نقش کمال گرایی در نتایج رفتار درمانی شناختی برای کودکان بالینی مضطرب

The main aim of this study was to determine whether pre-treatment levels of child perfectionism impacted on anxiety treatment outcomes for school-aged children. In addition, it was investigated whether child perfectionism decreased following treatment for anxiety. Participants were sixty-seven clinically anxious children aged 6–13 years (female = 34; majority Caucasian) who were enrolled in a group-based cognitive behaviour therapy program, and their parents. They completed self-report questionnaires on anxiety and depressive symptoms and were administered a diagnostic interview to determine the type and clinician rated severity of anxiety and related disorders pre- and post-treatment and at 6-month follow-up. Self- and parent-rated perfectionism were also measured pre-treatment, while a subset of children completed perfectionism measures post-treatment as well. Self-Oriented Perfectionism, but not Socially Prescribed Perfectionism, predicted poorer self-reported treatment outcome (higher levels of anxiety symptoms) immediately following treatment and at 6-month follow-up when using a multi-informant approach. Additionally, both Self-Oriented and Socially Prescribed child perfectionism significantly reduced immediately following treatment. Despite reductions in child perfectionism following anxiety treatment, higher Self-Oriented Perfectionism may impact negatively on child anxiety treatment outcome.

Anxiety disorders are the most prevalent childhood psychopathology, with community 3–6 month prevalence rates of 3–17.7% (Costello et al., 2003 and Green et al., 2004, pp. 23–70; Maughan, Brock, & Ladva, 1999, pp. 3–17), and are associated with considerable costs in terms of social, academic, and family impairment and distress (Bodden et al., 2008 and Green et al., 2004, pp. 23–70; Kessler, Foster, Saunders, & Stang, 1995). Cognitive behavioural treatment (CBT) programs have demonstrated reductions in both anxiety and its consequences, compared with wait list conditions (Cartwright-Hatton et al., 2004 and Reynolds et al., 2012) and general support conditions (Hudson et al., 2009). However, not all children demonstrate significant clinical benefits from treatment programs. That is, more than 40 percent of children are not diagnosis-free from their primary anxiety disorder immediately following CBT (Cartwright-Hatton et al., 2004). A more recent meta-analytic review also revealed that the post-treatment gains for CBT for childhood anxiety disorders do not seem to be maintained at 6- and 12-month follow-ups (Reynolds et al., 2012). That is, CBT treatment groups were not significantly different from waitlist and active controls at longer-term follow-ups. This limitation of existing treatment programs for child anxiety has led to consideration of issues that present obstacles to treatment and predict poor treatment outcome. A number of child and parent factors have been implicated in child anxiety treatment outcomes (see review by Rapee, Schniering, & Hudson, 2009). One factor generating considerable recent interest is perfectionism. Perfectionism typically refers to a trait that involves the pursuit of personally demanding and high standards for performance (Flett & Hewitt, 2002). Moreover, a multidimensional conceptualisation of perfectionism that distinguishes between Self-Oriented Perfectionism (SOP; perfectionistic standards directed by oneself towards oneself) and Socially-Prescribed Perfectionism (SPP; the belief that others have perfectionistic expectations and motives for oneself) is widely accepted and researched (e.g., see Hewitt, Flett, Besser, Sherry, & McGee, 2003). Researchers have also suggested that beyond these broad definitions, perfectionism may include dimensions such as excessive doubts about actions, concerns over mistakes, and elevated parental criticism and expectations (Frost, Marten, Lahart, & Rosenblate, 1990). Research in adults suggests that elements of perfectionism are related to maladjustment and psychopathology, including anxiety disorders (Flett and Hewitt, 2002, Frost and DiBartolo, 2002 and Shafran and Mansell, 2001). For instance, SPP and, to a lesser degree, SOP are linked to anxiety (Bieling et al., 2004 and Shafran and Mansell, 2001) and increased numbers of comorbid anxiety disorders (Bieling et al., 2004). Furthermore, patients with elevated pre-treatment severity of obsessive compulsive disorder (OCD; Frost and Steketee, 1997 and Sassaroli et al., 2008) and social anxiety (Antony et al., 1998, Juster et al., 1996 and Rosser et al., 2003) also report being more concerned about making mistakes and having more doubts about their actions, while social anxiety has been linked to SPP as well (Bieling & Alden, 1997). Given the link between psychopathology and perfectionism, the impact of perfectionism on treatment outcome is an important research question to address. Within the context of treatment for anxiety disorders, rigid adherence to irrational standards among highly perfectionistic individuals can be seen to pose significant obstacles to optimal treatment outcomes. A core component of CBT is the requisite ability to exhibit some cognitive flexibility to carry out thought challenging for distorted beliefs and interpretations. The rigidity and high standards imposed by perfectionistic individuals may mean that these individuals are less likely to be persuaded by the usefulness of cognitive therapy – that is, thought challenging. Moreover, they may also be less likely to believe that more realistic thoughts advocating for reduced standards in performance would be a feasible alternative belief to live by in order to reduce their anxiety. Individuals with high perfectionism also find problem-solving, which is an element of CBT for anxiety disorders, a challenge ( Hewitt & Flett, 2002). Furthermore, Nobel and colleagues have also noted that perfectionistic individuals tend to procrastinate and this can then interfere with CBT homework completion ( Nobel, Manasses, & Wilansky-Traynor, 2012). Finally, high levels of perfectionism can interfere with therapeutic alliance and the quality of social relationships ( Shahar, Blatt, Zuroff, Krupnick, & Sotsky, 2004), which may in turn contribute to poor gains during therapy. The potential adverse impact of perfectionism on anxiety treatment outcomes has gained research attention in the last decades. For example, several studies have now shown that elevated perfectionism is implicated as an obstacle to treatment of adult anxiety disorders (Ashbaugh et al., 2007, Chik et al., 2008 and Lundh and Öst, 2001). Specifically, perfectionism has been shown to affect treatment outcomes in adult social phobia (Ashbaugh et al., 2007, Lundh and Öst, 2001 and Rosser et al., 2003) and OCD (Chik et al., 2008). Lundh and Öst (2001) further demonstrated that treatment non-responders had higher initial perfectionism scores that did not reduce to subclinical levels, while treatment responders had lower initial perfectionism scores that became subclinical following treatment. Notable limitations of this study included the small sample of non-responders, and lack of regression analyses to determine whether perfectionism predicted outcome, after controlling for pre-treatment anxiety scores. Despite the established link between perfectionism and anxiety in adults and some evidence that perfectionism plays a role in adult anxiety treatment outcomes, less is known about perfectionism and anxiety and its treatment in children. Perfectionism has been suggested to emerge in childhood (Flett, Hewitt, Oliver, & Macdonald, 2002) and to be important in the aetiology of some childhood anxiety disorders, particularly OCD (Franklin, Piacentini, & D'Olio, 2007) due to the overlap between perfectionistic and ritualised or compulsive behaviours (Evans, Gray, & Leckman, 1999). Positive correlations have also been reported between perfectionism and OCD symptoms, depressive symptoms, and difficulties in peer relationships (Ye, Rice, & Storch, 2008). Outside of the OCD literature, the association between perfectionism and anxiety has largely been examined in the contexts of academically gifted children (Ablard & Parker, 1997), test anxiety (Tsui & Maziocco, 2007), and anxiety sensitivity (Dekryger, 2005, Flett et al., 2004 and Hewitt et al., 2002). Investigations have found that SOP and SPP, along with concerns over mistakes, doubts about actions, and perfectionistic parental expectations are positively correlated with child anxiety symptoms (Dekryger, 2005, Flett et al., 2004 and Hewitt et al., 2002), mirroring findings in adults. For instance, Hewitt et al. (2002) showed that SOP and SPP were positively associated with depression and anxiety in a group of non-clinical youths aged 10–15 years. Interestingly, Essau, Leung, Conradt, Cheng, and Wong (2008) also demonstrated that the association between perfectionism and anxiety is ubiquitous across cultures among ∼1000 adolescents sampled in Germany and Hong Kong. That is, a significant positive association was found between anxiety and perfectionism (SOP and SPP) in non-clinical adolescents regardless of culture (i.e., country of origin). In spite of the links between child anxiety and perfectionism, and suggestions it may be an important factor in child treatment (Oros, 2005 and Schuler, 2002), only one published study has investigated the impact of perfectionism on treatment outcomes for children with anxious and depressive symptoms (Nobel et al., 2012). In that study, SOP but not SPP significantly reduced following a school-based treatment program designed to target anxiety and depression. Moreover, Nobel et al. also demonstrated that greater pre-treatment SOP, but not SPP, significantly predicted worse post-treatment depression levels after controlling for pre-treatment depression, but the same relationship was not observed for anxiety. However, it is worth noting that this study was limited by the use of a sub-clinical sample, and the exclusion of clinically anxious or depressed children from the treatment. So far, no study has examined the impact of perfectionism on treatment outcomes for clinically-anxious children. Therefore, further examination is required to determine whether these results can be replicated in a clinical child sample. Improving our understanding of the link between perfectionism, anxiety, and treatment outcomes in children may aid in identification of children at risk for developing anxiety problems, and further enhance intervention programs designed to reduce the negative impact of anxiety. Given that childhood traits tend to become increasingly entrenched over the lifespan (Roberts and DelVecchio, 2000 and Sallquist et al., 2009), it seems particularly important to investigate traits and/or behaviours in a child sample to aid the development of effective child anxiety interventions. The paucity of research on perfectionism and anxiety treatment makes the current study an important contribution to the field. The current study capitalises on an ongoing large clinical trial, and differs from Nobel et al.’s (2012) study by recruiting a sample of clinically anxious children enrolled in treatment programs to determine the impact of perfectionism on child anxiety treatment. Given that previous research has shown that SOP but not SPP is related to anxiety difficulties in children (Nobel et al., 2012), we predicted that higher pre-treatment child SOP would predict worse outcomes following anxiety treatment. The current study also examined whether anxiety treatment brought about a change in child perfectionism. Child SOP but not SPP was expected to reduce following treatment.

Pre, post and 6-month follow-up treatment data (means, standard deviations, and Partial Eta2) for anxiety and depression measures are presented in Table 1. Scores on mother and child reports of child's anxiety and clinician rated child anxiety severity all decreased significantly from pre- to post-treatment with all pair-wise comparisons significant (all ps < .01). From post-treatment to 6-month follow-up, only child reports of anxiety decreased significantly (p = .002). There were no significant changes in mood across time (all ps > .05) for clinician severity ratings. However, both child- and maternal-ratings of low mood via the SMFQ revealed a significant decline over time. Follow-up comparisons revealed a significant decline from pre- to post-treatment for both child and maternal reports of mood symptoms (all ps < .01) but not from post-treatment to follow-up.