Dialectical Behavioral Therapy Skills Training for Adolescents with Attention Deficit Hyperactivity Disorder: A Pilot Study

This was a pilot feasibility study involving 10 × 2-hour treatment sessions carried out in two out-patient, community psychiatric clinics in Southern Sweden. Treatment involved a therapist manual and a workbook developed for use by the adolescents, with treatment delivered in a group format by two experienced CBT therapists, with a group size of up to six participants. Assessments occurred at pre-treatment, post-treatment (10 weeks after pre-treatment), and at 12 weeks after treatment (follow-up). Feasibility was assessed by analyzing the treatment acceptance rate, the drop-out rate once treatment was completed, and for participants who completed the treatment, their attendance at weekly sessions and patient-satisfaction questionnaire. The latter included open-ended questions (see below) that allowed preliminary investigation of the feasibility of treatment components. Outcome variables (discussed below) were assessed at pre-treatment, post-treatment and follow-up.

All participants and their parents/caregivers gave informed consent, and the study was approved by the Regional Ethics Committee of Lund (dnr: 2014/323).

Participants (N=12; 8 of whom completed treatment) were recruited via an out-patient clinic database of adolescents with a primary diagnosis of ADHD who were currently receiving treatment (medication only) and/or currently not receiving treatment (any) at two outpatient, community psychiatry clinics in Southern Sweden. The inclusion criteria were a current primary diagnosis of ADHD or ADD diagnosis; aged between 15–18 years; IQ > 80; no ongoing psychosocial treatment; and no current, comorbid psychiatric condition requiring psychosocial treatment. A total of 72 possible participants were invited to the treatment study and 12 (16.7%) expressed interest. Of these, two did not meet the inclusion criteria and were referred to individual treatment for a comorbid condition within the clinic at which they were already registered. Ten adolescents and their parents/guardians participated in an information meeting with detailed information about the study and a possibility to ask questions. At the end of the meeting, all 10 families gave written consent to participate. The participants and their guardians filled in rating scales at the end of this meeting (pre-treatment). The total completion time of all outcome measures was about 40 minutes.

All participants were previously diagnosed with ADHD/ADD according to a standardized assessment protocol at the clinics from which they were recruited. The protocol involved a clinical interview carried out by a psychiatrist assessing developmental history and diagnostic status using the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) criteria (17), a neuropsychological assessment performed by a licensed psychologist, a pedagogic evaluation report by a teacher, and rating scales filled out by parents/guardians, teachers, and the young person. All participants were tested with the fourth edition of the Whechsler Intelligence Scale for Children (WISC-IV) (18), the Integrated Visual and Auditory Continuous Performance Test IVA+Plus (19), and/or the Test of Everyday Attention for Children (TEA-Ch) attention test (20), and completed the Beck Youth Inventories (21). Parents/guardians and teachers filled in the Swanson, Nolan, and Pelham-IV (SNAP-IV) rating scales (22).

A workbook for adolescents and a therapist manual were produced based on the Swedish version of the group DBT treatment for adults with ADHD (23). In creating an adolescent workbook and therapist manual, the original treatment was reduced from 14 to 10 sessions, the written material in the workbook was reduced, and the language made more age-appropriate, and with more visually based materials included. The following topics were included in the adolescent workbook: an introduction with rules and treatment contract (one session); psychoeducation about neurobiology and ADHD symptoms (one session); mindfulness and acceptance (two sessions); chaos and control (one session); dysfunctional behavior and functional behavior analysis (two sessions); emotion theory and emotion regulation (one session); control of impulsive behavior (one session); and coping with stress (one session). The following topics from the adult workbook were not included in the adolescent version used in this study: substance abuse; self-esteem and relationships; review; and closure. Finally, an effort was made such that homework assignments were focused only on what was discussed in a particular session, and the overall number of homework assignments was reduced relative to the adult workbook.

The structure of each treatment session was as follows: agenda setting and homework review; a psychoeducational review of the skill to be trained in that session; therapist-led skill training involving role-play with group members; group discussion of the skill and its use; and discussion of planned homework exercises. All treatment sessions were held in the late afternoon (between 4–6 pm), so the participants missed nothing out of school. The participants were offered a break after one hour during which beverages and sandwiches were served. The two clinical psychologists carrying out the treatment were experienced in delivering CBT to children and adolescents with emotional and behavioral difficulties. One of the psychologists (lead author), had additional training and experience in DBT for individuals with borderline personality disorder and depression. Both psychologists lead each group.

Feasibility was assessed in the following ways: treatment acceptance rate (no a priori assumption), dropout rate (assumed <25%), attendance rate (assumed > 75%), and responses to the treatment satisfaction measure administered at post-treatment. This measure included five open questions regarding advantages/disadvantages of the group treatment, and whether the participants’ expectations were fulfilled. The preliminary efficacy of the treatment was assessed using standardized parent/guardian- and participant-completed measures.

The primary outcomes were pre- to post-treatment changes in parent/guardian-rated ADHD symptoms and participant-rated quality of life. Parent/guardian-rated ADHD symptoms were assessed via the Swanson, Nolan, and Pelham – IV Questionnaire (SNAP-IV) (22, 24), a DSM-IV based ADHD rating scale that assesses core symptoms of hyperactivity/impulsivity and inattention as well as symptoms of oppositional defiant disorder (ODD). The Swedish version of the SNAP-IV comprises 30 items, each scored on a 0–3 scale of symptom presence/absence (0 = “not at all”, 1 = “just a little”, 2 = “pretty much” and 3 = “very much”). Items 1 to 9 assess the criteria for inattention; items 11 to 19 assess impulsivity/hyperactivity; and items 21 to 28 assess the criteria for oppositional defiant disorder (ODD). Items 10 and 20 are summary questions for inattention and impulsivity/hyperactivity. Item 29 corresponds to the DSM-III-R (25) criteria for ODD, and item 30 is a summary question for ODD. Mean scores for each domain (inattention, impulsivity/hyperactivity, ODD) are computed by summing the item values and dividing by number of items in the respective domain. Clinical cut-offs for each domain indicates 95% significance. The SNAP-IV takes approximately 7 minutes to complete.

Participant-rated quality of life was assessed via the Global Quality of Life Scale (GQL), a single visual analog scale with a rating between 0 (the worst possible life situation) and 100 (the best possible life situation) (26). The GQL takes about 2 minutes to complete.

A range of secondary outcomes measures were administered to assess pre- to post-treatment change. Parents/guardians and the participants completed the Behavior Rating Inventory of Executive Function (BRIEF-P/T/SR) (27), a standardized rating scale that assesses the individual’s ability to regulate their behavior through appropriate inhibition control. The inventory is comprised of 86 items, forming eight clinical subscales: Inhibit, Shift, Emotional Control, Working Memory, Plan/Organize, Organization of Materials, and Monitor (28). Each item is scored on a 1–3 scale (1 = “never”, 2 = “sometimes” and 3 = “often”), with total score computed for each subscale and transformed into a T score (range from 30 to 100). Higher T score indicates greater difficulties with an average T score = 50, and T scores > 65 being considered above the threshold for clinical impairment. The BRIEF-SR and Parent/teacher Report Forms take approximately 15 minutes to complete.

Participants completed the Depression and Anxiety Scales from the Beck Youth Inventories (BYI) (21). Each scale is comprised of 20 items, scored on a 0–3 scale (0 = “never”, 1 = “sometimes”, 2 = “often” and 3 = “always”); summed to a total score with higher scores indicating more severe symptoms. Clinical cut-offs for the BYI Depression and Anxiety scales are above the 74th percentile compared to published norms. The Depression and Anxiety Scales take about 10 minutes to complete. Finally, participants completed the Five Facets of Mindfulness Questionnaire (FFMQ) (29), comprised of 29 items, scored 1–5 (1 = “never”, 2 = “seldom”, 3 = “sometimes”, 4 = “often” and 5 = “always”); summed to yield total and subscale scores. Only the total score is used in this study; higher scores indicate greater levels of mindfulness. The FFMQ takes about 5 minutes to complete.

Pre- to post-treatment changes on the primary and secondary outcomes were evaluated using Wilcoxon Signed Rank Tests in SPSS (IBM SPSS Statistics Version 28.0.1.0 (142) with a prespecified significance level (p <=.05). Wilcoxon Signed Rank Tests are nonparametric, i.e. they do not assume (or require) that data are normally distributed. The test statistic is the absolute value of the smaller of the sum of ranks of positive and negative differences. The formula is as follows:

Uncontrolled pre- to post-treatment effect sizes r were calculated using Z-score divided by the square root of the total number of observations. Analyses were first carried out in an intention-to-treat manner, i.e. based on all available data for the 10 participants who started treatment, and using the last observation carried forward method. The analyses were repeated based on all available data from the treatment completers; both sets of findings are reported.

The ten participants who started treatment and their parents/guardians filled out all measures at pre-treatment. One participant (who didn’t complete treatment) filled out a half of a questionnaire (FFMQ) which was not included in the outcome analyses. Of the eight participants who completed treatment, all eight and seven of their parent/guardians filled out all outcome measures at post-treatment and the follow-up. One parent/guardian of treatment completers failed to fill out the outcome measures at post-treatment but did complete the follow-up assessment. There were no missing items on the rest of the completed measures.

All ten adolescents were on stable stimulant medication at the time of enrollment to the study. One participant discontinued treatment after session 1 and another after session 3, leaving 5 participants in one group and 3 in the other group. There were three males and two females in one group and three males in the other group. All participants were between 15 and 17 years of age with a mean age of 16. Four participants were 15 and the other four were 17 years old. Four participants had a learning disability. Characteristics of the ten study participants are presented in Table 1.

Of the 72 potential participants who were invited to participate in the study treatment, 12 accepted and 10 met the study inclusion criteria. Eight of the 10 (80%) who were enrolled in the study completed the treatment. One participant wanted to try the treatment and dropped out after one session, while another attended three sessions and didn’t want to continue because he perceived that he had milder difficulties than the rest of the group (self-expressed reason). Of the eight participants who completed the treatment, they attended 8 out of the 10 planned group sessions. No adverse events were reported during the treatment phase or the 12-week post-treatment follow-up phase. According to the treatment satisfaction filled out at post-treatment by the eight participants who completed treatment, most were satisfied with the treatment. In response to the question, “What is your opinion about the treatment?”, seven participants indicated that the treatment was good or very good. Five participants reported that their expectations were fully met by the treatment, while two participants noted partial fulfillment of their expectations, and one participant did not respond to the question. They reported several advantages of the skills training, such as the possibility to learn new knowledge, to feel calm when practicing mindfulness, and to talk about their problems. The participants also reported the benefits of meeting other adolescents with ADHD. One participant indicated that the treatment was too time consuming, and the sessions and homework assignments competed with time needed for schoolwork; another adolescent wanted a shorter break in the middle of sessions; and another wanted more time to discuss their own problems in the group.

As planned, data were analyzed according to the intention-to-treat principle (all participants who started treatment) and per-protocol (treatment completers). Table 2 presents the medians, and Z-scores for the outcome measures at T1, T2, and T3, the p-values for the change score, and r (uncontrolled effect size) for primary and secondary outcome variables from T1 to T2, and T1 to T3 according to intention-to-treat analyses.

There was no improvement for the primary outcomes of parent/guardian-rated ADHD or participant-rated quality of life between pre-treatment (T1) and post-treatment (T2) according to the intention-to-treat analyses. However, there was a significant improvement for parent/guardian rated ADHD symptoms between pre-treatment (T1) and the 12-week follow-up (T3), with a medium effect size. A similar pattern was observed for the secondary outcomes with no significant improvements for the pre-treatment to post-treatment phase but significant improvements between pre-treatment and the 12-week follow-up (except participant-rated mindfulness), with the effect sizes in the moderate to large range.

Table 3 presents the medians, and Z-scores for the outcome measures at T1, T2, and T3, the p-values for the change score, and r (uncontrolled effect size) for primary and secondary outcome variables from T1 to T2, and T1 to T3 for the participants who completed treatment.

The completer analyses showed a similar pattern as the intention-to-treat analyses with no significant improvement for the primary outcomes of parent/guardian-rated ADHD or participant-rated quality of life between pre-treatment (T1) and post-treatment (T2). Nevertheless, there were significant improvements for these completer outcomes between pre-treatment (T1) and the 12-week follow-up (T3), with a medium effect size. Similarly, there were no significant improvements for the secondary outcomes for the completers between pre-treatment and post-treatment but significant improvements between pre-treatment and the 12-week follow-up (except participant-rated mindfulness), with effect sizes in the moderate to large range.