Antidepressants vs Placebo for Moderate to Severe Generalized Anxiety Disorder

Are antidepressants effective and tolerable for treatment of moderate to severe generalized anxiety disorder (GAD) in patients without serious medical comorbidities?

More patients with GAD achieve a 50% reduction in self-reported anxiety symptoms by using antidepressants compared with placebo (number needed to treat [NNT] for additional benefit = 7). Fewer participants who use antidepressants discontinue treatment due to lack of effectiveness (NNT = 27); more participants discontinue antidepressant treatment because of adverse effects (number needed to harm [NNH] = 17).¹ (Strength of Recommendation: A, consistent, good-quality patient-oriented evidence.)

GAD has a lifetime prevalence of 5.1%, according to the US National Comorbidity Survey.^2,3 GAD can cause significant morbidity and affects women twice as often as men.^2,3 Selective serotonin reuptake inhibitors (SSRIs) and serotoninnorepinephrine reuptake inhibitors (SNRIs) are considered first-line pharmacotherapy options for depression.^1,4 The authors of this Cochrane review sought to determine whether antidepressants are appropriate for patients with GAD. Known adverse effects of antidepressants include gastrointestinal, psychiatric, and sexual adverse effects.¹

This Cochrane review included 37 randomized controlled trials (RCTs) with 12,226 participants.¹ RCTs were selected if they compared monotherapy antidepressants (ie, tricyclic antidepressants, SSRIs, SNRIs, monoamine oxidase inhibitors, noradrenergic and specific serotonergic antidepressants, norepinephrine and dopamine reuptake inhibitors, noradrenergic reuptake inhibitors) with placebo for the treatment of GAD. Of the 37 RCTs, 15 trials were conducted in the United States; 14 were multinational; and single studies were conducted in China, Japan, the United Kingdom, and Greece. Participants were outpatient adults with the primary diagnosis of moderate to severe GAD classified by the Diagnostic and Statistical Manual of Mental Disorders, 4th ed. (DSM-IV); DSM-IV, text revision; or DSM 3rd ed., revised. Follow-up ranged from 4 to 28 weeks with no standardization of drug dosing or duration of therapy. Studies in which participants regularly used benzodiazepines or psychosocial therapies targeting GAD were excluded.

Use of any antidepressant increased the treatment response compared with placebo (risk ratio [RR] = 1.41; 95% CI, 1.29–1.55; 20 RCTs; n = 7,267; high-certainty evidence).¹ That is, treatment with any antidepressant increased the likelihood of a 50% response on the Hamilton Anxiety Rating Scale (HAMA) over 6 to 24 weeks (NNT = 7).

Withdrawal because of ineffective treatment was lower in the therapy group (RR = 0.41; 95% CI, 0.33–0.50; 29 RCTs; n = 11,007; high-certainty evidence). Risk of dropout due to adverse effects increased in the treatment group (RR = 2.18; 95% CI, 1.81–2.61; NNH = 17; 32 RCTs; n = 11,793; high-certainty evidence). The overall risk of bias was high in all included studies.

Treatment with an SSRI increased the likelihood of a 50% response on the HAM-A over 8 to 12 weeks (NNT = 6). Withdrawal due to adverse effects increased by 98% in the treatment group (NNH = 24; high-certainty evidence).

Treatment with an SNRI increased the likelihood of a 50% response on the HAM-A over 8 to 24 weeks (NNT = 8). Withdrawal due to adverse effects increased by 142% in the treatment group (NNH over 8–28 weeks = 10; high-certainty evidence).