Integrating elements of teddy bear therapy into cognitive behavioral therapy for a child with obsessive – compulsive disorder: A case study

Problem: Childhood obsessive – compulsive disorder (OCD) can chronically affect functioning across a multitude of areas. Cognitive behavioral therapy (CBT) is well ‐ evidenced as an effective treatment option, however, there is less research on how CBT for OCD can best be adapted to meet the specific needs of younger children. Integrating CBT with forms of therapy that incorporate play and externalization may be particularly appropriate for this age group. However, more research is needed detailing how this could be carried out in clinical settings. Methods: This study meets this need by describing the treatment of an 8 ‐ year ‐ old boy with OCD. An evidence ‐ based CBT approach was used integrated with teddy ‐ bear therapy (TBT). This study employs a single ‐ case A – B design to explore the acceptability and benefits of using an integrated CBT/TBT treatment approach. Findings and Conclusions: A reduction in ritualistic behavior and anxiety was seen following treatment, with qualitative feedback from the client and his family showing the inclusion of TBT to be experienced as acceptable and useful. All therapy goals were met by the end of treatment, though the parental scores on the Revised Child Anxiety and Depression Scale indicated ongoing clinically significant OCD symptoms. Implications for clinical practice and future research are discussed.


| Cognitive model of OCD
The cognitive model of OCD postulates that distress occurs when an individual catastrophically misinterprets normal intrusive thoughts as significant and threatening, prompting engagement in "neutralising" compulsions ("rituals"), which result in a short-term reduction in anxiety but are hypothesized to maintain symptoms by preventing disconfirmation of the misinterpretation (Salkovskis, 1999). Research has supported this model and shown it to be applicable with children (Barrett & Healy-Farrell, 2003;Farrell et al., 2011;March & Mulle, 1998).
These thinking styles have been linked to the catastrophic misinterpretation of intrusions which are characteristic of OCD and may make it harder for younger children to gain insight into their OCD (Adelman & Lebowitz, 2012), which has been shown to be predictive of worse treatment outcomes (Garcia et al., 2010;Lewin et al., 2010).
In addition, family accommodation, where family members collude with an individual's rituals, can be a key maintaining factor amongst younger children (Francazio et al., 2016). It has been suggested this is due to the appropriate reliance of younger children on parental figures to demonstrate what is dangerous (Adelman & Lebowitz, 2012). Although research has highlighted that family accommodation is often motivated by a desire to reduce distress, and thus is reflective of the individual's support network attempting to help, it significantly predicts symptom severity and persistence (Francazio et al., 2016;Meyer et al., 2018;O'Connor et al., 2020).

| OCD treatment
There is a well-developed evidence base showing the effectiveness of cognitive behavioral therapy (CBT) in the treatment of OCD amongst young people (March & Mulle, 1998;McGuire et al., 2015), with a recent meta-analysis reviewing 34 studies showing significant symptom reduction following CBT treatment, and lower rates of relapse as compared to selective serotonin reuptake inhibitor treatment (Öst et al., 2016). Currently, National Institute for Health and Care Excellence (NICE, 2005) guidelines recommend CBT to be offered to all young people with OCD as a front-line intervention (NICE, 2005).

| CBT for younger children
Although there is an emerging evidence-base that CBT can be effective with younger children Nakatani et al., 2011), it is a less developed area of research and more studies are needed (Freeman et al., 2007). Existing research has highlighted the need for CBT to be appropriately adapted to ensure it is accessible to younger clients (Stallard, 2004) through increased involvement of parents (Albert et al., 2010), incorporating developmentally appropriate play (Choate-Summers et al., 2008), and externalization (Banting & Lloyd, 2017;Barrett et al., 2008;White & Epston, 1990).
A well-developed evidence-based evidence the effectiveness of play therapy amongst younger children (Bratton et al., 2005). Play therapy can be defined as an intervention that uses play as a medium to support children to explore challenging situations. Play therapists analyze the play, with a particular focus on patterns, and respond in a way that might allow for a change in difficult feelings (Association of Play Therapy, 2008). Play therapy might offer a more natural vehicle for children to communicate difficult feelings, thus allowing children to develop a sense of mastery over challenging situations (Homeyer & DeFrance, 2005;Schaefer, 2011). As such, incorporating aspects of play into other therapeutic approaches, such as CBT, might be particularly indicated.

| Teddy bear therapy
Integrating elements of teddy bear therapy (TBT) into CBT may be one way in which the above adaptations could be made. TBT is a form of therapeutic intervention originally developed within psychoanalytic play therapy which involves mutual storytelling between the therapist and child about a teddy facing similar difficulties to those the child faces (Beyers et al., 2017(Beyers et al., , 2018. The therapist then supports the child to make changes to the teddy's story, in turn enabling change in relation to the child and their wider system (Beyers et al., 2017(Beyers et al., , 2018Baloyi, 2006). Within TBT, change is theorized to result from enabling children to problem-solve challenging situations within a safe, playful environment (Beyers et al., 2017(Beyers et al., , 2018. The therapist serves the role of scaffolding problemsolving, at times "guiding" the story towards an effective solution, such as by introducing characters who can assist the teddy (Beyers et al., 2017).
Incorporating elements of TBT into CBT treatment could serve as a framework within which play, externalization and systemic factors can be integrated with evidence-based cognitive and behavioral techniques. Currently, there is no research published exploring whether TBT might be a useful tool to facilitate CBT treatment of OCD amongst younger children. This study looks to meet this gap by exploring the acceptability and benefits of an integrated CBT/TBT treatment approach, and thus further develop the limited evidencebase for CBT treatment of OCD amongst younger children.

| Current study aims
The current study looks to explore how TBT can be integrated into CBT treatment of OCD by describing a single case report where integrated TBT and CBT were used in the treatment of a young boy with OCD. A detailed account of how TBT was integrated into treatment is presented, thus adding to the evidence-base regarding OCD treatment in children by showing how evidence-based CBT can be implemented in an innovative way within clinical practice.
2 | PRESENTING PROBLEMS 2.1 | Referral and assessment Chris (anonymized) is an 8-year-old White boy presenting with OCD symptoms and a diagnosis of anxiety referred by his care coordinator. Before this treatment episode, Chris had received multiple therapeutic interventions, including anxiety-management work informed by CBT theory at school and hild and adolescent mental health services (CAMHS), counseling and family therapy. Chris was not receiving medication at the time of intervention. During the intervention period, his case was discussed with a psychiatrist and it was agreed that therapeutic, as opposed to medical, intervention was most indicated at present. At assessment, Chris reported experiencing scary thoughts about "poo germs" for the past 3 years. Chris reported getting very scared when he went to the bathroom and would get thoughts like "if you don't do x, then you are going to get sick." Chris reported the "scared" feeling made his heart beat fast, his tummy feel wobbly and his breathing go fast, and rated this feeling as "more than 100%" in intensity. Chris reported believing the scary thoughts "around 97%" and that he would "do what the scary thoughts said" to "not get sick." Chris's mother reported Chris engaging in rituals around his toilet use, hand-washing, and bed-time routine, which meant it could take up to 4 h to get ready for bed/use the toilet, thus significantly impacting on functioning. These rituals included reassurance asking ("am I ok," "is there poo on me"), repetitive wiping/hand-washing, checking, and getting parents to complete tasks for him (such as wiping his bottom). These behaviors limited Chris's ability to spend time with friends and were negatively impacting his and his parents' relationship. However, Chris didn't feel the impact was significant and saw his internal anxious voice as his "friend" as "it helped keep him safe."

| Developmental history and risk
Chris was born early with kidney complications, resulting in a 3-week neonatal intensive care unit (NICU) stay. As such, there was understandable parental anxiety around his urine output resulting in parents more closely monitoring his bathroom functioning. Chris's mother reported that Chris met his developmental milestones, though he had more recently begun to struggle academically.
Queries had been raised regarding whether Chris had an autistic spectrum disorder. He was awaiting a formal assessment. Chris's risk to self, to others, and from others was low. However, Chris had developed an anal prolapse due to his rituals, thus there was some risk of unintentional harm to self.

| Treatment goals
Three goals were collaboratively set with Chris and his parents to be achieved by the end of treatment: (1) Chris to be able to go to the toilet for a "poo" by himself; (2) Chris to be able to wash his hands by himself following using the toilet; (3) Chris's bedtime routine to have reduced to a manageable length (under an hour).

| METHOD
A single-case A-B design was used (N = 1). Symptoms of OCD, general anxiety, and low mood were measured at baseline, Weeks 4, 9, 14, and 19. Idiosyncratic measures of ritualistic behavior were taken weekly.

| Outcome measures
The primary outcome measures were the Revised Child Anxiety and Depression Scale (RCADS) and RCADS-parent version (RCADS-P) (Chorpita et al., 2000). These measures assess child anxiety and depression and include an OCD subscale. They have been shown to T A B L E 1 Outcome measures used, rationale, and schedule of use

Measure Schedule of use Rationale
Revised Children's Depression and Anxiety Scale is a 47-item questionnaire with separate subscales exploring different anxiety and low mood presentations (including an OCD subscale) and has a self-report and parent-report version At assessment, 1/3 and 2/3 of the way through therapy, and at the final session Well-validated and established measures for anxiety in young people (Chorpita et al., 2000(Chorpita et al., , 2005. In line with CYP-IAPT procedures to use as key outcome measure. Normed from age 8. Idiosyncratic measure of ritual time Measure taken weekly Collaboratively constructed quantitative measure to capture changes in rituals by measuring the time taken for C to get ready for bed/go to the loo on a scale of 1-5 (5 = over 2 h, 4 = 1.5-2 h, 3 = 1-1.5 h, 2 = 0.5-1 h, 1 = less than 0.5 h). Scaling to reflect natural variation that might occur across a week.
Abbreviation: CYP-IAPT, children and young person's improving access to psychological therapies programme.

HARRIS ET AL.
| 3 have high consistency, validity, and reliability, and have been normed on children 8 and older (Chorpita et al., 2005). In addition, collaboratively designed idiosyncratic measures were used to explore change on a weekly basis, which looked at time taken completing rituals (see Table 1) to supplement less regular RCADS measures.
This appeared to reduce the burden on Chris and his parents and allow them to fill in the RCADS more reflectively when it was used.
Finally, qualitative feedback was gathered via an idiosyncratic questionnaire exploring the experience of including TBT in treatment from Chris and his mother using open-ended questions (Appendix 1).

| CONCEPTUALIZATION
A cross-sectional formulation was collaboratively developed based on the cognitive model of OCD (Salkovskis, 1999). This was tested and updated through therapy using information from TBT role-play and in vivo exposure. This formulation was developed in tandem with externalization work, which allowed Chris to consider his relationship to the OCD and begin to challenge his belief that the OCD was positive. Insight into the impact of the OCD was developed particularly by considering if the OCD acted as a "friend" or a "bully," and this language was built into the shared formulation ( Figure 1).
It was hypothesized that Chris was experiencing normal intrusive thoughts around germs triggered by items he saw as related to "poo" (toilets/his bottom) (Allsopp & Williams, 1996). It was hypothesized that the OCD was "telling" Chris that these thoughts were real, resulting in the catastrophic misinterpretation that the intrusions meant he was "in danger," which resulted in Chris feeling scared (OCCWG, 2001). Chris would then engage in a number of neutralizing compulsions, reassurance-seeking, and avoidance. This caused an initial decrease in anxiety, but prevented disconfirmation of his misinterpretation, thus maintaining his fear over time. It was hypothesized limited insight into the impact of his OCD was a further maintenance factor (Lewin et al., 2010).

| Family accommodation
At assessment, it became clear that there was a large level of family accommodation of Chris's rituals, avoidance, and reassurance-seeking from his parents. This was hypothesized to be a key maintenance factor that prevented Chris from updating his "obsessions" and confirmed the "reality" of the danger (Adelman & Lebowitz, 2012;Garcia et al., 2010). An additional cross-sectional formulation was developed in collaboration with Chris's parents to explore the role of the family (Figure 2).

I get normal scary thoughts about poo and germs
The bully tells me: and behavioral experiments, as well as emotion regulation skills (see Table 2 for additional details). Sessions were attended by Chris and his mother to enable family accommodation to be tackled across the therapy. In addition, a number of separate parent sessions were held to tackle family maintenance factors.

| Teddy bear therapy
The TBT aspect of treatment was introduced to Chris in Session 4, fol-

Physical sensaƟons
Racing heart Tired Tearful F I G U R E 2 Cross-sectional formulation of family accommodation using hot-cross bun (Greenberger & Padesky, 1995) HARRIS ET AL.

| 5
The teddy was used flexibly within sessions as a foil by which Chris could be empowered to take on the task of problem-solving his own difficulties. In addition, the teddy narrative provided a medium by which Chris and myself could explore family dynamics and what Chris thought/ felt during incidents from the week. It was noticed earlier in therapy that Chris at times found it difficult to express how he was feeling. The teddy seemed to provide Chris with an externalizing tool that allowed him to express how he felt in a way that was more manageable and perhaps felt safer.
This TBT narrative was integrated with the traditional CBT for OCD techniques detailed in Table 2 in a number of ways. Firstly, the teddy's story was used as a tool to facilitate psychoeducation around OCD and anxiety. Initially, this took the form of a "wise owl" toy (played by myself) helping to teach Chris's toy about these topics. However, as the weeks progressed, Chris took on the role of educator, either by directly teaching his own toy through acting as an "OCD consultant" or by his toy providing psychoeducation to my wise owl. Secondly, TBT was used as a tool to further embed learning regarding cognitive and behavioral strategies. The TBT narrative was also incorporated into relapsemanagement work. Within the TBT narrative, Chris and his teddy were portrayed as an "OCD-fighting team" who were continuing on to their next adventure together. This allowed space for us to consider how Chris could continue his fight against OCD, whilst also recognizing how far he had come already. It also empowered him to step out of the role of "patient" and into the role of a therapist who held responsibility for the continued progress of himself and his teddy.

| RESULTS
Chris's scores on the RCADS and RCADS-P were in the clinically significant range for anxiety and OCD at baseline. Throughout treatment, Chris and his parents were encouraged to fill in measures at home at regular time-points to reduce the chance of bias. When giving qualitative feedback, Chris and his mother declined to answer away from the therapist, as they did not feel this was necessary, which may have impacted on answers given.
Hypothesis 1. To test Hypothesis 1, time spent going to bed or the toilet was compared from assessment to final session. A clear reduction is seen across therapy, with time spent going to bed reducing from over 2 h to less than an hour, and going to the toilet from over 2 h to less than 30 min, from assessment to final session (Figure 3). Qualitative feedback in the final session from parents reflected that the remaining routine at bedtime was felt to be "normal and appropriate" (i.e., consisting of bedtime story, etc.) as opposed to anxiety-motivated. These findings support the initial hypotheses that ritualistic behavior would decrease through CBT therapy. Interestingly, we see an increase in ritual time in weeks 16. It is hypothesized this might have been due to an initial anxiety spike following parents fully disengaging from rituals or contextual factors, such as increased coverage of coronavirus.

Intervention session Focus of therapy
Session structure Sessions lasted 60-90 min in length, and were attended by Chris, his mother, and at times his wider family network (father and brother). Sessions were completed both at CAMHS and at Chris's house, to support with the generalization of skills into the home environment.
1-2 Psychoeducation (around how OCD works), externalization (consideration of OCD as "the bully" and integration with the specialist interest of Minecraft) and formulation 3-4 Cognitive restructuring (using surveys, evidence-gathering, behavioral experiments, and psychoeducation around germs) Introduction to TBT (picking teddy bear, setting up initial narrative, providing rationale to parents)

5-17
Cognitive restructuring via exposure-response prevention (hierarchy made, weekly goals selected, in vivo exposure work in session and as homework) TBT (exploration of teddy's narrative, Chris teaching teddy how to overcome his difficulties) Emotion regulation skills (muscle relaxation, diaphragmatic breathing) Psychoeducation (on how anxiety works, the impact of anxiety on the body and to support emotional literacy, and continued psychoeducation on OCD) Role-play (phone consultation with Chris as "creeper expert" to embed learning around how OCD works)

18-19
Relapse prevention work (cocreation of a video with key learning, ending of teddy's story, separate parent relapse management session) Abbreviations: CAMHS, child and adolescent mental health services; OCD, obsessive-compulsive disorder; TBT, teddy bear therapy.
Hypothesis 2. To test Hypothesis 2, scores on the RCADS and RCADS-P were compared from assessment to final session on the whole measure, anxiety subscales, and OCD subscale. A significant reduction was seen across the RCADS, from 67 to 9 on total score, 62 to 8 on total anxiety, and 16 to 3 on OCD (Figure 4). This indicates a change from clinically significant OCD, and borderline significant total symptoms and anxiety, to no clinically significant symptoms. Reductions were also seen on the RCADS-P, from 72 to 56 on total score, 58-45 on total anxiety, and 18-17 on OCD ( Figure 5). Although this indicates a change from clinically significant to borderline significant total symptoms and symptoms of anxiety, parental final scores indicate remaining clinically significant symptoms of OCD (Chorpita et al., 2005). Interestingly, not much change was reflected on the RCADS-P OCD subscale, despite qualitative feedback from parents suggesting significant improvements. It is possible this score partially reflects parents' own anxiety. It may also reflect the lack of sensitivity of the measure, with ceiling effects noticed due to the severity of Chris's initial symptoms. On reflection, the inclusion of an additional OCD-specific measure may have been useful to better capture change, such as the CY-BOCS (López-Pina et al., 2015). However, overall, these findings suggest some decrease in anxiety, supporting Hypothesis 2.
Hypothesis 3. Hypothesis 3 was tested by gathering qualitative feedback from Chris and his mother regarding their experience of using TBT within the intervention. Both written and verbal feedback was given, RCADS scores what we need to do to improve"). Chris also seemed to find the teddynarrative a useful tool by which to safely explore wider systemic factors (C: "I find it easier to talk about home because we have your toy and my toy and mum's toy to talk about things"). Finally, both Chris and his mother felt that TBT allowed therapy to be more fun, with C's mother feeling this enabled difficult conversations to happen (C's mother: "It made it much nicer for [Chris] to talk about things that are difficult"). However, Chris's mother did reflect that the TBT had not been as integrated into their homework, and this made it difficult to draw on as a tool outside of therapy.

| DISCUSSION
This case study aimed to explore the acceptability and benefit of incorporating elements of TBT into the CBT treatment of a boy with OCD. Results supported the initial hypotheses, with Chris reporting ritualistic behavior and anxiety decreasing through treatment, and the inclusion of TBT qualitatively described as acceptable and beneficial. All therapy goals were met by the end of treatment, however, the parental scores indicated ongoing clinically significant OCD symptoms.
Outcomes from this case study cohere with the wider research on OCD treatment in children, further adding to the evidence that CBT can effectively reduce symptoms in young children (McGuire et al., 2015;Ost et al., 2016). The inclusion of TBT was a unique aspect of this intervention which allowed externalization and play to be integrated into evidence-based CBT practice. This seemed to support skills development and generalization of learning by providing a fun space for Chris to practice skills and teach his teddy, thus further embedding cognitive behavioral learning in a manner that allowed Chris to move from the role of client to co-therapist.
Qualitative feedback from Chris and his family suggested that the TBT was positively experienced and seemed to reflect existing literature which suggests incorporating play and externalization into OCD treatment can be useful for this population (Banting & Lloyd, 2017;Choate-Summers et al., 2008). From the position of therapist, I found using the TBT very powerful. The TBT narrative provided a medium that was developmentally appropriate and engaging for Chris, allowing us to discuss difficult topics in a way that felt safe and contained. In addition, I found it useful as a tool to enable a more collaborative approach to our CBT-work, allowing Chris to step into a position of responsibility for his own recovery through taking on the role of therapist for his teddy. This appeared to allow a shift in Chris's role from "being looked after" to "looking after," which may have helped build self-efficacy and challenge the developing family narrative of Chris as the "difficult" child.
Another key finding highlighted in this case study is the critical role of parental empowerment in resulting symptom change. The addition of parent-only psychoeducational sessions appeared to be a turning point in treatment, enabling Chris's parents to understand their established patterns of behavior with Chris from a new perspective. This change in perspective appeared to facilitate the parents in moving from accommodating the OCD to supporting Chris in fighting the OCD (Barrett et al., 2008;Lebowitz, 2019;Rivett & Buchmüller, 2017). The inclusion of the TBT approach appeared to provide a language and a metaphor that parents could adopt as they moved into this supportive approach, thus increasing confidence and allowing them to step into the role of co-therapist. However, it was noted in feedback regarding the intervention that parents did find it difficult to continue to draw on the TBT approach outside of therapy sessions. Further research exploring why this is, and how to support greater generalization of the approach to outside of sessions, is therefore needed.
This reflects existing literature highlighting the key role family accommodation plays in OCD maintenance (Francazio et al., 2016) and the importance of including parents in therapy to address this (Rosa-Alcazar et al., 2019). Empowering parents to step into roles as co-therapists seemed to help diminish power imbalances which may have resulted in a professional-dependent approach to recovery, as well as enabling a more effective therapeutic alliance between therapist and parents.
Parental involvement may be particularly important when poorer insight means clients' own motivation for change fluctuates (Lewin et al., 2010).
Supporting parents to step into the roles of co-therapists may enable longer-term recovery by giving parents the tools to support children in future relapses. On reflection, increased effort to engage parents earlier in therapy might have resulted in quicker and more significant symptom changes, and this is a key learning point from this case study. In addition, more research exploring factors underlying parental accommodation, such as parental beliefs, is needed, particularly in the current climate of COVID-19.
It is important to note that the final scores suggest remaining OCD symptoms were of clinical significance. This is not surprising given the severe nature of symptoms at assessment and Chris's limited insight into his difficulties, both of which have been highlighted as factors that can predict worse treatment outcomes (Lewin et al., 2010;Rudy et al., 2014). This case study highlights the need for more research with severe cases of OCD in younger children, to consider if alternative treatment approaches (such as intensive treatment options) may be more effective.

| Limitations
This study makes a unique contribution to the evidence-base and extends research by showcasing an innovative and novel clinical application of an evidence-based approach to OCD treatment in children. However, there are a number of limitations it is important to highlight, including its single-case design, the inclusion of only one data point at baseline, and the failure to directly compare CBT for OCD with and without TBT. This means it is unclear how applicable findings are to the general population, it is less possible to attribute symptom changes to the therapy, and it is unclear how much TBT directly increased the acceptability of the CBT intervention. Significantly more research is therefore needed to build on these findings and allow the establishment of more evidence using more robust methodologies. Specifically, research using generalizable methodologies, direct comparisons, and establishing more clear baselines is needed.

| CONCLUSION
This study explores a novel and inventive application of CBT for OCD using integrated TBT with an 8-year-old boy. Findings show reductions in anxiety and ritualistic behavior following treatment, and that the TBT was experienced as acceptable and useful.

CONFLICT OF INTERESTS
The authors declare that there are no conflict of interests.

DATA AVAILABILITY STATEMENT
The anonymized, quantitative data that support the findings of this study are available from the corresponding author upon reasonable request. The qualitative data, and original score sheets, are not available to protect confidentiality.