Introduction
Obsessive compulsive disorder (OCD) affects up to 1% of young people and can cause significant impairment (Piacentini, Bergman, Keller and McCracken, Reference Piacentini, Bergman, Keller and McCracken2003). Cognitive behaviour therapy (CBT) is the recommended psychological treatment for young people with OCD (NICE, 2005). However, CBT is not always available. Accessibility of CBT can be increased by widening methods of delivery and optimizing resources through the use of stepped care models (NICE, 2005). Self-help interventions, with minimal therapist contact, have shown promise in reducing symptoms of OCD in adults (Mataix-Cols and Marks, Reference Mataix-Cols and Marks2006; Tolin et al., Reference Tolin, Hannan, Maltby, Diefenbach, Worhunsky and Brady2007). This is an area yet to be researched with young people. This study represents the first evaluation of the feasibility and acceptability of a self-help intervention for adolescents with OCD.
Method
Inclusion criteria
Inclusion criteria included: (a) Primary ICD-10 diagnosis of OCD; (b) Children's-Yale Brown Obsessive Compulsive Scale (CY-BOCS) score of <27; (c) access to a telephone; (d) parental consent and young person assent; (e) no medication, or stable SSRI medication for 12 weeks; (f) fluency in English. Participants were excluded if they had a diagnosis of autism spectrum or psychotic disorder, substance misuse, or recognized reading difficulties (self-report).
Participants
Eight young people (five females, three males) aged 11–16 years participated. All participants were on waiting lists for outpatient treatment at Child and Adolescent Mental Health Services in South London. Seven participants were white British, whilst P2 was Portuguese. P8 lived in a single parent household. The remainder lived with both birth parents. Four participants had received previous psychotherapy. P4, P6 and P7 had received previous CBT. P5 had received psychodynamic psychotherapy. Mean duration of OCD was 3 years (range 2–6 years). Young people had been clinically assessed for co-morbid diagnoses. P1 reported a history of depression, P3 reported anger and behavioural difficulties, and P7 described pre-existing generalized anxiety. P5 was on a stable dose of fluoxetine (SSRI). All others were medication free.
Measures and materials
The Children's Yale-Brown Obsessive-Compulsive Scale [CY-BOCS] (Scahill et al., Reference Scahill, Riddle, McSwiggin-Hardin, Ort, King and Goodman1997) is a clinician-administered, semi-structured interview to assess OCD symptom severity and impairment in young people. The Children's Obsessional Compulsive Inventory [CHOCI-R] (Shafran et al., Reference Shafran, Frampton, Heyman, Reynolds, Teachman and Rachman2003) is a self-report questionnaire for young people with OCD, with a parallel form for parents. The Strengths and Difficulties Questionnaire (SDQ) is a brief questionnaire designed to measure conduct, hyperactivity, emotional symptoms, peer problems and pro-social behaviour (Goodman, Reference Goodman1999). An Acceptability Questionnaire was developed for the study to assess opinions of the self-help book.
Self-help book
Development of Breaking Free from OCD (Derisley, Heyman, Robinson and Turner, Reference Derisley, Heyman, Robinson and Turner2008) was prompted by the NICE Guideline for OCD (2005), and written by clinicians experienced in the treatment of young people with OCD. It includes three sections: Part A: Understanding your OCD; Part B: How to Recover from Your OCD; and Part C: OCD and the Bigger Picture. Throughout the book, readers are encouraged to complete worksheets designed to assist them to understand OCD and support them to complete exposure with response prevention (ERP) tasks. Each chapter ends with a section for parents who might support the young person.
Ethics
Ethical approval was obtained from Wandsworth Research Ethics Committee, The Kings College London and Oxleas Research and Development departments and Royal Holloway, University of London.
Procedure
This study used a case series methodology with A–B design. Following an interview and completion of the measures to determine suitability, all participants were assigned to a no-intervention baseline phase of 3 weeks. Following this, baseline measures were again completed and participants received the book Breaking Free from OCD, with guidelines recommending chapters to read each week. Participants received weekly telephone calls, which allowed for administration of the CY-BOCS and opportunity to assess adherence (not for therapeutic support). The ChOCI and SDQ self-report measures were given to participants and parents at three time points (pre-intervention, mid-intervention, post-intervention). Acceptability questionnaires were completed post-treatment.
Approach to analyses
Stability of the baseline and variability of the data were examined using Morley and Adams’ (1991) recommendations for visual analyses. In order to effectively fit a trend line to the data set, the shift of central location over time was calculated with the “resistant line” method (Morley and Adams, Reference Morley and Adams1991). Wilcoxon-Signed Ranks were used to compare mean CY-BOCS scores of baseline and intervention phase. The mean CY-BOCS scores, as opposed to pre- and post intervention scores, were calculated as a more reliable and conservative measure of change over time. Effect size was calculated by dividing the z value by the square root of N (the number of observations over the two time points). The pre-, mid- and post intervention scores on the CHOCI and SDQ were analysed at the group level using Friedman tests.
Results
Unfortunately P4 accessed a copy of Breaking Free from OCD prior to completion of the baseline phase; thus intervention for this participant began early, and baseline information could not be reported. The results of P4 are included where possible in the analyses.
Visual analyses of participant responses identified relatively stable baselines, with a trend towards slight worsening of symptoms. Trend in CY-BOCS scores for individual participants over each phase are displayed in Figure 1. Wilcoxon-Signed Ranks demonstrated that mean CY-BOCS scores during intervention (18.2) were significantly lower than mean CY-BOCS scores during baseline (22.4), Z = 2.12, p = .034. A moderate effect size was identified (T = 26.5, p<.05, r = −.53).
Figure 1. Linear trend for all eight participants OCD symptoms during baseline and intervention phase. Symptoms were measured weekly using the Children's Yale-Brown Obsessive Compulsive Scale (CY-BOCs). Trend lines were calculated using Morley and Adams (1991) “resistant line” method. Note: Only intervention data are available for P4 who accessed the intervention prior to a baseline being completed.
Self-report measures
The results of Friedman tests indicated there were no statistically significant differences in self-report scores across the intervention.
Acceptability of the intervention
All eight participants recruited completed the study. Seven participants completed the acceptability questionnaire, with moderately positive responses: 100% judged the book as “just right” in terms of readability; 50% would “definitely” or “probably” recommend the book to a friend; 62% “liked” the book; 71% found the book “helpful” for fighting OCD; 85% felt they had learnt about OCD; and 62% found the book had helped them to conduct an ERP task and to resist a compulsion. Young people reported they found the information provided helpful (e.g. “It explained everything I needed to know”) and “user friendly”. Aspects of the book that were less positively rated included “some of the activities”. P1 highlighted that the book had not helped with her more difficult and long standing compulsions.
Self-reported adherence to the intervention
Three out of eight participants reported reading all chapters (P1, P3 and P6). Five of the eight participants (P1, P3, P4, P6 and P8) read more than 50% of the book. All participants completed the psych-education section, and with the exception of P7, all had begun the intervention chapters. Five participants (P1, P2, P4, P6 and P8) reported clear attempts at initiating ERP by the end of the 8 weeks.
Discussion
This study evaluated the feasibility and acceptability of a CBT self-help intervention for adolescents with OCD. The lack of attrition is notable and, combined with results of the feedback questionnaire, indicates a high level of acceptability. However, the level of adherence to the self-help material was variable. This raises some questions about the feasibility of the intervention and warrants further investigation.
Results of this case series suggest that the use of Breaking Free from OCD with minimal weekly telephone contact led to some reduction (18.5%) of OCD symptoms when measured by the CY-BOCS. The reduction in mean CHOCI and SDQ self-report scores was not statistically significant. These results are consistent with adult literature, which indicates some limited effectiveness for minimally supported self-help. Participants in the self-directed condition of Tolin's (2007) adult RCT of self-directed bibliotherapy versus therapist-led CBT for OCD showed a similar (18%) reduction in mean CY-BOCS scores, whilst the therapist-led condition led to a 46% reduction in CY-BOCS symptoms. It is possible that a more therapist-directed self-help intervention would achieve more clinically significant results. This is the recommended next step in evaluating the use of self-help for young people with OCD.
Adherence to the self-help material likely influenced the effectiveness of the intervention. The five participants who showed some improvement (P1, P2, P3, P4, P8) reported having read at least 50% of the book (P1, P3, P4, P8) and/or had made clear attempts at ERP (P1, P2, P4, P8). Over half the participants did not complete the workbook, therefore finding some change in CY-BOCS symptom severity with a very “light touch” intervention is promising. Had the intervention been further supported, or carried out over a greater time period, it is possible that participants may have covered more of the material, which may have benefited them further.
A number of limitations require acknowledgement. There were no concurrent controls for the natural passage of time and extraneous factors that may have influenced outcome (although baseline and repeated measurement improved internal validity). Neither could the possibility of the book working as a placebo be ruled out. The representativeness of clients seen in clinical practice was strengthened by the use of few exclusion criteria, but the sample was small, and participants were predominantly white British and from intact families. The sample could also be seen as having more complex and severe problems than the populations for which self-help interventions are recommended. Treatment naive participants and/or participants with milder symptoms may have responded differently to this low intensity intervention. Finally, measurements were not taken by an independent rater. Future research should address these limitations.
This has been the first study to examine the use of bibliotherapy for adolescents with OCD. It has contributed to the evidence base and highlighted some potential strengths and limitations of self-help for this client group. Given the prevalence of OCD and the need for accessible interventions, it would be beneficial to explore various levels of therapeutic guidance that might improve effectiveness of the book whilst maintaining the intervention as accessible and cost effective.
Disclosure of interests
Three of the authors were co-authors of the self-help book.
Acknowledgements
With grateful thanks to the patients and staff at CAMHS services in South London and Maudsley NHS Foundation Trust and Oxleas NHS Foundation Trust.
Comments
No Comments have been published for this article.