The Psychometric Properties of Children’s Yale-Brown Obsessive-Compulsive Scale in Tic Disorders

Background: Patients with Tic Disorders (TD) usually comorbid with Obsessive-compulsive disorder (OCD). The severity of obsessive-compulsive symptoms (OCS) in TD is widely evaluated via the Children’s Yale-Brown Obsessive-Compulsive Scale (CY-BOCS). However, the reliability and validity of CY-BOCS in TD patients are still unclear. Methods: In this study, we examined the psychometric properties of the Chinese version of the CY-BOCS in patients with tic disorders. Results: We examined the psychometric properties of the CY-BOCS in 367 patients with TD in China. We performed reliability and validity analyses on the CY-BOCS and found that the Cronbach’s Alpha and the test-retest reliability of CY-BOCS were 0.809 and 0.821 respectively. The CY-BOCS showed good consistency in the two-factor structure, namely the “obsession” and the “compulsive” factors (the comparative t index was 0.84). Conclusion: The CY-BOCS has good psychometric properties in children and adolescents with TD. It suggests that the CY-BOCS could be widely used in Chinese children and adolescents with TD, especially for the TS patients.

Chinese children and adolescents with TD. Further pieces of evidence are needed to address these issues.
Moreover, the east and west have shown cultural differences in many areas. For example, compared with Westerners, Chinese people prefer to be more concealing their feelings and less express inner feelings [23][24][25].
Therefore, this study aims to examine the reliability and validity of the CY-BOCS in patients with TD. We calculated the test-retest reliability, Cronbach's Alpha, as well as the concurrent validation. Moreover, we used con rmed factor analysis (CFA) to test its structure of CY-BOCS.

Participants
All the participants were recruited from the Department of Psychiatry in Beijing Children's Hospital in China from October 1, 2019, to January 1, 2020. Both inclusion and exclusion criteria were below: aged from 5 to 16 years, met the tic disorders diagnostic criteria according to DSM-5,no epilepsy, or any known comorbid brain medical conditions. TS is characterized by the presence of multiple motor tics and one or more vocal tics (not necessarily concurrently), which persists for at least one year [26]. Persistent (chronic) motor or vocal TD is characterized by the presence of single or multiple motor tics only or vocal tics only, which persist for at least one year [26]. Provisional TD is characterized by single or multiple motors with or without vocal tics that persists for less than one year [26].
The ethics committees approved this study of Beijing Children's Hospital of Capital Medical University, and we also obtained the written informed consent from the parent or guardian of the enrolled children and adolescents. Finally, we enrolled 367 children and adolescents diagnosed with TD. Furthermore, the 152 patients were diagnosed with TS based on the Statistical Manual of Mental Disorders' diagnostic criteria, fth Edition (DSM-5).
The total sample was 367 patients diagnosed with TD, with the percentage of male 79.8%. The mean age of included patients was 9.21 ± 2.06 years old (range 5-16 years old), and the mean duration of illness was 1.93 ± 1.72 years. And the 152 TS patients with male/female ratio 4.85 (126/26) were included. The mean age of TS patients was 9.46 ± 2.18 years old, and the mean duration of illness was 2.67 ± 1.65 years.
Age is an important factor for tics in the previous study. For example, the onset age of TD is typically 3 to 10 years; the highest estimated prevalence rates is 7 to 11 years; tic severity declines with age during adolescence; the age of most patients detected the premonitory urges of tics is above 10 years old; ticrelated OCD tends to have an earlier age of onset [3,27,28]. Therefore, we want to verify any differences in OCS between different age groups. The patients were classi ed to the "Younger group" (aged < 10 years, n = 284) and the "Older group" (aged > = 10 years, n = 83) to assess the validity and reliability of the CY-BOCS. We used a t-test to compare the differences between the younger and older groups to study age effects.
TS has higher tic severity and comorbidity score (compulsion, ordering, ADHD, anxiety, self-injurious behavior) than the chronic motor TD [29], and we want to know more about the characteristics of OCS in TS patients. The CY-BOCS has the same structure and items as the Y-BOCS. The Y-BOCS has been translated into Chinese and tested the psychometric properties [21]. Therefore, two translators translated the CY-BOCS to Chinese independently based on the Chinese version Y-BOCS. Two bilingual experts were invited to check the translated version and nish the back-translated version. The translated version was then modi ed until it was comparable with the original English version [30].

The Yale Global Tic Severity Scale (YGTSS)
The YGTSS is a semi-structured scale and rated by a clinician or trained interviewer. It is developed for assessing the tics observed within l week before the assessment [31]. Five dimensions included in the YGTSS are as follows: the number, frequency, intensity, complexity, and interference. The total score of YGTSS (range: 0-100) is derived by summing the tic severity ranged between 0 and 50 (motor tics range = 0-25 and vocal tics range = 0-25) and the impairment rating score (range = 0-50). The YGTSS is a widely used scale with excellent reliability and validity for assessing children and adolescents with TD

Premonitory Urge for Tics Scale (PUTS)
The PUTS is a nine-item scale for measuring premonitory sensations in patients with tics. Each item is rated using a 4-point scale (numerical value range = 1-4), and the total score ranged from 9 to 36 is yielded by summing the nine items-the higher scores of the PUTS, the greater severity of premonitory urges.
Besides, the three raters (Liping Yu, Fang Wen and Jingran Liu) nished the clinical assessments. The consistency coe cient between these raters was 0.873. All the participants were outpatients. The assessments were performed after diagnosis in the Psychiatric Department's psychology assessment room in Beijing Children's Hospital by one of the two psychiatrists. Furthermore, 30 participants were randomly selected for retesting one month later to verify the test-retest reliability of the CY-BOCS.

Statistical analysis
We used the Statistical Package for the Social Sciences for Windows (SPSS Inc., Chicago, IL, USA, v25.0) to perform the statistical analyses. First, we used Pearson's correlation to examine the item-total correlation of CY-BOCS. Second, we used Cronbach's alpha/test−retest reliability to test the reliability of CY-BOCS and assess its validation by calculating the correlation coe cient between the CY-BOCS scores and the total score of YGTSS as well as PUTS. Finally, we performed the CFA of CY-BOCS in R (version 3.5.3) using the package "Lavaan". Table 1 shows the mean and standard deviation (SD) of CY-BOCS, the obsession subscale score, and the compulsion subscale score for the different age groups. Signi cant differences between the younger and older groups were identi ed in CY-BOCS total score, the obsession subscale score, and the compulsion subscale score (p < 0.05). The score of the older group was higher in the CY-BOCS total score (p = 0.000), the obsession subscale score (p = 0.002), and the compulsion subscale score (p = 0.001) than the younger group (Table 1). In addition, the mean duration of illness for younger group was 1.68 ± 1.52 years, which was shorter than the older group (2.76 ± 2.06 years, p < 0.001).

Item correlations with the total score
We examined each item score's correlation and the total score of the CY-BOCS in patients with TD. All items, obsession score, compulsion score was signi cantly correlated with the total CY-BOCS score. The correlation between the total score of CY-BOCS and the obsession score, compulsion score, O1, O4, O5 was 0.827, 0.711, 0.716, 0.792, and 0.773 respectively (more details see Table 2).

Concurrent Validation of CY-BOCS with YGTSS and PUTS/Construct validity
We evaluate the concurrent validation of the CY-BOCS based on the correlation of the total tic severity, impairment, total YGTSS score, and PUTS with the obsession score, the compulsion score, and the total CY-BOCS score. The total CY-BOCS score was signi cant correlated with the total YGTSS score and the PUTS score (Table 3).

Con rmatory factor analysis of CY-BOCS
We calculated the construct validity of the CY-BOCS by the con rmatory factor analysis. The comparative t index (CFI) was 0.84 (Fig. 1).

Discussion
In the present study, the CY-BOCS showed good consistency in the two-factor structure. The "obsession" and the "compulsive" factors perfectly measured the OCS symptoms in Chinese children and adolescents with TD. The results suggested good reliability and validity for the CY-BOCS in patients with TD. The CFA results showed that the two factors structure of CY-BOCS was "stable" in patients with TD. Overall, CY-BOCS showed good psychometric properties in Chinses children with TD.
Some studies demonstrated ne psychometric properties of CY-BOCS in the assessment for the patients with OCD [11,12]. For instance, the internal consistency was high and had a good interrater agreement for subscale and total OCD [33]. However, few studies investigated the psychometric properties of CY-BOCS in patients with TD. We provide the evidence for the reliability and validity of CY-BOCS, which bene ts future studies for the TD with OCS.
TD and OCS are closely related and overlapped in some respects. For example, the altered corticostriato-thalamo-cortical circuits and dysregulations of the neurotransmitter systems (mainly serotonin and dopamine) are involved in the mechanism of OCS and TD [34][35][36][37]. Moreover, the genetic correlation between OCD and TD was higher than between OCD and ADHD [38]. Due to the close relationship between the OCS and TD, assessing OCS in patients with TD is an important issue in future studies [5].
Another aspect to consider is that the total score of Y-BOCS does not represent the severity of symptoms [39], and the total score of CY-BOCS served as an indicator for the severity of OCS may underestimate the degree [40]. For instance, if a child mainly suffers obsessions, the compulsion subscale scores could bring down the total score. Therefore, we should pay attention to both the total score of CY-BOCS and the subscale score. In this present study, we found a con rmed correlation between the obsessions/compulsion subscale and YGTSS and PUTS via the analysis of concurrent validation of CY-BOCS with YGTSS and PUTS. Therefore, we should pay close attention to the factor scores instead of merely the total score of the CY-BOCS applied in TD patients.
In this present study, the Obsessive-Compulsive Scale (OCS) severity was signi cantly higher in the older group than the younger group. This difference was probably due to the duration of TD in the older group being longer than the younger group. Notably, it was reported that the OCS often appeared at two years after the onset of tic symptoms [41]. Overall, age has an important in uence on the severity of OCS in TD patients. Also, few studies provide evidence of the reliability and validity for the CY-BOCS in young children aged less than 8 years [9,42]. In this study, we found that the CY-BOCS is a reliable and validated scale for assessing the symptom severity in patients with TD aged 5-8 years old. Larger sample studies are warranted to validate the CY-BOCS in the young group.
Several limitations should be clari ed in future studies. First, the YGTSS and PUTS were used as the concurrent validity measures, while other scales used to assess obsession and compulsion might be more suitable. Second, the sample only includes the children sample, but not the whole age range, such as the adult sample.

Conclusion
The CY-BOCS has good psychometric properties in children and adolescents with TD in our study. Thus, we suggest that the CY-BOCS could be widely used in Chinese children and adolescents with TD, especially for the TS patients.

Declarations
Ethics approval and consent to participate The ethics committees approved this study of Beijing Children's Hospital of Capital Medical University.

Consent for publication
We also obtained the written informed consent from the parent or guardian of the enrolled children and adolescents.
Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
None of the authors has any con icts of interest to disclose.

Competing interests
All of the authors declare that they have no competing interests.  Con rmatory factor analysis of CY-BOCS