Social Anxiety Disorder Treatments: Therapy, Drugs and the Data

Social anxiety disorder treatments fall into two groups, psychotherapy and medication, used alone or together, according to the National Institute of Mental Health (NIMH)^[6]. Cognitive behavioral therapy (CBT) is the best-studied option, and NIMH calls it the "gold standard" psychotherapy for social anxiety disorder^[6].

The medications NIMH lists are SSRI and SNRI antidepressants, beta-blockers and benzodiazepines^[6]. A 2014 network meta-analysis of 101 trials and 13,164 adults compared 41 interventions and control conditions and concluded that individual CBT should be regarded as the best initial treatment and that SSRIs show the most consistent evidence for people who decline psychological treatment^[1].

The evidence behind each option differs widely. Several treatments beat a waitlist but did not clearly beat a placebo in a 2014 analysis^[1], and beta-blockers showed no clear benefit over placebo in the two controlled trials that tested them for the disorder^[2].

Of the treatment classes that beat a waitlist in the 2014 Lancet Psychiatry analysis, individual CBT had the largest effect, with an SMD of 1.19^[1]. MAO inhibitors (1.01), benzodiazepines (0.96), group CBT (0.92) and SSRIs and SNRIs (0.91) also outperformed a waitlist^[1].

Effects look smaller against a placebo than against a waitlist. Only two classes also beat an appropriate placebo: individual CBT against a psychological placebo (SMD 0.56) and SSRIs and SNRIs against a pill placebo (SMD 0.44)^[1].

Individual CBT also outperformed psychodynamic psychotherapy, and it outperformed a combined group of interpersonal therapy, mindfulness and supportive therapy^[1]. The CBT for social anxiety review covers session structure, length and online formats in detail.

Prescribing labels that list social anxiety disorder as an FDA-approved use include paroxetine (Paxil) and sertraline (Zoloft), both SSRIs, and extended-release venlafaxine (Effexor XR), an SNRI^[7][8][9].

The 2017 Cochrane review of social anxiety medication included 66 randomized trials with 11,597 adults^[2]. Across 24 SSRI trials with 4,984 participants, about 523 per 1,000 people responded on an SSRI versus 317 per 1,000 on placebo, evidence the reviewers graded as very low quality^[2].

Side effects are the trade-off. Dropouts because of adverse events rose from about 40 per 1,000 on placebo to 105 per 1,000 on SSRIs^[2]. NIMH lists headaches, nausea and difficulty sleeping as possible antidepressant side effects and notes that SSRIs and SNRIs may take several weeks to start working^[6].

Other drug classes

MAO inhibitors (4 trials, 235 people) and reversible MAO-A inhibitors (8 trials, 1,270 people) also raised response rates over placebo, on low-quality evidence^[2]. Anticonvulsants that act as GABA analogues improved response in 3 trials with 532 people, on moderate-quality evidence^[2].

Benzodiazepines produced the largest response ratio in the review, but it rests on only 2 trials with 132 people^[2]. NIMH notes that they act quickly but that some people build tolerance or become dependent, so providers may prescribe them only for brief periods^[6]. Drug names and classes across all anxiety disorders are covered in the anxiety medication guide.

Beta-blockers can help control physical symptoms such as a rapid heart rate, sweating and tremors, and NIMH notes they are commonly the medication of choice for the "performance anxiety" type of social anxiety disorder^[6]. The beta-blocker trials in the Cochrane review include atenolol^[2].

Controlled evidence for the broader disorder is thin. In the Cochrane review, two trials with 97 participants found no evidence that beta-blockers increased response over placebo (risk ratio 1.09, very low quality)^[2].

Those trials do not settle the question for performance-only fears. The Cochrane authors note that forms of the disorder consisting of "pure" public speaking fears were not well represented in the trials they reviewed^[2].

Group CBT, exposure with social skills training, and guided self-help all beat a waitlist in the 2014 analysis, with SMDs of 0.92, 0.86 and 0.86^[1]. Self-help without therapist support (0.75) and psychodynamic psychotherapy (0.62) had smaller effects^[1].

Exposure therapy, a CBT method, works by progressively confronting the situations a person avoids, and NIMH notes that group CBT can offer unique benefits for social anxiety disorder^[6]. Acceptance and commitment therapy (ACT) uses mindfulness and goal setting, and NIMH describes a growing body of research supporting it for this condition^[6]. The ACT therapists explainer covers how that approach works.

Support groups are a complement rather than a substitute. NIMH says members can offer honest feedback about how others perceive them, while cautioning that peer advice does not replace a provider's treatment recommendations^[6]. Therapy for anxiety compares these approaches across other anxiety disorders.

One direct test is a 2010 placebo-controlled trial of 128 adults that compared cognitive behavioral group therapy (CBGT), the MAO inhibitor phenelzine, their combination and pill placebo^[3]. At week 24, 78.1% of the combined group had responded, compared with 52.9% on CBGT, 48.6% on phenelzine and 33.3% on placebo^[3].

Remission showed a wider gap: 53.1% with combined treatment versus 23.5% for CBGT and 25.7% for phenelzine at week 24^[3]. The authors concluded that combined treatment was superior to either treatment alone and to placebo^[3].

That result comes from one trial using group rather than individual CBT and an MAO inhibitor rather than an SSRI, so it does not show that every therapy-plus-SSRI pairing adds benefit. NIMH frames the choice between psychotherapy, medication or both as one based on a person's needs, preferences and medical situation^[6].

Antidepressants for social anxiety may take several weeks to start working, and NIMH notes that finding the best treatment may take trial and error^[6].

Among people who had responded, staying on an SSRI was associated with fewer relapses than switching to placebo: about 135 versus 397 per 1,000, in 3 trials with 389 participants^[2]. SSRIs were the only medication with moderate-quality evidence for reducing relapse in the Cochrane review^[2].

A 2017 BMJ meta-analysis of 28 relapse-prevention trials (5,233 patients) across anxiety disorders, OCD and PTSD found relapse rates of 36.4% after switching to placebo versus 16.4% on continued antidepressants, over follow-up of up to one year^[4]. Subgroup analyses, including by type of disorder, found no significant differences, and the authors wrote that the lack of data beyond one year should not be read as advice to stop at that point^[4].

The American Academy of Child and Adolescent Psychiatry's 2020 guideline covers anxiety disorders in children and adolescents as a group^[10]. It found that CBT and SSRIs have "considerable empirical support as safe and effective short-term treatments," and that SNRIs have some support as an additional option^[10].

The guideline also points to a "protracted severe shortage" of child and adolescent behavioral health specialists and lists long-term SSRI and SNRI effects as an open research question^[10]. Teen-specific figures are on the teen anxiety page.

Most people with the disorder go untreated in a given year. In WHO World Mental Health surveys of 142,405 respondents across 28 surveys, 38% of people with past-year social anxiety disorder received any treatment that year, counting everything from specialty mental health care to complementary and non-health-care services^[5].

Access tracked national income. Among people with any impairment, treatment rates ranged from 18.0% in low and lower-middle income countries to 44.2% in high-income countries^[5].

U.S. prevalence figures are on the social anxiety disorder page, and the mental health treatment gap analysis covers why care falls short across conditions.