Evidence Table 10Social Anxiety Disorder

STUDY:Authors: Allgulander C, et al.161
Year: 2004
Country: Multinational (Sweden, Denmark, Germany, Norway, France, Finland)
FUNDING:Wyeth Research
DESIGN:Study design: RCT
Setting: Multi-center
Sample size: 436
INTERVENTION:
Drug:Venlafaxine ERParoxetinePlacebo
Dose:75–225 mg/d20–50mg/dN/A
Duration:12 weeks12 weeks12 weeks
Sample size:129128132
INCLUSION:Over 18 years old with DSM-IV criteria for SAD for at least 6 months prior to study; score of ≥ 4 on CGI-S; 50 on LSAS, with 30% decrease between pre-study and baseline visits; pre-study Raskin depression total score ≤9, and a 17-item HAM-D score <15
EXCLUSION:Previous treatment with venlafaxine or venlafaxine ER within 6 months of study day 1; concurrent disorders that confounded the evaluation of treatment: substance disorders, personality disorders (except avoidant personality disorder), depression or other primary anxiety disorders
OTHER MEDICATIONS/ INTERVENTIONS:Not reported
POPULATION CHARACTERISTICS:Groups similar at baseline: No (differences in gender)
Mean age: Venlafaxine ER: 38.7; paroxetine: 38.8; placebo: 38.9
Gender (% female): Venlafaxine ER: 46%; paroxetine: 52%; placebo: 62%
Ethnicity: Not reported
Other population characteristics: Baseline LSAS score 86.6 for placebo, 83.2 for venlafaxine ER, 83.9 for paroxetine
OUTCOME ASSESSMENT:Primary Outcome Measures: LSAS
Secondary Outcome Measures: CGI-S; CGI-IM; SPIN; SDI
Timing of assessments: Baseline, and days 7, 14, 21, 28, 42, 56, 70 and 84
RESULTS:
  • No significant differences in any outcome measures between venlafaxine ER and paroxetine
  • Treatment with venlafaxine ER and paroxetine was associated with significantly greater improvement than treatment with placebo for all primary and secondary efficacy variables (p < 0.05)
  • LSAS total scores significantly improved for venlafaxine ER or paroxetine vs. placebo–primary endpoint, the baseline adjusted mean change in LSAS total score was −36.0 (SE 2.35) for venlafaxine, −35.4 (SE 2.46) for paroxetine and −19.1 (SE 2.40) for the placebo group
  • SPIN scores significantly improved for venlafaxine ER and paroxetine groups than for placebo group at weeks 3–12 (both p < 0.05 week 3; both p < 0.01 week 4; both p < 0.001 weeks 6–12)
ANALYSIS:ITT: Yes
Post randomization exclusions: Yes
ATTRITION:Loss to follow-up: 16.8%; venlafaxine ER: 16%; paroxetine: 16%; placebo: 18.5%
Withdrawals due to adverse events: 7.6%, venlafaxine: not reported; paroxetine: not reported
Loss to follow-up differential high: No
ADVERSE EVENTS:
  • During the double-blind treatment period, 90% venlafaxine ER, 89% paroxetine, and 82% placebo treated patients reported treatment emergent adverse events; the most common (incidence ≥5%) adverse events among venlafaxine ER treated patients were headache (10%), nausea (7%), dizziness (14%), insomnia (6%), and vertigo (10%); among paroxetine-treated patients were headache (12%), dizziness (13%), and insomnia (6%); among placebo treated patients, no taper/post study emergent adverse event occurred at an incidence of ≥5% and the differences between groups were not statistically significant
QUALITY RATING:Fair
STUDY:Authors: Davidson J, et al.162
Year: 2004
Country: US
FUNDING:National Institute of Mental Health grant
DESIGN:Study design: RCT
Setting: 2 academic medical centers
Sample size: 117 (295 total in arms including CCBT)
INTERVENTION:
Drug:FluoxetinePlacebo
Dose:10–60 mg/dayN/A
Duration:14 weeks14 weeks
Sample size:5760
INCLUSION:DSM-IV diagnosis of GSP; age between 18 and 65 years; fluency in English; provision of written informed consent
EXCLUSION:Primary comorbid anxiety disorder (defined by which disorder was the more debilitating and clinically salient); lifetime history of schizophrenia, bipolar disorder, or organic brain syndrome; major depression within the last 6 months; substance abuse or dependence within the past year; mental retardation or pervasive developmental disability; unstable medical condition; prior failure of response to fluoxetine at 60 mg/d for at least 4 weeks or to 12 weekly sessions of CCBT for GSP; concurrent psychiatric treatment or other psychoactive medications; positive urine drug screen results; inability to maintain 2 weeks’ psychotropic drug-free wash-out; pregnancy or lactation
OTHER MEDICATIONS/ INTERVENTIONS:NR
POPULATION CHARACTERISTICS:Groups similar at baseline: Yes
Mean age: fluoxetine: 36.3, placebo: 36.9
Gender (female %): fluoxetine: 42.9, placebo: 45.8
Ethnicity (% white): fluoxetine: 71.4, placebo: 82.8
OUTCOME ASSESSMENT:Primary Outcome Measures: CGI-I, CGI-S, BSPS
Secondary Outcome Measures: Social Phobia and Anxiety Inventory
Timing of assessments: baseline and weeks 4, 8 14
RESULTS:
  • CGI response rates at week 14 higher for fluoxetine (50.9% vs. 31.7%; p=0.03)
  • BSPS effect sizes (95% CI): 0.40 (0.02 to 0.77) for fluoxetine vs. placebo
  • CGI-S scale effect size (95% CI) for fluoxetine vs. placebo: 0.42 (0.04 to 0.80)
  • CGI-S score at baseline: 4.4 vs. 4.3; at week 14: 2.7 vs. 3.3; fluoxetine treatment superior to placebo (p<0.05)
  • SPAI score at week 14 69.3 vs. 94.8; fluoxetine superior to placebo (p<0.05)
ANALYSIS:ITT: Yes
Post randomization exclusions: yes (9)
ATTRITION:Loss to follow-up; fluoxetine: 32%; placebo: 40%
Withdrawals due to adverse events: fluoxetine: 8.8%; placebo: 3.3%
Withdrawals due to lack of efficacy: fluoxetine: 1.8%; placebo: 3.3%
Loss to follow-up differential high: No
ADVERSE EVENTS:TEAEs (fluoxetine vs. placebo)
  • Insomnia: 47.9 vs. 42.3; p=0.005
  • Headache: 31.2 vs. 38.5; p=0.008
  • Nausea: 18.8 vs. 15.4; p<0.04
  • Anorgasmia: 32.4 vs. 9.6; p<0.001
  • Erectile dysfunction: 10.4 vs. 1.9; p<0.02
QUALITY RATING:Fair
STUDY:Authors: Hansen et al., 163
Year: 2008
Country: multinational
FUNDING:Subcontract with the Center for Evidence-Based Policy; Oregon Health & Science University; first author supported by grant K12RR023248
DESIGN:Study design: systematic review and meta-analysis
Number of patients: 5,172
AIMS OF REVIEW:A systematic review and meta-analysis was conducted to evaluate the comparative efficacy and tolerability of second-generation antidepressants in social anxiety disorder.
STUDIES INCLUDED IN REVIEW18 trials: 3 head-to-head RCTs (Allgulander et al., 2004; Lader et al., 2004; Liebowitz et al., 2005) and 15 placebo-controlled trials (Allgulander, 1999; Baldwin et al., 1999; Blomhoff et al., 2001; Davidson et al., 2004; Kasper et al., 2005; Kobak et al., 2002; Lepola et al., 2004; Liebowitz et al., 2003; Rickels et al., 2004; Stein et al., 1999; Stein et al., 1998; Stein et al., 2005; Van Ameringen et al., 2001; Wagner et al., 2004; Westenberg et al., 2004) 15 placebo-controlled trials which compare one SSRI to placebo or which compare one SSRI to another;
TIME PERIOD COVERED:January 1980 through October 2006
CHARACTERISTICS OF INCLUDED STUDIES:3 head-to-head RCTs and 15 placebo-controlled trials which compare one second-generation antidepressant (SGAD) to placebo or which compare one SGAD to another;
duration 12–28 weeks;
CHARACTERISTICS OF INCLUDED POPULATIONS:All trials required a diagnosis of SAD consistent with the DSM-IV; baseline disease severity varying; mean age in most trials was between 35 and 45 years, with a relatively equal distribution of males and females. One study included children and adolescents (mean age, 13 years).
CHARACTERISTICS OF INTERVENTIONS:active and placebo-controlled trials ( Escitalopram, Fluoxetine, Fluvoxamine, Fluvoxamine CR, Paroxetine, Paroxetine CR, Sertraline, Venlafaxine ER)
MAIN RESULTS:Clinical response (meta-analyses, pooled results, effect sizes and relative risks)
  • Anxiety Severity (Liebowitz Social Anxiety Scale, LSAS): pooled weighted mean reduction: overall active treatments vs. placebo 10–16 point greater LSAS reduction than placebo: 10.3 (95%CI 5.9–14.6) for escitalopram, 12.3 (95% CI 8.2–16.3) for fluvoxamine, 16.1 (95%CI 13.1–19.1) for paroxetine, and 14.8 (95%CI 10.6–19.0) for venlafaxine. No significant differences in mean change in LSAS when directly comparing SGAD to another (escitalopram vs. paroxetine, paroxetine vs. venlafaxine).
  • Functional impairment (Sheehan Disability Scale, SDS): no significant difference in reducing disability between active treatments and placebo; statistical significant differences only in work domain, not family and social domains: compared with placebo, active treatment produced a 0.7 to 2.2 point greater reduction in the work domain (pooled difference 1.25; 95%CI 0.9 to 1.5); the social domain was significantly more improved for active treatment compared with placebo in all but one trial
  • Clinical Global Impression of Improvement Scale (CGI-I): (response “very much improved” or “improved” on CGI-I): pooled relative benefits for escitalopram (RB 1.31; 95%CI 1.17–1.46), paroxetine (RB 1.85; 95%CI 1.49–2.29), sertraline (RB 1.78; 95%CI 1.45–2.16), and venlafaxine (RB 1.68; 95%CI 1.47–1.93) were statistically significantly better than placebo. Fluvoxamine showed no significant improvement (RB 1.49; 95%CI 0.94–2.36).
  • Clinical Global Impression of Improvement Scale (CGI-I): No significant differences in response when directly comparing SGAD to another (escitalopram vs. paroxetine, paroxetine vs. venlafaxine). No significant differences in response in indirect comparisons.
ADVERSE EVENTS:
  • types of adverse events reported among patients with SAD similar to those reported in patients with other psychiatric disorders, but tendency towards higher frequencies (e.g. for nausea and insomnia). The most commonly reported adverse events were nausea, asthenia or fatigue, or changes in sleep.
  • The adverse events profile differed among SSRIs (mean incidence in percent with 95%CI)
NauseaAsthenia*SweatingSomnolenceInsomniaDry MouthAbnormal EjaculationLibido Decrease
Escitalopram;25 (19–32)14 (13–15)9 (3–15)11 (9–12)9 (NA)NR8 (4–12)6 (5–7)
Fluoxetine:NR30 (NA)NRNR47 (NA)NRNRNR
Fluvoxamine :39 (23–55)28 (NA)NR27 (18–35)32 (31–33)NR11 (9–13)9 (6–11)
Paroxetine :25 (21–29)19 (16–21)13 (9–17)15 (9–21)15 (11–18)15 (7–24)18 (12–25)9 (7–11)
Sertraline:27 (17–37)18 (17–19)11 (10–12)11 (NA)27 (22–33)14 (13–15)13 (10–16)7 (NA)
Venlafaxine :31 (27–34)16 (8–24)15 (8–22)22 (14–29)22 (16–28)17 (13–21)14 (10–17)8 (6–11)
COMPREHENSIVE LITERATURE SEARCH STRATEGY:Yes (MEDLINE®, Embase, The Cochrane Library, PsychLit, and the International Pharmaceutical Abstracts, Handsearch, pharma dossiers)
STANDARD METHOD OF APPRAISAL OF STUDIES:Yes
QUALITY RATING:Good
STUDY:Authors: Hedges D et al.164
Year: 2007
Country: Multinational
FUNDING:Brigham Young University, Department of Psychology
DESIGN:Study design: Systematic review
Number of patients: 3,361
AIMS OF REVIEW:To investigate the efficacy of SSRIs in social anxiety disorder
STUDIES INCLUDED IN REVIEW15 studies: van Vliet et al., 1994; Katzelnick et al., 1995; Stein et al., 1998; Allgulander, 1999; Baldwin et al., 1999; Stein et al., 1999; Blomhoff et al., 2001; Van Ameringen et al., 2001; Kobak et al., 2002; Liebowitz et al., 2002; Liebowitz et al., 2003; Davidson et al., 2004a; Davidson et al., 2004b; Lader et al., 2004, Lepola et al., 2004
TIME PERIOD COVERED:1966–2004
CHARACTERISTICS OF INCLUDED STUDIES:Double-blind, placebo-controlled trials ranging in duration from 10–24 weeks
CHARACTERISTICS OF INCLUDED POPULATIONS:Adults with social anxiety disorder (social phobia)
CHARACTERISTICS OF INTERVENTIONS:Citalopram, escitalopram, fluoxetine, fluvoxamine, paroxetine, or sertraline vs. placebo
MAIN RESULTS:
  • Effect sizes for the Liebowitz Social Anxiety Scale ranged from 0.029 to 1.214
  • Effect sizes for the Sheehan Disability Scale ranged from 0.203 to 0.480 for work, 0.237 to 0.786 for social function, and 0.118 to 0.445 for family function
  • The Θ log-odds ratios for CGI of change scores ranged from 0.644 to 3.267
  • SSRIs appear more effective than placebo for social anxiety disorder, with improvement extending into social and occupational function
ADVERSE EVENTS:NR
COMPREHENSIVE LITERATURE SEARCH STRATEGY:PubMed and PsychINFO were searched as well as the reference lists of pertinent articles.
STANDARD METHOD OF APPRAISAL OF STUDIES:NR
QUALITY RATING:Fair
STUDY:Authors: Kasper S, et al.165
Year: 2005
Country: Multinational
FUNDING:H. Lundbeck A/S
DESIGN:Study design: RCT
Setting: Multi-center
Sample size: 358
INTERVENTION:
Drug:EscitalopramPlacebo
Dose:10–20N/A
Duration:12 weeks12 weeks
Sample size:181177
INCLUSION:Outpatients with a primary diagnosis GSAD following DSM-IV criteria; 18–65 years old; a score of at least 70 on the LSAS; evidence of fear or avoidance traits in at least 4 social situations; otherwise healthy
EXCLUSION:Primary diagnosis of other Axis 1 disorders or a history of within the past 6 months; diagnosis of any Axis II cluster; substance abuse within 12 months; if investigator diagnosed a serious risk of suicide; MADRS >19; use of a depot antipsychotic within 6 months or any antipsychotic, anxiolytic or anticonvulsant within 2 weeks before start; known drug allergy or previous lack of therapeutic response to citalopram
OTHER MEDICATIONS/ INTERVENTIONS:Chloral hydrate for sleep
POPULATION CHARACTERISTICS:Groups similar at baseline: No – escitalopram group older (39 vs. 36) with greater duration of disease (24 vs. 21 years)
Mean age: 38
Gender (% female): 45%
Ethnicity: NR
Other population characteristics:
Baseline LSAS: placebo: 95.4, escitalopram: 96.3
Baseline CGI-S: placebo: 4.8, escitalopram: 4.8
OUTCOME ASSESSMENT:Primary Outcome Measures: LSAS total score

Secondary Outcome Measures: LSAS subscales; CGI-S; CGI-I; SDS; MADRS

Timing of assessments: Baseline and weeks 1, 2, 3, 4, 6, 8,12
RESULTS:
  • LSAS at 12 weeks: placebo 68.8, escitalopram 62.2 with a treatment difference of 7.3 (p < 0.01)
  • Mean reduction in LSAS fear/anxiety subscale: escitalopram −16.9, placebo −12.7 (p < 0.001)
  • Mean reduction in LSAS avoidance subscale: escitalopram −17.6, placebo −14.4 (p < 0.05)
  • Escitalopram showed significant improvements over placebo in CGI-S (p < 0.01); CGI-I responders 39% for placebo and 54% for escitalopram (p < 0.01)
  • Significantly more improvement in SDS work (p < 0.001) and social (p < 0.05) subscales
  • MADRS not reported
ANALYSIS:ITT: Yes
Post randomization exclusions: Yes-5 had no post-baseline assessment
ATTRITION:OverallPlaceboEscitalopram
Loss to follow-up:19%18%20%
Withdrawals due to adverse events:6.8%4.5%8.8%
Withdrawals due to lack of efficacy:4.2%6.2%2.2%
Loss to follow-up differential high:No
ADVERSE EVENTS:
  • Headache: placebo: 25%, escitalopram: 25%
  • Nausea: placebo: 12%, escitalopram: 22%
  • Fatigue: placebo: 9%, escitalopram: 14%
  • Somnolence: placebo: 5%, escitalopram: 10%
  • Diarrhea: placebo: 5%, escitalopram: 9%
  • Insomnia: placebo: 6%, escitalopram: 9%
QUALITY RATING:Fair
STUDY:Authors: Kobak KA, et. al.166
Year: 2002
Country: US
FUNDING:Eli Lilly & Co.
DESIGN:Study design: RCT
Setting: Single center
Sample size: 60
INTERVENTION:
Drug:FluoxetinePlacebo
Dose:20–60 mg/dN/A
Duration:14 weeks14 weeks
INCLUSION:DSM-IV criteria for social phobia for at least 6 months; a score of at least 50 on the Liebowitz Social Anxiety Scale (LSAS) before and after the lead–in; score could not decrease by more than 20%
EXCLUSION:Non-response to fluoxetine treatment; pregnancy; previous participation in a fluoxetine study; concurrent use of psychotropic or centrally acting drugs, anticonvulsants, corticosteroids, or tryptophan; serious illness; suicidal; concurrent Axis I disorders in past 12 months; psychotherapy; seizure disorder
OTHER MEDICATIONS/ INTERVENTIONS:Not reported
POPULATION CHARACTERISTICS:Groups similar at baseline: Not reported
Mean age: 39.5
Gender (% female): 58%
Ethnicity: Not reported
Other population characteristics: Not reported
OUTCOME ASSESSMENT:Measures: Liebowitz Social Anxiety Scale (LSAS) (primary), Social Phobia Subscale of Fear Questionnaire, CGI-S, CGI-I, Patient Global Improvement Scales, HAM-A, Brief Social Phobia Scale, HAM-D (did not report which scale), Global Assessment of Functioning Scale, QOL
Timing of assessments: Weeks 1, 2, 4, 6, 8, 10, 12, 14
RESULTS:
  • Fluoxetine was not significantly different from placebo on the LSAS score (p = 0.901)
  • Similar results in secondary outcome measures with no significant difference between fluoxetine and placebo
  • A significant change was found on all outcome measures from baseline to endpoint with both fluoxetine (p < 0.001) and placebo (p < 0.001)
ANALYSIS:ITT: Yes
Post randomization exclusions: No
ATTRITION:Loss to follow-up: 20%; fluoxetine 16%; placebo 23%
Withdrawals due to adverse events: 7%; fluoxetine 3%, placebo 10%
Loss to follow-up differential high: No
ADVERSE EVENTS:
  • For fluoxetine: headache, insomnia, asthenia, and nervousness
  • For placebo: headache, insomnia, nervousness, and myalgia
  • Significantly more fluoxetine than placebo patients had asthenia (p = 0.02)
  • Significantly more placebo than fluoxetine patients had myalgia (p = 0.04)
QUALITY RATING:Fair
STUDY:Authors: Lader M, et al.167
Year: 2004
Country: Multinational (11 countries)
FUNDING:H. Lundbeck A/S
DESIGN:Study design: RCT
Setting: Multi-center (47 centers)
Sample size: 839
INTERVENTION:
Drug:Escitalopram 5Escitalopram 10Escitalopram 20Paroxetine 20Placebo
Dose:5 mg/d10 mg/d20 mg/d20 mg/dN/A
Duration:24 weeks24 weeks24 weeks24 weeks24 weeks
Sample size:167167170169166
INCLUSION:Healthy female and male outpatients 18–65 years of age; primary diagnosis of generalized SAD according to DSM-IV criteria; score ≥ 70 on the Liebowitz Social Anxiety Scale (LSAS); score ≥ 5 on one or more of the Sheehan Disability Scale (SDS) subscales
EXCLUSION:Another Axis I disorder primary diagnosis within 6 months; MADRS total score ≥ 18; DSM-IV diagnosis of schizophrenia/ other psychotic disorder; Axis II Cluster B diagnosis; learning difficulties or other cognitive disorder; suicidal tendencies; no therapeutic response to SSRIs; drug hypersensitivities; taken a psychoactive drug within 2 weeks of screening; receiving formal psychotherapy
OTHER MEDICATIONS/ INTERVENTIONS:NR
POPULATION CHARACTERISTICS:Groups similar at baseline: Yes
Mean age: Escitalopram 5: 36.3; escitalopram 10: 37.2; escitalopram 20: 37; paroxetine 20: 37.4; placebo: 37
Gender (% female): Escitalopram 5: 50%; escitalopram 10: 57%; escitalopram 20: 53%; paroxetine: 54%; placebo: 49%
Ethnicity: 99.3% white
Other population characteristics: Mean duration of disorder (yrs): 19.5
OUTCOME ASSESSMENT:Primary Outcome Measures: Mean change from baseline to week 12 in LSAS total score (LOCF)
Secondary Outcome Measures: LSAS subscale scores; CGI-S; CGI-I; change in SDS
Timing of assessments: Baseline and after weeks 1,2,4,6,8,10,12,16,20,24,25, and 26.
RESULTS:
  • No significant difference observed between any escitalopram treatment groups and the paroxetine group in the LOCF analysis of LSAS total score.
  • At weeks 16, 20, and 24 (observed case analysis), compared to the paroxetine group (p < 0.05)the 20 mg/d escitalopram group had significantly superior LSAS scores
  • Escitalopram 20mg/d was superior to paroxetine 20mg/d on CGI-S at week 24
  • Escitalopram 20mg/d was superior to paroxetine 20mg/d on some SDS subscales during weeks 16 and 20, but no significant differences were noted at week 24
ANALYSIS:ITT: Yes
Post randomization exclusions: Not reported
ATTRITION:Loss to follow-up: 29%; escitalopram 5: 25.1%; escitalopram 10: 33.5%; escitalopram 20: 28.8%; paroxetine: 26.6%; placebo: 30.1%
Withdrawals due to adverse events: 9%; escitalopram 5: 4.8%; escitalopram 10: 9.6%; escitalopram 20: 11.8%; paroxetine: 13.6%; placebo: 6%
Loss to follow-up differential high: No
ADVERSE EVENTS:
  • Percentage patients experiencing any adverse effect: Escitalopram 5: 68.9%; escitalopram 10: 72.5%; escitalopram 20: 78.2%; paroxetine 20: 79.3%; placebo: 60.8%
  • Nausea: Escitalopram 5: 20.4%; escitalopram 10: 19.8%; escitalopram 20: 28.8%; paroxetine 20: 29%; placebo: 10.2%
  • Fatigue: 9% placebo; Escitalopram 5: 11.4%; escitalopram 10: 12%; escitalopram 20: 14.1%; paroxetine 20: 17.8%; placebo: 9%
  • Increased sweating: Escitalopram 5: 5.4%; escitalopram 10: 10.8%; escitalopram 20: 11.8%; paroxetine 20: 14.2%; placebo: 1.8%
QUALITY RATING:Fair
STUDY:Authors: Liebowitz MR, et al.168
Year: 2005
Country: US
FUNDING:Wyeth Research, Collegeville PA
DESIGN:Study design: RCT
Setting: Multi-center (26 centers)
Sample size: 440
INTERVENTION:
Drug:VenlafaxineParoxetinePlacebo
Dose:75–225 mg/d20–50 mg/dN/A
Duration:12 weeks12 weeks12 weeks
Sample size:146147147
INCLUSION:Outpatients ≥ 18 years who fulfilled DSM-IV criteria for SAD for ≥ 6 months at screening; LSAS ≥ 50 at screening and baseline with ≤ 30% decrease between prestudy and baseline; ≥ 4 on the CGI-S; Covi Anxiety Score total > Raskin Depression Scale total score; HAM-D < 15 with ≤ 2 on depressed mood item.
EXCLUSION:Patients with a clinically important Axis I or Axis II disorder other than SAD or avoidant personality disorder; history or current psychotic illness; Suicidal; history of drug or alcohol dependence within 1 year of the study; used anti-depressants (other than fluoxetine), anxiolytics, or herbal products within 14 days of the study; ECT within 6 months of the study; used antipsychotic medications or fluoxetine treatment within 30 days of the study; clinically significant abnormal findings on laboratory tests; pregnant or breastfeeding
OTHER MEDICATIONS/ INTERVENTIONS:NR
POPULATION CHARACTERISTICS:Groups similar at baseline: Yes
Mean age: venlafaxine: 35.7, paroxetine: 35.8, placebo: 37.3
Gender (% female): venlafaxine: 46.6%, paroxetine: 45.6%, placebo: 47.2%
Ethnicity: White: VX: 71.4% PX: 72.8% Placebo: 70.1%
African American: VX: 11.3% PX: 8.8% Placebo: 8.3%
Hispanic: VX: 15.0% PX: 12.5% Placebo: 13.2%
Other population characteristics:
Baseline LSAS: VX: 86.2 PX: 87.2 Placebo: 86.1
OUTCOME ASSESSMENT:Primary Outcome Measures: Reduction in Liebowitz Social Anxiety Scale (LSAS) total score
Secondary Outcome Measures: CGI-I; CGI-S; Social Phobia Inventory Scores, SDS
Timing of assessments: Weekly
RESULTS:
  • No significant difference in LSAS improvement was observed between the venlafaxine and paroxetine groups at endpoint. Both were significantly improved from placebo (p < 0.05).
  • No significant difference in CGI-I improvement was observed between the venlafaxine and paroxetine groups at endpoint. Both were significantly improved from placebo (p < 0.05)
  • No significant difference in Social Phobia Inventory improvement was observed between the venlafaxine and paroxetine groups at endpoint; both significantly improved from placebo (p < 0.05)
  • No significant difference in CGI-S improvement was observed between the venlafaxine and paroxetine groups at endpoint. Both were significantly improved from placebo (p < 0.05)
  • No significant differences in SDS domains between venlafaxine and placebo
ANALYSIS:ITT: Yes
Post randomization exclusions: Yes
ATTRITION:OverallVenlafaxineParoxetinePlacebo
Loss to follow-up:26%27.0%28.2%22.6%
Withdrawals due to adverse events:10.4%14.2%13.4%4.1%
Withdrawals due to lack of efficacy:2.3%0.7%0.7%5.5%
Loss to follow-up differential high:No
ADVERSE EVENTS:VenlafaxineParoxetinePlacebo
Nausea32.6%26.1%11.0%
Insomnia27.7%18.3%8.2%
Somnolence27%26.8%8.9%
Asthenia20.6%23.9%10.3%
Dry Mouth17.7%16.2%4.8%
Anorexia14.2%10.6%3.4%
Abnormal ejaculation (men)10.5%20.8%0%
QUALITY RATING:Fair
STUDY:Authors: Montgomery SA, et al.169
Year: 2005
Country: Multinational
FUNDING:H. Lundbeck A/S
DESIGN:Study design: Open label followed by randomized, double-blind, parallel group, placebo-controlled, fixed dose relapse prevention comparison
Setting: 76 private/hospital outpatient clinics & specialized clinical research centers (11 countries)
Sample size: 517 (open label); 372 (RCT)
INTERVENTION:
Drug:EscitalopramPlacebo
Dose:10 or 20 mg/dN/A
Duration:24 wks24 wks
Sample size:191181
INCLUSION:Outpatients between 18 and 80 yrs old; primary DSM-IV diagnosis of generalized social anxiety disorder (GSAD); total Liebowitz Social Anxiety Scale (LSAS) score ≥70 w/ exhibited fear or avoidance traits in ≥ 4 social situations; and score ≥ 5 on 1 or more Sheehan Disability Scale (SDS) subscales; RCT required CGI-I score of 1 or 2 after open-label treatment
EXCLUSION:Other Axis I diagnosis in previous 6 months; MADRS total score ≥ 18; score ≥ 5 on MADRS item 10 (suicidal thoughts); DSM-IV diagnosis of alcohol/drug abuse, eating disorder, major depressive disorder, panic disorder, obsessive-compulsive disorder, body dysmorphic disorder, schizophrenia, other psychotic disorder, mania or hypomania, or any Axis II diagnosis; known lack of response to SSRI; treatment with psychoactive drug in last 2 wks (or 5 wks if fluoxetine); formal psychotherapy in last 2 weeks.
OTHER MEDICATIONS/ INTERVENTIONS:NR
POPULATION CHARACTERISTICS:Groups similar at baseline: Yes
Mean age: Escitalopram: 36, Placebo: 37
Gender(% female): Escitalopram: 46%, placebo: 49%
Ethnicity: 95% white (both groups)
Other population characteristics: Mean BMI = 24.2; Mean age at GSAD onset = 17; Mean duration of GSAD = 19y (escitalopram) and 20y (placebo)
OUTCOME ASSESSMENT:Primary Outcome Measures: survival analysis estimate of time to relapse in the double-blind period. (Relapse defined as LSAS score increase ≥ 10 or withdrawal of patient due to lack of efficacy.)
Secondary Outcome Measures: LSAS total score; LSAS avoidance and fear/anxiety subscale; SDS
Timing of assessments: 1,2,4,8,12,16,20,& 24 weeks after randomization; also safety follow-up at 4 weeks after last dose of double-blind treatment
RESULTS:
  • Significant advantage in survival for escitalopram vs. placebo in primary efficacy analysis (log rank test p < 0.001)
  • Relapse rates = 22% (escitalopram) vs. 50% (placebo)
  • Risk of relapse was 2.8 times higher w/ placebo than escitalopram
  • Median time to relapse = 407 days (escitalopram) vs. 144 days (placebo)
  • Significant advantage for escitalopram on all secondary measures (LSAS, CGI-S, SDS, and MADRS)
  • Improvement on LSAS in escitalopram group (8.3 points), deterioration in placebo group (4.5 points)
  • Mean MADRS score change = +0.8 (escitalopram) and +2.6 (placebo)
  • Mean CGI-S score change = −0.3 (escitalopram) and +0.3 (placebo)
ANALYSIS:ITT: Yes, defined as all randomized patients who took at least 1 dose of double-blind medication and had at least 1 valid post baseline assessment of LSAS total score
Post randomization exclusions:
ATTRITION:Loss to follow-up: Escitalopram: 25 (13%), placebo: 15 (8.3%)
Withdrawals due to adverse events: Escitalopram: 5 (2.6%), placebo: 6 (3.3%)
Withdrawals due to lack of efficacy: N/A
Loss to follow-up differential high: No
ADVERSE EVENTS:
  • Assessed via spontaneous report, various clinical exam/lab reports, and 43-item Discontinuation Emergent Signs and Symptoms (DESS) checklist at randomization and 1 and 2 wks after.
  • Treatment emergent adverse events (TEAEs) with incidence ≥ 5 % in either group were: headache, dizziness, increased sweating, nervousness, fatigue, insomnia, nausea, rhinitis, and influenza-like symptoms
  • Incidence of TEAEs was lower in escitalopram group (62.6%) vs. placebo group (71.8%)
  • Dizziness, increased sweating, and nervousness were significantly higher in placebo group in 1st 2 weeks following discontinuation of escitalopram (p < 0.05). Excluding these TEAEs in 1st 2 weeks post-randomization, adverse events were similar in both treatment groups
  • After 1 and 2 weeks of double-blind treatment, mean total DESS score was significantly lower in -escitalopram group (week 1: escitalopram =1.17 vs. placebo = 2.61; week 2: escitalopram =1.02 vs. placebo = 1.78) (p < 0.01)
QUALITY RATING:Fair
STUDY:Authors: Muehlbacher M, et al.170
Year: 2005
Country: Multinational
FUNDING:NR
DESIGN:Study design: Randomized, double-blind, placebo controlled
Setting: Clinics
Sample size: 66
INTERVENTION:
Drug:MirtazapinePlacebo
Dose:30 mg/dN/A
Duration:10 wks10 wks
Sample size:3333
INCLUSION:Women aged 18 or older with DSM-IV diagnosed social phobia
EXCLUSION:Psychotic symptoms; use of mirtazapine or other psychotropic drug; psychotherapy; currently or planning to be pregnant (or no contraception use); severe somatic illness; currently suicidal; current drug / alcohol abuse; severe major depressive disorder.
OTHER MEDICATIONS/ INTERVENTIONS:NR
POPULATION CHARACTERISTICS:Groups similar at baseline: Cannot tell
Mean age: NR
Gender: NR
Ethnicity: NR
Other population characteristics: Both groups similar in percentage currently living in partnership, and with personality, panic, general anxiety disorders, OCDs
OUTCOME ASSESSMENT:Primary Outcome Measures: Change in social anxiety measured w/ social phobia inventory (SPIN) and LSAS
Secondary Outcome Measures: SF-36 Health Survey
Timing of assessments: Weekly for 10 weeks, although intermediate results were not analyzed
RESULTS:
  • Mirtazapine group experienced significantly greater rate of change on both SPIN and LSAS scales
  • Initial SPIN scores = 32.5 +/− 4.7 (mirtazapine) vs. 29.0 +/− 4.6 (placebo)
  • Final SPIN scores = 24.1 +/− 4.3 (mirtazapine) vs. 28.7 +/− 5.1 (placebo)
  • SPIN: Difference in change b/w both groups = −8.1 (95% CI −9.6 to 4.1; p < 0.001)
  • Initial LSAS scores = 71.9 +/− 8.3 (mirtazapine) vs. 72.5 +/− 8.0 (placebo)
  • Final LSAS scores= 46.3 +/− 7.0 (mirtazapine) vs. 67.1 +/− 7.4 (placebo)
  • LSAS: Difference in change b/w both groups = −20.2 (95% CI −27.5 to −4.1; p < 0.001)
  • Mirtazapine group experienced significantly greater rate of change on SF-36 (on general health perceptions, vitality, social functioning, role-emotional, and mental health scales)
ANALYSIS:ITT: No
Post randomization exclusions: Cannot tell
ATTRITION:Loss to follow-up: NR
Withdrawals due to adverse events: NR
Withdrawals due to lack of efficacy: NR
Loss to follow-up differential high: NR
ADVERSE EVENTS:
  • Most frequently reported adverse events in mirtazapine vs. placebo were: dry mouth (21.2% vs. 12.1%), drowsiness (18.2% vs. 9.1%), sedation (18.2% vs. 6.1%), increased appetite (12.1% vs. 3.0%), and weight gain (21.2% vs. 6.1%)
QUALITY RATING:Fair
STUDY:Authors: van der Linden et. al.171
Year: 2000
Country: South Africa, the Netherlands
FUNDING:MRC Research Unit on Anxiety and Stress Disorders; Harry Crossley Trust; Cochrane review collaborators
DESIGN:Study design: Meta-analysis
Number of patients: 1482
AIMS OF REVIEW:To review all available SSRI studies for social anxiety disorder
STUDIES INCLUDED IN META-ANALYSISVan Vliet et al., 1994, Katzelnick et al., 1995, Stein et al., 1998, Stein et al., 1999, Baldwin et al., 1999, Pfizer
Pharmaceutical Group data on file, 1999, SmithKlineBeecham data on file, 1998
TIME PERIOD COVERED:Not reported (included studies for dates 1994 to 2000)
CHARACTERISTICS OF INCLUDED STUDIES:RCTs (placebo controlled); 18 trials; 2 unpublished
CHARACTERISTICS OF INCLUDED POPULATIONS:Patients with social anxiety disorder
CHARACTERISTICS OF INCLUDED INTERVENTIONS:RCT data were analyzed for fluvoxamine, paroxetine, and sertraline
MAIN RESULTS:Odds ratio of responder status for SSRI vs. placebo varied between 2.1 and 26.2
The NNT varied from 1.6 to 4.2
LSAS effect size varied from 0.3 to 2.2
No difference in efficacy between SSRIs was reported
ADVERSE EVENTS:Not reported
COMPREHENSIVE LITERATURE SEARCH STRATEGY:Not defined in article but described to be consistent with methods of a Cochrane review
STANDARD METHOD OF APPRAISAL OF STUDIES:Not defined in article but described to be consistent with methods of a Cochrane review
QUALITY RATING:Fair
STUDY:Authors: Van Ameringen M, et al.172
Year: 2007
Country: Canada
FUNDING:Bristol-Myers Squibb
DESIGN:Study design: RCT
Setting: Outpatient anxiety clinics (4)
Sample size: 105
INTERVENTION:
Drug:NefazodonePlacebo
Dose:100–600 mg/dayN/A
Duration:14 weeks14 weeks
Sample size:5253
INCLUSION:Psychiatric outpatients; 18–65 yrs; met DSM-IV criteria for GSP for >1 year; be of at least moderate illness severity based on CGI-S rating; patients with comorbid secondary MDD could participate if MADRS baseline score ≤ 19, no risk of suicidality, and onset of social phobia predated MDD by at least 5 years.
EXCLUSION:Current comorbid Axis I disorders such as panic disorder with agoraphobia, OCD, body dysmorphic disorder, or alcohol/substance abuse; lifetime history of bipolar affective disorder, schizophrenia, psychoses, delirium, dementia, or other cognitive disorders; reporting 2 previous treatment failures for GSP.
OTHER MEDICATIONS/ INTERVENTIONS:Chloral hydrate up to 1000 mg/night for sleep
POPULATION CHARACTERISTICS:Groups similar at baseline: Yes
Mean age: nefazodone: 34.6, placebo: 37.0
Gender (female %): nefazodone: 53.8%, placebo: 50.9%
Ethnicity (%white): nefazodone: 86.5%, placebo: 83.0%
Other population characteristics:
OUTCOME ASSESSMENT:Primary Outcome Measures: CGI-I responders at endpoint; mean change in LSAS score
Secondary Outcome Measures: CGI-S, Social Phobia Inventory, SPS, Social interaction Anxiety Scale, Beck Depression Inventory, Beck Anxiety Scale, Sheehan Disability Scale, RAND 36-Item Health Survey
Timing of assessments: weeks 1, 2,3,5,7,9,12, and 16
RESULTS:
  • Higher % of nefazodone patients were CGI-I responders (CGI-I score of 1 or 2) at endpoint: 31.4% vs. 23.5%; p=0.38
  • With the exception of the Social Phobia Scale, no significant differences found in measures of social phobia between treatment groups
ANALYSIS:ITT: Yes (N=102)
Post randomization exclusions:
ATTRITION:Loss to follow-up: 23.8%; nefazodone 30.8%, placebo 17.0%
Withdrawals due to adverse events:
Withdrawals due to lack of efficacy:
Loss to follow-up differential high: No
ADVERSE EVENTS:
  • Headache: 35.3% vs. 29.4%; p=0.53
  • Fatigue: 19.6% vs. 11.8%; p=0.28
  • Dizziness/lightheadedness; p<0.01
  • Nausea/vomiting: 23.5% vs. 7.8%; p=0.03
  • Somnolence/drowsiness: 19.6% vs. 11.8%; p=0.28
  • Dry mouth: 23.5% vs. 2.0%; p<0.01
  • Indigestion: 11.8% vs. 9.8%; p=0.75
  • No significant differences between groups in liver function tests
QUALITY RATING:Fair

From: Evidence Tables

Cover of Drug Class Review: Second-Generation Antidepressants
Drug Class Review: Second-Generation Antidepressants: Final Update 5 Report [Internet].
Gartlehner G, Hansen RA, Reichenpfader U, et al.
Portland (OR): Oregon Health & Science University; 2011 Mar.
Copyright © 2011 Oregon Health & Science University.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.