SSRI for Anxiety: Which Ones Work Best and What Trials Show

Selective serotonin reuptake inhibitors (SSRIs) are a first-line medication for panic disorder, generalized anxiety disorder (GAD) and social anxiety disorder, alongside SNRIs^[1]. FDA approval to market an SSRI for anxiety is disorder-specific: escitalopram is approved for GAD, sertraline for panic and social anxiety disorder, paroxetine for all three, and fluoxetine for panic disorder^[1].

NIMH notes that for panic disorder and social anxiety disorder, clinicians typically start with SSRIs or other antidepressants because they have fewer side effects than other medications^[3]. How common anxiety disorders are is covered on our pillar page; this guide covers how SSRIs perform for them.

Citalopram and fluvoxamine have no FDA approval for these three disorders but are used off-label for all of them^[1]. Escitalopram's current label, revised in 2023, covers GAD in adults and in children 7 years and older^[4].

DSM-5 places obsessive-compulsive disorder and PTSD outside the anxiety disorders chapter^[5], so this guide focuses on panic disorder, GAD and social anxiety disorder.

No single SSRI has been shown to be the best for anxiety. A 2019 meta-analysis of 57 placebo-controlled trials with 16,056 adults found no significant efficacy differences between drugs within the SSRI class, or between SSRIs and SNRIs^[2]. Both classes showed their largest benefits in social anxiety disorder^[2].

For GAD, a 2019 Lancet network meta-analysis of 89 trials and 25,441 patients found escitalopram more effective than placebo with relatively good acceptability, improving Hamilton Anxiety Scale (HAM-A) scores by 2.45 points more than placebo^[6]. Sertraline and fluoxetine were also efficacious and well tolerated, but those findings rested on small samples^[6].

Paroxetine is the only SSRI with FDA approval for all three disorders^[1]. In the same Lancet analysis it was effective for GAD but poorly tolerated compared with placebo^[6]. A 2020 clinical review notes that escitalopram and duloxetine, an SNRI, may have the largest effect sizes in GAD^[1].

For social anxiety disorder, a 2017 Cochrane review pooled 24 SSRI trials with 4,984 participants and found SSRIs raised the likelihood of treatment response over placebo (risk ratio 1.65), although the evidence was rated very low quality^[7]. SSRIs were the only medication class with moderate-quality evidence for preventing relapse in that review^[7]. More detail is in our guide to social anxiety disorder treatments.

NIMH says antidepressants usually take 4 to 8 weeks to work, and that sleep, appetite, energy and concentration often improve before mood lifts^[3]. The three adult trials behind escitalopram's GAD approval each ran 8 weeks^[4].

Pooled weekly data show SSRI gains over placebo building at a similar rate across the acute treatment phase, while SNRI gains came mostly early in treatment^[2].

Dose matters more for SSRIs than for SNRIs: higher SSRI doses within the therapeutic range were associated with greater symptom improvement and higher odds of response, a relationship not seen for SNRIs^[2]. Higher doses of both classes were also linked to more dropouts due to side effects^[2].

Escitalopram's label sets 10 mg once daily as the adult starting dose for GAD, with a maximum of 20 mg once daily that can be reached no sooner than one week later^[4].

Common side effects in anxiety trials

In escitalopram's placebo-controlled adult GAD trials, nausea affected 18% of patients vs. 8% on placebo, insomnia 12% vs. 6%, and fatigue 8% vs. 2%^[4]. Eight percent of patients on escitalopram stopped because of an adverse event, compared with 4% on placebo^[4].

Sexual side effects showed the widest gap: 14% of men reported ejaculation disorder vs. 2% on placebo, and 6% of women reported anorgasmia vs. under 1%^[4]. A 2020 clinical review describes sexual dysfunction as more durable than nausea, headache, dry mouth, diarrhea and constipation, which tend to be short-lived^[1].

NIMH describes SSRI side effects as generally mild and likely to fade with time^[3]. It also notes that combining SSRIs with other serotonin-acting products, such as triptans or St. John's wort, can cause serotonin syndrome, a rare but life-threatening reaction^[3].

Can SSRIs make anxiety worse at first?

Early worsening of anxiety, agitation and irritability after starting an antidepressant is called jitteriness/anxiety syndrome^[8]. A 2009 systematic review of 107 articles found published incidence rates ranging from 4% to 65% of people starting antidepressants, and no validated rating scale for the syndrome^[8].

The same review found no robust evidence that the syndrome is more common with SSRIs than with tricyclic antidepressants, or more common in anxiety disorders^[8]. Slower dose increases and added benzodiazepines are common ways to manage it, but conclusive evidence for either is lacking^[8].

Antidepressants, SSRIs included, carry a boxed warning for increased suicidal thoughts and behaviors in pediatric and young adult patients^[4]. In pooled placebo-controlled trials of about 77,000 adults and 4,500 pediatric patients, the incidence was higher on antidepressants than placebo in patients 24 and younger^[4].

In absolute terms, the pooled data showed 14 additional patients with suicidal thoughts or behaviors per 1,000 treated under age 18, and 5 additional per 1,000 at ages 18 to 24^[4]. The direction reversed with age, to 1 fewer per 1,000 at ages 25 to 64 and 6 fewer at 65 and older^[4].

NIMH notes the increase can occur especially in the first few weeks after starting a medication or changing the dose^[3]. For children and teens, the American Academy of Child and Adolescent Psychiatry's 2020 guideline concludes that CBT and SSRIs both have considerable support as safe and effective short-term treatments for anxiety^[9]. How therapy compares is covered in therapy for anxiety.

Recommended treatment length varies, from as short as 3 to 6 months to 1 to 2 years or longer^[1]. International guidelines are consensus based and advise continuing for 6 to 24 months before tapering^[5].

The strongest data on stopping come from a 2017 BMJ meta-analysis of 28 relapse-prevention trials with 5,233 patients who had responded to an antidepressant^[5]. Those switched to placebo relapsed at a pooled rate of 36.4%, compared with 16.4% of those who kept taking the drug^[5].

Discontinuation raised the odds of relapse about threefold (odds ratio 3.11), and follow-up in these trials ran up to one year, so the data say little about stopping after longer treatment^[5]. In a separate Cochrane review, SSRIs were the only drug class with moderate-quality evidence for preventing relapse of social anxiety disorder^[7].

Stopping an SSRI, particularly abruptly, can bring discontinuation symptoms such as dizziness, irritability, electric shock sensations, headache and insomnia, which escitalopram's label describes as generally self-limiting^[4]. NIMH notes that a provider can decrease the dose slowly and safely, and that stopping a medication too soon may cause unpleasant or harmful side effects^[3].

SNRIs such as duloxetine and venlafaxine are the other first-line class for anxiety disorders^[1]. Pooled trials found no significant efficacy difference between SSRIs and SNRIs, though SNRI gains tended to arrive earlier^[2].

Benzodiazepines are used as needed for severe short-term anxiety, but NIMH notes that long-term use may lead to tolerance or dependence, so they are usually prescribed for short periods^[3]. In the 2019 Lancet analysis, benzodiazepines were effective for GAD but poorly tolerated compared with placebo^[6].

Buspirone, a non-benzodiazepine option, must be taken daily for 3 to 4 weeks to reach full effect and does not work on an as-needed basis^[3]. The full list of drug classes is in our guide to anxiety medication, and prescribing trends are in anxiety medication statistics.