Two Cases of Early Childhood Selective Mutism

1 Theoretical and Research Basis for Treatment

Selective mutism (SM) is a condition in which children fail to speak in situations when speaking is expected, despite having the ability to speak. SM affects approximately 0.7% of children and is most often recognized when children enter school (Chavira, Bailey, Stein, & Stein, 2004). There is evidence that children develop symptoms of SM between 2.7 and 4.1 years of age; however, they may not experience functional impairment prior to the social demands of preschool or elementary school (Schwartz, Freedy, & Sheridan, 2006). SM has been related to significant social skills deficits, early academic skills deficits, and poor peer relations (Cunningham, McHolm, Boyle, & Patel, 2004).

Although the etiology for SM is unknown, operant factors likely contribute to the establishment and maintenance of nonspeaking behaviors (e.g., Nolan & Pence, 1970). SM has been related to social anxiety disorder, but may be unique in that children with SM display more mild oppositional behaviors than children with social anxiety disorder, but rarely meet criteria for oppositional defiant disorder (Black & Uhde, 1992; Yeganeh, Beidel, & Turner, 2006). Still there is evidence that children with SM experience less physiological arousal in social situations than children with social anxiety disorder, suggesting that their lack of speech serves an avoidant function during social interactions (Young, Bunnell, & Beidel, 2012). Based on these and similar findings, Diagnostic and Statistical Manual of Mental Disorders (DSM-V; currently under development) may consider SM to be a subtype of social anxiety disorder (Bögels et al., 2010; Rao, 2011).

Given the early age of onset in most SM cases, there is some evidence that caregiver behavior may play a role in the development and maintenance of SM (Yeganeh et al., 2006). Parents may allow escape or avoidance of speaking in anxiety provoking situations (e.g., speaking for children), increasing the likelihood of a child not speaking via negative reinforcement. Nonspeaking behaviors may also be inadvertently positively reinforced (e.g., comforting the child who does not speak). For these reasons, parenting may play an important role in helping children with SM and may be targeted in interventions.

Despite the impact of SM on young children’s development, there has been a surprising lag in the establishment of effective treatments. Building on evidence from other anxiety disorders predominantly employing exposure to feared stimuli (e.g., Beidel, Turner, & Morris, 2000; Gillian & Rachman, 1974), there is preliminary evidence that exposure treatment may be effective for young children with SM. Previous case studies have demonstrated child improvements in SM using modular interventions that included exposure (Reuther, Davis, Moree, & Matson, 2011; Vecchio & Kearney, 2009), similar to the treatment of social anxiety disorder. Oerbeck, Johansen, Lundahl, and Kristensen (2011) utilized graduated exposures in home and school environments in combination with reinforcement to encourage an increase in and generalization of speaking behaviors in children ages 3 to 5. Alternatively pharmacological interventions, most commonly antidepressant medications, have not shown clinically significant reductions in symptoms (Black & Uhde, 1994; Wong, 2010).

Building on previous literature, the treatment of the presented cases was selected based on the efficacy of behavioral interventions for anxiety disorders in children and based on preliminary evidence that behavioral interventions may be efficacious in children with SM (Cohan, Chavira, & Stein, 2006; Davis, May, & Whiting, 2011). Key components of behavioral treatments for anxiety disorders in children include exposure treatment, contingency management, and psychoeducation for parents and children. The following cases employed parents and utilized exposures in multiple contexts to treat SM in early childhood.

2 Case Introduction

This article presents two cases of SM with the aim of demonstrating treatment complexities and flexible implementation of behavioral techniques with young children and their families. Brian and Zoe (pseudonyms) are two 5-year-old European American children who presented with their parents to a university hospital–based anxiety specialty clinic. At the time of assessment, neither child was speaking outside of the home.

3 Presenting Complaints

Brian—At the time Brian presented for treatment, Brian’s teachers reported that he had not spoken to peers or his teacher since he entered the classroom approximately 3 months earlier. Brian’s mother reported that she would occasionally volunteer in his classroom and Brian would whisper to her infrequently but would not speak to anyone else. Brian’s mother stopped volunteering after noticing Brian became very “clingy” and wanting to leave with her when she was finished volunteering. Brian’s parents recalled that prior to enrollment in preschool he had limited interactions with others in the community. They noted that Brian had engaged in parallel play with children from his neighborhood or play group. In contrast, at home Brian was reportedly a “chatterbox” and had no difficulty interacting or speaking with family members.

Brian had not received previous treatment for SM, although he had been evaluated for speech difficulties and was found to be delayed in speech articulation and production. Brian had received limited speech remediation and early intervention services through the school system for several months and parents reported dramatic improvements, although he continued to refuse speaking in school and community settings. Brian’s parents accommodated his lack of speech by allowing him to whisper in their ears when out in public and then speaking for him.

Brian’s parents’ primary concerns for Brain was that he had difficulty making friends and engaging in community activities such as intramural sports or summer camps due to not speaking. They were also concerned that his progress academically may begin to be affected by his lack of participation in classroom instruction. Brian’s family’s primary goal for treatment was for Brian to communicate and participate in the classroom and with other children in the community independent of his parents’ presence.

Zoe—At the time Zoe presented for treatment, her parents reported that she had been speaking less frequently and more quietly outside of the home over the past year. Zoe was not speaking at all in school and was only whispering to a limited number of close peers outside of school. Zoe’s parents reported that she spoke to them and her grandmother at a typical volume when at home. Zoe’s parents attempted to accommodate her lack of speech when in public by allowing her to whisper in their ears and then speaking for her. On the rare occasions that Zoe communicated in school, she whispered to a close peer who spoke for her. Despite these behavioral difficulties, Zoe presented as a happy, cheerful, and well-behaved child with no evidence of dysphoria or anxiety outside of social situations.

Zoe’s parents’ primary concern was that Zoe would fail to meet academic and social milestones. For example, Zoe’s parents and her teachers expressed concerns that Zoe would be unable to complete required achievement tests in school due to the required verbal responses. Zoe’s parents were also concerned about Zoe’s ability to make new friends and develop age appropriate peer relationships. Zoe’s parents’ primary goal for treatment was for Zoe to be able to communicate in school using either a whisper or a typical speaking voice.

4 History

Brian—Brian resided with his mother, father, and older brother on the outskirts of a large metropolitan area. Brian was midway through his first year of preschool and had one year remaining in preschool prior to attending kindergarten. Brian’s parents reported that he had reached early developmental milestones within normal limits, with the exception of his speech articulation and production difficulties which were reportedly remediated. Brian’s parents reported that he had always been a shy child, similar to his father who also reported being very shy in social interactions as a child and as an adult. Brian had no significant medical history and was not taking any medications throughout the course of treatment.

Zoe—Zoe resided with her mother, father, and younger sister and was scheduled to begin kindergarten a few months after seeking treatment. Her parents reported that psychoeducational testing suggested that she was at least of average intelligence for her age and had demonstrated above-average reading capabilities. Zoe’s parents reported that she had reached early developmental milestones within normal limits, but that her speaking behaviors declined in settings outside of the home beginning around age 2. Zoe displayed developmentally appropriate speech in her home and with her parents during this time period and there was no history of delays in language development. As Zoe’s speech declined, she continued to participate in social play, but rarely spoke to other children beginning in pre-kindergarten (age 4). She displayed an ability to form friendships and spoke with peers one-on-one in her home with her mother or father present. Her parents reported that Zoe had always been shy and had experienced nausea and headaches related to anxiety. Zoe’s parents also reported a family history of anxiety and learning disability. Zoe had no significant medical history and was taking 10 ml of fluoxetine at the time of treatment as prescribed by her child psychiatrist for anxiety. Her parents reported that her anxiety regarding speaking appeared to be worsening despite 3 months on medication.

5 Assessment

Brian—Brian and his parents completed a clinical interview and parent-report questionnaires to assess Brian’s current behaviors.

The clinical interview with both of Brian’s parents revealed that Brian had always been a “shy” child, but the level of impairment of his shyness was not salient to the family until Brian entered preschool when he did not interact or engage with his peers or teachers. Prior to preschool Brian had interacted primarily with his brother and cousins. Brian’s parents denied any history of a traumatic social event or teasing/bullying, although they did note that Brian appeared to “feel badly” when people had difficulty understanding him due to his speech difficulties. Brian also refused to speak during his speech therapy sessions, making it difficult for him to make significant progress on his articulation. Information from the interview and records review provided no indication of suicidal or homicidal ideation. Assessment information also provided no indication of depressed mood, inattentive/hyperactive behavior, tics, obsessive-compulsive disorder, other anxiety disorders, autism spectrum disorder, or oppositional defiant disorder. Brian was born full-term without any complications. His parents reported that he met all early developmental milestones within normal limits. Parents denied any history of concussion, seizures, or major illness. Brian’s parents endorsed of family history of anxiety and his mother and father reported being very shy.

Brian’s mother also completed the Selective Mutism Questionnaire (SMQ; Bergman, Keller, Piacentini, & Bergman, 2008) and Spence Preschool Anxiety Scale (SPAS; Spence, Rapee, McDonald, & Ingram, 2001). The SMQ is a 17-item parent-report measure of the child’s social verbal behavior at school, at home/with family, and in public/social settings. Consistent with her verbal report, Brian’s mother reported significant elevations in school and in public/social settings, but not at home/with family. The SPAS is a broader 32-item parent-report measure that assesses the frequency of symptoms of social anxiety, separation anxiety, generalized anxiety, obsessive-compulsive disorder, and fears of physical injury. Brian’s mother reported significant elevations on the social anxiety scale and the separation anxiety scale. When questioned about separation anxiety, she denied impairing separation anxiety, but reported some distress during initial school drop-off.

Zoe—Zoe’s parents participated in a clinical interview and provided comprehensive assessment records for review. During the clinical interview, information was gathered regarding family, social, and academic history as well as the history and current nature of Zoe’s speaking behavior.

Based on information obtained during the clinical interview and record review, Zoe’s parents described a gradual decline in Zoe’s speaking behaviors that began shortly following the birth of her younger sister. Zoe’s lack of speaking behavior became most prominent in her pre-kindergarten classroom when Zoe stopped speaking to her teacher and to most peers. For a short time, Zoe communicated in her classroom by whispering to a peer spokesperson but eventually ceased speaking at school altogether. Zoe was also hesitant to whisper in her mother’s ear in the presence of her teacher or classmates. During her pre-kindergarten year, Zoe’s parents reported that Zoe also began to withdraw from social activities outside of school. At the time that Zoe presented for treatment, Zoe would only speak in the close proximity of her parents in a private setting. Zoe would occasionally communicate in public settings by whispering directly into her parents’ ears using a volume that was inaudible to other listeners. At home, Zoe’s parents described her as “chatterbox.” Zoe’s parents denied any history of a traumatic social event or teasing/bullying at school. Information from the interview and records review provided no indication of suicidal or homicidal ideation. She met all early developmental milestones within normal limits. Parents denied any history of concussion, seizures, or major illness. Assessment information also provided no indication of depressed mood, inattentive/hyperactive behavior, tics, obsessive-compulsive disorder, other anxiety, autism spectrum disorder, or oppositional behavior.

6 Case Conceptualization

Building on existing literature on the phenomenology of SM, an idiographic approach was taken to tailor treatment to the specific needs of each child. The following case conceptualizations seek to demonstrate similarities and differences in the presentations of SM.

Brian—Brian’s presentation was consistent with a diagnosis of SM and, after several sessions with Brian, social anxiety disorder was ruled out due to the proscribed nature of his symptoms to the initiation of speech. Brian displayed a behaviorally inhibited temperament and speech articulation difficulties as a toddler. His parents frequently accommodated Brian’s reticence to speak by acting as his spokesperson in public. Brian’s teachers felt that Brian was just a “shy boy” who would eventually grow out of his SM and, consequently, were not directly addressing his behavior in the classroom. Brian’s refusal to speak at school and in the community appeared to be resulting in functional impairment in his academic and social growth, warranting intervention. Given the primacy of SM, treatment targeted Brian’s anxiety regarding speech initiation using systematic graduated exposure. Due to Brian’s young age and lack of motivation to engage in anxiety provoking tasks, shaping and differential reinforcement procedures were used. Parents, primarily his mother, were involved in treatment procedures with the aim of exposing them to their child feeling anxious and as dispensers of reinforcement (as in Patterson, Littman, & Hinsey, 1964).

Zoe—Given Zoe’s longstanding history of temperamental inhibition, shyness, and anxiety, Zoe’s SM was conceptualized as a part of social anxiety disorder due to the fact that SM symptoms was directly related to social situations, the presence (or lack thereof) of her parents, and to new situations in which she was not yet comfortable. Zoe’s symptoms of social anxiety included increased anxiety when meeting new peers, refusal to give small presentations to her classmates (even using gestures), and increased anxiety in and avoidance of extracurricular activities (e.g., Girl Scouts). The clinical interview also revealed that Zoe’s parents frequently accommodated Zoe’s SM by acting as her spokesperson in public. Despite a lack of distress regarding her failure to speak in most situations, the limited nature of Zoe’s speaking behavior and the functional impairment in academic settings warranted systematic graduated exposure intervention. The treatment also focused on modifying Zoe’s parents’ ability to address Zoe’s anxiety surrounding speaking given their involvement and accommodation of symptoms.

7 Course of Treatment and Assessment of Progress

Brian—The therapist met with the child and his mother or father for 15 weekly 1hr sessions. Therapy consisted primarily of shaping and differential reinforcement procedures and could best be conceptualized as 4 phases of treatment: (a) psychoeducation and implementation of reward system for oppositional behaviors, (b) graduated exposures with adults, (c) graduated exposures with peers, and (d) skills generalization. The rapport building and psychoeducation portion of therapy took place in the first session. This session was following by the development of a fear hierarchy and the initiation of graduated adult exposures, lasting five sessions over the course of 5 weeks. Graduated exposures with peers then began, lasting for seven sessions over the course of approximately 4 months. The final two sessions conducted with the family focused on skills generalization and preparation for termination.

Graduated Exposures with Adults

During the second session, the therapist met with both Brian and his mother to begin socializing Brian to therapy and providing psychoeducation regarding session structure and rules and expectations for therapy. The therapist explained to the family the inclusion of a fear hierarchy in treatment and assigned them homework to begin developing a hierarchy specific to Brian for situations involving familiar and unfamiliar adults and peers. During the session, Brian engaged in an initial exposure task consisting of facing his body toward familiar adults and making eye contact during interactions and communication.

Sessions three through six consisted of graduated exposure tasks based on Brian’s fear hierarchy (see figure 1). Earlier exposure tasks focused on answering close-ended and open-ended questions from adults in the clinic office during sessions and with familiar adults, including his teachers and neighbors out-of-session for daily homework. The therapist emphasized the importance of involving Brian’s preschool teachers and classroom peers in exposures given that this environment was reported to be the most difficult environment for Brian to interact. The family had difficulty getting Brian’s teacher’s assistance with exposures and reported that his teacher seemed to feel the family was pushing him too hard to engage with others. The therapist encouraged the family to call a parent–teacher meeting where the therapist could be invited to provide psychoeducation to Brian’s teacher and school administrator’s regarding SM, the importance of early intervention, and the importance of Brian engaging in exposures in the school environment. Although Brian’s teacher and school staff were receptive to the therapist’s request for Brian to continue with exposures throughout the day at school, they remained only partially compliant with encouraging Brian to complete exposures during the day (e.g., answering questions during Circle Time, sharing a fun fact with a teacher, etc.) throughout the course of treatment. As Brian became more proficient initiating communication with adults, exposures shifted to social interactions with unfamiliar adults.

OPEN IN VIEWER

Figure 1.

Examples from exposure hierarchies.

8 Complicating Factors

Brian—Although Brian’s treatment progressed smoothly, the lack of consistency across contexts may have limited the effectiveness of the intervention. While the therapists acknowledge the difficult job that preschool teachers have balancing the needs of many children, Brian’s treatment may have been more effective had teachers been able to monitor his progress speaking with peers and providing rewards and praise within the classroom environment. It is well established that shorter reinforcement intervals produce stronger behavior change (Bijou, 1957), especially with younger children. Without consistent teacher involvement, Brian reported his successes to his mother at the end of the school day and was provided rewards in the evening. The reliability of Brian’s report was questionable given his young age but was verified by his mother’s observations at drop-off and pick-up. In addition, Brian did not speak in the context of speech pathology, making it difficult for him to make progress with his articulation. Despite the therapist’s efforts to coordinate rewards and exposure tasks across contexts, the competing demands in each context likely limited Brian’s progress.

Zoe—Treatment progressed with few complications during the acute phase. However, during the transition from the acute phase to follow-up phase and generalization of treatment to the school setting, several difficulties were encountered. Although Zoe made rapid progress with speaking behaviors in the therapy setting, she made relatively little progress using whispered or spoken communication in the school setting during the first few months of the transition. After assessing treatment through interview with Zoe’s parents, it became apparent that school-based exposures were not occurring on a regular basis (i.e., not weekly). The reason for the lack of consistency was largely due to the fact that Zoe’s parents were present at all school-based exposures and thus, the occurrence of exposures was dependent on her parents’ schedules. To assist in the transition of school-based exposures from assistance by Zoe’s parents to Zoe’s teachers, the psychologist met with Zoe’s parents and teachers to provide education about anxiety and exposure treatment to Zoe’s teachers and to develop a hierarchy for school-based exposures. Following this meeting, Zoe’s teachers were able to implement exposure practice on a daily basis in school and Zoe’s speaking behavior in school progressed rapidly.

9 Access and Barriers to Care

As noted, treatment was provided in a university hospital–based anxiety specialty clinic where patients were given options to bill insurance, pay out-of-pocket, or pay a reduced fee. Given this range of options and physical accessibility of the clinic to patients, we feel that there were not significant barriers to accessing care. Travel to and from exposure sites (e.g., playgrounds) was not billable time and therefore may make treatment difficult to implement within private practice or community mental health settings. In addition, treatment required a high level of parental time and effort. This may be more difficult for families of lower socioeconomic status. To this end, we have attempted to demonstrate the functional features of the interventions so that treatment may be implemented flexibly to meet the different needs across settings and family situations.

10 Follow-Up

Brian—In the months following treatment, Brian’s mother continued to engage Brian in a variety of novel social settings and he became increasingly comfortable speaking to strangers. As he interacted more with peers, it became evident that Brian had some social skills deficits, likely as a result of not having practiced social skills in the previous years. Although eye contact and smiling had been a focus of treatment, he still had difficulty with knowing what to say to children and how to start out a conversation, even though he was now more comfortable. Although many children with SM would likely benefit from explicit inclusion of social skills training in formal treatment (e.g., Beidel et al., 2000), Brian’s mother opted to work with Brian on social skills training less formally. She was provided educational information on social skills development and materials based on the Skillstreaming in Early Childhood program (McGinnis & Goldstein, 2003). Skillstreaming in Early Childhood is an evidence-based curriculum that provides parents and teachers with a checklist for prioritizing 40 different skills and provides lessons for explicit and direct instruction in the identified skill areas.

Zoe—Zoe’s progress was followed regularly with visits every 1 to 2 months for the last 9 months of her treatment. Over the course of 9 months and a new school year, Zoe continued to make progress speaking in school and with peers outside of school. She encountered difficulty using whispering and speaking behaviors in school with the start of a new school year. However, with the continued guidance of her parents and with teacher-facilitated school-based exposures, Zoe finally achieved the ability to use a normal speaking voice on a regular basis throughout the day in school. During follow-up, Zoe’s parents continued to communicate regularly with her teachers at school and design new exposures.

11 Treatment Implications of the Case

Specifically, these cases highlight the importance of in-vivo exposure in extinguishing anxiety and generalizing skills. In both cases, children practiced approaching new peers either in the clinic or in naturalistic settings like playgrounds or in the school. In-vivo exposure tasks have been shown to be superior to imaginal exposure scripts in extinguishing fears and generalizing treatment gains (Menzies & Clarke, 1993). In both cases, when exposure tasks were conducted in naturalistic settings, patients experienced the greatest gain.

In addition, these cases demonstrate the importance of parent involvement in treatment. As discussed, parents play a unique role in delivering reinforcement to children, but in the case of children with SM, it may be most important to expose parents to their children tolerating anxiety during exposure tasks. In Brian’s case, rewards were used in the context of oppositional behavior. In Zoe’s case, rewards were used in the context of avoidant behavior. However, in both cases, they were necessary for treatment compliance and progress. As suggested by Ginsburg, Silverman, and Kurtines (1995), parental involvement in treatment sessions may be paramount based on the transfer of control model of change which allows for greater maintenance of treatment effects.

12 Recommendations for Clinicians and Students

The two cases of Brian and Zoe illustrate the adaptation of empirically supported behavioral principles. Understanding the underlying principles in reward programs and exposure hierarchies allows the therapist to teach parents to flexibly implement treatment outside and beyond the session. In addition, when working with young children in particular, it is important to adapt treatment to the current needs of the child. As in these cases, the therapists modified the reward systems and the level of difficulty of exposures to continue to engage and motivate the patients and families.

Finally, therapists should not hesitate to plan exposures in the context of the child’s presenting problems; utilizing nearby locations that may serve as analogues, and planning ahead with families for practicing new skills, allows families to make more relevant, generalizable gains.