C OMORBID ADHD AND CBT FOR ANXIETY i Comorbid ADHD: Implications for Cognitive-Behavioral Therapy of Youth with a Specific Phobia Thorhildur Halldorsdottir Dissertation submitted to the faculty of the Virginia Polytechnic Institute and State University in partial fulfillment of the requirements for the degree of Doctor of Philosophy In Psychology Thomas H. Ollendick, Chair Martha Ann Bell Susan W. White Bradley A. White Jyoti Savla January 24, 2014 Blacksburg, VA Keywords: Attention-deficit/hyperactivity disorder, specific phobia, comorbidity, one-session treatment Copyright 2014, Thorhildur Halldorsdottir COMORBID ADHD AND CBT FOR ANXIETY ii Comorbid ADHD: Implications for Cognitive-Behavioral Therapy of Youth with a Specific Phobia Thorhildur Halldorsdottir ABSTRACT Objective: Although findings have been mixed, accumulating evidence suggests that co- occurring attention-deficit/hyperactivity disorder (ADHD) diagnoses and symptoms negatively predict cognitive-behavioral therapy (CBT) outcomes for anxious youth. The current study extends past research by examining the association of not only ADHD but also other features of ADHD with treatment outcomes of youth who received an intensive CBT for a specific phobia. Method: 135 youth (ages 6-15; 52.2% female; 88.2% white) were randomized to either an individual or parent-augmented intensive CBT targeting a specific phobia. Latent growth curve models were used to explore the association of ADHD symptoms, effortful control, sluggish cognitive tempo, maternal depression and the two treatment conditions (i.e., individual versus parent-augmented) with pre-treatment severity of the specific phobia and the trajectory of change in the severity of the specific phobia from pre-treatment to the 6-month follow up after the intervention. Results: As expected, higher levels of ADHD symptoms were associated with lower levels of effortful control and increased maternal depression at pre-treatment. Contrary to expectations, ADHD symptoms and its associated difficulties were not significantly associated with treatment outcomes. Conclusion: Overall, the findings lend support to the generalizability of intensive CBT for a specific phobia to youth with comorbid ADHD and associated difficulties. Implications and limitations of the study are discussed. COMORBID ADHD AND CBT FOR ANXIETY iii Acknowledgements I would like to acknowledge a number of individuals for their support throughout my graduate studies. First, I would to express appreciation to my supervisors and colleagues. I am especially thankful for the privilege to work with my advisor, Prof. Thomas Ollendick, for his encouragement, enthusiasm, and excellent guidance in all my academic endeavors. I would also like to thank other committee members, Martha Ann Bell, Susan White, Brad White and Tina Savla, for their insightful suggestions and comments. I am also grateful to my lab members and friends in Blacksburg who became my family away from home. Secondly, I am very thankful for a supportive and encouraging husband and family. I cannot envision going through graduate school without my husband. Throughout my life, I have also been fortunate to have strong female role models in my family. To my grandmother, Asta Bjarnadottir, you are dearly missed. COMORBID ADHD AND CBT FOR ANXIETY iv Table of Contents Acknowledgements iii Table of Contents iv List of Tables v List of Figures vi I. Introduction 1 Attention-Deficit/Hyperactivity Disorder (ADHD) 1 Comorbid ADHD and Anxiety 2 Comorbidity Treatment Outcome Studies 3 ADHD and CBT Outcomes 6 Current Study 11 Specific Aims 12 Hypotheses 12 II. Method 14 Participants 14 Interventions 14 Measures 16 Procedure 18 Data Analysis 19 III. Results 22 Attrition 22 Pretreatment Differences 23 Latent Growth Curve Models 23 IV. Discussion 27 V. References 35 VI. Tables 47 VII. Figures 57 COMORBID ADHD AND CBT FOR ANXIETY v List of Tables Table 1. Breakdown of pre-treatment characteristics of comorbid ADHD group 47 Table 2. Comparison of completers at each time point on key demographic variables 48 Table 3. Pre-treatment descriptives for participants with and without a co-occurring ADHD diagnosis 49 Table 4. Pre-treatment descriptives broken down by participants with and without elevated ADHD symptoms 50 Table 5. Comparison of treatment outcomes across time within participants with comorbid ADHD broken down by medication status 51 Table 6. Means, standard deviations, skewness, kurtosis, and correlations of study variables 52 Table 7. Unstandardized results of the latent growth curve analyses for all models 53 Table 8. Standardized results of the latent growth curve analyses for all models 54 Table 9. Unstandardized and standardized results of the latent growth curve analyses for all models with gender as a covariate 55 Table 10. Standardized results of the latent growth curve analyses for all models with gender as a covariate 56 COMORBID ADHD AND CBT FOR ANXIETY vi List of Figures Figure 1. Flow chart of all participants through the study 57 Figure 2. Unconditional growth curve model of treatment outcomes at posttreatment and at the 1-month and 6-month follow ups for participants across treatment conditions 58 Figure 3. Growth curve model with ADHD as a predictor of treatment outcomes at posttreatment and at the 1-month and 6-month follow ups for participants across treatment conditions 59 Figure 4. Growth curve model with effortful control as a mediator of ADHD symptoms and treatment outcomes at posttreatment and 1-month and 6-month follow ups 60 Figure 5. Growth curve model with symptoms of sluggish cognitive tempo and ADHD symptoms as predictors of treatment outcomes at posttreatment and at the 1-month and 6-month follow ups 61 Figure 6. Growth curve model with maternal depression and ADHD symptoms as a predictor of treatment outcomes at posttreatment and 1-month and 6-month follow ups 62 Figure 7. Growth curve model with ADHD as a moderator of treatment outcomes at posttreatment and at the 1-month and 6-month follow ups 63 Figure 8. Clinician severity rating means for the treated specific phobia across time derived from the unconditional model 64 COMORBID ADHD AND CBT FOR ANXIETY 1 I. Introduction Several evidence-based treatments have been identified for a variety of disorders in children and adolescents (Ollendick & King, 2012; Pelham & Fabiano, 2008). However, a significant number of youth do not respond to these treatments (Ollendick & Sander, 2012). Given high rates of comorbidity (Angold, Costello, & Erkanli, 1999; Costello, Mustillo, Erkanli, Keeler, & Angold, 2003, Merikangas et al., 2010), it is possible that some youth do not respond because of co-occurring disorders that limit the effectiveness of these interventions. Indeed, accumulating evidence suggests that co-occurring attention-deficit/hyperactivity (ADHD) negatively predicts outcomes in children and adolescents treated for a variety of anxiety disorders (Halldorsdottir & Ollendick, 2013); however, it is poorly understood at present how ADHD impedes successful treatment outcomes in anxious youth. In an attempt to address this gap, the current study expands on the literature by exploring various facets of ADHD that may shed light on how the disorder may hinder optimal treatment gains in children and adolescents with a specific phobia, a common anxiety disorder, but one that has not yet been examined in terms of comorbidity with ADHD. Toward that end, various difficulties associated with ADHD in addition to the diagnosis itself will be examined as predictors and moderators of treatment outcome. Predictors inform what baseline characteristics or factors facilitate or hinder successful treatment outcomes across treatment conditions, whereas moderators indicate which specific treatment works best for whom (Kraemer, Wilson, Fairburn, & Agras, 2002). Attention-Deficit/Hyperactivity Disorder (ADHD) ADHD is among the most common psychiatric disorders in youth (Merikangas et al., 2010). There are three different subtypes of ADHD: ADHD predominately inattentive presentation (ADHD-I), ADHD predominately hyperactive/impulsive presentation (ADHD-H), COMORBID ADHD AND CBT FOR ANXIETY 2 and ADHD combined presentation (ADHD-C, APA, 2000). ADHD-I is characterized by developmentally inappropriate inattention, whereas ADHD-H is characterized by excessive hyperactivity/impulsivity. Children with ADHD-C present with a combination of the inattentive and hyperactive/impulsive symptoms seen in the other two subtypes. Lifetime prevalence rates of ADHD subtypes range from 8.5-19% (Kessler, Berglund, Demler, Jin, & Walters, 2005; Merikangas et al., 2010). In addition to the core symptoms, children with ADHD are known to have deficits with executive functioning, emotion regulation, and social skills (Barkley, 1997; Hinshaw, 2003; Nigg, 2006). Treatments building on operant strategies (i.e., behavior modification techniques) have been identified as evidence based treatments for ADHD (Pelham & Fabiano, 2008). These interventions involve teaching caregivers and/or teachers to manage the child’s behavior systematically through attending to and reinforcing appropriate behaviors and ignoring or punishing inappropriate behaviors (MTA Cooperative Group, 1999). Behavioral interventions have been shown to lead to significant reductions in core ADHD symptoms (MTA Cooperative Group, 1999); although the treatment gains are frequently transitory and not maintained over time (Molina et al., 2009). In regard to barriers to long-term outcomes, parental psychopathology, including depression, has been found to interfere with achieving an optimal treatment response (Chronis, Chacko, Fabiano, Wymbs, & Pelham, 2004). Comorbid ADHD and Anxiety Rates of comorbid ADHD and anxiety disorders (AD) are greater than would be expected by chance (Angold et al., 1999), ranging from 11-25% (Biederman, Newcorn, & Sprich, 1991; Jensen et al., 2001, Kendall et al., 2010; Larson, Russ, Kahn, & Halfon, 2011). Children with dual diagnoses of ADHD and AD evince greater impairment than youth with ADHD or AD COMORBID ADHD AND CBT FOR ANXIETY 3 alone (Bowen, Chavira, Bailey, Stein, & Stein, 2008; Jensen et al., 2001). Specifically, when compared to youth with either ADHD or AD alone, youth with comorbid ADHD and AD present with greater symptom severity, more attentional problems, more school fears, and decreased social competence (Bowen et al., 2008). Furthermore, youth with comorbid ADHD and AD are more likely to display symptoms of sluggish cognitive tempo (i.e., daydreaming, being slow to respond, and easily confused) and poor response inhibition (i.e., inhibiting an automatic response for a more task appropriate response) when compared with youth with ADHD alone (Lee, Burns, Snell, & McBurnett, 2013; Pliszka, 1989; 1992; Schatz & Rostain, 2006; Skirbekk, Hansen, Oerbeck, & Kristensen, 2011). Consequently, identifying and, if needed, devising effective treatments for this population is of clinical importance (Jarrett & Ollendick, 2008). Comorbidity Treatment Outcome Studies Cognitive-behavioral treatments (CBTs) have been shown to be effective in treating youth with various ADs. CBT traditionally includes components of psychoeducation, teaching the child to identify and challenge dysfunctional cognitions, and exposure to the anxiety- provoking stimuli along with graduated exposure inside and outside the therapy session. Using CBT principles, the One-Session Treatment (OST; Öst, 1989; 1997) – the focus of the present study - has been shown to be effective in treating youth with SPs (Ollendick et al., 2009). The intervention involves psychoeducaton, intensive graduated in vivo exposures, cognitive challenges, participant modeling and reinforced practice. During the intervention, change is thought to be facilitated through habituation, the elicitation and challenging of dysfunctional cognitions, skill acquisition, and reduction in behavioral avoidance (Davis & Ollendick, 2005). Although EBTs have been identified for ADHD and ADs, treating youth with comorbid ADHD and anxiety may pose difficulties in need of further consideration. As noted, evidence is COMORBID ADHD AND CBT FOR ANXIETY 4 beginning to accumulate that anxious youth with co-occurring ADHD diagnoses and symptoms, when examined independent of other disruptive behavior disorders, respond less well to CBT treatments (Garcia et al., 2010; Halldorsdottir et al., 2013; Storch et al., 2008). Still, some studies have found comparable outcomes between anxious youth with and without ADHD (Manassis et al., 2002; Southam-Gerow, Kendall, & Weersing, 2001). Among these studies, Storch et al. (2008) examined the influence of comorbid ADHD on treatment outcomes in 96 children (ages 7 to 19; 45% females) with a primary diagnosis of obsessive-compulsive disorder (OCD) after receiving family-based CBT with exposure and response prevention (ERP). Approximately 26% of the participants had comorbid ADHD (n = 24). Response rates were significantly lower and remission rates were marginally lower (p = .09) in participants with comorbid ADHD than those without this comorbid disorder. In explaining their findings, Storch and colleagues indicated that youth with comorbid ADHD may have been unable to attend long enough to habituate to anxiety-provoking stimuli and engage in deliberate cognitive restructuring during the sessions (Storch et al., 2008). They also postulated that the comorbid children may have had difficulties with planning and implementing exposures or other therapeutic tasks outside of the treatment session. To accommodate for these difficulties, the authors suggested teaching parents specific behavior modification techniques prior to pursuing CBT for OCD and implementing these adjunctive procedures throughout treatment. Similar findings were shown in the multisite Pediatric OCD Treatment Study (POTS, 2004). In the POTS trial, treatment outcomes were evaluated for 112 youth (ages 7 to 17; 50% females) with a primary diagnosis of OCD who were randomly assigned to receive individual CBT, medication, the combination of the individual CBT and medication regimen, or a pill placebo. Approximately 15% of the sample (n = 17) met diagnostic criteria for ADHD (Garcia et