FDA-approved vs off-label: what psychiatric medications are actually approved for
a great deal of ordinary psychiatric prescribing is off-label, and that is legal, common, and often reasonable — but it changes what the evidence behind your prescription actually is. trazodone is approved for depression, not for the low-dose insomnia use it is mostly prescribed for. lamotrigine is approved to prevent bipolar mood episodes, not to treat an acute one. sertraline is approved for several anxiety disorders but not for generalized anxiety disorder. below is the label-verified table for 21 medications, read off the current FDA labels rather than summarised from a secondary source.
the table
last verified: august 18, 2026; labels read august 17, 2026. the "approved for" column is the indications section of the current FDA structured product label. the "commonly assumed, not approved" column is a negative read of that same label — it lists conditions where the label is silent, which is a statement about what FDA has reviewed, not a claim that the drug does not work. numbered references link to the sources at the bottom of the page.
| medication | class | FDA-approved for | commonly assumed, not approved | sources |
|---|---|---|---|---|
| sertralineZoloft | SSRI |
| generalized anxiety disorder. this is the single most common misconception about sertraline: it is approved for four anxiety-spectrum conditions and GAD is not one of them. prescribing it for GAD is off-label. | [1] |
| escitalopramLexapro, Cipralex | SSRI |
| OCD, panic disorder, PTSD, social anxiety disorder — all off-label for escitalopram. | [1] |
| fluoxetineProzac, Sarafem | SSRI |
| generalized anxiety disorder, PTSD, social anxiety disorder. | [1] |
| paroxetinePaxil, Pexeva, Brisdelle | SSRI |
| nothing on the anxiety spectrum — paroxetine carries the broadest set of anxiety indications of any SSRI. that breadth is a labelling fact, not a ranking of effectiveness. | [1] |
| venlafaxineEffexor | SNRI |
| PTSD and OCD. note the formulation split: three of the four indications belong to the extended-release form only, so immediate-release venlafaxine for GAD is off-label. | [1] |
| duloxetineCymbalta | SNRI |
| panic disorder, PTSD, OCD, social anxiety disorder. | [1] |
| bupropionWellbutrin, Zyban, Aplenzin, Forfivo | atypical antidepressant |
| any anxiety disorder, and ADHD. bupropion is frequently prescribed for both; neither is on the label. | [1] |
| mirtazapineRemeron | atypical antidepressant |
| insomnia and appetite stimulation — two of the most common reasons it is actually prescribed. both are off-label. | [1] |
| trazodoneDesyrel, Oleptro | atypical antidepressant |
| insomnia. trazodone is the most commonly prescribed antidepressant for insomnia in the US and that use is off-label — see the detail below. | [1] [2] [3] |
| quetiapineSeroquel, Seroquel XR | antipsychotic |
| generalized anxiety disorder and insomnia — two very common off-label uses. see the detail below for what the trials show. | [4] [5] [6] |
| aripiprazoleAbilify | antipsychotic |
| anxiety disorders, PTSD, and monotherapy for unipolar depression (the MDD indication is adjunctive only). | [7] |
| lamotrigineLamictal | mood stabiliser |
| acute bipolar depression and acute mania. this is the highest-value distinction on the page: lamotrigine’s bipolar indication is maintenance — preventing the next episode — not treating the one you are in. | [8] |
| lithiumLithobid, Eskalith | mood stabiliser |
| unipolar depression augmentation — a long-standing and evidence-supported off-label use that is not on the label. | [9] |
| alprazolamXanax | benzodiazepine |
| insomnia, and long-term anxiety management. note the word "acute" in the GAD indication — it is doing real work. | [10] |
| lorazepamAtivan | benzodiazepine |
| insomnia. the label itself states that effectiveness beyond 4 months has not been systematically established — the tablet label carries no long-term claim at all. (the injectable form carries separate indications, including status epilepticus, which were not verified here.) | [11] |
| clonazepamKlonopin, Rivotril | benzodiazepine |
| generalized anxiety disorder — clonazepam’s only psychiatric indication is panic disorder. GAD, insomnia and situational anxiety are all off-label. | [12] |
| buspironeBuspar | other anxiolytic |
| depression, including as an add-on to an antidepressant — the use tested in the government-funded STAR*D trial, and off-label. | [13] |
| hydroxyzineVistaril, Atarax | other anxiolytic |
| insomnia as a standalone indication, and any specific DSM anxiety disorder by name. | [14] |
| zolpidemAmbien, Edluar, Intermezzo | z-drug |
| sleep-maintenance insomnia (for the standard formulation), long-term use, and anxiety. the label carries a boxed warning added in april 2019 for complex sleep behaviours — sleepwalking, sleep-driving — including cases resulting in death, and a sex-specific starting dose of 5 mg for women. | [15] [16] |
| amphetamineAdderall, Mydayis, Evekeo | stimulant |
| depression, cognitive enhancement, and weight loss. note the formulation split: the narcolepsy indication is on the immediate-release tablet label; Adderall XR and Mydayis carry ADHD only. | [17] |
| methylphenidateRitalin, Concerta, Metadate, Daytrana, Quillivant, Jornay | stimulant |
| depression and cognitive enhancement. again the formulation split matters — some extended-release brands, including Concerta, carry ADHD only. | [18] |
what "off-label" actually means
the FDA approves a drug for specific uses, in specific populations, based on trials a manufacturer submitted. that approval governs what the manufacturer may market. it does not govern what a licensed clinician may prescribe. once a drug is on the market for anything, a prescriber can use it for any condition in their clinical judgement — that is what "off-label" means, and it is legal and routine.
what changes off-label is the evidence you are standing on. an approved indication means FDA reviewed trials designed to answer that exact question and judged them adequate. an off-label use might be backed by a large independent trial (buspirone augmentation for depression was tested in the government-funded STAR*D trial [13]), a decent Cochrane review (quetiapine for generalized anxiety [5]), a modest one (trazodone for insomnia [2][3]), or almost nothing at all (quetiapine for insomnia: one randomised trial, thirteen patients [6]). the label will not tell you which. only asking will.
the three distinctions that get blurred most often
trazodone: approved for depression, prescribed for sleep. its label lists major depressive disorder and nothing else [1]. the low-dose insomnia use that accounts for most trazodone prescriptions in the US is off-label. the evidence is genuinely mixed rather than absent: pooling 3 trials in 370 people, Cochrane found a moderate short-term improvement in self-rated sleep (SMD −0.34) while the trials that measured sleep in a lab found little or no change in sleep efficiency [2]. a larger 2024 meta-analysis of 44 trials found about 28 minutes of measured sleep versus placebo — and more than twice as many people stopping because of side effects [3]. worth knowing on both sides of the ledger.
lamotrigine: approved to prevent episodes, not to treat one. the bipolar indication reads "maintenance treatment of bipolar I disorder to delay the time to occurrence of mood episodes... after acute stabilisation" [8]. acute bipolar depression and acute mania are not on the label. if you were started on lamotrigine while in an episode, that is off-label use — common and often defensible, but a different evidentiary situation from what the label supports.
sertraline: approved for several anxiety disorders, not for GAD. panic disorder, PTSD and social anxiety disorder are all on the label; generalized anxiety disorder is not [1]. escitalopram and duloxetine carry the GAD indication; paroxetine and venlafaxine XR do too. this is the query where the internet is least helpful, because "approved for anxiety" gets answered as if anxiety were one condition.
the formulation trap
indications attach to products, not molecules — and the split is invisible in ordinary conversation. venlafaxine immediate-release is approved for major depressive disorder only; the GAD, social anxiety and panic indications belong to the extended-release form [1]. the adjunctive-depression indication belongs to Seroquel XR, not immediate-release Seroquel [4]. narcolepsy is on the immediate-release amphetamine tablet label while Adderall XR and Mydayis carry ADHD only [17], and among methylphenidate products some extended-release brands including Concerta carry ADHD only [18]. "that drug is approved for this" can be simultaneously true and false depending on which bottle is in front of you.
where approved and off-label meet the evidence
an approved indication is not a claim of superiority. the largest independent comparison of antidepressants — 522 trials, 116,477 adults, publicly funded — found all 21 drugs beat placebo for acute major depression, but with meaningful differences in effectiveness and dropout between them, and trazodone ranked among the least effective and highest-dropout of the set for depression itself [19]. approval tells you FDA reviewed evidence for that use. it does not tell you the drug is the best option for you, and it does not rank it.
the reverse also holds. an off-label use can rest on better evidence than some approved uses, which is why "is it FDA-approved?" is the beginning of a conversation with your prescriber rather than the end of one.
related evidence pages
for substances that are not prescriptions at all — ashwagandha, kratom, CBD, St John’s wort, melatonin and eleven more, none of which is FDA-approved for any mental health condition — see supplements and non-prescription substances for mental health. for the drugs still moving through the FDA, see the psychedelic therapy FDA status tracker, and for the ketamine question specifically, ketamine vs esketamine: what is actually approved.
how we verify
how we verify this page. every indication in the table was read off the current FDA structured product label on DailyMed on 2026-08-17, drug by drug — not summarised from a secondary source, a drug database, or another article. where the specific label setid was recorded during verification, it is cited and linked in the sources list so you can open the same document we read. where it was not, the source line says so plainly rather than linking a label we did not confirm.
the "not approved for" column is a negative read. it lists conditions the label does not mention, chosen because they are commonly assumed to be approved uses. it is not a claim that the drug does not work for them, and several of those uses are supported by good trials — which is exactly the point of the page.
every trial cited was verified independently. each study on this page was fetched from PubMed via NCBI E-utilities on 2026-08-17 and checked for retractions; funding was recorded only where the funding statement could actually be read, and marked UNKNOWN otherwise rather than inferred from the journal or the authors.
we are not clinicians. this page reports what FDA labels say. it is not medical advice and it is not a reason to change or stop anything you are taking.
this is not medical advice — talk to your prescriber. finding out a prescription is off-label is a reason to ask a question at your next appointment, never a reason to stop taking it: stopping antidepressants, benzodiazepines and mood stabilisers abruptly can be genuinely dangerous. if you're in crisis, call or text 988 (u.s.), 24/7, free.
questions
is sertraline FDA-approved for anxiety?
partly — and the distinction matters. sertraline’s label covers panic disorder, post-traumatic stress disorder and social anxiety disorder, all of which are anxiety disorders. it does not cover generalized anxiety disorder. so "is sertraline approved for anxiety" has no single yes/no answer: for GAD specifically, prescribing sertraline is off-label, even though it is a very common prescription. escitalopram, duloxetine, paroxetine and extended-release venlafaxine all do carry a GAD indication.
is trazodone FDA-approved for insomnia?
no. trazodone’s FDA label lists major depressive disorder and nothing else. it is nonetheless the most commonly prescribed antidepressant for insomnia in the US, at doses far below the antidepressant range. the evidence for that use is real but modest: a Cochrane review of 3 trials in 370 people found a moderate short-term improvement in how people rated their sleep, while the trials that used sleep-lab measurement found little or no difference in sleep efficiency.
what is lamotrigine FDA-approved for?
maintenance treatment of bipolar I disorder — delaying the time to the next mood episode after acute stabilisation — plus three epilepsy indications. it is not approved for acute bipolar depression or acute mania. that is the distinction most summaries blur: lamotrigine is a prevention drug on its label, not a rescue drug.
is quetiapine approved for anxiety or sleep?
neither. quetiapine’s label covers schizophrenia, bipolar mania, bipolar depression, bipolar maintenance, and adjunctive treatment of major depression (XR only). for generalized anxiety disorder the Cochrane evidence is actually reasonably good — about twice the response rate of placebo across 4 trials in 2,265 people — but the trade is more weight gain, sedation and dropout, and the use is off-label. for insomnia, the evidence is close to nonexistent: a systematic review found exactly one randomised trial in primary insomnia, with 13 patients, and it showed nothing.
is off-label prescribing legal?
yes. FDA regulates what a manufacturer may market a drug for, not what a licensed clinician may prescribe it for. once a drug is approved for anything, a prescriber may use it for any condition in their clinical judgement. off-label does not mean unstudied — buspirone augmentation for depression was tested in a large government-funded trial — and it does not mean unreasonable. it means the FDA has not reviewed the evidence for that specific use, so the quality of that evidence varies from excellent to nearly absent.
why does the formulation matter?
because indications attach to a specific product, not to a molecule. three of venlafaxine’s four indications belong to the extended-release form only. the narcolepsy indication sits on immediate-release amphetamine tablets while Adderall XR and Mydayis carry ADHD only, and some extended-release methylphenidate brands including Concerta carry ADHD only. the adjunctive-depression indication belongs to Seroquel XR, not immediate-release Seroquel. "the drug is approved for that" is often true of one product and false of the one in your hand.
what should I ask my prescriber about an off-label prescription?
three things. first: is this use on the label, or off-label? second: what is the evidence for this specific use — a large trial, a small one, or clinical experience? third: what would tell us in six to eight weeks that it is working, and what is the plan if it is not? none of those questions challenge the prescription; they just make the reasoning visible.
sources
- FDA structured product labels (DailyMed) for sertraline, escitalopram, fluoxetine, paroxetine, venlafaxine, duloxetine, bupropion, mirtazapine and trazodone — indications sections read individually on 2026-08-17. Where a specific label setid was recorded during verification it is cited and linked in that drug’s own source entry; these nine were read without a recorded setid, so they are cited as label reads rather than linked to a document we did not confirm.
- Everitt H, et al. Antidepressants for insomnia in adults. Cochrane Database Syst Rev 2018;CD010753. PMID 29761479. 3 trials, 370 participants: low-dose trazodone gave a moderate short-term improvement in self-rated sleep (SMD −0.34, 95% CI −0.66 to −0.02, moderate certainty), while 2 trials with sleep-lab measurement found little or no difference in sleep efficiency. Funded by the NIHR School of Primary Care Research and the University of Southampton. Accessed 2026-08-17. https://doi.org/10.1002/14651858.CD010753.pub2 read our summary of this study →
- Kokkali M, et al. Effects of trazodone on sleep: a systematic review and meta-analysis. CNS Drugs 2024. PMID 39123094. 44 trials, 3,935 participants, mean dose ≈179 mg/day: improved rated sleep quality and ≈28 minutes of measured sleep vs placebo, no change in perceived sleep duration, and more than twice as many people stopped for side effects (RR 2.30). Funding statement paywalled — recorded UNKNOWN. Accessed 2026-08-17. https://doi.org/10.1007/s40263-024-01110-2 read our summary of this study →
- FDA structured product labels (DailyMed) for Seroquel immediate-release, setid 0584dda8-bc3c-48fe-1a90-79608f78e8a0, and Seroquel XR, setid 473a3ac4-67f4-4782-baa9-7f9bdd8761f4 — both fetched 2026-08-17. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=0584dda8-bc3c-48fe-1a90-79608f78e8a0
- Depping AM, et al. Second-generation antipsychotics for anxiety disorders. Cochrane Database Syst Rev 2010;CD008120. PMID 21154392. 11 trials, 4,144 participants: people with GAD were about twice as likely to respond to quetiapine as to placebo (OR 2.21 across 4 trials, n=2,265), with more weight gain, sedation and dropout from side effects — and this use is not FDA-approved. Accessed 2026-08-17. https://doi.org/10.1002/14651858.CD008120.pub2 read our summary of this study →
- Thompson W, et al. Atypical antipsychotics for insomnia: a systematic review. Sleep Med 2016. PMID 27544830. Despite quetiapine’s wide off-label use as a sleep aid, this review found only one randomised trial ever conducted in primary insomnia — 13 patients — showing no significant improvement in sleep time, latency or satisfaction versus placebo. Accessed 2026-08-17. https://doi.org/10.1016/j.sleep.2016.04.003 read our summary of this study →
- FDA structured product label (DailyMed) for ABILIFY (aripiprazole), Otsuka/Bristol-Myers Squibb, label revision 2016-08-06 — indications section fetched 2026-08-17.
- FDA structured product label (DailyMed) for LAMICTAL (lamotrigine), setid d7e3572d-56fe-4727-2bb4-013ccca22678 — fetched 2026-08-17. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=d7e3572d-56fe-4727-2bb4-013ccca22678
- FDA structured product label (DailyMed) for lithium carbonate extended-release, setid 611b0ea2-6378-4c8f-953d-87f6b35b0634 — fetched 2026-08-17. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=611b0ea2-6378-4c8f-953d-87f6b35b0634
- FDA structured product label (DailyMed) for alprazolam tablets, setid 0cf8b567-201b-44fa-8fd3-c74023aff44f — fetched 2026-08-17: "acute treatment of generalized anxiety disorder (GAD) in adults; treatment of panic disorder (PD), with or without agoraphobia in adults." https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=0cf8b567-201b-44fa-8fd3-c74023aff44f
- FDA structured product label (DailyMed) for lorazepam tablets, ANI Pharmaceuticals, setid 787240e9-67b5-4af3-b16d-7cf314e3f2e3 — fetched 2026-08-17: "management of anxiety disorders or for the short-term relief of the symptoms of anxiety or anxiety associated with depressive symptoms," with the label noting effectiveness beyond 4 months has not been systematically established. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=787240e9-67b5-4af3-b16d-7cf314e3f2e3
- FDA structured product label (DailyMed) for clonazepam — indications section read 2026-08-17.
- Trivedi MH, et al. Medication augmentation after the failure of SSRIs for depression. N Engl J Med 2006 (STAR*D). PMID 16554526. 565 adults whose SSRI had not worked were randomised to add buspirone or bupropion; about 3 in 10 in each group reached remission, with buspirone slightly less well tolerated. NIMH contract N01MH90003; NCT00021528. This trial tested buspirone as an add-on for depression — an off-label use. Accessed 2026-08-17. https://doi.org/10.1056/NEJMoa052964 read our summary of this study →
- FDA structured product label (DailyMed) for hydroxyzine — indications section read 2026-08-17.
- FDA structured product label (DailyMed) for AMBIEN (zolpidem), setid c36cadf4-65a4-4466-b409-c82020b42452 — fetched 2026-08-17, carrying the boxed warning on complex sleep behaviours and the 5 mg initial dose for women. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=c36cadf4-65a4-4466-b409-c82020b42452
- FDA drug safety communication — boxed warning added for the risk of serious injuries caused by sleepwalking with certain prescription insomnia medicines, april 2019. Accessed 2026-08-17. https://www.fda.gov/drugs/drug-safety-and-availability/fda-adds-boxed-warning-risk-serious-injuries-caused-sleepwalking-certain-prescription-insomnia
- FDA structured product label (DailyMed) for Adderall tablets (Teva), setid f22635fe-821d-4cde-aa12-419f8b53db81 — fetched 2026-08-17. Adderall XR and Mydayis carry ADHD only. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=f22635fe-821d-4cde-aa12-419f8b53db81
- FDA structured product label (DailyMed) for Ritalin (methylphenidate, Novartis), setid c0bf0835-6a2f-4067-a158-8b86c4b0668a — fetched 2026-08-17. Some extended-release brands, including Concerta, carry ADHD only. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=c0bf0835-6a2f-4067-a158-8b86c4b0668a
- Cipriani A, et al. Comparative efficacy and acceptability of 21 antidepressant drugs for the acute treatment of adults with major depressive disorder: a systematic review and network meta-analysis. Lancet 2018. PMID 29477251. 522 trials, 116,477 adults; publicly funded (NIHR Oxford Health BRC BRC-1215-20005, JSPS 17K19808), PROSPERO CRD42012002291. Accessed 2026-08-17. https://doi.org/10.1016/S0140-6736(17)32802-7
related on resolv
- ketamine vs esketamine: what is actually FDA-approved
- supplements and non-prescription substances: the evidence for 17 of them
- psychedelic therapy FDA status tracker
- depression support — free, anonymous peer support
- insomnia and anxiety — free, anonymous peer support
- who paid for the research behind your medication? we counted all 235 studies
- how long were these drugs actually tested? every trial, by length
- the medication research library — approvals, warnings and the trials behind them
- how we score evidence