Cognitive behavioural group therapy for social phobia: Evidence of transportability to community clinics
Highlights
► CBGT for social phobia was compared across research and community settings. ► Community patients were more severe and complex than research patients. ► Symptoms and life interference improved to a similar degree across settings. ► Being younger, single, and having depression were associated with dropout. ► CBGT is transportable to real world community clinics.
Introduction
Social phobia is characterised by a fear of negative evaluation within social or performance situations, where individuals believe that they are under scrutiny and may be embarrassed (American Psychiatric Association, 1994). Social phobia is common, persistent, debilitating, and has one of the earliest onsets of all anxiety disorders, resulting in it being temporally primary to many subsequent comorbid anxiety, affective and substance use disorders (Andrews, Henderson, & Hall, 2001; McEvoy, Grove, & Slade, 2011; Wittchen & Fehm, 2003). Cognitive behavioural individual (CBIT) and group (CBGT) therapy have been shown to be efficacious for social phobia in randomised controlled trials (RCTs, Butler, Chapman, Forman, & Beck, 2006; Jørstad-Stein & Heimberg, 2009; Rapee, Gaston, & Abbott, 2009). However, RCTs are difficult to conduct within real world clinics and the transportability of efficacious treatments to community populations cannot necessarily be assumed.
Efficacy trials prioritise internal validity and are a crucial first step to establishing the credentials of treatment protocols. However, there are a number of potential threats to the generalisability (i.e., external validity) of outcomes from well-controlled efficacy trials to community clinics. Research trials typically select patients based on strict inclusion criteria, use highly trained and closely supervised specialist clinicians, and closely follow manualised treatment protocols. In contrast, effectiveness studies, which examine whether efficacious treatments result in comparable outcomes in naturalistic settings, use clinically representative patients (e.g., highly comorbid, severe, referred by health practitioners rather than through advertisements or self-referral), therapists (e.g., not formally trained or supervised by the protocol developers, broad caseload rather than specialising in the target disorder, various levels of experience, unmonitored protocol adherence), and services (e.g., private practice, community mental health clinics, primary care; Stewart & Chambless, 2009). Dissemination of evidence-based treatments has probably been impeded by the perception that RCTs are not representative of real world circumstances (Barlow, Levitt, & Bufka, 1999), so it is important to demonstrate that efficacious treatments are robustly effective across various treatment settings.
Evidence for the transportability of CBT for social phobia to community clinics is accumulating for CBIT (Lincoln et al., 2003) and CBGT (Marom, Gilboa-Schechtman, Aderka, Weizman, & Hermesh, 2009; McEvoy, 2007), and there is evidence that CBGT is equally effective in research and private practice settings (Gaston, Abbott, Rapee, & Neary, 2006). Moreover, some effectiveness researchers have found that applying exclusion criteria used in efficacy trials has no effect on treatment outcomes (Lincoln et al., 2003; McEvoy, 2007), thus supporting the generalisability of efficacy trial outcomes. In contrast, two meta-analyses of CBT for social phobia across research and naturalistic settings have found an association between the number of laboratory treatment conditions and effect sizes, with studies using an array of efficacy trial restrictions having better outcomes (Lincoln & Rief, 2004; Stewart & Chambless, 2009). This finding suggests that treatment within clinically representative settings may be less effective than within well controlled research settings, and reinforces the need to directly compare outcomes across settings to demonstrate equivalence.
The evidence for the transportability of CBT for social phobia is promising, but limitations in the existing effectiveness literature remain. One limitation is that some effectiveness studies have failed to directly compare outcomes to previous benchmarks (e.g., Haug et al., 2000), so it is unclear whether outcomes vary across settings. A second limitation is that some effectiveness studies have used protocols without demonstrated efficacy within well controlled trials. Whilst modifying evidence-based protocols to better fit patient or service requirements (e.g., inpatient vs. outpatient, session duration, program length, clinician availability) is consistent with the philosophy of effectiveness research, it nonetheless makes it difficult to directly attribute treatment gains to the strategies in the manual. For instance, McEvoy (2007) used a benchmarking strategy to compare effect sizes from an uncontrolled study in a community mental health clinic (CMHC) to previous efficacy and effectiveness trials of both CBIT and CBGT. This study found that effect sizes were within the range of previous efficacy and effectiveness studies for both treatment formats. However, while the treatment protocol in McEvoy's (2007) study contained similar core components to efficacious CBGT protocols, the manual had not been previously evaluated within an efficacy trial and no comparison group was used. Likewise, Marom et al. (2009) found that effect sizes from CBGT in a naturalistic setting compared well to previous studies and meta-analyses, with gains maintaining at one-year follow-up. While a theory- and evidence-guided protocol was used (i.e., cited as being based on an efficacious manual), it was unclear whether the protocol had been previously evaluated in an RCT. A third limitation is that many effectiveness studies contain efficacy trial qualities, such as excluding individuals with comorbidities or recruiting via self-referral or newspaper advertisements (e.g., Haug et al., 2000; Marom et al., 2009). In addition to these limitations recent research has demonstrated that combining theory-driven components with more traditional CBT strategies (referred to as enhanced-CBGT) led to somewhat stronger effects than use of a more traditionally-based CBGT only (Rapee et al., 2009). However, this enhanced protocol is yet to be evaluated within a naturalistic setting. The current study sought to address these limitations by using a benchmarking strategy to directly compare outcomes in community and research settings when the same efficacious, CBGT protocol was used.
The first aim of this study was to compare treatment outcomes across research and community samples using a CBGT manual with demonstrated efficacy (Rapee et al., 2009). Patient, clinician and service factors within a CMHC may reduce treatment effects (Lincoln & Rief, 2004; Stewart & Chambless, 2009). Alternatively, consistent with benchmarking studies conducted to date (Lincoln et al., 2003; Marom et al., 2009; McEvoy, 2007), it may be that CBGT is robustly effective despite differences that exist across service settings. The second aim of this study was to identify sociodemographic and clinical factors associated with (a) treatment attrition, (b) any observed differences in outcomes between the two settings, and (c) post-treatment symptoms. In addition to basic demographic variables (age, gender), other factors previously found to be associated with treatment outcome were examined, including pre-treatment symptoms, pre-treatment impairment, medication use, and comorbid depression.
Section snippets
Participants
Participants in the research unit (RU) comprised the 71 individuals from an RCT comparing the “enhanced” CBGT protocol that was used in this study to standard CBGT and non-specific treatment (Rapee et al., 2009). Five participants did not provide pre-treatment data, so were excluded from all subsequent analyses in this study (total N = 66). All patients had been allocated to the enhanced-CBGT condition and all had a principal diagnosis of social phobia (see Table 1). Most were born in Australia
Demographic comparisons
The two groups were first compared on demographic variables to assess for any significant pre-treatment differences using independent-samples t-tests and chi-square analyses. Bonferroni corrections were not initially applied for these analyses because they would have favoured the hypothesis that the two samples were similar and thus may have obscured important differences. The samples did not significantly differ on age, marital status, or employment status, but they did significantly differ on
Discussion
The first aim of this study was to benchmark outcomes from CBGT for social phobia in a community mental health clinic (CMHC) to outcomes from a research unit (RU) using the same efficacious treatment protocol. There were several indications that the CMHC sample was more complex and debilitated by their social anxiety than those in the RU sample. Compared to the RU sample, the CMHC sample was more likely to use medication for their anxiety, to have more comorbid disorders, and to have a comorbid
References (36)
Diagnostic and statistical manual of mental disorders
(1994)- et al.
Prevalence, comorbidity, disability and service utilisation: overview of the Australian National Mental Health Survey
British Journal of Psychiatry
(2001) - et al.
The dissemination of empirically supported treatments: a view to the future
Behaviour Research and Therapy
(1999) - et al.
Reliability of DSM-IV anxiety and mood disorders: implications for the classification of emotional disorders
Journal of Abnormal Psychology
(2001) - et al.
The empirical status of cognitive-behavioral therapy: a review of meta-analyses
Clinical Psychology Review
(2006) - et al.
Defining empirically supported therapies
Journal of Consulting and Clinical Psychology
(1998) - et al.
Predictors of response to cognitive-behavioral group therapy for social phobia
Journal of Anxiety Disorders
(1997) A power primer
Psychological Bulletin
(1992)- et al.
A comparison of social phobia outcome measures in cognitive-behavioral therapy
Behavior Modification
(1998) - et al.
Anxiety disorders interview schedule for DSM-IV-lifetime version
(1994)
Pre-treatment patient variables as predictors of dropout and treatment outcome in cognitive behavioural therapy for social phobia: a systematic review
Nordic Journal of Psychiatry
Do empirically supported treatments generalize to private practice? A benchmark study of a cognitive-behavioural group treatment programme for social phobia
British Journal of Clinical Psychology
The treatment of social phobia in general practice. Is exposure therapy feasible?
Family Practice
Treatment attrition during group therapy for social phobia
Journal of Anxiety Disorders
Social phobia: an update on treatment
Psychiatric Clinics of North America
Clinical significance: a statistical approach to defining meaningful change in psychotherapy research
Journal of Consulting and Clinical Psychology
The Mini International Neuropsychiatric Interview (MINI). A short diagnostic structured interview: reliability and validity according to the CIDI
European Psychiatry
Impact of depressive symptoms on the treatment of generalized social anxiety disorder
Depression and Anxiety
Cited by (50)
Internet videoconferencing delivered cognitive behavioral therapy for social anxiety disoder: Protocol for a randomized controlled trial
2023, Contemporary Clinical TrialsBrief cognitive behavior group therapy for social anxiety among medical students: A randomized placebo-controlled trial
2021, Asian Journal of PsychiatryCitation Excerpt :Despite available treatments for SAD, their use are limited (Wang et al., 2005). Randomized controlled trials (RCT) (Leichsenring et al., 2013; McEvoy et al., 2012) and meta- analyses (Barkowski et al., 2016; Mayo-Wilson et al., 2014) have shown cognitive- behavioral therapy (CBT) in both individual and group formats to be the treatment of choice for SAD. Cognitive behavioral group therapy (CBGT) has demonstrated; a) statistical equivalence to individual CBT in meta-analyses (Barkowski et al., 2016; Mayo-Wilson et al., 2014; Powers et al., 2007) b) maintenance of treatment gains for an extended period (4–6 years after the conclusion of treatment) (Fogarty et al., 2019; Heimberg et al., 1993); and c) cost-effectiveness (McCarthy et al., 2013).
Do depressive symptoms affect the outcome of treatments for SAD? A meta analysis of randomized controlled trials
2020, Clinical Psychology ReviewTransportability of imagery-enhanced CBT for social anxiety disorder
2018, Behaviour Research and Therapy