1.1. Comorbidity
There has been considerable interest over the last 20 years in the causes and maintaining mechanisms of obsessive-compulsive disorder (OCD) and its association with other psychiatric disorders. There is an emerging consensus in the clinical literature that there are a range of genetic and environmental risk factors for OCD, which may be shared with other disorders (Pauls, Abramovitch, Rauch, & Geller, 2014) and estimates suggest that over 50% of children with OCD also meet criteria for other psychiatric disorders (Geller et al., 2003). These include other anxiety disorders (55%; Franklin et al., 2011), attention deficit hyperactivity disorder (18%; Geller et al., 2003) and tic disorders (TD), including Tourette syndrome. Estimates of the prevalence of TD amongst OCD populations vary, with figures of between 9 and 59% reported in the literature (Stewart et al., 2004). Similarly, OCD is over-represented amongst children with TD, with estimates suggesting over 50% of children with TD also meet criteria for OCD (Lebowitz et al., 2012). This relationship has been well established in both the adult and child clinical literature and has recently been recognised in diagnostic classification systems, with the addition of a “tic-related” specifier to the obsessive compulsive disorders in the Diagnostic and Statistical Manual of Mental Disorders, 5th Edition (American Psychiatric Association, 2013).
1.2. Clinical features & phenomenology
In recent years the clinical research literature has highlighted differences in the phenotypic expression of OCD in adults with no tic history compared to those with comorbid tic disorders (OCD + TD). It is now well established that there is a greater proportion of males and earlier OCD symptom onset amongst those with a comorbid tic disorder compared to those without tics (Leckman et al., 1994). Adults with OCD + TD typically endorse higher rates of violent and sexual thoughts and imagery, counting rituals, hoarding, intrusive sounds and tic-like compulsions (Cath et al., 2001 and Diniz et al., 2006). Furthermore, OCD + TD individuals are more likely to meet criteria for a range of other comorbid conditions, including other anxiety disorders, attention deficit hyperactivity disorder, trichotillomania and body dysmorphic disorder (Diniz et al., 2006).
Such questions have received far less research attention in the paediatric literature and findings are inconsistent. As in adults, some studies find a higher proportion of males with OCD + TD in comparison to OCD alone (e.g. Storch et al., 2008 and Zohar et al., 1997), although some find no gender differences (Conelea et al., 2014, Hanna et al., 2002 and Scahill et al., 2003). Scahill et al. (2003) found a trend towards greater levels of externalising symptoms and significantly greater levels of attention problems in those with OCD + TD. Similarly Lewin, Chang, McCracken, McQueen and Piacentini (2010) found higher rates of Oppositional Defiant Disorder and ADHD in those with OCD + TD compared to OCD alone, although no group differences in overall rates of comorbidities. Conelea et al. (2014) also found no group differences in comorbidity rates.
Most studies find no difference in overall OCD severity in those with and without TD (Conelea et al., 2014, Hanna et al., 2002, Storch et al., 2008 and Zohar et al., 1997). Some studies report fewer of some types of obsessions in OCD + TD groups compared to OCD alone (overall – Lewin et al., 2010; contamination and sexual obsessions – Storch et al., 2008; religious – Conelea et al., 2014); whereas some find more (e.g. aggressive and sexual images; Zohar et al., 1997) and others find no differences (Hanna et al., 2002). Young people with OCD + TD may also be less likely to be able to identify triggers for obsessions (Leckman et al., 1994). Some studies find more compulsions in the comorbid TD groups (e.g. washing and ordering – Conelea et al., 2014; repetitive routine behaviours, ordering and arranging, unrelated to harm avoidance; Scahill et al., 2003; counting - Storch, 2008; harm-avoidance – Zohar et al., 1997) whereas others find fewer (ordering, hoarding, washing – Hanna et al., 2002; contamination, washing, reassurance seeking – Scahill et al., 2003).
1.3. Evidence for treatment of OCD
The typical natural course of tics is to peak in late childhood and enter remission during adolescence (Leckman, 2003). In contrast, OCD symptoms continue to increase in adolescence, typically peaking 2 years after the greatest severity of tics (Kichuk et al., 2013). Furthermore, OCD related difficulties are more likely to persist into adulthood (Bloch et al., 2006). Given this relationship, prioritising early treatment of OCD is particularly important in those with OCD + TD as it may prevent persistence to adulthood.
There has been a significant amount of research on the effectiveness of treatments for OCD in young people. Two treatments have been shown to be efficacious, namely cognitive behaviour therapy (CBT) incorporating exposure and response prevention (ERP) and serotonin re-uptake inhibitor/selective serotonin re-uptake inhibitors (SRI/SSRI) medications. Two meta-analyses (Abramowitz et al., 2006 and Watson and Rees, 2008) have demonstrated that CBT/ERP demonstrates significantly improved outcomes over pharmacotherapy, which is itself more effective than placebo. Meanwhile, other studies (Pediatric OCD Treatment Study, 2004) suggest that combined treatment modalities offer the best overall treatment response. CBT is often recommended as a first line treatment for paediatric OCD (e.g. National Institute for Clinical Excellence, 2005).
1.4. Treatment of OCD in the presence of tic disorders
Whilst there is good evidence supporting the treatment of childhood OCD, the degree to which these treatments are effective for OCD + TD groups is less clear. Whilst individuals with TD are often included in randomised controlled trials (RCTs) examining the effectiveness of CBT for paediatric OCD, their small number commonly precludes examination of any differences in their response to treatment, or differences in the treatment received. It has been suggested that young people with OCD + TD may require longer or more sessions of ERP to increase practice due to elevated ‘just right’ compulsions, or more home-based practice if obsessions cannot be elicited in the therapy room due to an inability to identify specific triggers (Mansueto & Keuler, 2005).
Given the high rates of comorbidity between OCD and TD, there is a relative paucity of studies examining the effectiveness of first line treatments in the OCD + TD population. A review of the literature identified only six studies examining possible moderating effects of a TD on OCD treatment (Ginsburg, Kingery, Drake, & Grados, 2008). Of these six studies, three focussed on medication, two on CBT and one a combination study. Amongst the medication studies the results were somewhat mixed, but overall suggest that comorbid tic disorders were associated with poorer SSRI treatment response. In contrast, neither of the two CBT studies found any impact of a tic disorder on the effectiveness of either individual (n = 14 with comorbid TD; Piacentini, Bergman, Jacobs, McCracken, & Kretchman, 2002) or group OCD treatment (n = 8 with comorbid TD; Himle, Fischer, Van Etten, Janeck, & Hanna, 2003). Interestingly, the final study (n = 17 with comorbid TD: CBT n = 3, CBT + medication n = 4, placebo n = 5, medication n = 5; March et al., 2007) examined the effectiveness of both medication and CBT on OCD treatment amongst children with OCD + TD. These results demonstrated that sertraline use was equivalent to placebo amongst individuals with comorbid tic disorders, but CBT remained effective irrespective of the presence of tics. More recently, Conelea et al. (2014) in the US similarly examined the treatment outcome of CBT for OCD alongside medication (SRI) in 124 children (n = 66 with comorbid tics: medication management n = 25, medication management plus instructions in CBT n = 20, medication management + CBT n = 21). The study found no differences in OCD severity, impairment or comorbidity, nor in treatment outcome, in the group of children with tics compared to children without. No CBT studies explicitly mention modification to the CBT intervention to account for the presence of tics and the CBT delivered was the same as that delivered to young people without comorbid tics.
Of note, the aforementioned studies are restricted by small sample sizes and they may therefore have been underpowered to detect possible effects of comorbid TD on response to CBT for OCD. They are also limited by site differences (Conelea et al., 2014 and March et al., 2007). Furthermore, previous studies included different formats of intervention, including both individual and group CBT protocols, with and without medication, making comparisons between studies difficult. Finally, none of the studies focus on UK samples, which limits generalisability to a UK population. The reduced SSRI response rates and higher relapse rates in OCD + TD also strengthens the need to further investigate the effectiveness of traditional CBT amongst those with comorbid tic disorders.
1.5. Summary & aim
In summary, there is considerable evidence demonstrating a high prevalence of comorbidity between OCD and TD and, moreover, that OCD + TD may form a distinct subset of the wider OCD population. Furthermore, evidence to date suggests that traditional pharmacotherapy is less effective in those with a comorbid tic disorder. There is a striking lack of research investigating the effectiveness of traditional CBT/ERP in this population, with previous studies being limited by small sample sizes. Given these factors, the aim of the current study was to evaluate the effectiveness of a standardised, developmentally sensitive OCD treatment protocol amongst individuals with OCD + TD, compared to a matched group of children with OCD and no tics.
3. Results
3.1. Group comparison of OCD symptom profile as baseline
3.1.1. Group differences in baseline obsessions and compulsions scores
There was no significant main effect of group on CY-BOCS obsessions and compulsions scores, F(1,55) = .24, p = .62. There was a significant effect of symptom type F(1,55) = 10.68, p = .002, with greater scores for compulsions in comparison to obsessions. There was no significant interaction effect F(2,55) = 1.08, p = .30.
3.1.2. Group differences in categories of obsessions and compulsions at baseline
Frequencies of the categories of obsessions and compulsions outlined in the CY-BOCS were available for all of the OCD + TD group (n = 29) and all but one of the OCD-TD group (n = 28). a greater proportion of participants in the OCD-TD reported (24/28) cleaning compulsions in comparison to the OCD + TD group (16/29), however this difference was not significant following correction for multiple comparisons. There were no significant differences in any other compulsion category (cleaning; checking; repeating; counting; ordering/arranging; hoarding/saving; games/superstitions; rituals involving others; mental rituals; need to tell/ask/confess; measures to prevent harm; ritualized eating; list-making; touching/tapping/rubbing; blinking/staring; ‘just right’; trichotillomania; self-harming; miscellaneous), nor in categories of obsessions reported (contamination; aggressive; sexual; hoarding/saving; magical thoughts/superstitious; somatic; religious; needing to know/remember; fear of saying certain things; fear of not saying the right thing; intrusive-non-violent images; intrusive sounds, words; music or numbers; and miscellaneous obsessions). Table 2 and Table 3 present the comparisons for each obsession and compulsion category, respectively.
Table 2.
Chi-squared comparisons between OCD + TD and OCD-TD groups for each obsession category.
OCD + TD (n present of 29) OCD-TD (n present of 29) χ2 (p)*
Contamination 19 24 3.14 (.08)
Aggressive 26 21 (.18)
Sexual 9 6 .68 (.41)
Hoarding 10 6 1.2 (.27)
Magical/Superstitious 9 13 1.42 (.23)
Somatic 8 11 .88 (.35)
Religious 10 13 .85 (.36)
Need to know/remember 6 4 .40 (.53)
Fear of saying certain things 9 3 3.54 (.06)
Fear of not saying the right thing 3 4 (.71)
Intrusive (non-violent) image 4 0 (.11)
Intrusive sounds, words, music or numbers 3 1 (.61)
Miscellaneous 5 2 (.42)
Table options
Table 3.
Chi-squared comparisons between OCD + TD and OCD-TD groups for each compulsion category.
OCD + TD (n present of 29) OCD-TD (n present of 29) χ2 (p)*
Cleaning 16 24 6.35 (.01)ns
Checking 22 19 .45 (.50)
Repeating 21 14 3.02 (.08)
Counting 16 12 .86 (.35)
Ordering/Arranging 17 13 .85 (.36)
Hoarding/Saving 10 6 1.20 (.27)
Games/Superstitious 8 10 .44 (.51)
Rituals involving others 22 23 .34 (.56)
Mental Rituals 14 9 1.54 (.22)
Need to tell/ask/confess 6 6 .01 (.95)
Measures to prevent harm 2 5 (.25)
Ritualised eating 2 8 (.04)
Touching/Tapping/Rubbing 13 10 .49 (.48)
Blinking/Staring 9 4 2.27 (.13)
Just right 9 13 1.42 (.23)
Trichotillomania 1 0 .98 (.32)
Self-harm 1 1 .001 (.98)
Other 3 3 .002 (.96)
ns Not significant following correction for multiple comparisons.
*Fisher's exact test is reported where any expected cell count <5. This does not compute a test statistic and therefore only p values are reported.
Table options
3.2. Group comparison of treatment received
3.2.1. Medication usage
At assessment, 8 out of 29 young people in the OCD + TD group and 5 out of 29 in the OCD-TD group were using OCD targeted SSRI medication; these proportions were not significantly different (p = .51).
3.2.2. Number of sessions
Information on number of CBT sessions received was available for 28 of the OCD-TD group and 25 of the OCD + TD group. Young people in the OCD no tics group had a significantly greater number of CBT sessions (M = 15.71, SD = 3.13) than did the OCD + TD group (M = 13.56, SD = 3.85) (Z = −2.20, p = .03).
3.3. Group differences in pre- and post-treatment CY-BOCS scores
Table 4 provides mean CY-BOCS scores at pre-intervention, post-intervention and follow-up time-points. There was no significant main effect of Group on CY-BOCS scores, F(1,56) = 1.57, p = .22. As expected, there was a significant main effect of Time, with scores decreasing between baseline and post-treatment F(1,56) = 147.55, p < .001. Importantly, there was no significant interaction effect between Group and Time F(1,56) = 1.54, p = .22, indicating that the two groups responded equally well over the acute phase of treatment.
Table 4.
Pre, post and follow-up CY-BOCS scores in the OCD + TD and OCD-TD groups.
OCD + TD OCD-TD
Mean pre-intervention CY-BOCS (SD) 24.76 (4.79) 24.86 (4.44)
Mean post-intervention CY-BOCS (SD) 11 (7.65) 13.66 (5.52)
Mean follow-up CY-BOCS (SD) 10.43 (7.82) 11 (7.66)
Table options
3.4. Group differences in pre- and post-treatment and follow-up CY-BOCS scores
The same pattern of results was observed when follow-up data were included in the analysis (Table 4). There was no significant main effect of Group on CY-BOCS scores (F(1,47) = .68, p = .42), but there was a significant main effect of Time, with scores decreasing from baseline through to follow-up, F(1,47) = .124.59, p < .001. Again, there was no significant interaction effect between Group and Time, F(1,47) = .01, p = .95, indicating that the two groups respond equally well to treatment in the short-term and at follow-up. Fig. 1 demonstrates that the OCD + TD and the OCD-TD groups responded similarly to intervention across time.
Full-size image (12 K)
Fig. 1.
CY-BOCS scores at pre-intervention, post-treatment and follow-up, for OCD + TD and OCD-TD groups.
Figure options
3.5. Response and remission rates
21 of 29 participants in both groups responded to treatment post-intervention (72%; p > .99). At follow up, 17 of 21 of the OCD + TD group (81%) and 23 of 28 of the OCD-TD group (82%) responded (p > .99).
18 of 29 of the young people in OCD + TD group (62%) and 14 out of 29 in the OCD-TD group had remitted symptoms post-intervention (48%). These proportions are not significantly different (p = .344). At follow-up (3 or 6 months), 13 of 21 in the OCD + TD group (62%) and 18 of 28 in the OCD-TD group (64%) had remitted symptoms; again, these proportions are not significantly different (p > .99).
