Tourette’s disorder and the chronic tic disorders (CTDs) are common, heritable, neurological conditions occurring in 1.5% to 3% of youth (Kurlan et al. 2001; Scahill et al. 2009) with males more commonly affected than females (i.e., 4:1; Freeman et al. 2000; Kadesjo and Gillberg 2000; Mol Debes et al. 2008; Roessner et al. 2007). CTDs are characterized by tics, which are sudden, repetitive movements or vocalizations, with usual onset between 5 and 7 years of age and with peak symptom severity occurring between the ages of 10 and 12 years. For most youth, tic symptom severity and related impairment decline in late adolescence and early adulthood (Bloch and Leckman 2009; Coffey et al. 2005).

Population-based studies and clinical samples consistently show that youth with CTDs often have co-occurring psychiatric conditions that contribute substantially to impairment. Population-based studies and clinically ascertained samples suggest that attention-deficit/hyperactivity disorder (ADHD) is the most common co-occurring condition: affecting 35% of youth with CTDs in community samples (Khalifa and von Knorring 2006) and up to 60% in large clinical samples (Scahill et al. 2009). These studies also suggest that obsessive-compulsive disorder (OCD), non-OCD anxiety disorders, major depressive disorder, as well as other externalizing disorders, while often present, are less common than ADHD. Given the varying methodologies for ascertainment and assessment, rates of co-occurring conditions in youth with CTDs vary considerably in population-based epidemiological studies (Khalifa and von Knorring 2006), clinical registries (Freeman et al. 2000; Roessner et al. 2007), and large population-based surveys of identified cases (Scahill et al. 2009). In clinical populations, rates of co-occurring conditions may vary by provider specialty (i.e., neurology, psychiatry, or primary care) or by age group (i.e., child or adult). Also the profile of co-occurring conditions may depend on whether families of children with CTDs seek assessment and treatment for the tics themselves or for co-occurring conditions.

Rates of co-occurring conditions in children with CTDs enrolled in randomized controlled trials (RCTs) may provide an estimate of co-occurring conditions in youth seeking treatment to reduce tic severity. However, most published RCTs have focused on medication interventions and have reported rates based on small, highly selected and/or not well-characterized samples. As a result, the rates of co-occurring psychiatric disorders in those seeking treatment to reduce tic severity are also variable (Gilbert et al. 2004; Nicolson et al. 2005; Sallee et al. 2000; Scahill et al. 2003; Silver et al. 2001; Toren et al. 2005).

Although most youth with CTDs seeking care have mild tics, which do not require treatment (Scahill et al. 2009), their lack of tic severity does not mean that they are not otherwise impaired by co-occurring conditions. Thus, clarifying the rates of co-occurring conditions in youth with CTDs has important implications for diagnosis and treatment. Specifically, clinicians treating children with CTDs need to know the type and prevalence of co-occurring conditions to assure that diagnoses and associated impairment are not overlooked. This is especially critical as there are increasing numbers of evidenced based treatments for the co-occurring conditions commonly presenting in youth with CTDs (Allen et al. 2005; Jensen et al. 2001; Kurlan 2002; Pediatric OCD Treatment Study [POTS] Team 2004; Scahill et al. 2001; Treatment for Adolescents with Depression Study [TADS] Team 2004; Walkup et al. 2008).

Although most treatment seeking youth with CTDs have mild tic symptoms, some youth report being impaired by their tics. Given emerging evidence that particular aspects of tic severity may contribute to impairment (Boudjouk et al. 2000; Himle et al. 2007; Storch et al. 2007; Woods 2002; Woods et al. 1999), it may be useful for clinicians to understand the relationship between particular aspects of tic severity (e.g., intensity, complexity, and interference) and impairment in youth with CTDs seeking treatment to reduce tic severity.

The current study examined the clinical characteristics and rates of co-occurring psychiatric conditions in a large, well-characterized sample of children and adolescents with CTDs enrolled in a RCT to reduce tic severity. A secondary aim was to describe the pharmacological treatments for tics and co-occurring conditions in this sample. Exploratory aims were to: (a) examine relationships among gender, age, co-occurring conditions, and tic symptom severity; and (b) replicate prior research which has evaluated the relative contributions of co-occurring conditions and tic severity to impairment (Himle et al. 2007).

Method

Design

Participant’s ages 9–17 years were recruited to participate in the two-phase, National Institute of Mental Health-funded Comprehensive Behavioral Intervention for Tics (CBIT) randomized controlled trial (RCT; Piacentini et al. 2010) designed to compare the efficacy of a comprehensive behavioral intervention to reduce tic severity to a supportive therapy and psychoeducation control. Following screening and clinical evaluation, participants were randomized (1:1) and began the initial phase of the study (10 weeks). Treatment responders at week 10 were followed naturalistically during the second 6-month follow-up phase. Additional design information can be found in Piacentini et al. (2010). The current study is an examination of the pre-treatment, baseline clinical characteristics of participants prior to randomization.

Data collection was conducted at three academic institutions (i.e., the Johns Hopkins University, the University of Wisconsin–Milwaukee, and the University of California, Los Angeles). Four collaborating institutions (i.e., Yale University, Massachusetts General Hospital/Harvard Medical School, and the University of Texas Health Sciences Center at San Antonio) provided administrative support, data management/analysis, rater training, and quality assurance. The Tourette’s Syndrome Association (TSA) provided administrative support and assistance with recruitment. A data safety monitoring board was established to independently oversee the study. Institutional review board (IRB) approval was obtained at each participating site.

Participants

Participants were recruited from various sources (i.e., clinics, primary care physicians, mental health providers, schools, churches, community organizations, paid/public service notices, and ads in local media) during a 30-month period (December 2004—May 2007). Outreach efforts were dedicated to enhance minority enrollment. A total of 178 children and adolescents were screened for participation yielding a sample (N = 126) with a primary CTD diagnosis that met all study inclusion criteria. Enrollment was equally distributed across the three enrolling sites.

To be eligible, participants had to have a diagnosis of Tourette’s or chronic motor or vocal tic disorder; Yale Global Tic Severity Scale scores of at least 13 (≥9 for participants with only motor or vocal tics); fluency in the English language; and average or above average intellectual functioning on the Wechsler Abbreviated Scale of Intelligence (Standard Score >80). Children with co-occurring psychiatric conditions—with the exception of psychosis, mania, and substance abuse/dependence—were included in the study unless the condition required immediate treatment or change in current treatment. Seventeen (n = 17) potential participants were excluded due to the presence of a primary psychiatric disorder other than a CTD requiring immediate treatment: five (n = 5) with OCD, four (n = 4) with ADHD, four (n = 4) with multiple co-occurring disorders, two (n = 2) with a non-OCD anxiety disorder, and two (n = 2) with a mood disorder. Participants with a stable psychotropic medication regimen for tics and/or co-occurring conditions (i.e., at least 6 weeks) were included; however, individuals with an unstable medication regimen, or those who had previously received at least four sessions of habit reversal training (HRT), were excluded from participation.

Assessment

Participants completed a demographics form. Medication status was established via a medication history form. Diagnostic status was systematically evaluated by a team led by well-trained Master’s or Doctoral-level evaluators using semi-structured interviews and validated assessment measures. The following measures were administered as part of a broader assessment battery.

The presence of tic disorders as well as co-occurring psychiatric conditions were evaluated via the Anxiety Disorders Interview Schedule-Research Lifetime Version (ADIS-RLV; Silverman and Albano 2002). The ADIS-RLV is a semi-structured diagnostic interview which was adapted from the ADIS-IV (Silverman and Albano 1996) to include the assessment of a number of Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR; American Psychiatric Association [APA] 2000) diagnoses not included in the original measure, including tic disorders. A trained evaluator completed the ADIS-RLV interview with the parent(s) and child/adolescent, concurrently. The ADIS-RLV possesses good convergent validity (Kendall et al. 2010), while its precursor, the ADIS-IV has demonstrated good to excellent temporal stability and validity (Silverman et al. 2001; Wood et al. 2002).

Tic severity and related impairment was assessed via the Yale Global Tic Severity Scale (YGTSS; Leckman et al. 1989) a semi-structured clinician-rated interview. A trained evaluator conducted a joint interview with the parent(s) and child and then scored the YGTSS. The YGTSS yields a motor tic score (YGTSS-Motor, range: 0–25), a vocal tic score (YGTSS-Vocal, range: 0–25), as well as a Total Tic Severity (YGTSS-TTS) score and an Impairment (YGTSS-IMP) score. The YGTSS-TTS score corresponds to the number, frequency, intensity, complexity, and interference of tics in the past week (range: 0–50). The YGTSS-IMP score (range: 0–50) reflects the overall impact of the tic disorder on the participant. The YGTSS has demonstrated acceptable psychometric properties (Leckman et al. 1989).

Global tic symptom severity was assessed via the Clinical Global Impressions—Severity scale (CGI-S; Leckman et al. 1988). The CGI-S is a clinician rating scale. Variants of the CGI have been used in studies with tic disorder patients (Scahill et al. 2001). In the current study, trained evaluators rated the global tic symptom severity on a 7-point scale (1 = not ill to 7 = extremely ill).

Global functioning over the past month was assessed by the Children’s Global Assessment Scale (CGAS; Shaffer et al. 1983). The CGAS is a 0 to 100 scale rated by a clinician based on all available information (higher scores reflect higher functioning). A score below 60 is often used as the threshold to mark overall impairment. The CGAS possesses adequate inter-rater (r = .84) and test-retest (r = .85) reliability as well as discriminant validity (Green et al. 1994).

General psychosocial (e.g., depression, inattention, hyperactivity, aggression), adaptive, and social functioning were assessed via the Child Behavior Checklist (CBCL; Achenbach 1991), a 118-item parent report measure. The CBCL includes three broad scales (i.e., internalizing, externalizing, and total problems), which subsume eight problem area scales (e.g., depression, inattention, hyperactivity, aggression, etc.). It also includes three competency scales (social, school and activities) and a total competence scale (i.e., CBCL-TC). The CBCL provides age and gender standard scores (i.e., T-scores; M = 50, SD = 10) for each problem and competency scales. T Scores ≥60 on the problem scales are suggestive of moderate or greater psychopathology. By contrast, T scores ≤35 on the competency scales indicate suboptimal adaptive functioning. The CBCL has demonstrated excellent psychometric properties (Achenbach 1991)

The extent to which tics or psychiatric symptoms had a negative impact on the family was assessed via the Caregiver Strain Questionnaire (CSQ; Brannan et al. 1997). Parents rated each of the 21 items on a 5-point scale. The CSQ provides a 10-item subscale (range: 0–50) measuring objective strain including demands on time, and financial strain (CSQ-Direct Impact) and an 11-item subscale (range: 0–55) assessing subjective strain including worry and embarrassment (CSQ-Subjective Strain).

Statistical Analysis

Unless otherwise indicated, all data analyses were based on baseline data from the total randomized sample (N = 126). All analyses pertaining to co-occurring conditions include dichotomous diagnostic status based on the ADIS-IV. Frequencies and percentages were used for categorical variables; means and standard deviations were used for dimensional variables. Bivariate correlations were used to examine relationships between variables. One-way analysis of variance (ANOVA) was used to examine between group differences with respect to key dimensional variables. Chi-square (χ 2) statistics were used to evaluate difference between observed and expected distributions. Simultaneous multiple linear regressions were used to evaluate the unique contribution of continuous independent variables with respect to key dependent variables. Statistical analyses were performed with Statistical Package for the Social Sciences (SPSS) Version 17.0 (Chicago, Il.).

Results

Demographics and Clinical Characteristics

Participants were predominantly male (78.6%; male-to-female ratio = 3.67:1), Caucasians (84.9%) with a mean age of 11.7 years old (SD = 2.3). Most were from intact (84.9%) and upper-middle-class homes (80.2%). Participants met DSM-IV-TR diagnostic criteria for Tourette’s Disorder (93.7%), chronic motor or vocal tic disorder motor only (5.6%) or vocal only (0.8%). YGTSS total tic severity generally fell in the moderate-to-marked tic severity range (M = 24.7, SD = 6.1) as tic tic-related impairment (M = 23.7, SD = 8.4). Tic disorder severity of participants, based on CGI-S ratings, were moderate (60%), marked (37%), or severe (3%). Overall adjustment as measured on the CGAS fell just below the clinical cut off (M = 58.97, SD = 7.9).

Co-Occurring Psychiatric Conditions

A majority of the participants (64%) had at least one co-occurring diagnosis, as shown in Table 1. The average number of co-occurring conditions for the sample was one (Mdn = 1.00, M = 1.1, SD = 1.1). The most common co-occurring condition was ADHD (26%) followed by social phobia (21%), generalized anxiety disorder (20%) and OCD (19%), as shown in Fig. 1. Less frequently observed were separation anxiety disorder (9%), oppositional defiant disorder (6%), and enuresis (5%). Of those with any co-occurring condition about one half had only one (54%, n = 44/81), a one third had two (32%, n = 26/81), and fewer (14%, n = 11/81) had three or more conditions.

Table 1 Gender difference in tic severity and co-occurring psychiatric conditions
Full size table
Fig. 1
figure 1

Percent of participants meeting current DSM-IV-TR Diagnostic Criteria for each co-occurring psychiatric condition. Figure does not account for multiple co-occurring diagnoses. No participants were diagnosed with a co-occurring mood disorder. ADHD = attention-deficit/hyperactivity disorder; Soc Phobia = social phobia; GAD = generalized anxiety disorder; OCD = obsessive-compulsive disorder; Sep. Anx. = separation anxiety disorder; ODD = oppositional defiant disorder

Full size image

Approximately one half of the entire sample (48%, n = 60/126) met criteria for at least one anxiety disorder, OCD included. To examine the extent to which non-OCD anxiety disorders occurred alone or in combination with OCD, we created three groups: (a) those with OCD, but no other anxiety disorder; (b) those with OCD and at least one non-OCD anxiety disorder; and (c) those with only non-OCD anxiety disorders. About one fourth of those with an anxiety disorder presented with OCD only (23%, n = 14/60): most presented with only a non-OCD anxiety disorder (60%, n = 36/60) and approximately one fifth presented with both OCD and a non-OCD anxiety disorder (17%, n = 10/60).

Medication Treatment

Exactly one half (50%) of the participants were adhering to a stable medication regimen. Just over one third of the sample (36%) was on a stable medication for tics tic symptoms (i.e., alpha-agonist or antipsychotic): one fifth (20%) were on a selective serotonin reuptake inhibitor (SSRI); approximately 10% were on a psychostimulant. Of those with co-occurring ADHD, one fourth (24%, n = 8/33) were on a psychostimulant medication and 15% (n = 5/33) were on a non-stimulant alternative (i.e., alpha-agonist). Of those with a co-occurring anxiety disorder, approximately one fourth (24%, n = 14/60) had an active medication regimen that included an SSRI. Approximately one third of those with OCD only (36%, n = 5/14) had a regimen that included an SSRI. Of those with only a non-OCD anxiety disorder, one fifth (19%, n = 7/36) had a regimen that included an SSRI. Likewise, one fifth (20%, n = 2/10) of those with both OCD and another anxiety disorder had an SSRI included in their medication regimen.

Tic Severity, Co-Occurring Conditions, Age, and Gender

There were no significant differences between males and females with respect to tic disorder diagnosis, tic severity, indices of tic severity (i.e., YGTSS number, frequency, intensity, complexity, and interference), tic-related impairment, or number of co-occurring conditions including ADHD, as shown in Table 1.

To explore relationships between age, presence of another psychiatric disorder, and tic symptom severity, we divided the sample by age (i.e., age range: 9–13 and 14–17 years) based on prior reports suggesting that tic symptom severity is greatest before age 13 years (Leckman et al. 1998; Bloch and Leckman 2009). Statistically, more of 9–13 year-olds had a co-occurring condition than did not, χ 2 (1, n = 105) = 4.20, p = .04, as shown in Table 2. Similarly, more of the 14–17 year-olds had a co-occurring condition than did not χ 2 (1, n = 21) = 10.71, p < .001. The most common co-occurring condition in 9–13 year-olds was ADHD (29%, n = 30/105). The most common in co-occurring condition in 14–17 year-olds was social phobia (43%, n = 9/21), followed closely by GAD (38%, n = 8/21).

Table 2 Age differences in tic severity, and co-occurring psychiatric conditions
Full size table

Global Tic Severity and Impairment

Individuals with co-occurring psychiatric conditions did not differ from those without with respect to total tic severity (i.e., YGTSS-TTS score), F(1,124) = 0.09, ns, or tic-related impairment (i.e., YGTSS-IMP score), F(1,124) = 0.71, ns. The YGTSS-TTS and the YGTSS-IMP scores were moderately correlated, r(124) = .42, p < .001. The YGTSS-TTS was inversely correlated with global psychosocial functioning as measured on the CGAS, r(117) = −.46, p < .001, but not social and adaptive competence as measured the CBCL-TC composite score, r(118) = −.13, ns. The correlation of YGTSS-TTS score with family burden as measured on the CSQ was small in magnitude but significant: CSQ-Direct Impact, r(124) = .23, p = .008 and CSQ-Subjective Strain, r(124) = .27, p = .003. The YGTSS-IMP score was inversely correlated with the CGAS, r(117) = −.49, p < .001 and CBCL-TC, r(118) = −.28, p = .002. The correlation of YGTSS-IMP score with family burden as measured on the CSQ-Direct Impact, r(124) = .34, p < .001 and CSQ-Subjective Strain scores, r(124) = .28, p = .001, were modest.

Motor Tics, Vocal Tics, Aspects of Tic Severity, and Impairment

The relationships between motor tics, vocal tics, and aspects of tic severity (i.e., YGTSS tic number, frequency, intensity, complexity, interference) with various forms of impairment including tic-related impairment (YGTSS-IMP), global psychosocial functioning (CGAS), subjective strain/direct impact on the family (CSQ-Subjective Strain and CSQ-Direct Impact), and adaptive/social functioning (CBCL-TC) were also explored.

The YGTSS-Motor and YGTSS-Vocal tic scores were not correlated, r(124) = .05, ns. The YGTSS-Motor tic score was moderately correlated with YGTSS-IMP, r(124) = .38, p < .001, and inversely correlated with CGAS, r(117) = −.40, p < .001. The YGTSS-Motor tic score was also positively correlated with CSQ-Subjective Strain, r(124) = .19, p = .03, and CSQ-Direct Impact, r(124) = .18, p = .04). The YGTSS-Vocal tic score was correlated with YGTSS–IMP, r(124) = .25, p = .005), as well as CSQ-Subjective Strain, r(124) = .19, p = .03, and was inversely correlated with CGAS, r(117) = −.27, p = .003.

The unique contributions of motor versus vocal tics with respect to various forms of impairment were also evaluated. The YGTSS-Motor and YGTSS-Vocal tic scales were entered into a series of simultaneous multiple regressions to predict each measure of impairment. The YGTSS-Motor and YGTSS-Vocal scales significantly predicted YGTSS-IMP, R 2 = .20, F(2, 123) = 15.08, p < .001, 95% CIs [.47, 1.18] and [.13, .72], respectively; both the YGTSS-Motor, β = .37, t(123) = 4.56, p < .001, 95% CI [.47, 1.18], and YGTSS-Vocal scales, β = .23, t(123) = 2.82, p = .006, 95% CI [.13, .72], independently predicted YGTSS-IMP. Likewise, YGTSS-Motor and YGTSS-Vocal scales inversely predicted CGAS scores, R 2 = .23, F(2, 116) = 16.90, p < .001, 95% CIs [−1.14, −.47] and [−.71, −.16], respectively; both the YGTSS-Motor, β = −.39, t(116) = −4.76, p < .001, 95% CI [−1.14, −.47], and YGTSS-Vocal scales, β = −.26, t(116) = −3.12, p = .002, 95% CI [−.71, −.16], independently predicted CGAS scores. The YGTSS-Motor and YGTSS-Vocal tic scales predicted CSQ-Subjective Strain scores, R 2 = .07, F(2, 123) = 4.71, p = .01, 95% CIs [.02, .69] and [.02, .57], respectively; both the YGTSS-Motor, β = .18, t(123) = 2.11, p = .04, 95% CI .02, .69], and YGTSS-Vocal tic scales, β = .18, t(123) = 2.11, p = .04, 95% CI [.02, .57], independently predicted CSQ-Subjective Strain scores. Although, the YGTSS-Motor and YGTSS-Vocal tic scales together predicted CSQ-Direct Impact scores, R 2 = .06, F(2, 123) = 3.63, p = .03, 95% CIs [−.01, .76] and [.04, .76], neither were significant independent predictors at the p < .05 significance level.

Regarding the relationships between aspects of tic severity (e.g., number, frequency, intensity) and impairment, the YGTSS tic intensity was correlated with all measures of impairment: CGAS, r(117) = −.31, p = .001; CBCL-TC, r(118) = −.21, p = .02; YGTSS-IMP, r(124) = .35, p < .001; CSQ-Direct Impact, r(124) = .20, p = .02); and CSQ-Subjective Strain, r(124) = .27, p = .003) score. Tic complexity was correlated with four of five impairment measures: CGAS, r(117) = −.32, p < .001; YGTSS-IMP, r(124) = .33, p < .001; CSQ-Direct Impact, r(124) = .19, p = .04; and CSQ-Subjective Strain, r(124) = .23, p = .01. Tic number was also correlated with four of five impairment measures: CGAS, r(117) = −.33, p < .001; YGTSS-IMP, r(124) = .29, p = .001; CSQ-Direct Impact, r(124) = .25, p = .006; and CSQ-Subjective Strain, r(124) = .29, p = .001. Tic interference was correlated with two of four impairment measures: CGAS, r(117) = −.45, p < .001, and YGTSS-IMP, r(124) = .43, p < .001. Tic frequency was correlated with only one impairment measure, the CGAS, r(117) = −.24, p = .008.

The unique contributions of YGTSS tic severity dimensions with respect to various forms of impairment were also evaluated. All tic severity dimensions were entered into a simultaneous multiple regression to predict each impairment measure. The total model (i.e., number, frequency, intensity, complexity, and interference) significantly predicted YGTSS-IMP scores, R 2 = .26, F(5, 120) = 8.20, p < .001, 95% CIs [−1.06, 1.19], [−1.26, .42], [−.13, 2.09], [−.01, 1.74] and [.69, 2.54], respectively; however, only interference, β = .32, t(120) = 3.46, p = .001, 95% CI [.69, 2.54] and complexity, β = .19, t(120) = 1.97, p = .05, 95% CI [−.01, 1.74], independently predicted YGTSS-IMP scores. The total model also significantly predicted CGAS scores, R 2 = .25, F(5, 113) = 7.51, p < .001, 95% CIs [−1.26, .94], [−1.13, .50], [−1.22, .95], [−1.65, .06] and [−2.54,−.73], respectively; however, only interference, β = −.34, t(113) = −3.57, p = .001, 95% CI [−2.54,−.73], independently predicted CGAS scores. The total model also significantly predicted CSQ-Subjective Strain scores, R 2 = .13, F(5, 120) = 3.51, p = .005, 95% CIs [−.04, 2.07], [−1.42, .15], [.03, 2.11], [−.55, 1.09] and [−.96, .77], respectively; however, only intensity, β = .22, t(120) = 2.04, p = .04, 95% CI [.03, 2.11], independently predicted CSQ-Subjective Strain scores. The total model failed to significantly predict CBCL-TC scores, R 2 = .07, F(5, 114) = 1.71, ns, 95% CIs [−1.63, 1.57], [−.37, 1.91], [−3.33,−.26], [−1.7., .68] and [−1.01, 1.56], respectively; however, intensity independently predicted of CBCL-TC scores, β = −.27, t(114) = −2.31, p = .02, 95% CI [−3.33,−.26]. The total model also failed to predict CSQ-Direct Strain scores, R 2 = .08, F(5, 120) = 2.01, ns, 95% CIs [−.26, 2.22], [−1.30, .55], [−.55, 1.89], [−.72, 1.20] and [−.92, 1.12], respectively; none of the tic dimensions remained as an significant independent predictor. The previously mentioned tic severity dimensions remained significant independent predictors of impairment while controlling for the contributions of co-occurring conditions (i.e., ADHD, OCD, non-OCD anxiety disorder); however, ADHD also remained a significant independent predictor of CBCL-TC scores, β = −.34, t(115) = 3.99, p < .001, 95% CI [−11.59,−3.89]; CSQ-Subjective Strain scores, β = .31, t(121) = 3.73, p < .001, 95% CI [2.39, 7.81]; and CGAS scores, β = −.25, t(114) = −3.06, p = .003, 95% CI [−7.11,−1.53].

Discussion

The profile of tic symptom severity in this sample was in the moderate to severe range and in keeping with clinical reports of treatment seeking youth with CTDs was similar to that reported in previous medication trials (Gilbert et al. 2004; Nicolson et al. 2005; Sallee et al. 2000; Scahill et al. 2003; Silver et al. 2001; Toren et al. 2005). However, in contrast to the majority of medication RCTs, which required participants to be off tic suppressing medications at baseline, approximately one third of participants in the present study were being treated with tic-suppressing medications. The overall high baseline tic severity, and a subgroup with high tic severity despite pharmacological treatment, suggests that participants in this study were appropriately selected to pursue treatment to reduce tic severity.

The current study confirms the high rate of co-occurring conditions in youth with CTDs despite study entry criteria that excluded youth with co-occurring conditions in imminent need for treatment. The percentage of participants meeting DSM-IV diagnostic criteria for co-occurring ADHD was much lower than clinically-ascertained case series (Scahill et al. 2009; Spencer et al. 1998; Sukhodolsky et al. 2003) but was still the most common co-occurring condition in this sample. ADHD was the most common concomitant condition in the 9–13-year-old participants, was more common in 9–13 year-olds than 14–17 year olds, and appeared to contribute to poorer global psychosocial functioning, poorer adaptive/social functioning, and perceived strain on family members when tic severity and other Axis I conditions were statistically controlled. Contrary to epidemiological studies, we did not identify a male preponderance of ADHD in this sample. The lack of a gender difference in ADHD suggests that treatment-seeking females with tics, like their male counterparts, should be evaluated for ADHD.

Nearly half the sample met DSM-IV-TR diagnostic criteria for an anxiety disorder. Although 23% of those with any co-occurring anxiety disorder were diagnosed with OCD only, the majority (77%) were diagnosed with a non-OCD anxiety disorder. Social phobia and generalized anxiety disorder were among the most common co-occurring conditions in the entire sample. Social phobia was the most common co-occurring condition in the 14–17-year-old participants followed closely by generalized anxiety disorder. The emergence of social phobia as the prominent co-occurring condition in late adolescence likely reflects the later age of onset for this condition. The somewhat higher than expected rates of social phobia in comparison to prior reports (e.g., Spencer et al. 1998) likely results from our use of the ADIS-RLV which, by design, provides a systematic assessment of anxiety disorders including social phobia.

Despite high rates of anxiety disorders in this sample, there were low rates of SSRI usage at study entry. The reasons for the low utilization of an evidenced based treatment for childhood anxiety disorder (Walkup et al. 2008) cannot be determined from this study; however, there are several possibilities. One possibility is that that in children with multiple presenting problems (e.g., severe tics, ADHD, OCD), treatment of other anxiety disorders may seem less urgent. In fact, by design, participants in this study were included only if tics were more pressing than the co-occurring conditions identified during screening; therefore, despite the high rate of non-OCD anxiety disorders in the current sample, anxiety disorders were not the most impairing condition(s). It is also possible that SSRI treatments had been tried in some participants and were found to be unhelpful. Finally, some families may have previously sought and obtained non-pharmacological treatment such as cognitive behavioral therapy. Nonetheless, the high rates of anxiety disorders in the current sample speak to the importance of careful evaluation of non-OCD anxiety disorders in youth with CTDs.

Prior research is inconsistent with respect to the prevalence of mood disorders in youth with CTDs. However, in the current sample no participants with a co-occurring mood disorder were enrolled in the study. The younger average age of the study population may account for the higher-than-anticipated rates of anxiety disorders and the absence of mood disorders. However, it is also possible that youth with a mood disorder were excluded from participating due to an imminent need for treatment. The specific number of participants excluded from participation due to a mood disorder is not known.

In the current sample, individuals with co-occurring psychiatric conditions did not systematically exhibit greater tic severity, nor was tic symptom severity associated with a greater number of co-occurring conditions. While prior reports suggest that impairment is more likely the result of co-occurring disorders and/or symptoms rather than tics (Bawden et al. 1998; Carter et al. 2000; Coffey et al. 2004), emerging evidence suggests more complex relationships exist between tics, comorbidity, and impairment (Boudjouk et al. 2000; Himle et al. 2007; Woods 2002; Woods et al. 1999). The current findings lend support to the notion that CTDs and co-occurring psychiatric conditions coexist but do not necessarily influence one another. However, because our sample was sufficiently restricted with respect to tic severity the current findings may not apply to those with milder tic symptom presentations.

Given that youth in the current sample were seeking treatment for tics despite the presence of a co-occurring psychiatric conditions suggests that families seek treatments based, at least in part, on availability and salient need. Indeed, global tic severity was associated with increased tic-related impairment, poorer gross psychosocial functioning, and perceived family burden. Our results also suggest that motor and vocal tics contribute equally to impairment. Also, loud vocalizations and/or forceful movements (i.e., high tic intensity), bouts or prolonged/orchestrated patterns of tics (i.e., high tic complexity), and tics that impede one’s intended actions (i.e., high tic interference) may contribute to impairment more than the frequency and/or number of tics and may result in treatment seeking even in the presence of co-occurring conditions.

Limitations

Despite recruiting participants from geographically and ethnically/culturally diverse populations, only a moderate number of participants were from racial/ethnic minorities and lower socio-economic (SES) groups. Although CTDs have been identified in most racial and ethnic groups, it is possible that minority and/or lower SES groups may encounter barriers to involvement in mental health services (Hoberman 1992; Scahill et al. 2009) and treatment research. It may also be that mild to moderate tic severity is not a primary concern in families with psychosocial and economic stressors. It is also possible that study entry criteria (e.g., fluency in the English language) may have resulted in decreased ethnic/cultural diversity of the current sample.

The sampling frame for this RCT was restricted to children 9–17 years of age to ensure sufficient cognitive maturity for full participation in the study treatment. As a result, the current findings may not be applicable to treatment-seeking children younger than 9 years of age.

Although co-occurring psychiatric conditions and current medication status were thoroughly evaluated, we did not inquire about historical treatment of co-occurring conditions. As a result, we were unable to determine the extent to participants previously sought treatment for co-occurring conditions. Clearly, this information would have been helpful in further characterizing our sample.

In this study, we attempted to replicate prior research (i.e., Himle et al. 2007) examining the relative contribution of aspects of tic severity to impairment. It should be noted, however, that the YGTSS dimensions: number, frequency, intensity, complexity, and interference are 0 to 5 scales that may not meet usual assumptions for parametric statistics. These findings are exploratory and offered to generate hypotheses about the relationship between tic severity and impairment.

Conclusions

Clinicians treating children and adolescents with CTDs should consider non-OCD anxiety disorders in addition to OCD and ADHD. Although there is little research to guide clinicians in sequencing or combining treatments for children with CTDs, it is generally recommended to initiate treatment for the most immediate source of distress or impairment. However in children with CTDs and prominent co-occurring conditions this may not be a straightforward task. Even in the presence of diagnosable co-occurring conditions parents and clinicians may feel compelled to focus on tics. With the development of a safe and effective behavioral treatment for reducing tic severity, parents and children now have a non-medication option for reducing tic severity. Future research is needed to determine the best approach to sequencing and combining treatments to address the complex needs of children with CTDs and co-occurring conditions.