Evaluation of the Duration of Action and Comparative Effectiveness of Lisdexamfetamine Dimesylate and Behavioral Treatment in Youth With ADHD in a Quasi-Naturalistic Setting

Abstract

Objective: This study compared the relative effects of three treatment conditions: long-acting stimulant medication (MED), behavior modification, and medication/behavioral treatments combined (COM) in children with ADHD. Method: A total of 25 children, aged 6 to 12 years, received the three treatment conditions during a 7-week Summer Treatment Program in an alternating treatments design. Counselors completed behavioral ratings from 0.5 to 12.5 hr post dose, and parents completed nighttime ratings. Results: Ratings for SKAMP (Swanson, Kotkin, Agler, M-Flynn, and Pelham) and for following instructions indicated COM and MED improved symptoms over BEH treatment beginning 3 hr post dose (p = .008), with ratings maintained 12.5 hr post dose (p = .001 and .006). Results for frustration tolerance indicated significant improvement in all three conditions until 9 hr post dose. Conclusion: MED and COM separated from BEH at 3 hr post dose, and sustained benefit was observed across the day for two of three measures. BEH appears to have an additive effect, extending the duration of frustration tolerance.

Keywords

ADHD psychosocial treatment multimodal treatment extended-release stimulant medication

ADHD is the most prevalent psychiatric disorder of childhood, affecting an estimated 9.5% of school-aged children (Centers for Disease Control and Prevention [CDC], 2010). ADHD is typically chronic, with adverse effects extending into adolescence and adulthood in many patients (Biederman, 2005). Children with ADHD struggle with impairments in numerous domains of functioning, including academic performance, self-efficacy, and interpersonal relationships (Adesman, 2001). Evidence indicates that patients diagnosed with ADHD are at higher risk for mental health problems, such as mood disorders, oppositionality, and substance use (Biederman, 1998; Biederman et al., 1996; Biederman, Newcorn, & Sprich, 1991; Spencer, Biederman, & Mick, 2007). Families of children with ADHD experience considerable emotional and financial stressors as well (Johnston & Mash, 2001; Swensen et al., 2003). The adverse outcomes associated with ADHD make a strong case that it is a public health concern requiring appropriate and effective treatment approaches. Furthermore, the pervasiveness of dysfunction and the effects on family quality of life (for reviews, see Kaplan & Newcorn, 2011; Limbers, Ripperger-Suhler, Boutton, Ransom, & Varni, 2010; Vaughan, March, & Kratochvil, 2011) support the need for treatments that improve symptoms and behavior across settings and throughout the course of the day. This latter area—investigation of effects of different treatment strategies across the course of a day—is lacking in the study of ADHD as it is typically not a focus of treatment in randomized clinical trials (RCTs).

To date, treatment of ADHD has focused on reducing ADHD symptoms through pharmacological and behavioral treatment programs applied independently and in combination (Multimodal Treatment Study of Children With ADHD [MTA]; MTA Cooperative Group, 1999) almost exclusively RCTs. Medication is generally considered the treatment of choice for many children with ADHD and stimulants are the gold standard for pharmacotherapy (American Academy of Pediatrics, 2001; S. Pliszka & AACAP Work Group on Quality Issues, 2007; S. R. Pliszka et al., 2006). Nonetheless, recent 8-year follow-up data indicate that the long-term benefit of stimulant medication may taper over time (Molina et al., 2009), possibly due to treatment nonadherence. Although these findings and practice parameters are instructive, they are based on results derived from studies using older, shorter acting stimulant delivery systems of methylphenidate and amphetamine and may not be applicable to newer, longer acting delivery systems that may have notable impact across the course of a day. Lisdexamfetamine dimesylate (LDX) is a long-acting stimulant prodrug that has been shown in initial clinical studies to have good safety and tolerability, and long duration of action (Biederman, Boellner, et al., 2007; Biederman, Krishnan, Zhang, McGough, & Findling, 2007; Findling, Childress, Krishnan, & McGough, 2008; Wigal, Kollins, Childress, Squires, & 311 Study Group, 2009). Despite favorable safety and tolerability data, empirical evidence regarding the duration of LDX in naturalistic type settings and across the course of a day compared with controlled clinical trials remains unclear.

Combining short-acting stimulants with behavioral intervention has been consistently demonstrated to reduce symptoms of ADHD (Pelham et al., 2005) and appears to add benefit to some children beyond medication alone (March et al., 2000). Although data support the primacy of stimulant treatment for decreasing ADHD symptoms (MTA Cooperative Group, 1999; Van der Oord, Prins, Oosterlaan, & Emmelkamp, 2008), to our knowledge, no published studies have examined the effects of long-acting stimulant medication behavioral intervention, and their combination; thus, the clinical relationship of combined long-acting stimulant treatment and behavioral intervention remains undocumented.

The present study addresses an important gap in the literature by examining the duration and contribution of effect of long-acting medication alone relative to intensive behavior therapy alone relative to the combined effect of behavior therapy and long-acting medication. In contrast to RCTs, we used a small-N alternating-treatments design (ATD; Hersen & Barlow, 1976; Kazdin, 1982) in a quasi-naturalistic setting. We examined the relative effectiveness of a long-acting ADHD stimulant treatment, intensive behavioral intervention, and combined treatment (long-acting medication and behavioral intervention) in an effort to compare active treatments rather than treatment versus placebo alone. We were interested in when the effect of behavioral intervention ends and what if any augmentative effects emerge when behavioral intervention is combined with long-acting medication. A small-N design allows us to determine the separation and duration effects across short-time intervals and permits collection of symptom ratings and behavioral observation data. The ATD rapidly alternates treatments across days within small groups of participants to provide sensitive examination of treatment effects. The design removes variability through improved experimental control of treatment conditions. Most previous efficacy studies focus exclusively on global improvement of symptoms as measured by clinician- and parent-rated scales that do not measure the impact of long-acting stimulants on real-world behaviors that affect home functioning. Thus, the present study adds to previous pharmacologic and BEH studies of ADHD in that it compares the relative effectiveness of a long-acting stimulant (MED), intensive behavioral intervention (BEH), and combined medication and behavioral intervention (COM). Treatments are implemented in a camp environment that approximates daily settings children experience in play and classrooms. Data include clinician-rated global symptom measures and parent-rated measures for following instructions and tolerating frustration (i.e., the ability to calm oneself). These latter two behaviors are important aspects of ADHD that are not explicitly accounted for in symptom ratings but are highly relevant to everyday functioning. As such, the present study represents an effectiveness extension to previously conducted efficacy trials (Frazier et al., 2010; Gartlehner, Hansen, Nissman, Lohr, & Carey, 2006). In this study, effectiveness measures may generalize to stimulant naïve children and to previous stimulant responders but not necessarily to poor responders who were excluded from the study.

This study seeks to answer two questions:

What is the relative effectiveness of intensive behavioral and long-acting medication to the treatment of ADHD?

What is the duration of effect of extended-release medication and behavior intervention across the day in a quasi-naturalistic setting?

We hypothesized that postdose MED and COM treatments would result in improvement in ADHD symptoms and that these improvements would persist throughout the day (>9 hr post dose) as compared with BEH. Furthermore, we anticipated that adjunctive behavioral intervention would enhance the effectiveness of medication by suppressing the expression of global symptoms of ADHD as measured by the Swanson, Kotkin, Agler, M-Flynn, and Pelham (SKAMP) rating scale, increasing compliance with instructions, and enhancing emotional control as indicated by reduced expression of frustration on the Behavior Monitoring Scale (BMS; Manos, Short, & Findling, 1999).

Method

Participants

A total of 34 children (26 male, 8 female) were screened for participation, and 30 were enrolled in the study (Figure 1). Study procedures were implemented after parents signed consent and children provided assent. A member of the research team explained all aspects of the study. The hospital Institutional Review Board (IRB) approved the study protocol, informed consent procedures, and assent procedures.

Figure 1.

Participant accountability.

Participants were recruited from clinics in a hospital-based pediatric medical and psychological practice and through advertisements placed in local newspapers to participate in a 7-week Summer Treatment Program (STP; Pelham & Hoza, 1996). Study staff reviewed medical records and confirmed the presence of symptom criteria for ADHD, confirmed the absence of exclusion criteria in each participant, and determined eligibility for the study. In the initial screening visit, parents and children were consented/assented and each participant proceeded into a dose-optimization phase starting with a baseline visit. For those patients who were undergoing pharmacotherapy currently, a 1-week washout from ADHD stimulant treatment and a 2-week washout from nonstimulant treatment (i.e., atomoxetine) were necessary prior to the baseline visit. A total of 19 participants were previously treated with psychostimulants, and 1 was treated with a nonstimulant medicine. These 19 participants entered the washout phase; 5 participants had not been previously treated and immediately entered titration (Figure 1). In all, 18 participants were naïve to LDX. Of the 3 participants who entered titration and withdrew due to side effects, sleep delay was the primary side effect reported. No severe adverse events were reported.

Titration visits were conducted for 5 weeks: Initial dose administered began at LDX 30 mg and was increased to 50 mg or 70 mg until optimization was achieved. Optimization was determined as a Clinical Global Impressions–Improvement (CGI-I) scale of 1 or 2 and a T-score of 60 or less on the ADHD Rating Scale–IV (ARS-IV; DuPaul, Power, McGoey, Ikeda, & Anastopoulos, 1998). Although the protocol allowed for one downward titration of dose if a child experienced significant side effects, none of the children required the downward dosing. Following optimization, the optimized dose was maintained until the child started the 7-week STP and throughout the program.

Inclusion Criteria

To be eligible for inclusion, participants met criteria at initial screening and baseline in that (a) written consent was signed by the participant’s parent or legal guardian, with assent given by the participant; (b) the child was aged 6 to 12 years; (c) females of childbearing age had negative pregnancy test at screening and baseline, and were not at risk for becoming pregnant; (d) participants met Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.; DSM-IV-TR; American Psychiatric Association, 2000) criteria for a primary diagnosis of ADHD based on a detailed psychiatric evaluation that included ARS-IV parent report; (e) each participant had IQ scores in at least the average range (Kaufman Brief Intelligence Test [K-BIT] range = 85-115); (f) symptom criteria for a comorbid mental health condition that could affect safety or tolerability of medication, or interfere with the participant’s participation in the study were not in evidence; (g) blood pressure measurements were within the 95th percentile for age, gender, and height at screening and baseline; (h) electrocardiogram (ECG) results were within the normal range and not clinically elevated at screening and baseline as determined by the study physician; and (i) caregivers were able to comply with the requirements of the study protocol.

Exclusion Criteria

At screening or baseline, eligibility was declined if the participant (a) had a current, controlled, or uncontrolled comorbid psychiatric diagnosis with significant symptoms, including posttraumatic stress disorder (PTSD), psychosis, bipolar illness, pervasive developmental disorder, severe obsessive compulsive disorder, severe depressive or severe anxiety disorder that, in the opinion of the examining physician, contraindicated treatment or assessment; (b) was suspected of substance abuse or dependence disorder within the past 12 months in accordance with DSM-IV-TR criteria; (c) had a positive urine drug result of a substance other than the currently prescribed stimulant; (d) had a history of seizures during the last 2 years, a severe tic disorder, and a current diagnosis or family history of Tourette’s syndrome; (e) had conduct disorder; (f) participated in any other research study 30 days prior to onset of this study or any time during this trial; (g) had clinically significant laboratory abnormalities at screening or baseline in the opinion of the study physician; (h) had a known history of structural cardiac abnormality; (i) had a concurrent chronic or acute medical illness that would prohibit the participant from completing the study or would not be in the best interest of the participant; and (j) had a previous history of failure to respond to stimulant medication treatment or stimulant in which side effects outweighed benefit.

Study Design

An ATD was used to examine the effects of intensive BEH, MED, and COM treatment in children as a function of age (6- to 9-year-olds and 9- to 12-year-olds). The logic of the ATD is that it allows the rapid and randomly counterbalanced alternation of conditions across days within the week. If there is a clear and constant separation between measures at multiple time points that is consistently associated with the three different treatment conditions, then we will be positioned to make stronger statements about how treatments affect performance. The ATD also allows many observations of a small number of participants repeatedly across days of the week. This strategy facilitates more and varied data collection than typically occurs in RCT designs as they usually allow for few observations of many participants. The ATD alternates treatment conditions across days within a single group of participants and provides a sensitive examination of the effects of different interventions. The design removes variability through improved experimental control of treatment conditions (Hersen & Barlow, 1976), with participants acting as their own controls. The three treatment conditions—BEH, MED, and COM—were randomized and counterbalanced (to control for order effects) across days (Monday through Thursday) during Weeks 4 through 6 of the STP. Weeks 1 through 3 and Week 7 involved open-label treatment with COM. In Weeks 4 through 6, the three treatment conditions were randomly assigned without replacement to the first 3 days of the week starting with Week 4. Next, a single treatment condition was randomly chosen for the 4th day of that week. After treatment conditions had been randomly assigned for Week 4, treatment conditions were assigned for Weeks 5 and 6 by incomplete counterbalancing, similar to a Latin square design, to ensure that no treatment condition occurred on the same day as the previous week and no two treatments occurred consecutively. The orders for each week are outlined in Table 1. Day 5 (Friday) was not included as part of the design because the STP employed a modified schedule atypical of the Monday through Thursday schedule (i.e., on Friday, children went on earned field trips). Time-series data were collected to assess changes in behavior across days of alternating treatments. Collection of time-series data permitted the assessment of ongoing treatment-related changes across each day and week, including duration of action and side-effect profiles.

Table 1.

Randomized, Counterbalanced Schedule of Alternating Treatments for Weeks 4 to 6.

	Monday	Tuesday	Wednesday	Thursday
Week 4	COM	MED	BEH	COM
Week 5	MED	BEH	COM	MED
Week 6	BEH	COM	MED	BEH

Note: COM = combined treatment (MED + BEH); MED = long-acting stimulant medication (lisdexamfetamine treatment only); BEH = behavior modification only.

Figure 2 presents the study design. Following baseline, each patient attended weekly titration visits for up to 5 weeks allowing for one downward titration of dose, although, in this study, none were necessary. When patients achieved dose optimization, they were enrolled in the STP (Pelham, Greiner, & Gnagy, 1998).

Figure 2.

Study design.

In Weeks 1 to 3, the optimized MED dose was maintained and children participated in the behavioral contingencies of the STP. Weeks 1 to 3 served to orient children to the contingencies and procedures of the behavioral program so that the novelty of program participation in the STP did not influence results. In Weeks 1 to 3 and 7, the optimal dose was administered between 8:00 a.m. and 8:30 a.m. on-site by the study coordinator in the presence of the patient’s parents prior to the start of the morning’s activities. During Weeks 4 to 6, the procedures used in Weeks 1 to 3 for administering medicine were followed, that is, medicine dose (COM and MED) or placebo capsule (BEH) was administered between 8:00 a.m. and 8:30 a.m., and data were collected as described. On Fridays and during the 7th week of the STP, participants received the combined treatment approach in that the standard behavioral contingencies of the STP were in effect and children’s optimized MED dose was administered in an unblinded manner; data obtained on Fridays were not used in the analyses. The present article examines data from Days 1 to 4 (i.e., Monday through Thursday) of Weeks 4 to 6, as this represents the repeated-measures, alternating treatments data-collection phase.

In this study, a full placebo condition was not included in the alternating treatments as the aim was to compare active treatment conditions with already established efficacy. This decision was based partly on ethical considerations: When all treatments are withheld during STP activities, children’s behavior deteriorates substantially, so much so that safety may be compromised. In addition, the use of active treatment provides a powerful examination of the relative effectiveness of MED, BEH, and COM treatment conditions in managing behavior in a quasi-naturalistic setting.

Procedures

Treatment setting

The STP is a 7-week, intensive summer academic-sports camp designed for children with ADHD and their families. The quasi-naturalistic, camplike setting differed from typical academic-sports camps in that it provided intensive behavior modification as part of daily activities. It approximated natural settings in that children play baseball, complete schoolwork at desks with other students, eat lunch, go on field trips, have recess, and the like. It was held on the campus of a public elementary school in Northeast Ohio. Children attended from 8:30 a.m. to 5:30 p.m. weekdays, and parents and children also participated from 7:00 p.m. to 9:00 p.m. on Wednesdays. During the evening session, parents attended a parent training program and children attended child care where behavioral contingencies were not in effect. The STP is administered in the context of a day treatment program with a broad treatment focus—each child had a clinical treatment plan with specific, individualized goals. Social skills training, relationship building, and the development of self-regulatory skills were emphasized in recreational activities. Children also worked in the academic learning centers to develop their ability to follow through with instructions, complete tasks, and comply with adults’ requests in the classroom.

Children were assigned to one of two age groups based on birth date: young (6-9 years, n = 12) and old (9.1-12 years, n = 13). A recently turned 9-year-old was assigned to the younger age group, and all other children 9 years and older were assigned to the older age group. Procedures were conducted in accordance with the Children’s Summer Treatment Program manual (Pelham et al., 1998) and as implemented in the MTA (MTA Cooperative Group, 1999). Five clinical staff members (i.e., one lead counselor, typically a graduate student, and four group counselors, usually undergraduate students) supervised each group of children during daily activities. The groups of children and counselors stayed together throughout the summer, so that children received intensive experience in functioning as a group, making friends, and interacting appropriately with adults. Recreational activities included age-appropriate games (e.g., soccer, basketball, and softball) and activities (e.g., swimming), group problem-solving discussions, social skills training, anger management training, dyadic friendship training, group cooperative tasks, and other individualized programs. The program is described in detail by Pelham and Hoza (1996).

In addition to the children’s involvement in the day treatment program, parents participated in a once weekly STP parent training session on how to implement behavior modification programs at home. The social learning approach to behavior management in the home complemented the behavior management program used in the STP. Study children remained on-site supervised by STP staff, whereas their parents attended the once weekly sessions.

Integrity and fidelity of treatment

To confirm treatment fidelity in the three treatment conditions (BEH, MED, and COM), all staff participated in 2 months of self-study and a 2-week intensive training program that included lectures, opportunities for role-play, and small-group coaching throughout each day. Clinical supervisors used fidelity checks to maximize consistency of behavioral observations across staff in all three treatment conditions; Pearson correlations were above .8 (Chronis et al., 2004). After fidelity checks, supervising staff noted discrepancies in treatment presentation and reviewed these discrepancies with staff in team meetings, with the team correcting these discrepancies after the meeting. Treatment fidelity was high (90% consistent with behavioral protocol) by the end of Week 1. After an overview of the study design and methods, staff members were instructed on the specifics of the protocols that were to be followed for the manipulation of treatment conditions during Weeks 4 through 6. Counselors and teachers followed the standardized protocol according to the STP manual (Pelham et al., 1998) on days that were randomized to the COM and BEH treatment conditions. For days that were randomized to the MED treatment condition, staff members suspended the behavioral contingencies of the STP following a revised protocol designed to mimic a typical summer camp environment described elsewhere (Chronis et al., 2004). In the MED condition, the behavior modification point contingencies were suspended and counselors were unblinded to treatment. Protocol fidelity was assessed by confirming that counselors called behaviors but did not assign points. That is, they did not impose contingent response cost or positive reinforcement (i.e., removing or awarding points) to called behaviors.

Investigators and camp counselors were blinded to medication treatment in the BEH and COM conditions during alternating treatments, but not during the MED-only condition. In this condition, children were told that the behavior management program was suspended, and counselors did not award points although they still called behaviors as they occurred. Children were aware of the suspension of reinforcement, replicating previous studies (Chronis et al., 2004). Thus, the MED condition was similar to that which occurs in everyday life where adults give instructions and children are expected to follow these instructions without overt contingencies in place.

Measures

Two rating scales captured evidence of behavior change in this study. The Swanson, Kotkin, Agler, M-Flynn, and Pelham (SKAMP) rating scale was the primary outcome measure (Wigal, Gupta, Guinta, & Swanson, 1998). It consists of 13 items with each item scored using a 0 to 6 “degree-of-impairment” anchored rating scale (i.e., 0 = normal, 1 = slight, 2 = mild, 3 = moderate, 4 = severe, 5 = very severe, and 6 = maximal). SKAMP items are not conducive to monitoring home behaviors, however, and do not assess everyday behaviors directly relevant to home and school functioning. For these reasons, we also used the BMS (Manos et al., 1999) as a secondary outcome measure to assess nine everyday behaviors. The BMS uses the Clinical Global Impressions–Severity (CGI-S) scale (1 = normal, 7 = most extreme). Two BMS ratings, “following instructions” (Item 5) and “frustration tolerance” (Item 6), were chosen a priori to evaluate the possible effects of LDX and behavioral intervention on attention-related compliance (following instructions) and emotion regulation (frustration tolerance).

Counselors completed the SKAMP and BMS at 9:00 a.m., 9:30 a.m., 11:30 a.m., 1:30 p.m., 3:30 p.m., and 5:30 p.m. Mondays through Thursdays (Days 1-4) to track behavior during the day and at 7:00 p.m. and 9:00 p.m. on Wednesdays (Day 3) to track behavior during the evening. (Counselors rated child behavior during child care, which was provided for parents who attended Wednesday parent trainings. A total of 17 children regularly attended child care as some parents did not bring their child to every parent training session.) Parents completed the BMS at 7:00 p.m. and 9:00 p.m. on Mondays, Tuesdays, and Thursdays. (Wednesday was excluded for parents due to their participation in evening training.) All ratings were based on children’s behavior observed during the 30-min period prior to the assigned rating time. Counselor ratings at 9:00 a.m., 9:30 a.m., and 11:30 a.m. evaluated onset of action of treatment. Previous data have shown strong correspondence between teacher SKAMP ratings and ratings of ADHD symptoms (Murray et al., 2009), making this a brief and useful measure of behavioral impairment resulting from ADHD symptoms. Instruction for completing the SKAMP was conducted during the 2-week pre-STP training.

Statistical Analyses

To examine treatment group differences in the onset and duration of MED treatment, a series of mixed-effects regression models were computed with each outcome measure included in separate analyses (Peugh & Enders, 2005; Tabachnick & Fidell, 2007). SPSS Version 19 was used. For each model, time (9:00 a.m., 9:30 a.m., 11:30 a.m., 1:30 p.m., 3:30 p.m., 5:30 p.m., 7:00 p.m., and 9:00 p.m.) and treatment condition (MED, BEH, and COM) were repeated-measures (time-varying) covariates. Time was also modeled as a random effect to evaluate whether significant individual differences in patient slope were present across the day. Age and sex were also included in the model as time-invariant covariates to examine the influence of these variables on outcomes. Because counselors 7:00 p.m. and 9:00 p.m. ratings were only obtained 1 day each week, separate analyses were computed adding these additional ratings. However, results indicated consistent findings whether evening times were included or excluded. Therefore, results are presented including the 7:00 p.m. and 9:00 p.m. times. Dependent variables included the SKAMP score and BMS ratings for frustration tolerance and following instructions. The interaction of Time × Treatment condition is the effect of primary interest because the interaction evaluates whether treatment conditions show a different pattern across the course of the day.

Mixed-effects regression models are advantageous relative to standard repeated-measures analysis of variance because of their ability to accommodate missing data (not all youth were able to be rated at all time points due to absences), to explicitly model relationships between repeated measures (rather than assuming sphericity), and to directly model the effects of time. Alternative covariance structures were investigated and model fit was evaluated using the Akaike’s Information Criterion. Model fit was similar between diagonal and autoregressive covariance structures. Results are presented using an autoregressive structure.

Power analyses

A priori power analyses were computed to determine the ability to detect significant differences in SKAMP ratings across time points. To be conservative, a repeated-measures ANOVA model was used with the treatment group by time interaction being the effect of interest. Mixed-effects regression models should be more powerful given improved accuracy for modeling time and tolerance for missing observations. Sample size was assumed to be 25 participants based on expected recruitment to the STP and the probability level was set to α = .05. Repeated-measures correction of effect sizes was used because the functional effect size for a within-participants design is larger than for a between-participants design. The larger functional effect size occurs because participants serve as their own controls, thereby eliminating between-participants error variance. Given the proximity of time points, a large correlation between symptoms was assumed (r = .50). However, the assumed correlation is likely to be smaller than the actual correlation, conservatively underestimating the functional effect size. Power to detect a large effect (Cohen’s d = 0.80, functional d = 1.13) between time points was excellent (0.98). Power to detect a medium effect (Cohen’s d = 0.50, functional d = 0.71) was good (0.83). These results indicate that primary mixed-effects regression models should be powerful enough to detect treatment difference for the primary outcomes. Secondary analyses involved only two time points (7:00 p.m. and 9:00 p.m.) with three outcomes. A conservative multiple-comparison correction (p < .01) was used to reduce Type 1 error for these secondary analyses. For these analyses, power to detect large effects remained very good (0.87) but was much weaker for medium effects (0.43).

Results

The final study sample included 25 participants (20 males, 5 females) aged 6 to 12 years (mean age = 8.84, SD = 1.70) with significant ADHD symptoms by parent report, baseline total raw score on ARS-IV-Parent: M (SD) = 38.76 (9.19). During the STP, participants were grouped by age into a younger group of 13 children aged 6 to 9 years (M = 7.42, SD = 0.99) and an older group of 12 children aged 9.1 to 12 years (M = 10.15, SD = 0.98). Table 2 presents baseline sample characteristics by age groups.

Table 2.

Baseline Characteristics of Study Sample by STP Age Grouping.

	Younger	Older	t/χ² (p)
n	13	12
Age range (years)	6-9	9-12
Sex	11 males,	9 males,
	2 females	3 females
Race/ethnicity
White, non-Hispanic	10	10
Black	2	1
Other	1	1
ARS-IV (M)
Inattention	8.42 (0.90)^a	7.85 (1.91)	0.94 (.356)
H/I	7.33 (2.15)	5.08 (3.50)	1.92 (.067)

Note: STP = Summer Treatment Program; ARS-IV = ADHD Rating Scale–IV; H/I = hyperactivity/impulsivity. ARS-IV scores represent the mean number of symptoms endorsed at 2 or 3.

Mean (SD).

There were no significant relationships between age and sex, χ²(10) = 0.36, p = .548, or between age and baseline ADHD symptom ratings on the ARS-IV, largest r(24) = −.19, p = .370. However, when examining symptom counts, as expected there was a trend toward greater hyperactive/impulsive symptoms in the younger age group (p = .067; Table 2). A total of 17 participants attended child care on every Wednesday evening when parent training programs were held; 8 participants attended sporadically. To ensure that attrition in the evening sessions did not influence analyses examining counselor SKAMP and BMS ratings of evening time points, participants who consistently and inconsistently attended child care were compared on baseline characteristics. No differences between consistent and inconsistent attendees emerged for a variety of demographic and behavioral indices, including age, t(23) = 1.36, p = .189; gender, χ²(1) = 0.18, p = .668; or baseline ADHD symptom ratings, t(23) = 0.98, p = .336. Consistent with intent-to-treat modeling, data from inconsistent attendees were included to most accurately estimate treatment effects.

SKAMP Ratings

Figure 3 presents counselor SKAMP ratings. Significant main effects for time, F(1, 71) = 12.38, p = .001, and treatment condition, F(2, 372) = 9.13, p < .001, were observed. Across all treatment conditions, behavior deteriorated as the day progressed, with slight improvements at the end of the day. A significant interaction between time and treatment condition was also observed, F(2, 378) = 29.36, p < .001. Medication conditions (COM and MED) were substantially better at maintaining lower levels of behavioral symptoms than was the BEH group alone. There was no significant difference between COM and MED at any time point (all ps > .50). By 3 hr post treatment (11:30 a.m.) absence of medicine (BEH) resulted in significant behavioral deterioration relative to medicine conditions (COM and MED). Furthermore, this difference was maintained throughout the 9:00 p.m. assessment (at 9:00 p.m.: COM M = 0.29, 95% confidence interval [CI] = [0.11, 0.47]; MED M = 0.21, 95% CI = [0.03, 0.38]; and BEH M = 0.71, 95% CI = [0.53, 0.89]). The overall magnitude of differences between medicine conditions and behavior therapy alone was very large (COM vs. BEH, Cohen’s d = 1.14; MED vs. BEH, Cohen’s d = 1.16).

Figure 3.

Counselor SKAMP ratings across all time periods, separately for each treatment condition.

Counselor BMS: Following Instructions

Figure 4 presents results for the counselor BMS ratings of following instructions. The results are similar to those for counselor SKAMP ratings. Following instructions deteriorates slightly, but nonsignificantly, across the day, time: F(1, 59) = 2.42, p = .125. Medicine conditions (COM and MED) significantly outperformed behavioral treatment alone, treatment condition: F(2, 382) = 8.49, p < .001. A significant interaction between time and treatment condition was also observed, F(2, 387) = 21.78, p < .001, with behavioral deterioration occurring by 3 hr post treatment in the BEH condition relative to medicine conditions and remaining through 12.5 hr post treatment. The overall magnitude of differences between medicine conditions and behavior therapy alone was large (COM vs. BEH, Cohen’s d = 0.92; MED vs. BEH, Cohen’s d = 0.87).

Figure 4.

Counselor BMS “following instructions” ratings across all time periods, separately for each treatment condition.

Counselor BMS Frustration Tolerance

Figure 5 presents results for counselor BMS ratings of frustration tolerance. There was a significant worsening of frustration tolerance throughout the day, time: F(1, 25) = 11.44, p = .002. The treatment condition effect was significant, F(2, 475) = 8.13, p < .001, indicating better maintenance of frustration tolerance in medication conditions relative to behavioral treatment alone. However, the significant time by treatment condition interaction indicated that medication effects were only present at the end of active behavioral intervention at the 9-, 10.5-, and 12.5-hr (5:30 p.m., 7:00 p.m., and 9:00 p.m.) posttreatment time points, F(2, 479) = 17.00, p < .001. The overall magnitude of differences between medicine conditions and behavior therapy alone was moderate (COM vs. BEH, Cohen’s d = 0.50; MED vs. BEH, Cohen’s d = 0.51).

Figure 5.

Counselor BMS “frustration tolerance” ratings across all periods, separately for each treatment condition.

Individual Differences in Treatment Response

There were no significant individual differences in initial (9:00 a.m.) SKAMP ratings or BMS following instructions (largest Wald z = 1.64, p = .102), consistent with the notion that all youth began the day with good behavioral control. There were significant individual differences in BMS frustration tolerance (Wald z = 2.42, p = .008), indicating that some individuals were showing initial difficulties with frustration at the beginning of the day. Significant individual differences were observed in the slope of SKAMP, BMS following instructions, and frustration tolerance ratings across the day (smallest Wald z = 1.92, p = .027). Inspection of individual slope data indicated that significant individual differences in slope were primarily due to a variable loss of behavioral control throughout the day in the BEH treatment condition.

Age and Sex Effects

Across all three measures (counselor SKAMP, counselor BMS “following instructions,” and counselor BMS “frustration tolerance”) and treatment conditions, males showed worse behavior than did females, smallest F(1, 104) = 3.44, p = .066. Younger children had worse symptoms/behaviors across all time points compared with older children, smallest F(1, 114) = 18.20, p < .001. There were no significant interactions involving age and sex (all ps > .200).

Parent BMS

These ratings at the evening time points indicated significant improvement with medication conditions (MED and COM) for following instructions and frustration tolerance at 10.5 and 12 hr post dose, smallest F(2, 21) = 4.22, p = .029, but no evidence of improvement for behavioral intervention. In addition, females were reported to have better frustration tolerance than did males across both time points, F(1, 22) = 7.44, p = .012, regardless of treatment condition.

Conclusion

In this investigation, we sought to evaluate the duration of action of extended-release stimulant medication (LDX), and we compared medicine’s relative effectiveness with behavioral intervention and a combined treatment protocol in the management of ADHD. Symptom impairment was measured by global parent and counselor ratings on the SKAMP and on two everyday behaviors—following instructions and managing frustration—as measured by counselor and parent ratings of the BMS. This study provides an innovative look at medication and behavior management effectiveness by (a) comparing three active treatments rather than to placebo alone, (b) using an ATD where individuals serve as their own controls, (c) implementing and measuring the effects of treatment in a setting that approximates a naturalistic setting, (d) using global and specific behavioral measures, (e) using a long-acting rather than immediate-release stimulant formulation, and (f) providing intensive rather than nonintensive behavioral intervention approaches. Together, these innovations provide key tests of the relative effectiveness of modern ADHD treatments in a quasi-naturalistic setting.

Behavioral intervention is the other side of the adage “Pills don’t teach skills.” From an ecological perspective, few could argue that psychosocial intervention is unimportant to treating ADHD in everyday settings such as classrooms, playgrounds, and sports events. Previous effectiveness studies provide evidence that behavioral intervention contributes to medical intervention by affecting co-occurring conditions (consider comorbid anxiety; Jensen, 2001). In two of the three domains of observation in this study—global symptom impairment and following instructions—medicine conditions improved behavior for a longer duration relative to behavioral intervention alone and this relative improvement tended to occur quickly (at 2.5 hr post dose for global symptom impairment and at 4.5 hr post dose for following instructions) and lasted throughout the day (12 hr post dose). These results are consistent with myriad previous reports that confirm medicine effects (Kaplan & Newcorn, 2011; Vaughan et al., 2011). Behavioral advantages were also observed, however, as frustration tolerance in the BEH condition was no different from medicine conditions until behavioral intervention was no longer present (after 5:00 p.m.); only then did separation of the conditions occur.

The observation that applied behavioral intervention as monotherapy sustained frustration tolerance as effectively as medicine suggests interesting differential effects of behavioral treatment. In this study, for example, this difference may be due to the variable schedule of behavioral contingencies in the STP for a free operant; frustration intolerance, a form of emotional dysregulation, occurs at irregular, unpredictable intervals, and in the STP it elicits significant aversive contingencies (i.e., time out and higher loss of points) at each occurrence. This variable schedule of aversive consequation may make behavioral control of frustration more durable, thus placing it under more rigorous and vigilant self-control by the child. Although behavioral intervention showed sustained effects with a free operant, it did not produce the same durability of effect in a discriminant operant, that is, behavior that is consequated on a more predictable schedule. Following instructions is a discriminant operant, and long-lasting effects were only noted when medicine was present. Behavioral treatment sustained self-control for a free operant (i.e., frustration intolerance) but not for a discriminant operant (following instructions). This may indicate possible selective effects of behavioral intervention as a monotherapy for some children with ADHD.

Ratings for SKAMP and for following instructions began to deteriorate after 3 hr post dose in the BEH condition. Studies of self-control (Baumeister, Bratslavsky, Muraven, & Tice, 1998) indicate that teaching an individual to discern stimulus conditions for contingencies of reward and punishment does not insure the person sustains the capacity to engage in behaviors that produce the discriminated outcome; rather a child may know what to do and still not do it. In typical adults, self-control has been demonstrated to have limits (Inzlicht & Gutsell, 2007) as it becomes “worn out” and less available to executive functions. Self-control may be exerted to produce desired outcomes, but the limits of behavioral intervention may be associated with the limits of the child’s access to effortfully directed attention, that is, to continue to use self-control. This limitation in individuals with ADHD (Barkley, 1997) may account for the early deterioration of behavioral effects in the BEH condition apart from frustration tolerance.

The overall magnitude of differences between medicine conditions (COM and MED) and behavior therapy alone (BEH) was quite large, with effect sizes approximating, and in some cases exceeding, a full standard deviation difference. This indicates the significant relative contribution of long-acting stimulant delivery systems to the treatment regimens for children with ADHD and gives support to recommendations of pharmacotherapy as monotherapy when the cost in time and money and the capacity of caregivers to deliver behavioral interventions is limited. The small sample size did not permit examination of whether unique subgroups of children may have shown better effects with behavior therapy. Further evaluation of the contribution that behavioral intervention makes to long-acting pharmacotherapy in managing specific everyday behavior in real-world settings is needed to guide treatment decisions and resource allocation.

In summary, the present study provided strong evidence for real-world effectiveness of MED in treating ADHD symptom impairments and at least one everyday behavior associated with executive function (i.e., following instructions). Onset was within 3 hr and duration was long-lasting (>12 hr). The study also suggests that behavioral intervention alone, when consistently applied, may be as useful in managing emotion as medication treatment. Using a small-N approach and applying repeated measures to children’s activity allowed us to elucidate treatment effects in a quasi-naturalistic setting across typical days, similar to viewing effects of the three intervention arms of the MTA across days within the week. Additional effectiveness data using large samples and discrete behavioral measurements are needed to further evaluate possible differential treatment effects.

Clinical Significance

To our knowledge, this is the first study to compare extended-release pharmacotherapy with behavioral intervention in the treatment of ADHD under conditions that are similar to a naturalistic camp-like setting. It compared the relative effects of behavioral intervention only, pharmacotherapy only, and behavioral intervention combined with pharmacotherapy on global behavior change, ability to follow instructions, and frustration tolerance often associated with emotional regulation. Comparing active treatments with repeated measures allowed nuances of treatment differences to emerge. In particular, frustration tolerance may be better treated with a combined behavioral and medical intervention in home settings. The addition of behavior therapy appears helpful for managing emotional responses and may be indicated for youth with significant emotional dysregulation.

Although scores on the SKAMP show statistical differences, differences between groups were clinically small with BEH rated at about 1 (slight impairment) and the scores on the other two groups rated at about 0.25 (normal). This observed small clinical effect does not negate the obvious statistical effect but clarifies the need to use further measures that are more sensitive to behavior change.

Limitations and Future Directions

The primary limitations of the present study were a modest sample size and the inability to blind the pharmacotherapy-only intervention (MED). In this condition, counselors modified their own behavior intentionally, thus they were aware of which treatment was delivered. This may have skewed ratings on the SKAMP throughout the day but is unlikely to have skewed ratings on the BMS as this was completed by parents who were not aware what treatment condition was in effect during the day. The lack of blinding of MED is certainly suboptimal but it may not have had as pronounced an effect because a similar pattern of observation was noted across measures and raters. Although the modest sample size provided more than adequate power in a repeated-measures design in which children acted as their own controls, it did not permit evaluation of differential treatment effects for subgroups, such as age or ADHD subtype. Studies combining larger samples and ATDs may explore possible differential treatment effects. Although multiple treatment designs, such as the ATD, provide unique “snapshots” of treatment effects across experimental conditions, they are limited by the types of behaviors that can be investigated, may be confounded by the ability of participants to discern the presence of different treatments (as is the case in this study), and treatment conditions may interfere with each other across presentation (McGonigle, Rojahn, Dixon, & Strain, 1987). In this study, treatment fidelity for all conditions was monitored to reduce but not to eliminate these effects. Between-participant investigations can then be used to further distinguish the variety of differential treatment effects in naturalistic settings.

Footnotes

Declaration of Conflicting Interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Dr. Short has received federal funding or research support from National Institute on Drug Abuse (NIDA), National Science Foundation (NSF), and National Institute of Mental Health (NIMH) and has acted as a consultant to Shire Development, Inc. Dr. Giuliano, Ms. Raleigh, Ms. Pucci, and Mr. Caserta report no financial or potential conflicts of interest.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by an unrestricted grant for investigator-initiated research from Shire Development, Inc. Dr. Manos has received grant/research support from, consulted for, and/or participates or participated in speakers’ bureaus for Shire Development, Inc., Cortex, Shionogi, and Novartis. Dr. Frazier has received federal funding or research support from, acted as a consultant to, received travel support from, and/or received a speaker’s honorarium from Forest Laboratories, Ecoeos, Integragen, Shire Development, Bristol-Myers Squibb, National Institutes of Health, and the Brain and Behavior Research Foundation.

Author Biographies

Michael J. Manos, PhD, is head of the Center for Pediatric Behavioral Health, Pediatric Institute, Cleveland Clinic. He is the founding clinical and program director of the pediatric and adult ADHD Center for Evaluation and Treatment of Cleveland Clinic in Cleveland, Ohio. He is adjunct faculty in Case Western Reserve University School of Medicine, Department of Psychiatry.

Donald A. Caserta, MSSA, is a clinical social worker in the Center for Pediatric Behavioral Health at the Children’s Hospital, Cleveland Clinic. He provides behavioral consultation, diagnostic assessments, and individual and group therapy for children, adolescents, adults, and their families.

Elizabeth J. Short, PhD, is a professor in the Department of Psychological Sciences and codirector of Childhood Studies at Case Western Reserve University. Her primary research and clinical interests are in the cognitive, emotional, and behavioral factors contributing to successful social adjustment and academic achievement in preschool and school age populations.

Kristina L. Raleigh, MEd, was the program director for the Summer Treatment Program (STP), ADHD Center for Evaluation and Treatment, Cleveland Clinic. Currently, she facilitates social skills training and parent coaching groups in the public schools and in private practice.

Kimberly C. Giuliano, MD, is a general pediatrician specializing in ADHD at the Cleveland Clinic, collaborating with Pediatric Behavioral Health as part of the ADHD Medication Monitoring Program.

Nicole C. Pucci, BA, is currently a doctoral student in the Department of Psychological Sciences, Case Western Reserve University.

Thomas W. Frazier, PhD, is with Cleveland Clinic Children’s Hospital Center for Autism and Center for Pediatric Behavioral Health. He is also jointly appointed in the Genomic Medicine Institute and an assistant professor in the Lerner College of Medicine. His primary clinical interest includes the assessment and behavioral treatment of individuals with autism spectrum disorder and he also specializes in the assessment and differentiation of pediatric bipolar disorder, ADHD, and other disruptive behavior disorders in his clinical practice.

References

Adesman

A. R.

(2001). The diagnosis and management of attention-deficit/hyperactivity disorder in pediatric patients. Primary Care Companion to the Journal of Clinical Psychiatry, 3, 66-77.

American Academy of Pediatrics, Subcommittee on Attention-Deficit/Hyperactivity Disorder and Committee on Quality Improvement. (2001). Clinical practice guideline: Treatment of the school-aged child with attention-deficit/hyperactivity disorder. Pediatrics, 108, 1033-1044.

American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., text rev.). Washington, DC: Author.

Barkley

R. A.

(1997). Behavioral inhibition, sustained attention, and executive functions: Constructing a unifying theory of ADHD. Psychological Bulletin, 121, 65-94.

Baumeister

R. F.

Bratslavsky

Muraven

Tice

D. M.

(1998). Ego depletion: Is the active self a limited resource? Journal of Personality and Social Psychology, 74, 1252-1265.

Biederman

(1998). Attention-deficit/hyperactivity disorder: A life-span perspective. Journal of Clinical Psychiatry, 59(Suppl. 7), 4-16.

Biederman

(2005). Attention-deficit/hyperactivity disorder: A selective overview. Biological Psychiatry, 57, 1215-1220.

Biederman

Boellner

S. W.

Childress

Lopez

F. A.

Krishnan

Zhang

(2007). Lisdexamfetamine dimesylate and mixed amphetamine salts extended-release in children with ADHD: A double-blind, placebo-controlled, crossover analog classroom study. Biological Psychiatry, 62, 970-976.

Biederman

Faraone

Milberger

Guite

Mick

Chen

(1996). A prospective 4-year follow-up study of attention-deficit hyperactivity and related disorders. Archives of General Psychiatry, 53, 437-446.

10.

Biederman

Krishnan

Zhang

McGough

J. J.

Findling

R. L.

(2007). Efficacy and tolerability of lisdexamfetamine dimesylate (NRP-104) in children with attention-deficit/hyperactivity disorder: A phase III, multicenter, randomized, double-blind, forced-dose, parallel-group study. Clinical Therapeutics, 29, 450-463.

11.

Biederman

Newcorn

Sprich

(1991). Comorbidity of attention deficit hyperactivity disorder with conduct, depressive, anxiety, and other disorders. American Journal of Psychiatry, 148, 564-577.

12.

Centers for Disease Control and Prevention. (2010). Increasing prevalence of parent-reported attention-deficit/hyperactivity disorder among children—United States, 2003 and 2007. MMWR. Morbidity and Mortality Weekly Report, 59, 1439-1443.

13.

Chronis

A. M.

Fabiano

G. A.

Gnagy

E. M.

Onyango

A. N.

Pelham

W. E.

Lopez-Williams

Seymour

K. E.

(2004). An evaluation of the Summer Treatment Program for children with attention-deficit/hyperactivity disorder using a treatment withdrawal design. Behavior Therapy, 35, 561-585.

14.

DuPaul

G. J.

Power

T. J.

McGoey

K. E.

Ikeda

M. J.

Anastopoulos

A. D.

(1998). Reliability and validity of parent and teacher ratings of attention-deficit hyperactivity disorder symptoms. Journal of Psychoeducational Assessment, 16, 55-68.

15.

Findling

R. L.

Childress

A. C.

Krishnan

McGough

J. J.

(2008). Long-term effectiveness and safety of lisdexamfetamine dimesylate in school-aged children with attention-deficit/hyperactivity disorder. CNS Spectrums, 13, 614-620.

16.

Frazier

T. W.

Weiss

Hodgkins

Manos

M. J.

Landgraf

J. M.

Gibbins

(2010). Time course and predictors of health-related quality of life improvement and medication satisfaction in children diagnosed with attention-deficit/hyperactivity disorder treated with the methylphenidate transdermal system. Journal of Child and Adolescent Psychopharmacology, 20, 355-364.

17.

Gartlehner

Hansen

R. A.

Nissman

Lohr

K. N.

Carey

T. S.

(2006). A simple and valid tool distinguished efficacy from effectiveness studies. Journal of Clinical Epidemiology, 59, 1040-1048.

18.

Hersen

Barlow

(1976). Single-case experimental designs: Strategies for studying behavior change. New York, NY: Pergamon.

19.

Inzlicht

Gutsell

J. N.

(2007). Running on empty: Neural signals for self-control failure. Psychological Science, 18, 933-937.

20.

Jensen

P. S.

(2001). Introduction—ADHD comorbidity and treatment outcomes in the MTA. Journal of the American Academy of Child & Adolescent Psychiatry, 40, 134-136.

21.

Johnston

Mash

E. J.

(2001). Families of children with attention-deficit/hyperactivity disorder: Review and recommendations for future research. Clinical Child and Family Psychology Review, 4, 183-207.

22.

Kaplan

Newcorn

J. H.

(2011). Pharmacotherapy for child and adolescent attention-deficit hyperactivity disorder. Pediatric Clinics of North America, 58, 99-120, xi.

23.

Kazdin

A. E.

(1982). Single-case research designs. New York, NY: Oxford University Press.

24.

Limbers

C. A.

Ripperger-Suhler

Boutton

Ransom

Varni

J. W.

(2010). A comparative analysis of health-related quality of life and family impact between children with ADHD treated in a general pediatric clinic and a psychiatric clinic utilizing the PedsQL. Journal of Attention Disorders, 15, 392-402.

25.

Manos

M. J.

Short

E. J.

Findling

R. L.

(1999). Differential effectiveness of methylphenidate and Adderall in school-age youths with attention-deficit/hyperactivity disorder. Journal of the American Academy of Child & Adolescent Psychiatry, 38, 813-819.

26.

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.

27.

McGonigle

J. J.

Rojahn

Dixon

Strain

P. S.

(1987). Multiple treatment interference in the alternating treatments design as a function of the intercomponent interval length. Journal of Applied Behavior Analysis, 20, 171-178.

28.

Molina

B. S.

Hinshaw

S. P.

Swanson

J. M.

Arnold

L. E.

Vitiello

Jensen

P. S.

MTA Cooperative Group . (2009). The MTA at 8 years: Prospective follow-up of children treated for combined-type ADHD in a multisite study. Journal of the American Academy of Child & Adolescent Psychiatry, 48, 484-500.

29.

MTA Cooperative Group. (1999). A 14-month randomized clinical trial of treatment strategies for attention-deficit/hyperactivity disorder. Multimodal Treatment Study of Children With ADHD. Archives of General Psychiatry, 56, 1073-1086.

30.

Murray

D. W.

Bussing

Fernandez

Hou

Garvan

C. W.

Swanson

J. M.

Eyberg

S. M.

(2009). Psychometric properties of teacher SKAMP ratings from a community sample. Assessment, 16, 193-208.

31.

Pelham

W. E.

Burrows-Maclean

Gnagy

E. M.

Fabiano

G. A.

Coles

E. K.

Tresco

K. E.

Hoffman

M. T.

(2005). Transdermal methylphenidate, behavioral, and combined treatment for children with ADHD. Experimental and Clinical Psychopharmacology, 13, 111-126.

32.

Pelham

W. E.

Greiner

Gnagy

(1998). Summer Treatment Program manual. Buffalo, NY: Comprehensive Treatment for Attention Deficit Disorders.

33.

Pelham

W. E.

Hoza

(1996). Intensive treatment: A Summer Treatment Program for children with ADHD. In Hibbs

Jensen

(Eds.), Psychosocial treatments for child and adolescent disorders: Empirically based strategies for clinical practice (pp. 311-340). New York, NY: APA Press.

34.

Peugh

J. L.

Enders

C. K.

(2005). Using the SPSS mixed procedure to fit cross-sectional and longitudinal multilevel models. Educational and Psychological Measurement, 65, 717-741.

35.

Pliszka

, & AACAP Work Group on Quality Issues. (2007). Practice parameter for the assessment and treatment of children and adolescents with attention-deficit/hyperactivity disorder. Journal of the American Academy of Child & Adolescent Psychiatry, 46, 894-921.

36.

Pliszka

S. R.

Crismon

M. L.

Hughes

C. W.

Conners

C. K.

Emslie

G. J.

Jensen

P. S.

Texas Consensus Conference Panel on Pharmacotherapy of Childhood Attention Deficit Hyperactivity Disorder. (2006). The Texas Children’s Medication Algorithm Project: Revision of the algorithm for pharmacotherapy of attention-deficit/hyperactivity disorder. Journal of the American Academy of Child & Adolescent Psychiatry, 45, 642-657.

37.

Spencer

T. J.

Biederman

Mick

(2007). Attention-deficit/hyperactivity disorder: Diagnosis, lifespan, comorbidities, and neurobiology. Journal of Pediatric Psychology, 32, 631-642.

38.

Swensen

A. R.

Birnbaum

H. G.

Secnik

Marynchenko

Greenberg

Claxton

(2003). Attention-deficit/ hyperactivity disorder: Increased costs for patients and their families. Journal of the American Academy of Child & Adolescent Psychiatry, 42, 1415-1423.

39.

Tabachnick

B. G.

Fidell

L. S.

(2007). Multilevel linear modeling. In Using multivariate statistics (5th ed., pp. 781-857). Boston, MA: Pearson / Allyn & Bacon.

40.

Van der Oord

Prins

P. J.

Oosterlaan

Emmelkamp

P. M.

(2008). Efficacy of methylphenidate, psychosocial treatments and their combination in school-aged children with ADHD: A meta-analysis. Clinical Psychology Review, 28, 783-800.

41.

Vaughan

B. S.

March

J. S.

Kratochvil

C. J.

(2011). The evidence-based pharmacological treatment of paediatric ADHD. International Journal of Neuropsychopharmacology, 14, 1-13.

42.

Wigal

S. B.

Gupta

Guinta

Swanson

J. M.

(1998). Reliability and validity of the SKAMP rating scale in a laboratory school setting. Psychopharmacology Bulletin, 34, 47-53.

43.

Wigal

S. B.

Kollins

S. H.

Childress

A. C.

Squires

, & 311 Study Group. (2009). A 13-hour laboratory school study of lisdexamfetamine dimesylate in school-aged children with attention-deficit/hyperactivity disorder. Child & Adolescent Psychiatry & Mental Health, 3, 17.