Print version ISSN 1516-4446
Rev. Bras. Psiquiatr. vol.34 no.1 São Paulo Mar. 2012
Uma revisão sobre fatores de predição de resposta ao tratamento no transtorno de ansiedade social
Sara Costa Cabral MululoI; Gabriela Bezerra de MenezesI; Paula VigneI; Leonardo F. FontenelleI,II
IPrograma de Pesquisa em Ansiedade e Depressão, Instituto de Psiquiatria, Universidade Federal do Rio de Janeiro (IPUB/UFRJ), Rio de Janeiro, Brazil
IIInstituto D'Or de Pesquisa e Ensino, Rio de Janeiro & Departmento de Psiquiatria e Saúde Mental, Instituto de Saúde da Comunidade, Universidade Federal Fluminense (MSM/UFF), Niterói, Brazil
OBJECTIVE: This article aims to review the clinical features and therapeutic characteristics that may predict treatment response in patients with social anxiety disorder (SAD).
METHODS: A systematic review of trials identified through databases of ISI, Medline, PsycInfo, Cochrane, LILACS, Current Controlled Trials, and in references of previously selected articles published in English up to December 2010. In our literature search, we used the words prediction/predictors and social anxiety disorder or social phobia.
RESULTS: Early onset, greater disease severity, comorbidity with other anxiety disorders (including generalized anxiety disorder and simple phobia), and high expectations about the role of the therapist emerged as potential predictors of less effective treatment in SAD.
CONCLUSIONS: Knowledge of various clinical and treatment features may help professionals to predict possible responses to therapeutic interventions in patients with SAD. However, given the diversity of measures used to assess response, further studies should be performed with standardized methods to investigate the aspects related to treatment resistance in SAD.
Descriptors: Phobic Disorders; Predictors; Treatment Outcome; Review; Clinical Trials.
OBJETIVO: Este artigo tem por objetivo descrever as principais características clínicas e terapêuticas que possam predizer resposta ao tratamento em pacientes com transtorno de ansiedade social (TAS).
MÉTODOS: Revisão sistemática de ensaios clínicos identificados através das bases de dados ISI, Medline, PsycInfo, Cochrane, LILACS, Current Controlled Trials e em referências bibliográficas de artigos previamente selecionados publicados em inglês até dezembro de 2010. As seguintes palavras-chave foram utilizadas em nossa busca bibliográfica: prediction/predictors e social anxiety disorder ou social phobia.
RESULTADOS: Início precoce, maior gravidade da doença, comorbidade com outros transtornos de ansiedade (incluindo o transtorno de ansiedade generalizada e fobia simples) e alta expectativa sobre o papel do terapeuta emergiram como potenciais fatores de predição menor eficácia do tratamento do TAS.
CONCLUSÕES: O conhecimento de uma variedade de características clínicas e de tratamento pode auxiliar os profissionais a preverem possíveis respostas às intervenções terapêuticas nos pacientes com TAS. No entanto, devido à diversidade de medidas utilizadas para avaliar a resposta, novos estudos com o objetivo de investigar aspectos relacionados à resistência ao tratamento do TAS devem ser realizados com métodos mais padronizados.
Descritores: Transtornos fóbicos; Fatores de predição; Resultado de tratamento; Revisão; Ensaios clínicos.
Social anxiety disorder (SAD) is characterized by intense and persistent fear of being negatively evaluated by others, which leads to significant avoidance or intense psychological distress in various social situations.1 SAD has high prevalence rates, early and insidious onset, and a chronic course.2 It is associated with low levels of quality of life; significant feelings of incapacity; and notorious social, educational, and occupational disability.3 Despite remaining largely underdiagnosed,4 recent studies are beginning to unveil SAD neurobiological underpinnings, thus leading to greater knowledge regarding its pathophysiology.5
Currently, the most empirically effective treatments of SAD include pharmacotherapy with selective serotonin reuptake inhibitors (SSRIs) or serotonin and norepinephrine reuptake inhibitors (SNRIs), and cognitive behavioral therapy (CBT), either together or separately.6,7,8 Although these strategies lead to a therapeutic response in a large proportion of individuals, a significant number of patients with SAD (up to 50%) does not respond at all or remain significantly symptomatic after being submitted to first-line treatments.9,10,11
This article aims to review the clinical and therapeutic characteristics that have been reported to be associated with response to treatment prescribed to patients with SAD. Potentially, the characterization of these features may save patients' time by avoiding ineffective treatments, which may be sometimes associated with severe side effects and/or economic burden. It could also help to identify factors that should be taken into account if one aims at elaborating more specific treatment for SAD individuals who do not respond to conventional strategies.
A systematic review of clinical trials that describe the profile of patients with SAD who respond to treatment was performed in the following databases: ISI/Web of science, Medline, PsycInfo, Cochrane Library, LILACS, and Current Controlled Trials (articles not published yet). An additional search for articles of interest was performed in the bibliographic list of selected articles. The following keywords were used in the bibliographical search: prediction/predictors and social anxiety disorder or social phobia. Clinical trials originally published in English up to December 2010 were assessed.
Inclusion criteria consisted of studies composed of subjects with a primary diagnosis of SAD according to the DSM-III-R or DSM-IV and describing an assessment of treatment response predictors in this condition. We included papers that addressed the following aspects as potential response predictors: (i) clinical features (age of onset, duration of SAD, severity of symptoms, generalized subtype of SAD, family history of SAD, and psychiatric comorbidity); (ii) sociodemographic characteristics (age, gender, educational level, and marital status); and (iii) treatment features (treatment history, duration of current treatment, expectancy regarding treatment, and group cohesion [in case of group therapy]). For the sake of space, other potential predictors (e.g., personality traits) were not addressed in our review.
Medline search identified 392 articles, of which 10 met the inclusion criteria. In PsycInfo, from the 58 studies identified, 5 were included. The ISI/Web of Science database led to the identification of 679 articles, of which 4 were included. A Cochrane search resulted in 158 studies, of which 11 were selected, including 10 that had already been localized on Medline and Psyinfo, and one that appeared for the first time.
A LILACS search led to only one article, which did not meet the inclusion criteria. Unpublished articles searched on Current Controlled Trials were not found as well. Four additional articles were identified in the reference lists of previously selected articles. These combined search strategies led to a total of 24 selected articles. Most studies (n = 19) reported independent predictors of response using regression analysis, while the remaining ones12-16 used univariate analysis.
Studies describing the impact of different clinical characteristics (including age of onset, duration, symptom severity, subtype, and family history) on SAD treatment outcomes are listed in Table 1.
Age of onset
The impact of the age of SAD onset on treatment outcome was investigated in four studies. In one of them, no impact of age of onset on the response of 57 patients treated with group CBT was found.17 In the second trial with 204 patients treated with sertraline or placebo, patients with late-onset (> 19 years) generalized SAD tended to have a better response to treatment with sertraline than those with early-onset generalized SAD.18 Accordingly, in two additional studies; e.g., one including 102 patients with generalized SAD submitted to individual or group CBT19 and the other including 80 SAD patients treated with individual and group cognitive or interpersonal therapy,20 early (childhood) onset emerged as a predictor of poor response to treatment.
Duration of illness
Duration of SAD was not associated with a worse therapeutic outcome in four studies, including trials with moclobemide,21 brofaromine and fluvoxamine,22 paroxetine,23 and group CBT.17 Only one placebo-controlled trial found baseline duration of generalized SAD to be a significant predictor of poor response to treatment (in this case, sertraline).24 In this study, none of the independent variables were significant as predictors of response to placebo.
The impact of SAD severity on treatment outcome was analyzed in three studies.22,23,25 Increased severity of symptoms predicted worse response in 30 patients treated with brofaromine or fluvoxamine (with higher Symptom Checklist-90 interpersonal sensitivity scores and heart rate predicting 92% of non-responders),22 and in 76 patients submitted to individual or group CBT25 (with greater impairment or Global Clinician rating of Social Phobia severity predicting a slight proportion [4-8%] of treatment outcome). In the third study,23 the severity levels (according to the baseline scores on the Liebowitz Social Anxiety Scale) of 829 SAD patients did not predict differences in outcome after treatment with paroxetine or placebo.
No differences in terms of response to treatment was found between generalized and non generalized SAD patients in three clinical trials: one article with brofaromine or fluvoxamine,22 one involving group CBT,17 and one comparing atenolol to exposure psychotherapy (EXP) with flooding.12 However, in other four studies using group CBT,26,27 moclobemide,21 or a set of interventions (including group CBT, educational support therapy, phenelzine, and placebo pill),28 generalized SAD patients exhibited worse response to treatment compared to non-generalized patients in the post-test. In one of these studies, the generalized subtype was characterized by a trend towards worse short-term response in group CBT, although this difference reached statistical significance after one-year of follow up.26
Only one article suggested that the presence of a family history of SAD predicts worse response to treatment with brofaromine or fluvoxamine in SAD patients.22 Therefore, the relationship between family history and response to treatment should be considered relatively fragile at this moment.
Studies describing the impact of different psychiatric comorbidities (including personality, anxiety and mood disorders, and substance abuse) on SAD treatment outcomes are depicted in Table 2.
Eleven trials assessed the influence of avoidant personality disorder (PD) on treatment response of patients with SAD. In five of these studies, patients with SAD and avoidant PD displayed worse outcome after treatment (including moclobemide,21 atenolol or EXP with flooding,12 group CBT,29 EXP,13 and moclobemide or individual cognitive therapy)30 compared to patients with SAD without avoidant PD. Nonetheless, in one of these studies,30 this difference did not remain significant after 15 months of treatment.
In five additional articles, the comorbidity of SAD with avoidant PD did not predict differences in response to interventions with sertraline,18 EXP,14 individual or group CBT,25 individual or group cognitive therapy, individual or group interpersonal therapy20 and a set of interventions (including group CBT, educational support therapy, phenelzine and placebo pill).28 Finally, in contrast to the results described above, a study with 295 generalized SAD patients reported that patients with avoidant personality disorder (APD) had exhibited greater changes in SAD early in treatment than those without APD.15
Two studies reported that other types of PD (e.g., obsessive-compulsive and dependent) had no influence in response of 84 SAD patients who received either atenolol or EXP with flooding treatment,12 and 93 SAD patients who were treated with moclobemide.21 Histrionic PD co-morbidity with SAD, on the other hand, had a negative impact on response to treatment with atenolol or EXP with flooding.12
A study with 52 SAD patients treated with group CBT plus video feedback of exposures revealed that co-morbidity with any anxiety disorder was associated with a worse therapeutic outcome.31 Similarly, the presence of simple phobia had a negative impact on treatment response of 84 patients with SAD treated with atenolol or EXP with flooding.12 Finally, generalized anxiety disorder (GAD) was associated with worse treatment response of 93 patients with SAD treated with moclobemide,21 and of 84 patients treated with atenolol or EXP with flooding.12
The influence of comorbid mood disorders on the treatment outcome of patients with SAD was associated with mixed results. For instance, in at least three studies,25,26,29 the presence of major depression was associated with worse response of SAD patients treated with group and individual CBT25 or group CBT,26,29 respectively. Similarly, comorbid dysthymia had a negative impact on response to treatment with atenolol or EXP with flooding12 and with moclobemide.21 However, in one of these articles, the negative influence of depression on the therapeutic outcome was not maintained in the 18-month follow up.25
In two trials27,32 no difference was found in the post-test of SAD patients with comorbid depression treated with group CBT. However, in one of these studies,27 patients with SAD and comorbid depression displayed an increased relapse rate after one year. On the other hand, the presence of previous episode of depression in patients treated with moclobemide did not predict any difference in treatment outcome.21
In one of the studies described above,21 the occurrence of alcohol abuse co-morbidity was by far the strongest predictor of worse therapeutic outcome in 93 SAD patients treated with moclobemide.
Studies describing the impact of different sociodemographic features (including age, gender, educational level, and marital status) on SAD treatment outcomes are listed in Table 3.
According to several aforementioned studies, the patient's age did not predict the therapeutic response of SAD patients to sertraline,18 moclobemide,21 paroxetine,23 brofaromine or fluvoxamine,22 group CBT,17,32 and group or individual CBT.19
Similarly, no impact of gender on treatment response of SAD patients was found in most trials reviewed above, including studies that used group CBT,17 group or individual CBT,19 paroxetine,23 moclobemide,21 brofaromine or fluvoxamine.22 In one naturalistic study, no difference regarding treatment outcomes was found when CGI scores of both genders were compared at the end-point after treatment with SSRIs and/ or benzodiazepines.16 However, male gender was associated with worse response of SAD patients to sertraline or placebo18 and to group plus video feedback CBT.31
Educational and marital status had no impact on the response of 57 SAD patients in group CBT in one study,17 and of 102 patients treated with group or individual CBT in another one.19
Studies describing the impact of different therapeutic features on the outcome of SAD (including history, duration and expectancy of current treatment, and group cohesion [in case of group therapy]) are listed in Table 4.
In three aforementioned studies, previous treatment (either pharmacotherapy or psychotherapy) had no impact on the response of SAD patients to moclobemide,21 brofaromine and fluvoxamine,22 and in group CBT.17
Duration of treatment
In a study with 829 SAD patients,23 treatment with paroxetine for 12 weeks was associated with increased response compared to treatment for only 8 weeks. Conversely, another study that used group CBT with a protocol of 11-20 sessions found that the number of treatment sessions did not predict response of 57 patients with SAD.17
Expectancy regarding treatment effectiveness
Low expectancy regarding treatment effectiveness and reduced confidence concerning treatment predicted worse response in studies that comprised 6229 and 234 patients33 with SAD assigned to group CBT, and 80 patients with SAD assigned to cognitive therapy or interpersonal therapy.20
Expectancy regarding the therapist role
In one study, the expectations of 49 SAD patients about the role of the therapist in their treatment did not have any influence on their response in group CBT in the short-term.34 However, in the three-month follow up, patients who attributed greater responsibility for change to their therapists had worse therapeutic response.
Group cohesion in group CBT
The perception of group cohesion of individuals attending group CBT and its impact on treatment outcome were studied in 34 patients with SAD.35 In this study, increases in group cohesion ratings over the course of treatment significantly predicted therapeutic response.
The present work aimed at reviewing the overall profile of individuals with SAD who respond and not respond to standard treatment, with the ultimate purpose of contributing to future research on the development of more effective interventions.36 We found that early onset, increased duration and severity of illness, presence of generalized subtype, positive family history of SAD, and male gender were all predictors of worse response to treatment in at least one study. Among these, early onset and severity of illness were the most replicated findings.
In terms of co-morbidity, major depression, dysthymia, generalized anxiety disorder, simple phobia, avoidant and histrionic PD, and alcohol abuse predicted negative responses to treatment in at least one trial. Among these conditions, anxiety disorders emerged as a predictor of poor therapeutic response in more than one study.
Finally, regarding treatment characteristics, low expectancy regarding treatment efficacy, high expectancy regarding the role of the therapist in treatment, and decreased group cohesion (in case of group cognitive behavioral treatment) were associated with worse therapeutic responses. Among the later, low expectancy regarding treatment efficacy was the most replicated finding.
Importantly, we found a handful of studies suggesting the association between early onset of SAD and poor response to treatment. While this finding provides some support to the view that the early-onset of SAD represents a valid subtype of SAD, with greater psychiatric comorbidity, higher levels of incapacitation, and worse prognosis,19 there is still controversy on this subtype definition, as different studies used different age criteria for "early" or "late" onset. Previous reviews suggested that the unwelcome influence of both age at onset and severity of symptoms on treatment outcome holds true for other anxiety disorders.37
Furthermore, we also found some evidence that greater severity of social anxiety symptoms predicts decreased efficacy in the response of patients with SAD to different types of treatment (both pharmacotherapy and psychotherapy). This finding is particularly relevant when one considers that all studies reporting them focused their analysis on patients who completed treatment (and not on an intention-totreat sample), thus suggesting that decreased treatment efficacy cannot be ascribed to the tendency that patients with more severe symptoms have to discontinue treatment more frequently.
Data supporting an effect of comorbid psychiatric disorders in SAD treatment response are mixed, particularly for depression. It is difficult to explain these findings because, while major depression and dysthymia could perpetuate and strengthen SAD patients negative beliefs about themselves, some of the SAD treatments (including SSRIs and cognitive re-structuring) may effectively treat some forms of depression, particularly those that are secondary. Accordingly, it is interesting to note that many of the comorbid conditions found to predict worse response of SAD to treatment (e.g., alcohol abuse,21 specific phobia,12 histrionic PD12) are not clearly responsive to treatment (e.g., antidepressants) themselves. It is worth noting that findings regarding the role of several comorbid conditions were assessed in only one study, which restrains us to consider these results as conclusive.
Finally, we also found that low expectancy regarding treatment effectiveness,29 which might be related to comorbid depression,38 was associated with worse response to treatment in different studies. Clearly, patients with SAD might believe that the treatment will not work in their case, that they will not be able to follow the therapeutic steps, or that another person will be totally responsible for his improvement. These cognitive distortions should be identified and managed early in the treatment.
Conversely, lower expectations about the role of the therapist were related to better treatment outcomes, particularly in the follow-up sessions.34 Likewise, in one study, increased feelings of group cohesion between participants and therapist in group CBT was associated with better outcomes.35 It would be interesting to test whether add-on strategies (aimed at increasing treatment expectancy, attributing responsibility for change to patients, and stressing the importance of group cohesion in group CBT) increase the effectiveness of treatment provided to patients with SAD.
Indeed, our review has a number of limitations. First, several sociodemographic and clinical variables associated with worse outcomes (e.g., previous treatment, dependent and obsessive compulsive PD comorbidity) had their relationship with outcome assessed by a very small number of studies. Second, as there is no current consensus on what should be the optimal criteria for treatment response in SAD, one should examine our results with caution.12,25,29 It has been argued that the concept of response and remission in SAD must be multidimensional,39 embracing the reappraisal of the diagnosis,40 social anxiety symptoms, and comorbid conditions.36 Of note, none of the reviewed studies adopted such criteria. Third, there is a clear methodological heterogeneity among selected studies, including study design,15,17,23 type of intervention,21,25-27,29,32 and duration of treatment (varying from five or seven days26 to two years).21 Accordingly, one must consider that, regardless of treatment, longer follow-ups are usually associated with substantial reduction in anxiety symptoms, a phenomenon that can be explained by the tendency of SAD individuals to orientate their life choices so that anxiogenic situations become less prominent.41
The recent increase in the number of studies in the field of response and resistance in SAD is complicated by the difficulty of replicating and comparing these studies. Even when they are conducted with similar patients, treatment, and methods, a considerable variety of treatment response and resistance definitions is used.25 As a consequence, the current sociodemographic and clinical profile of SAD patients who respond to treatment remains elusive, as it does not point to a very clear direction.
To date, early onset, severity of illness, comorbid anxiety disorders (particularly generalized anxiety disorder and simple phobia), and low expectancy regarding treatment efficacy emerged as potential correlates and/or predictors of low therapeutic response. Clearly, the field of treatment response needs conceptual and methodological redefinitions if it aims to lead to more solid results. A better knowledge about the profile of the significant proportion of SAD patients who remain resistant to treatment is important for discussing the emergence of novel therapeutic alternatives, both in terms of psychotherapy and pharmacotherapy.
1. Chagas MH, Nardi AE, Manfro GG, Hetem LA, Andrada NC, Levitan MN, Salum GA, Isolan L, Ferrari MC, Crippa JA. Guidelines of the Brazilian Medical Association for the diagnosis and differential diagnosis of social anxiety disorder. Rev Bras Psiquiatr. 2010;32(4):444-52. [ Links ]
2. Kessler RC, Chiu WT, Demler O, Merikangas KR, Walters EE. Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey. Arch Gen Psychiatr. 2005;62:617-27. [ Links ]
3. American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th ed. Washington (DC): Author; 1994. [ Links ]
4. Crippa JA. Does social phobia remain the 'Ugly Duckling' of anxiety disorders? Rev Bras Psiquiatr. 2009;31(4):297-9. [ Links ]
5. Freitas-Ferrari MC, Hallak JE, Trzesniak C, Filho AS, Machadode-Sousa JP, Chagas MH, Nardi AE, Crippa JA. Neuroimaging in social anxiety disorder: a systematic review of the literature. Prog Neuropsychopharmacol Biol Psychiatr. 2010;34(4):565-80. [ Links ]
6. Jørstad-Stein EC, Heimberg RG. Social phobia: an update on treatment. Psychiatr Clin North Am. 2009;32(3):641-63. [ Links ]
7. Rodebaugh TL, Holaway RM, Heimberg RG. The treatment of social anxiety disorder. Clin Psychol Rev. 2004;24(7):883-908. [ Links ]
8. Mululo SC, De Menezes GB, Fontenelle L, Versiani M. Eficácia do tratamento cognitivo e/ou comportamental para o transtorno de ansiedade social. Rev Psiquiatr RS. 2009;31(3):1-10. [ Links ]
9. Blanco C, Schneier FR, Schmidt A, Blanco-Jerez C, Marshal RD, Sanchez-Lacay A, Liebowitz MR. Pharmacological treatment of social anxiety disorder: a meta-analysis. Depress Anxiety. 2003;18(1):29-40. [ Links ]
10. Van Ameringen M, Mancini C, Pipe B, Bennett M. Optimizing treatment in social phobia: a review of treatment resistance. CNS Spectr. 2004;9(10):753-62. [ Links ]
11. Aarre TF. Phenelzine efficacy in refractory social anxiety disorder: a case series. Nord J Psychiatry. 2003;57(4):313-15. [ Links ]
12. Turner SM, Beidel DC, Wolff PL, Spaulding S, Jacob RG. Clinical features affecting treatment outcome in social phobia. Behav Res Ther. 1996;34(10):795-804. [ Links ]
13. Feske U, Perry KJ, Chambless DL, Renneberg B, Goldstein AJ. Avoidant personality disorder as a predictor for treatment outcome among generalized social phobics. Eur Neuropsychopharmacol. 2007;17:S517-S8. [ Links ]
14. Van Velzen CJ, Emmelkamp PM, Scholing A. The impact of personality disorders on behavioral treatment outcome for social phobia. Behav Res Ther. 1997;35(10):889-900. [ Links ]
15. Huppert JD, Strunk DR, Ledley DR, Davidson JRT, Foa EB. Generalized social anxiety disorder and avoidant personality disorder: structural analysis and treatment outcome. Depress Anxiety. 2008;25(5):441-8. [ Links ]
16. De Menezes GB, Fontenelle LF, Versiani M. Gender effect on clinical features and drug treatment response in social anxiety disorder (Social Phobia). Int J Psychiatry Clin Pract. 2008;12(2):151-5. [ Links ]
17. Chen J, Nakano Y, Ietzugu T, Ogawa S, Funayama T, Watanabe N, Noda Y, Furukawa TA. Group cognitive behavior therapy for japanese patients with social anxiety disorder: preliminary outcomes and their predictors. BMC Psychiatry. 2007;10(7):59-69. [ Links ]
18. Van Ameringen M, Oakman J, Mancini C, Pipe B, Chung H. Predictors of response in generalized social phobia: effect of age of onset. J Clin Psychopharmacol. 2004;24(1):42-8. [ Links ]
19. Dalrymple KL, Herbert JD, Gaudiano BA. Onset of illness and developmental factors in social anxiety disorder: preliminary findings from a retrospective interview. J Psychopathol Behav Assess. 2007;29(2):101-10. [ Links ]
20. Borge FM, Hoffart A, Sexton H. Predictors of outcome in residential cognitive and interpersonal treatment for social phobia: do cognitive and social dysfunction moderate treatment outcome? J Behav Ther Exp Psychiatry. 2010;41(3):212-9. [ Links ]
21. Versiani M, Amrein R, Montgomery SA. Social phobia: long-term treatment outcome and prediction of response - a moclobemide study. Int Clin Psychopharmacol. 1997;12(5):239-54. [ Links ]
22. Slaap BR, van Vliet IM, Westenberg HM, Den Boer JA. Responders and non-responders to drug treatment in social phobia: differences at baseline and prediction of response. J Affect Disord. 1996;39(1):13-9. [ Links ]
23. Stein DJ, Stein MB, Pitts CD, Kumar R, Hunter B. Predictors of response to pharmacotherapy in social anxiety disorder: an analysis of 3 placebo-controlled paroxetine trials. J Clin Psychiatry. 2002; 63(2):152-5. [ Links ]
24. Van Ameringen M, Mancini C, Chung H, Batzar E, Yang R. Predictors of remission in patients treated with sertraline for moderate to severe generalized social phobia. Eur Neuropsychopharmacol. 2002;12(3):S342. [ Links ]
25. Scholing A, Emmelkamp PM. Prediction of treatment outcome in social phobia: a cross-validation. Behaviour Research and Therapy. 1999;37:659-70. [ Links ]
26. Lincoln TM, Riefa W, Hahlwegb K, Frankc M, von Witzlebend I, Schroederb B, Fiegenbaume W. Who Comes, Who Stays, Who Profits? Predicting refusal, dropout, success, and relapse in a short intervention for social phobia. Psychother Res. 2005;15(3):210-25. [ Links ]
27. Marom S, Gilboa-Schechtman E, Aderka IM, Weizman A, Hermesh H. Impact of depression on treatment effectiveness and gains maintenance in social phobia: a naturalistic study of cognitive behavior group therapy. Depress Anxiety. 2009;26(3):289-300. [ Links ]
28. Brown EJ, Heimberg RG, Juster HR. Social phobia subtype and avoidant personality disorder: effect on severity of social phobia, impairment, and outcome of cognitive behavioral treatment. Behav Ther. 1995;26(3):467-86. [ Links ]
29. Chambless DL, Tran GQ, Glass CR. Predictors of response to cognitive-behavioral group therapy for social phobia. J Anxiety Disord. 1997;11(3):221-40. [ Links ]
30. Oosterbaan DB, van Balkom AJ, Spinhoven P, de Meij TG, van Dyck R. The influence on treatment gain of comorbid avoidant personality disorder in patients with social phobia. J Nerv Ment Dis. 2002;190(1):41-3. [ Links ]
31. Chen JW, Furukawa TA, Nakano Y, Ietsugu T, Ogawa S, Funayama T, Watanabe N, Noda Y, Rapee RM. Video feedback with peer ratings in naturalistic anxiety-provoking situations for social anxiety disorder: preliminary report. J Behav Ther Exp Psychiatry. 2010;41(1):6-10. [ Links ]
32. Alfano CA, Pina AA, Villalta IK, Beidel DC, Ammerman RT, Crosby LE. Mediators and moderators of outcome in the behavioral treatment of childhood social phobia. J Am Acad Child Adolesc Psychiatry. 2009;48(9):945-53. [ Links ]
33. Erwin BA, Heimberg RG, Schneier FR, Liebowitz MR. Anger experience and expression in social anxiety disorder: pretreatment profile and predictors of attrition and response to cognitive-behavioral treatment. Behav Ther. 2003;34(3):331-50. [ Links ]
34. Delsignore A, Carraro G, Mathier F, Znoj H, Schnyder U. Perceived responsibility for change as an outcome predictor in cognitive-behavioural group therapy. Br J Clin Psychol. 2008;47(3):281-93. [ Links ]
35. Taube-Schiff M, Suvak MK, Antony MM, Bieling PJ, McCabe RE. Group cohesion in cognitive-behavioral group therapy for social phobia. Behav Res Ther. 2007;45(4):687-98. [ Links ]
36. Laguna LB, Healey EC, Hope DA. Successful interdisciplinary intervention with an initially treatment-resistant social phobic. Behav Modif. 1998;22(3):358-71. [ Links ]
37. Serreti A, Chiesa A, Calati R, Perna G, Bellodi L, De Ronchi D. Common genetic, clinical, demografic and psychosocial predictors of response to pharmacotherapy in mood and anxiety disorders. Int Clin Psychopharmacol. 2009;24(1):1-18. [ Links ]
38. 3Carter JD, Luty SE, McKenzie JM, Mulder RT, Frampton CM, Joyce PR. Patient predictors of response to cognitive behaviour therapy and interpersonal psychotherapy in a randomised clinical trial for depression. J Affect Disord. 2011;128:252-61. [ Links ]
39. De Menezes GB, Fontenelle LF, Mululo S, Versiani M. Treatment-resistant anxiety disorders: social phobia, generalised anxiety disorder and panic disorder. Rev Bras Psiquiatr. 2007;29(2):S55-S60. [ Links ]
40. Davies RD, Dubovsky SL, Gabbert S, Chapman M. Treatment resistance in anxiety disorders. Bull Menninger Clin. 2000;64(3):A22-A36. [ Links ]
41. Ramsawh HJ, Raffa SD, Edelen MO, Rende R, Keller MB. Anxiety in middle adulthood: effects of age and time on the 14-year course of panic disorder, social phobia and generalized anxiety disorder. Psychol Med. 2009;39(4):615-24. [ Links ]
Leonardo F. Fontenelle
Rua Visconde de Pirajá 547 Sala 719
22410-003, Ipanema, Rio de Janeiro, Brazil
Fax: (.55 21) 2239-4919
Received on July 4, 2011; accepted on September 3, 2011