Study one

The Australian Tourette’s Survey (Additional file 1) was designed by the primary author for the purpose of a larger study of Australian youth with TS. In addition to demographic questions, the survey included questions that asked parents to identify their child’s formally diagnosed comorbid disorder and the experience of stigma. It also employed three widely used and well-validated psychometric measures (parent proxy versions) relevant to the current project to assess social functioning, peer relationship problems and security of peer attachment of youth in the study.

The single-item Attachment Questionnaire for Children (AQC) [54] was employed to assess peer attachment. The use of this age downward version of the widely employed single-item measure of self-reported adult attachment [55] was necessary due to the absence of multi-item measures of peer attachment suitable for use in middle childhood. Parents were asked to choose which of three narrative descriptions based on the three attachment styles constituting Ainsworth’s [56] typology (secure, insecure-ambivalent and insecure-avoidant) best described their child’s peer attachment. The reliability of the AQC and Hazan and Shaver’s measure has been demonstrated in multiple studies [54, 57, 58]. As the current research sought to determine the impact of secure versus insecure attachment, the two insecure attachment styles were combined to create a dichotomous variable evaluating secure or insecure attachment.

Qualitative studies

All interviews were conducted via telephone by the principal researcher once consent had been obtained. Interviews varied in duration from one hour to 90 min. Participation was limited to one interview per parent, one parent per family, and one child or adolescent with TS. Each recorded interview was then transcribed in full for the purposes of data analysis.

Study one

The principal researcher conducted a content analysis of the survey, employing a method described by Biddle and colleagues [60]. A deductive approach to coding was adopted, with the text read multiple times by the researcher to identify major themes, which were code-named as they emerged.

Data were further reduced by classifying each item related to the youth’s personality according to the “Big Five” (FFM) traits of extraversion, agreeableness, openness, conscientiousness and neuroticism [34, 35]. The validity of the findings regarding personality classification was examined by conducting third party trustworthiness checks (by a rater blind to the purposes of the study) on all of the coded personality data. This resulted in 100 % agreement.

Study two

The researcher employed both an inductive approach to detect emerging themes, as theoretically described by Strauss and Corbin [61], and a deductive approach to content analysis (guided by open-ended research questions grounded in the relevant literature) as described by Miles and Huberman [62]. The major analytic category of the current report was parental perception of the child’s peer attachments. Manual coding of verbatim transcripts was conducted with multiple line-by-line reviews to create a hierarchy from lower to higher order themes [60]. Manual coding with the aim of data reduction continued in an iterative process over multiple months until completion. Data dictionaries, which provided definitions and examples of all of the emerging codes, were composed for each study.

Inter-rater reliability

Two trained raters blind to the purposes of the study conducted inter-rater reliability checks. Using three randomly selected cases from each study, both raters checked all of the coded text data that emerged from Study One and all of the coded interview data from Study Two by referencing the respective data dictionaries. Percentage of agreement and Krippendorff’s alpha coefficient- KALPHA [63] were calculated using ReCAL [64]. 100 % agreement was determined for the overall sample, and all of the examined codes in Study One and Study Two achieved 100% agreement with KALPHA α =1. The high degree of agreement between raters may have been attributable to the clarity and simplicity of the variables in the study, which required minimal interpretation.

Statistical procedure

Data were analysed using SPSS Version 19 (IBM [65]). All continuous outcome and predictor variables were assessed for their suitability for parametric analyses by testing for normality and homogeneity of variance (Levene’s test). This was shown to be adequate for all variables. Outliers were examined and a per comparison critical significance level of α = .01 was applied for all comparisons. For all inferential test results, an R²-type effect size measure is reported (i.e., η_p² for ANOVA analyses).

In Study One, differences between the TS and control groups on the PedsQL and the SDQ subscales were evaluated with univariate ANOVAs. Chi-square analysis was used to test for significant relationships between group membership and responses to the AQC, demographic variables and the ability to form friendships.

The qualitative findings from both studies were cross-tabulated to generate frequency data. Z-tests of proportions were conducted to reveal between group differences in the qualitative findings from Study One, with a critical significance level of α = .01.

Study one

Characteristics of the friendships of youth with TS

Coding and analysis of data in Study Two provided maternal descriptions of the friendships of the current sample of youth with TS (n = 22), with findings suggesting fewer than 20 % enjoyed a ‘typical’ social life. Mothers defined ‘typical’ as having at least one close friend, several peripheral friends and the ability to socialise with classmates and acquaintances. Mothers attributed reduced motivation for peer interaction to factors associated with TS symptoms including social anxiety (n = 3, 14 %), fear of bullying, teasing and rejection (n = 4, 18 %), difficulty maintaining friendships (n = 5, 23 %), inability to spend long periods of time with friends due to efforts to supress tics (n = 6, 27 %) and having a low level of interest in classmates (n = 2, 32 %). Almost a third of the youth in the sample had overtly expressed the desire for more friends to their mother (n = 7, 32 %). Motivation for friendship and romantic relationships was reported to increase for three of the four older adolescents.

Factors impacting peer attachment – parental perspectives

Findings from both studies revealed multiple factors impacting the peer relationships of youth. Parents in Study One identified twenty-two factors; The first being ‘Personality’ (FFM Traits) followed by twenty-one discrete ‘Non-Personality’ factors. Study Two revealed eighteen factors including ‘Personality’ that enhanced peer attachment and seventeen that negatively impacted peer attachment. There was a high degree of homogeneity across the findings from both studies in the factors identified by parents of youth with TS.

Between group differences in factors impacting peer relationships

Non-personality factors

Parents identified many ‘non-personality’ factors that they believed impacted the quality of their children’s peer relationships. These included the adverse effects of increased tic severity and the presence of comorbid disorders on the peer relationships of youth, although standardised assessment of tic severity and the quantitative impact of these variables were not goals of the current study. This is consistent with previous research [4–8, 11, 66, 68].

Current findings also revealed a highly complex role for tics and increased tic severity in impairment in peer relationships. Whilst this research found that simply having tics and increased tic severity are detrimental to peer relationships, distress and dysfunction in peer relationships was more closely related to the youth’s negative cognitive appraisal and their affective and behavioural responses to their tics. Specifically, the degree of self-consciousness experienced in regard to tics, rather than tic severity, appears to be most damaging to the youth’s peer relationships. This is an important finding as it may help to explain some of the individual variability in results of the impact of tic severity on social functioning evident in prior studies.

In addition, parents in Study Two linked increased self-consciousness and an inability to adjust to or accept a diagnosis with highly adverse behavioural and psychological consequences that further alienated individuals with TS from their peers. These included responses such as denial, rage, depression, social anxiety and social withdrawal. Supporting the acuity of these parental observations is the finding that a diagnosis of chronic disorder places youth at a significantly increased risk of adjustment disorder [69].

Parents also identified three more factors with negative impacts directly attributable to the ‘non-tic’ symptoms of their child’s TS and comorbid diagnoses. These included maladaptive symptoms of TS such as aggressiveness, impulsivity, a tendency to dominate peers, to behave bizarrely, incongruently, or to withdraw from or fail to participate socially. These factors have all been shown to have notable adverse effects on interpersonal relationships in previous TS studies [16, 17].

Parents also indicated that some of the behaviours associated with comorbid diagnoses, including inattention, impulsivity, anxiety, obsessiveness, defiance and antisocial behaviour had a significant impact on their child’s relationships. High rates of comorbid diagnoses were evident in both studies. Disorders such as OCD and ADHD have previously accounted for a disproportionate amount of social adversity for individuals with TS [4]. It is however important to note that disentangling behaviours attributable to TS from comorbid disorders, the child’s underlying personality traits, and behaviours that would generally be construed as misbehaviour, was reported as being very challenging for parents in the current study (Study Two). This is also a challenge in both clinical and research settings.

As identified in prior studies [6], negative peer behaviours such as bullying, teasing, social rejection and stigmatisation were reported as being a major impediment to secure peer relationships for many TS youth. The concerns of youth regarding peer responses towards them in this study also extended to a fear of being seen as “different”, “uncool”, “weird”, receiving unwanted peer attention and anxiety that they may frighten or irritate peers with their symptoms. For youth who experience socially embarrassing tics such as coprolalia, these self-perceptions are understandable. However, there is also some suggestion that some individuals with TS may “self-stigmatise” by internalising negative TS stereotypes and adverse social experiences, including being stigmatised by others [70].

Other less frequently identified factors to negatively affect peer relationships included poor social skills and competence, which parents often attributed to comorbid disorders and low social interest, that is a manifestation of the youth’s “eccentricity”, “shyness” or comorbidity. The ability to participate in sport was also important and appears to be of great cultural significance in the Australian context, particularly for males [71]. Non-participation for youth with TS appears to be associated with low interest, physical limitations associated with diagnosis, psychological barriers (e.g., social anxiety), the stress of competition and the youth’s cognitive rigidity. Parents also identified barriers to participation, such as social exclusion by peers, other parents, schools and social organisations.

Both studies also identified multiple ‘non-personality’ factors that enhanced the ability to form secure peer relationships. The most common factor was the youth’s ability to cope with their tics. This referred to a suite of factors including the youth’s successful psychological adjustment to the diagnosis of TS, as well as a lower level of self-consciousness in regard to their tic symptomology. Parents reported that the ability to camouflage, suppress or otherwise manage tics at important times (such as during class or whilst playing with other children), and being in a waning phase of the tic cycle both positively impacted the development of secure relationships with peers.

Other important positive factors included the youth’s ability to be open with others about their TS and comorbid disorders, to self-advocate and educate peers regarding their TS and their ability to defend themselves against the adverse behaviour of others (such as bullying teasing and social rejection). Some of these abilities form components of recent interventions designed to improve outcomes for children and adolescents with TS, particularly in the school setting [72, 73].

Finally, external factors including the acceptance, understanding and support of peers emerged as a key determinate of positive peer relationships, with parents in Study Two identifying this factor more frequently than any other. Some participants also acknowledged that peers needed occasional respite and support in order to preserve friendship with youth with TS. The importance of the attitudes of classmates to the relationships and wellbeing of youth with TS has been demonstrated during early trials of school-based interventions [73].

Personality factors

Personality was the most frequently identified factor impacting peer relationships to emerge from the current research, with over 80 % of Study One respondents attributing the quality of their child’s peer relationships to at least one personality dimension. Although there were no between-group differences in the frequency with which parents nominated personality, the control group were more likely to refer exclusively to FFM personality traits.

Of the FFM traits, Extraversion and Agreeableness, and to a limited degree Openness (to experience), appear to help youth with TS to overcome the significant barriers to friendship that they experience, as well as counterbalance the stigmatising, alienating and disturbing impact of their symptoms. As noted in the literature, Extraversion and Agreeableness have direct positive effects on peer relationships, with both linked to more positive peer representations. Extroverts are valued for their sociability, drive and energy, and this trait is the strongest predictor of friendships longitudinally [36, 37]. Those high in Agreeableness are appreciated for their caring, loving and empathic qualities, and Agreeableness predicts the highest rate of peer acceptance and reciprocity of friendship [36, 37].

Both Extraversion and Agreeableness may also indirectly enhance peer relationships by increasing resilience, adaptive coping and attracting social support. An extrovert’s assertiveness and social skills may help them to overcome the often reported negative behaviour of peers experienced by individuals with TS [6, 11, 15, 19], and enhance self advocacy skills. Agreeableness may help youth to recruit peer support and understanding [36, 37]. However, extroversion was also linked to problems with peers by some TS group parents. Furthermore, it has been suggested that the increased drive of extroverts for friendship may expose individuals high in this trait to increased psychological risk associated with rejection [74]. Openness to experience may extend the social networks of youth, which were found to be limited in this study, and the opportunities these provide for relationships facilitated by shared values and interests [36, 37, 75].

Consistent with the literature, the results of the current study indicated that Neuroticism was identified by parents as being commonly associated with the inability to form friendships in individuals with TS. It appears to share the closest theoretical link to insecure attachment, as both insecure attachment and Neuroticism are characterised by emotional dysregulation and negative affect [27, 28, 34, 35]. Previous research has also suggested that Neuroticism has the most adverse impact on the ability to maintain friendships [37], a problem that was identified for a significant minority of youth in Study Two. Emotional dysregulation, as well as a decreased ability to interpret social cues associated with Neuroticism, has also been shown to result in increased relationship conflict [76]. Finally, Neuroticism has been associated with decreased resilience and the impaired ability to cope with adversity in the context of chronic disorders [41], both of which may adversely affect social functioning and peer relationships.