
an alternative method of administering the DISC (Baker et al., 2010; Edelbrock, Crnic, & Bohnert, 1999) in which six modules
appropriate for young children (Social Phobia, Separation Anxiety, Major Depressive Disorder, Dysthymic Disorder, ADHD
and ODD) were targeted. The interviewer began by providing a brief summary of each diagnosis and asked the mother to
select the first diagnostic area to be covered. Standard administration of the module was followed. After the first module, the
interviewer asked the mother to select the next diagnostic area. This continued until the mother said that there were no
other areas relevant to her child. The interviewer then reviewed the remaining areas to confirm further that the child did not
have problems represented in any remaining module. This alternate administration procedure has been found to increase
reliability, decrease attenuation (reporting fewer symptoms later in the interview), and decrease administration time as
compared to the standard procedure (Edelbrock et al., 1999; Jensen et al., 1999). This administration procedure has been
successfully implemented to assess mental health in children with or without DD (Baker et al., 2010; Edelbrock et al., 1999;
Neece, Baker, Crnic, & Blacher, 2012).
2.4. Data analysis
2.4.1. Matching procedure
To determine whether rates of behavior problems and mental disorders differed between children with DD and TD
controls of equivalent MA, participants were individually matched according to MA. The mental age for each participant was
calculated using the following formula: MA = (IQ/100 CA). IQs for the DD group were determined using the WISC-IV scores
obtained at age 9 and for the TD group using the SB-IV scores obtained at age 5. After calculating MAs, each participant in the
DD group at age 9 was individually matched to the participant in the TD group at age 6 with the closest MA. The resulting IQ
range for the TD participants was 90125. The difference in MA between matched pairs ranged from .00 to .24 years. The two
matched groups did not differ in MA (t = .06, p = .95); this p-value is well above the recommended level to consider groups
matched on a target variable (p > 50; Mervis & Klein-Tasman, 2004). To validate prior findings indicating elevated rates of
behavior problems and mental disorders in children with DD as compared to CA-matched controls, the same participants
were compared using data collected at the same assessment time (age 9).
2.4.2. Analysis of group differences
Independent samples t-tests were conducted on CBCL scale scores to determine group differences in ratings of DSM-
oriented behaviors, syndrome-related behaviors, as well as internalizing, externalizing and total problems. To compare rates
B. Caplan et al. / Research in Developmental Disabilities 37 (2015) 143151 147
of mental disorders, as measured by meeting diagnostic criteria for disorders according to the DISC, chi-square statistics
were utilized. The CA-match analyses examined group differences (TD vs. DD) at child age 9, while the MA-match analyses
compared the TD sample at age 6 to the DD sample at age 9.
3. Results
3.1. Preliminary analyses
The distributions of the behavior problems variables were examined within each group (DD, CA-Match Control, and MA-
Match Control). Five data points that were more than three standard deviations above the mean of a variable were
considered to be outliers and as recommended by Cohen, Cohen, West, and Aiken (2002), these were set equal to plus or
minus 3 standard deviations from the mean.
3.2. Analyses using chronological age-match control group
3.2.1. Rates of behavior problems: DD vs. TD samples
Table 2 shows the rates of behavior problems for the DD and CA-matched control groups. Children with DD were rated as
having significantly more total problems as well as marginally significantly more externalizing problems than the CA-
matched control group. With CBCL syndrome scales, there were significant group differences (DD > TD) on the Social
Problems, Attention Problems and Aggression subscales as well as a marginally significant difference in Rule Breaking
behavior. With the DSM-oriented subscales of the CBCL, children were DD were reported to have higher levels of ADHD, ODD
and Conduct Disorder behaviors. All significant group comparisons demonstrated medium to large effect sizes according to
the conventions for Cohens d (Cohen, 1988).
3.2.2. Rates of mental disorder: DD vs. TD samples
Table 3 depicts the prevalence of DSM-IV diagnoses assessed in the DD and CA-matched control groups. The proportion of
children with DD meeting criteria for any mental disorder assessed was marginally significantly higher than that of children
with TD. The prevalence of any type of ADHD was significantly higher in the DD than the TD sample, with the Inattentive sub-
type over two times as prevalent in the DD group. Separation Anxiety Disorder was significantly more prevalent in the DD
sample as well, with 11.4% of children in the DD sample meeting diagnostic criteria; however, risk and odds ratios could not
be calculated for Separation Anxiety due to empty cells (i.e., no children in the TD sample met criteria). No significant group
differences were found for ADHD-Hyperactive/Impulsive subtype, ODD and Social Phobia, although prevalence rates for
these disorders were all greater in the DD group than the CA-Matched TD group.
able 2
hild behavior problems.
CBCL Scale DD samplea CA-match comparison MA-match comparison
TD samplea
(age 9)
t-Value Cohens d TD samplea
(age 6)
t-Value Cohens d
Broad band
Internalizing 52.7 50.2 1.35 .25 50.2 1.07 .25
Externalizing 54.5 50.1 1.93y .46 50.3 1.76y .42
Total Problems 56.3 49.8 2.66* .63 50.2 2.59* .62
DSM-oriented
ADHDb 58.1 54.4 2.17* .52 53.8 2.69** .64
Oppositional defiant disorder 57.1 54.1 2.07* .50 55.0 1.56 .37
Conduct disorder 57.4 53.4 2.55* .61 53.7 2.44* .58
Affective problems 54.8 54.5 0.23 .06 54.4 0.29 .07
Anxiety problems 55.7 54.7 0.64 .15 54.5 0.76 .18
Somatic problems 55.0 53.4 1.17 .28 54.2 0.53 .13
Other
Social problems 59.3 53.7 3.37** .83 53.4 3.98*** .97
Attention problems 60.6 55.2 2.98** .71 54.0 4.21*** 1.00
Rule breaking 56.1 53.7 1.69y .40 52.8 2.83** .68
Aggression 57.4 53.7 2.32* .55 54.4 1.78y .43
Anxious/depressed 54.0 54.7 0.46 .11 54.4 0.27 .06
Withdrawn/depressed 55.9 54.1 1.15 .27 54.4 0.93 .22
Thought problems 56.7 56.1 0.34 .08 54.9 1.14 .27
y p < .10.
* p < .05.
** p < .01.
*** p < .001.
a n = 35.
b ADHD = Attention Deficit Hyperactivity Disorder.
T
C
Table 3
Child mental disorder.
DISC modules DD samplea CA-match analyses MA-match analyses
TD samplea
(age 9)
X2 Relative risk
(DD:TD)
OR TD samplea
(age 6)
X2 Relative risk
(DD:TD)
OR
Any mental disorder 57.1 37.1 2.81y 1.54:1 2.26 34.3 3.68y 1.66:1 2.56
ADHDb (any subtype) 42.9 20.0 4.24* 2.14:1 3.00 14.3 7.00** 3.00:1 4.50
ADHDb-inattentive subtype 40.0 17.1 4.48* 2.34:1 3.22 11.4 7.48** 3.50:1 5.17
ADHDb-hyperactive subtype 25.7 14.3 1.43 1.80:1 2.08 5.7 5.29* 4.50:1 5.71
Oppositional defiant disorder 31.4 20.0 1.20 1.57:1 1.83 20.0 1.20 1.57:1 1.83
Separation anxiety disorder 11.4 0 4.24* 8.6 0.16 1.32:1 1.38
Social phobia 17.1 8.6 1.15 1.99:1 2.21 0 6.56*
y p < .10.
* p < .05.
** p < .01.
a n = 35.
b ADHD = Attention Deficit Hyperactivity Disorder.
Not listed due to empty cells: Major Depressive Disorder, Dysthymia Disorder.
B. Caplan et al. / Research in Developmental Disabilities 37 (2015) 143151148
3.3. Analyses using mental age-match control group
3.3.1. Rates of behavior problems: DD vs. TD samples
Table 2 also shows findings using the MA-matched control group. These were highly similar to results when matched by
CA. Children with DD were reported to have significantly more total problems and marginally significantly more
externalizing problems. Here too, Social Problems and Attention Problems were significantly higher in the DD group than the
MA-matched control sample. While the increased risk for Rule Breaking in the DD group was marginally significant in the
CA-match analyses, this difference was significant in the MA-match analyses. On the other hand, Aggression was
significantly higher in the DD group for the CA-match analyses and marginally significant in the MA-match analyses. In
examining the DSM-oriented subscales, the DD sample received significantly higher ratings for ADHD and Conduct Disorder
behaviors than the TD comparisons. As with the CA-matched analyses, significant group differences for the DD and MA-
matched samples on the CBCL subscales demonstrated medium to large effect sizes (Cohen, 1988). Additionally, ADHD,
Attention Problems, Social Problems and Rule Breaking had higher effect sizes and lower p-values in the MA-matched
analyses than the CA-matched analyses, noting stronger group differences in these areas.
3.3.2. Rates of mental disorder: DD vs. TD samples
Table 3 also shows the prevalence rates of DSM-IV diagnoses in the DD and MA-matched control groups. The proportion
of children who met criteria for any mental disorder was marginally significantly greater in the DD sample than the MA-
matched control sample. Significant group differences were found for ADHD, most notably the inattentive subtype, which
was over three times as prevalent in the DD group. In contrast to the CA-match analyses, in the MA-match analyses, the DD
group demonstrated significantly higher rates of the hyperactive subtype. Additionally, the prevalence of Social Phobia was
significantly higher in the DD than the TD sample for the MA-match analyses. Risk and odds ratios for Social Phobia could not
be calculated due to empty cells. Conversely, no group differences were found for Separation Anxiety in the MA-matched
analyses.
In sum, the DD group was rated to have more problems than both TD comparison groups for all of the scales of the CBCL
assessed except for the Anxious/Depressed scale. Of 16 scales, these group differences were significant for six scales in the
CA-matched analyses, and seven scales in the MA-match analyses. Of note, no significant group differences were found for
the Internalizing, Affective Problems, Anxiety, Anxious/Depressed, Withdrawn/Depressed and Somatic Problems scales of
the CBCL in either set of analyses. In addition, a larger proportion of the DD group than the TD groups met criteria for all
mental disorders assessed and group differences for most DSM diagnoses (with the exception of Separation Anxiety) were
stronger (i.e., higher chi-square values, greater relative risk for DD group) in the MA-matched than the CA-matched analyses.
While group differences were found for Separation Anxiety in the CA-, but not MA-matched analyses, the opposite was true
for Social Phobia, which was significant in the MA-, but not CA-matched analyses. Neither the CA- nor MA-matched analyses
revealed significant group differences for ODD, though descriptively, the proportion of children that met criteria was higher
in the DD group in both sets of analyses.
4. Discussion
We examined rates of behavior problems and the prevalence of psychopathology among children with DD compared to
both CA and MA-matched comparison groups in order to obtain a more complete picture of the heightened risk for
psychopathology among children with DD. To the best of our knowledge, this is the first study to use a MA-matched
comparison group to assess the relative risk of children with DD for developing a range of behavior problems and mental
disorders. Additionally, the design is unique in that we utilized select data from a longitudinal investigation to
B. Caplan et al. / Research in Developmental Disabilities 37 (2015) 143151 149
simultaneously compare children with DD to the same comparison children with TD, matched for both MA and CA. This
matching procedure allowed us to assess the risk for psychopathology in children with DD relative to CA- and MA-matched
comparisons, controlling for individual differences in the TD sample other than developmental level that might contribute to
the relative risk in one DD-TD comparison versus another.
Consistent with previous findings, children with DD were reported to have more behavior problems and mental disorder
diagnoses compared to their same-aged peers. Importantly, we also found that this elevated risk for behavior problems and
psychopathology in children with DD was maintained, or even accentuated, when contrasted with an MA-matched control
group. These findings substantiate the increased risk for psychopathology and problem behaviors in youth with DD, above
and beyond mental age. But if mental age does not account for the increased risk, what does? There is some evidence to
suggest that variables related to DD other than mental age, such as regulatory abilities, could play a key role in the
development of psychopathology.
Childrens regulatory capacities, specifically those involving emotional regulation, have long been implicated in the
emergence of childrens behavior problems and mental disorders (Crnic, Hoffman, Gaze, & Edelbrock, 2004). Emotion
regulation involves the ability of an individual to control, modify, change and direct emotional reactions and expression to
achieve ones goals and effectively manage interpersonal relationships (e.g., Cole, Martin, & Dennis, 2004), a set of abilities
likely to be less developed in children with DD (Crnic et al., 2004). The few studies examining emotion regulation in children
with DD consistently report deficits in these skills among this group (e.g., Berkovits & Baker, 2013; Wilson, 1999).
Furthermore, studies increasingly have suggested that emotion dysregulation may serve as one mechanism through
which early developmental delays influence later behavior and social problems. Gerstein et al. (2011) found childrens
regulatory abilities mediated the relation between early developmental status and later behavior problems. Additionally,
Baker, Fenning, Crnic, Baker, and Blacher (2007) found that global emotion dysregulation at age four not only predicted
childrens social skills at age six, but also partially mediated
