Your Instructor will assign a specific disorder for you to research for this Assignment. ? Use the Walden library to research evidence-based treatments for

Learning Disorder Confers Setting-Specific Treatment Resistance for Children with ADHD, Predominantly

Inattentive Presentation

Lauren M. Friedman, Keith McBurnett, and Melissa R. Dvorsky Department of Psychiatry, University of California, San Francisco

Stephen P. Hinshaw Department of Psychiatry, University of California, San Francisco and Department of Psychology,

University of California, Berkeley

Linda J. Pfiffner Department of Psychiatry, University of California, San Francisco

Attention deficit/hyperactivity disorder–predominantly inattentive presentation (ADHD-I) and specific learning disorder (SLD) are commonly co-occurring conditions. Despite the considerable diagnostic overlap, the effect of SLD comorbidity on outcomes of behavioral interventions for ADHD-I remains critically understudied. The current study examines the effect of reading or math SLD comorbidity in 35 children with comorbid ADHD-I+SLD and 39 children with ADHD-I only following a behavioral treatment integrated across home and school (Child Life and Attention Skills [CLAS]). Pre- and posttreatment outcome measures included teacher-rated inattention, organizational deficits, and study skills and parent-rated inattention, organizational deficits, and homework problems. A similar pattern emerged across all teacher-rated measures: Children with ADHD-I and comorbid ADHD-I+SLD did not differ significantly at baseline, but between-group differences were evident following the CLAS intervention. Specifically, children with ADHD-I and comorbid ADHD-I+SLD improved on teacher-rated measures following the CLAS intervention, but children with ADHD-I only experienced greater improvement relative to those with a comorbid SLD. No significant interactions were observed on parent-rated measures—all children improved following the CLAS intervention on parent-rated measures, regardless of SLD status. The current results reveal that children with ADHD-I+SLD comorbidity benefit significantly from multimodal behavioral interventions, although improvements in the school setting are attenuated significantly. A treatment-resistant fraction of inattention was identified only in the SLD group, implying that this fraction is related to SLD and becomes apparent only when behavioral intervention for ADHD is administered.

Attention deficit/hyperactivity disorder (ADHD) and speci- fic learning disorders (SLDs) are two of the most prevalent disorders in childhood, affecting approximately 7% and 9% of children worldwide, respectively (Altarac & Saroha,

2007; Thomas, Sanders, Doust, Beller, & Glasziou, 2015). ADHD and SLD are also commonly co-occurring—chil- dren with ADHD are almost 5 times more likely to be diagnosed with an SLD relative to their typically develop- ing peers (DuPaul, Gormley, & Laracy, 2013), and recent estimates suggest that approximately 45% of children with ADHD meet criteria for an SLD (DuPaul et al., 2013).

Comorbidity of any two disorders may be worse than the sum of its parts. For example, children with ADHD and

Correspondence should be addressed to Lauren M. Friedman, Linda J. Pfiffner, Department of Psychiatry, University of California, San Francisco 401 Parnassus Avenue, San Francisco, CA 94143. E-mail: [email protected]; [email protected]

Journal of Clinical Child & Adolescent Psychology, 49(6), 854–867, 2020 Copyright © Society of Clinical Child & Adolescent Psychology ISSN: 1537-4416 print/1537-4424 online DOI: https://doi.org/10.1080/15374416.2019.1644647

conduct disorder, compared to children with only one of these disorders, have been found to have an earlier age of symptom onset, greater persistence of problem behaviors, worse aca- demic problems, and increased severity of ADHD and con- duct symptoms (Loeber & Keenan, 1994). An additive effect may explain some findings, but simple addition cannot explain the synergistic effect that comorbid ADHD has on the severity of conduct disorder symptomatology, and vice versa. In a related vein, both inattention and learning difficul- ties are often more severe for children with ADHD and SLD than for children diagnosed with only one disorder (McNamara, Willoughby, & Chalmers, 2005; Purvis & Tannock, 2000; Wei, Yu, & Shaver, 2014). Comorbid ADHD/SLD is also associated with greater educational, neu- rocognitive, and social impairments relative to children with only ADHD, including more severe executive functioning deficits, higher rates of grade-retention, increased likelihood of placement in special education classes, greater use of in- school tutoring services, and poorer social skills (Bental & Tirosh, 2007; Seidman, Biederman, Monuteaux, Doyle, & Faraone, 2001; Wei et al., 2014; Willcutt et al., 2007, 2010; Willcutt, Pennington, Olson, Chhabildas, & Hulslander, 2005). The greater symptom load associated with comorbidity is difficult to explain solely on the basis of additive effects of ADHD and SLD.

The question thus arises: If having an accompanying condi- tion such as SLD confers more impairment than ADHD alone, will ADHD interventions prove less effective for children with ADHD/SLD comorbidity as a result of the inattentive sequela related to SLD? This question must be framed in the context of specific effects of treatment, because the best information will come from using a treatment that is known to preferentially reduce ADHD rather than SLD. If treatment targeted at one domain reduced impairments related to ADHD and SLD, we would not be able to distinguish the improvement of ADHD proper from the improvement in inattention that overflows from SLD. Recent evidence, however, suggests that treatments tar- geted toward one disorder do not substantially affect the other. Tamm et al. (2017) examined the effectiveness of intensive reading instruction, ADHD treatment (behavioral parent train- ing and medication management administered concomitantly), and combined treatment (reading instruction, parent training, andmedication) for childrenwith comorbidADHDand reading disorder. Children assigned to the ADHD and combined treat- ment conditions improved in parent- and teacher-reported ADHD symptoms, whereas those receiving reading instruction did not. In addition, children assigned to the reading instruction and combined conditions showed improvement on standardized reading measures, whereas children receiving Behavioral Parent Training (BPT)/medication therapy only did not show significant reading gains. Furthermore, there was no added benefit to combined versus mono-domain therapy. Thus, Tamm et al. demonstrated specific effects of treatments designed for each diagnosis.

One of the most difficult differential decisions in child psychopathology, for children with weaknesses in both atten- tion and learning, is ascertaining how symptoms and impair- ment might be attributable to each disorder. On the continuum of learning problems, even mild difficulty with reading or math may manifest as inattention, particularly when the child is engaged in academic endeavors and when the effort demanded requires additional attentional resources for those with already- reduced attention spans, sapping energy and motivation. Therefore, during academic tasks children with ADHD/SLD comorbidity may appear inattentive phenotypically partially because they lose focus, engage in off-task behaviors, and become frustrated because of the arduous nature of learning- related tasks (Pennington, Groisser, &Welsh, 1993). This frac- tion of the total inattention symptomatology (the part emanating from SLD) may be relatively intractable; that is, treatments that are effective for primary inattention may be considerably less effective for inattention that is secondary to learning difficulties, particularly in settings requiring increased learning demands (e.g., school, homework completion). Such an interpretation is consistent with evidence that childrenwithADHDand SLDare poorer responders to psychostimulant medications than those with ADHD alone (Grizenko, Bhat, Schwartz, Ter-Stepanian, & Joober, 2006).

Indeed, recent evidence suggests that deficits in learning adversely affect response to behavioral interventions. Breaux et al. (2019) examined predictors of treatment response among middle school adolescents with ADHD who received either a contingency-management or skill- based intervention for homework problems. Across a range of predictors examined, baseline math and reading achievement scores were the most consistent predictors of parent- and teacher-rated treatment response. Those with low to below-average academic achievement (i.e., reading or math achievement standard scores less than 95) were less likely to have reductions in homework problems and improved homework completion following treatment. However, findings from the multimodal treatment study for ADHD (MTA) did not support these results, as youth with a comorbid SLD did not differ on treatment-related improvement in homework problems (Langberg et al., 2010). It is important to note that whether comorbid SLD moderates or predicts treatment-related improvements in inattention and other related impairments (e.g., organiza- tional and study skills) has not been examined but warrants scrutiny given the potential synergistic effect of SLD comorbidity on ADHD-related sequelae.

No study to date has examined varying responses to behavioral intervention outcomes among children with ADHD–predominantly inattentive presentation (ADHD-I). Extrapolating conclusions regarding treatment response from children with clear hyperactivity and impulsivity to children with ADHD-I is questionable, given that ADHD-I is uniquely associated with different attention and

LEARNING DISORDER CONFERS SETTING-SPECIFIC TREATMENT RESISTANCE FOR CHILDREN 855

neurocognitive profiles, psychopathological correlates (e.g., less oppositionality, greater sluggish cognitive tempo and substance use), and social skills deficits than is the com- bined presentation (Bauermeister et al., 2005; Huang- Pollock, Mikami, Pfiffner, & McBurnett, 2007; McBurnett, Pfiffner, & Frick, 2001; Milich, Balentine, & Lynam, 2001; Sobanski et al., 2008). Furthermore, at least one longitudinal study indicates that academic impairments for youth with ADHD-I presentation are more profound and persistent than those found in other presentations of the disorder (Massetti et al., 2008). Given the unique impairments and academic difficulties faced by children with ADHD-I, it is especially important to examine the impact of SLD in this presentation of ADHD.

Most behavioral interventions for ADHD target proble- matic behaviors typically associated with ADHD–com- bined presentation. That is, most behavioral interventions emphasize reducing hyperactivity, impulsivity, and defiance that are either absent in or less relevant to children with ADHD-I. To our knowledge, only one validated behavioral treatment exists currently for children with ADHD-I: The Child Life and Attention Skills program (CLAS; Pfiffner et al., 2014). CLAS is a multicomponent intervention that combines behavioral parent training, child skills training, and classroom consultation strategies tailored to address the cross-setting challenges specific to children with ADHD-I. In a randomized, controlled trial, our team (Pfiffner et al., 2014) found that CLAS was associated with significant improvements in teacher-rated attention, social skills, orga- nization, and global functioning, as well as parent-rated organizational skills, relative to parent training alone and to treatment as usual. CLAS also demonstrated superior results relative to treatment as usual on parent-rated atten- tion, social skills, and global functioning. Whether SLD comorbidity affects response to CLAS among children with ADHD-I, however, remains unknown.

In sum, no study to date has examined whether the presence of SLD predicts differential response to beha- vioral intervention for treatments designed specifically for ADHD-I. Herein, the effect of SLD comorbidity was assessed across several outcome domains (e.g., ADHD symptoms, organizational deficits, study skills, and home- work problems) using both parent and teacher informants. We hypothesized a significant interaction between treat- ment and comorbid SLD status, such that children with ADHD-I (without SLD) would exhibit greater treatment- related improvements on multiple domains, including inat- tention severity, relative to those with ADHD-I/SLD. The hypothesized interaction is based on the greater symptom severity, educational impairments, and cognitive challenges among children with comorbid ADHD/SLD, compared to those with only ADHD (whose inattention is less likely to be secondary to learning-related difficulties; Bental & Tirosh, 2007; Seidman et al., 2001; Willcutt et al., 2010,

2005). This fraction of the symptom profile emanating from learning difficulties is hypothesized to be less responsive when treated with interventions targeting ADHD singly, such as CLAS. It is also based on contemporary etiological models of ADHD/SLD comorbidity suggesting that chil- dren with comorbid ADHD/SLD evince more severe and/or numerous neurocognitive (DuPaul et al., 2013; Purvis & Tannock, 2000; Willcutt et al., 2005, 2007) and neural morphology (Hynd, Semrud-Clikeman, Lorys, Novey, & Eliopulos, 1990; Jagger-Rickels, Kibby, & Constance, 2018; Kibby, Kroese, Krebbs, Hill, & Hynd, 2009) deficits than those with an ADHD monodiagnosis, features that are not directly addressed through the CLAS (ADHD-focused) intervention.

METHOD

Participants

The current study comprises a secondary analysis of a larger, randomized, controlled clinical trial (Pfiffner et al., 2014). Briefly, participants ages 7 to 11 with a diagnosis of ADHD-I were randomly assigned to one of three treatment conditions: CLAS program, behavioral par- ent training only, and treatment as usual. We examine the CLAS group (n = 74; age M = 9.21, SD = 1.10) exclusively herein. First, CLAS demonstrated superior results relative to parent training alone and treatment as usual in previous studies (Pfiffner et al., 2014). Second, it was the only intervention associated with improvements across all of the outcome domains assessed (e.g., inattention, organiza- tional skills, social skills, and overall functioning)—and it is unlikely to find moderation effects in the absence of treatment effects barring any suppression effects (Hayes, 2017). Third, it was the only intervention that improved performance in the school setting, which is particularly relevant for children with learning disabilities.

Participants were recruited at two treatment sites: University of California, San Francisco and University of California, Berkeley. Children were recruited or referred from school personnel including principals, school mental health professionals, and learning specialists; pediatricians; and child psychiatrists and psychologists. In addition, recruitment flyers were posted in online parent networks and professional organizations. Across 4 years (2009– 2012), six cohorts of children participated, with a mean number of 12 children in each cohort (range = 10–15).

To be considered for inclusion, children met the follow- ing criteria: (a) primary Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV; American Psychiatric Association, 1994) diagnosis of ADHD-I, as confirmed by the Kiddie Schedule for Affective Disorders and Schizophrenia (KSADS-PL) clinical interview (see next); (b) ages 7–11 (Grades 2–5); (c) attending school

856 FRIEDMAN ET AL.

full time in a regular classroom; (d) Full Scale IQ greater than 80, as confirmed on the Wechsler Intelligence Scale for Children, Fourth Edition (Wechsler, 2003); (e) living with at least one parent for 1 year prior to study recruit- ment; (f) family schedule that permitted participation in CLAS groups; and (g) school proximity within 45 min of either treatment site to allow study personnel to conduct teacher consultation meetings. Children were excluded if they were planning to initiate or change medication (stimu- lant or otherwise) in the near-term. Children taking non- stimulant psychoactive medications were also excluded because of the difficulties of withholding medication to confirm ADHD-I symptoms among raters potentially unfa- miliar with children’s behavior while not taking medica- tions (i.e., classroom teachers), as required to confirm cross-setting impairment required for diagnosis. Children with pervasive developmental disorders or other neurologi- cal illnesses were also excluded.

Demographic data for the participants in this study (i.e., children receiving CLAS, n = 74) are as follows: Mean child age was 9.21 years (range 7–11) with 18% in the second grade, 21% in third grade, 21% in fourth grade, and 14% in fifth grade. Boys comprised 51.4% of the sample. 55.4% were Caucasian, 12.2% were Latinx, 9.5% were Asian American, 5.4% were African American, and 17.6% identified as mixed-race. Total household income was below $50,000 for 12.2% of families, $50,000-$100,000 for 31.1%; $100,000-$150,000 for 24.3%, and more than $150,000 for 27.0% of families. Income data was missing from 5.4% of families. 84.9% of parents reported graduating from college and 9.5% of chil- dren were living in single-parent homes. Note that only 6.8% of children were taking medication for ADHD.

Procedure

A detailed description of participant screening, flow, attri- tion, diagnostic procedures, treatment fidelity, and therapist qualifications are provided elsewhere (Pfiffner et al., 2014). In short, participant screening was conducted using a successive, three-wave approach. First, telephone screen- ing calls were conducted with parents and teachers to assess initial eligibility regarding demographics and medication status. Next, those meeting initial screening criteria were invited to complete rating scale packets containing the par- ent- and teacher- versions of the Child Symptom Inventory (CSI-IV, Gadow & Sprafkin, 2002) and the Impairment Rating Scale (IRS, Fabiano et al., 2006). Third, children who met the following criteria were invited for a full diag- nostic assessment: (a) at least five symptoms rated as occur- ring “often” or “very often” by parents or teachers on the CSI, with each informant endorsing at least two symptoms; (b) five or fewer hyperactive/impulsive symptoms endorsed as occurring “often” or “very often” by parents and teachers

on the CSI; and (c), at least one area of functioning rated as � 3 on the IRS by both parent and teacher, thereby indicat- ing evidence of impairment across settings. Diagnostic status was ascertained using clinical interviews that consisted of detailed questions regarding children’s developmental, med- ical, clinical, and school history, as well as the Kiddie Schedule for Affective Disorders and Schizophrenia (K-SADS-PL; Kaufman, Birmaher, Brent, Rao, & Ryan, 1997). The K-SADS is a semi-structured interview that assesses the presence and impairment of psychopathology including ADHD, oppositional defiant disorder, conduct dis- order, anxiety disorders, mood disorders, and psychosis based on DSM-IV criteria. Its psychometric properties are well-established (cf., Kaufman et al., 1997).

To be considered for study entry, children were required to meet full DSM-IV criteria for ADHD-I based on K-SADS interview—viz., six or more inattention symp- toms and fewer than six hyperactive/impulsive symptoms. Parents also completed a battery of questionnaires, and children were administered the Wechsler Intelligence Scale for Children, Fourth Edition (Wechsler, 2003), select subtests from the Woodcock–Johnson Test of Achievement, Third Edition (Woodcock, Mather, McGrew, & Schrank, 2001), and a questionnaire battery.

Study procedures were approved by the Committee on Human Research at University of California, San Francisco and University of California, Berkeley. All participating parents and children provided their informed written con- sent and assent, respectively. Families were compensated for measure completion at posttreatment ($50). Teachers were also compensated for competing measures at baseline ($50) and posttreatment ($75) and provided a total of $100 for their participation in teacher consultation meetings. Treatment was provided to participants at no cost. Immediately following treatment, laboratory visits were scheduled with families and rating scales were sent to teachers to collect posttreatment ratings.

Intervention

CLAS consists of three empirically supported behavioral interventions adapted for children with ADHD-I: beha- vioral parent training, child skills training, and daily report card with teacher consultation. For a detailed description of CLAS intervention skills and modules, see Pfiffner et al. (2014). The size of each CLAS group ranged between six and eight families.

Parent component

The parent training consisted of ten 90-min weekday groups, along with up to six 30-min individual family meetings (parent, child, and therapist). The curriculum was adapted from extant parent training programs (Barkley, 1997b; Forehand & McMahon, 1981) and

LEARNING DISORDER CONFERS SETTING-SPECIFIC TREATMENT RESISTANCE FOR CHILDREN 857

modified to include modules targeting challenges specific to ADHD-I. Parent stress management skills were also included.

Child component

The child skill component consisted of ten 90-min weekday groups that ran concurrently with the parent group sessions. Modules were adapted from a social skills program for children with ADHD (Pfiffner & McBurnett, 1997) and focused on building independence, organization, emotion regulation, assertiveness, and social skills. Parents reinforced skills using a token economy outside of the child group to encourage generalization of the skills across contexts.

Teacher component

Teachers were taught evidence-based classroom man- agement strategies to scaffold and support attention and use of the child skills in the classroom (DuPaul, Weyandt, & Janusis, 2011; Fabiano et al., 2010; Pfiffner et al., 2011). Teachers also implemented a customized school–home daily report card whereby teachers rated students three times daily on up to four personalized treatment goals. Up to five meetings were conducted with teachers, parents, children, and study personnel to discuss daily report card goals, classroom accommodations, and the skills taught within the child component to encourage generalization of group skills across contexts.

Measures

Specific learning disorder

SLD Status was assessed a posteriori and did not affect participant inclusion or exclusion. Children were consid- ered to have a suspected SLD if they received a standard score of 85 or lower (i.e., 16th percentile) on any of the following subtests of the Woodcock–Johnson Test of Educational Achievement–III (Woodcock et al., 2001): Passage Comprehension, Reading Fluency, Calculation, or Math Fluency. The psychometric properties of this test are well-established, including concurrent validity with other measures of academic achievement (Woodcock et al., 2001).1

Although SLD definitions vary widely in the literature wherein delineation scores range from 80 to 90 (cf. Brueggemann, Kamphaus, & Dombrowski, 2008, for a review), a cutoff score of 85 was chosen, as it indicates the presence of a basic skill deficit that may require

intervention, reliably identifies children with poor school performance and functional impairments (Brueggemann et al., 2008), and is associated with the lowest rates of reading growth following intervention (Vellutino, Scanlon, & Reid Lyon, 2000). In addition, the “low achievement model” (i.e., below-average academic achievement) was chosen over alternative models of SLD definition, such as the “IQ-achievement discrepancy model,” as the latter is associated with limited reliability, questionable validity, poor sensitivity and positive predictive power, and limited incremental validity over the low-achievement definition (Brueggemann et al., 2008; Dombrowski, Kamphaus, & Reynolds, 2004; DuPaul et al., 2013). Both fluency and ability subtests were considered, consistent with the current conceptualization of SLD within the DSM-5 (American Psychiatric Association, 2013), which recognizes an uneven profile of abilities wherein deficits can be observed in accurate and fluent calculation/reading, either indepen- dently or concomitantly. Based on this definition, 47.3% (n = 35) met criteria for an SLD. Specifically, 41.9% (n = 31) met criteria for a disability in math, 13.5% (n = 10) met criteria for a disability in reading, and 8.1% (n = 6) met criteria for a learning disability in both reading and math.

Outcome Measures

Inattention

Parent- and teacher-rated symptom count2 from the Inattention scale of Child Symptom Inventory (CSI; Gadow & Sprafkin, 2002) was used to assess ADHD- related inattention symptomatology and had good internal consistency in the present sample (αs = .77–.82). The CSI measures inattention consistent with ADHD DSM-IV cri- teria on a 4-point scale from 0 (never) to 3 (very often). Inattention symptoms were considered present if they were rated as occurring often or very often. The Inattention scale of the CSI has normative data, acceptable test–retest relia- bility, and predictive validity for a categorical diagnosis of ADHD (Gadow & Sprafkin, 2002).

Organizational deficits

Parents and teachers completed respective versions of the Children’s Organizational Skills Scale (COSS; Abikoff & Gallagher, 2003). Age-corrected T-scores of the COSS Total composite score served as the dependent variable to assess children’s deficits in organization, planning, and time management skills and had good internal consistency in the present sample (α = .91–.97). The parent and teacher ver- sions have adequate psychometric properties including high

1 It is important to recognize that SLD diagnosis is usually conferred following full psychoeducational or neuropsychological evaluation and that the presence of significant academic achievement deficits indicates a suspected but not confirmed SLD diagnosis.

2 Alternative summary scores, such as symptom severity scores, were also analyzed but did not change the pattern or interpretation of results.

858 FRIEDMAN ET AL.

test–retest reliability (rs = .94–.99 and .88–.93, respec- tively), and evidence of structural, convergent, and discri- minant validity (Abikoff & Gallagher, 2003). Items are rated on a 4-point scale from 1 (hardly ever/never) to 4 (just about all the time) and assess the extent to which children have difficulties with planning tasks effectively; engaging in organizational behaviors such as list creation, routines, and reminders; and managing materials and sup- plies necessary for task completion.

Study skills

Teacher-rated age-corrected decile scores on the Study Skills subscale of the Academic Competence Evaluation Scale (DiPerna & Elliott, 2001) served as the dependent variable to measure children’s study skills and had adequate internal consistency in the present sample (α = .88–.90). The Academic Competence Evaluation Scale has excellent psychometric properties including test–retest reliability (r = .96) and evidence of predictive and concurrent validity (DiPerna & Elliott, 2001). Items are rated on a 5-point scale ranging from 1 (never) to 5 (almost always); they assess the extent to which children are able to prepare for and manage tests and class assignments, with higher scores indicating greater functioning in study skills.

Homework problems

Average parent-rated scores on the Homework Problems Checklist (Anesko, Schoiock, Ramirez, & Levine, 1987) served as the dependent variable to measure children’s challenges with managing and completing homework and showed high internal consistency in the present sample (α = .89–.91). The Homework Problems Checklist has adequate psychometric properties, including test–retest reliability and predictive validity with children’s academic perfor- mance (Anesko et al., 1987). Items are rated on a 4-point scale ranging from 1 (never) to 4 (very often) and assess difficulties with the management of homework materials, knowledge and organization of homework tasks, homework completion, and homework independence.

Data Analytic Plan

All statistical analyses were performed using SPSS (Version 25; IBM Corp, 2017). Preliminary analyses involved inves- tigation of missing data and assessment of baseline charac- teristics by SLD status (see Table 1). We analyzed outcomes in the four domains that were the primary focus of our investigation: inattention, organizational deficits, study skills, and homework problems. For measures that included both parent and teacher ratings (i.e., inattention and organi- zation deficits), separate analyses were performed for each rater. Primary analyses involved mixed model analyses of variance (ANOVAs) examining within (pretreatment,

posttreatment) and between (ADHD-I, ADHD-I+SLD) group comparisons. Analyses were initially completed with- out covariates. We then performed follow-up ANCOVAs adjusting for the following pretreatment variables: child’s age, gender, race, medication status, and oppositional defi- ant disorder symptoms, as well as education level of the primary parent. However, each of these covariates were either nonsignificant or did not change the pattern of inter- pretation of results when included within the analyses. Simple mixed model ANOVAs without covariates are there- fore presented. Consistent with recommendations (Dennis et al., 2009; Miller & Chapman, 2001), participant’s Full Scale IQ score was not examined as a covariate. That is, current etiological models of ADHD (Barkley, 1997a; Castellan