Characteristics of Children With ADHD and Comorbid Anxiety

Abstract

Objective: The following comorbid subgroups of ADHD have been proposed: ADHD Only, ADHD + anxiety disorders (ANX), ADHD + oppositional defiant disorder/conduct disorder (ODD/CD), and ADHD + ODD/CD + ANX. The current study examined a subset of these groups. Method: A total of 134 children and adolescents (M age = 9.92; range = 6-17) from a clinic-referred sample (n = 407) were grouped based on a semistructured diagnostic interview: ADHD only (n = 41), ADHD + ANX (n = 31), and ANX Only (n = 62). Results: Findings supported greater parent-reported anxiety symptoms in anxiety groups, and greater parent- and teacher-reported attention problems in ADHD groups. ADHD groups performed worse on a continuous performance test, whereas ADHD + ANX performed worse on working memory than ADHD Only. ADHD + ANX reported more physical anxiety symptoms than ADHD Only. Conclusion: Comorbid anxiety should be considered in ADHD assessment and treatment.

Keywords

ADHD anxiety comorbidity children

ADHD is often accompanied by a comorbid disorder, but relatively little is known about how comorbidity affects ADHD symptomatology and related deficits (e.g., neuropsychological functioning). Of research on comorbidity that is available, particular comorbid disorders have received more empirical attention than others. For example, given that externalizing disorders such as oppositional defiant disorder (ODD) and conduct disorder (CD) are present in approximately 50% of children with ADHD, a large body of research has addressed the characteristics of these children (Newcorn, Halperin, & Miller, 2009). Although comorbidity with these disorders is clearly very high, it is also important to note that internalizing disorders such as anxiety disorders are comorbid with ADHD in approximately 25% of children in the general population and around 30% to 40% of clinically referred children (MTA Cooperative Group, 1999; Tannock, 2009). Unfortunately, only a limited number of studies have examined the characteristics of children with ADHD and comorbid anxiety disorders (see Jarrett & Ollendick, 2008, for a recent review) despite this high prevalence of comorbidity.

One of the questions in this literature has been whether the presence of anxiety alters the symptomatology and neurocognitive characteristics of children with ADHD. Recently, Tannock (2009) noted that the effects of anxiety on ADHD symptomatology have been inconsistent. In studies involving the Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV; American Psychiatric Association, 1994) criteria, the presence of anxiety has been associated with somewhat greater inattention relative to impulsivity (Jensen et al., 2001; Newcorn et al., 2001); however, such differences have not been found for objective behavior ratings (Abikoff et al., 2002) or when ODD/CD is also present (Newcorn et al., 2001). Although most research has focused on how anxiety affects ADHD symptoms, surprisingly little research has examined whether the presence of ADHD affects the nature of anxiety symptoms. One reason for this lack of research is that most studies of ADHD have not included comparison group with anxiety disorders only. In one study that did use such a comparison group, Manassis, Tannock, and Barbosa (2000) found that children with ADHD and anxiety showed reduced auditory emotion recognition using a dichotic listening task (i.e., recognizing the emotional tone of a word target). In a subsequent study, Manassis, Tannock, Young, and Francis-John (2007) found that children with ADHD and anxiety had less auditory perception of anger. Although these studies addressed emotional perception differences, they did not address differences in types of anxiety symptomatology. These two studies also reported mean scores for general measures of child anxiety but did not examine specific aspects of anxiety (e.g., physical symptoms, social anxiety, etc.). As noted by Tannock (2009), future studies are needed comparing children with ADHD and anxiety to children with pure anxiety disorders.

In addition to effects on symptomatology, there has also been substantial interest in how anxiety affects neuropsychological functioning in children with ADHD. This literature has also been mixed. Some studies have found that children with ADHD and anxiety tend to perform better on tasks measuring inhibitory control such as continuous performance tests (Manassis et al., 2000; Pliszka, 1992; Pliszka, Hatch, Borcherding, & Rogeness, 1993), whereas other studies have not supported or provided only partial support for this finding (Epstein, Goldberg, Conners, & March, 1997; Newcorn et al., 2001; Sorensen, Plessen, Nicholas, & Lundervold, 2011). In addition to inhibitory control, studies have examined other neuropsychological functions such as sustained attention and working memory. For example, one study found that children with ADHD and anxiety showed enhanced sustained attention and selective attention on laboratory tasks (Vloet, Konrad, Herpetz-Dahlmann, Polier, & Gunther, 2010). In relation to working memory, there is some evidence to indicate that children with ADHD and anxiety tend to show greater working memory impairments and/or working memory variability (Pliszka, 1989; Skirbekk, Hansen, Oerbeck, & Kristensen, 2011; Tannock, Ickowicz, & Schachar, 1995) along with poorer working memory changes following a stimulant medication trial (Bedard & Tannock, 2008; Tannock et al., 1995). Overall, one possibility is that children with ADHD and anxiety tend to perform worse on more cognitively complex and mentally effortful tasks. For example, anxiety at moderate levels may serve to enhance vigilance and regulate impulses but may disrupt more cognitively effortful processes such as working memory (Tannock, 2009).

One confound in many of these past studies is that groups with anxiety have frequently included children with and without comorbid ODD/CD. This confound is important, as the presence of ODD/CD has unique effects on symptomatology and neurocognitive functioning (Newcorn et al., 2001). Based on the Multimodal Treatment of ADHD Study (MTA), Jensen et al. (2001) advocated for the examination of more fine-grained comorbid subgroups of ADHD in future studies: ADHD Only, ADHD + anxiety (ANX), ADHD + ODD/CD, and ADHD + ODD/CD + ANX. Although potentially valuable for understanding subgroup characteristics, this nosology poses significant assessment challenges. For example, many studies of ADHD rely on parent and teacher report for the assessment of ADHD and other co-occurring symptom domains. Although these practices are consistent with evidence-based assessment guidelines for ADHD (Pelham, Fabiano, & Massetti, 2005), guidelines for anxiety emphasize the importance of including child report when assessing anxiety (Silverman & Ollendick, 2005). Reports of parents and children might lead to different conclusions. This assessment conundrum was observed in the MTA. For example, diagnostic groups involving anxiety in the MTA Study were defined solely by parent report of anxiety, an approach that may have compromised the validity of the anxiety findings (March et al., 2000). March et al. (2000) reexamined MTA data and found that parent-reported anxiety on the Diagnostic Interview Schedule for Children (DISC; Shaffer et al., 1996), the diagnostic interview used for defining MTA groups, appeared to measure aspects of broader negative affectivity rather than fear and phobic anxiety. March et al. (2000) noted that a diagnostic tool oriented toward anxiety disorders such as the Anxiety Disorders Interview Schedule for Children (ADIS-C; Silverman & Albano, 1996) would have been valuable for anxiety assessment. In addition, the ADIS-C is a semistructured interview that allows for greater clinical clarification regarding the nature of symptomatology (e.g., differentiating frustration/irritability from fear/phobic anxiety). These assessment issues are critical in clearly identifying the presence of fear and phobic anxiety in children with ADHD and better understanding how the presence of true anxiety affects ADHD and related deficits.

Given some of the limitations of past studies of ADHD and anxiety comorbidity, the current study sought to clarify relationships among a subset of comorbid ADHD subgroups proposed by Jensen et al. (2001). In particular, we used the ADIS-C to guide diagnostic decisions, as suggested by March et al. (2000). Although the ADIS-C has shown evidence for the valid assessment of child anxiety, it should be noted that it has also shown reliability and validity in the assessment of ADHD (Jarrett, Wolff, & Ollendick, 2007; Silverman & Albano, 1996). In turn, such a measure may be particularly valuable for examining ADHD and anxiety comorbidity.

The current study had a series of goals. The first goal of the study was to accurately assess child anxiety disorders to draw sharper conclusions regarding the effects of fear and phobic anxiety on ADHD and related impairments. This goal was accomplished by using the ADIS-C, a measure that has been recommended for the evidence-based assessment of child anxiety (Silverman & Ollendick, 2005). To avoid ODD/CD as a potential confound (March et al., 2000; Newcorn et al., 2001), we excluded ODD/CD from all groups in our study. Second, we sought to examine whether parent- and teacher-reported symptoms of ADHD (e.g., attention problems), and internalizing problems varied based on the presence of comorbid anxiety. Third, we sought to evaluate differences in neurocognitive characteristics on tasks measuring sustained attention, inhibitory control, and working memory. Finally, we sought to evaluate the nature of internalizing symptoms in children with ADHD and whether these symptoms differ from children with pure anxiety disorders. Although our primary study question was related to anxiety symptom domains, we also included measures of depression. Tannock (2009) noted that very few studies have considered the presence of comorbid depression, which has also been linked to working memory deficits (Ladouceur et al., 2005). In turn, our study used measures of depression to determine whether differences on depression symptomatology might explain the differences between our groups. Strengths of the current study included the use of the ADIS-C for defining the presence of anxiety, the use of more fine-grained diagnostic groups that do not include ODD/CD, and the use of an anxiety disorders only group, a group that has rarely been examined in past studies of ADHD comorbidity.

Method

Children and their parent(s) presented at an outpatient clinic in southwestern Virginia for a psychoeducational assessment. Children were consecutive referrals from community pediatricians, schools, and mental health professionals. Two clinicians were assigned to each case. One clinician interviewed the parent(s) and administered parent measures, whereas the second clinician separately and independently completed the child assessment. All children taking stimulant medication were instructed to not take stimulant medication on the day of assessment. Families were pursuing clinical assessments, but all agreed to let their de-identified data be used for research purposes. No specific inclusion criteria were used for assessment participation. Institutional Review Board approval, child assent, and parent consent were obtained prior to the start of the assessment. Data were collected over approximately 8 years.

Consensus diagnoses during a consensus team meeting¹ were established based on independent child and parent Anxiety Disorders Interview Schedule for DSM-IV–Child and Parent Versions (ADIS-C/P) interviews (see Grills & Ollendick, 2003, for details). For diagnoses of ADHD, teacher report of ADHD symptoms on the Teacher Report Form (TRF) was used to supplement ADIS-C findings from parent and child interviews for determining the diagnosis of ADHD, given the critical role that teachers play in the diagnosis of ADHD (Pelham et al., 2005). A total of 134 children were selected from 407 clinically referred children based on the following criteria: diagnosis of DSM-IV ADHD, Combined Type (ADHD-C) and/or a diagnosis of generalized anxiety disorder, social phobia, separation anxiety disorder, or a specific phobia, Full Scale IQ (FSIQ) ≥80, and no diagnosis of pervasive developmental disorder, schizophrenia, or bipolar disorder.²

Mean age of the sample was 9.92 (SD = 2.73; range = 6-17). In all, 61.9% of the sample were male. Mean IQ was 99.98 (SD = 12.43) and family income was US$53,047 (SD = US$40,374). A total of 94% of the sample were Caucasian. Three groups were formed for study purposes: ADHD (n = 41; 30.6%), ADHD + ANX (n = 31; 23.1%), and ANX Only (n = 62; 46.27%). For children in the ADHD + ANX group (n = 31), the following DSM-IV anxiety diagnoses were present: obsessive-compulsive disorder (n = 2; 6.45%), specific phobias (n = 11; 35.48%), posttraumatic stress disorder (PTSD; n = 2; 6.45%), social phobia (n = 7; 22.58%), separation anxiety (n = 7; 22.58%), and generalized anxiety (n = 17; 54.84%).³ For children in the ANX Only group (n = 62), the following diagnoses were present: obsessive-compulsive disorder (n = 5; 8.06%), specific phobias (n = 11; 17.74%), PTSD (n = 0; 0%), social phobia (n = 26; 41.94%), separation anxiety (n = 7; 11.29%), and generalized anxiety (n = 36; 58.06%). The ADHD + ANX group was compared with the ANX Only group on the rate of specific phobias, social phobia, separation anxiety, and generalized anxiety. No differences were found across groups, χ²(2) = 6.55, p = .09. All three groups had similar rates of clinically diagnosed mood disorders (i.e., major depressive disorder or dysthymia): ADHD = 2 of 41 (4.88%), ADHD + ANX = 1 of 31 (3.23%), and ANX = 3 of 62 (4.84%). The low base rate of depression precluded a statistical test of these group differences. Given the high comorbidity of learning disorders with ADHD, groups were also compared on the presence of a learning disorder (i.e., reading disorder, mathematics disorder, disorder of written expression, and learning disorder, not otherwise specified). Rates for ADHD Only (n = 12; 29.27%), ADHD + ANX (n = 7; 22.58%), and ANX Only (n = 21; 33.87%) were comparable with no significant differences found across groups, χ²(2) = 1.27, p = .53. Finally, rates of psychotropic medications were also compared. Stimulant medication rates differed across groups, χ²(2) = 9.22, p < .05, with the following rates identified: ADHD Only (n = 10; 24.39%), ADHD + ANX (n = 11; 35.48%), and ANX Only (n = 6; 9.68%). Although the two ADHD groups did not differ on stimulant usage, the ADHD + ANX group differed from the ANX Only group (odds ratio = 5.0). Rates of usage for the nonstimulant, atomoxetine, were comparable with somewhat higher representation in the ADHD Only group (n = 6; 14.63%) in comparison to ADHD + ANX (n = 1; 3.23%) and ANX Only (n = 2; 3.23%). Rates of medication for anxiety and mood-related medications (e.g., selective-serotonin reuptake inhibitors, other antidepressants, etc.) were generally low: ADHD Only (n = 2; 4.88%), ADHD + ANX (n = 7; 22.58%), and ANX Only (n = 6; 9.84%). ADHD + ANX had a higher rate of usage of these medications. Children taking anxiety and mood-related medications were not instructed to discontinue medication for the assessment.

ADIS-C/P

The ADIS-C/P (Silverman & Albano, 1996) versions are semistructured interviews for diagnosis of most child psychiatric disorders (including all child disorders except pervasive developmental disorders, schizophrenia, and bipolar disorder). Symptoms and ratings are used to identify diagnostic criteria and create a clinician’s severity rating (CSR). A CSR of 4 or above (0-8) indicates a diagnosable condition. As noted earlier, a team consensus diagnosis approach was used for defining groups in the current study. Recent examination of the ADIS-C/P (for DSM-IV) has yielded acceptable to excellent test–retest reliability estimates regarding child and parent diagnoses (κ = .61-1.00; Silverman, Saavedra, & Pina, 2001). Trained clinical psychology doctoral students, blind to purposes of the study, conducted the interviews. Acceptable levels of interrater agreement for video review were found for child (κ = .71) and parent (κ = .77) interviews (Grills & Ollendick, 2003).

Child Behavior Checklist (CBCL)

The CBCL (Achenbach, 2001a) is a 113-item parent-report measure with acceptable reliability and validity (Achenbach, 2001a). T-scores were used for the current study (clinical cutoff = 65).

Children’s Depression Inventory (CDI)

The CDI (Kovacs, 1985) is a 27-item child self-report questionnaire assessing symptoms of depression. The scale has acceptable reliability and validity (Kovacs, 1985). T-scores were used for the current study (clinical cutoff = 65).

Conners’ Continuous Performance Test (CPT)

The CPT (Conners, 1994) is a vigilance task in which respondents are asked to press the space bar when any letter other than “X” appears on the computer screen. Children completed the task off of stimulant medication for ADHD (if applicable). The overall index score was used for the study (clinical cutoff = 11) along with T-scores for mean reaction time, reaction time standard deviation, and response variability.

Multidimensional Anxiety Scale for Children (MASC)

The MASC (March, 1998) is a 45-item child self-report questionnaire for symptoms of anxiety. The scale has acceptable reliability and validity (March, 1998). T-scores were used for the current study (clinical cutoff = 65).

TRF

The TRF (Achenbach, 2001b) is a 113-item teacher-report measure with acceptable reliability and validity (Achenbach, 2001b). T-scores were used for the current study (clinical cutoff = 65).

Wechsler Intelligence Scale for Children (WISC)

The WISC (Wechsler, 1991, 2003) is an individually administered clinical instrument for assessing the intellectual ability of children aged 6 years through 16 years and 11 months. The mean score for the FSIQ is 100 with a standard deviation of 15. Evidence for adequate reliability and validity of the WISC has been documented (Prifitera, Saklofske, & Weiss, 2005; Wechsler, 1991). The FSIQ was used in the current study to screen for overall cognitive functioning (i.e., IQ <80). In addition, the Digits Span Backward subtest scaled score was also used as a measure of working memory. Given the time frame of the study, which extended over several years (2001-2007), the WISC-III (1991) and WISC-IV (2003) were used to assess intellectual ability.

Results

Data were analyzed for all participants who met inclusion criteria (n = 134). Missing data were handled using listwise deletion. Missing data for parent- and child-report data were generally limited. The MASC was added to the assessment protocol later in the study, so smaller cell sizes for the MASC reflect this later addition rather than missing data. Missing data were more substantial for teacher report due primarily to assessments that were conducted over the summer months: 30 of 41 (73.17%) teachers returned the measure in the ADHD group, 29 of 31 (93.55%) for ADHD + ANX, and 47 of 61 (77.05%) for ANX. Return rates did not differ by group, χ²(2) = 0.36, p = .83. Study analyses included chi-square analyses for categorical data and ANOVA for continuous data. ANOVAs involving measures with subfactors used a Bonferroni correction for determining significance level. For ANOVA results, significant differences were followed with post hoc comparisons using Hochberg’s GT2 given its strength for samples with unequal cell sizes (Field, 2009). Effect sizes for analyses are reported in terms of η² in Table 1 and in terms of Cohen’s d in the text for post hoc comparisons.⁴

Table 1.

Means, SDs, and Tests Comparing ADHD Only, ADHD + ANX, and ANX Only.

ADIS-C	ADHD (n = 39)	ADHD + ANX (n = 30)	ANX (n = 61)	H	df	p	η²
ADHD–Parent CSR	4.18 (3.13)^b	4.73 (2.68)^b	1.21 (2.42)^a	32.74	2	.01	NA
CBCL	ADHD (n = 40)	ADHD + ANX (n = 29)	ANX (n = 61)	F	df	p	η²
Anxiety/depression	59.83 (8.52)^a	66.59 (9.29)^b	62.93 (10.21)^b	4.26	2, 127	.02	.06
Withdrawn/depressed	58.10 (7.30)	61.14 (9.92)^a	61.80 (8.63)	2.36	2, 127	.10	.04
Attention problems	70.73 (9.85)^b	72.24 (9.87)^b	66.08 (8.73)^a	5.39	2, 127	.01	.08
TRF	ADHD (n = 30)	ADHD + ANX (n = 25)	ANX (n = 47)	F	df	p	η²
Anxiety/depression	58.33 (9.60)	58.44 (10.91)	57.53 (9.69)	0.09	2, 99	.91	.01
Withdrawn/depressed	55.17 (6.89)	57.20 (7.77)	58.66 (8.25)	1.86	2, 99	.16	.04
Attention problems	66.33 (8.75)^b	64.12 (10.17)^b	60.00 (5.96)^a	6.14	2, 99	.01	.11
CPT	ADHD (n = 34)	ADHD + ANX (n = 29)	ANX (n = 57)	F	df	p	η²
Overall index	14.99 (6.52)^b	14.10 (6.82)^b	7.33 (7.44)^a	15.97	2, 121	.01	.21
Hit rate	49.47 (14.10)	52.13 (16.13)	47.75 (13.71)	0.93	2, 127	.40	.01
Hit rate standard error	71.36 (16.11)^b	72.37 (13.05)^b	61.66 (15.75)^a	7.15	2, 127	.01	.10
Variability	66.52 (14.38)^b	67.74 (12.66)^b	57.47 (12.22)^a	8.88	2, 127	.01	.12
Working memory	ADHD (n = 32)	ADHD + ANX (n = 24)	ANX (n = 49)	F	df	p	η²
Digit Span Backward	7.38 (3.35)^b	4.58 (2.80)^a	5.98 (2.77)^a,b	6.14	2, 102	.01	.11
CDI	ADHD (n = 36)	ADHD + ANX (n = 29)	ANX (n = 59)	F	df	p	η²
Total score	49.17 (10.36)	53.14 (12.42)	52.41 (10.04)	1.39	2, 121	.25	.02
MASC	ADHD (n = 34)	ADHD + ANX (n = 25)	ANX (n = 53)	F	df	p	η²
Physical symptoms	47.65 (10.00)^a	57.36 (14.94)^b	51.91 (10.60)^a,b	5.11	2, 109	.01	.09
Social anxiety	49.74 (12.74)	56.76 (14.17)	55.30 (12.46)	2.68	2, 109	.07	.05
Separation	53.06 (11.62)	60.12 (13.82)	53.96 (13.61)	2.45	2, 109	.09	.04
Harm avoidance	48.21 (11.87)	49.96 (13.44)	49.72 (12.10)	0.20	2, 109	.82	.01

Note. ADIS-C = Anxiety Disorders Interview Schedule for Children; ANX = anxiety disorder; CBCL = Child Behavior Checklist; CDI = Children’s Depression Inventory; CPT = Conners’ Continuous Performance Test; CSR = clinician’s severity rating; MASC = Multidimensional Anxiety Scale for Children; TRF = Teacher Report Form. Groups do not total to 134 due to missing data. Letters in superscript indicate statistically significant differences between groups (p < .05). Cohen (1988) recommended the following interpretation of η²: small = .01 to .05, medium = .06 to .13, large = .15 or greater.

Groups did not differ by gender, χ²(2) = 4.24, p = .12. One-way ANOVAs failed to show differences on IQ, F(2, 126) = 1.07, p = .35, or family income, F(2, 106) = 0.10, p = .91. Group differences did emerge for age, F(2, 131) = 4.84, p < .05. Post hoc tests revealed that the ANX Only group (M = 10.68, SD = 2.91) was significantly older (d = 0.60) than was the ADHD + ANX group (M = 9.10, SD = 2.30).

Table 1 reflects results of the primary study analyses. Although groups did differ on age, we chose to not control for age given that ANCOVA is appropriate only in those situations where a covariate is not confounded with group assignment (e.g., in situations of random assignment). In addition, our outcome variables involved age-corrected T-scores. For all analyses in Table 1, Levene’s test was first used for ANOVA to examine the homogeneity of variance assumption of ANOVA. With the exception of one analysis, no violations of homogeneity of variance were noted. The one exception that did violate homogeneity was the comparison of groups on the parent interview CSR for ADHD from the ADIS-C. For this analysis, a nonparametric Kruskal–Wallis test was used followed by pairwise comparisons with a Bonferroni correction.

On the CBCL, differences were found among groups for attention problems, F(2, 127) = 5.39, p = .01. Post hoc tests revealed significant differences between ANX Only and ADHD + ANX (d = 0.15) and between ANX Only and ADHD Only (d = 0.50). ADHD + ANX and ADHD Only did not differ. It should be noted, though, that all three groups were in the clinical range (i.e., T > 65; see Table 1). On the CBCL, differences also emerged for anxiety/depression, F(2, 127) = 4.26, p < .05, but not withdrawn/depressed, F(2, 127) = 2.36, p = .10. Post hoc analyses revealed that the ANX Only and the ADHD + ANX groups significantly differed from the ADHD Only group (d = 0.76 and d = 0.33, respectively). ADHD + ANX and ANX Only did not differ, although only ADHD + ANX was in the clinical range.

On the TRF, differences were found among groups for attention problems, F(2, 99) = 6.14, p < .01. Post hoc tests revealed significant differences between ADHD + ANX and ANX Only (d = 0.49) and between ADHD Only and ANX Only (d = 0.14). ADHD + ANX and ADHD Only did not differ. It should be noted that only the ADHD Only group was rated in the clinical range (i.e., T > 65). Differences did not emerge for anxiety/depression, F(2, 99) = 0.09, p = .91, or withdrawn/depressed, F(2, 99) = 1.86, p = .16.

On the neurocognitive tasks, an interesting set of findings was observed. On the CPT, differences were found among groups on the overall index score, F(2, 121) = 15.97, p < .01. Post hoc tests revealed significant differences between ADHD + ANX and ANX Only (d = 0.95) and ADHD Only and ANX Only (d = 1.10). ADHD + ANX and ADHD Only did not differ. It should also be noted that the ADHD Only group and ADHD + ANX group were in the clinical range (i.e., overall index score > 11). Although no differences emerged on the overall hit rate, F(2, 127) = 0.93, p = .40, differences did emerge for the overall hit rate standard error, F(2, 127) = 7.15, p < .01, and response variability, F(2, 127) = 8.88, p < .01, indices. Post hoc tests revealed significant differences between ADHD + ANX and ANX Only (d = 0.74) and ADHD Only and ANX Only (d = 0.61) on overall hit rate standard error. Similarly, post hoc tests revealed significant differences between ADHD + ANX and ANX Only (d = 0.83) and ADHD Only and ANX Only (d = 0.68) on response variability. Finally, on the WISC, differences between groups were examined for the Digit Span Backward scaled score as a measure of working memory. Significant differences were found among the groups, F(2, 102) = 6.14, p < .01. Post hoc tests revealed a significant difference between ADHD + ANX and ADHD Only (d = 0.91).

Group differences on child-reported internalizing symptoms were also examined. On the CDI, no differences were found among groups on the CDI Total Score, F(2, 121) = 1.39, p = .25. On the MASC, specific anxiety domains were examined, as we were interested in whether the nature of anxiety symptomatology is affected by the presence of ADHD. No differences were found for harm avoidance, F(2, 109) = .20, p < .05, social anxiety, F(2, 109) = 2.68, p = .07, or separation/panic, F(2, 109) = 2.45, p = .09. However, group differences were found for physical symptoms, F(2, 109) = 5.11, p < .01. Post hoc analyses revealed a significant difference between ADHD + ANX and ADHD Only (d = 0.76). No group was in the clinical range. To further explore differences in physical symptoms, the subfactors of tense/restless and somatic/autonomic were examined. No differences were found for somatic/autonomic, F(2, 109) = 2.97, p = .06, but differences were found for tense/restless, F(2, 109) = 5.58, p < .01. Post hoc analyses revealed a significant difference between ADHD + ANX and ADHD Only (d = 0.81). No group was in the clinical range on tense/restless or somatic/autonomic.

In addition to the previously mentioned differences, groups were also compared on the CSR for ADHD-C from the parent interview only, given the limited utility of child report for ADHD. Significant differences were noted among the groups, H(2) = 32.74, p < .05. Post hoc tests revealed significant differences between ADHD Only and ANX Only (d = 1.12) and ADHD + ANX and ANX Only (d = 1.27). ADHD Only and ADHD + ANX did not differ, suggesting that the groups did not differ on ADHD severity and impairment.

Discussion

Overall, the current study sought to examine differences in clinical characteristics within a subset of comorbid subgroups of ADHD-C proposed by Jensen et al. (2001). Unique aspects of the study included the use of a multi-informant evidence-based assessment tool for child anxiety (i.e., the ADIS-C), the examination of groups without comorbid ODD/CD, and an ANX Only comparison group, a group that has rarely been used in past studies of ADHD comorbidity.

In relation to ADHD symptomatology, little evidence was found for the contention that children with ADHD + ANX differ from children with ADHD Only on ADHD-related symptom severity and impairment. Both groups showed clinically significant (or very close to significant) attention problems on the CBCL and TRF and differed from the ANX Only group. In addition, the ADHD Only and ADHD + ANX groups did not differ on clinician-rated ADHD severity and impairment from the ADIS-C parent interview. Although past research has shown some mild effects for greater inattention relative to impulsivity in children with ADHD + ANX, our results generally coincide with the conclusion by Tannock (2009) that past studies “have provided little evidence to date that the presence of an anxiety disorder alters the clinical presentation of ADHD in a systematic way” (p. 136).

For internalizing symptoms, although we found some differences between our ADHD Only group and our ADHD + ANX group, we found limited evidence that children with ADHD + ANX systematically differ from children with ANX Only on anxiety and depression symptomatology. In relation to comorbid anxiety disorders, no significant differences were found between ADHD + ANX and ANX Only on diagnosis rates of specific comorbid ANXs. In addition, mood disorder diagnosis was generally low (i.e., 3% for ADHD + ANX; 4% for ANX), and neither group showed clinically significant mean T-scores on the CDI and did not differ. Thus, our findings do not appear to be confounded by differences in mood disorders.

On other ratings of internalizing symptomatology, children with ADHD + ANX did not significantly differ on anxiety and depression symptoms on the CBCL or the TRF, and limited evidence was found for differences in child-reported anxiety symptoms on the MASC. The only exception was a significant difference emerging between ADHD + ANX and ADHD Only for physical symptoms of anxiety, with the ADHD + ANX group reporting more symptomatology. At the same time, ADHD + ANX and ANX Only did not differ. In turn, these findings, much like the findings for ADHD, generally support the notion that children with ADHD + ANX and ANX Only do not appear to differ on anxiety symptoms in a systematic way. In relation to all groups showing subclinical MASC levels, this finding is not uncommon. For example, child awareness of anxiety may be limited prior to treatment, and many child anxiety studies that have diagnosed children with anxiety based on a diagnostic interview show pretreatment MASC levels in the subclinical range (Ollendick et al., 2009; Wood, Piacentini, Southam-Gerow, Chu, & Sigman, 2006).

Finally, our study also sought to examine differences in neurocognitive characteristics.

Overall, groups with ADHD appeared to be more impaired in terms of their performance on the CPT. Both groups with ADHD showed clinically elevated overall index scores, whereas the ANX Only group did not. Although the three groups did not differ on mean reaction time, groups with ADHD were much more variable in their responses, consistent with research showing variable CPT responding in children with ADHD (Epstein et al., 2003). Although evidence was not found for ADHD Only and ADHD + ANX differences on this measure of sustained attention and inhibitory control, we did find group differences on a measure of working memory, Digit Span Backward on the WISC. The group with ADHD + ANX differed from ADHD Only, and the effect was in the large range (d = 0.91). This finding converges with past evidence showing that children with ADHD + ANX show greater working memory impairments than children with ADHD Only (Pliszka, 1989; Tannock et al., 1995). It should be noted, though, that mean scores were well below average for all three of the groups, suggesting that working memory impairments may be present in ADHD and anxiety disorders. Interestingly, however, our ADHD + ANX group and ANX Only group did not differ on working memory, suggesting that anxiety alone impairs working memory. Although working memory deficits have long been noted in ADHD (Martinuseen, Hayden, Hogg-Johnson, & Tannock, 2005), few studies have examined the neurocognitive characteristics of children with anxiety. Working memory may be an important area to examine further in future studies of child anxiety.

Overall, our results indicate that children with ADHD + ANX do not appear to differ from children with ADHD Only in a systematic way on symptoms of ADHD and that children with ADHD + ANX do not appear to differ in a systematic way from children with pure anxiety disorders on symptoms of anxiety. At the same time, we did find evidence that children with ADHD + ANX exhibited greater working memory deficits than did children with ADHD Only. It appears that children with ADHD + ANX are a particularly impaired group of children in terms of cognitive functioning with additional deficits in working memory. Although some past studies have not found such differences (e.g., Manassis et al., 2007), our study removed the confound of comorbid ODD/CD in our diagnostic groups and used an anxiety assessment process that is more in line with evidence-based assessment recommendations for child anxiety (Silverman & Ollendick, 2005). In addition, some other studies that have not found differences in working memory deficits have used parent- and teacher-report measures of working memory such as the Behavior Rating Inventory of Executive Functioning (Sorensen et al., 2011). Given the mixed literature to date on the effects of anxiety on neurocognitive functioning in relation to ADHD, it may be important for future studies to also consider state-related anxiety and its effects on performance-based tasks. A large body of research on normal populations has supported an important role for state anxiety in cognitive processing (Eysenck, Derakshan, Santos, & Calvo, 2007). At moderate levels, as shown many years ago, anxiety may serve to enhance effort and facilitate enhanced performance, but at more extreme levels, anxiety may overload cognitive resources, which would lead to decreased performance. Importantly, as noted by Eysenck et al. (2007), anxiety effects may not always appear on task performance but may be reflected in greater allocation of cognitive resources, which could be detected via psychophysiological measures. Future studies will need to incorporate such measures to better understand the effect of anxiety on ADHD and related impairments. Finally, anxiety type may be important to consider in neurocognitive task performance. For example, Epstein et al. (1997) found that self-reported physiological anxiety was associated with enhanced response inhibition, whereas cognitive anxiety (e.g., worry) was associated with decreased response inhibition. Future research might combine such methods as self-reported anxiety in combination with physiological indicators during laboratory task performance.

Limitations of the current study include a relatively small, largely Caucasian, clinical sample. In addition, our sample was relatively young, so results might not generalize to older children with ADHD and/or anxiety disorders. Our study also focused only on ADHD-C. It will be important for future studies to examine comorbid subgroups of other ADHD subtypes. Our study also did not include pure measures of inattention and hyperactivity/impulsivity, so we were unable to examine differences on these specific symptom domains of ADHD. In addition, our study only focused on a subset of comorbid ADHD groups proposed by Jensen et al. (2001). Future studies will be needed to compare additional comorbid subgroups of ADHD. Despite these limitations, the study provides needed additional data on the characteristics of children with ADHD and anxiety, and may provide important clues for future research.

Finally, it will be important for future studies to further examine treatment response for ADHD in relation to the presence of anxiety. As noted by Jensen et al. (2001) in the MTA Study, the presence of anxiety was rarely associated with differences in baseline characteristics but frequently interacted with other variables in relation to treatment response. An important treatment target for children with ADHD and anxiety may be cognitive deficits. For example, the current study suggests that children with ADHD and anxiety are a particularly impaired subgroup of children with ADHD in terms of cognitive functioning. In turn, interventions designed to treat underlying cognitive deficits such as working memory training (Klingberg et al., 2005) may prove beneficial in treating the neuropsychological impairments of children with ADHD and anxiety.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Matthew A. Jarrett is an Assistant Professor in the Department of Psychology at the University of Alabama.

Jennifer C. Wolff is an Assistant Professor (Research) at the Warren Alpert Medical School of Brown University, Staff Psychologist at Rhode Island Hospital, and Director of the Adolescent Mood Clinic.

Thompson E. Davis, III, is an Associate Professor in the Department of Psychology at Louisiana State University.

Maria J. Cowart is a Postdoctoral Research Associate at the Virginia Tech Child Study Center.

Thomas H. Ollendick is a University Distinguished Professor in the Department of Psychology at Virginia Tech. Dr. Ollendick is also the Director of the Child Study Center.

References

Abikoff

H. B.

Jensen

P. S.

Arnold

L. L.

Hoza

Hechtman

Pollack

. . . Wigal

(2002). Observed classroom behavior of children with ADHD: Relationship to gender and comorbidity. Journal of Abnormal Child Psychology, 30, 349-359.

Achenbach

T. M.

(2001a). Manual for the Child Behavior Checklist 4-18 and 2001 Profile. Burlington: University of Vermont, Department of Psychiatry.

Achenbach

T. M.

(2001b). Manual for the Teacher Report Form 4-18 and 2001 Profile. Burlington: University of Vermont, Department of Psychiatry.

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.

Bedard

A. C.

Tannock

(2008). Anxiety, methylphenidate response, and working memory in children with ADHD. Journal of Attention Disorders, 11, 546-557.

Cohen

(1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum.

Conners

C. K.

(1994). The Conners Continuous Performance Test. Toronto, Ontario, Canada: Multi-Health Systems.

Epstein

J. N.

Erkanli

Conners

C. K.

Klaric

Costello

J. E.

Angold

(2003). Relations between continuous performance test performance measures and ADHD behaviors. Journal of Abnormal Child Psychology, 31, 543-554.

Epstein

J. N.

Goldberg

N. A.

Conners

C. K.

March

J. S.

(1997). The effects of anxiety on continuous performance test functioning in an ADHD clinic sample. Journal of Attention Disorders, 2, 45-52.

10.

Eysenck

M. W.

Derakshan

Santos

Calvo

M. G.

(2007). Anxiety and cognitive performance: Attentional control theory. Emotion, 7, 336-353.

11.

Field

(2009). Discovering statistics using SPSS (3rd ed.). London, England: Sage.

12.

Grills

A. E.

Ollendick

T. H.

(2003). Multiple informant agreement and the Anxiety Disorders Interview Schedule for Parents and Children. Journal of the American Academy of Child and Adolescent Psychiatry, 42, 30-40.

13.

Jarrett

M. A.

Ollendick

T. H.

(2008). A conceptual review of the comorbidity of attention-deficit/hyperactivity disorder and anxiety: Implications for future research and practice. Clinical Psychology Review, 28, 1266-1280.

14.

Jarrett

M. A.

Wolff

J. C.

Ollendick

T. H.

(2007). Concurrent validity and informant agreement of the ADHD module of the Anxiety Disorders Interview Schedule for DSM-IV. Journal of Psychopathology and Behavioral Assessment, 29, 159-168.

15.

Jensen

P. S.

Hinshaw

S. P.

Kraemer

H. C.

Lenora

Newcorn

J. H.

Abikoff

H. B.

. . . Vitiello

(2001). ADHD comorbidity findings from the MTA Study: Comparing comorbid subgroups. Journal of the American Academy of Child & Adolescent Psychiatry, 40, 147-158.

16.

Klingberg

Fernell

Olesen

Johnson

Gustaffson

Dahlstrom

. . . Westerberg

(2005). Computerized training of working memory in children with ADHD—A randomized, controlled trial. Journal of the American Academy of Child & Adolescent Psychiatry, 44, 177-186.

17.

Kovacs

(1985). The Children’s Depression Inventory (CDI). Psychopharmacology Bulletin, 21, 995-998.

18.

Ladouceur

C. D.

Dahl

R. E.

Williamson

D. E.

Birmaher

Ryan

N. D.

Casey

B. J.

(2005). Altered emotional processing in pediatric anxiety, depression, and comorbid anxiety-depression. Journal of Abnormal Child Psychology, 33, 165-177.

19.

Manassis

Tannock

Barbosa

(2000). Dichotic listening and response inhibition in children with comorbid anxiety disorders and ADHD. Journal of the American Academy of Child & Adolescent Psychiatry, 39, 1152-1159.

20.

Manassis

Tannock

Young

Francis-John

(2007). Cognition in anxious children with attention deficit hyperactivity disorder: A comparison with clinical and normal children. Behavioral and Brain Functions, 3, 4.

21.

March

(1998). Manual for the Multidimensional Anxiety Scale for Children (MASC). Toronto, Ontario: Multi-Health Systems.

22.

March

J. S.

Swanson

J. M.

Arnold

L. E.

Hoza

Conners

C. K.

Hinshaw

S. P.

. . . Pelham

W. E.

(2000). Anxiety as a predictor and outcome variable in the Multimodal Treatment Study of Children With ADHD (MTA). Journal of Abnormal Child Psychology, 28, 527-541.

23.

Martinuseen

Hayden

Hogg-Johnson

Tannock

(2005). A meta-analysis of working memory impairments in children with attention-deficit/hyperactivity disorder. Journal of the American Academy of Child & Adolescent Psychiatry, 44, 377-384.

24.

MTA Cooperative Group. (1999). Moderators and mediators of treatment response for children with attention-deficit/hyperactivity disorder: The Multimodal Treatment Study of Children With Attention-Deficit Hyperactivity Disorder. Archives of General Psychiatry, 56, 1088-1096.

25.

Newcorn

J. H.

Halperin

J. M.

Jensen

P. S.

Abikoff

H. B.

Arnold

L. E.

Cantwell

D. P.

. . . Vitiello

(2001). Symptom profiles in children with ADHD: Effects of comorbidity and gender. Journal of the American Academy of Child & Adolescent Psychiatry, 40, 137-146.

26.

Newcorn

J. H.

Halperin

J. M.

Miller

C. J.

(2009). ADHD with oppositionality and aggression. In Brown

T. E.

(Ed.), ADHD comorbidities: Handbook for ADHD complications in children and adults (pp. 157-176). Washington, DC: American Psychiatric Publishing.

27.

Ollendick

T. H.

Öst

L. G.

Reuterskiöld

Costa

Cederlund

Sirbu

. . . Jarrett

M. A.

(2009). One-session treatment of specific phobias in youth: A randomized clinical trial in the United States and Sweden. Journal of Consulting and Clinical Psychology, 77, 504-516.

28.

Pelham

W. E.

Fabiano

G. A.

Massetti

G. M.

(2005). Evidence-based assessment of attention deficit hyperactivity disorder in children and adolescents. Journal of Clinical Child & Adolescent Psychology, 34, 449-476.

29.

Pliszka

S. R.

(1989). Effect of anxiety on cognition, behavior, and stimulant response in ADHD. Journal of the American Academy of Child & Adolescent Psychiatry, 28, 882-887.

30.

Pliszka

S. R.

(1992). Comorbidity of attention-deficit hyperactivity disorder and overanxious disorder. Journal of the American Academy of Child & Adolescent Psychiatry, 31, 197-203.

31.

Pliszka

S. R.

Hatch

J. P.

Borcherding

S. H.

Rogeness

G. A.

(1993). Classical conditioning in children with attention deficit hyperactivity disorder (ADHD) and anxiety disorders: A test of Quay’s model. Journal of Abnormal Child Psychology, 21, 411-423.

32.

Prifitera

Saklofske

D. H.

Weiss

(Eds.). (2005). WISC-IV: Clinical use and interpretation. New York, NY: Elsevier Science.

33.

Shaffer

Fisher

Dulcan

M. K.

Davies

Piacentini

Schwab-Stone

M. E.

. . . Regier

D. A.

(1996). The NIMH Diagnostic Interview Schedule for Children Version 2.3 (DISC-2.3): Description, acceptability, prevalence rates, and performance in the Methods for the Epidemiology of Child and Adolescent Mental Disorders (MECA) Study. Journal of the American Academy of Child & Adolescent Psychiatry, 35, 865-877.

34.

Silverman

W. K.

Albano

A. M.

(1996). Manual for the ADIS-IV-C/P. New York, NY: Psychological Corporation.

35.

Silverman

W. K.

Ollendick

T. H.

(2005). Evidence-based assessment of anxiety and its disorders in children and adolescents. Journal of Clinical Child & Adolescent Psychology, 34, 380-411.

36.

Silverman

W. K.

Saavedra

L. M.

Pina

A. A.

(2001). Test-retest reliability of anxiety symptoms and diagnoses with the Anxiety Disorders Interview Schedule for DSM-IV: Child and parent versions. Journal of the American Academy of Child & Adolescent Psychiatry, 40, 937-944.

37.

Skirbekk

Hansen

B. H.

Oerbeck

Kristensen

(2011). The relationship between sluggish cognitive tempo, subtypes of attention-deficit/hyperactivity disorder, and anxiety disorders. Journal of Abnormal Child Psychology, 39, 513-525.

38.

Sorensen

Plessen

K. J.

Nicholas

Lundervold

A. J.

(2011). Is behavioral regulation in children with ADHD aggravated by comorbid anxiety disorders? Journal of Attention Disorders, 15, 56-66.

39.

Tannock

(2009). ADHD with anxiety disorders. In Brown

T. E.

(Ed.), ADHD comorbidities: Handbook for ADHD complications in children and adults (pp. 131-155). Washington, DC: American Psychiatric Publishing.

40.

Tannock

Ickowicz

Schachar

(1995). Differential effects of methylphenidate on working memory in ADHD children with and without comorbid anxiety. Journal of the American Academy of Child & Adolescent Psychiatry, 34, 886-896.

41.

Vloet

T. D.

Konrad

Herpetz-Dahlmann

Polier

G. G.

Gunther

(2010). Impact of anxiety disorders on attentional functions in children with ADHD. Journal of Affective Disorders, 124, 283-290.

42.

Wechsler

(1991). Manual for the Wechsler Intelligence Scale for Children–Third Edition, Revised. San Antonio, TX: The Psychological Corporation.

43.

Wechsler

(2003). Manual for the Wechsler Intelligence Scale for Children–Fourth Edition. San Antonio, TX: The Psychological Corporation.

44.

Wood

J. J.

Piacentini

J. C.

Southam-Gerow

Chu

B. C.

Sigman

(2006). Family cognitive behavioral therapy for child anxiety disorders. Journal of the American Academy of Child & Adolescent Psychiatry, 45, 314-321.