Antidepressant comparison

Class-by-class comparison

The table below lists the antidepressant classes used most often in adult outpatient practice. Examples use international nonproprietary names (INN), the generic chemical name shared across countries. Brand names vary by country and are not included to keep the table portable. Mechanism descriptions are simplified; the relationship between receptor pharmacology and clinical effect is more nuanced than the early "chemical imbalance" framing suggested. Withdrawal severity refers to the typical intensity of discontinuation symptoms on abrupt stop or rapid taper after several weeks of treatment, and is driven largely by half-life.

Common dose ranges and pediatric considerations are not listed here; prescribing details belong with a clinician and the FDA label for each medication.

How clinicians actually pick one

The decision is rarely about which medication is "best." It is about matching side-effect profile, prior response, other medical conditions, and life context to the person. Most experienced prescribers move through a short mental checklist within the first visit.

1. Has this person taken an antidepressant before? A medication that worked in the past, without intolerable side effects, is usually the right first choice again. A medication that failed (after an adequate dose for at least six to eight weeks) is usually skipped.
2. What does the symptom picture look like? Anxious depression often favors an SSRI (escitalopram, sertraline) or an SNRI. Low energy, hypersomnia, and low motivation often favor bupropion. Severe insomnia or weight loss often favors mirtazapine. Cognitive symptoms (sometimes called brain fog) sometimes favor vortioxetine.
3. What else is on the medication list? SSRIs and SNRIs interact with tramadol, triptans, MAOIs, linezolid, and other serotonergic agents. Citalopram has a dose ceiling on QT-prolonging combinations. Bupropion is avoided in seizure history or active eating disorder.
4. What are the other medical conditions? Duloxetine has FDA approvals for fibromyalgia, diabetic peripheral neuropathy, and chronic musculoskeletal pain. Bupropion helps smoking cessation. Mirtazapine is often chosen in older, frail patients with poor appetite. Tricyclics are avoided in people with cardiac conduction disease.
5. Which side effects are deal-breakers? Sexual side effects, weight gain, sedation, and insomnia are the four most common reasons patients stop antidepressants. Asking up front saves a wasted trial.
6. What did a first-degree relative respond to? Family response is a reasonable tiebreaker among first-line options. It is not strong evidence on its own, but it is real-world information that often holds up.

How long to give a medication

An adequate trial is six to eight weeks at a therapeutic dose. Some change is usually visible by week two to four (sleep, appetite, anxiety often shift first). Mood lift often lags by another two to four weeks. The full benefit may not appear until week eight to twelve.

About one in three people reach full remission on the first medication tried. Roughly half show a meaningful response (a 50 percent or greater symptom drop on the PHQ-9). Most of the remaining people respond to a second medication or to augmentation. The STAR*D trial mapped this out: by the fourth treatment step, about 67 percent of patients had reached remission, though reanalyses with stricter outcome criteria report lower numbers (Pigott, 2010). The practical takeaway is that the plan is built around iteration, not a single attempt.

Once remission is reached, the standard continuation phase is at least six to twelve months for a first episode, with longer maintenance after two or more episodes. See stopping antidepressants for tapering.

Augmentation versus switching

When a medication produces a partial response (improvement, but not full remission), the next decision is whether to add a second agent (augmentation) or to change the first agent (switching). The choice is individual. The patterns below summarize how most outpatient prescribers think about it.

When switching tends to be preferred

• The first medication produced little or no benefit after six to eight weeks at a therapeutic dose.
• The first medication is causing intolerable side effects that augmentation will not solve (sexual side effects, weight gain, sedation, GI intolerance).
• The diagnosis is being reconsidered (for example, possible bipolar depression, anxiety disorder, or substance use as a major driver).
• The first medication is in the same class as a prior failed agent.

When augmentation tends to be preferred

• There has been a partial response: some symptoms are clearly better, others are not.
• Side effects of the first medication are acceptable.
• A targeted second agent fits the residual symptom (bupropion added for low energy or SSRI-induced sexual side effects, mirtazapine added for insomnia, an atypical antipsychotic added for prominent agitation or psychotic features).
• The patient prefers not to lose the gains of the first medication.

The most evidence-supported augmentation strategies are lithium and triiodothyronine (T3), both used off-label, and the FDA-approved adjuncts aripiprazole, brexpiprazole, cariprazine, quetiapine extended-release, and the olanzapine and fluoxetine combination. Atypical antipsychotic adjuncts add metabolic monitoring (weight, glucose, lipids) and the risk of movement side effects. Lithium adds blood level, kidney, and thyroid monitoring. Augmentation decisions involve trade-offs in side-effect profile and are best made with a prescriber who knows the full picture.

FDA boxed warning

This warning applies across the class. It is not specific to any one medication. The clinical implication is close follow-up in the first weeks and after any dose change, with same-day contact for any new or worsening suicidal thoughts. Untreated depression also carries meaningful suicide risk; the warning is a basis for monitoring, not a reason to avoid treatment.

If you may be in danger, call or text 988 in the United States, call 911, or go to the nearest emergency department. See crisis resources.

Drug interactions worth knowing

• Serotonin syndrome. Combining serotonergic drugs (SSRIs, SNRIs, tramadol, triptans, MAOIs, linezolid, methylene blue, some recreational drugs including MDMA) can produce serotonin syndrome. Most cases are mild. Severe cases involve high fever, agitation, tremor, clonus, and autonomic instability, and are a medical emergency.
• QT prolongation. Citalopram has a maximum dose of 40 mg in adults and 20 mg in older adults or in patients on other QT-prolonging medications. Escitalopram and sertraline have minimal QT effect at standard doses.
• Bleeding risk. SSRIs and SNRIs modestly increase bleeding risk, especially with NSAIDs, aspirin, or anticoagulants. Worth flagging if you take daily aspirin or a blood thinner.
• MAOI rules. A strict low-tyramine diet and a two-week medication washout (five weeks for fluoxetine) are required when switching to or from most other antidepressants. Hypertensive crisis can be fatal if these rules are not followed.
• Bupropion seizure risk. Avoid in seizure disorder, current eating disorder, or abrupt alcohol or benzodiazepine withdrawal. Total daily dose limits keep risk low in standard use.
• CYP2D6 and CYP3A4 inhibitors. Fluoxetine, paroxetine, and bupropion are strong CYP2D6 inhibitors and can raise levels of tamoxifen, opioids metabolized by 2D6 (codeine, tramadol), and several antiarrhythmics. Fluvoxamine is a strong CYP1A2 inhibitor and affects clozapine, theophylline, and caffeine. The prescriber checks for these.

What about generics?

Generic antidepressants are equivalent to their brand-name counterparts in clinical use. The FDA requires bioequivalence within a tight range. Cost should not steer the choice between brand and generic. The choice between specific molecules (sertraline vs. escitalopram, for example) is driven by fit, not by brand.

What this page is not

• It is not a prescription, a recommendation, or a substitute for a clinical evaluation. Specific medication decisions belong with a prescriber who has reviewed your history, current medications, other diagnoses, and goals.
• It is not a complete drug reference. Dose ranges, pediatric considerations, pregnancy and lactation guidance, hepatic and renal dosing, and full interaction lists are not included. The FDA label and a clinician are the reference for those.
• It is not a comparison of efficacy by name. Cipriani 2018 found small differences between agents, and most of those differences disappear when fit, tolerability, and prior response are taken into account.
• It is not a guide to stopping. See stopping antidepressants for tapering, discontinuation symptoms, and the FINISH mnemonic.