Abstract
This case study describes the use of cognitive-behavioral therapy (CBT) for a 7-year-old girl, “Lauren,” who was diagnosed with social anxiety disorder (SAD) and was experiencing significant family stress. Manualized CBT was conducted and substantial reductions in SAD symptoms and impairment were demonstrated. Given additional difficulties remaining in the home setting, booster sessions were continued following manualized treatment to target Lauren’s functioning at home. These follow-up sessions were associated with maintained functioning, and only minimal additional benefits emerged. This case study calls attention to the importance of considering family stress in individual CBT for children with SAD, as well as the difficulties associated with addressing family stress in CBT for SAD. Existing treatments do not routinely address family stress in the context of child SAD, despite the potential to enhance CBT outcomes for children with SAD.
1 Theoretical and Research Basis for Treatment
Social anxiety disorder (SAD) affects at least 12% of the population and is characterized by persistent fear of one or more social or performance situations that involve exposure to unfamiliar people, possible scrutiny, or public embarrassment (Ruscio et al., 2008). Although SAD most commonly presents during adolescence, likely due to the high salience of social functioning and social evaluation during this developmental period, it has also been clearly identified as a relevant disorder for children. In childhood, SAD is associated with substantial impairments, in both academic and social functioning, which can persist or worsen if not adequately treated (Mohatt, Bennett, & Walkup, 2014).
Much empirical support documents the efficacy of cognitive-behavioral therapy (CBT) for children and adolescents with SAD (Silverman & Field, 2011). Several large-scale trials have documented notable improvements in anxiety symptoms and related impairment from treatment in group, family, and individual settings. However, nearly 40% of children who initiate treatment do not improve (James, Cowdrey, Soler, & Chocke, 2013), and several reports suggest that children with SAD, compared with other anxiety disorders, are at highest risk for the poor CBT outcomes (Hudson et al., 2015; Kerns, Read, Klugman, & Kendall, 2013). However, the reason that SAD, in particular, is associated with poor outcomes is not well understood. Thus, research attending specifically to complicating issues in CBT for childhood SAD is greatly needed.
Family stressors such as divorce, parental psychopathology, and family conflict have also been shown to meaningfully contribute to the development of SAD (Rapee, 2012). Family stress, particularly the accumulation of multiple family stressors, has also been linked to negative treatment outcomes (Bagner & Graziano, 2012) and may be particularly important in understanding the risk for poor CBT outcomes related to SAD. Yet manualized treatments for CBT do not routinely target these difficulties, despite evidence that treatment of family stress improves child functioning (Cobham, Dadds, Spence, & McDermott, 2010).
2 Case Introduction
The present report describes a case study of Lauren, a 7-year-old White girl presenting with SAD. Lauren presented for treatment in the summer before second grade. She attended a suburban public school in the northeastern United States in a regular education classroom. At the time of treatment, Lauren was experiencing significant family stress. She split time between her biological parents’ homes, as her father had moved out within the past 2 years to pursue another romantic relationship. Her biological mother had recently remarried, and Lauren’s stepfather and his three children recently moved in to her mother’s home, causing significant distress to the family. In particular, one of Lauren’s new stepsiblings was diagnosed with autism, and his often disruptive and oppositional behaviors caused family conflict and distress. As will be described below, it was believed that Lauren’s social discomfort and predisposition to shyness were exacerbated with these major family stressors.
3 Presenting Complaints
Lauren was referred for treatment by her primary care physician due to persistent stomachaches and headaches. Despite substantial medical testing, no physical reason was identified to explain her frequent somatic symptoms. Lauren’s mother described that her stomachaches occurred nearly every day, and typically lasted several hours. Her headaches usually occurred in her mother’s home, most frequently on the weekends, and before and after school. Her mother described that Lauren’s stomachaches often occurred with headaches and back pain, which required her to lie down for hours during the day. Furthermore, her eating was disturbed due to her stomachaches, causing Lauren to miss meals or, on other occasions, eat more than expected. At the time of the initial therapy intake, Lauren’s weight was within the expected range. She frequently was absent from school because of these somatic complaints and attended numerous medical appointments related to them.
According to Lauren’s mother, the onset of Lauren’s somatic symptoms occurred within the 3 months prior to starting treatment. Her mother questioned whether these symptoms were related to stress, as the onset of the symptoms coincided with her stepfather and stepsiblings moving in to her home full-time, a transition Lauren’s mother described as markedly stressful for the whole family. Lauren’s mother described that her current husband had recently obtained full custody of his children, but was still involved in legal proceeding related to maintaining full custody. She described that she and her husband were experiencing difficulty merging their families and adjusting to parenting five children. In addition, Lauren’s stepbrother was diagnosed with autism. He often displayed disruptive and aggressive behavior, which Lauren’s mother reported caused considerable yelling and conflict in the home. Although Lauren and her mother described that she was closest in age to and often played with her stepbrother in the house, at other times he would direct his anger and yell at her, which caused her to have “meltdowns” as her mother described, during which Lauren would cry and run away to her room. Lauren’s mother reported that Lauren would sometimes need to stay in her room for several hours to calm down and manage her distress after these negative interactions with her stepbrother. Although her mother reportedly encouraged Lauren to use coping strategies to calm down, such as laying down, deep breathing, and distracting herself, it was not clear that Lauren used these strategies consistently or sufficiently. Her mother described that Lauren was particularly sensitive to criticism and would often become extremely upset from seemingly minor interactions with her stepsiblings. For example, she would start crying, telling her mother that her younger stepsister looked at her in a particular way. Prior to her stepfamily moving in, Lauren lived with her mother and younger biological sister. They reported very little conflict in the home at that time. According to Lauren’s mother, Lauren did not experience behavior problems and had not required any sort of discipline.
4 History
Lauren’s mother reported that during her pregnancy with Lauren, she was hospitalized for preterm labor. Lauren was born 3 weeks early, but there were no concerns with Lauren’s development. At the time of her initial therapy intake, Lauren was prescribed Prilosec due to frequent stomachaches for which no identifiable cause was found. She was also prescribed Zantac 75 mg and was followed by gastroenterology due to persistent stomachaches and concerns heightened by her younger sister’s diagnosis of eosinophilic esophagitis. Lauren was diagnosed with chronic idiopathic urticarial which was controlled with Zyrtec 10 mg. She also had a history of hives in reaction to sulfa-based medication. Her family history was significant for bipolar disorder (paternal aunt), but no other family psychiatric illness was reported.
Lauren’s mother reported that nearly 2 years before initiating the current treatment, when Lauren’s father suddenly left home to pursue another romantic relationship, Lauren had experienced a period of a few months with persistent and impairing stomachaches. However, once she had reestablished consistent and predictable contact with her father, these somatic complaints had diminished considerably. At the time of the initial intake, she spent time with her biological father every other weekend, partially due to her parents’ availability (i.e., father at work, mother unemployed). Both parents shared custody of Lauren and her younger sister. Communication between parents was reportedly cordial, and no conflicts were reported. Notably, Lauren rarely described somatic complaints at her father’s home. Her mother described a large and supportive network of family on both sides.
According to both her mother and father, Lauren was not experiencing any academic difficulties in school. Lauren reported loving school and earning good grades. She and her mother denied that she experienced any difficulties with peers and reported that she had a number of friends at school. However, it appeared that she had few close friends at school and, instead, spent most of her time with her best friend, who was also her cousin. She did not have frequent social contact with other peers individually outside of school. However, Lauren was involved in church groups, soccer, and reported wanting to be a girl scout. Lauren had never previously been involved in psychological treatment. Her mother and stepfather were involved in treatment related to managing disruptive child behavior, which her mother described was somewhat beneficial. They also regularly attended religious programming, addressing issues related to blending families, which Lauren’s mother also reported as beneficial in improving expectations about their family functioning.
5 Assessment
A clinical interview was conducted reviewing all symptoms of Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV; American Psychiatric Association[APA], 1994) diagnoses relevant to children. Lauren provided little information in this evaluation, and when probed by her mother to answer questions, she would remain quiet, hide her head in her shirt, and sometimes whisper to her mother that she was tired. Lauren’s mother described that Lauren experienced marked discomfort in social situations, concerns about how she is perceived by others, and anxiety around adults and other people she does not know well. For example, in the previous year at school, Lauren had an accident during school because she had been too afraid to ask a new teacher at recess whether she could use the bathroom. She also described that Lauren had frequent somatic complaints, trouble separating from caregivers, and a fear of heights. Lauren’s mother described a number of depressive symptoms, including often feeling sad and crying, sleep difficulties, and feeling worthless, inferior, and unloved. Although these symptoms became most noticeable within the last 3 months prior to beginning treatment, upon further discussion, Lauren’s mother described a more chronic pattern of discomfort in social situations that had persisted from early childhood. Additional rating scales were completed by Lauren and her mother to better characterize the severity of these symptoms. At intake, her father also completed selected forms.
Impairment Rating Scale (IRS) for Children
Lauren’s mother rated Lauren’s impairment in peer, sibling, and parent relationships, academics, self-esteem, family functioning, and overall need for treatment on the Impairment Rating Scale for Children (IRS-C; Fabiano et al., 2006). The rater places an “X” on a line to indicate the level of impairment in each domain and provides a brief description for each domain. The line is divided into seven equal portions, to rate impairment from 0 (no problem, does not need treatment or special services) to 6 (extreme problem, definitely needs treatment or special services). The IRS-C has good concurrent, convergent, and discriminant validity and has been shown to be highly accurate in identifying impairment in children with behavior problems (Fabiano et al., 2006), with scores below 3 indicating subclinical functioning. At the time of the intake, maternal ratings indicated significant impairment related to Lauren’s relationships with siblings and her family in general. Significant self-esteem impairment and overall adjustment difficulties were also reported.
Short Moods and Feelings Questionnaire (SMFQ)
Depressed mood was assessed by parent and child report on the SMFQ. The SMFQ includes 13 items that assess depressive symptomology within the last 2 weeks on a 3-point scale (0 = not true, 1 = sometimes true, 2 = true). The SMFQ has been shown to discriminate between children with and without depressive disorders, and Lauren’s responses indicated a score of 7, which is indicative of moderate levels of depressive symptomology (Rhew et al., 2010). Items she indicated were crying a lot, feeling lonely, feeling as though she was no good anymore, and moderate levels of feeling like she could never be as good as other kids. Her mother’s ratings yielded a score of 4, also less than the suggested clinical cut-off. Items her mother rated (at the “sometimes true” level) included feeling miserable or unhappy, crying a lot, feeling lonely, and feeling as if she could never be as good as other kids. Her father’s ratings indicated a score of 0.
Screen for Child Anxiety Related Emotional Disorders (SCARED)
The SCARED (Birmaher et al., 1997) was used to assess symptoms of Diagnostic and Statistical Manual of Mental Disorders (DSM) anxiety disorders. It includes 41 items that are rated on a 3-point scale (0 = not true or hardly ever true, 2 = somewhat true or sometimes true, 2 = very true or often true) to yield scores on five subscales and a total anxiety score. Lauren, her mother, and her father completed the SCARED. Her ratings yielded clinically significant overall (25), separation anxiety (6), and SAD (12) scores. Her mother’s ratings indicated a significant generalized anxiety score (11), and her father indicated significant overall anxiety (26) and SAD (10).
Disruptive Behavior Disorders Rating Scale (DBD)
The DBD (Pelham, Gnagy, Greenslade, & Milich, 1992) was used to assess for any behavioral problems perceived by Lauren’s mother. The DBD includes DSM symptoms of attention deficit hyperactivity disorder (ADHD), oppositional defiant disorder (ODD), and conduct disorder (CD). The DBD includes Diagnostic and Statistical Manual of Mental Disorders (3rd ed., rev.; DSM-III-R; APA, 1987) and DSM-IV symptoms of ADHD, and has been shown to be a reliable and valid measure of ADHD and related DBD (Pelham, Fabiano, & Massetti, 2005) with normative data. Lauren’s mother did not report any clinically impairing symptoms.
Altogether, Lauren met diagnostic criteria for SAD. Despite elevated levels of generalized anxiety disorder symptoms rated by her mother on the SCARED, items that were most elevated were those that involved social concern (e.g., worrying about other people liking her, being as good as other children), and it was not clear from Lauren’s report or her mother’s report that concerns beyond her social competence were relevant. Of note, some symptoms of depression were indicated, particularly crying and feelings of loneliness, although these symptoms did not appear sufficient to warrant a clinical diagnosis.
6 Case Conceptualization
Somatic symptoms are a common manifestation of anxiety disorders, particularly in children (Campo, 2012). Lauren’s headaches and stomachaches were the primary concern at the time of initial intake, as they were causing impairment in her daily functioning. At that time, anxiety, although suspected at least by her mother, was not an initial presenting concern of her mother’s. Lauren’s mother’s initial ratings of SAD and total anxiety (see Figure 1) were actually below the clinical cut-off at the time of treatment, likely due to her not realizing the extent to which Lauren’s current problems were due to psychological rather than physical reasons.
Course of treatment outcome.
Lauren’s mother described that Lauren had always been a shy child. Although she reported that Lauren was not particularly assertive and was always sensitive to criticism, her social inhibition had never been problematic until recently. In fact, there were few times when Lauren needed to assert herself, as her mother would often accommodate Lauren’s anxiety and speak up for her. For example, as observed during the therapy intake, Lauren’s mother often answered questions directed at Lauren, thus eliminating opportunities for Lauren to speak to people, such as the clinicians, with whom she was not familiar. In addition, Lauren was compliant, thoughtful, and helpful at home, and according to her mother, she never needed to use discipline or provide negative feedback with Lauren. Likewise, at school she was well-liked and excelled academically, and did not receive critical feedback. She was involved in sports and extracurricular activities, and difficulties, at least initially, were not apparent.
Lauren’s shy temperament represented a vulnerability, which only became problematic when she was experiencing overwhelming family stress. Parental divorce, remarriage, living with a sibling with autism (Pollard, Barry, Freedman, & Kotchick, 2013), and large family size are all individual risk factors related to child internalizing problems (Rapee, 2012). Altogether, these family risk factors exerted cumulative risk for the development of Lauren’s anxiety. In particular, it was believed that her stepbrother, who often directed verbally aggressive behavior toward Lauren, exacerbated her anxiety. For example, she had extreme difficulty coping with the verbally aggressive behavior of her stepbrother, as she had little experience coping with criticism from others and was markedly uncomfortable with such feedback. These interactions were unpredictable, as on other occasions she would enjoy her time with her stepbrother. His verbally aggressive behavior also appeared to present distress that was too overwhelming for Lauren to manage.
At the same time, the family stressors that Lauren experienced also negatively affected the family more generally. Lauren’s mother was no longer able to spend as much time with Lauren as she had three additional stepchildren to parent. In particular, Lauren’s stepbrother required much of her mother’s time and effort, because her mother reported that Lauren’s stepfather had significant difficulty consistently and effectively managing his disruptive behavior. In fact, this inequity in child-rearing responsibilities appeared to cause some conflict in their relatively new marriage. Given the greater parenting responsibilities associated with raising five children, Lauren’s mother had fewer opportunities to discuss Lauren’s upset reactions, and also had fewer opportunities to step in to defend Lauren in these uncomfortable, unfamiliar interactions with her stepsiblings and stepfather. Thus, although she had previously been able to frequently accommodate Lauren’s social discomfort by speaking for Lauren or reassuring her in uncomfortable situations, there were now fewer opportunities to do so. In fact, Lauren was left to assert herself with her stepbrother, which she had difficulty doing. On other occasions, she was left with her stepfather and did not feel comfortable making requests of her stepfather or informing him of problems with her stepbrother. Moreover, on other occasions, her stepfather would raise his voice toward Lauren only because he was upset with his son’s disruptive behavior. This further intensified her distress at home.
Relatedly, it is important to point out that Lauren’s difficulties primarily occurred at her mother’s home. Few difficulties were noted at her father’s home or school, with the exception of some instances of not speaking up at school and frequently missing school due to medical appointments. In her mother’s home, in addition to disruptive child behavior, large family size, and potential marital discord, other stressors, including her mother’s unemployment, financial strain, and potential parental psychopathology, which was discussed in later appointments (described below), also likely added to the general level of distress.
Given the large to very large immediate and long-term effects of CBT for childhood anxiety disorders (Silverman & Field, 2011), it was chosen as the most appropriate treatment for Lauren. Specifically, psychoeducation, cognitive restructuring, and exposure were deemed relevant to help Lauren better control her symptoms and impairment related to SAD. The individual CBT manual, Coping Cat (Kendall & Hedtke, 2006), was selected to guide treatment, given its wide use and large evidence base. Of note, Coping Cat is typically suggested for children ages 7 through 13. CBT skills may be challenging for some children to understand, as they require self-awareness and the ability to think abstractly, and emerging research suggests that parent–child based treatment may optimize treatment outcome for your children (e.g., Hirshfeld-Becker et al., 2010). However, individual CBT was selected for Lauren for several reasons. First, given the aforementioned family stressors that Lauren’s family experienced, it did not appear feasible that her mother would be able to consistently bring Lauren to the treatment sessions. For example, she described on several upcoming appointments that another family member would be bringing Lauren to treatment, due to existing commitments related to other family members (e.g., court related to husband’s custody proceedings) that Lauren’s mother would need to attend. Second, Lauren’s mother reported that she excelled academically, which suggested she may understand and benefit from the somewhat abstract thinking skills inherent in CBT. Third, individual CBT has been used and associated with notable improvement for children even younger than Lauren (e.g., Michael, Payne, & Albright, 2012) with some developmentally appropriate modifications (Beidas, Benjamin, Puleo, Edmunds, & Kendall, 2010). Furthermore, given that Lauren experienced distress in situations away from her mother (e.g., school), we believed individual therapy that aimed to equip Lauren with skills she could apply independently across situations would be worthwhile.
7 Course of Treatment and Assessment of Progress
As mentioned above, treatment was based on the CBT protocol, Coping Cat (Kendall & Hedtke, 2006). This manual includes 16 weekly individual sessions, as well as one parent session (between Sessions 4 and 5) to provide psychoeducation to parents to help to reinforce and generalize treatment. The goal of treatment, initially, was to reduce Lauren’s SAD symptoms and impairment, by providing her and her mother with psychoeducation about anxiety and skills to manage her anxiety and distress in social situations, along with coping strategies to alleviate her somatic symptoms. More specifically, Lauren and her mother discussed the following goals: to reduce the frequency of headaches and stomachaches, particularly so that visits to the doctor and absences from school were significantly reduced; to increase Lauren’s assertiveness so that she was able to approach teachers with concerns (e.g., needing to use the bathroom); and managing her emotions more appropriately at home. These goals were generated by Lauren’s mother and the therapist; Lauren agreed with these goals but did not offer additional treatment goals. Each session focused on specific skills to help Lauren manage her anxiety and provided relevant homework exercises to reinforce treatment topics outside of therapy. Specific components of the manualized treatment are described in detail below.
Rapport Building
Considerable time and attention during the initial sessions were devoted to improving rapport with Lauren. Session 1 was devoted exclusively to developing rapport and trust in the therapists. Therapeutic alliance has been related to therapy outcomes for child CBT (McLeod et al., 2014), and given the significant difficulties that Lauren faced even responding to questions during the initial appointment intake, it was deemed an important aspect of the treatment. In fact, given the severity of Lauren’s SAD symptoms, social interactions such as playing games were conceptualized as therapeutic exposures, as it took her some time to be able to talk openly with the therapists. Lauren’s treatment was conducted in a training clinic for child and adolescent psychiatry fellows, supervised by a clinical psychologist (the first author). Thus, the presence of both therapists in the room also provided a relevant initial exposure.
Psychoeducation
The first five treatment sessions focused on orienting Lauren and her mother to the nature of CBT. In addition to discussing her treatment goals and the course of treatment, psychoeducation was provided to Lauren and her mother. Psychoeducation related to the somatic manifestation of anxiety symptoms was particularly important for Lauren and her mother who were at times not clearly aware, at least initially, that her headaches and stomachaches were related to worry, despite no medical reasons being identified. Session topics also included discussion of negative cognitions and practice of progressive muscle relaxation and deep breathing. Lauren appeared to understand these techniques well. In this treatment phase, Lauren and her mother displayed increased understanding of the nature and manifestation of anxiety, particularly the cognitive and somatic manifestations of anxiety. Lauren and her mother tracked her somatic complaints throughout this treatment phase, and after 3 weeks of treatment, somatic complaints were no longer occurring on a weekly basis, and thus were no longer tracked. This phase focused on increasing Lauren’s awareness of her thoughts, emotions, and distress.
Cognitive Restructuring
Consistent with the Coping Cat manual, Sessions 6 through 10 were devoted to learning cognitive restructuring. This included teaching Lauren to recognize negative and anxious self-talk and to develop more rational alternative thoughts. These sessions also discussed the importance of Lauren evaluating her own progress and providing herself rewards for her progress toward her therapy goals. Lauren appeared to understand the concept of cognitive restructuring. She was able to identify negative thoughts and develop more balanced thoughts. However, she appeared to have difficulty applying this skill outside of the therapy setting, as her homework related to cognitive restructuring was never complete, and only with prompting from the therapists did she appear able to both identify negative thoughts and generate new alternative thoughts. During this treatment phase, self-talk appeared more helpful to Lauren, as she developed several mantras (e.g., “I can do this!”) which she used frequently to motivate her to continue working through stressful situations.
Behavioral Exposure
Alongside discussion and practice of cognitive restructuring in Sessions 9 and 10, Lauren also began exposure exercises. These practices were initially intended to provide Lauren with exposure practice in situations provoking only a low level of anxiety, but given Lauren’s rapid acclimation to these situations and her desire to engage in more difficult exposures, she attempted exposures that provoked considerable anxiety (see Table 1). These exposures were planned by the therapists. Lauren rated her fears using a Subjective Units of Distress Scale (SUDS) from 0 to 10, where 0 = no anxiety and 10 = high anxiety, and, at least initially, almost all exposures presented to her were rated as causing very high levels of anxiety. For example, items such as standing up in the therapy room and talking with staff member in the clinic, whom she had never met and calling a restaurant to inquire about the hours of business were all rated in the very high anxiety range (i.e., 9-10 out of 10 on the SUDS). Additional time to discuss and better calibrate Lauren’s SUDS rating was used, and it was thought that starting in exposures, such as standing up in the room, although rated highly, would not be as anxiety-provoking as talking to people she did not know. Therefore, to begin exposure practices, several exposures thought to be mildly anxiety-provoking in the therapists’ view were presented. Following practice in Sessions 9 and 10, a fear hierarchy was developed, and Sessions 11 to 15 involved additional, more formal exposure treatment. Lauren had some difficulty generating exposures for her hierarchy. The therapists offered a list of ideas and also talked with Lauren’s mother to generate a list. Her SUDS ratings for different exposures were reassessed in subsequent sessions, and as exposure treatment progressed, Lauren was able to provide more input in developing meaningful exposures that would challenge her anxiety most directly. See Table 1 for a description of exposure activities with before and after SUDS ratings.
Exposure Exercises.
Note. Numerical ratings listed above are based on the SUDS rating, for which scores range from 0 to 10 with higher scores indicating greater anxiety and distress. SUDS = Subjective Units of Distress Scale.
Parental Involvement
Consistent with the manualized treatment protocol, Lauren’s parents attended one session without Lauren present. This visit occurred during the psychoeducation phase and was designed to provide her parents with more information about her treatment, collect additional information about her difficulties, and discuss the importance of reinforcing her treatment gains and not accommodating her anxiety. This session was particularly important for Lauren’s parents, as they were divorced and were reporting somewhat different patterns of Lauren’s behavior. Her father stated that he generally did not see Lauren’s worry as impairing, whereas her mother reported that these symptoms were primarily occurring in her home. However, both reported understanding how the treatment would target her anxiety. Her father denied any concerns about how he could support her treatment. However, her mother reported some concern, as she reported that she had not realized how Lauren’s difficulties were related to her social interactions. Her mother reported that she herself had substantial social anxiety and would engage in many of the same behaviors as Lauren (e.g., not using the phone, preferring to be alone). Thus, her mother reported some concern that she would be able to sufficiently change her own behavior and support Lauren to do the same. Also, reflected in her apparent increase in symptoms and impairment at this time, it appeared that Lauren’s mother learned more about the manifestations of social anxiety and became more aware of these difficulties in Lauren’s life.
We believed that Lauren’s treatment would be enhanced by including her parents beyond the one session specified in the Coping Cat manual. Given that a large portion of Lauren’s difficulties were occurring in her mother’s home, we were particularly interested in including her mother in the sessions, and, on the days when she would bring Lauren in for treatment, we would briefly discuss the session content, the skills we had discussed and practiced with Lauren, as well as the homework to help Lauren’s mother support her progress outside of the treatment setting. In addition, ratings were routinely collected from Lauren’s mother every 2 weeks to assess her perception of treatment outcome, and ratings were discussed with her. She was encouraged to discuss new concerns she had noticed related to Lauren’s anxiety as well as progresses and achievements. Furthermore, she was frequently reminded to encourage rather than accommodate Lauren’s anxiety.
Phase 1 Treatment Outcome
After the 16 weekly sessions, Lauren experienced clinically significant reductions in anxiety (see Figure 1). Her overall SCARED score by maternal report reduced from 23 to 11 and from 25 to 12 by self-report. Specific scores on subscales were also reduced: the SCARED SAD score dropped from 7 to 5 by maternal report and from 12 to 11 by self-report; the generalized anxiety disorder (GAD) score dropped from 23 to 11 by maternal report and from 1 to 0 by child report. In addition, impairment rated by Lauren’s mother also was significantly reduced from an average of 3 overall to 2.56, which is below the clinical level (Fabiano et al., 2006). Depression ratings dropped from 7 to 0 by self-report and 4 to 0 by parent report. The initial treatment goals of psychoeducation and reducing headaches and stomachaches were also met, and she had learned and implemented a range of effective coping strategies to manage her anxiety (e.g., exposure, muscle relaxation, cognitive restructuring). With the exception of Lauren’s own ratings of social anxiety on the SCARED, all ratings of anxiety were below the clinical cut-off. Given that Lauren and her mother consistently described improvement in her social functioning, assertiveness, and confidence in various settings including at school and extracurricular activities, it may have been that the SCARED, a screening measure, was not sufficiently sensitive to detect treatment differences. For example, items asked about her thoughts and comfort level in social situations but did not necessarily tap into whether or not she was avoiding social situations and whether or not social situations were causing distress, which would be better markers of SAD.
Figure 1 shows Lauren’s improvements in bweekly ratings of social anxiety, generalized anxiety, and total anxiety as rated on the SCARED throughout the course of treatment. From psychoeducation, substantial initial improvements were seen in her mother’s report of total anxiety symptoms. However, measures of social anxiety and generalized anxiety did not show clear improvement during this treatment phase. This is likely due to Lauren and her mother’s earlier confusion regarding the nature of her somatic symptoms and behavior. They had initially believed the symptoms were organic and not related to her psychological functioning, and her mother had remarked in discussing these ratings that she had not been cognizant of these behaviors and had not necessarily viewed them as problematic until recently, especially after the psychoeducation treatment phase. This is likely why during the cognitive restructuring phase, when Lauren was learning to identify thoughts and generate more realistic interpretations of situations, Lauren’s mother rated her social anxiety as increasing relative to her initial baseline ratings. Although Lauren rated reductions in her generalized and total anxiety symptoms during this time, observations of her behavior in session were not clearly indicative of this progress. Although she was able to discuss thought records, she generally did not report using cognitive strategies in anxiety-provoking situations when they would be most beneficial, and although her symptoms of generalized and total anxiety were below the clinical cut-off during that treatment phase, her symptoms of social anxiety, the primary target of treatment, were still in the clinical range by both her own and her mother’s report. In contrast, dramatic improvement by maternal report was seen during the exposure treatment phase. Clear improvements in Lauren’s assertiveness and mood were also observable during the treatment session.
Despite marked improvement and having met all initial treatment goals, Lauren and her mother both expressed reservation about terminating CBT. Lauren described that she really enjoyed treatment and would be sad to leave. Of note, Lauren was typically brought to session with her mother, allowing her special time with her mother away from the chaos of their home. In fact, both she and her mother reported enjoying this time together, and they often enjoyed breakfast at a restaurant together or spent time in some other activity after the session. Lauren’s mother also noted that Lauren was still having some difficulty asserting herself at home, particularly with her stepfather. She reported that, although less frequently than when she first presented for treatment, Lauren would often become upset after interactions with her stepfather. According to Lauren’s mother, Lauren would often come to her mother with complaints about her stepfather making an unfair request or providing unjust consequences to her just because another sibling had engaged in a negative behavior. Based on Lauren’s mother’s report, it appeared that these interactions with Lauren’s stepfather were partially related to SAD. Thus, additional treatment specifically targeting Lauren’s responses and assertiveness in interactions with her stepfather was deemed worthwhile.
Based on her mother’s report, it also seemed as though attention to family behavior management strategies would further support Lauren’s well-being and would potentially reduce levels of parent strain and family stress. For example, clear and consistent house rules would help to eliminate Lauren seeking her mother’s response after given another response from her stepfather and increase her trust and confidence in her stepfather. In addition, positive preventative family strategies such as spending one-on-one time together may improve Lauren’s relationship with her stepfather and also help to solidify her relationship with her mother, which was affected by her family’s transition. Lauren’s mother agreed with treatment targeting parenting strategies, and she was encouraged to bring her husband to these appointments. She stated that he would likely be unable to attend due to his work schedule, although she felt confident in her ability to convey the session material to her husband. These maintenance sessions to enhance the benefits seen from the first phase of treatment (i.e., manualized CBT) were conducted every other week.
There was substantial discussion regarding ways that Lauren could challenge her social anxiety at home with her stepfather, and Lauren provided several examples, such as spending time playing a game with her stepfather and making reasonable requests of her stepfather. Lauren appeared eager to apply the same strategies she had mastered from the manualized treatment phase. In one situation, Lauren had asked her stepfather for another serving of banana as a way to challenge her social discomfort. She described that her stepfather told her that it would not be fair to give her more food as there was not enough food left for all of her siblings to have more. Lauren then asserted that she was the oldest, deserved more to eat, was hungry, and there were more bananas remaining. Lauren reported that she became upset, but managed her emotion well, as she had not run to her room. After her interaction with her stepfather, she told her mother what had occurred, and her mother gave her another banana. In this situation, Lauren had attempted to assert herself with her stepfather, and, when her request had not been granted, she subsequently proceeded to obtain her goal (in this case the banana) by approaching her mother. Although we believed this example illustrated meaningful improvement in Lauren’s social discomfort, this example also exemplified ongoing challenges within Lauren’s family more generally. Specifically, there was a lack of agreement between her mother and stepfather about how to respond to her requests and how to delegate parenting responsibilities in their blended family. Lauren also discussed several examples of attempted exposures in the home setting that were not successful, mostly because her mother appeared to step in and eliminate the need for Lauren to assert herself. Despite discussion of ways to enhance the utility of these exposures at home, Lauren and her mother continued to report that exposures were difficult to conduct. Even more, these discussions began to involve significant discussion of Lauren’s mother’s frustrations with her husband. For example, she revealed that she believed he did not understand Lauren’s needs. However, it was not clear that he was given sufficient opportunity to collaborate in Lauren’s care. The tone of these conversations became increasingly negative and seemed inappropriate to discuss in front of Lauren. Furthermore, these conversations detracted from all of Lauren’s progress in treatment and were not clearly relevant to her care. It was discussed that these concerns would likely be more appropriately treated in family therapy or marital therapy, and Lauren’s mother was accepting of these recommendations. Thus, Lauren’s CBT, given her notable progress and maintenance of progress during these parent–child booster sessions, was terminated by mutual agreement of the therapists, Lauren, and her mother.
8 Complicating Factors
It was believed that family stressors impacted the course of CBT in several ways. Lauren consistently did not complete homework assignments, which has been shown to impact treatment outcome (Cummings, Kazantzis, & Kendall, 2014). Despite numerous discussions about problem-solving issues related to her completing her therapy homework and discussions that included her mother, her homework was complete only one time throughout the entire course of therapy. Although there are likely numerous explanations for her incomplete homework, including her own avoidance of anxiety-provoking assignments, family stress likely affected her ability to complete assignments for a number of reasons. While Lauren’s mother primarily brought her to each session, Lauren often spent time with other family members during the week, which she reported often resulted in her forgetting to complete her homework. On other occasions, when she was with her mother for the entire week, Lauren described that she often did not have enough time with her mother to complete the homework assignment together. Furthermore, at other times, Lauren was brought to treatment by her father or other family member, and it was not clear that the homework or session content was communicated to Lauren’s mother.
It was also not clear that Lauren’s treatment was prioritized consistently at home. Although there was little doubt that Lauren’s treatment was important to her mother, other family stressors often required more immediate involvement of her mother. The extent to which she and her mother could work on therapy homework often depended on the intensity of her stepbrother’s behavior problems, as her mother needed to manage those first. It also became clear that her mother was struggling with marital difficulty and her own anxiety and stress, which likely further negatively affected the treatment outcomes. Much research on family stress suggests that it is not one stressor per se which leads to negative treatment outcomes, but rather the accumulation of these problems and their compounding effect on family maladjustment (Bagner & Graziano, 2012). Given the numerous family stressors, it is likely that Lauren’s mother was experiencing burnout and addressing her stressors would likely have a positive impact on Lauren’s as well as the entire family’s adjustment.
9 Access and Barriers to Care
Lauren and her parents were consistently on time for appointments and did not cancel any appointments. Lauren’s family did not report any financial burden from the cost of treatment, which was covered by insurance. They traveled approximately 45 min each way to the treatment sessions which occurred during the morning. Thus, Lauren did miss some school to attend appointments, which her mother reported was not excused by her school even with a note and explanation for her absence. Her mother described that she was not concerned with Lauren’s absences given her continued academic progress and the importance of her treatment.
10 Follow-Up
Given her age and the remaining stressors in the family, the possibility of relapse was discussed at length prior to terminating treatment. Lauren and her mother had not returned to the clinic for difficulties after 1 year. In fact, while Lauren has experienced some stomachaches, they have been mild, and she has reported to her primary care physician that she has felt successful in her ability to manage her distress and approach new social situations. The CBT skills she learned in therapy, including psychoeducation, cognitive restructuring, positive self-talk, and behavioral exposure, as well as the skills her mother learned likely reduced the need for her to seek additional treatment. As further evidence of Lauren’s and her mother’s satisfaction with treatment, they have highly recommended the treatment to their primary care staff.
11 Treatment Implications of the Case
This case study illustrates the use of CBT for a 7-year-old girl with SAD complicated by substantial family stress. Using a commonly used manualized treatment, she experienced clinically significant improvements in symptoms and impairment, and her improvement was demonstrated relatively quickly upon initiating treatment. Given that the majority of Lauren’s difficulties occurred in the home setting, some difficulties persisted in her mother’s home after the completion of the manualized treatment, despite marked improvement in her overall functioning. Maintenance booster sessions, specifically targeting home difficulties, were continued to reinforce the gains made during the manualized treatment phase. These sessions were continued with minimal additional benefit. CBT was discontinued at that time, as Lauren had met her goals, although family stress had persisted.
Overall, Lauren demonstrated a marked decrease in her symptoms and impairment over the course of the manualized treatment. While her symptom ratings of total anxiety just missed the clinical level at intake, ratings on the social anxiety and generalized anxiety subscales denoted areas of clinical concern. By both self and parent report, her total anxiety symptoms were reduced by more than 50% over the course of treatment, and reductions of a similar magnitude were demonstrated on self and parent reports of generalized anxiety symptoms. Furthermore, impairment (i.e., IRS) no longer reached the clinical cut-off at the end of treatment. Although Lauren’s mother’s ratings of social anxiety were below the clinical threshold at the end of the manualized treatment, as described above, Lauren’s own report did not show meaningful improvement in SAD symptoms. However, this finding stands in contrast to the multiple indicators of treatment progress, including her own ratings of total and generalized anxiety symptoms, her mother’s report, and the overall descriptions of improvement both she and her mother discussed. Thus, it is likely that the SAD subscale on the SCARED was not sufficiently sensitive to detect treatment changes. The SCARED was used as a brief and free measure of treatment outcome, which may better reflect trait rather than state anxiety.
As depicted clearly in Figure 1, Lauren appeared to experience the most benefit from exposure treatment. Although she generally showed a reduction in symptoms from other treatment modules, behavioral exposure yielded the most striking treatment changes. This fits with much research showing the large effects of exposure therapy and the unclear benefits of cognitive components of therapy (Feske & Chambless, 1995). Especially for a young child such as Lauren, who did not clearly describe anxious thinking patterns at least initially, cognitive restructuring is an abstract concept, and exposure provides a more proximal way to combat anxiety.
Biweekly sessions to enhance her gains were delivered, and with the exception of Lauren’s own report of social anxiety, all other indicators of treatment outcome demonstrated continued improvement, although at a smaller magnitude than the initial treatment gains she first experienced. Booster sessions focused on Lauren’s home life, as she and her mother discussed ways to spend quality time together and ways to consistently respond to stress related to her interactions with her new stepfamily. However, the utility of these sessions was likely limited by unclear involvement of her stepfather, marital discord, and additional, more overt child difficulties related to Lauren’s stepbrother with autism. Thus, termination of treatment was most relevant for Lauren at that time.
It is quite notable that Lauren displayed improvement in symptoms and functioning, despite treatment that, at times, did not fully map onto Lauren’s specific feared situations within the family, and yet her symptoms (at least by parent report) improved. This generalization from the situations worked on in session to more naturalistic situations is not uncommon in CBT for childhood anxiety disorders (e.g., Cummings et al., 2014). We had selected many of the exposure exercises to generally target her social discomfort with others. Although we had considered more naturalistic feared situations, such as including Lauren’s stepfather and exposing her to his reprimands, or exposure to her stepbrother’s shouting after the Coping Cat exposure sessions, she and her mother did not believe these would be necessary. Lauren’s mother also believed that she and her new husband would need to discuss further how to set up exposures for Lauren in the home setting and not accommodate her anxiety. Her mother stated that Lauren had made great progress in treatment, and she believed clear family discipline and structure were needed that did not necessarily map onto Lauren’s treatment.
Although it is clear that family stress is relevant to many families with SAD, it is often not addressed by existing manualized treatments, and previous research to understand the utility of parental involvement in CBT has not clearly shown additional benefit to individual therapy for the child (Silverman & Field, 2011). Furthermore, research on CBT for children with SAD suggests that children with SAD compared with those with other anxiety disorder may be at risk for less optimal treatment outcomes (Hudson et al., 2015; Kerns et al., 2013). However, very little research exists to explain these negative treatment outcomes. This case study exemplifies the impact of significant family stress on CBT outcomes for a young girl with SAD and suggests that attention to family functioning may be a critical focus to enhance outcomes for children with SAD. Understanding how to optimize treatment outcomes represents a critical area of research, as SAD can persist into adulthood and is associated with numerous impairments in daily life functioning and additional psychopathology (Mohatt et al., 2014).
The results of a case study clearly limit generalizability and controlled studies are needed to further evaluate the efficacy of SAD for children facing substantial family stress. Other limitations include longer than expected treatment and assessment measures that may not have been sufficiently sensitive to treatment effects. Also, other potentially meaningful treatment outcome variables, including assessments of parental psychopathology, and Lauren’s own relationship with her mother and stepfather, were not collected, although our assessment battery was fairly comprehensive and is excessive for clinical care. Furthermore, the extent of assessment at year follow-up was qualitative in nature, and empirically validated measures of functioning are needed. However, given that little is known about more complicated cases of childhood SAD, we believe this case study extends our knowledge base at this time in a meaningful direction.
12 Recommendations to Clinicians and Students
The current treatment was implemented in a CBT training clinic that was part of a child and adolescent psychiatry program. The current case provided an ideal introduction for a clinician new to the implementation of CBT, as the case conceptualization fit well with existing, commonly used manualized treatments for childhood anxiety. In addition, Lauren generally responded quite well to CBT with minimal need, at least initially, to tailor the course of treatment beyond the manualized sessions. Although CBT is now a required component of child and adolescent psychiatry fellowship training (Pearl, Mahr, & Friedberg, 2013), a cognitive-behavioral approach offers components that may be new or challenging to a psychiatry trainee, such as longer and more frequent patient interaction, tolerating negative affect, and developing a collaborative relationship (Pearl et al., 2013). Lauren’s case also presented opportunities to personalize the manualized treatment protocol. Given that Lauren was at the younger age range for conducting individual CBT, the therapists designed CBT in ways that were appropriate with Lauren’s developmental level (e.g., playing games to highlight skills rather than entirely focusing on the treatment workbook, developing and scaffolding exposures until she was able to initiate them herself). These components were a priority in supervision discussions of this case.
Lauren’s case also provided an ideal teaching opportunity to highlight the importance of psychological treatment in the interdisciplinary care of childhood anxiety disorders. As described above, Lauren’s primary initial presenting problem was frequent and impairing somatic complaints. Her treatment could have been managed, potentially by a gastroenterologist, and this was actually her mother’s first consideration in Lauren’s care. However, this treatment would likely have been insufficient to produce meaningful improvement in her symptoms and impairments, and the treatment that had been pursued initially through gastroenterology had been associated with minimal benefit. Alternatively, Lauren could have been treated through psychiatric medication management, but pharmacological intervention would not have equipped her with skills to actively manage her distress. It was believed that her difficulties were most proximally improved by CBT, and comprehensive evaluation and continued frequent assessments helped to provide clear evidence that CBT was appropriate and meaningful to her improvement. The referral for psychological treatment made by Lauren’s primary physician was integral in providing meaningful interdisciplinary care to optimize her functioning.
Altogether, these results confirm that CBT is a powerful intervention for childhood SAD that can be implemented well with minimal background in CBT. Even for a child experiencing numerous family stressors, CBT was associated with meaningful improvements in symptoms and impairment that continued over time and equipped Lauren with useful strategies to manage future difficulties. However, it was believed that family stress was a critical factor in the development as well as the treatment of SAD for Lauren. Attention to family factors, including family stress, has not been a frequent focus of studies of CBT for children with SAD. However, family stress likely exerts a negative impact on CBT and may at least partially explain the risk for poor CBT outcomes that is associated with SAD (Hudson et al., 2015; Kerns et al., 2013). This case study calls attention to the way that family stress can complicate the course of CBT and highlights the importance of considering family factors in the treatment of childhood SAD. Treatments that consider the larger family context may enhance CBT gains for children with SAD.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.