Brief ACT Protocol in At-risk Adolescents with Conduct Disorder and Impulsivity

Background and objectives

Conduct disorder (CD) is a persistent and difficult-to-treat psychological problem in childhood and adolescence, characterized by persistent patterns of problematic behavior, including defiant and oppositional behaviors, antisocial activities, lying, theft, running away, physical violence, and sexually coercive behaviors. CD generates serious consequences: physical harm, reduced legal, occupational and educational opportunities, and higher prevalence of serious physical disease. Traditional treatments (parent management training, multisystemic therapy, functional family therapy) have important limitations: limited efficacy, difficulty generalizing changes, and poor long-term maintenance.

Acceptance and Commitment Therapy (ACT) is a contextual psychological intervention that promotes psychological flexibility: the ability to maintain contact with present private experiences without needing to avoid or escape from them, while adjusting behavior according to personal values and goals. The main objective of this study was to explore the effect of a brief, individualized ACT protocol (four 90-minute sessions implemented over two weeks) specifically designed for at-risk adolescents with CD and impulsivity who had failed previous treatments. The ACT protocol was designed to specifically address the inflexible pattern of experiential avoidance and to promote a values-based self-control repertoire.

The brief protocol was based on ACT literature (Hayes et al., 1999; Wilson & Luciano, 2002), previous experimental and clinical analogues, and was designed to change the adolescent's rationale for psychological treatment. For the first time, adolescents were made responsible for their own decisions, and the intervention was centered on what really mattered in their lives.

Method

Participants

Five adolescents (15 to 17 years old; three boys) who met criteria for conduct disorder and impulsivity participated in this study. All came from dysfunctional and low-income social contexts. They presented a wide range of disruptive behaviors in the classroom, home, and neighborhood, including consumption of tobacco, alcohol, and other drugs. All had legal problems and were sentenced to perform community service and to receive psychological treatment and rehabilitation programs. At the time the study began, they had been receiving psychological treatment for 2 to 3 years through school psychology teams and community services. Specific participant characteristics were:

P1 (Juan, 15 years old, boy): Dysfunctional family, low income. Legal issues for vandalism, theft. Aggressive, oppositional, defiant behaviors. Impulsivity problems.
P2 (Luis, 15 years old, boy): Functional family, low-to-middle income. Legal issues for assault, robbery. Tobacco, cannabis, alcohol consumption. Classroom defiance to teachers. Rights violations with peers.
P3 (Isaac, 13 years old, boy): Functional family, low income. Legal issues for theft, vandalism. Tobacco, alcohol consumption. Class truancy. Participation in vandal and group acts.
P4 (Laura, 16 years old, girl): Dysfunctional family. Divorced parents living temporarily in precarious settlement. Economic dependence. Nutritional, abuse, violent behavioral problems. Unpredictable relationships with teachers.
P5 (Ana, 16 years old, girl): Functional family, low income. Legal issues for aggressive behavior, neighborhood delinquency, drug consumption. Low self-control.

A functional analysis conducted at the beginning of the study revealed that all adolescents had a history of problematic conflict resolution characterized by an inflexible and persistent pattern of experiential avoidance. All showed a repertoire of deliberate problematic actions such as aggressive verbal reactions, physical violence, drug use, etc., directed toward immediate reduction or elimination of affect (anger, tension, thoughts/feelings of inferiority, frustration, etc.). The persistence of these repertoires limited opportunities for school success and social integration. Analysis revealed that they had become resistant to and uninterested in psychological interventions.

Design

A multiple case study with five participants was conducted to assess the effect of a brief, individualized, four-session ACT protocol implemented over two weeks. The protocol was delivered by a psychologist who had previously implemented cognitive-behavioral interventions with limited success. Disruptive and desirable behaviors were assessed at pretreatment, during treatment, and at posttreatment by the adolescents themselves and their teachers. Three self-report instruments assessing impulsivity, self-control, and psychological flexibility were also administered at pretreatment and posttreatment. Data from teachers, peers, family members, and neighbors were collected at posttreatment and at 1-year follow-up.

Intervention / Conditions

A brief ACT protocol was specifically designed to functionally suit the resistance to, or lack of interest in, psychological treatment after years of unsuccessful results. The protocol consisted of four individual 90-minute sessions conducted twice weekly over two weeks. All sessions were audio-recorded with participants' authorization.

The protocol was based on ACT books (Hayes et al., 1999; Wilson & Luciano, 2002), information from experimental and clinical analogues, and a previously published English protocol with other protocols designed for adolescents. A detailed description of the protocol was published in Spanish in the article's Annex.

The protocol focused on four core objectives: (a) to establish a context between therapist and adolescents to promote a sense of personal responsibility for behavior change; (b) to confront adolescents with the effect of their behavior regulation (pros and cons) and the experience of creative hopelessness of the limited behavior repertoire; (c) to clarify personally valued directions; and (d) to promote defusion skills so adolescents could take charge of their private thoughts and emotions and choose actions according to their values.

Procedure and session content:

Two days before protocol implementation, teachers were given records of disruptive and desirable behaviors. The four sessions were then implemented over two weeks:

Session 1 (Day 3): Establish context for therapeutic interaction and analysis of behavior effectiveness in class. Objectives were to generate a new context between therapist and adolescents and to begin analysis of pros and cons of what they do in class. Therapist asked permission to audio-record sessions.
Session 2 (Day 3): Experience of creative hopelessness through multiple examples of problematic classroom behavior. Multiple examples of problematic behavior controlled by a wide range of private aversive events (anger, humiliation, frustration, etc.) were analyzed. Emphasis was placed on feelings of powerlessness and the rules driving their aggressive reactions.
Session 3 (Day 8): Clarification of personally valued directions. Adolescents were facilitated to discern the things they wanted to achieve in their lives. Emphasis was placed on the participant's ability to choose freely, strengthening personal responsibility in the actions they chose.
Session 4 (Day 10): Training in defusion skills and sense of self as context. Defusion exercises were taught through multiple-exemplar training to allow adolescents to experience difficult or unwanted private events as context in which to act. These exercises were designed to establish regulatory functions of experience through defusion.

At posttreatment (Days 15-16), data were collected from teachers and participants. One year after intervention completion, the psychologist contacted the adolescents' natural contexts and the adolescents themselves to obtain relevant information about their progress.

Outcome measures

Naturally observed behavioral measures:

Disruptive behaviors: Defined as actions that would typically be under the direct control of reducing negative emotions or producing immediate pleasure, such as breaking objects, smoking in class, stealing objects from peers, fighting and starting arguments with peers, interrupting school equipment and furniture, missing or not appearing in classes, hiding to escape school, threatening and bullying behaviors, cursing when questioning teachers' matters, etc.
Desirable behaviors: Defined as actions that would typically be under the control of directions valued by adolescents. Behaviors included classroom attendance, arriving on time, complying with teacher demands, doing homework, participating in group activities, showing frustration tolerance, going to exams, interacting calmly with teachers and peers, persevering in tasks, etc. Teachers provided this information on eight days: two days before intervention and two days after intervention. Two independent observers rated the frequency of behaviors during the eight days.

Self-report measures:

Acceptance and Action Questionnaire-II (AAQ-II): General measure of experiential avoidance and psychological inflexibility. Consists of 7 items rated on a 7-point Likert-type scale. Items reflect unwillingness to experience unwanted thoughts and emotions (e.g., "I am afraid of my feelings," "I worry about not being able to control my worries and feelings") and inability to be in the present moment and behave according to value-directed actions when experiencing a psychological event (e.g., "My painful memories and thoughts make it difficult for me to live a life that I would value"). The initial 10-item version of the AAQ-II was used, and items were scored so that higher scores reflect greater psychological inflexibility. The Spanish version by Capafons et al. (2013) showed good psychometric properties and one-factor structure.
Computerized Magallanes Impulsivity Scale (EMIC; Servera & Llabrés, 2000): Computerized version of the Matching Familiar Figure Test (MFFT; Kagan, 1965) that measures reflexivity-impulsivity cognitive style. Consists of 16 items in which the participant observes a sample figure centered in the upper part of the computer screen and then six comparison figures in the lower part of the screen separated by a horizontal line. The participant must select the comparison figure that is identical to the sample. In this study, the impulsivity score provided by the EMIC after test completion was used. The EMIC has shown high internal consistency and good test-retest reliability.
Self-Control Scale (SCS; Rosenbaum, 1980): Measures the tendency to apply self-control methods to solve behavioral problems. Consists of 36 items rated on a 6-point Likert scale. The SCS covers four aspects related to self-control: (a) the use of cognitions and self-statements to control emotional and behavioral responses, (b) application of problem-solving strategies, (c) ability to delay immediate gratifications, and (d) perceived self-efficacy. The Spanish version by Capafons & Barreón (1989) was used.

Qualitative information collection:

Qualitative information was obtained from the adolescents themselves as well as from other people in the school and community. Adolescents provided information during evaluation, treatment process, and follow-up. Information provided by significant others (peers, family members, friends, neighbors) was collected at posttreatment and follow-up. Adolescents were asked about the presence or absence of problematic behaviors as well as prosocial actions and behaviors important to them.

Data analysis

Frequencies of disruptive and desirable behavior based on teachers' daily reports were analyzed. Paired t-tests were used to compare changes from pretreatment to posttreatment on self-report measures: impulsivity (EMIC), self-control (SCS), and psychological flexibility (AAQ-II). Individual line graphs were examined for each participant showing changes in disruptive and desirable behavior across pretreatment, treatment, and posttreatment.

Results

Disruptive and desirable behaviors in class:

Figure 1 shows the frequency of disruptive and desirable behaviors in class based on teachers' daily reports. Considering the whole sample, the frequency of disruptive behaviors at pretreatment was clearly higher than the frequency of desirable behaviors (disruptive: M = 7.70, SD = 3.59; desirable: M = 3.1, SD = 1.37; t(9) = 4.27, p = .002). However, during intervention, all participants showed significant decreases in disruptive classroom behavior and increases in desirable behavior. At posttreatment, the number of disruptive behaviors was near zero (M = .40, SD = .52) and the preposttreatment difference was statistically significant (t(9) = 6.48, p < .001, d = 2.85). Similarly, participants showed a clear change in behavioral tendency (a statistically significant increase in desirable behaviors from pretreatment to posttreatment: M = 13.30, SD = 2.31; t(9) = -10.71, p < .001, d = 5.37). In fact, the frequency of desirable behaviors at posttreatment was clearly higher than the frequency of disruptive behaviors (t(9) = -17.50, p < .001).

Self-report measures:

Results of self-report measures are shown in Table 3 and Figure 2. All participants showed decreases in impulsivity as measured by the EMIC. The preposttreatment difference was statistically significant (pretreatment: M = 57.60, SD = 12.40; posttreatment: M = 41.40, SD = 15.85; t(4) = 5.54, p = .005, d = 1.14). All participants also showed increases in self-report as measured by the SCS, with the pre-posttreatment difference being statistically significant (pretreatment: M = 11.60, SD = 10.24; posttreatment: M = 26.00, SD = 11.96; t(4) = -4.84, p = .008, d = 1.29). Finally, 3 of 5 participants increased their scores on psychological flexibility as measured by the AAQ-II, while the remaining 2 showed small increases. The pre-posttreatment difference was not statistically significant (pretreatment: M = 33.40, SD = 9.10; posttreatment: M = 43.40, SD = 4.39; t(4) = -2.29, p = .083, d = 1.40).

Presence or absence of problematic and desirable behaviors:

Table 4 shows the presence or absence of problematic and desirable behaviors in a general sense and beyond school context according to reports from participants, teachers, family members, peers, and neighborhood. At posttreatment, all five participants showed an increase in the number of actions connected to positive progress, while negative reactions decreased. This difference was much greater at 1-year follow-up, with only one participant showing problematic behavior and all participants showing the presence of desirable behaviors.

One-year follow-up:

Psychological flexibility skills and vital changes experienced in many other relevant areas were expansive beyond the school context and daily coexistence with peers. For example, Juan and Luis found good jobs and were working well. Luis began studying music in a very demanding community conservation program, and all of them were participating in new group activities and maintaining good relationships. For example, Laura had a new healthy partner, Ana maintained a good relationship with her sister, and Juan was collaborating with his family not only in daily tasks but also economically. Four of the five adolescents using alcohol, tobacco, or cannabis quit during the one-year follow-up as reported by significant others. No illegal behavior such as vandalism, robbery, or aggressive behaviors occurred either within or outside school during the entire year. Furthermore, they reoriented their lives toward more valued goals and engaged in healthy activities and relationships.

Discussion and conclusions

A brief, individualized, four-session ACT protocol was implemented with 5 at-risk adolescents who had legal issues due to law violations. All of them had been receiving psychological treatment for 2 to 3 years with limited results. Due to this lack of positive results, a radical change was proposed in which adolescents were given the role of taking responsibility for their own decisions. All participants improved their behaviors quickly, over a span of two weeks, which surprised the therapist as well as other people (e.g., teachers, peers, family). Changes were measured in class during treatment and in general contexts, and 1-year follow-up informants showed that changes expanded across different areas in their lives.

With respect to the former assessment, a clear difference was observed in the adolescents' classroom behaviors. The frequency of problematic behavior (e.g., fighting with classmates, oppositional responding to teachers, running away from classes and exams, etc.) decreased to a near zero point by the end of sessions and, more importantly, these behaviors no longer interfered with teacher-peer-student relationships or academic progress. Additionally, desirable behaviors in class increased significantly (e.g., classroom attendance, group participation, persistence in difficult tasks, ability to tolerate frustration and tension, and reduction of violence, etc.). Furthermore, participants also showed changes in self-report measures at posttreatment. In this regard, it is important to note that the rapid change obtained only two weeks after pretreatment was by request that participants respond under temporal control (i.e., in the AAQ-II and SCS).

Most importantly, at 1-year follow-up, participants strengthened their psychological flexibility skills and experienced vital changes in many other relevant areas. For example, Juan, Luis, and Laura were able to complete Mandatory Secondary Education, whereas Isaac and Ana completed the following academic year. It could be said that treatment impacted the functioning of this group of youths, which changed in ways that placed them within normative levels of functioning. This has been a clearly relevant clinical criterion for change. All four adolescents using alcohol, tobacco, or cannabis quit during the 1-year follow-up as reported by significant others.

Despite the limitations associated with single-case studies, which do not allow isolation of intervention effects from other possible sources of change, it is important to highlight that, for the first time in their lives, this small group of at-risk adolescents who had a very disabled profile had an opportunity to change their lives. The characteristics of severe impulsive and antisocial repertoire, as well as resistance to previous treatment, defined these adolescents as "lost cases." However, this changed after implementation of the brief ACT protocol. Further studies with better measures are required to isolate the relevant components of the ACT protocol as well as to compare it with other treatment conditions.

However, in this study, naturalistic observation and consensual definition of specific problematic and desirable behaviors in class were a persuasive source of data that was sensitive to the important difference in the youths' lives as they themselves expressed it and was experienced by others they were interacting with. The brief ACT protocol appears to be effective with a population that typically drops out of treatment early. Important characteristics of the protocol are the motivational effect of the first session, the brevity of the protocol, its minimal intrusiveness, and the ecological validity of the intervention and measures used. In conclusion, the brief ACT protocol deserves further research.

Significance and contribution

This study contributes significantly to the field of conduct disorder treatment in adolescents by demonstrating that a brief, individualized, low-cost ACT protocol (four 90-minute sessions over two weeks) can produce clinically significant and long-term maintained changes in a highly at-risk group of adolescents who had failed previous treatments for 2-3 years. Unlike traditional approaches that emphasize external behavior modification or require active parent/caregiver participation, this ACT protocol places adolescents in control of their own decisions and changes, which appears to be especially motivating and effective for a treatment-resistant population. The rapid (two weeks) and maintained (one year) change suggests that the protocol has important motivational and ecological validity characteristics that make it particularly suitable for adolescents with conduct disorder and impulsivity who resist conventional treatments.