Skip to content

Major depressive disorder — overview

TL;DR — MDD is a highly prevalent (US lifetime 20.6%), highly disabling episodic mood disorder whose treatments all work modestly and none reliably: all 21 studied antidepressants beat placebo (ORs 1.37–2.13), but sequenced real-world remission falls from ~37% at step 1 to ~13% at step 4, and the field's most-cited cumulative-remission figure (67%) is under active dispute after a 2023 reanalysis. The last decade delivered the first mechanistically new treatments since the 1980s — esketamine, dextromethorphan-bupropion, accelerated TMS, and (investigationally) psilocybin. The biggest unknowns: what MDD actually is biologically (the serotonin story has collapsed as a sufficient account; GWAS points at 100+ small-effect loci), and how to match the right treatment to the right patient instead of by trial-and-error.

Scope. This page covers unipolar major depressive disorder. Bipolar depression — depressive episodes within bipolar disorder, where treatment logic differs fundamentally — is covered in the bipolar-disorder condition.

Definition and diagnostic anchors

  • MDD is diagnosed clinically (DSM-5 / ICD-11): ≥2 weeks of depressed mood or anhedonia plus associated cognitive/neurovegetative symptoms with functional impairment. There is no confirmatory biomarker.
  • DSM-5 severity and specifier structure (anxious distress, mixed features) was characterized epidemiologically in NESARC-III: the anxious/distressed specifier applied to 74.6% of US cases and mixed features to 15.5%, both marking worse course (Hasin 2018, PMID 29450462).
  • Diagnostic heterogeneity is a core problem: two patients can share a diagnosis with almost no overlapping symptoms, which bedevils trials and biomarker work (see diagnostic criteria and heterogeneity).

Epidemiology and burden

Quantity Estimate Source
US lifetime prevalence (DSM-5 MDD) 20.6% Hasin 2018, PMID 29450462
US 12-month prevalence 10.4% Hasin 2018, PMID 29450462
Global DALYs, all mental disorders (2019) 125.3 million (4.9% of all DALYs); depressive disorders a leading component GBD 2019 Mental Disorders Collaborators 2022, PMID 35026139
Trend 1990–2019 Age-standardized DALY rate essentially flat (1581→1566 per 100,000) PMID 35026139
Minimally adequate treatment, 21 countries Received by a small minority of those with MDD Thornicroft 2016, PMID 27908899
  • Twelve-month MDD odds are higher in women (men OR 0.5), young adults (18–29: OR 3.0), and low-income groups (OR 1.7) (Hasin 2018, PMID 29450462).
  • Mortality: mental disorders carry roughly a doubling of all-cause mortality risk in meta-analysis, with a large potential contribution to global deaths not captured in cause-of-death statistics (Walker 2015, PMID 25671328). GBD 2019 counts almost no direct YLLs for mental disorders — a known accounting gap flagged by the authors themselves (PMID 35026139).

Mechanism sketch

  • The classical monoamine (serotonin) hypothesis is no longer tenable as a sufficient causal account: a 2022 systematic umbrella review found no consistent evidence of lowered serotonin concentration or activity in depression (Moncrieff 2022, PMID 35854107). Antidepressants engaging monoamine transporters still outperform placebo (Cipriani 2018, PMID 29477251) — mechanism of illness and mechanism of treatment need not coincide.
  • Genetics: GWAS meta-analysis of 807,553 individuals identified 102 independent variants, enriched in prefrontal-brain-expressed genes and synaptic pathways — many tiny effects, no "depression gene" (Howard 2019, PMID 30718901).
  • The rapid (hours) antidepressant action of the NMDA-receptor antagonist ketamine (Berman 2000, PMID 10686270; Zarate 2006, PMID 16894061) reoriented mechanistic thinking toward glutamatergic signaling and synaptic plasticity.
  • Current working models (stress–HPA axis, neuroplasticity, inflammation, circuit dysfunction) are compared in neurobiology and mechanisms.

Treatment landscape

Antidepressants — efficacy and the effect-size debate

  • What works, on average: 522-trial network meta-analysis (116,477 participants): every one of 21 antidepressants was more effective than placebo for acute adult MDD, ORs from 2.13 (amitriptyline) to 1.37 (reboxetine); only agomelatine and fluoxetine beat placebo on acceptability (Cipriani 2018, PMID 29477251).
  • Why it's disputed: FDA-dataset meta-analysis found drug–placebo differences below conventional clinical-significance criteria except at the most severe baseline, driven by falling placebo response rather than rising drug response (Kirsch 2008, PMID 18303940); a patient-level analysis replicated severity-dependence (Fournier 2010, PMID 20051569). Selective publication inflated apparent effect sizes ~32%: 94% of published trials appeared positive vs 51% by FDA reckoning (Turner 2008, PMID 18199864).
  • Continuation matters: continuing antidepressants after response cuts relapse odds ~70% (average relapse 41% on placebo vs 18% on drug) (Geddes 2003, PMID 12606176).

Sequenced treatment — STAR*D and its contested legacy

  • STAR*D (N=3,671 entering step 1): QIDS-SR remission 36.8% (step 1), 30.6% (step 2), 13.7% (step 3), 13.0% (step 4); theoretical cumulative remission 67%; relapse rates rose with the number of steps required (Rush 2006, PMID 17074942).
  • A 2023 reanalysis of the patient-level data applying the original protocol's prespecified criteria reported substantially different (lower) remission/response figures, reopening debate about the trial's headline claims (Pigott 2023, PMID 37491091). Unresolved — both readings currently circulate in the literature.

Rapid-acting and novel agents

  • Intranasal esketamine + newly initiated oral antidepressant beat antidepressant + placebo spray in treatment-resistant depression at day 28 (MADRS difference −4.0 points), with dissociation/nausea/vertigo as characteristic adverse events (Popova 2019, PMID 31109201); FDA approval followed in 2019.
  • Dextromethorphan–bupropion (AXS-05) showed phase-3 efficacy in MDD (GEMINI) (Iosifescu 2022, PMID 35649167); FDA-approved 2022.
  • Psilocybin (2×25 mg) vs 6 weeks of escitalopram: no significant difference on the primary endpoint; secondary outcomes favored psilocybin but uncorrected for multiplicity (Carhart-Harris 2021, PMID 33852780). Blinding integrity is a live methodological objection.

Neuromodulation

  • ECT remains the most effective acute short-term treatment on meta-analytic evidence: real vs sham SES −0.91; ECT vs pharmacotherapy SES −0.80 (UK ECT Review Group 2003, PMID 12642045).
  • Accelerated intermittent theta-burst stimulation (SAINT/SNT): high open-label remission in treatment-resistant patients (Cole 2020, PMID 32252538), followed by a positive double-blind randomized controlled trial (Cole 2021, PMID 34711062). Small samples; durability unknown.

Psychotherapy, combination, and measurement-based care

  • Network meta-analysis (101 trials): psychotherapy ≈ pharmacotherapy head-to-head; the combination beats either alone (RR ~1.25–1.27 for response) and is better accepted than drugs alone (Cuijpers 2020, PMID 31922679).
  • Measurement-based care — systematic rating-scale-guided dose/switch decisions — outperformed standard care in a randomized trial with blind raters (Guo 2015, PMID 26315978), and systematic, personalized management is a central emphasis of the CANMAT 2023 guideline update, which organizes treatment by lines with patient-partner input (Lam 2024, PMID 38711351).

Frontier glimpse

Precision-psychiatry stratification (whom to give esketamine vs TMS vs combination first), inflammation-defined subgroups, and the collision between rising placebo response and drug development are the live fronts; the SAINT-style compressed-course TMS model and psychedelic trials are the most-watched near-term developments (Cole 2021, PMID 34711062; Carhart-Harris 2021, PMID 33852780). The active pipeline is mapped in clinical trials landscape.

Quantitative orientation beyond the headline estimates

The headline prevalence, GWAS, and acute-treatment estimates describe different denominators and should not be collapsed into one model of “depression.” Screening accuracy, episode course, treatment availability, and excess mortality each change what the observed burden means.

Domain Quantitative anchor Interpretation
Screening At PHQ-9 ≥10, pooled sensitivity was 0.88 (95% CI 0.83–0.92) and specificity 0.85 (0.82–0.88) against semistructured interviews in 17,357 participants A positive screen is not a diagnosis; predictive value changes with prevalence (Levis 2019, PMID 30967483)
Familial liability First-degree relatives had 2.84-fold odds (95% CI 2.31–3.49); twin-model heritability was 37% (31–42%) Genetic influence is material but leaves most liability variance outside additive inherited effects (Sullivan 2000, PMID 11007705)
Molecular scale A trans-ancestry GWAS of 688,808 cases and 4,364,225 controls found 697 associations at 635 loci; polygenic scores explained at most 5.8% of liability variance in Europeans Discovery-scale biology is not individual diagnosis or treatment selection (PGC 2025, PMID 39814019)
Inflammation A cumulative meta-analysis found higher IL-6, TNF-α, and CRP on average, with substantial between-study heterogeneity The signal supports subgroup research, not a universal inflammatory disease model (Haapakoski 2015, PMID 26065825)
Continuation Across 40 discontinuation trials, continued antidepressant treatment reduced relapse compared with placebo substitution Enriched responder samples and withdrawal after substitution limit transportability (Kato 2021, PMID 32704061)
Global shock The COVID-19 burden analysis estimated substantial 2020 increases in depressive and anxiety disorders, with larger effects in women and younger groups Prevalence and burden respond to social conditions; they are not fixed biological constants (COVID-19 Mental Disorders Collaborators 2021, PMID 34634250)
Scalable psychotherapy Individual-participant evidence from LMICs found task-shared psychological interventions reduced depressive symptoms Delivery model is part of efficacy when specialists are scarce (Karyotaki 2022, PMID 35319740)
Mortality A 2025 large-scale synthesis found elevated all-cause and multiple cause-specific mortality in people with depression Suicide is only one pathway; cardiometabolic disease, substance use, and inequitable care contribute (Chan 2025, PMID 40948054)

Four distinctions that prevent overclaiming

  1. Syndrome validity is not biomarker validity. A reproducible clinical syndrome can be useful even when no laboratory test identifies it; conversely, a group biomarker difference can be real without being diagnostically discriminative.

  2. Relative efficacy is not absolute recovery. Odds ratios from acute trials do not state how many people regain function, how durable remission is, or how many discontinue because treatment is unacceptable.

  3. Treatment mechanism is not disease cause. Monoamine transporter blockade, NMDA modulation, psychotherapy, and stimulation can all alter a final common state without implying that untreated MDD has one corresponding deficiency.

  4. Population burden is not service reach. Effective interventions do not reduce disability where diagnosis, continuity, affordability, language access, or trained staff are absent; task-sharing evidence therefore belongs in the central treatment map (Karyotaki 2022, PMID 35319740).

Controversies carried through the wiki

Controversy Evidence supporting the first position Evidence constraining it
Antidepressants are clinically important vs statistically superior but often modest All studied drugs beat placebo in the large network meta-analysis (Cipriani 2018, PMID 29477251) Severity analyses and unpublished FDA data reduce estimated average benefit (Kirsch 2008, PMID 18303940; Turner 2008, PMID 18199864)
Depression has biological signatures vs no clinically useful biological test Large GWAS and immune meta-analyses identify reproducible group signals (PGC 2025, PMID 39814019; Haapakoski 2015, PMID 26065825) Explained variance and discrimination remain too small for diagnosis or routine selection
Sequential care yields high cumulative remission vs attrition makes that estimate misleading Original STAR*D analysis reported theoretical 67% cumulative remission (Rush 2006, PMID 17074942) Protocol-faithful reanalysis reports materially lower rates (Pigott 2023, PMID 37491091)
Treatment access is mainly a discovery problem vs a delivery problem Novel agents and compressed stimulation expand options Task-shared psychotherapy already has randomized evidence while coverage remains low (Karyotaki 2022, PMID 35319740; Thornicroft 2017, PMID 27908899)

Open questions

  • What is the true cumulative remission rate under sequenced antidepressant treatment — the 67% of the original STAR*D report or the substantially lower protocol-faithful estimate? (Rush 2006, PMID 17074942; Pigott 2023, PMID 37491091)
  • Are average antidepressant effects clinically meaningful across severity strata, or concentrated in severe depression via placebo-response decay? (Kirsch 2008, PMID 18303940; Fournier 2010, PMID 20051569; Cipriani 2018, PMID 29477251)
  • If serotonin deficiency is not the mechanism (Moncrieff 2022, PMID 35854107), what explains the efficacy of monoaminergic drugs — and can the glutamatergic rapid-acting pathway (Zarate 2006, PMID 16894061) be separated from dissociative/abuse liability?
  • Do accelerated TMS remission rates survive large multi-site replication and long-term follow-up? (Cole 2021, PMID 34711062)
  • Can the global treatment gap be closed when even high-income countries deliver minimally adequate treatment to a minority? (Thornicroft 2016, PMID 27908899)

References

  1. Rush AJ, et al. Acute and longer-term outcomes in depressed outpatients requiring one or several treatment steps: a STAR*D report. Am J Psychiatry. 2006;163(11):1905ff. PMID 17074942
  2. Pigott HE, et al. What are the treatment remission, response and extent of improvement rates after up to four trials of antidepressant therapies in real-world depressed patients? A reanalysis of the STAR*D study's patient-level data with fidelity to the original research protocol. BMJ Open. 2023;e063095. PMID 37491091
  3. 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;391:1357-1366. PMID 29477251
  4. Kirsch I, et al. Initial severity and antidepressant benefits: a meta-analysis of data submitted to the Food and Drug Administration. PLoS Med. 2008;5:e45. PMID 18303940
  5. Fournier JC, et al. Antidepressant drug effects and depression severity: a patient-level meta-analysis. JAMA. 2010. PMID 20051569
  6. Turner EH, et al. Selective publication of antidepressant trials and its influence on apparent efficacy. N Engl J Med. 2008;358:252-60. PMID 18199864
  7. Geddes JR, et al. Relapse prevention with antidepressant drug treatment in depressive disorders: a systematic review. Lancet. 2003;361:653-61. PMID 12606176
  8. Berman RM, et al. Antidepressant effects of ketamine in depressed patients. Biol Psychiatry. 2000. PMID 10686270
  9. Zarate CA Jr, et al. A randomized trial of an N-methyl-D-aspartate antagonist in treatment-resistant major depression. Arch Gen Psychiatry. 2006;63(8). PMID 16894061
  10. Popova V, et al. Efficacy and Safety of Flexibly Dosed Esketamine Nasal Spray Combined With a Newly Initiated Oral Antidepressant in Treatment-Resistant Depression: A Randomized Double-Blind Active-Controlled Study. Am J Psychiatry. 2019;176:428-438. PMID 31109201
  11. Iosifescu DV, et al. Efficacy and Safety of AXS-05 (Dextromethorphan-Bupropion) in Patients With Major Depressive Disorder: A Phase 3 Randomized Clinical Trial (GEMINI). J Clin Psychiatry. 2022;21m14345. PMID 35649167
  12. Carhart-Harris R, et al. Trial of Psilocybin versus Escitalopram for Depression. N Engl J Med. 2021;384:1402-1411. PMID 33852780
  13. Cole EJ, et al. Stanford Accelerated Intelligent Neuromodulation Therapy for Treatment-Resistant Depression. Am J Psychiatry. 2020. PMID 32252538
  14. Cole EJ, et al. Stanford Neuromodulation Therapy (SNT): A Double-Blind Randomized Controlled Trial. Am J Psychiatry. 2021. PMID 34711062
  15. UK ECT Review Group. Efficacy and safety of electroconvulsive therapy in depressive disorders: a systematic review and meta-analysis. Lancet. 2003;361:799-808. PMID 12642045
  16. Guo T, et al. Measurement-Based Care Versus Standard Care for Major Depression: A Randomized Controlled Trial With Blind Raters. Am J Psychiatry. 2015. PMID 26315978
  17. Cuijpers P, et al. A network meta-analysis of the effects of psychotherapies, pharmacotherapies and their combination in the treatment of adult depression. World Psychiatry. 2020;19:92-107. PMID 31922679
  18. Lam RW, et al. Canadian Network for Mood and Anxiety Treatments (CANMAT) 2023 Update on Clinical Guidelines for Management of Major Depressive Disorder in Adults. Can J Psychiatry. 2024;69:641-687. PMID 38711351
  19. GBD 2019 Mental Disorders Collaborators. Global, regional, and national burden of 12 mental disorders in 204 countries and territories, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet Psychiatry. 2022;9:137-150. PMID 35026139
  20. Hasin DS, et al. Epidemiology of Adult DSM-5 Major Depressive Disorder and Its Specifiers in the United States. JAMA Psychiatry. 2018;75:336-346. PMID 29450462
  21. Thornicroft G, et al. Undertreatment of people with major depressive disorder in 21 countries. Br J Psychiatry. 2016/2017. PMID 27908899
  22. Howard DM, et al. Genome-wide meta-analysis of depression identifies 102 independent variants and highlights the importance of the prefrontal brain regions. Nat Neurosci. 2019. PMID 30718901
  23. Moncrieff J, et al. The serotonin theory of depression: a systematic umbrella review of the evidence. Mol Psychiatry. 2022/2023. PMID 35854107
  24. Walker ER, et al. Mortality in mental disorders and global disease burden implications: a systematic review and meta-analysis. JAMA Psychiatry. 2015. PMID 25671328
  25. Major Depressive Disorder Working Group of the Psychiatric Genomics Consortium, et al. Trans-ancestry genome-wide study of depression identifies 697 associations implicating cell types and pharmacotherapies. Cell. 2025;188:640-652.e9. PMID 39814019
  26. COVID-19 Mental Disorders Collaborators. Global prevalence and burden of depressive and anxiety disorders in 2020 due to the COVID-19 pandemic. Lancet. 2021;398:1700-1712. PMID 34634250
  27. Levis B, et al. Accuracy of Patient Health Questionnaire-9 for screening to detect major depression. BMJ. 2019;365:l1476. PMID 30967483
  28. Sullivan PF, et al. Genetic epidemiology of major depression: review and meta-analysis. American Journal of Psychiatry. 2000;157:1552-1562. PMID 11007705
  29. Haapakoski R, et al. Cumulative meta-analysis of IL-6, IL-1β, TNF-α and CRP in major depressive disorder. Brain Behavior and Immunity. 2015;49:206-215. PMID 26065825
  30. Kato M, et al. Discontinuation of antidepressants after remission in MDD: systematic review and meta-analysis. Molecular Psychiatry. 2021;26:118-133. PMID 32704061
  31. Karyotaki E, et al. Task-shared psychological interventions and depression outcomes in LMICs. JAMA Psychiatry. 2022;79:430-443. PMID 35319740
  32. Chan JKN, et al. All-cause and cause-specific mortality in people with depression. World Psychiatry. 2025;24:404-421. PMID 40948054