An international perspective on Tourette syndrome: selected findings from 3500 individuals in 22 countries Roger D Freeman* MD, Clinical Professor of Psychiatry, Associate in Pediatrics; Diane K Fast MD PhD, Clinical Professor of Psychiatry, Associate in Pediatrics, University of British Columbia, Vancouver, BC, Canada; Larry Burd PhD, Clinical Associate Professor of Pediatrics and Neuroscience; Jacob Kerbeshian MD, Clinical Professor of Neuroscience, University of North Dakota School of Medicine, Grand Forks, USA; Mary M Robertson MD FRPsych, Professor of Neuropsychiatry, University College London Medical School, London, UK; Paul Sandor MD, Assistant Professor of Psychiatry, University of Toronto, Canada; Tourette Syndrome International Database Consortium. *Correspondence to first author at Neuropsychiatry Clinic (C-4), British Columbia’s Children’s Hospital, Vancouver, BC, V6H 3V4, Canada. E-mail:
[email protected]
We have established a multisite, international database of 3500 individuals diagnosed with Tourette syndrome (TS). The male:female ratio is 4.3:1 for the total sample, with wide variation among sites; the male excess occurs at every site. Anger control problems, sleep difficulties, coprolalia, and self-injurious behavior only reach impressive levels in individuals with comorbidity. Anger control problems are strongly correlated with comorbidity, regardless of site, region, or whether assessed by neurologists or psychiatrists. The mean age at onset of tics is 6.4 years. At all ages, about 12% of individuals with TS have no reported comorbidity. The most common reported comorbidity is attentiondeficit–hyperactivity disorder. Males are more likely to have comorbid disorders than females. The earlier the age at onset, the greater the likelihood of a positive family history of tics. An understanding of the factors producing these and other variations might assist in better subtyping of TS. Because behavioral problems are associated with comorbidity, their presence should dictate a high index of suspicion of the latter, whose treatment may be at least as important as tic reduction. The established database can be used as the entry point for further research when large samples are studied and generalizability of results is important.
436
Developmental Medicine & Child Neurology 2000, 42: 436–447
Tourette syndrome (TS) is described as a complex, usually familial, neuropsychiatric disorder with an onset of multiple waxing and waning motor and vocal tics (not necessarily simultaneous) at a mean age of about 6 to 7 years and a duration of at least 12 months (Table I) (Leckman et al. 1997, 1998). Prevalence figures range from 0.05% (Burd et al. 1986a, b; Caine et al. 1988) to 3% (Comings et al. 1990, Mason et al. 1998). The mode of inheritance may be autosomal dominant, but could also be accounted for by a mixed model (Walkup et al. 1996, Barr and Sandor 1998, Rutter et al. 1999). Phenotypic definition has been problematic in unraveling the genetics of TS (Sadovnick and Kurlan 1997, Barr and Sandor 1998, Barr et al. 1999). Although the detailed mechanism of tic expression is still unknown, dopaminergic systems in the frontal/subcortical tracts are presumably involved because of the response to dopamine blocking agents and certain other findings (Leckman et al. 1997). Symptoms typically increase until preadolescence and then gradually decrease (Kerbeshian and Burd 1992, Leckman et al. 1998). Recently a hypothesis of ‘molecular mimicry’ (pediatric autoimmune neuropsychiatric disorders associated with streptococcal infection or PANDAS) has been proposed for a subset of individuals, in which an abrupt onset of tics and/or obsessive–compulsive disorder (OCD) occurs some weeks after an infection with Group A β-hemolytic streptococci. This proposed autoimmune mechanism is still under investigation (Allen et al. 1995; Kurlan 1998, 1999; Perlmutter et al. 1999). It is generally agreed that a large but undetermined proportion of individuals with TS remains undiagnosed in the community (Robertson and Gourdie 1990, Bruun and Budman 1997, Peterson and Cohen 1998). Recently, the subject of comorbidity has grown in importance. It has been reported that attention-deficit–hyperactivity disorder (ADHD) (Pauls et al. 1993, Yeates and Bornstein 1994, Walkup et al. 1995, Nolan et al. 1996), OCD (George et al. 1993, Santangelo et al. 1994, Zohar et al. 1997), self-injurious behaviors (SIBs) (Robertson 1992, Robertson et al. 1989), sleep disorder (Allen et al. 1992), anger control problems and ‘rage’ (Budman et al. 1998) are more common in persons with TS, at least in clinical samples. Whether this is so in community samples is unknown. Baron-Cohen et al. (1999a, b) recently reported a much higher than expected rate of TS in two screened samples of individuals with autism in the UK (one sample included over 400 pupils with autism). In general, the nature of the comorbidity relationships has been unclear (Carter et al. 1994). Method BACKGROUND
This project was established in 1992 with the aim of finding out more about the variability in clinical samples of TS among sites. At that time a collaboration was set up among sites in Vancouver, BC; Grand Forks, North Dakota; and Winnipeg, Manitoba. The TS Foundation of Canada sponsored a meeting in Montreal at which the data entry form was designed. The form is brief, to enable use in standard clinical practice without needing specific research funding. During 1993 and 1994 the form was piloted and slightly modified. A rapid recruitment of sites has occurred since a meeting in Quebec City in October 1997. Currently 22 countries are represented in the consortium (see Appendix 1), with 65 participating
sites. A variety of medical specialists are involved, including psychiatrists, neurologists, pediatricians, and medical geneticists. The composition of the medical staff varies at different sites. This study reports on clinical information from 3500 individuals with TS whose details were entered into the database. STUDY DESIGN
The data entry form (see Appendix 2) was distributed to all participating clinicians who agreed to record the results of their usual assessments. All consecutive patients who met the DSM-III-R (American Psychiatric Association 1987) diagnostic criteria for TS were included. Thus the form could uniformly record clinical diagnostic impressions using standard diagnostic criteria. No attempt was made to establish identical assessment procedures or to determine the interrater reliability of the diagnostic categories. To protect patient confidentiality, patients’ names were not put on the form (see Appendix 2). The data were entered into a database by one of the authors (RDF) at the central site. He also performed descriptive data analyses which were distributed to the participants. Each site followed its own local requirements for informed consent and patient confidentiality. DIAGNOSTIC CATEGORIES
Because of difficulties with the impairment criterion in the
Table I: Criteria differences for diagnosis of Tourette syndrome DSM-IV (1994)
TS Classification Study Group (1993) ‘Definite Tourette Syndrome’
Both multiple motor and one or more vocal tics, not necessarily concurrently
Both multiple motor and one or more vocal tics have been present at some time, although not necessarily concurrently
Occurring many times a day, nearly every day or intermittently throughout a period of more than 1 year; there was never a tic-free period of more than 3 months
Tics occur many times a day, nearly every day or intermittently throughout a period of more than 1 year
DSM-IV (American Psychiatric Association 1994, Freeman et al. 1995, Erenberg and Fahn 1996, Kurlan 1997), the modified DSM-III-R consensus criteria developed by the Tourette Syndrome Classification Study Group (1993) are generally used (see Table I). We also used DSM-III-R criteria for diagnosing TS. For other disorders, DSM-IV criteria were used. Because children, adolescents, and adults were included in the study, and as oppositional defiant disorder (ODD) can be construed as a subset and often a precursor of conduct disorder (CD), we combined the two (CD/ODD). Any firm conclusions based upon this combined construct must be tentative until further elaboration and confirmation of its validity. COMORBIDITY SCORE
For the comorbidity score, we adopted the simple technique of Spencer et al. (1998) which assigns a score of one for each psychiatric comorbidity, giving a score from 0 (‘TS only’) to 7 (when TS and seven comorbid conditions are present). Although this method assumes each comorbidity to be equivalent, it provides a rough estimate of complexity of the patient’s disorder(s). STATISTICAL ANALYSIS
The mean and standard deviation of continuous variables such as age at onset and age at diagnosis were calculated for each site and for the total sample. Similarly the frequency of the categorical variables, e.g. the presence of various comorbid conditions, was determined and tabulated for each site and for the total sample. Data from the largest sites were analyzed separately before aggregate data were used. Results Of the 3500 individuals studied, 42% were from Canada, 27% from the United States, and 31% from outside North America. As can be seen in Table II, 18 out of 64 sites provided 50 or more individuals with TS, accounting for a large proportion of the total sample. This is due to the varying duration of membership to the Tourette Syndrome International Database Consortium (TIC). AGE AT ONSET AND AGE AT DIAGNOSIS
Figure 1 shows that the mean age at onset, ranging between 6 and 7 years, is remarkably consistent among sites (Leckman and Cohen 1999). Forty-one percent of individuals had an onset of tics below the age of 6 years, and 93% were
The anatomic location, number, frequency, complexity, type, severity of tics change over time Criterion C. Marked distress or significant impairment in important areas of functioning
Table II: Reported individuals by site n
Onset before age 18 years
Onset before age 21 years
Not due to direct physiological effects of a substance or a general medical condition
Involuntary movements and noises cannot be explained by other conditions Motor and/or vocal tics must be witnessed by a reliable examiner directly at some point, or be recorded by videotape or cinematography
0–24 25–49 50–74 75–99 100–124 125–149 150–199 200–299 300–399 ≥ 400 Total
Nr of sites 30 16 4 2 3 4 2 1 1 1 64
Tourette Syndrome Roger D Freeman et al. 437
reported in SIB, TS only, and trichotillomania. These sexrelated associations are summarized in Table IV.
symptomatic by age 10 years. The diagnosis had been made by the age of 10 years in 56%. Only 16% of individuals were not diagnosed until adulthood (after age 18 years) (see Table III).
DIFFERENCES AMONG SITES
SEX DIFFERENCES
Table V shows the extensive intersite variation in most parameters. (Only data from sites reporting 50 or more
There was considerable variation in the ratio of females to males, ranging from 1:10 to 1:3. However, the mean ratio of 1:4.3 is in line with previous reports (Bruun and Budman 1997). Male excess is evident at every site, and is more pronounced in children (1:5.2) than in adults (1:3) (p
