Depression Drugs: Medications That Cause or Treat It

First-Line Antidepressants: SSRIs and SNRIs

Selective serotonin reuptake inhibitors are the standard initial pharmacotherapy for moderate-to-severe major depressive disorder. The APA 2023 Clinical Practice Guideline recommends SSRIs or SNRIs as monotherapy for most adults, given their favorable risk-benefit ratio compared to older tricyclic agents.

Six SSRIs are FDA-approved for MDD: fluoxetine (20-80 mg/day), sertraline (50-200 mg/day), paroxetine (20-50 mg/day), citalopram (20-40 mg/day), escitalopram (10-20 mg/day), and fluvoxamine (used more often for OCD). Escitalopram demonstrated a small but statistically significant advantage over other SSRIs in the CINeMA network meta-analysis (Cipriani et al., 2018; N=116,477) for both efficacy and tolerability.

SNRIs add norepinephrine reuptake inhibition. Venlafaxine (75-225 mg/day) and duloxetine (60-120 mg/day) show slightly higher remission rates than SSRIs in some head-to-head comparisons, though the clinical difference is modest. Duloxetine carries a dual indication for depression and chronic pain conditions, making it a logical choice when both coexist [1].

Common side effects across both classes include nausea in the first 1-2 weeks, sexual dysfunction in 30-40% of patients, weight gain (particularly with paroxetine), and sleep disruption. These effects drive discontinuation more than lack of efficacy does.

The STAR*D Lesson: Why Switching and Augmentation Matter

The Sequenced Treatment Alternatives to Relieve Depression (STAR*D) trial remains the largest real-world antidepressant effectiveness study conducted. Funded by NIMH, it enrolled 4,041 outpatients with nonpsychotic MDD across 41 clinical sites.

Results were sobering. Only 36.8% achieved remission with citalopram monotherapy at step 1. Cumulative remission reached approximately 67% after all four treatment levels, but each subsequent step showed declining response rates and rising dropout. The trial established that treatment resistance is common, not exceptional.

"After two adequate antidepressant trials, the likelihood of remission with a third agent drops below 15%," noted Dr. A. John Rush, STAR*D principal investigator, in the study's summary publication.

Clinical takeaway: providers should reassess response at 4-6 weeks, switch medication class if response is absent, and consider augmentation strategies for partial responders rather than simply increasing the dose of a poorly tolerated first agent [2].

Bupropion: The Activating Alternative

Bupropion (Wellbutrin) operates through norepinephrine-dopamine reuptake inhibition with no serotonergic activity. This distinct mechanism produces a side-effect profile that differs markedly from SSRIs. Sexual dysfunction rates match placebo. Weight gain does not occur; many patients lose 1-3 kg over the first 12 weeks [3].

The drug comes in three formulations: immediate-release (150-450 mg/day in divided doses), sustained-release (150-400 mg/day in twice-daily dosing), and extended-release (150-450 mg/day once daily). Seizure risk is dose-dependent and clinically relevant only above 450 mg/day or in patients with eating disorders, alcohol withdrawal, or prior seizure history.

Bupropion is particularly suited for patients with fatigue-predominant depression, hypersomnia, weight concerns, or SSRI-related sexual dysfunction. It is also FDA-approved for smoking cessation (as Zyban), offering dual benefit in depressed patients attempting to quit tobacco. The drug is less effective for anxiety-predominant presentations and may worsen insomnia if dosed in the evening.

Medications That Can Cause Depression

A 2018 JAMA study (Qato et al.; N=26,192) found that 37.2% of U.S. adults use at least one prescription medication listing depression as a potential adverse effect. Polypharmacy with three or more such agents was associated with a 15% prevalence of concurrent depression, compared to 5% in those not taking any depressogenic drugs.

The highest-risk medication classes include:

Corticosteroids. Prednisone at doses above 40 mg/day produces mood disturbance in up to 30% of patients. The Boston Collaborative Drug Surveillance Program reported psychiatric symptoms in 1.3% at <40 mg/day versus 18.4% at doses exceeding 80 mg/day. Both mania and depression occur, often within the first two weeks of initiation.

Beta-blockers. Propranolol crosses the blood-brain barrier readily and has the strongest historical association with depressive symptoms among beta-blockers. However, a 2021 Hypertension meta-analysis challenged this assumption, finding no statistically significant increase in depression with beta-blockers versus placebo (OR 1.02; 95% CI 0.87-1.20). The clinical signal may be fatigue misattributed as depression.

Isotretinoin (Accutane). The FDA added a black-box warning for suicidality in 1998, though subsequent large cohort studies have not confirmed a causal relationship after controlling for baseline acne severity and psychosocial burden. Monitoring remains standard practice.

Interferons. Interferon-alpha for hepatitis C produced major depression in 20-30% of treated patients in the pre-DAA era. Prophylactic SSRI use reduced incidence to approximately 10% [4].

Hormonal contraceptives. A Danish nationwide cohort study (Skovlund et al., 2016; N=1,061,997) found that combined oral contraceptive users had a relative risk of 1.23 for first antidepressant use compared to non-users. Adolescents showed higher vulnerability (RR 1.8 for combined pills).

Newer Agents: Ketamine, Esketamine, and Rapid-Acting Options

Traditional antidepressants require weeks to produce therapeutic effects. The glutamate hypothesis opened an alternative pathway. Ketamine, an NMDA receptor antagonist, produces antidepressant effects within hours in treatment-resistant patients.

The FDA approved esketamine (Spravato) nasal spray in March 2019 for treatment-resistant depression as an adjunct to oral antidepressants. The SUSTAIN-1 trial demonstrated that continued esketamine plus oral antidepressant reduced relapse risk by 51% compared to antidepressant plus placebo nasal spray over 16 weeks [5].

Dosing is 56 mg or 84 mg intranasally, administered twice weekly for the first month, then weekly or biweekly during maintenance. The drug requires administration in a certified healthcare setting with 2-hour post-dose monitoring due to dissociation and sedation. Cost remains a barrier: approximately $590-$885 per session before insurance.

Brexanolone (Zulresso), a synthetic allopregnanolone, gained approval for postpartum depression in 2019. It requires 60-hour continuous IV infusion in a REMS-certified facility. Zuranolone (Zurzuvae), an oral neuroactive steroid, received FDA approval in August 2023 for postpartum depression at 50 mg nightly for 14 days, representing the first oral rapid-acting antidepressant for any depressive indication [6].

Augmentation Strategies for Partial Responders

When a patient achieves 25-49% symptom reduction on an adequate antidepressant trial (defined as at least 4-6 weeks at therapeutic dose), augmentation is preferred over switching because it preserves partial gains.

The strongest evidence supports three augmentation approaches:

Atypical antipsychotics. Aripiprazole (2-15 mg/day, typically 2-5 mg), quetiapine XR (150-300 mg/day), and brexpiprazole (1-3 mg/day) all carry FDA approval as adjunctive therapy for MDD. The aripiprazole augmentation trials showed a number needed to treat (NNT) of 7-10 for remission. Weight gain and akathisia are the primary limitations.

Lithium. Augmentation with lithium 300-600 mg/day (targeting levels of 0.4-0.8 mEq/L) has meta-analytic support showing a response rate roughly double that of placebo augmentation. Thyroid and renal monitoring are required.

Thyroid hormone. T3 (liothyronine 25-50 mcg/day) augmentation dates to the 1960s. The STAR*D trial used it in step 3, finding comparable efficacy to lithium with better tolerability. It is particularly considered in patients with subclinical hypothyroidism or those on lithium who develop elevated TSH [7].

"Augmentation should be considered first-line for partial responders, reserving full medication switches for non-responders or patients with intolerable side effects," states the 2023 APA Practice Guideline.

Tricyclics, MAOIs, and When They Still Apply

Tricyclic antidepressants (amitriptyline, nortriptyline, imipramine, desipramine) preceded SSRIs by three decades. Their efficacy equals or exceeds SSRIs for severe melancholic depression, but anticholinergic effects, cardiac conduction delays, and lethality in overdose pushed them to second- or third-line status.

Nortriptyline is the best-tolerated TCA, with a defined therapeutic window (50-150 ng/mL serum level). It remains useful in elderly patients with depression and neuropathic pain who cannot tolerate duloxetine. ECG screening is mandatory before initiation.

MAOIs (phenelzine, tranylcypromine, selegiline patch) are the most effective antidepressants for atypical depression characterized by mood reactivity, leaden paralysis, rejection sensitivity, and hypersomnia. A Columbia University meta-analysis showed phenelzine response rates of 71% versus 50% for imipramine in atypical depression.

The tyramine dietary restriction and drug interaction profile limit MAOI prescribing to specialists. The selegiline transdermal patch (EMSAM) at 6 mg/24h bypasses first-pass gut MAO inhibition and does not require dietary modification at that dose, though higher doses (9-12 mg/24h) still carry restrictions [8].

Drug-Induced Depression: Recognition and Management

Identifying medication-induced depression requires temporal correlation. Key clinical clues: depressive symptoms appearing within days to weeks of starting a new medication, absence of prior psychiatric history, improvement after dose reduction or discontinuation, and recurrence upon rechallenge.

The management algorithm follows a stepwise approach. First, confirm the temporal relationship. Second, evaluate whether the offending agent can be discontinued or substituted. Beta-blockers can often be switched to a cardioselective agent (metoprolol, bisoprolol) with less CNS penetration. Corticosteroids may be tapered or replaced with steroid-sparing agents. Hormonal contraceptives can be changed to non-hormonal alternatives or progestin-only formulations.

If the offending medication cannot be stopped (for example, interferon in active hepatitis treatment, or chronic immunosuppression post-transplant), concurrent antidepressant therapy is appropriate. SSRIs are first-line in this scenario, with sertraline preferred for cardiac patients given its SADHART trial safety data in post-MI depression [9].

A complete medication reconciliation should be performed in every patient presenting with new depressive symptoms. The 2018 Qato study found that patients rarely associate non-psychiatric medications with mood changes unless directly asked.

Pharmacogenomics and Personalized Antidepressant Selection

CYP2D6 and CYP2C19 polymorphisms affect metabolism of most SSRIs, SNRIs, and tricyclics. Poor metabolizers at CYP2D6 experience higher plasma drug levels and more side effects; ultrarapid metabolizers may fail to reach therapeutic concentrations.

The GUIDED trial (N=1,167) randomized treatment-resistant patients to pharmacogenomic-guided prescribing versus treatment as usual. At week 8, the guided group showed higher response (26% vs. 20%; P=0.013) and remission rates (15% vs. 10%; P=0.007). The effect was most pronounced in patients whose prior medications had gene-drug interactions identified by the panel.

The Clinical Pharmacogenetics Implementation Consortium (CPIC) publishes specific dosing guidelines for SSRIs based on CYP2D6/CYP2C19 status. For CYP2C19 poor metabolizers, escitalopram and citalopram require 50% dose reductions. For CYP2D6 poor metabolizers, paroxetine and fluvoxamine alternatives are recommended.

Cost for commercial pharmacogenomic panels ranges from $300-$500 with variable insurance coverage. Medicare covers testing when supported by clinical need. The best candidates for testing are patients who have failed two or more adequate antidepressant trials or experienced severe adverse effects at standard doses [10].

Antidepressant Discontinuation: Timing and Tapering

The APA recommends continuation therapy for 4-9 months after achieving remission to prevent relapse. Patients with three or more lifetime episodes, residual symptoms, or psychosocial stressors benefit from indefinite maintenance therapy.

Discontinuation syndrome occurs in approximately 40-50% of patients who stop SSRIs or SNRIs abruptly, based on a systematic review by Davies and Read (2019). Symptoms include dizziness, electric shock sensations ("brain zaps"), irritability, nausea, insomnia, and flu-like malaise. Paroxetine and venlafaxine carry the highest discontinuation risk due to short half-lives.

Tapering schedules should be individualized. A general approach: reduce dose by 25% every 2-4 weeks for patients on treatment <12 months, with slower tapers (10% reductions every 4-6 weeks or longer) for patients on multi-year therapy. Fluoxetine's 4-6 day active metabolite half-life makes it the exception; many patients can stop without a prolonged taper [11].

For patients struggling with the final dose reduction, liquid formulations allow precise micro-tapering. Switching from a short-half-life SSRI to fluoxetine before discontinuation is another evidence-based strategy for difficult tapers.