Exposure-Based Cognitive-Behavioral Therapy for Disruptive Mood Dysregulation Disorder: An Evidence-Based Case Study

Abstract

Severe, chronic irritability is one of the most frequently reported problems in youth referred for psychiatric care. Irritability predicts adult depressive and anxiety disorders, and long-term impairment. Reflecting this pressing public health need, severe, chronic, and impairing irritability is now codified by the DSM-5 diagnosis of disruptive mood dysregulation disorder (DMDD). Since DMDD has only recently been added as its own nosological class, efficacious treatments that specifically target severe irritability as it presents in DMDD are still being developed. In a recent pilot study, we described the general concept of exposure-based cognitive-behavioral therapy (CBT) for irritability. This mechanism-driven treatment is based on our pathophysiological model of irritability that postulates two underlying mechanisms, which potentiate each other: (1) heightened reactivity to frustrative nonreward, and (2) aberrant approach responses to threat. In this case report, we describe and illustrate the specific therapeutic techniques used to address severe irritability in an 11-year-old boy with a primary diagnosis of DMDD. Specific techniques within this CBT include motivational interviewing to build commitment and target oppositionality; creation of an anger hierarchy; in-session controlled, gradual exposure; and parent training focusing on contingency management to counteract the instrumental learning deficits in irritable youth. Parents learn to tolerate their own emotional responses to their youth’s irritability (e.g., parents engage in their own exposure) and increase their adaptive contingencies for their youth’s behavior (e.g., withdraw attention during unwanted behavior, praise desirable behavior). Future directions in the context of this CBT, such as leveraging technology, computational modeling, and pathophysiological targets, are discussed.

SEVERE CHILDHOOD IRRITABILITY, DEFINED as an increased proneness to anger (Brotman, Kircanski, Stringaris, Pine, & Leibenluft, 2017), predicts later-life unipolar depression, anxiety (Vidal-Ribas, Brotman, Valdivieso, Leibenluft, & Stringaris, 2016), and functional impairment (Copeland, Shanahan, Egger, Angold, & Costello, 2014). Despite its significant implications and high prevalence in both internalizing and externalizing disorders (Toohey & DiGiuseppe, 2017), chronic and severe irritability was only recently introduced as a primary diagnosis of its own: disruptive mood dysregulation disorder (DMDD). DMDD is characterized by a combination of (a) severe and age-inappropriate temper outbursts that occur at least three times a week, and (b) persistently irritable or angry mood between these outbursts (American Psychiatric Association, 2013). To meet criteria for DMDD, these symptoms must be present for at least 1 year across at least two settings (home, school, peers). Although these criteria must be fulfilled before the age of 10, DMDD cannot be diagnosed before the age of 6, as irritability shows a normative peak in preschoolers and kindergarteners (Wakschlag et al., 2015). Prevalence estimates of DMDD in school-age children and adolescents range from 0.12 to 3.00% (Althoff et al., 2016; Copeland, Angold, Costello, & Egger, 2013). Given these numbers, it is alarming that evaluated, established, efficacious, and specific psychological or pharmacological treatments specifically for DMDD do not yet exist (Stringaris, Vidal-Ribas, Brotman, & Leibenluft, 2018).

Currently, several groups are developing and evaluating behavioral interventions targeting irritability. A form of cognitive-behavioral therapy (CBT) for aggression that focuses on emotion regulation, social problem solving, and social skills, and involves both parents and teachers to support the child during anger management (Tudor, Ibrahim, Bertschinger, Piasecka, & Sukhodolsky, 2016), is currently being tested in a randomized controlled trial (RCT; Sukhodolsky et al., 2016). Further, the utility of interpersonal psychotherapy in treating irritability was confirmed in a small pilot trial (N = 19; Miller et al., 2018). In addition, positive effects of a combination of CBT (e.g., emotion recognition, problem solving, emotion regulation), parent management training (PMT; e.g., consequences and antecedents of behavior, supporting child’s attempts to regulate negative emotions), and stimulant medication on parent-reported irritability were demonstrated via an RCT with 68 patients (Waxmonsky et al., 2016). Finally, an adaptation of dialectical behavior therapy for preadolescent children (DBT-C) was successfully used to treat youth with DMDD (Perepletchikova et al., 2017). The DBT-C consisted of up to 32 weekly sessions that included individual child, PMT, and joint parent–child components, and showed higher remission rates and treatment satisfaction in youth and parents compared to treatment as usual in an RCT (N = 43).

Recently, Kircanski, Clayton, Leibenluft, and Brotman (2018) published initial feasibility data on a novel adaptation of exposure-based CBT for severe irritability as seen in DMDD conceptualized as an augmentation to any existing medication treatment. This exposure-based CBT was derived from Brotman et al.’s (2017) translational model of irritability, which posits two core mechanisms of the phenotype: (a) exaggerated responses to frustrative nonreward, and (b) aberrant approach behavior toward threatening stimuli. Both mechanisms are purportedly mediated by abnormalities in fronto-amygdalar circuitry. Studies show that high irritability is associated with decreased activation in regulatory prefrontal regions during frustration (Grabell et al., 2018; Perlman et al., 2015). Thus, we hypothesize that impaired cognitive control may contribute to (a) increased frustration in response to nonreward as reflected in symptoms of mood dysregulation, and (b) lowered threshold for temper outbursts (angry approach behavior) in response to nonreward or threat. Based on these assumptions, an intervention that increases prefrontal control of amygdala responses in the context of frustration should be particularly potent in treating irritability. We believe that an approach focusing on this specific mechanism will advance the field. Of note, increased inhibitory control of the prefrontal cortex over the amygdala seems to be one of the key mechanisms underlying exposure techniques (Craske et al., 2008). Some evidence suggests that exposure might not be successful in treating proactive aggression, which Sukhodolsky, Kassinove, and Gorman (2004) defined as planful and goaloriented aggression motivated by external reward (Dodge, 1991). However, irritability is characterized by reactive aggression (Leibenluft & Stoddard, 2013), meaning aggressive responses to others’ behavior that is perceived as threatening or intentional. Based on the current knowledge regarding the pathophysiology of irritability and mechanisms underlying exposure, Kircanski et al. (2018) proposed targeting symptoms of irritability with exposure-based CBT. Indeed, pilot data in 10 patients with DMDD demonstrate preliminary support for continued research on exposure-based CBT (Kircanski et al., 2018).

The translational model of irritability (Brotman et al., 2017) indicates that instrumental learning deficits may also play an important role in irritability. This is defined as difficulties with learning associations between one’s behaviors and their consequences in the environment. First, evidence suggests that irritable youth have difficulties learning when certain behaviors do not lead to positive consequences anymore (Adleman etal., 2011). In the home setting, for example, a child might be unable to stop a certain behavior despite repeated negative parental feedback. Whether these difficulties actually manifest in terms of irritable behavior also depends on the content of instrumental learning. In settings where problematic behaviors are consistently ignored and only positive behaviors are reinforced, aberrancies in the process of instrumental learning should be less impairing. Thus, parental contingency management might counteract deficits in the process of instrumental learning in irritable youth (Kircanski et al., 2018).

The main goal of this case report is to detail the specific mechanism-driven techniques used in exposure-based CBT for chronic, severe irritability in the context of DMDD. We focus on interventions targeting both neurobiological (e.g., heightened response to frustrative nonreward, abnormal approach toward threat) and behavioral (e.g., parental contingency management) factors that play a role in the maintenance of DMDD.

Presenting Problem

“Ethan” was a 11-year-old boy who was referred to us by his psychiatrist. Ethan and his family reported daily intense and disproportionate irritable reactions to a range of situations (e.g., yelling, hitting his parents and on occasion siblings and peers, slamming doors, calling himself and others names, and breaking things). These outbursts typically lasted for 30 minutes but could persist for up to 2 hours. After each outburst, Ethan felt significant remorse, often cried and apologized to his parents. Triggers included being asked to pause the use of electronic devices, to do his homework and household chores, and disagreements with siblings and peers. Ethan’s parents noted that he was “cranky” most of the time, leading the family to act very cautiously around him. At school, Ethan often refused to comply with instructions and was physically aggressive toward his peers, despite accommodations, such as a “flash pass” to exit the class and see the counselor when he became upset. Recently, Ethan was asked to switch soccer teams due to the frequency of his outbursts during practices.

Case History

Ethan lived with his mother (41 years), father (43 years), 13-year-old sister, and 8-year-old brother in a middle-class suburb of Washington, DC. Both of Ethan’s siblings performed well in school with no accommodations or behavioral problems. The father reported a personal history of attention-deficit/hyperactivity disorder (ADHD). For the other family members, no history of psychiatric illness was reported.

Ethan’s parents first had concerns about fussiness and difficulty being soothed in infancy. Ethan was assessed at 17 months by the county and qualified for speech and occupational therapy due to a variety of delays (slow to babble and to develop single words and phrase speech, sensory issues, weak core, and poor muscle coordination). He has been diagnosed with an expressive language disorder by a speech pathologist. Although he had been receiving speech therapy since age 2, mild symptoms (e.g., difficulties finding the right words or cutting to the point) were present throughout the therapy and might have negatively contributed to his irritability by increasing the frequency of frustrating situations. He received occupational therapy from ages 2 to 9 and qualified for an individualized education plan (IEP) beginning in preschool due to his developmental delays.

Ethan was diagnosed with ADHD at age 7 by a psychiatrist. Since then, he has taken stimulants, which improved the symptomatology. Yet, his mother endorsed ongoing mild symptoms of inattention, hyperactivity, difficulties with organization, and impulsivity, for which he was receiving some accommodations in school, such as extended time on tests and assignments.

Ethan’s parents recalled that he had been irritable since preschool, showing oppositionality toward family members and teachers. His irritable mood and outbursts remained stable throughout his life. At home, the family was very accommodating to Ethan’s irritability. His parents commented that the family typically went on camping trips several times a year, but had to stop these outings due to Ethan’s severe outbursts during long car rides. Although Ethan maintained a small core group of friends over time, his irritability, need for control, and rigidity impeded his ability to get along with his larger peer group. His parents recalled that classmates declined to schedule playdates with Ethan because he was “generally grumpy and may get mad while we play if things don’t go his way.” Ethan had never received psychotherapy for his irritability but was highly motivated to learn how to “control his anger,” as he wanted to be reinstated as a member of his soccer team, have friends, and have “things run more smoothly” at home.

Assessment

diagnostic interview

A clinical psychologist conducted the Kiddie Version of the Schedule for Affective Disorders and Schizophrenia—Present and Lifetime Version (K-SADS-PL; Kaufman et al., 1997) and the DMDD module (Wiggins et al., 2016) with Ethan and his mother. Ethan met the fifth edition of Diagnostic and Statistical Manual of Mental Disorders (DSM-5; American Psychiatric Association, 2013) criteria for DMDD and ADHD-combined type, although symptoms were only minimally impairing under his daily medication (8 mg of atomoxetine, 1.5 mg of methylphenidate). Ethan had no history of anxiety, depression, hypomania, suicidality, psychosis, or any other DSM-5 diagnoses.

TARGET SYMPTOMS

The primary target of the intervention were symptoms of irritability that were assessed pre-, mid-, and posttreatment with the modified Clinical Global Impressions—Improvement (CGI-I) scale (Dickstein et al., 2009) and the Clinician Affective Reactivity Index (CL-ARI; Haller et al., submitted for publication). The scores of the modified CGI-I range from 1 to 8, with lower scores reflecting greater levels of improvement: 1 (completely recovered), 2 (much improved), 3 (improved), 4 (slightly improved), 5 (unchanged), 6 (slightly worse), 7 (worse), and 8 (much worse). The CL-ARI is a semistructured interview designed to be administered by a trained clinician to parents and children separately. Its 12 items query the frequency, duration, and severity of mild, moderate, and severe temper outbursts and irritable mood between outbursts, as well as functional impairment in home, school, and peer settings. Parent and child ratings are synthesized by the clinician into a consensus rating. The total score of CL-ARI ranges between 0 and 100. The clinicians rating Ethan’s symptoms were not blinded, which might have biased the ratings.

Ethan and his father rated symptoms of irritability weekly using the self- and parent-report ARI, respectively (Stringaris et al., 2012). The overall ARI score range is 0–12. Higher scores reflect higher levels of irritability. The ARI has good to excellent internal consistency (α = .88–.92) and tes–-retest reliability (.88–.90; Tseng et al., 2017).

intellectual functioning

Per study protocol, Ethan completed the Wechsler Abbreviated Scale of Intelligence (Wechsler, 1999). This test indicated a Full-Scale IQ of 118. This score implies that Ethan’s intelligence fell slightly above the average range of functioning.

Ratings of Severity, Depression, and Anxiety

The Children’s Global Assessment Scale (CGAS; Shaffer et al., 1983) is a clinician-rated measure of global functioning, which scores range from 1 to 100 and shows good interrater reliability (intraclass correlation coefficient [ICC] = 0.88; Bird, Canino, Rubio-Stipec, & Ribera, 1987). Higher CGAS scores reflect better overall functioning. Ethan’s score on the CGAS pretreatment was 41, reflecting the substantial impairment resulting primarily from his irritability. Unfortunately, we were unable to obtain a posttreatment CGAS rating for Ethan.

Depressive symptoms were measured with the Children’s Depression Rating Scale (CDRS; Poznanski, Cook, & Carroll, 1979), showing good internal consistency (.79–.92; Mayes, Bernstein, Haley, Kennard, & Emslie, 2010).Prior to treatment, Ethan’s CDRS score was 30, suggesting the absence of clinically significant depressive symptoms (cutoff ≥ 40).

The Pediatric Anxiety Rating Scale (PARS), which comprises 50 items that are scored by a clinician as present or absent based on parent and child report, measures anxiety symptoms and related functional impairment in youth as continuous outcome (Research Units on Pediatric Psychopharmacology Anxiety Study Group, 2002). The sum score is calculated based on subscales 2 (frequency of symptoms), 3 (severity of distress associated with anxiety symptoms), 5 (avoidance), 6 (interference at home), and 7 (interference outside of home) and ranges from 0 to 25 with higher scores reflecting greater levels of anxiety. The PARS shows satisfactory internal consistency (ICC = 0.97) and moderate test–retest reliability (.55; Research Units on Pediatric Psychopharmacology Anxiety Study Group, 2002). Before treatment Ethan had a score of 15, indicating the presence of clinically relevant anxiety levels (cutoff = 11.5). Ethan’s PARS score was reduced to a subthreshold score of 8 mid- and posttreatment.

Case Conceptualization

Based on research evidence regarding core mechanisms underlying and maintaining DMDD, a cognitive-behavioral conceptualization was developed (see Figure 1). We began with an initial maintenance formulation. Further inquiry showed that Ethan’s outbursts were triggered when (a) goal attainment was blocked (e.g., he was asked to stop a favorite activity), (b) he was confronted with difficult or boring tasks (e.g., folding laundry), and (c) he perceived others as threats (e.g., sports competition). These events triggered an intense neurophysiological response (e.g., heightened heart rate, breathing, muscle tension), which impaired his judgment. For example, he experienced frequent negative thoughts: “This is unfair,” “XY hates me,” “I always get the short end of the stick,” or “Everyone is against me.” These thoughts led to feelings of frustration, anger, and disappointment, which in turn motivated aggressive behavior against objects, peers, and adults (e.g., yelling, hitting, throwing, and kicking). His aggressive behaviors were inconsistently reinforced. For example, asking Ethan to put his tablet away regularly triggered reactions such as screaming, hitting objects, and looking menacingly at his parents. Sometimes his parents would wait, while Ethan continued using the tablet (positive reinforcement). On other occasions, they might tell him that if he would not comply, he would not get to use the tablet the next day (negative punishment) or wrestle the tablet from him (negative punishment). Long term, the daily outbursts led to several accommodations (e.g., “flash” pass at school, siblings were not allowed to enter his room or use his toys) and increased the family’s stress level (e.g., cancelation of pleasant family activities, such as vacations, eating out, visiting friends). In terms of protective factors, Ethan had a supportive family and a small core group of friends. In addition, he was intelligent, achieving well academically, and enjoyed soccer and drawing.

FIGURE 1.

A diagram of Ethan’s formulation of DMDD within a cognitive-behavioral framework. Gray coloring denotes relevant neurobiological mechanisms for irritability.

Over the course of therapy, we developed a longitudinal conceptualization. Harsh parenting—defined as impatience, anger, and frustration with the child, including verbal and physical aggression—has been associated with externalizing behavior (Wiggins, Mitchell, Hyde, & Monk, 2015) and increased irritability (Oliver, 2015). Ethan’s father’s own ADHD symptoms contributed to his failure to graduate high school, which led to several hardships that he wanted to protect his son from. Driven by his worries, Ethan’s father became easily impatient and frustrated when Ethan did not function like a “reasonable human being.” He would then utter harsh critiques, yell, and occasionally snatch Ethan’s toys away. Ethan’s mother, in an attempt to reconcile father and son, exhibited an inconsistent parenting style. She switched irregularly between a harsh and an accommodating style. Thus, we hypothesized that harsh and inconsistent parenting styles served to maintain Ethan’s symptoms of irritability.

Collaboratively Identified Treatment Targets

It was postulated that Ethan’s irritability was being maintained through his enhanced neurobiological response to frustration and threat, combined with negative and inconsistent parenting. Ethan’s goal for therapy was to train his brain to better cope with his anger when asked to put his tablet away, turn off the TV or the computer, complete difficult or boring tasks, and do household chores. His parents were motivated to change their parenting styles to be more consistent and foster more positive interactions between Ethan and themselves.

TREATMENT

Ethan was seen for 12 weekly CBT sessions by a licensed clinical psychologist. Every session comprised 30–45 minutes with Ethan and his parents separately; some sessions also included a joint parent–child component. The treatment started with a joint session and comprised psychoeducation on irritability and setting of treatment goals. Throughout the therapy, the work with Ethan focused on exposure and included motivational interviewing to address ambivalence regarding change, generation of an anger hierarchy (see Figure 2), and gradual in session exposure. Initial sessions with both parents comprised psychoeducation on instrumental learning in parent–child relationships. The parents were coached to praise and acknowledge Ethan’s adaptive behaviors, actively ignore Ethan’s maladaptive behaviors, give commands, set limits, increase time spent on positive joint activities, and deliver intermittent rewards. In the later sessions, the therapist worked particularly with Ethan’s father to tolerate his emotional distress in response to Ethan’s irritability symptoms. In between sessions, Ethan was asked to practice tolerating his anger through exposure and his parents were asked to support Ethan employing techniques and strategies they had learned in session. The following section describes the two core treatment elements: exposure and PMT. For an overview of the sessions see Table 1.

FIGURE 2.

The anger hierarchy developed by Ethan.

Table 1.

An Overview of the Interventions Delivered to Ethan and His Parents

Session	Child	Parent
1–2	Psychoeducation; motivational interviewing; assessment of parental psychopathology and family history; assessment of levels of irritability in different situations; anger hierarchy
3	Breathing retraining; rationale for exposure; imaginal exposure	Praise/acknowledgment (psychoeducation, role plays)
4	In-session exposure to a frustrating situation (folding clean laundry) while monitoring and tolerating irritability levels	Active ignore (psychoeducation, role plays)
5	In-session exposure to a frustrating situation (folding clean laundry) while monitoring and tolerating irritability levels	Dealing with outbursts (psychoeducation, role plays)
6	In-session exposure to a frustrating situation (Ethan having the tablet taken away while he is playing a game on it) while monitoring and tolerating irritability levels	Commands (psychoeducation, role plays)
7	In-session exposure to a frustrating situation (Ethan having the tablet taken away by therapist while he is playing a game on it) while monitoring and tolerating irritability levels; father is watching, tolerating his own discomfort and practices praise/acknowledgment and active ignore
		Setting limits (psychoeducation, role plays)
8	In-session exposure to a frustrating situation (Ethan having the tablet taken away by his father while he is playing a game on it) while monitoring and tolerating irritability levels; father is watching, tolerating his own discomfort and practices commands, praise/acknowledgment, and active ignore
		Positive interactions and rewards (psychoeducation, role plays)
9–11	In-session exposure to a frustrating situation (Ethan having the tablet taken away and switching to the nonpreferred activity of folding laundry or continuing to sit with toys and games in front of him) while monitoring and tolerating irritability levels; father is tolerating his own discomfort and practices commands, praise/acknowledgment, and active ignore
11–12	Termination, including discharge planning and generation of an “accomplishments and goals drawing”

Note. The different treatment modules are shown in italic to enhance clarity. Out-of sessions exposures were assigned as homework from Session 4 onward.

Psychoeducation and Therapeutic Alliance

Developing a consistent language was crucial at the start of therapy. The therapist first defined irritability and anger and learned the vocabulary that Ethan and his family were using to describe Ethan’s symptoms (e.g., being grumpy, exploding). Further, joining with the parents and establishing a positive relationship was key. This was achieved by recognizing and validating the challenges and the stress of raising a severely irritable child. Further, the therapist demonstrated interest in the “parents’ side,” in assessing parental mood, psychiatric history, treatment, cognitive function, organizational abilities, current and upcoming stressors, the couple’s functioning, and the consequences of Ethan’s behavior on his siblings. Psychoeducation focused on the prevalence of irritability symptoms; normalizing anger in the context of the evolutionary utility of the fight-or-flight reaction; reviewing somatic, cognitive, and behavioral responses associated with anger; and discussing adaptive (“okay”) and maladaptive (“not okay”) ways to show anger.

Exposure

The central piece of the first two sessions with Ethan consisted of motivational interviewing techniques and in vivo exposures (e.g., drawing his anger, drawing a frustrating situation as a movie clip to reexperience the situation with different outcomes, and describing his experience when asked to do undesirable household chores, such as laundry). To fully leverage these exercises, the therapist introduced two tools: (a) the “readiness ruler,” a 0–10 scale to elicit motivational change talk (e.g., when readiness for change was rated with a 3, the therapist would ask Ethan why he was not a 0 and what it would take for him to get to a 4), and (b) “temperature ratings,” a color-coded scale ranging from 1 to 10, which was used to measure Ethan’s anger in vivo (see Figure 2). Further, the therapist introduced that breathing affects emotions and that “a normal size breath in, followed by a very slow exhale” is relaxing as an immediate coping skill, which was practiced intensively. Of note, the introduction of additional skills, such as muscle relaxation, might be indicated for other cases depending on the individual. Finally, the therapist explained the rationale of the graded exposure hierarchy using an analogy of training for a sport: a weight lifter starts training with a smaller weight and builds up muscle to be able to handle a heavier weight. Similarly, they were going to build “Ethan’s brain muscle” to tolerate anger.

In the third session, a collaborative anger hierarchy was developed (see Figure 2). As Ethan felt less inclined to talk about these situations, he drew them. Over the next week, Ethan recorded situations in which he felt angry and rated his feelings with the thermometer. Gradual exposures began with the mildly frustrating situation of putting his dirty laundry in the laundry basket, which he rated with a 4. Sessions 4–12 focused on exposure tasks (see Table 1). The successful completion of the in-session exposures (see Figure 3A) showed Ethan that his “brain muscle was growing stronger” and strengthened the confidence in his ability to control his anger, which allowed him to start exposure outside of the sessions. This is particularly important given that compliance with homework is associated with improved treatment outcomes (Mausbach, Moore, Roesch, Cardenas, & Patterson, 2010). Ethan’s parents were actively involved in his out-of-session exposures through eliciting frustration (e.g., asking Ethan to do the laundry, put the tablet away, turn off the computer), and subsequently actively ignoring his irritable behaviors and praising (attempts to show) appropriate behaviors.

FIGURE 3.

Illustration of anger ratings during exposure and scores on the Affective Reactivity Index (ARI) A. Shows anger ratings of six exposure sessions, where the tablet was taken away. Sessions typically lasted between 5 and 25 minutes. B. Displays ARI scores across the 12 sessions. Note. Children with DMDD are characterized by ratings ˃3 for the parent report and ˃ 2 for the child report (Kircanski et al., 2017).

During exposure, Ethan was not allowed to engage in alternative rewarding or distracting activities as these would have decreased the exposure to frustration. Throughout all of the sessions, the therapist provided generous verbal praise as positive reinforcement for Ethan’s attempts to show appropriate behaviors regardless of the actual outcome. Further, it was essential that the therapist created a calm atmosphere and modeled the goal of not avoiding intense emotions, while at the same time not escalating the temper outburst. This was achieved by maintaining contact with the child in a nonthreatening, noncontrolling manner.

Challenging Situations While Working With Ethan

There were two types of challenging situations that occurred during the exposure sessions: (a) Ethan experiencing a temper outburst during the first exposure, where he had to stop a game on his tablet, and (b) Ethan not wanting to start the following two sessions. During Ethan’s outburst, it was key that the therapist was not scared by this intense reaction and that she conveyed through her composure and demeanor that the session was a safe place for Ethan to experience these intense emotions. This included talking slowly in a quiet tone, engaging in eye contact, increasing the use of nonverbal communication, and sitting down on the floor with Ethan. The therapist remained calm and refocused Ethan’s attention (“I see you are exploding. We are going to monitor your feelings now. I am going to ask you for your temperature rating every 5 minutes. Until then, I will work on my computer”) without addressing the reason or the content of his anger. Further, Ethan’s cooperation in monitoring his anger and providing ratings throughout his outburst was positively reinforced. When Ethan did not want to start the following session, the therapist engaged him in a conversation about sports, which she then steered toward his soccer team. This led Ethan to discuss why he initially wanted to work on his anger (e.g., being reinstated as a member of his soccer team, having friends), thereby motivating him to start the session.

Parent Management Training

PMT, as used here, is based on principles of instrumental learning and parent–child interaction (Patterson, 1992). Consistent with PMTs described for oppositional behavior and aggression in youth (Kazdin, 2010; Sukhodolsky & Scahill, 2012), the main goal was to facilitate parental use of consistent positive reinforcement (e.g., selective attention, praise) for adaptive child behaviors, and consistent nonreinforcement (e.g., active ignoring) or mild negative consequences (e.g., time-out) for maladaptive child behaviors.

Ethan’s parents met with the therapist alone after Sessions 2–9 for 30–45 minutes, while a research assistant engaged in play activities with Ethan in a separate room. Ethan’s father was involved in all sessions, while Ethan’s mom attended sessions 1–6 only. An overview of the different techniques addressed during sessions with Ethan’s parents is provided in Table 1.

The PMT started with psychoeducation covering instrumental learning and active-ignore parenting techniques for Ethan’s low-level, nondangerous irritable behaviors (e.g., talking back, drumming with his fingers on furniture, fidgeting, fumbling with his hair). After active ignore was established for the low-level irritable behaviors, it was introduced as a response to Ethan’s outbursts. The therapist worked with Ethan’s parents on monitoring their own level of anxiety and anger during Ethan’s outbursts and on tolerating these emotions. It was conveyed that during an outburst Ethan’s parents must “pull back” and should not engage in discussions about the specific issue. Next, the therapist introduced commands to shape Ethan’s behavior. In role plays, parents practiced how to let Ethan know that they wanted him to change his behavior in 5 minutes (e.g., making eye contact, coming close to Ethan, neutral tone, not phrasing the command as a question). Once skills such as recognizing and praising appropriate behaviors and actively ignoring disruptive behaviors were established, discussions focused on how to set limits (e.g., regarding screen time, interactions with siblings, chores). It was emphasized that consistency is key, and Ethan’s parents practiced during role plays how to set limits empathetically and to resist the temptation to be punitive. In line with previous reports linking increases in pleasant and positive, nondirected parent–child interactions with lower externalizing behaviors (Eisenberg et al., 2005), Ethan’s parents committed to setting aside 5–10 minutes daily to attend positively to some of Ethan’s regular activities (e.g., playing the piano, drawing, practicing soccer moves). Last, periodic positive consequences for appropriate behaviors were introduced, which should ideally be enjoyable, brief, and inexpensive activities (e.g., cuddling, playing a favorite song very loud, going to the park, riding the bike around the block). Starting at Session 9, these parenting skills were practiced, but no new content was introduced. During the last two sessions, it was discussed how to best support Ethan in his continued practice of breathing skills and anger tolerance in other frustration-eliciting situations.

Complicating Factors in Working With Ethan’s Parents

Two challenging aspects in working with Ethan’s parents involved their expectations regarding Ethan’s behavior and factors associated with his father’s own psychopathology. With regard to the expectations, it was very difficult for Ethan’s parents to praise any of Ethan’s behaviors that were not “perfect.” To overcome this complicating factor, the therapist worked intensely with Ethan’s father on his cognitions regarding failure and learning and practiced the verbalization of praise during role-plays.

Further, both parents showed a high need for control and therefore experienced high levels of discomfort during Ethan’s irritable behaviors. His father worried that Ethan would never be successful in life if he continued fidgeting, talking back, and having outbursts. While Ethan was working hard on tolerating and monitoring his anger during exposure, his father became visibly agitated and relentlessly corrected and criticized Ethan’s behaviors (e.g., “Look at the therapist,” “Stop fidgeting,” “The therapist just asked you a question. You need to answer”). In a first step, the therapist verbalized and validated the discomfort of Ethan’s father. Next, a combination of psychoeducation and cognitive restructuring techniques was used to address the dysfunctional assumptions of Ethan’s father (e.g., Socratic dialogue about what success in life actually means, pie diagram to illustrate determinants of success in life, reframing of his interventions from helpful as intended to preventing Ethan from developing appropriate behaviors/skills to cope with anger by himself). Further, the father’s exposure to Ethan’s irritability while monitoring and tolerating his own negative feelings became an integral element of the therapy.

Treatment Outcome

At the beginning of treatment, Ethan’s DMDD symptoms significantly interfered with his everyday functioning, including his family and peer relationships and his schoolwork. At the end of treatment, Ethan continued to meet criteria for DMDD, but at a milder level of clinical severity than at the start of treatment. He showed less irritable mood between outbursts. Ethan’s overall level of impairment was moderate at the end of treatment.

Ethan’s scores on the clinician ARI and CGI indicate an improvement throughout therapy (see Table 2 and Figure 3B). This improvement was also evident clinically, including positive outcomes on his goals: to be able to handle his laundry without temper outbursts, and to be able to tolerate his anger when his parents interfere with his use of electronic devices.

Table 2.

Summary of Scores on Assessment Measures Completed Across Therapy

	Pretreatment	Midtreatment	Posttreatment
CGI-S a
Overall	5	4	4
Mood	4	4	4
Outbursts	5	4	4
CGI-I a
Overall	-	4	4
Mood	-	5	4
Outbursts	-	5	5
Clinician ARI
Impairment b	6	3	3
Mood c	10	9	9
Outbursts d	14	12	13
Total e	81	58	61

Note. CGI-S = Clinical Global Impressions Severity; CGI-I = Clinical Global Impressions Improvement; ARI = Affective Reactivity Index.

^aIn this study, we used a modified CGI with the following scores: 1 = completely recovered, 2 = much improved, 3 = improved, 4 = slightly improved, 5 = unchanged, 6 = slightly worse, 7 = worse, 8 = much worse.

^bImpairment is rated in the family, school, and with peers on a scale from 0 to 2. Thus, the scale ranges from 0 to 6.

^cMood is measured with three items (0–4) assessing frequency, severity, and duration of cranky, grouchy mood. Thus, the scale ranges from 0 to 12.

^dFor mild, moderate, and severe outbursts, duration and frequency were assessed on a scale from 0 to 4. Thus, outburst scores could range between 0 and 24.

^eThe total score represents the mean of the percentages of the Impairment, Mood, and Outburst Scale: Total = ([∑ Impairment / 6] * 100 + [∑ Mood / 12] * 100 + [∑ Outbursts / 24] * 100).

REFLECTIONS, THEORY, AND LIMITATIONS

This case report illustrates how symptoms of irritability can be treated with 12 sessions of exposure-based CBT. The case of Ethan, an 11-year-old boy, highlights the transdiagnostic utility of exposure techniques. In the following, we highlight some important aspects regarding safety, implementation of rewards, and familial factors as potentially complicating factors.

Safety

In our view, this treatment is not appropriate for children who have instrumental aggression and/or high levels of callous/unemotional traits. Instead, this treatment targets reactive, disruptive behaviors. However, if the child’s irritability manifests in terms of unusually long and intense outbursts or tends to take a dangerously aggressive form, this approach is not appropriate. Therefore, prior to treatment, we found it helpful to determine how Ethan’s typical outbursts manifest and, in addition, how the most severe outbursts have presented in the recent past. Ethan’s family reported no evidence of behaviors causing serious injury to others or physically intensive outbursts, which represent counterindications for this exposure-based treatment.

Ambivalence and Oppositionality

Temper outbursts can be shaped by instrumental learning when they become very efficient ways to exert control over interaction partners and obtain desired benefits. Thus, despite their negative long-term effects, outbursts also have beneficial short-term effects and so ambivalence regarding a change in this behavior may be expected. We found it crucial to resolve this ambivalence without provoking oppositionality. Therefore, we used motivational interviewing techniques (Miller & Rollnick, 1991) throughout the therapy. This empathic, supportive, yet directive counseling style is ideally suited to establish a collaborative alliance between child and therapist, engaging the child’s intrinsic motivations and values to actually do the exposures and change his or her behavior.

The Therapeutic Use of Rewards

Given the instrumental learning deficit in DMDD (Adleman et al., 2011), we purposefully refrained from implementing reward programs or token economies. In working with irritable youth, we found that some youth tend to focus on the rewards per se and exhibit difficulties disengaging attention toward the therapeutic task needed to achieve the rewards. Further, the occurrence of a temper outburst can be tied to triggering stimuli characterized as frustrative nonreward, which is defined as resulting from instrumental conditioning when a reward expectancy is violated (Amsel, 1958). Thus, irritable youth might show severe irritable reactions when they fail to meet the condition for the reward. With regard to the unexpected rewards, which we introduce last during the parent sessions, we emphasize that these might trigger outbursts in very rigid children who easily become upset where there is any deviation from the expected. It is important for parents to understand that the reward must be given after the behavior (e.g., once the child successfully finished doing his or her homework independently). Sometimes parents try to motivate their children to continue a behavior by “dangling” a reward. However, in our experience, severely irritable children tend to focus on this reward instead on the behavior and therefore might not finish the behavior with the consequence of no reward, which, however, will likely lead to an outburst. In this regard, we also emphasize that parents should avoid “cumulative” rewards over a series of behaviors (“If you can do that again tomorrow, we’ll go for a treat”).

Parental Psychopathology

As in Ethan’s case, parental psychopathology might interfere with the delivery of parent training. Parental ADHD symptoms have been previously related to overreactive (Chen & Johnston, 2007) and inconsistent parenting (Chronis-Tuscano et al., 2008), less positive reinforcement (Chen & Johnston, 2007), and more negative commands and critical statements (Chronis-Tuscano et al., 2008). This disrupted parenting might mediate the association between parental ADHD symptoms and disruptive behaviors (Breaux, Brown, & Harvey, 2017). In Ethan’s case, a driving factor that motivated overreactivity was the father’s thought of being responsible for Ethan’s symptoms, accompanied by feelings of guilt and the desire to “fix the damage.” Thus, it was important to work on these thoughts as they were related to negative feelings, which impaired the parental ability to support Ethan during the exposure.

Family Processes as Complicating Factors

We assessed parental expectations of the therapy process prior to treatment (McMahon & Forhand, 2001).Ethan’s parents agreed with the philosophy of parent training and were willing to work on their parenting efficacy. However, some parents may become surprised or defensive upon learning they will need to modify their behavior. Parents might also have unrealistic expectations about the developmental appropriateness of their children’s behavior or, similar to Ethan’s parents, will be very focused on the negative interactions with their child. This is expected and should be explicitly addressed through psychoeducation, increasing awareness for the child’s positive behaviors and fostering positive nondirected parent–child interactions.

Marital difficulties were not evident in Ethan’s case, but might complicate the implementation of the PMT. Parents engaging in high levels of conflict may be less likely to provide positive reinforcement or might be less consistent in their parenting (McMahon & Forhand, 2001). Moreover, higher levels of marital conflict at the start of parent training have been shown to interfere with maintenance of treatment gains over time (Dadds, Schwartz, & Sanders, 1987). Thus, marital conflict might be another complicating factor in treating youth irritability, which should be assessed and acknowledged during treatment.

Limitations

Some limitations should be considered. First, due to the design of the intervention it is not clear which specific techniques generated improvement in irritability. Alternative mechanisms of change include general factors, such as the alliance between Ethan and his therapist, which is a consistent predictor of outcome (Wampold, 2015). In the ongoing multiple baseline trial, we will additionally measure the therapeutic allegiance and adherence to the manual. Second, the treatment was conducted in the context of a study at a research facility, which entailed free service and monetary compensation. If Ethan’s family had to shoulder the treatment costs by themselves, this might have limited their ability to complete the treatment.

Clinical and Research Implications

DISRUPTIVE MOOD DYSREGULATION DISORDER AND ADHD

Ethan’s diagnostic assessment resulting in comorbid DMDD and ADHD is very characteristic, as these diagnoses are highly comorbid with a 3-month DMDD prevalence of 31% in ADHD youth and at least one symptom of irritability in almost every ADHD child (Eyre et al., 2017). Irritability may be an early marker of mood problems in children with ADHD. However, longitudinal studies are necessary to test this hypothesis. Further, little is known regarding neural and behavioral similarities and differences across DMDD and ADHD. Recent studies indicate reduced gray matter volume in the prefrontal cortex in DMDD compared to HV and ADHD (Gold et al., 2016), but similar blunting of prefrontal activity during an attention task in both disorders (Pagliaccio et al., 2017). More studies parsing overlapping and distinct neural mechanisms of ADHD and irritability are needed to refine etiological models and inform mechanism-based treatments. Future studies should also investigate the validity of the translational model of irritability in the context of other DSM-5 disorders that list irritability among the symptoms (Toohey & DiGiuseppe, 2017).

Active Components of the Exposure-Based CBT

The efficacy of the exposure-based CBT is currently being investigated in a multiple baseline trial that employs functional magnetic resonance imaging (fMRI) during tasks probing core mechanisms of DMDD (e.g., frustrative nonreward, threat processing, reward learning) pre- and posttreatment. On a neural level, positive effects of exposure in anxiety are thought to be mediated by increased inhibitory control of the prefrontal cortex over the amygdala (Craske et al., 2008). We theorize that the downstream effects of such increased control also extend to other emotional reactions, such as exaggerated frustration. However, while in anxiety, the effects of exposure are understood within the framework of Pavlovian conditioning; for irritability, instrumental learning (Amsel, 1958) might be the more relevant process. Thus, we propose that in irritability, exposure exerts its effects through continued learning of stimulus–reinforcer contingencies, which will reduce the intensity of the emotional response allowing for alternative nonaggressive behaviors. While this may occur in parallel to habituation, reduction in frustration/anger is not obligated to accompany strengthening of prefrontal inhibitory control. Future studies should focus on mediators and moderators of the response to exposure in DMDD.

Further, parenting techniques are an integral module not only within this exposure-based CBT but also in other behavioral treatments proposed for DMDD (Perepletchikova et al., 2017; Sukhodolsky & Scahill, 2012; Waxmonsky et al., 2016). All parent training modules within the behavioral treatments currently proposed for DMDD focus on instrumental learning and particularly the reinforcement of positive, adaptive behaviors, limit setting, and realistic expectations regarding the child’s behavior. The focus on parental discomfort seems unique to the DBT (Perepletchikova et al., 2017) and exposure-based CBT (Kircanski et al., 2018) model, which, however, emphasize different techniques to address the parental discomfort (skills training vs. in vivo exposure, respectively). Parent management modules also differ with regard to length and support of the child’s attempts to regulate negative emotions. Future dismantling studies could help identifying the active ingredients of these complex interventions.

Parent–Child Interactions

The current pathophysiological model of irritability emphasizes the relevance of environmental factors in the emergence and maintenance of irritability (Brotman et al., 2017). Irritable children experience environments where rewards and punishments are inconsistently delivered, leading parents to unintentionally reinforce (and maintain) the child’s disruptive behavior. A large body of literature links inconsistent and negative parenting to disruptive behaviors (Paulussen-Hoogeboom, Stams, Hermanns, Peetsma, & van den Wittenboer, 2008). Another well-studied association involves parental psychopathology and increased levels of inconsistent and negative parenting (Chen & Johnston, 2007; Chronis-Tuscano et al., 2008). However, relatively little is known about the neurobiology of parenting, particularly beyond infancy, when rodent models are not feasible. Negative parenting during childhood and adolescence might reflect low levels of self-regulation, instrumental learning deficits, and exaggerated responses to threat and frustrative nonreward. Future studies may seek to tailor PMT by investigating the associations of these processes with parenting styles and characteristics of parent–child interactions.

Summary

In this case study, exposure therapy was applied to a preadolescent boy with the primary diagnosis of DMDD. This report highlights the benefits of exposure in the context of severe irritability. Further, important questions for clinicians working with DMDD youth were raised. While different areas for future research were identified, it seems foremost to test the efficacy of the exposure-based therapy. This is currently being investigated in a multiple baseline trial focusing on a more detailed assessment of the therapeutic process and irritability in a naturalistic setting via ecological momentary assessment, as well as neurobiological mechanisms of change.