free · from the FDA record · 33 of 37 medications state a trial length

How long were psychiatric medications tested before approval?

for the antidepressants below, the trials that earned FDA approval ran 4 to 12 weeks — most of them six to eight. every row here is the pivotal-trial record from the medication's own label and Drugs@FDA: how long, how many people, compared against what, and the application number so you can check it. this is what approval rested on. it is not evidence about years of use, and the page says so wherever that matters.

the short version

  • an approval trial answers one question: does the drug beat placebo on a symptom scale over an acute episode. six to eight weeks is enough to measure that.
  • what it cannot answer is what happens over years. that evidence, where it exists, comes from relapse-prevention extension studies, post-marketing reports, and observational data — each medication page walks through what its label actually says before you stop or change a dose.
  • where the public record does not state a number, the cell below says so. nothing is estimated.

antidepressants

medicationapprovedapproved fortrial lengthpeoplecompared againstapplication
sertraline1991major depressive disorder in adults6 and 8 weeksnot statedPlacebo and amitriptylineNDA 019839
fluoxetine1987major depressive disorder, acute and maintenance treatment5 and 6 weeks (acute adult trials)not statedPlaceboNDA 018936
paroxetine1992major depressive disorder in adults6 weeksnot statedPlaceboNDA 020031
citalopram1998major depressive disorder in adults4 to 6 weeks1,063 patients on citalopram in the pooled placebo-controlled trialsPlaceboNDA 020822
escitalopram2002major depressive disorder in adults8 weeks715 patients on escitalopram in the placebo-controlled trialsPlacebo; citalopram as an active comparator in two studiesNDA 021323
venlafaxine1997major depressive disorder in adults; indications for generalized anxiety disorder, social anxiety disorder, and panic disorder were added later8 and 12 weeks705 patients on the extended-release capsules across MDD studiesPlaceboNDA 020699
desvenlafaxine2008the treatment of adults with major depressive disorder8 weeks1,834 patients on desvenlafaxine, 1,116 on placeboPlaceboNDA 021992
duloxetine2004major depressive disorder in adults; indications for generalized anxiety disorder, diabetic peripheral neuropathic pain, fibromyalgia, and chronic musculoskeletal pain were added later8 to 9 weeks3,779 patients on duloxetine across placebo-controlled MDD trialsPlaceboNDA 021427
bupropion1996the treatment of major depressive disorder; this anchor is the 1996 sustained-release formulation, whose efficacy rests on trials of the earlier immediate-release form4 to 6 weeksDrug arms of 34 to 110 patientsPlaceboNDA 020358
mirtazapine1996the treatment of major depressive disorder in adults6 weeks453 patients on mirtazapine, 361 on placebo (pooled U.S. 6-week trials)PlaceboNDA 020415
trazodone1981the treatment of major depressive disorder in adultsnot stated in the public record142 inpatients and 157 outpatients on trazodone in the controlled studiesPlaceboNDA 018207
vortioxetine2013major depressive disorder in adults6 to 8 weeks2,616 patients on vortioxetine in the short-term studiesPlaceboNDA 204447
esketamine2019treatment-resistant depression in adults, in conjunction with an oral antidepressant; the monotherapy option and the indication for depressive symptoms in major depressive disorder with acute suicidal ideation or behavior were added later4 weeks114 on esketamine vs 109 on placebo sprayPlacebo nasal spray; both arms on a newly initiated oral antidepressantNDA 211243

antipsychotics

medicationapprovedapproved fortrial lengthpeoplecompared againstapplication
aripiprazole2002schizophrenia, and acute manic and mixed episodes associated with bipolar I disorder4 to 6 weeks367 to 420 patients per positive trialPlacebo; risperidone or haloperidol arms in three trialsNDA 021436
olanzapine1996treatment of schizophrenia; the 1996 approval covered adults, with the bipolar I disorder and adolescent indications added later6 weeks (acute trials)149, 253, and 326 patientsPlacebo; haloperidol arm in one trialNDA 020592
quetiapine1997schizophrenia6 weeks286 to 618 patients per trialPlacebo in two trials; dose comparison in oneNDA 020639
risperidone1993schizophrenia in adults4 to 8 weeks160 to 1,356 patients per trialPlacebo in three trials; dose comparison in oneNDA 020272
lurasidone2010treatment of schizophrenia; the 2010 approval was followed later by the adolescent and bipolar-depression indications6 weeks145 to 489 patients per studyPlacebo; olanzapine or quetiapine XR in two studiesNDA 200603
ziprasidone2001schizophrenia; acute bipolar mania4 and 6 weeks (short-term trials)139 to 419 patients per short-term trialPlacebo; haloperidol arm in one trialNDA 020825

mood stabilizers

medicationapprovedapproved fortrial lengthpeoplecompared againstapplication
lithium1970acute manic and mixed episodes of bipolar I disorder, and maintenance treatment, in patients 7 years and older; the 1970 approval predates by decades the pediatric trial that today's label describes8 weeks81 patients aged 7 to 17PlaceboNDA 016782
lamotrigine2003maintenance treatment of bipolar I disorder to delay the time to occurrence of mood episodes (depression, mania, hypomania, mixed episodes) in patients treated for acute mood episodes with standard therapy; the epilepsy indications came first, in 1994Up to 18 months404 and 171 patients in the 2 trialsPlaceboNDA 020241 SUPPL-17

benzodiazepines and other anxiolytics

medicationapprovedapproved fortrial lengthpeoplecompared againstapplication
alprazolam1981the acute treatment of generalized anxiety disorder and the treatment of panic disorder, with or without agoraphobia, in adults4 weeks (anxiety); up to 10 weeks (panic)565 patients on alprazolam vs 505 on placebo (anxiety trials)PlaceboNDA 018276
clonazepam1975seizure disorders (Lennox-Gastaut syndrome, akinetic and myoclonic seizures, and absence seizures unresponsive to succinimides); the panic disorder indication was added later6 to 9 weeks574 patients on clonazepam vs 294 on placeboPlaceboNDA 017533
diazepam1963the management of anxiety disorders or the short-term relief of the symptoms of anxiety; the label also covers acute alcohol withdrawal, skeletal muscle spasm, and adjunctive use in convulsive disordersnot stated in the public recordnot statednot statedNDA 013263
lorazepam1977the management of anxiety disorders or the short-term relief of the symptoms of anxiety or anxiety associated with depressive symptomsnot stated in the public recordAbout 3,500 patients in the anxiety samplenot statedNDA 017794
buspirone1986the management of anxiety disorders or the short-term relief of the symptoms of anxiety3 to 4 weeksAbout 2,200 patients in the efficacy trialsPlaceboNDA 018731
hydroxyzine1956symptomatic relief of anxiety and tension associated with psychoneurosis, and as an adjunct in organic disease states in which anxiety is manifested; the label also covers pruritus from allergic conditions and pre- and post-anesthetic sedationnot stated in the public recordnot statednot statedNDA 010485

ADHD medications

medicationapprovedapproved fortrial lengthpeoplecompared againstapplication
amphetamine2001the treatment of attention deficit hyperactivity disorder (ADHD) in adults and pediatric patients 6 years and older3 to 4 weeks584 children, 327 adolescents, 255 adultsPlaceboNDA 021303
lisdexamfetamine2007the treatment of attention deficit hyperactivity disorder (ADHD), initially in children aged 6 to 12; the adolescent, adult, and binge eating disorder indications were added later4 weeks (1 week per treatment in the crossover studies)290 children in the parallel-group trialPlacebo; Adderall XR in one classroom studyNDA 021977
methylphenidate2000the treatment of attention deficit hyperactivity disorder (ADHD); the current label covers patients aged 6 to 65 years old1 to 7 weeks416 children, 177 adolescents, 627 adultsPlacebo and immediate-release methylphenidateNDA 021121
atomoxetine2002the treatment of attention-deficit/hyperactivity disorder (ADHD) in adults and pediatric patients 6 years of age and older6 to 10 weeks1,613 pediatric patients and 541 adults received atomoxetine in acute placebo-controlled studiesPlacebo; methylphenidate in two pediatric studiesNDA 021411
guanfacine2009the treatment of attention deficit hyperactivity disorder (ADHD) in children and adolescents 6 to 17, as monotherapy and as adjunctive therapy to stimulant medications5 to 8 weeks345 and 324 in the fixed-dose trialsPlaceboNDA 022037

sleep medications

medicationapprovedapproved fortrial lengthpeoplecompared againstapplication
zolpidem1992the short-term treatment of insomnia characterized by difficulties with sleep initiation1 night to 5 weeks35 to 462 patients per trialPlaceboNDA 019908
temazepam1981the short-term treatment of insomnia (generally 7 to 10 days)2 weeks1,076 patients on temazepam vs 783 on placeboPlaceboNDA 018163

other

medicationapprovedapproved fortrial lengthpeoplecompared againstapplication
buprenorphine2010treatment of opioid dependence, as part of a complete treatment plan that includes counseling and psychosocial support (buprenorphine/naloxone sublingual film)4 and 16 weeks (safety studies)Over 3000 opioid-dependent subjectsPlacebo, in the 4-week tablet studyNDA 022410
gabapentin1993adjunctive therapy for partial-onset seizures in adults and pediatric patients 3 years and older with epilepsy, and management of postherpetic neuralgia in adults; the label lists no psychiatric indication12 weeks705 patients aged 12 and older, plus 247 childrenPlacebo, added to existing antiepileptic drugsNDA 020235
modafinil1998improving wakefulness in adult patients with excessive sleepiness associated with narcolepsy; the obstructive sleep apnea and shift work disorder indications were added later9 weeks558 patientsPlaceboNDA 020717

straight answers

How long were antidepressants tested in clinical trials before approval?

For the 12 antidepressants on this page whose FDA record states a trial length, the placebo-controlled trials that earned approval ran 4 to 12 weeks, with most between six and eight weeks. That is the acute-treatment trial the label reports. Longer relapse-prevention studies exist for some drugs and are described on each medication's own page; none of the approval trials listed here ran for years.

What are the long-term effects of taking an antidepressant like Zoloft for 10 years?

The approval record cannot answer that, and this page does not pretend it can. The trials that approved sertraline (Zoloft) ran six and eight weeks; no drug on this page was approved on a decade-long controlled trial. What exists for longer use is relapse-prevention extension studies, post-marketing adverse-event reports (FAERS), and observational data — different kinds of evidence with different limits. Each medication page lists what its label says under "Before you stop or change your dose", and the decision about staying on a medication belongs with you and your prescriber, not a table.

Where do these numbers come from?

Every row is hand-curated from the medication's FDA label (the Clinical Studies section on DailyMed) and Drugs@FDA, and carries the application number so you can pull the approval documents yourself. Where the public record does not state a figure, the cell says so rather than filling it in. Nothing here is estimated or modelled.

Why do trials run only a few weeks?

Because the regulatory question is whether the drug beats placebo on a symptom scale over an acute episode, and that can be measured in six to eight weeks. It is a reasonable way to test efficacy and a poor way to learn about years of use — which is why the label, the FAERS reports, and the relapse-prevention studies all have to be read together. This page shows the first of those; the medication pages show the rest.

related

how long the trials in our evidence library ran — a different dataset: the 72 randomised trials we score, not the approval record · side-effect rates, drug vs placebo from the same labels · the medication library with each drug's approval walked through step by step · how we score evidence. veisund is free peer support and public-record information — not medical advice. in crisis in the US, call or text 988.