1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusion
  7. References
  8. Additional co-authors

The aims of this descriptive study were to examine the prevalence and associations of coprophenomena (involuntary expression of socially unacceptable words or gestures) in individuals with Tourette syndrome. Participant data were obtained from the Tourette Syndrome International Database Consortium. A specialized data collection form was completed for each of a subset of 597 consecutive new patients with Tourette syndrome from 15 sites in seven countries. Coprolalia occurred at some point in the lifetime of 19.3% of males and 14.6% of females, and copropraxia in 5.9% of males and 4.9% of females. Coprolalia was three times as frequent as copropraxia, with a mean onset of each at about 11 years, 5 years after the onset of tics. In 11% of those with coprolalia and 12% of those with copropraxia these coprophenomena were one of the initial symptoms of Tourette syndrome. The onsets of tics, coprophenomena, smelling of non-food objects, and spitting were strongly intercorrelated. Early onset of coprophenomena was not associated with its longer persistence. The most robust associations of coprophenomena were with the number of non-tic repetitive behaviors, spitting, and inappropriate sexual behavior. Although coprophenomena are a frequently feared possibility in the course of Tourette syndrome, their emergence occurs in only about one in five referred patients. Because the course and actual impact of coprophenomena are variable, additional prospective research is needed to provide better counseling and prognostic information.

Tourette syndrome has a highly variable neurodevelopmental clinical pattern of waxing and waning tics, beginning in childhood and usually stabilizing and lessening in severity during and after adolescence.1 There is a strong male preponderance, and most patients have marked neurodevelopmental comorbidity.2

Coprophenomena are traditionally considered to be tics that are socially inappropriate verbal expressions (coprolalia) or gestures (copropraxia), often having a vulgar (obscene), religious (profane), or racially or ethnically insulting content, not expressed out of immediate anger or frustration, or solely for purposes of emphasis.3 It is an important but curious observation that, although coprophenomena are the most widely recognized clinical feature of Tourette syndrome by the public, the research on coprophenomena is very limited.4–6

Gilles de la Tourette’s seminal paper7 described eight of his own patients. Only two of them manifested coprolalia; in one it abated over time, and in the other the onset was 4 years after tic onset and coprolalia persisted. The historical forces that influenced his (and others’) conclusions about the importance of coprophenomena have been discussed by Kushner.8,9

The neurobiological underpinnings of tics are not well established, but brain imaging, neurophysiological, and post-mortem studies indicate involvement of cortical–striatal–thalamocortical pathways.10–12 Further, variations in tic phenotypes may imply disparate brain mechanisms that cause tics. For example, in a functional neuroanatomy study using positron emission tomography, Stern et al.13 demonstrated that simple motor tics (e.g. anatomically isolated, sudden, rapid movements, such as eye-blinking) were associated with the sensorimotor cortex, whereas more complex tics (such as coprolalia and clear vocal tics) were associated with activity in prerolandic and postrolandic language regions, insula, caudate, thalamus, and cerebellum.

Kurlan et al.14 described a more broadly encompassing definition of coprophenomena that includes noticeable, repeated, and interfering ‘non-obscene socially inappropriate’ patterns (e.g. insults) that also fall into this category. Occasionally coprolalia occurs outside of Tourette syndrome, in post stroke, postencephalitic, or hemispherectomized patients and, rarely, in patients with obsessive–compulsive disorder without tics15–17 or patients with frontotemporal dementia.

The paucity of investigation into coprophenomena leaves large gaps in our understanding of their epiphenomena and endophenotypes. Unanswered questions include when coprophenomena typically emerge in relation to tic onset, how long symptoms last, whether coprolalia and copropraxia represent unique neurological entities, how coprophenomena and tic severity co-vary, what the relationship might be with other repetitive behaviors, and how responsive the disorders are to clinical intervention. As a result, management and counseling have severe limitations. The present study is an initial attempt to fill some of the knowledge gaps and to highlight the need for further research. Answers to these questions may provide much-needed guidance in clinical management of coprophenomena and in understanding the neurobiology of tic disorders.

Arthur and Elaine Shapiro, pioneers in the modern era of Tourette syndrome research,5 cited prevalence rates of 32% for coprolalia and 12.8% for copropraxia among patients with Tourette syndrome, with a range at onset from 1 to 36 years. A subsequent review by Singer6 indicated an onset of coprophenomena at an average of 4 to 7 years after tic onset. Selected findings from 15 other studies are summarized in Table I.

Table I.   Studies on coprophenomena
AuthorsnMale: female ratioAge at tic onset (y)Age at onset of coprophenomena (y)Delay between tic and coprophenomena onset (y) Coprolalia (%) Copropraxia (%)Coprolalia: copropraxia ratio
  1. aIncludes some mental coprolalia; badults only (age ≥16y); –, data not provided; TIC, Tourette Syndrome International Database Consortium (the present study).

Cardoso et al. 199618323:17.128251.12
Chang et al. 200419437.6:19.744.1
Comings & Comings 1985202504:16.933
Eapen et al. 200421911.8:17.314.67.342.923.11.86
Fen et al. 200122581.6:127.620.11.37
Goldenberg et al. 1994231124.1:
Jankovic & Rohaidy 1987241123.8:17.344a19.02.32
Kano et al. 199725646.1:16.950
Lees et al. 1984265371139211.86
Miranda et al. 199927706.48.5
Robertson et al. 19884902.3:1715.
Robertson et al. 1999282803.2:129
Robertson et al. 200229573.8:1640.025.01.60
Shapiro et al. 198856663:16.58-123212.82.50
Teive et al. 200130331.75:1
TIC, 20085584:15.911.35.418.45.73.23


  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusion
  7. References
  8. Additional co-authors

The Tourette Syndrome International Database Consortium (TIC) began compiling descriptive data in 1996. Over the past 10 years the TIC has recruited 61 member sites in 27 countries and incorporates about 800 new cases annually. TIC cases are registered in the database only once. Selected findings of the first 3500 cases have been published,2 and details about the database project with a discussion of its advantages and limitations have been presented elsewhere.31,32 This initial data set had a coprophenomena lifetime prevalence of 14%, and, on the basis of that result, a revised TIC template was designed to refine the details of information gathered. One of the TIC’s purposes was to facilitate the participation of interested member sites in special projects. The current report is the first from such a project.

Fifteen self-selected TIC sites in seven countries (Canada, Germany, Israel, Japan, Poland, UK, and USA) participated in the more detailed coprophenomena study. Each site used a specialized data collection form, which was an expanded version of the TIC data collection form. Detailed coprophenomena data from 648 prospectively entered consecutive patients with Tourette syndrome, representing 8.1% of the full TIC data set, were acquired over a 4-year period (2005–2008) from the participating sites and entered into a structured database. Tourette syndrome was diagnosed according to the Tourette Syndrome Classification Study Group criteria,33 which are an elaboration of the DSM-III (revised) criteria.34 The older criteria were used because, when data collection began, the 1994 criteria35 were unsatisfactory and were subsequently modified in 2000.36 Other disorders were diagnosed using the DSM-IV (text revision) criteria.36 Information collected met respective member sites’ local requirements of their institutional review boards (including for informed consent) and was anonymized. Patients with intellectual disability (n=15), pervasive developmental disorder (n=32), and psychosis (n=4) were excluded because of small numbers and peculiarities in interpreting their repetitive behavior. Our methodology excluded consideration of ‘mental’ coprolalia (i.e. coprolalia only occurring internally and not perceived by others).16

All data were forwarded to the project data center manager (RDF), who reviewed each form for completeness, errors, or indecipherable responses. Problem forms were returned to the clinical sites for correction.

The final sample comprised 597 patients, 506 children under 18 years of age and 91 adults. Those having tics as the only clinical manifestation of Tourette syndrome were termed ‘TS-only’, and those with comorbid disorders ‘TS+’. For some comparisons a comorbidity score was calculated, representing simply the individual’s number of comorbid disorders. Data collected included a history of repetitive behaviors not usually considered to be tics (e.g. nail-biting, skin-picking, nonfunctional nose-picking, bruxism, self-cutting and self-hitting, joint-cracking, rocking, smelling non-food objects, and urges to take specific risks after being warned about them). The total number of these patterns was calculated for each participant. Sample characteristics with significant differences between children and adults are shown in Table II.

Table II.   Selected characteristics that differ between children and adults with Tourette syndrome
CharacteristicsChildrena n=506 (85%)Adults n=91 (15%)p OR (95% CI)
  1. Except where otherwise indicated, data are numbers of patients (%). aChildren defined as age <18y; bt-test on means with equal variance not assumed (Levene test): t=7.39, mean difference 2.75, SE of difference 0.372 (95% CI 2.01–3.48); NA, not applicable (2×3 table).

Coprophenomena93 (18.4)26 (28.6)0.0321.11 (1.00–1.22)
Males:females, n (ratio)429:77 (5.6:1)65:26 (2.6:1)0.0042.23 (1.33–3.73)
Mean age at tic onset, y (SD)5.57 (2.4)8.32 (3.1)<0.001bNA
Tics rated severe76 (15)32 (35) NA
Tics rated moderate222 (43.9)46 (51)<0.001 (combined)NA
Tics rated mild208 (41.1)13 (14) NA
TS-Only122 (24.1)8 (9)0.0013.30 (1.55–7.00)
Specific learning disability102 (20.2)9 (10)0.0192.31 (1.12–4.76)
Mood disorder31 (6)29 (32)<0.0011.71 (1.34–2.19)
Pre-/perinatal problems 84 (17)9 (11)0.1961.66 (0.80–3.45)
Medication for tics (ever)170 (33.7)59 (65)<0.0011.23 (1.13–1.34)
Obsessive-compulsive disorder73 (15)27 (30)0.0011.19 (1.05–1.35)
Self-injurious behavior137 (27.1)44 (48)0.0011.17 (1.07–1.28)
Attention-deficit–hyperactivity disorder243 (48.0)41 (45)0.6491.13 (0.72–1.76)
Anxiety disorder100 (19.8)26 (29)0.0691.09 (0.99–1.20)
Developmental coordination disorder64 (13)2 (2)0.006NA

Statistical analysis

Statistical analysis was performed using SPSS version 16.0 (SPSS Inc., Chicago, IL, USA). Bivariate analyses of history of coprophenomena against each possible explanatory variable were performed using Pearson’s χ2 test for categorical variables (or Fisher’s exact test for tables that have cells with 5 or fewer numbers), one-way analysis of variance (ANOVA), or nonparametric correlations where appropriate. The history of coprophenomena was then entered as the dependent variable into a binary logistic regression analysis against all explanatory variables. Stepwise (alternating) variable selection was performed with the probability to enter the regression equations set at ≤0.05, and the probability to remove set at ≥0.10. The purpose of the variable selection procedure was not to develop a predictive equation; the intent was to remove the effects of spurious correlations resulting from partial confounding of correlated variables, so that relationships between history of coprophenomena and various other measurements could be made clearer than a sequence of bivariate analyses would allow. Missing data were at a very low level and resulted in listwise exclusion in statistical calculations.

Bootstrapping was used to quantify the uncertainty inherent in stepwise model selection in logistical regression as follows.37,38 Ten thousand random samples were drawn with replacement from the original sample. Each resample was subjected to the same stepwise variable selection procedure and the final model was recorded. For each explanatory variable, the proportion of resamples in which it was selected into the final model was calculated. These proportions are indications of strength of the relationship between each variable and the history of coprophenomena, accounting for all other available variables. To explore the effects of known multicollinearity among certain groups of the explanatory variables, this procedure was repeated on different pools of variables. Specifically, comorbid disorders were considered either as an enumerated list of disorders or as a summary score, or both together; a similar approach was taken with repetitive behaviors.


  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusion
  7. References
  8. Additional co-authors

Sites differed in their proportion of TS-only cases (range 3.1–50.0%), children (range 37–100%), and tics rated severe (3–67%). Mean ages at onset of tics ranged from 5 years 2 months to 8 years 7 months (SD 2y 7mo), and ages at onset of coprophenomena from ranged from 8 years 11 months to 16 years 8 months (SD 6y 8mo). Some variations across large sites were consistently small, others varied, as can be seen in Table III.

Table III.   Characteristic variations across study sites in patients with Tourette syndrome
CharacteristicSite A n=215Site B n=136Site C n=94All others n=152
  1. Data are numbers of patients (%). TS, Tourette syndrome.

Females28 (13.0)22 (16.2)25 (26.6)28 (18.4)
Children209 (97.2)134 (98.5)56 (59.0)107 (70.4)
TS-only53 (24.7)32 (23.5)15 (16.0)30 (19.7)
‘Severe’ tics7 (3.3)24 (17.6)21 (22.3)56 (36.8)
Coprophenomena20 (9.3)19 (14.8)30 (31.9)49 (32.2)
Obsessive-compulsive disorder38 (17.8)19 (14.0)14 (14.9)29 (19.1)
Attention-deficit–hyperactivity disorder99 (46.0)69 (50.7)38 (40.4)78 (51.3)
Specific learning disability40 (18.8)42 (30.9)7 (7.4)22 (14.5)
Anxiety disorder50 (23.3)35 (25.7)19 (20.2)22 (14.5)
Mood disorder3 (1.4)19 (14.0)17 (18.1)21 (13.8)
Sleep problems52 (24.2)32 (23.5)21 (22.3)38 (25.0)
Anger control problems36 (16.7)30 (22.1)40 (42.6)42 (27.6)
Social skill deficits43 (20.0)23 (16.9)18 (19.1)20 (13.2)
Sexually inappropriate behavior7 (3.3)4 (2.9)0 (0)10 (6.6)
Spitting28 (13.0)16 (11.8)1 (1.1)36 (23.8)
Nail-biting65 (30.4)42 (31.1)14 (15.2)54 (35.8)
Skin-picking40 (18.6)27 (19.9)14 (14.9)27 (17.9)

The overall male to female ratio in our sample was 4:1. The prevalence of coprophenomena was 19.3% (17.6% of children and 28.6% of adults). In this sample the rate of coprolalia ever was 18.5% (16.9% in children, 27.5% in adults), of copropraxia ever was 5.7% (5.8% in children, 5.5% in adults), and of both symptoms was 4.0%. Thus, coprolalia was roughly three times as frequent as copropraxia.

Coprolalia had been present in 19.3% of males and 14.6% of females, and copropraxia in 5.9% of males and 4.9% of females (these sex differences are not statistically significant). There were 91 adults, of whom 26 (28.6%) had had coprophenomena (i.e. 71.4% did not report a history of any coprophenomena); of those 26 adults, five also had copropraxia. (Examples of copropraxia were making masturbatory gestures, pelvic thrusts, and ‘giving the finger.’)

Mean age at onset of coprolalia was 11 years 8 months (SD 7y 1mo, median 9y 6mo, range 3–49y; n=97). Mean age at onset of copropraxia was 10 years 6 months (SD 5y 6mo, median 10y, range 2–35y; n=34). A minority of patients developed coprophenomena after the age of 18 years (coprolalia n=9, 9.3%; copropraxia n=2, 6.1%).

Only 15 of 220 individuals (7.3%) with tics rated mild had coprolalia, whereas, among the 108 patients for whom tics were rated severe, this was much more common (n=46, 42.6%, p<0.001).

In relation to tic onset, there was a moderately strong (positive) relationship with coprophenomena onset (Spearman’s rho correlation coefficient 0.575, p<0.001, n=99). Age at tic onset correlated with age at onset of both spitting (rho 0.442, p<0.001, n=73) and smelling non-food objects (rho 0.416, p<0.001, n=103). Other repetitive behavior patterns analyzed also showed significant associations between ages at onset of the repetitive behavior and coprophenomena onset: spitting (rho 0.732, p<0.001, n=31); bruxism (0.703, p<0.001, n=25); self-hitting (0.652, p<0.001, n=36); joint-cracking (0.470, p=0.013, n=27); and smelling (0.441, p=0.007, n=36).

Mean age at onset for coprolalia was 5 years 4 months after onset of tics (SD 5y 10mo, median 4y, range 0–41y; n=94). For copropraxia, the mean time lag was 4 years 10 months (SD 5y 5mo, median 4y, range 0–31y; n=32). In about half of the child patients, coprophenomena were reported to have run an intermittent course or had ceased, and in the other half the pattern had been present continuously. Continuous or intermittent occurrence of coprophenomena did not make a significant difference in the proportion that had taken tic-suppressing medication (70%). Of the 118 with a history of any coprophenomena, coprolalia occurred without any history of copropraxia in 84 patients (71%), and copropraxia without a history of coprolalia in 10 patients (9%). In 12 of 110 patients (10.9%) coprolalia was present at time of tic onset; for copropraxia this was true for four of 34 patients (11.8%).

An important question is whether coprophenomena that start early are more likely to persist. With a cross-sectional data set we could only examine reported duration of coprophenomena in those in whom it had stopped, in relation to age at onset of their symptoms (e.g. <5y, 5–10y, 11–15y). The numbers available for this analysis were relatively small, but no difference was found using this crude indicator.

There was an important positive correlation between coprophenomena and the number of comorbid conditions, for both children and adults (one-way intergroup ANOVA df=1, F=23.56, MS=45.12, p<0.001). The relationship in children is shown in Figure 1. Coprophenomena patients had a mean comorbidity score of 2.31 (SD 1.49, median 2.0); patients without coprophenomena had a mean comorbidity score of 1.62 (SD 1.36, median 1.0). A t-test (equal variances not assumed) showed t=–4.59 (df=170, p<0.001), with a mean difference of –0.688 (standard error [SEM] 0.150, 95% confidence intervals [CI] –0.984 to –0.392).


Figure 1.  Relationship of number of repetitive behaviors (black) and comorbidity score (white) to coprophenomena in children with Tourette syndrome.

Download figure to PowerPoint

The cross-tabulation associations of several comorbid diagnoses and behavior patterns with coprophenomena in children and adults are shown in Table IV. The associations found across both age ranges were spitting and self-injurious behavior. In children only, sexually inappropriate behaviors, obsessive–compulsive disorder, smelling non-food objects, anger control problems and oppositional-defiant disorder/conduct disorder were significant, whereas attention-deficit–hyperactivity disorder and sleep problems were significant in adults.

Table IV.   Association of comorbid diagnoses and behavior patterns with coprophenomena in children and adults with Tourette syndrome
ItemWith coprophenomenaWithout coprophenomenap (2-tailed)OR95% CIPhib
  1. Except where otherwise indicated, data are numbers (%) of patients. aChildren defined as age <18y; bmeasure of strength of association (and effect size) for 2×2 contingency tables; cCramer’s V; NA, not applicable. OR, odds ratio; CI, confidence interval.

Sexually inappropriate behaviors11 (12)6 (1)<0.0019.243.32–25.680.23
Medication for tics (ever)63 (69)107 (29)<0.0016.213.80–10.160.35
Spitting30 (33)37 (9)<0.0015.012.88–8.710.27
Smelling33 (36)64 (16)<0.0013.061.85–5.060.20
Obsessive-compulsive disorder24 (26)49 (12)<0.0012.621.51–4.560.16
Self-injurious behavior40 (44)97 (23)<0.0012.511.57–4.030.17
Anger control problems35 (38)91 (22)0.0012.181.35–3.530.14
Oppositional-defiant disorder/conduct disorder16 (17)38 (9)0.0212.081.11–3.930.10
Attention-deficit–hyperactivity disorder52 (57)191 (46)0.0711.520.96–2.390.08
Sleep problems24 (26)94 (23)0.4881.200.72–2.020.03
Tics severe33 (36)43 (10)<0.001NANA0.31c
Adults (n=91)n=26n=65    
Spitting8 (31)6 (9)0.0104.371.34–14.260.27
Attention-deficit–hyperactivity disorder18 (69)23 (35)0.0034.111.55–10.900.31
Sleep problems12 (46)13 (20)0.0123.431.28–9.150.27
Self-injurious behavior17 (65)27 (42)0.0402.661.03–6.850.22
Sexually inappropriate behaviors2 (8)2 (3)0.3222.630.35–19.700.10
Smelling9 (35)13 (20)0.1412.120.77–5.820.15
Obsessive-compulsive disorder10 (39)17 (26)0.2461.770.67–4.630.12
Medication for tics (ever)19 (73)40 (62)0.2981.700.62–4.610.11
Anger control problems8 (31)14 (22)0.3531.620.58–4.500.10
Tics severe16 (62)16 (25)0.003NANA0.36c

Anger control problems at the time of evaluation were associated with copropraxia in 18 of 34 patients (53%), which is much higher than with coprolalia (38 of 110 patients, 34.5%). The rate of occurrence when both coprolalia and copropraxia had been present was 12 of 24 patients (50%, p=0.001, phi=0.18).

Of the 129 TS-only group, just 13 reported coprolalia (10.1%), and three reported copropraxia (2.4%); by contrast, 98 of 466 TS+ group (21.0%) had had coprolalia, and 31 (6.7%) had had copropraxia (p=0.028).

The differences in prevalence of inappropriate sexual behaviors were highly significant (21.9% in those with copropraxia, but only 9.2% in those with coprolalia, and 1.9% in Tourette syndrome patients without coprophenomena, p<0.001, Cramer’s v=0.24).

The number of non-tic repetitive behaviors was strongly positively associated with coprophenomena (one-way intergroup ANOVA df=1, F=44.03, MS=174.95, p<0.001; Fig. 1). Coprophenomena cases had a mean number of repetitive behaviors of 3.24 (SD 2.55, median 3.0, interquartile range 4); those without coprophenomena had a mean number of repetitive behaviors of 1.87 (SD 1.84, median 1.0, interquartile range 2); a t-test (equal variances not assumed) showed t=–5.46 (df=145, p<0.001), with a mean difference of –1.37 (SEM 0.251, 95% CI –1.87 to –0.87).

For patients with Tourette syndrome with a history of spitting, 38 of 81 (46.9%) had had coprophenomena, compared with 15.3% in those without spitting (p<0.001, phi=0.272, odds ratio [OR] 4.9, 95% CI 3.0–9.0). Similarly, in those with a history of smelling non-food objects, 42 of 119 (35%) had had coprophenomena, compared with 5.9% of those without this history (p<0.001, phi 0.20, OR 2.9, 95% CI 1.8–4.5).

From the bootstrap logistic regression variable selection analysis, coprophenomena were correlated with tic severity, number of non-tic repetitive behaviors, spitting, and sexually inappropriate behavior, and less strongly with smelling non-food objects, self-hitting, social skill deficits, and comorbidity. The number of repetitive behaviors was strongly related to coprophenomena when used as a surrogate for the individual behaviors. Stuttering, trichotillomania or hair-pulling, attention-deficit–hyperactivity disorder, specific learning disability, anxiety, and mood disorder were not commonly selected into the bootstrap models, indicating that they do not have strong relationships with coprophenomena. The bootstrapping model selection results are shown in Table V.

Table V.   Results of bootstrapping validation of coprophenomena associations
VariableMean percentage selection in 10 000 resamples
  1. aWhen component variables are excluded from analysis; bwhen relevant summary measure (comorbidity score or number of repetitive behaviors) is excluded from the analysis.

Tic severity (mild, moderate, severe)100
Non-tic repetitive behavior29 (96a)
Spitting85 (95b)
Inappropriate sexual behavior78
Self-hitting44 (57b)
Smelling non-food objects36 (51b)
Social skill deficits45
Comorbidity score33 (43a)
Obsessive–compulsive disorder24 (47b)
Urges (to contradict commands)26
Specific learning disability15 (5b)
Anger control problems10
Sleep problems9
Attention-deficit–hyperactivity disorder8
Oppositional-defiant disorder7
Mood disorder6
Child or adult6
Age at tic onset6
Anxiety disorder3


  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusion
  7. References
  8. Additional co-authors

Gilles de la Tourette apparently believed that Tourette syndrome was a progressive condition, with coprolalia a cardinal symptom: all patients would eventually display it.39 Although some of that concept may survive today in clinical practice, when inexperienced clinicians are reluctant to make the diagnosis in its absence, our data are consistent with other contemporary studies of phenomenology, finding that most people with Tourette syndrome never experience coprophenomena. Furthermore, Tourette syndrome symptom severity is usually not progressive in adult life.40,41

Higher prevalence figures for obsessive–compulsive disorder or behavior, mood disorder, self-injurious behavior, and medication use in adults may be accounted for by their lifetime prevalence, as shown in Table II. Lower adult figures for specific learning disability, oppositional-defiant disorder/conduct disorder, and developmental coordination disorder are probably related to lack of familiarity among patients, and perhaps clinicians, with diagnostic criteria applied to adults, as well as historically less well established awareness of these disorders when the adult patients in our study were children. The reduction in the male:female ratio of Tourette syndrome in adult life merits additional research. Children are brought by their parents (who provide most of the clinical information), whereas adults typically bring and speak for themselves, so referral biases (as well as clinical problems) may be expected to differ.

Age at coprophenomena onset in our series was within that cited by Shapiro et al.5 and other studies listed in Table I. We found that coprolalia could start as late as 33 and 49 years of age in the two extreme cases, but the range for copropraxia was much narrower (the latest was at 19 years); this could be due to the smaller number of copropraxia cases. Coprolalia was much more likely than copropraxia to occur as the exclusive expression of coprophenomena. No adult had copropraxia alone, a finding also reported by Robertson et al.42

Our data enhance a limited body of multinational research supporting an understanding that age at tic onset in Tourette syndrome is similar worldwide (6–7y), a finding already illustrated by an earlier analysis of the TIC database.2 The age at onset of coprolalia, however, is much broader (range 8–15y). Reported frequencies of the two categories of coprophenomena (8–50% for coprolalia and 4.5–25% for copropraxia) are quite variable. Our own findings are on the low side of these ranges, perhaps reflecting less stringent referral criteria among some of the larger participating sites and the trend toward identification of milder cases.

In Table IV, several major differences between those with and without coprophenomena merit consideration. Separate analysis by sex and age range did not account for these differences. Higher rates of self-injurious behavior and anger control problems may be, at least partly, associated with the general pattern of increase in behavioral problems positively correlated with number of comorbid conditions.2 Increases in the presence of coprophenomena associated with obsessive–compulsive disorder are more speculative, but might reflect a state of heightened arousal in patients with the latter symptoms.

We were interested in determining whether coprophenomena have a usual or typical course that could be identified. An identifiable course may help shape patient counseling and structure intervention programs. Our finding of the latency in coprophenomena onset after initial tic onset (5y 7mo for coprolalia, 5y 4mo for copropraxia) is similar to that reported by both Robertson et al.4 and Goldenberg et al.23 As coprophenomena probably represent severe tic expressions, this temporal latency is consistent with recent research showing that, during the preadolescent period, tics tend to be most severe and expressively fluctuating.40,41 However, there is considerable variability in latency and duration of presence of coprophenomena, with some reported to have occurred intermittently or to have ceased.

The high rate of inappropriate sexual behaviors may relate to multiple factors associated with the decrease in inhibition seen with severe tics, including inappropriate touching.42

The association of coprophenomena with tic severity is not surprising. These symptoms could reflect more complex tics and comorbidity patterns, but, because participating clinicians were not blind to the presence (or history) of these phenomena, some confounding could also have occurred. Severe tics were reported in a higher proportion of adults, whose longer experience with Tourette syndrome confers a higher probability of experiencing these patterns than those typical patients whose tics remit in and after adolescence.

Coprophenomena had remitted by adulthood in about one-third of Shapiro’s group,5 consistent with our findings. Increased association with medication use for tic suppression is to be expected in this group, because of the association of coprophenomena with tic severity. Perceived severity of tics depends on their social impact: when coprophenomena are present, tics probably seem more severe and, therefore, are more likely to be treated.

The association with the number of repetitive behaviors may reflect greater dysregulation of inhibition among those with coprophenomena. Spitting, the most strongly associated pattern, is perceived as highly socially unacceptable and may be considered a copropraxia-equivalent.5 The association with sexually inappropriate behavior suggests the existence of a cluster of patterns that may reflect social disinhibition.

Our finding of an association of anger control problems with copropraxia is similar to that of Robertson et al.4 with aggressive behavior.


The study used cross-sectional data (one-time registration) that were not obtained in a blinded manner or subjected to a determination of interrater reliability. Coprophenomena per se have rather loose definitional boundaries. So-called ‘mental coprolalia’ was specifically excluded. Tic severity is a composite figure, and coprophenomena may represent merely a severe tic manifestation along a spectrum of severity, because coprophenomena were not excluded from the severity definition. No satisfactory global indicator of tic severity (over time) has been developed, and the effects of medication on coprophenomena could not be determined with this methodology. Some data were retrospective, based on reports from parents or patients and were, therefore, subject to recall bias, although arguably less so for coprophenomena, because of their social salience.

An important consideration is how representative this sample was of the population of persons with Tourette syndrome. As these were all cases referred to specialists (psychiatrists, neurologists, and pediatricians), each of whom had different referral patterns, one can assume that milder cases of Tourette syndrome were under-represented. The prevalence of coprophenomena is likely to be lower in a non-referred community sample than reported here. The frequencies of TS-only cases and of tics rated severe and the proportions of children may reflect variable clinical heterogeneity. However, our sample representation was strengthened by the diversity and number of our clinical expert participants, encompassing 15 sites in seven countries and from three different medical specialties.


  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusion
  7. References
  8. Additional co-authors

Although coprophenomena can start early in the course of Tourette syndrome or in later adult life, more typically the symptoms manifest first during the preadolescent period around the time that tics are at their peak severity and greatest instability. Coprophenomena can persist, remit, or resume, but most adults with Tourette syndrome never display them. Coprophenomena are associated with reported tic severity, spitting, number of non-tic repetitive behaviors, and comorbidity (especially obsessive–compulsive disorder). Coprolalia is approximately three times as common as copropraxia. That coprophenomena represent complex tic symptoms is suggested by their associations found in this study as well as their broad ages at onset and variable course.

To our knowledge, ours is the first published report of an important association of coprophenomena with the presence and total number of non-tic repetitive behaviors.

Clinical implications

Our data indicate that a significant minority of patients with Tourette syndrome will develop coprophenomena, usually emerging before adulthood, which may persist. Our findings of a positive association between the occurrence of coprophenomena, number of non-tic repetitive behaviors, and extent of comorbidity may constitute a first step in discovering predictors of these disturbing symptoms. Clinicians should inquire about coprophenomena and repetitive behaviors as dimensions of disorder complexity, to assist in clarifying experiential, functional, and management implications.

We do not have data on the treatment of coprophenomena specifically, nor have we found any in the literature, with the exception of botulinum toxin for severe vocal tics and coprolalia,43,44 and a recent case report on the use of aripiprazole.45 It would be of great practical significance to determine the effect of medication, but that is not feasible with this methodology and will require a carefully controlled prospective study.


  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusion
  7. References
  8. Additional co-authors
  • 1
    Coffey BJ, Biederman J, Geller D, et al. Reexamining tic persistence and tic-associated impairment in Tourette’s disorder: findings from a naturalistic follow-up study. J Nerv Ment Dis 2004; 192: 77680.
  • 2
    Freeman RD, Fast DK, Burd L, et al. An international perspective on Tourette syndrome: selected findings from 3500 individuals in 22 countries. Dev Med Child Neurol 2000; 42: 43647.
  • 3
    Van Lancker D, Cummings JL. Expletives: neurolinguistic and neurobehavioral perspectives on swearing. Brain Res Rev 1999; 31: 83104.
  • 4
    Robertson MM, Trimble MR, Lees AJ. The psychopathology of the Gilles de la Tourette syndrome: a phenomenological analysis. Br J Psychiatry 1988; 152: 38390.
  • 5
    Shapiro AK, Shapiro ES, Young JG, Feinberg TE, editors. Gilles de la Tourette Syndrome, 2nd edn. New York: Raven Press, 1988.
  • 6
    Singer C. Tourette syndrome: coprolalia and other coprophenomena. Neurol Clin 1997; 15: 299308.
  • 7
    Gilles de la Tourette G. Étude sur une affection nerveuse caractérisée par de l’incoordination motrice accompagnée de écholalie et de coprolalie (jumping, latah, myriachit). Archives de Neurologie 1885; 9: 1942, 158–200.
  • 8
    Kushner HI. A Cursing Brain? The Histories of Tourette Syndrome. Cambridge, MA: Harvard University Press, 1999.
  • 9
    Kushner HI. Medical fictions: the case of the cursing marquise and the (re)construction of Gilles de la Tourette’s syndrome. Bull Hist Med 1995; 69: 22454.
  • 10
    Mink JW. The basal ganglia and involuntary movements: impaired inhibition of competing motor patterns. Arch Neurol 2003; 60: 136568.
  • 11
    Singer HS, Minzer K. Neurobiological issues in Tourette’s syndrome. In: KurlanR, editor. Handbook of Tourette’s and Related Tic and Behavioral Disorders, 2nd edn. New York: Marcel Dekker, 2005, 273317.
  • 12
    Singer HS. Tourette’s syndrome: from behavior to biology. Lancet Neurol 2005; 4: 14959.
  • 13
    Stern E, Silbersweig DA, Chee K-Y, et al. A functional neuroanatomy of tics in Tourette syndrome. Arch Gen Psychiatry 2000; 57: 74148.
  • 14
    Kurlan R, Daragjati C, Como PG, et al. Non-obscene complex socially inappropriate behavior in Tourette’s syndrome. J Neuropsychiatry Clin Neurosci 1996; 8: 31117.
  • 15
    Pitman RK, Jenike MA. Coprolalia in obsessive-compulsive disorder: a missing link. J Nerv Ment Dis 1988; 176: 31113.
  • 16
    Eapen N, Yakely JW, Robertson MM. Obsessive-compulsive disorder and self-injurious behavior. In: KurlanR, editor. Handbook of Tourette’s and Related Tic and Behavioral Disorders, 2nd edn. New York: Marcel Dekker, 2005, 3988.
  • 17
    Jankovic J, Kwak C. Tics in other neurological disorders. In: KurlanR, editor. Handbook of Tourette’s and Related Tic and Behavioral Disorders, 2nd edn. New York: Marcel Dekker, 2005, 17394.
  • 18
    Cardoso F, Veado CC, De Oliveira JT. A Brazilian cohort of patients with Tourette’s syndrome. J Neurol Neurosurg Psychiatry 1996; 60: 20912.
  • 19
    Chang HL, Tu MJ, Wang HS. Tourette’s syndrome: psychopathology in adolescents. Psychiatry Clin Neurosci 2004; 58: 35358.
  • 20
    Comings DE, Comings BG. Tourette syndrome: clinical and psychological aspects of 250 cases. Am J Hum Genet 1985; 37: 43550.
  • 21
    Eapen V, Fox-Hiley P, Banerjee S, Robertson M. Clinical features and associated psychopathology in a Tourette syndrome cohort. Acta Neurol Scand 2004; 109: 25560.
  • 22
    Fen CH, Barbosa ER, Miguel EC. [Gilles de la Tourette syndrome: a clinical study on 58 cases]. Arq Neuropsiquiatr 2001; 59: 72932 (In Portuguese).
  • 23
    Goldenberg JN, Brown SB, Weiner WJ. Coprolalia in younger patients with Gilles de la Tourette syndrome. Mov Disord 1994; 9: 62225.
  • 24
    Jankovic J, Rohaidy H. Motor, behavioral and pharmacologic findings in Tourette’s syndrome. Can J Neurol Sci 1987; 14: 54146.
  • 25
    Kano Y, Ohta M, Nagai Y. Differences in clinical characteristics between Tourette syndrome patients with and without ‘generalized tics’ or coprolalia. Psychiatry Clin Neurosci 1997; 51: 35761.
  • 26
    Lees AJ, Robertson M, Trimble MR, Murray NM. A clinical study of Gilles de la Tourette syndrome in the United Kingdom. J Neurol Neurosurg Psychiatry 1984; 47: 18.
  • 27
    Miranda M, Menendez P, David P, Troncoso M, Hernandez M, Chano P. [Tics disease (Gilles de la Tourette syndrome): clinical characteristics of 70 patients] Rev Med Chil 1999; 127: 148086 (In Spanish).
  • 28
    Robertson MM, Banerjee S, Kurlan R, et al. The Tourette Diagnostic Confidence Index: development and clinical associations. Neurology 1999; 53: 210812.
  • 29
    Robertson MM, Banerjee S, Eapen V, Fox-Hiley P. Obsessive-compulsive behaviour and depressive symptoms in young people with Tourette syndrome: a controlled study. Eur Child Adolesc Psychiatry 2002; 11: 26165.
  • 30
    Teive HA, Germiniani FM, Della Coletta MV, Werneck LC. Tics and Tourette syndrome: clinical evaluation of 44 cases. Arq Neuropsiquiatr 2001; 59: 72528.
  • 31
    Burd L, Freeman RD, Klug MG, Kerbeshian J. Tourette syndrome and learning disabilities. BMC Pediatr 2005; 5: 34.
  • 32
    Burd L, Freeman RD, Klug MG, Kerbeshian J. Variables associated with increased tic severity in 5,500 participants with Tourette syndrome. J Dev Phys Disabilities 2006; 18: 1324.
  • 33
    Tourette Syndrome Classification Study Group. Definitions and classification of tic disorders. Arch Neurol 1993; 50: 101316.
  • 34
    American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 3rd Edn (revised). Washington, DC: American Psychiatric Press, 1987.
  • 35
    American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th Edn. Washington, DC: American Psychiatric Press, 1994.
  • 36
    American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th Edn (text revision). Washington, DC: American Psychiatric Press, 2000.
  • 37
    Austin PC, Tu JV. Automated variable selection methods for logistic regression produced unstable models for predicting myocardial infarction mortality. J Clin Epidemiol 2004; 57: 113846.
  • 38
    Austin PC, Tu JV. Bootstrap methods for developing predictive models. Am Stat 2004; 58: 13137.
  • 39
    Lajonchere C, Nortz M, Finger S. Gilles de la Tourette and the discovery of Tourette syndrome. Includes a translation of his 1884 article. Arch Neurol 1996; 53: 56774.
  • 40
    Leckman JF, Zhang H, Vitale A, et al. Course of tic severity in Tourette syndrome: the first two decades. Pediatrics 1998; 102: 1419.
  • 41
    Bloch MH, Peterson BS, Scahill L, et al. Adulthood outcome of tic and obsessive-compulsive symptom severity in children with Tourette syndrome. Arch Pediatr Adolesc Med 2006; 16: 6569.
  • 42
    Robertson MM. Tourette syndrome, associated conditions and the complexities of treatment. Brain 2000; 123: 42562.
  • 43
    Porta M, Maggioni G, Ottaviani F, Schindler A. Treatment of phonic tics in patients with Tourette’s syndrome using botulinum toxin type A. Neurol Sci 2004; 24: 42023.
  • 44
    Trimble MR, Whurr R, Brookes G, Robertson MM. Vocal tics in Gilles de la Tourette syndrome treated with botulinum toxin injections. Mov Disord 1998; 13: 61719.
  • 45
    Djebara MB, Worbe Y, Schüpbach M, Hartmann A. Aripiprazole: a treatment for severe coprolalia in ‘refractory’ Gilles de la Tourette syndrome. Mov Disord 2008; 23: 43840.

Additional co-authors

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusion
  7. References
  8. Additional co-authors

Additional co-authors from the Tourette Syndrome International Database Consortium include: Joseph Jankovic MD, Baylor College of Medicine, Houston, TX, USA; Thomas Loughin PhD, Simon Fraser University, Surrey, BC, Canada; Piotr Janik MD, Medical University of Warsaw, Poland; Gary Shady PhD, Department of Clinical Health Psychology, University of Manitoba, Winnipeg, MB, Canada; Mary M Robertson MD, St Georges Hospital & Medical School, London, UK; Anthony E Lang MD, University of Toronto, ON, Canada; Cathy Budman MD, New York University School of Medicine, NY, USA; Alida Magor MD, Child & Adolescent Outpatient Community Clinic, Kupat Holim Meuhedet, Israel; Ruth Bruun MD, New York University Medical Center, NY, USA; Cheston M Berlin Jr MD, Milton S Hershey Medical Center College of Medicine, Hershey, PA, USA.