Abstract

A depressive disorder is a mood disorder marked by persistent low mood or loss of interest or pleasure, and the term names a family of diagnoses rather than one illness. DSM-5 and ICD-11 organize the class in substantially similar ways, and whether depression is one condition or a dimension remains disputed, with taxometric evidence leaning dimensional and no data-driven subtype replicated. Risk is spread across many genetic variants of small effect, early adversity, neuroticism and stress; depressive disorders are common, a leading cause of disability burden, and carry a raised risk of suicide that remains hard to predict. Negative processing biases and executive impairment are well documented, and medication, psychotherapy, brain stimulation and prevention in high-risk groups show benefit while mechanisms remain open. Most of the evidence comes from clinical, Western and major depressive disorder samples.

Keywords: depressive disorder, unipolar depression, classification, heterogeneity, mood disorders

A depressive disorder is a mood disorder in which low mood, or lost interest and pleasure in usual activities, is both prominent and lasting (MeSH, D003866). The label covers a family of diagnoses, from major depressive disorder and persistent depressive disorder to categories tied to the menstrual cycle, to childhood irritability and to substance use, and the older names unipolar depression, endogenous depression and melancholia survive among its MeSH entry terms. Major depression, the best-studied member, is common and severely disabling, and its presentation, course and response to treatment all vary from person to person (Malhi & Mann, 2018). No established mechanism explains all aspects of it, and its heritability is estimated at about 35% (Otte et al., 2016). This page covers depressive disorders as a class of diagnoses; low mood as a symptom and depression in general are covered on the depression page, and the most common single diagnosis is covered on the major depressive disorder page.

The article first defines the class and traces the history of the concept, from melancholia to explicit diagnostic criteria. It then sets out how the two current manuals classify the depressive disorders, how symptoms are diagnosed and measured, and what qualitative research reports about living through depression. Each type, category and specifier follows with its own cognitive findings, and the evidence on whether depression is one disorder or many is weighed before selected disorders are compared side by side. Later sections cover prevalence, course and burden; depression across the lifespan; the leading cognitive account and its rivals; cognitive processes and the tests of causal claims about them; causes, biology, brain networks and sleep; how laboratory measures relate to everyday life; co-occurring conditions; suicide and self-harm; and culture and society. The treatment sections report the evidence on medication and psychological therapy, how treatments are thought to work, brain stimulation and other biological treatments, emerging treatments and prevention. The page closes with living with a depressive disorder, related conditions and how they differ, a worked example, a discussion and current directions.

Key Takeaways

  • Depressive disorders are a family of diagnoses that share a persistent disturbance of mood or of interest and pleasure. A comparison of DSM-5 and ICD-11 found their organization substantially similar and closer than before, though each system has categories the other lacks; mixed depressive and anxiety disorder, for example, is an ICD-11 category but not a DSM-5 one.
  • Classification moved from cause-based labels such as endogenous and neurotic depression, whose features a factor-analytic review found to vary independently without establishing two aetiological types, to explicit criteria first proposed in 1972 for 14 illnesses. The DSM-5 field trials found reliability mostly good to very good across the diagnoses tested, but six fell in a questionable band and three in an unacceptable one.
  • One diagnosis can look very different from patient to patient. In interviews with more than 1,500 patients, 170 of the 227 symptom combinations that satisfy the criteria occurred, and 3,703 STAR*D outpatients showed 1,030 unique symptom profiles. Taxometric work leans toward a dimension rather than a category, although the result depends on which items are analysed.
  • The types of depressive disorder include manual categories, specifiers and research constructs. Melancholic, atypical, peripartum and seasonal presentations are specifiers of an episode, not separate disorders, and in STAR*D baseline data the melancholic and atypical specifiers did not reduce symptom-profile heterogeneity.
  • Most prevalence figures describe major depressive disorder or the major depressive episode only. US surveys put lifetime prevalence at 16.2% (2003) and 20.6% (NESARC-III, DSM-5 criteria), surveys in 18 countries found lifetime episodes in 14.6% of adults in high-income and 11.1% in low- and middle-income countries, and the Global Burden of Disease Study 2010 ranked depressive disorders a leading cause of disability burden.
  • Familial risk is mostly genetic, with a heritability of liability of about 37% in pooled twin data, and it is spread across many common variants of small effect. Preregistered tests in large samples did not support 18 candidate genes, and a meta-analysis of 31 data sets found no stress by 5-HTTLPR interaction. Early adversity, neuroticism and stress add risk shared with other disorders.
  • Depression is associated with negative biases in attention, memory and interpretation and with broad executive impairment (d = 0.32 to 0.97 in one meta-analysis), and some deficits persist after mood episodes end. Nearly all of this comes from major depressive disorder samples, and whether it holds for the other depressive disorders is largely unstudied.
  • All 21 antidepressants in a 522-trial network meta-analysis beat placebo in acute treatment, at moderate to very low certainty, and the main psychotherapies outperformed usual care with few differences among them. Preventive psychological interventions cut incident depression by 19% in high-risk groups, while no mediator of psychotherapy has been firmly established.
  • Qualitative syntheses of the experience of depression report that seeking help can be experienced as a threat to identity, that alternative coping strategies are the main factor in treatment delay, and that young people most often describe sadness, social withdrawal and loneliness.
  • Much of the evidence comes from clinical samples and from high-income, Western populations. Direct prevalence data are missing for most countries, and the expression of depression differs across cultures, with Chinese outpatients in one comparison more somatic and Euro-Canadian outpatients more psychological.
  • Psychological autopsy studies found mental disorder to be the factor most strongly tied to suicide, and within depression the correlates of suicide include male gender, family psychiatric history, a previous attempt, severe depression and hopelessness. Prediction remains weak: across 365 studies spanning 50 years it stayed only slightly above chance, and sorting psychiatric patients into high-risk and lower-risk groups gave a sensitivity of 56% and a specificity of 79%.
  • In children and adolescents, fluoxetine was the only one of 14 antidepressant treatments to beat placebo in a 34-trial network meta-analysis, with very low quality evidence for most comparisons. Whether antidepressants raise suicidality in the young is contested.
  • Living with a depressive disorder brings losses beyond symptoms. Depressed people had more unemployment, absences and deficits at work, and in an eight-country survey with self-reported diagnoses presenteeism cost 5 to 10 times more than absenteeism. About 1 in 5 members of depression and bipolar organisations reported moderate or high self-stigma, stigma showed a small negative association with help-seeking, and non-pharmacological interventions reduced caregiver depression mainly in the short term.
  • Depression is linked to hyperactivity of the hypothalamic-pituitary-adrenal axis, to higher interleukin-6 and tumor necrosis factor-alpha, and to lower serum BDNF, but these are group differences that cannot show cause, and the BDNF difference is smaller in newer and larger studies. Baseline insomnia predicted later depression (odds ratio 2.10 after outlier adjustment), although whether treating insomnia would prevent depression is still open.
  • In a phase 2 trial of 59 patients, psilocybin did not differ significantly from escitalopram on the primary outcome, and its secondary advantages were not corrected for multiple comparisons. Exercise lowered depression with moderate effects across 218 studies, though confidence in the estimates was low or very low, and discontinuation symptoms occurred more often after an antidepressant than after placebo, while how frequent and lasting withdrawal is remains disputed.

What Depressive Disorder Is

MeSH defines depressive disorder as an affective disorder in which either a dysphoric mood or a loss of interest or pleasure in usual activities is present, with a mood disturbance that is prominent and relatively persistent (MeSH, D003866). MeSH has used the descriptor since 1981, files it under mood disorders, and gives eight entry terms in its own inverted form: depression disorder; depression, endogenous; depression, neurotic; depression, unipolar; depressive syndrome; melancholia; neurosis, depressive; and unipolar depression.

A Class of Diagnoses, Not a Single Illness

The term names a family of diagnoses rather than one illness. What the members share is a persistent disturbance of mood or of interest and pleasure, which separates them from a passing low mood or the everyday sense of being depressed. This page treats the class; the Depression page treats the mood state and symptom, and the lead of this article marks that division.

Within the class, the best-studied member is major depression. Reviews describe it as common and severely disabling, with effects on psychosocial functioning and quality of life (Malhi & Mann, 2018). The same review notes that presentations vary, that the course and prognosis are unpredictable, and that response to treatment is variable (Malhi & Mann, 2018). Loss of pleasure, the core of the MeSH definition, is the territory of anhedonia.

Explanation is as plural as the presentation. A review of the biology of depression concluded that the disorder is multifactorial, with genetic, stress-related and psychosocial factors each contributing, and that the monoamine theory accounts for how antidepressants act rather than for the whole illness (Belmaker & Agam, 2008). A disease primer reached the same position on major depressive disorder: no established mechanism explains all aspects of it, and heritability is estimated at about 35% (Otte et al., 2016).

Where the Class Sits

For clinical purposes the depressive disorders are set out in two manuals, the American Psychiatric Association's fifth-edition diagnostic manual, DSM-5 (American Psychiatric Association, 2013; American Psychiatric Association, 2022), and the World Health Organization's ICD-11 guidance on mental, behavioural and neurodevelopmental conditions, known as the CDDR (World Health Organization, 2024).

A comparison of the two systems found their organization substantially similar, and that ICD-11 and DSM-5 have moved closer together (First et al., 2021). The manuals' treatment of the depressive disorders, set against the older indexing term unipolar depression, is taken up in How Depressive Disorders Are Classified.

History of the Concept

The modern depressive disorders descend from melancholia, a term that meant something quite different to the earliest authors who used it. Freud's theory of depression, set out in Mourning and Melancholia, is the best-known psychoanalytic account of it. A conceptual analysis argued that the theory needs testing as a whole mechanism, and described the resulting schema as preliminary and in need of empirical adaptation (Desmet, 2013). The author added that the clinical component of his own analysis consists of unsystematic observations, and that the schema he presents is not an established body of knowledge but a starting point for empirical research (Desmet, 2013).

A historical review of twelve authors writing between the 1780s and the 1880s traced a change in what melancholia was taken to be: a disorder of the intellect in the earlier texts, and a disorder of mood in the later ones (Kendler, 2020). Five authors from 1780 through the 1830s treated it as an intellectual disturbance. Five more, writing from the 1860s to the 1880s, treated it as a primary disturbance of mood. Two texts from the 1850s fall between them, describing melancholia without delusions as a matter of mood and not of intellect (Kendler, 2020). The reviewer cautioned that the route to the modern category was more complicated than commonly assumed. The history he sketched shows two complications. One is that delusional melancholia first moved from intellect toward mood and was later reversed, so that mood came to produce the disturbance of intellect. The other is that judgments of understandability were given causal roles in some accounts (Kendler, 2020).

By the middle of the twentieth century the argument had shifted to a different division. Endogenous depression was the label for a form thought to arise from within the person, and neurotic depression, also called depressive neurosis or reactive depression, for a form thought to follow from circumstances and personality. A review of seven factor analytic studies found that the endogenous and reactive features loaded on independent factors. The endogenous factor resembled the classical depressive syndrome, whereas the reactive factor combined depression with features that were not depressive (Mendels & Cochrane, 1968). The authors concluded that the findings might not indicate two aetiological types, and the etiological distinction had not been established (Mendels & Cochrane, 1968).

The disagreement left the field with competing schemes. Kendell's narrative review described classifications of depressive illness as proliferating while rival schools disputed one another, although areas of agreement were beginning to emerge (Kendell, 1976). He treated methodological problems as large and advocated no particular solution (Kendell, 1976). Disputes of the same kind recur in later work. A review in 2000 judged both the unitarian and the binarian model unsatisfactory, proposed a hierarchical model in their place, and called for the models to be tested against each other (Parker, 2000). The remedy that eventually displaced such rivalry was the explicit diagnostic criterion set, which the section How Depressive Disorders Are Classified describes.

The old aetiological labels have not wholly disappeared. Depression, Endogenous, Depression, Neurotic, Neurosis, Depressive and Melancholia survive chiefly as MeSH entry terms, together with Depressive Syndrome, Unipolar Depression and Depression, Unipolar, and the notion of melancholia persists in the melancholic specifier discussed under Types of Depressive Disorder.

How Depressive Disorders Are Classified

Classification decides which presentations count as one illness, which as several, and which as none. For depression the answer has moved from clinical description, through explicit criteria, to two international manuals that now overlap heavily but are not identical. The debates over where the boundaries fall, covered in the sections on heterogeneity and on types, rest on the history set out here.

From Causes to Criteria

The operational turn began with a proposal of explicit diagnostic criteria for 14 psychiatric illnesses, which its authors offered as the best available at the time and expected to be revised (Feighner et al., 1972). They stated that the system was not final or closed, and that the criteria needed further testing (Feighner et al., 1972). A historical essay traces where those criteria came from, with dates of 1967 and 1972, and credits them with three key contributions to psychiatry. It also follows their later influence through the Research Diagnostic Criteria and DSM-III (Kendler et al., 2010). The essay reconstructs, as far as possible, how the criteria were developed, and for three diagnoses, major depression among them, it identifies the original criteria from which the group worked (Kendler et al., 2010).

The Research Diagnostic Criteria extended this approach to the functional psychiatric disorders. Initial reliability studies of diagnoses made with them showed high reliability, though the evidence at that stage was preliminary (Spitzer et al., 1978).

Criteria have drawn a historical critique. A comparison of textbooks published between 1900 and 1960 with the DSM-5 symptomatic criteria for major depression found that the older books gave more weight to cognitive, physical and psychomotor features and described a broader range of mood and cognitive content. The DSM, by contrast, gave more weight to neurovegetative features (Kendler, 2016). The same analysis counted eighteen symptoms and signs in the textbooks, 10 of them covered by the DSM criteria and five absent from them, and warned that some phenomena are subtle and hard to assess (Kendler, 2016). Its conclusion is that the DSM should index disorders rather than constitute them, a warning against treating the manual's definition as the illness itself (Kendler, 2016).

Unitarian, Binarian and Hierarchical Models

Three models of the whole class compete. The unitarian model treats depression as one condition varying in severity, and the binarian model splits it into two kinds. A review of the field judged both unsatisfactory, and argued that a hierarchical model with psychotic, melancholic and nonmelancholic classes is better, which the author presented as his own view (Parker, 2000). The review argued that the final common pathway model had cemented a unitarian framework in later DSM and ICD revisions, favoured a spectrum model for nonmelancholic subclasses, and called for comparative testing of neurobiological determinants and antidepressant response (Parker, 2000). Parker (2000) concluded that deciding which model fits best requires comparative testing of the two (Parker, 2000). The question remains debated: in a STAR*D analysis, the melancholic and atypical specifiers did not produce more homogeneous groups than the undivided diagnosis (Lorenzo-Luaces et al., 2021).

The case for melancholia as a distinct mood disorder has been argued directly under the heading of issues for DSM-5 (Parker et al., 2010). A narrative review of a different kind proposes that the bipolar spectrum revives Kraepelin's broad concept of manic-depressive illness. It weighs the evidence for and against the spectrum, attributes its exclusion from DSM categories to a wish to avoid overdiagnosis, and reflects the author's own clinical view (Ghaemi, 2013). That review also reports very little treatment research on the spectrum (Ghaemi, 2013).

The older vocabulary survives in the indexing terms. Melancholia and endogenous depression name the presentations that the binarian and hierarchical models treat as a separate class, whereas unipolar depression marks the contrast with bipolar illness. The manuals' response is described in the section on types: major depression carries specifiers such as with anxious distress, with mixed features and with perinatal onset (Uher et al., 2014). Whether a binary or hierarchical split should replace this arrangement remains open, because Parker (2000) held that the models required comparative testing (Parker, 2000) and the spectrum remains outside DSM categories (Ghaemi, 2013).

The Two Current Classifications

Two systems define the class in current practice. The DSM-5 is published by the American Psychiatric Association (American Psychiatric Association, 2013; American Psychiatric Association, 2022). The ICD-11 chapter on mental, behavioural and neurodevelopmental disorders has its own clinical descriptions and diagnostic requirements, published by the World Health Organization (World Health Organization, 2024).

A comparative review of 103 diagnostic entities shared by the two systems found them substantially similar in organization and now closer than before. Of the 103, 31 (30.1%) were essentially identical, and 42 (40.8%) differed in minor definitional ways. A further 10 (9.7%) differed minimally, with greater specification in DSM-5, and 20 (19.4%) showed major differences (First et al., 2021). The comparison also counted nineteen ICD-11 categories absent from DSM-5 against seven DSM-5 categories absent from ICD-11 (First et al., 2021). The authors added that the remaining divergence permits empirical comparison of the validity of the two approaches, and that differing interpretations of the evidence lie behind some of it (First et al., 2021).

One proposal sits outside both lists. A review of psychometric evidence in patient and normal samples found that anxiety and depression can be separated despite their shared distress. General distress was shared by the two, hyperarousal was specific to anxiety, and anhedonia was specific to depression. On that tripartite structure the authors proposed a mixed anxiety-depression diagnosis (Clark & Watson, 1991).

How Reliable the Categories Are

The DSM-5 field trials tested whether two clinicians, using their usual interview methods, reach the same diagnosis on separate occasions. The multisite test-retest field trial ran at eleven academic centers across the United States and Canada, and it covered 15 adult diagnoses along with eight diagnoses of childhood and adolescence (Regier et al., 2013). The authors concluded that reliability was mostly good to very good, and adequate for clinical populations assessed by the usual methods (Regier et al., 2013).

The kappa bands show a spread. Five diagnoses fell at 0.60 to 0.79 and nine at 0.40 to 0.59. Six fell at 0.20 to 0.39, which the trial called questionable, and three fell below 0.20, which it called unacceptable (Regier et al., 2013). Eight diagnoses lacked adequate sample sizes, so the bands describe the tested diagnoses as a whole (Regier et al., 2013). Across eleven academic centres in the United States and Canada, clinicians assessed each patient twice with their ordinary interview approach; most DSM-5 diagnoses with enough cases showed good to very good test-retest reliability, while six fell in the questionable kappa range, three were unacceptable and eight had samples too small to evaluate (Regier et al., 2013).

Reliability is one concern; the changes DSM-5 made to major depression are another. A review of those changes described them as small but consequential (Uher et al., 2014). The bereavement exclusion was replaced by clinical judgment, which made the diagnosis less objective and complicated research on adversity and depression. The core mood criterion gained hopelessness, and the perinatal-onset specifier added pregnancy (Uher et al., 2014). The review also flagged that the link between persistent depressive disorder and major depression is ambiguous, and that the relation of the mixed-features specifier to the bipolar-related category needs clarification (Uher et al., 2014).

Symptoms, Diagnosis and Measurement

Many Ways to Meet the Criteria

Because the diagnostic definition is polythetic, a patient needs only some of the nine criterion symptoms, and 227 different combinations qualify. Semi-structured interviews with more than 1,500 patients meeting DSM-IV criteria for major depressive disorder showed that 170 of those combinations occurred and 57, about a quarter, did not. The most frequent single pattern, found in 10.1% of patients (n = 157), was all nine criteria together. Nine combinations each accounted for more than 2% of patients, and between them they covered over 40% of diagnoses (Zimmerman et al., 2015). The authors read this as significant heterogeneity, with few possible diagnostic prototypes.

Two People, One Diagnosis

Tick the symptom domains for two hypothetical people. The symptom count is met with five or more domains, at least one of them one of the two core domains shown in bold.

Person A

5 of 9 selected: meets the symptom count

Person B

5 of 9 selected: meets the symptom count

Shared domains: 1. Both meet the symptom count while sharing one domain. Counting every set of five or more domains that includes a core domain gives 227 possible combinations.

Simplified illustration. The symptom labels are plain-language summaries, and a diagnosis also depends on duration, distress or impairment and the exclusion of other causes, which this demonstration leaves out.

A secondary analysis of STAR*D trial data points the same way. Among 3,703 depressed outpatients, with symptoms scored present or absent, 1,030 unique symptom profiles appeared. Of these, 864 (83.9%) were shared by five or fewer participants, and 501 profiles belonged to a single person each. The commonest profile covered only 1.8% of the sample (Fried & Nesse, 2015). Section 7 takes up what this variety implies for the validity of the diagnosis.

Rating Scales

Scales that claim to measure depression severity do not measure the same thing. A content analysis of 7 widely used depression scales catalogued 52 distinct symptoms across them and quantified the overlap with the Jaccard index. Little symptom content was shared between scales, and many symptoms were idiosyncratic to one instrument. The analysis coded conservatively, so the heterogeneity it found is probably an underestimate, and it covered only 7 scales, so its symptom count needs confirmation with more. The author warned that results specific to one scale threaten replicability (Fried, 2017). A severity score from one instrument is therefore not interchangeable with a score from another.

Instruments for Screening and Severity

Several instruments dominate screening and severity rating, and they differ in role and format. The Patient Health Questionnaire (PHQ-9) was validated in a study of 6,000 patients drawn from 8 primary care and 7 obstetrics-gynecology clinics, 580 of whom were interviewed by a mental health professional. The study judged it a valid and reliable measure of depression severity. At a cutoff of 10, the PHQ-9 identified major depression with 88% sensitivity and 88% specificity, and this held much the same in primary care as in obstetrics-gynecology settings (Kroenke et al., 2001).

Later evidence bears on how well that cut-off screens. An individual participant data meta-analysis of 58 of 72 eligible studies (n = 17,357, including 2,312 cases of major depression) found a cut-off of 10 or above gave the best balance of sensitivity and specificity. Against semistructured interviews (29 studies, 6,725 participants), sensitivity was 0.88 (95% CI 0.83 to 0.92) and specificity 0.85 (0.82 to 0.88). Sensitivity was 5 to 22% higher against semistructured than against fully structured interviews, and 2 to 15% higher than against the Mini International Neuropsychiatric Interview, while specificity was similar across interview types. Younger patients may be less specific than older ones, although sensitivity was similar across age (Levis et al., 2019).

For clinician ratings, the Montgomery-Åsberg scale was built from a 65-item psychopathology scale and cut to 10 items, chosen from the 17 most common symptoms of depressive illness. It was designed to be sensitive to treatment change, which implies smaller samples for clinical trials. Inter-rater reliability was high, its correlation with the Hamilton Rating Scale was significant, and it discriminated responders from non-responders better than the Hamilton scale did (Montgomery & Åsberg, 1979).

The self-report version of the Quick Inventory of Depressive Symptomatology, QIDS-SR(16), was evaluated in 596 adult outpatients with chronic major depressive disorder. Its internal consistency was alpha = .86, it correlated r = .96 with the 30-item Inventory of Depressive Symptomatology (self-report) and r = .86 with the 24-item Hamilton scale, and its sensitivity to change matched both. Multiplying its total by 1.3 predicted the 17-item Hamilton total, and the authors judged it a useful brief severity rating (Rush et al., 2003). The STAR*D trial used the QIDS-SR(16) and the 17-item Hamilton scale (HRSD(17)) as its measures (Rush et al., 2006).

Course Definitions

A task force consensus paper fixed shared terms for the points at which the course of major depressive disorder changes: remission, recovery, relapse and recurrence (Frank et al., 1991). Their purpose was to replace the inconsistent definitions that varied from one research report to the next. The authors described their operational criteria as tentative and in need of testing against data.

Dimensional Assessment

Categorical diagnosis has dimensional rivals. The Hierarchical Taxonomy of Psychopathology (HiTOP) arranges syndromes, spectra and components or subtypes along continuous dimensions, and a narrative review judged its dimensions better than traditional categories at handling diagnostic boundaries and comorbidity. The same review called the approach promising but still a work in progress, with only some domains ready for clinical use (Kotov et al., 2017). Taxometric work on unipolar depression, listed in MeSH also as Depression, Unipolar, bears on the same question. A reanalysis of survey data from n = 4,577 respondents endorsing depressed mood or anhedonia, using two nonredundant procedures on two indicator sets, favoured a dimensional solution (Prisciandaro & Roberts, 2005). Item choice matters, however. Across 4 samples totalling 984 undergraduates, distress items looked more dimensional than taxonic, whereas somatic items tied to homeostatic mechanisms looked more taxonic than dimensional (Beach & Amir, 2003). The wider debate over whether depressive disorder is one condition or several is set out in the heterogeneity section.

The Experience of Depressive Disorders

Qualitative research asks what living with depression is like from the inside, and it answers in themes rather than in rates or effect sizes. Two syntheses frame this section. Both pool accounts from people described as having depression, and neither sorts those accounts by diagnostic subtype. What they report is therefore the shared experience of depression broadly. Experience that belongs to one subtype is covered on that subtype's page, under Types of Depressive Disorder.

Help-Seeking and Identity

A meta-ethnography of 20 papers, drawing on the accounts of 474 people with depression, concentrated on the route into care (Doblyte & Jiménez-Mejías, 2017). Its main conclusion was that asking for help can feel like a threat to a person's sense of self, and it pointed to reliance on other ways of coping as one contributor to delays before treatment begins. The authors also identified other contributors to the treatment gap. Turning to alternative coping strategies was likewise associated with later treatment. Social networks pulled in both directions, since they could generate conflict over help-seeking or supply support for it (Doblyte & Jiménez-Mejías, 2017). Delay, on this account, is partly a matter of what a diagnosis would say about the person, and not only a matter of access to services.

Young People's Accounts

A qualitative meta-synthesis covered 39 studies and 884 adolescents and young people aged 10 to 24 years across 16 countries, and the review team screened 23,424 unique records to reach that set (Viduani et al., 2024). Sadness, social withdrawal and loneliness were the features reported most often. The synthesis produced three themes, and its constructs included how young people made sense of their experience, the influence of culture and context, and access to support and care. Some of the features young people described fall outside DSM and ICD criteria (Viduani et al., 2024).

The synthesis carries limits of its own. Qualitative work yields no effect sizes. Most of the included countries were high-income, most participants were female, and the literature search ran only to March 2023 (Viduani et al., 2024). Whether distress is voiced in bodily or psychological terms also varies between cultures, a point developed in Culture and Society.

Types of Depressive Disorder

The word type covers three different kinds of thing on this page, and the entries keep them apart. Some are diagnostic categories in a current manual. Some are specifiers, which describe features of an episode within a category. The rest are research constructs, MeSH descriptors and boundary presentations that sit outside the class or at its edge. MeSH lists eight children of the descriptor, and the manuals add categories and specifiers that MeSH does not list.

Cognitive findings exist for some of these types, chiefly major depressive disorder and its psychotic and melancholic forms, and are thin or absent for most of the others. Where no study on this page reports findings specific to a type, its Cognitive Implications entry says so in a single sentence.

Table 1

Types of Depressive Disorder and Related Presentations

TypeStandingPage
Major Depressive DisorderDSM-5-TR category–
Single Episode and Recurrent Depressive DisorderICD-11 categories–
Persistent Depressive Disorder (Dysthymic Disorder)DSM-5-TR categoryPersistent Depressive Disorder
Double DepressionResearch construct; now within persistent depressive disorder–
Premenstrual Dysphoric DisorderDSM-5-TR categoryPremenstrual Dysphoric Disorder
Disruptive Mood Dysregulation DisorderDSM-5-TR category–
Substance- or Medication-Induced Depressive DisorderDSM-5-TR category–
Depressive Disorder Due to Another Medical ConditionDSM-5-TR category–
Other Specified Depressive DisorderDSM-5-TR residual category–
Unspecified Depressive DisorderDSM-5-TR residual category–
Mixed Depressive and Anxiety DisorderICD-11 category; not a DSM-5 category–
Specifiers of a Depressive EpisodeSpecifiers–
With Anxious DistressSpecifier–
With Mixed FeaturesSpecifier–
With Melancholic FeaturesSpecifier–
With Atypical FeaturesSpecifier–
With Psychotic FeaturesSpecifier–
With CatatoniaSpecifierCatatonia
Postpartum Depression (With Peripartum Onset)Specifier; MeSH descriptorPostpartum Depression
Seasonal Affective Disorder (With Seasonal Pattern)Specifier; MeSH descriptorSeasonal Affective Disorder
Course and Severity SpecifiersSpecifiers–
Treatment-Resistant DepressionMeSH descriptor; not a manual categoryTreatment-Resistant Depression
Vascular DepressionMeSH descriptor; research constructVascular Depression
Subthreshold DepressionResearch construct; boundary condition–
Bipolar DepressionBoundary condition–
Prolonged Grief DisorderDSM-5-TR boundary conditionProlonged Grief Disorder

Table 1 follows the order of the entries below: categories first, then specifiers, then constructs and presentations at the boundary of the class. Standing is the status of the label, not a judgement of how well it is supported.

What Kind of Type Is It?

Choose a kind of standing to see which of the types on this page have it.

A diagnosis in a current manual. A person receives the diagnosis itself. 8 types on this page:

  • Major depressive disorder
  • Persistent depressive disorder
  • Premenstrual dysphoric disorder
  • Disruptive mood dysregulation disorder
  • Substance- or medication-induced depressive disorder
  • Depressive disorder due to another medical condition
  • Other specified depressive disorder
  • Unspecified depressive disorder

Categories and specifiers follow the current American classification; the two manuals differ in places, as the classification section explains.

Major Depressive Disorder

Major depressive disorder is the most common and most studied category. In a national survey of 36 309 US adults, 12-month prevalence was 10.4% and lifetime prevalence 20.6%, and the authors described the condition as prevalent, comorbid and disabling while noting that its specifiers need much more study (Hasin et al., 2018). A Lancet seminar likewise calls major depression common and severely disabling, with varied presentations and an unpredictable course (Malhi & Mann, 2018). The category is also heterogeneous: 3,703 STAR*D outpatients showed 1,030 unique symptom profiles, which led the authors to question its status as a consistent syndrome (Fried & Nesse, 2015). The debate is taken up in One Disorder or Many?

Cognitive Implications

Major depressive disorder is associated with broad impairment of executive function (Snyder, 2013), and a meta-analysis found that cognitive deficits persist beyond mood episodes (Rock et al., 2014).

Single Episode and Recurrent Depressive Disorder

The ICD-11 names single-episode and recurrent depressive disorder as separate categories, where the DSM keeps one category and records the course with a specifier. The ICD-11 descriptions appear in the clinical descriptions and diagnostic requirements volume (World Health Organization, 2024). A comparative review of the two systems counted nineteen ICD-11 categories with no DSM-5 counterpart and seven DSM-5 categories with none in the ICD-11, among 103 shared entities (First et al., 2021). The course specifiers themselves are described under Course and Severity Specifiers.

Cognitive Implications

No study on this page reports cognitive findings by the number of episodes.

Persistent Depressive Disorder (Dysthymic Disorder)

Persistent depressive disorder is the chronic category that brought dysthymia and chronic major depression together. A narrative review judged the validity of dysthymia doubtful, noting that the diagnosis fails to predict illness outcome, and expected the merged category to be heterogeneous as well (Rhebergen & Graham, 2014). Before the merger, one proposal would have replaced the DSM depressive categories with two dimensions, severity and chronicity, although the best form of those dimensions was left undetermined (Klein, 2008). A later paper by Parker and Malhi takes up the argument over whether the category should persist (Parker & Malhi, 2019). DSM-5 merged the two conditions, and the validity of the merged category remains debated.

Cognitive Implications

No study on this page reports cognitive findings specific to chronic depression.

Double Depression

Double depression denotes a major depressive episode superimposed on dysthymia. In pooled baseline data from 11 treatment trials, 85% of patients with chronic or double depression had clinically severe quality-of-life impairment, against 63% with major depressive disorder and 56% with dysthymic disorder, although the data came only from trial entrants (Rapaport et al., 2005). The pattern now falls within persistent depressive disorder, whose validity and heterogeneity are questioned (Rhebergen & Graham, 2014).

Cognitive Implications

No study on this page reports cognitive findings specific to double depression.

Premenstrual Dysphoric Disorder

An expert review argued that premenstrual dysphoric disorder merits a full place in the DSM-5 (Epperson et al., 2012). A philosophical paper makes the opposite case on bioethical grounds: it holds that calling the condition a mental disorder is unethical, while accepting that premenstrual distress is real lived experience (Browne, 2015). The condition is a DSM-5 category (American Psychiatric Association, 2013), and its status as a disorder is still argued.

Cognitive Implications

No study on this page reports cognitive findings specific to premenstrual dysphoric disorder.

Disruptive Mood Dysregulation Disorder

Disruptive mood dysregulation disorder is a childhood category introduced in DSM-5. Pooling three community studies (3,258 children, 7,881 observations, ages 2 to 17), one analysis found three-month prevalence of 0.8% to 3.3%, highest in preschoolers, with frequent comorbidity (Copeland et al., 2013). A meta-analysis of 14 studies of youths put pooled prevalence at 3.33% in community samples and 21.88% in clinical samples, and urged cautious interpretation because of heterogeneity and data quality (Benarous et al., 2025). In 706 outpatient children aged 6 to 12, the diagnosis could not be separated from oppositional defiant and conduct disorders; the post hoc analysis gave relative risks of 3.9 and 4.5 for those two disorders (Axelson et al., 2012). Chronic irritable mood in youth, the construct behind the category, predicts later anxiety and depression (Leibenluft, 2011). A meta-analysis found odds ratios of 1.80 for depression and 1.72 for anxiety, with no significant link to bipolar, conduct or attention-deficit outcomes (Vidal-Ribas et al., 2016). Whether the category is distinct remains unresolved.

Cognitive Implications

No study on this page reports cognitive findings specific to disruptive mood dysregulation disorder.

Substance- or Medication-Induced Depressive Disorder

The induced category rests on a distinction between depressive episodes that follow substance use and those that arise independently. In a family-study sample of 2,548 men and women, the two kinds had distinct correlates, and among alcoholics about half of lifetime depressive episodes were substance-induced (Schuckit et al., 2007). Induced episodes were more common in alcoholic probands, men and nonwhite participants, and the worst-episode symptoms of the two groups were similar. The authors concluded that the distinction is valid.

Cognitive Implications

No study on this page reports cognitive findings specific to substance-induced depression.

Depressive Disorder Due to Another Medical Condition

This category covers depression arising from a medical condition, and post-stroke depression is the best-studied example. A narrative review of post-stroke depression, which belongs to this category, reports that it occurs in a significant number of patients, that antidepressants treat and prevent it, and that inflammation, white matter disease and genetic variation may contribute (Robinson & Jorge, 2016). The cerebrovascular construct of late-life depression is treated under Vascular Depression below.

Cognitive Implications

The same review reports that major post-stroke depression carries more cognitive impairment than minor depression (Robinson & Jorge, 2016).

Other Specified Depressive Disorder

Other specified depressive disorder is the residual category for presentations that do not meet the criteria of a full category, and it is defined in the DSM-5 (American Psychiatric Association, 2013). Recurrent brief depression is one example; a review described it as well defined and prevalent, with a raised suicide risk, and blamed false negative treatment trials partly on highly selected samples (Pezawas et al., 2005). Depressive episodes with too few symptoms are a second example and are discussed under Subthreshold Depression.

Cognitive Implications

No study on this page reports cognitive findings specific to other specified depressive disorder.

Unspecified Depressive Disorder

Unspecified depressive disorder is the residual category used when the clinician does not state why the criteria for a specific category are unmet. It too appears in the DSM-5 (American Psychiatric Association, 2013).

Cognitive Implications

No study on this page reports cognitive findings specific to unspecified depressive disorder.

Mixed Depressive and Anxiety Disorder

The tripartite model separated anxiety from depression despite their shared general distress, with hyperarousal higher in anxiety and anhedonia higher in depression, and it proposed a mixed anxiety-depression diagnosis (Clark & Watson, 1991). A narrative review argued that mixed anxiety and depression is very common and as impairing as the full disorders, and noted that its proposed DSM-5 criteria were not sufficiently reliable (Möller et al., 2016). The category is kept in the ICD-11 (World Health Organization, 2024) but was not included in DSM-5 because its proposed criteria were not sufficiently reliable (Möller et al., 2016). The DSM's own route to the same territory is the anxious distress specifier, covered under With Anxious Distress.

Cognitive Implications

No study on this page reports cognitive findings specific to mixed depressive and anxiety disorder.

Specifiers of a Depressive Episode

Specifiers describe features of an episode. They do not create separate disorders, and one episode can carry several.

An analysis of STAR*D baseline data in 2,496 outpatients found that the melancholic and atypical specifiers did not reduce symptom-profile heterogeneity, and found no evidence that either specifier homogenises the diagnosis (Lorenzo-Luaces et al., 2021). A review of the DSM-5 changes to major depressive disorder called them small but consequential, and the changes included the specifiers with anxious distress, with mixed features and with perinatal onset (Uher et al., 2014).

Cognitive Implications

Cognitive findings, where any exist, are given under the individual specifier entries below.

With Anxious Distress

In a clinical interview study of 260 patients with major depressive disorder, the anxious distress specifier was common, affecting about three-quarters, and its validity was supported (Zimmerman et al., 2019). Patients with it had more anxiety and drug use disorders and poorer functioning and coping than those without. The specifier was added in DSM-5 (Uher et al., 2014).

Cognitive Implications

No study on this page reports cognitive findings specific to the anxious distress specifier.

With Mixed Features

A meta-analysis of 17 studies and 20 samples put the prevalence of DSM-5 mixed features at 11.6% of major depressive episodes and 26.8% of manic or hypomanic episodes, with moderate to high risk of bias and unexplained heterogeneity (Na et al., 2021). A historical and systematic review judged the DSM-5 criteria poorly grounded and possibly harmful for treatment choices (Koukopoulos & Sani, 2014). An editorial by Koukopoulos and colleagues went further, calling the criteria scientifically invalid and preferring agitation-based criteria, though it reported no new data (Koukopoulos et al., 2013). The criteria remain contested. The boundary with bipolar disorder is described under Related Conditions and How They Differ.

Cognitive Implications

No study on this page reports cognitive findings specific to mixed features.

With Melancholic Features

Melancholia has a long endogenous lineage, traced under History of the Concept. One paper made the case for classifying it as a distinct mood disorder (Parker et al., 2010), and a review of competing models proposed a hierarchy of psychotic, melancholic and nonmelancholic classes (Parker, 2000). It remains a specifier in the current manuals (American Psychiatric Association, 2013). In STAR*D, 1,053 patients (42.19%) were melancholic, and the specifier did not significantly reduce the number of distinct symptom profiles (Lorenzo-Luaces et al., 2021); see Specifiers of a Depressive Episode.

Cognitive Implications

A six-week prospective study compared melancholic (n = 142), atypical (n = 76) and undifferentiated (n = 91) depression with 200 healthy controls and found that the melancholic and atypical subtypes differed in processing speed and verbal fluency, although the evidence was preliminary (Lin et al., 2014).

With Atypical Features

Atypical depression is a contested subtype with no treatment guidelines of its own. A narrative review noted that prevalence varies with the criteria used and that mood reactivity is obligatory yet disputed (Łojko & Rybakowski, 2017). A systematic review of 34 latent variable analyses found no conclusive evidence for data-driven symptom subtypes, a result discussed in the section on heterogeneity (van Loo et al., 2012).

Cognitive Implications

The atypical subtype differed from the melancholic in processing speed and verbal fluency (Lin et al., 2014); see With Melancholic Features.

With Psychotic Features

Delusions mark this specifier. In a ten-year follow-up of 452 patients with major depression, delusional patients took longer to recover, used more antipsychotic medication (p < .0001 for each) and had more depressive morbidity (p < .05), yet the 10-year outcome did not differ significantly (Maj et al., 2007).

Cognitive Implications

A meta-analysis of 12 studies found poorer cognition in psychotic than non-psychotic major depression, with an overall effect size of 0.38 and 0.69 in drug-free patients (Zaninotto et al., 2015). An earlier review and meta-analysis also examined cognitive impairment in psychotic and nonpsychotic depression (Fleming et al., 2004).

With Catatonia

A meta-analysis of 74 studies and 110764 patients put the pooled prevalence of catatonia across conditions at 9.0% (95% CI 6.9 to 11.7). Results were highly heterogeneous (I2 = 98%) and publication bias was present, so the authors advised cautious interpretation (Solmi et al., 2018).

Cognitive Implications

No study on this page reports cognitive findings specific to catatonia in depression.

Postpartum Depression (With Peripartum Onset)

The classification carries a peripartum-onset specifier, and MeSH lists postpartum depression as a child descriptor. A review described postpartum depression as common and as affecting the whole family (Stewart & Vigod, 2016). A narrative review argued that it may be distinct from major depression: anxiety, psychomotor symptoms, obsessive thoughts and fatigue were higher, and sad mood and suicidal ideation lower, so the authors held that the diagnostic criteria should be modified (Radoš et al., 2024). Whether it is a distinct disorder remains debated.

Cognitive Implications

No study on this page reports cognitive findings specific to postpartum depression.

Seasonal Affective Disorder (With Seasonal Pattern)

The seasonal-pattern specifier and the MeSH descriptor for seasonal affective disorder cover depression that recurs in winter. A randomised trial of 177 community adults with recurrent winter episodes compared cognitive-behavioural therapy with light therapy over two winters. The groups did not differ in the first winter; in the second, recurrence was lower after cognitive-behavioural therapy (27.3% against 45.6%) and remission higher (68.3% against 44.5%) (Rohan et al., 2016).

Cognitive Implications

No study on this page reports cognitive findings specific to seasonal affective disorder.

Course and Severity Specifiers

These specifiers record whether an episode is single or recurrent, how severe it is, and whether it is in partial or full remission. A task force proposed consensus definitions of remission, recovery, relapse and recurrence, calling its operational criteria tentative (Frank et al., 1991); the Symptoms section gives them. A systematic review found that clinical factors predict recurrence best: previous episodes and residual symptoms raised the risk, while gender, civil status and socioeconomic status were unrelated to it (Hardeveld et al., 2010).

Cognitive Implications

Cognitive deficits persist into remission, although memory deficits in remitted samples were not significant (Rock et al., 2014).

Treatment-Resistant Depression (Treatment-Resistant Depressive Disorder)

Treatment-resistant depression is a MeSH child descriptor and not a manual category. In the STAR*D sequenced-treatment trial of 3,671 outpatients, more treatment steps went with lower remission and more relapse (Rush et al., 2006). An expert consensus review reported rapid-onset efficacy for ketamine and esketamine, while noting safety concerns and uncertainty about their place in treatment algorithms (McIntyre et al., 2021); see Brain Stimulation and Other Biological Treatments.

Cognitive Implications

No study on this page reports cognitive findings specific to treatment-resistant depression.

Vascular Depression

Vascular depression is a MeSH child descriptor and a research construct. The idea started as a proposal that cerebrovascular disease may give rise to some cases of depression in later life; the founding paper backed this with comorbidity and lesion evidence and set out questions for future work rather than testing them (Alexopoulos et al., 1997).

Cognitive Implications

The founding paper reports no cognitive finding; the Vascular Depression page covers that literature.

Subthreshold Depression

Subthreshold depression means depressive symptoms below the full threshold. A meta-analysis of 16 longitudinal cohorts (n = 67 318) found a raised incidence of major depression afterwards (IRR = 1.95, 95% CI 1.28 to 2.97), with no significant difference by age group or by community versus primary care sample, and the authors concluded that the result favours indicated prevention (Lee et al., 2019). Prevention is discussed in its own section.

Cognitive Implications

No study on this page reports cognitive findings specific to subthreshold depression.

Bipolar Depression

Bipolar depression lies outside the class: it is the depressive episode of bipolar disorder. A narrative review found broad parallels between bipolar and unipolar depression, and suggested that mania and depression may be separate, comorbid syndromes (Cuellar et al., 2005). Bipolar onset came earlier and episodes were more frequent, while anxiety and somatisation were higher in unipolar depression. The differential is set out under Related Conditions and How They Differ.

Cognitive Implications

The review lists a negative cognitive style as a predictor of depression, but direct comparisons were sparse and its conclusions need caution (Cuellar et al., 2005).

Prolonged Grief Disorder

Prolonged grief disorder is a boundary condition added to DSM-5-TR. A validation study across three bereaved samples (Yale, Utrecht and Oxford) supported its criteria and the PG-13-Revised scale, and found the diagnosis distinct from major depressive disorder (phi = 0.25) (Prigerson et al., 2021). Grief and depression are compared under Related Conditions and How They Differ.

Cognitive Implications

No study on this page reports cognitive findings specific to prolonged grief disorder.

One Disorder or Many? Heterogeneity and the Dimensional Debate

A diagnosis that can be reached in many ways invites the question of whether it names one condition. Four lines of evidence bear on it: how much patients differ, whether the difference is one of kind or of degree, whether data-driven methods find subgroups, and whether a different architecture than the diagnostic categories describes the data better. Each position below is reported with its standing.

Symptom Heterogeneity

The diagnostic rule for major depression is polythetic, so patients who share a diagnosis can share few symptoms. Analysis of 3,703 STAR*D outpatients found 1,030 unique symptom profiles, which led the authors to question whether the diagnosis denotes a consistent syndrome (Fried & Nesse, 2015). Interviews with more than 1,500 patients showed the same spread from the other side: 170 of the 227 possible ways to meet the criteria occurred, and the polythetic definition was the stated reason for the diversity (Zimmerman et al., 2015). The profile counts are reported in the Symptoms section.

Specifiers were designed partly to carve out more uniform groups, but they did not do so. In STAR*D baseline data, melancholic and non-melancholic patients showed no significant difference in the number of unique profiles, and the atypical specifier likewise yielded no reduction (Lorenzo-Luaces et al., 2021).

Taxometric Evidence

Taxometric methods test whether a construct is a category with a boundary in nature or a continuum. A quantitative review of 177 articles and 311 findings concluded that most latent variables studied are dimensional, and that earlier taxonic findings were likely spurious: taxonic results were rarer in recent studies and in those using stronger methods (Haslam et al., 2012). The review treated the evidence for some other constructs, including schizotypy, autism and substance use, as less definitive. For depression specifically, a reanalysis of National Comorbidity Survey data from 4,577 respondents who endorsed depressed mood or anhedonia, using two nonredundant taxometric procedures, favoured a dimensional solution for unipolar depression (Prisciandaro & Roberts, 2005).

The answer depends on which indicators are analysed. In four samples totalling 984 undergraduates, distress items were more dimensional than taxonic, while somatic homeostatic items were more taxonic than dimensional, so item selection determined whether depression appeared to be a taxon (Beach & Amir, 2003). That study used a student sample, not a clinical one, and the later dimensional findings do not remove its warning that the result is partly a product of the items chosen.

Category or Dimension?

Move the cut-off and switch views to see how a single severity dimension turns into a yes-or-no diagnosis.

0 (no symptoms)100 (most severe)

16 of 40 people fall at or above the cut-off and would be counted as cases. 4 sit within 5 points of it, where a small change in score flips the label.

Illustrative only. The scores are invented and evenly spread; they show how a cut-off works, not how severity is distributed in any real sample.

Data-Driven Subtypes and Biotypes

Statistical subtyping has not delivered. A systematic review of 34 latent variable analyses in 20 articles and 7,684 patients found no conclusive evidence for symptom-based dimensions or subtypes (van Loo et al., 2012). Latent classes tracked severity more than symptom-specific clusters, and 62 of 71 significant class differences fitted a severity account. Factor structure also varied with the questionnaire used, and the studies differed too much to pool. A later systematic review of 20 unique samples (about 4,000 subjects) located suggestive biological leads, among them biomarker profiles, functional connectivity and a genetic pattern, but reported lack of replication and inconsistent neuroimaging results, and drew no definitive conclusions (Beijers et al., 2019).

The biological claim rests on a multisite fMRI clustering analysis of 1,188 patients that identified four biotypes, each marked by its own pattern of limbic and frontostriatal connectivity; membership in a biotype predicted response to transcranial magnetic stimulation, and the four carried differing symptom profiles even though clinical features alone did not tell them apart (Drysdale et al., 2017). A methodological replication in 187 patients with depression or anxiety did not reproduce it. The canonical correlation and the clusters were not significant against permutation nulls, and cross-validated connectivity predicted symptoms at about zero (Dinga et al., 2019). Its authors flagged limits: it was not a direct replication, used a different clinical instrument and preprocessing pipeline, and had a heterogeneous sample with site effects limiting power. The result is best read as a failure to confirm under altered conditions, calling for caution about the biotype evidence.

A newer study derived six biotypes from fMRI in 801 treatment-free patients, found that they differed in symptoms, cognitive performance and treatment response, and tested prediction in 250 randomized patients (Tozzi et al., 2024). The authors presented it as a promising route to precision psychiatry. It is a single study that has not yet been replicated, and the history of the four-biotype result argues for waiting.

Hierarchical and Network Models

One alternative drops the diagnostic categories in favour of dimensions arranged in a hierarchy. Earlier structural work argued that mood and anxiety disorders form a single emotional class with three subclasses, and proposed that an empirically based structure replace the DSM-IV system of grouping disorders into classes, while leaving the placement of obsessive-compulsive disorder unclear (Watson, 2005). The Hierarchical Taxonomy of Psychopathology (HiTOP) develops that line. A narrative review argued that its dimensions repair flaws of the traditional categories, including their boundaries and comorbidity, and called the system promising but a work in progress, with only some domains ready for clinical use (Kotov et al., 2017). A critical review judged that HiTOP lacks verisimilitude, cannot handle multifinality and equifinality, risks algorithmic bias against underrepresented groups, and is not clinic-ready, so that the DSM should be kept clinically (Haeffel et al., 2022). A reply held that the critics' concerns were mostly unwarranted, while granting some real current limits that the HiTOP consortium is addressing (DeYoung et al., 2022). HiTOP is therefore a proposed, actively contested taxonomy.

The network account makes a different claim, that a disorder is a set of symptoms in causal interplay and not the expression of one underlying factor. Its proponents argued that this approach explains the limited success of research built on a common underlying factor and that network methods might guide intervention, which they described as a possibility (Borsboom & Cramer, 2013). A replication across two epidemiological samples (n = 9,282 and n = 8,841) found conditional independence networks poorly replicable, with unstable edge and centrality estimates, and called for higher evidence standards (Forbes et al., 2017). A reanalysis of the same two datasets rejected that conclusion, reporting that network models replicated very well, and argued that skip structure in the interviews distorts correlations and that the replicability metrics used were debatable (Borsboom et al., 2017). The network account is a live, unsettled proposal.

Where the Question Stands

The taxometric evidence leans dimensional, with the caveat that item selection can produce a categorical-looking result. No data-driven subtype, symptom-based or biological, has replicated, and the largest biotype claim failed an independent test. The hierarchical and network alternatives are promising but contested, each facing published critique and a reply. The clinical manuals still use categories, so a diagnosis of a depressive disorder remains a category applied to a population that the evidence suggests varies along continua.

Comparing the Depressive Disorders

A reader deciding which subtype page to open next needs a way to set the conditions side by side, and Table 2 offers one. It should be read as a map of what each body of work reports, not as a ranking. No single study measured all seven conditions with one instrument in one sample. The cells instead draw on a national household survey, pooled community cohorts, a meta-analysis of youth samples, a family-genetic interview study, a randomized trial and several narrative or expert reviews. Figures from different designs, populations and years do not line up, and a gap in the table usually marks a question the cited work did not address, not a finding of no difference.

Three contrasts stand out. Major depressive disorder is the best characterized: a survey of 36,309 US adults put its 12-month prevalence at 10.4% and its lifetime prevalence at 20.6%, and found it comorbid and disabling (Hasin et al., 2018). Women carried higher odds than men in two meta-analyses of national samples, with the gap widest in adolescence (Salk et al., 2017). Persistent depressive disorder and disruptive mood dysregulation disorder, by contrast, are the two categories whose independence from neighbouring diagnoses is most openly questioned. A narrative review judged the validity of dysthymia doubtful and expected the persistent form to be heterogeneous as well (Rhebergen & Graham, 2014), while pooled community data left the distinctiveness of the childhood category from existing disorders unresolved (Copeland et al., 2013). That doubt connects to the proposal that severity and chronicity dimensions could replace the DSM categories (Klein, 2008), taken up in the section on heterogeneity and the dimensional debate.

Table 2

Selected Depressive Disorders Compared

DisorderTypical onset or populationCourseDistinguishing featureEvidence on this page
Major depressive disorderUS adults surveyed; higher odds at ages 18 to 29 and in womenSpecifiers linked to early onset, poor course and suicidality (initial findings)Prevalent, comorbid and disabling; lifetime prevalence 20.6%Hasin et al., 2018; Salk et al., 2017
Persistent depressive disorder––Dysthymia validity doubtful; the persistent form likely the same and heterogeneousRhebergen & Graham, 2014; Klein, 2008
Premenstrual dysphoric disorder––Expert review argued it merits a full DSM-5 categoryEpperson et al., 2012
Disruptive mood dysregulation disorderYouths under 18–Chronic irritability and temper outbursts; uncommon and highly comorbid; pooled prevalence 3.3% in community and 21.9% in clinical samplesBenarous et al., 2025; Copeland et al., 2013
Substance- or medication-induced depressive disorderMore often in men, nonwhite participants and those with less education (family study of 2,548 adults)–Distinct correlates from independent episodes, similar worst-episode symptoms; about half of lifetime episodes among alcoholicsSchuckit et al., 2007
Postpartum depression––Common; affects the whole familyStewart & Vigod, 2016
Seasonal affective disorderCommunity adults with recurrent winter seasonal affective disorderRecurrent across winters; recurrence 27.3% after CBT-SAD versus 45.6% after light therapy in winter twoWinter pattern; CBT-SAD more durable than light therapy by the second winterRohan et al., 2016

Note. An en dash marks a cell for which no source cited on this page reports the information. Entries come from different study designs and samples, so the table is a guide to reading order and not a head-to-head comparison. The sources are Hasin et al. (2018), Salk et al. (2017), Rhebergen and Graham (2014), Klein (2008), Epperson et al. (2012), Benarous et al. (2025), Copeland et al. (2013), Schuckit et al. (2007), Stewart and Vigod (2016) and Rohan et al. (2016).

Prevalence, Course and Burden

Depressive disorders are common, they tend to recur, and they carry a large share of the world's disability and economic cost. The figures below come from different surveys with different scopes, so each is given with its source. Most are for major depressive disorder (MDD) or the major depressive episode (MDE) alone; they do not describe every depressive disorder.

How Common

In a face-to-face household survey of 9,090 US adults, 16.2% met criteria for lifetime MDD and 6.6% for MDD in the past 12 months, equivalent to roughly 32.6 to 35.1 million and 13.1 to 14.2 million adults respectively (Kessler et al., 2003). A later national survey, NESARC-III, interviewed 36,309 noninstitutionalized US adults and applied DSM-5 criteria. It gave a higher 12-month prevalence of 10.4% and a lifetime prevalence of 20.6% for MDD (Hasin et al., 2018). The two surveys differ in sample, instrument and diagnostic manual, so the gap between them is not a measured rise in the disorder.

Within the NESARC-III sample, the odds of 12-month MDD were lower in men (OR 0.5) and in African American, Asian/Pacific Islander and Hispanic respondents than in white respondents (ORs 0.6, 0.6 and 0.7). The odds were higher at ages 18 to 29 (OR 3.0) and with an income of $19,999 or less (OR 1.7) (Hasin et al., 2018).

Outside the United States, the World Mental Health Survey Initiative interviewed 89,037 adults in 18 countries with the Composite International Diagnostic Interview and reported on MDE. The lifetime figure reached 14.6% across the high-income nations surveyed and 11.1% across the lower- and middle-income ones. Twelve-month prevalence was similar across the two groups, at 5.5% and 5.9%. Mean age of onset was 25.7 years in high-income and 24.0 in low- and middle-income countries, a figure the authors note rests on retrospective report (Bromet et al., 2011).

Lifetime prevalence varies widely across countries, from 1.5% in Taiwan to 19.0% in Beirut. A review of the cross-national surveys attributes part of that spread to methodology, including differences in diagnostic thresholds, and notes that direct data are missing for most countries (Kessler & Bromet, 2013). The same review observes that the highest lifetime rates were mostly in high-income countries and the four lowest were in low- and middle-income ones, while 12-month rates were similar across income groups. Onset age and chronicity were broadly consistent across settings.

The COVID-19 pandemic raised prevalence. A systematic review and meta-regression screened 5,683 data sources, included 48 studies (46 of them on MDD), and estimated 53.2 million additional MDD cases worldwide, a 27.6% increase. Locations with higher SARS-CoV-2 infection rates and greater reductions in mobility saw larger increases. Females and younger people were more affected than males and older people (Santomauro et al., 2021).

Sex Differences

Women are diagnosed with depression more often than men. Two meta-analyses of national samples, covering over 90 nations, found an odds ratio of 1.95 (95% CI 1.88 to 2.03) for diagnoses and an effect of d = 0.27 for symptoms (Salk et al., 2017). The gap is not constant across the lifespan. It is larger in adolescence than in adulthood, with an odds ratio of 2.37 at age 12 and 3.02 at ages 13 to 15, and a symptom effect of d = 0.47 at age 16 (Salk et al., 2017). The diagnostic gap was also larger in countries with higher gender equity, though the same pattern was not significant for symptoms. The World Mental Health surveys put the female-to-male ratio of MDE at about 2 to 1 (Bromet et al., 2011).

A review of five classes of explanation concluded that none accounts for the size of the difference, and it proposed that a response set, in which women ruminate where men distract themselves, contributes (Nolen-Hoeksema, 1987). That account is weighed with its rivals in the section on rival theories.

Course

Depression recurs. A systematic review found that the cumulative recurrence of MDD ran from 35% to 85% after 15 years, with higher recurrence in samples from specialised mental healthcare than in the general population; data from the general population and primary care were scarce (Hardeveld et al., 2010). Clinical factors were the strongest predictors. Recurrence became more likely as the count of earlier episodes rose and when subclinical residual symptoms persisted, while gender, civil status and socioeconomic status showed no relation to it (Hardeveld et al., 2010).

Research on course depends on shared definitions. A task force produced consensus definitions of remission, recovery, relapse and recurrence in MDD, because reports had used the terms inconsistently. The authors called their operational criteria tentative and in need of testing against data (Frank et al., 1991). How episodes and their severity are identified is covered in the section on symptoms, diagnosis and measurement.

Burden

The Global Burden of Disease Study 2010 estimated the burden of 291 diseases and injuries across 187 countries and 21 regions using Bayesian meta-regression (DisMod-MR). It ranked depressive disorders as a leading cause of global disability burden, with major depressive disorder and dysthymia as the constructs analysed. The authors flagged sparse data from developing countries and uncertain validity of the disability weights, and the uncertainty intervals overlapped widely between regions (Ferrari et al., 2013).

Later GBD work shows the burden has not eased. The 2019 analysis of 12 mental disorders in 204 countries, which is not specific to depression, found no evidence of a global reduction in burden since 1990. Absolute disability-adjusted life-years were higher in 2019, while the age-standardised rate was unchanged, and the age-standardised rate was higher in females. The analysis counted deaths only for eating disorders, so it did not reflect premature mortality from the other disorders (GBD 2019 Mental Disorders Collaborators, 2022).

The economic burden of MDD in the United States grew from $173.2 billion in 2005 to $210.5 billion in 2010, an increase of 21.5%. Direct costs made up 45% of the 2010 total, suicide-related costs 5% and workplace costs 50%. Presenteeism cost more than absenteeism. Comorbid conditions drove most of the rise, and only 38% of the total was attributed to MDD itself. The study combined survey and claims data, imputed costs for those aged 65 and over, and had no markers of severity (Greenberg et al., 2015).

Depressive Disorders Across the Lifespan

Presentation, treatment response and treatment risk all shift with age. The evidence is densest at the two ends of the age range, in youth and in later life, and it is argued most heatedly over the safety of antidepressants in the young.

Children and Adolescents

The female excess in depression is largest in adolescence. Two meta-analyses of national samples from more than 90 nations found that women were diagnosed and scored higher on symptoms than men overall (OR = 1.95), with the gap widest around puberty: OR = 3.02 at ages 13 to 15, against 2.37 at age 12 (Salk et al., 2017).

In young people, irritability is a recognised presentation. A narrative review with a meta-analysis of 12 studies (9 cohorts, 7,594 participants) found that irritability predicted later depression (OR = 1.80) and anxiety (OR = 1.72). It did not predict later bipolar disorder, conduct disorder, ADHD or substance disorders, and its link to oppositional defiant disorder was probably inflated by overlapping items (Vidal-Ribas et al., 2016). The same review noted that the stability of a categorical diagnosis of disruptive mood dysregulation disorder was poor; that diagnosis is taken up under Types of Depressive Disorder.

Medication evidence in youth is thin. A network meta-analysis of 34 trials (5,260 children and adolescents with major depressive disorder) compared 14 antidepressant treatments. Fluoxetine alone beat placebo on efficacy (SMD -0.51, 95% CrI -0.99 to -0.03), and the other drugs did not separate from placebo. Imipramine, venlafaxine and duloxetine each led to more discontinuations from adverse events than placebo did, and fluoxetine was better tolerated than duloxetine and imipramine. The authors found no clear advantage for any drug and judged fluoxetine probably the best option, though the quality of evidence for most comparisons was very low (Cipriani et al., 2016).

Psychotherapy shows an age gradient. Across 366 trials (36,702 patients), the pooled effect was g = 0.75, with very high heterogeneity and suboptimal study quality. Effects were smaller in children (g = 0.35) and adolescents (g = 0.55) than in middle-aged adults (g = 0.77), and largest in young adults (g = 0.98). The authors judged the smaller effect probable in children and possible in adolescents. Older adults (g = 0.66) did not differ significantly from the oldest group (g = 0.97) (Cuijpers et al., 2020).

The Black-Box Warning Debate

Whether antidepressants raise suicidality in the young is contested. One side holds that the warning rests on trial data. A critical narrative review found that antidepressants produced more suicidality events than placebo in youth trials (pooled OR = 1.71, 95% CI 1.14 to 2.77, across 23 randomised trials), while rating claims that the warning itself caused harm as weak. The review also noted that the trials recorded no completed suicides and usually excluded active suicidality, and that national suicide statistics were not necessarily accurate (Spielmans et al., 2020).

The other side examines what followed the warnings. A systematic review of 34 studies, of which 11 had rigorous designs, reported less mental health care and more psychotropic poisonings after the warnings of 2003, 2005 and 2007. Poisonings serve as a proxy for suicide attempts in those studies. The reviewers concluded that the warnings' risks and benefits needed reevaluation (Soumerai et al., 2024).

Adult trials show why age matters to the argument. Among 99,231 adults in 372 trials, the odds of suicidal behaviour were higher on antidepressants than placebo under age 25 (OR 2.30, 95% CI 1.04 to 5.09), although the combined ideation and behaviour estimate for that group was not significant. Odds were lower on drug than on placebo at ages 25 to 64 (OR 0.79) and at 65 and over (OR 0.37) (Stone et al., 2009). The standing of the warning is therefore unsettled: the trial evidence supports its basis, while the population-level evidence on its consequences has prompted calls for review.

Older Adults

Antidepressant classes differ in adverse effects in later life. A systematic review of adults aged 65 and over with major depressive disorder (19 randomised trials and two observational studies) found no excess of overall adverse events with SSRIs relative to placebo, but a raised rate with SNRIs (high strength of evidence). Withdrawals due to adverse events exceeded placebo for SSRIs (low strength) and for SNRIs (moderate strength), and duloxetine raised the rate of falls (moderate strength). The trials were not powered for adverse events, head-to-head comparisons were few, and falls were underreported (Sobieraj et al., 2019).

Late-life depression also tracks later dementia. A meta-analysis of 23 cohort studies (49,612 older adults) reported higher risk of all-cause dementia in depressed participants than in controls (1.85, 95% CI 1.67 to 2.04), with elevated risk for Alzheimer's disease (1.65) and a larger one for vascular dementia (2.52; difference P = 0.03, from only five studies). The authors treated the result as an association. Their caveats included follow-up ranging from 1 to 17 years, no structured diagnostic interviews, pooled hazard and odds ratios, and no separate analysis of recurrent episodes (Diniz et al., 2013).

The larger vascular estimate links to Vascular Depression, a late-onset presentation treated in its own article.

Beck's Cognitive Model

Beck's cognitive model is the best-known psychological account of how depression arises and persists, and it has been revised in stages for more than 40 years. In his 2008 theoretical review, Beck placed dysfunctional beliefs, biases in how information is processed, cognitive distortions and automatic thoughts within one framework (Beck, 2008). The model treats these as the working parts of depressive thinking. The sections that follow set out what each part is said to do and how far later research supports it.

The parts sit at different levels. Dysfunctional beliefs are the stored, enduring layer, and they are formed in the course of development. Automatic thoughts and cognitive distortions are the layer a person experiences in the moment, the thoughts and misreadings of events that appear once the beliefs are in play. Information-processing biases are the layer beneath both, the habitual tilt in how material is taken in and handled. The 2008 account does not, in the fields available here, specify how each part sustains low mood, so the maintenance claims below are limited to what later reviews report.

The model also describes a sequence of vulnerability and onset. Trauma early in life and the development of dysfunctional beliefs count among the factors that predispose a person to depression. A later stressor that matches the content of those beliefs then acts as the precipitant (Beck, 2008). Depression is therefore not triggered by stress alone. The stressor has to be congruent with a belief the person already holds, and that matching is what the sequence adds to a plain stress account. In this order, early adverse experience lays down the beliefs, a congruent stressor activates them, and the automatic thoughts, distortions and processing biases are the expression of that activation.

The 2008 version added a biological layer. It linked the cognitive model to a hypersensitive amygdala, hypoactive prefrontal regions and hypercortisolemia (Beck, 2008). Two caveats come with that integration. By the author's own account, the integration was demonstrated only recently, and the cortisol pathway was probably cognitively mediated (Beck, 2008). Neural correlates and the causal direction between thought and biology are treated again in the Biology of Depressive Disorders section and in the section on testing causal claims about cognition.

A later integrative revision changed the account's emphasis. Beck and Bredemeier (2016) proposed that depression is an energy-conserving adaptation that follows a perceived loss, and that severe depression is an anachronistic evolutionary program (Beck & Bredemeier, 2016). The symptoms were presumably adaptive once and are now maladaptive. In their model, negative information processing and strong stress reactivity each raise the likelihood of depressogenic beliefs, and those beliefs in turn raise the likelihood of loss appraisals. Here the model does state a route by which the parts feed one another: biased processing strengthens the beliefs, and the beliefs strengthen the appraisal of loss. The beliefs concern the self, the world and the future, and the proposed outcomes include anhedonia and anergia, with emotion regulation networks also part of the account (Beck & Bredemeier, 2016). This is a theory paper, so its predictors are proposed relations, not tested effect sizes.

Recent reviews bear on the model's current standing, and they support its central claim of biased cognition while changing the explanation. Drawing on roughly 50 years of research, a narrative review linked depression to heightened negative biases and diminished positive ones in how people process material about themselves, interpret events, direct attention and remember (LeMoult & Gotlib, 2019). The authors concluded that deficits in cognitive control probably underlie these biases. Difficulty with inhibition predicted more rumination, more attention bias and more memory bias, and less reappraisal (LeMoult & Gotlib, 2019). Such difficulty may also impede the flexible choice of emotion regulation strategies. An earlier narrative review had reached a similar picture, with biased negative processing tied to cognitive control, working memory, inhibition and rumination (Gotlib & Joormann, 2010). That review noted that neural and genetic factors still had to be brought in, and that integrative investigations were still required.

The standing of the model is therefore mixed. Its core claim that depression involves biased negative cognition is well supported by the reviews. Its older emphasis on beliefs and distortions as the first cause has been joined by an account in which control processes carry much of the weight. Both reviews are narrative, so neither provides a pooled estimate.

This evidence comes almost entirely from samples with major depressive disorder, and its reach to the other depressive disorders in this article is largely untested. The fuller treatment of cognitive models in that disorder is on the Depression page.

Rival Theories and Accounts

Several accounts compete with, or complement, the cognitive model of depression. Each is summarised below with its current standing.

Learned Helplessness and Hopelessness

The attributional reformulation of helplessness theory recast learned helplessness in terms of attribution theory, distinguishing universal from personal helplessness and sorting attributions as stable or unstable, global or specific, and internal or external. The authors conceded that the original hypothesis lacked explanatory scope and that the implications for depression were only outlined (Abramson et al., 1978). Abramson and colleagues later proposed a hopelessness theory of depression (Abramson et al., 1989). Neuroscience then revised the original account. Fifty years after the first helplessness work, a review argued that passivity is the default response and that control is what is learned, so the original theory had it backward. The same paper called its implications for depression speculative (Maier & Seligman, 2016).

Prospective data bear on the cognitive-vulnerability side of these accounts. In 347 college freshmen with no initial disorders, followed for 2.5 years, negative cognitive styles predicted both first onsets and recurrences of depression, with 3.5 to 6.8 times greater odds in the high-risk group. Specificity was only partial, because the styles also predicted anxiety that was comorbid with depression (Alloy et al., 2006). A systematic review of 67 studies concluded that the core components of hopelessness theory are supported and that the clinical evidence is thin. Many findings were cross-sectional, clinical samples were scarce, the prospective depression evidence was preliminary, and the theory is often conflated with learned helplessness theory (Liu et al., 2015). Its standing is therefore that of a theory with supported core components and a thin clinical evidence base.

Response Styles and Rumination

Response styles theory began as a proposal about why unipolar depression is more common in women than in men. A narrative review sorted the explanations into five classes and found that none of them accounted for the size of the difference; the author proposed instead that a habitual response set separates the sexes (Nolen-Hoeksema, 1987). On that proposal, men tend to distract themselves while women tend to ruminate, and the men's more active response may be more adaptive on average, a point the review itself hedged as only a possibility (Nolen-Hoeksema, 1987). Longitudinal analyses later linked rumination to depressive disorders and to new onsets of depressive episodes. The link to chronicity vanished once baseline symptoms were controlled, and the tie to mixed anxiety and depressive symptoms was tentative (Nolen-Hoeksema, 2000). Its standing is therefore that of a supported theory of onset, with the sex-difference explanation resting on a proposal and the chronicity claim unsupported.

The Tripartite Model

A review of psychometric evidence from patient and normal samples concluded that anxiety and depression are separable despite the distress they share (Clark & Watson, 1991). In the tripartite structure, general distress is common to both conditions, physiological hyperarousal is higher in anxiety, and anhedonia, with low positive affect, is higher in depression (Clark & Watson, 1991). The same review proposed a mixed anxiety-depression diagnosis (Clark & Watson, 1991). A later structural paper went further and argued that mood and anxiety disorders form roughly one emotional class, with three subclasses, and that an empirical structure should replace the DSM-IV categories. It left the placement of obsessive-compulsive disorder unclear and called for more research (Watson, 2005). The model remains a live structural proposal rather than a settled classification.

Diathesis-Stress and Its Critics

Diathesis-stress theories hold that a vulnerability combines with life stress to produce depression. A theoretical review of two decades of life-stress research argued that the findings challenge these models, and that broader and more specific frameworks are possible. It noted that the effect of the diathesis on the person's life is ignored and that stress assessment may be biased (Monroe & Simons, 1991). Differential susceptibility is the rival framing. A narrative review proposed that individuals differ in plasticity, not only in vulnerability, and that temperament, endophenotypic attributes and specific genes may mark that plasticity. The same review acknowledged the limits of the evidence and unknowns in the susceptibility equation (Belsky & Pluess, 2009). Diathesis-stress is thus contested without having been replaced, and differential susceptibility is proposed rather than established. What Causes Depressive Disorders sets out the causal evidence in full.

Social Signal Transduction

The social signal transduction theory traces a path from social threat to proinflammatory cytokines and then to depression. Social stress, rejection and early life stress are each treated as positive predictors of depression, and the review presents the account as a biologically plausible hypothesis in which inflammation may be a key driver, one that might yield explanatory and treatment insights (Slavich & Irwin, 2014). It is a theory under investigation, not an established mechanism.

The Monoamine Hypothesis

A systematic umbrella review of 17 studies, drawn from 361 publications and covering six research areas, found no consistent association between serotonin and depression (Moncrieff et al., 2023). The review flagged limits in the evidence it covered: prior antidepressant use was not reliably excluded, the 5-HT 1A and serotonin transporter evidence was weak, no tryptophan depletion review had appeared since 2007, and 11 of 17 reviews scored under 50% on quality (Moncrieff et al., 2023). A critical commentary replied that the null conclusion was overstated. It pointed to effect sizes such as g = −0.45 and g = −0.84, and to PET imaging of the 5-HT 1A receptor, and it noted wide confidence intervals in the tryptophan depletion findings and that its own authors hold pharmaceutical industry ties (Jauhar et al., 2023). The standing is that no simple serotonin deficit is supported, while serotonin's involvement remains argued. Biology of Depressive Disorders covers the neurochemical evidence.

Cognitive Biases and Executive Function

Cognitive research on depressive disorders has been organised for roughly 50 years around one observation: people with depression process information in a biased way. Reviews describe a negative bias that runs through attention, memory and the interpretation of ambiguous material, together with fewer positive biases than are seen in people who are not depressed (LeMoult & Gotlib, 2019). An earlier review reached the same overall picture, describing depression as marked by negatively biased cognitive processing (Gotlib & Joormann, 2010). The bias is therefore not confined to one stage: it touches how attention is allocated, how experience is encoded and recalled, and how events are construed. The attention component links to the wider literature on attentional bias. Beck's own account of such biases is taken up in the section on his cognitive model.

Negative Biases in Attention, Memory and Interpretation

Memory bias has a specific form in overgeneral autobiographical memory, which a model-elaborating review linked to emotional disorder (Williams et al., 2007). That account ties the memory pattern to capture and rumination, to functional avoidance, and to executive capacity and control, within a hierarchical search model. The support for it is at the level of the abstract only.

Interpretation bias has the clearest quantitative estimate. A meta-analysis of 87 studies (N = 9,443) found depression associated with interpretation bias at a medium effect size, g = 0.72 (95% CI 0.62 to 0.82) (Everaert et al., 2017). Clinical depression, remitted depression and elevated symptoms did not differ in the size of the effect. Self-referential stimuli produced larger effects than other stimuli, and direct measures larger effects than indirect ones, for which the effect was not significant. Imagery instructions made no significant difference. The review also found a negative bias and a lack of positive bias to be similar in size, and it cautioned that methodological factors shape the conclusions.

Cognitive Control as the Proposed Basis

A recurring proposal is that the biases rest on a weakness in cognitive control. The 2019 review concluded that cognitive control deficits probably underlie the biases (LeMoult & Gotlib, 2019). In its account, difficulty with inhibition goes with more rumination, with less reappraisal, and with stronger attention and memory biases. The same review noted that such a deficit potentially impedes the flexible selection of emotion regulation strategies. The earlier review pointed to the same family of constructs: cognitive control, working memory and inhibitory processes, alongside rumination, emotion dysregulation and the elaboration of negative information (Gotlib & Joormann, 2010). That review also stated that neural and genetic factors must be brought in and that integrative investigations are still required, so the proposal remains a hypothesis about mechanism and not a settled finding.

Reward Processing and Anhedonia

Anhedonia is usually treated as a single symptom, but one narrative review argued that it should be split into loss of pleasure and loss of motivation, with consummatory and motivational reward as separate constructs (Treadway & Zald, 2011). That review reported no new data, found the evidence linking dopamine to anhedonia contradictory, and questioned the framing of anhedonia as a steady-state mood.

Imaging work points to the striatum. In an fMRI study using a monetary incentive delay task, 30 unmedicated people with major depressive disorder showed weaker responses to gains than 31 healthy comparison participants, in the left accumbens and bilateral caudate (Pizzagalli et al., 2009). Group differences during anticipation were weaker, and the authors read the pattern as possible basal ganglia dysfunction in consummatory reward. Anhedonia and depression severity were each associated with smaller caudate volume, though the morphometric analyses were secondary.

A later integrative review proposed that anhedonia arises from stress-related dysfunction in reward processing (Pizzagalli, 2014). The model is a heuristic, and its supporting evidence is described as initial. The review also noted mixed evidence on loss of pleasure, limited specificity of anhedonia as an endophenotype, an unclear split between state and trait, and a stress-dopamine link that rests on animal data.

Executive Function

The strongest quantitative evidence concerns executive function. A meta-analysis of 113 studies, with 3,936 patients with major depressive disorder and 3,771 healthy controls, found broad impairment across executive domains, including inhibition, shifting, updating, planning and verbal fluency (Snyder, 2013). The effect sizes ranged from d = 0.32 to 0.97. The review set limits on what the result shows: a causal link was not established, specificity to major depressive disorder was not established, and the evidence on age, severity and medication was weaker or only suggestive. Its author also flagged task impurity, the restriction to published English-language studies, and coding of the studies by a single author as limitations.

Deficits Beyond the Episode

Impairment is not limited to periods of low mood. A systematic review and meta-analysis found significant deficits in executive function and attention that persist in people with remitted depression, and non-significant deficits in memory (Rock et al., 2014). Two caveats apply: the memory deficits in remitted samples were not significant, and the included studies relied on the Cambridge Neuropsychological Test Automated Battery (CANTAB) alone.

Rumination

Rumination is the cognitive process most often linked to the course of depression, and the section on rival theories and accounts treats it as a theory in its own right. In longitudinal analyses, rumination was associated with new onsets of depressive disorders (Nolen-Hoeksema, 2000). The same analyses carried caveats: a link to chronicity vanished once baseline symptoms were controlled, and a link to mixed anxiety and depressive symptoms was tentative.

How Far the Evidence Reaches

Much of this work comes from samples with major depressive disorder, as in the executive function meta-analysis above (Snyder, 2013). Whether the same biases and deficits appear in persistent depressive disorder, premenstrual dysphoric disorder, disruptive mood dysregulation disorder or substance- and medication-induced depressive disorder is largely unstudied. The findings are best read as describing the major depressive episode, and their extension to the other depressive disorders is an open question and not a result.

Testing Causal Claims About Cognition

Cognitive accounts of depressive disorders say that biased ways of processing information help to produce the disorder, and a correlation between negative thinking and low mood cannot settle that. Three kinds of evidence bear on the causal question: prospective data, experimental manipulation of the bias, and studies of early adversity that try to separate exposure from confounding. None is decisive, and each has a published counterweight.

Prospective data come first. Rumination predicted new onsets of depressive disorders in longitudinal analyses (Nolen-Hoeksema, 2000). Two of the findings in that report were weaker, however: the link to chronicity vanished once baseline symptoms were controlled, and the link to mixed anxiety and depressive symptoms was described as tentative. The rumination evidence is taken up in the Cognitive Processes section.

The experimental test is cognitive bias modification, in which a training procedure changes an attentional or interpretive bias so that any effect on mood can be read as causal. A meta-analysis of 49 randomized trials, reported in 44 articles, found effects that were small and possibly null in clinical terms (Cristea et al., 2015). For depression, 17 trials gave g = 0.43, which fell to g = 0.33 once two outliers were removed. A trim-and-fill correction for publication bias, which imputed seven missing studies, reduced it to g = 0.09, and the result was non-significant after that adjustment. Trial quality was a further limit: 33 of the 49 trials met fewer than three quality criteria, and only 5 met all of them. The authors concluded that the effects were small or absent.

A reanalysis of the same studies argued that the null reading overlooked a process question (Grafton et al., 2017). Where attentional or interpretive bias was changed successfully, emotional vulnerability changed with it. Where attentional change failed, vulnerability did not change significantly. Online delivery succeeded in changing bias in 11% of studies against 74% for laboratory delivery. On this reading, a weak average effect may come from procedures that often fail to move the bias, so it does not show that the bias is causally idle. The reanalysis reused the data of the earlier meta-analysis, and it found no significant heterogeneity.

Early adversity is the third line, and it relies on observation. In a retrospective cohort survey of 9,460 health-plan adults, a higher count of adverse childhood experiences was associated with depressive disorders in a graded way, and the risk persisted for decades (Chapman et al., 2004). Childhood emotional abuse was likewise linked to depression. The design is retrospective, and the dose-response relation weakened slightly after adjustment for a mentally ill household member. A meta-analysis of 34 quasi-experimental studies (54,646 participants, 29 cohorts) addressed confounding directly by using twin, sibling, adoption and related designs (Baldwin et al., 2023). The association between maltreatment and mental health problems fell from d = 0.56 unadjusted to d = 0.31 adjusted, a reduction of 45%. The authors judged the result consistent with a small causal contribution. They also noted that small-study bias was driven by one study programme and that propensity-based methods leave unmeasured confounding possible. The outcome in that analysis was mental health problems broadly, not depressive disorders alone.

The standing of the question can be stated plainly. Causal roles for cognitive biases are plausible: rumination precedes onset, successful bias change tracks vulnerability change, and adversity retains a small association after adjustment. They are not established, because the manipulation trials are weak on average, and the process reanalysis reanalyses the same disputed data.

What Causes Depressive Disorders

No single cause accounts for depressive disorders. The evidence points to many small genetic influences, a set of developmental pathways, early and ongoing adversity, and a personality profile shared with other disorders, all of which interact over time. The subsections below set out each strand, with its evidence weight and its limits.

Heritability and Genes

Depression runs in families, and the familial pattern is mostly genetic. A review and meta-analysis of five family studies and five twin studies found that relatives of affected probands carried a higher risk of major depression than relatives of comparison subjects (odds ratio 2.84, 95% CI 2.31 to 3.49). The twin data put the heritability of liability at 37% (95% CI 31% to 42%). Shared environment contributed 0% (95% CI 0% to 5%), and individual-specific environment accounted for 63% (95% CI 58% to 67%) (Sullivan et al., 2000). The authors noted that no adoption study met their inclusion criteria, and that measurement error is confounded with the individual-specific environmental estimate.

Molecular work has since identified specific variants. A genome-wide meta-analysis of three cohorts, with 246,363 cases and 561,190 controls, found 102 independent variants and 269 genes associated with depression, with enrichment in prefrontal brain regions. Of the 102 variants, 87 replicated in a further 1,306,354 individuals. SNP heritability was 0.089. The phenotypes were broad and often self-declared, and 15 of the 102 variants did not replicate (Howard et al., 2019). A later trans-ancestry meta-analysis of 109 cohorts from 29 countries (688,808 individuals with major depression and 4,364,225 controls) reported 697 associations at 635 loci, 293 of them novel. The associated genes were enriched in neuronal cell types and in targets of antidepressant drugs. Polygenic scores explained up to 5.8% of liability variance in European ancestry samples, but their transferability to other ancestries was limited, and the sample was 76.6% European (Adams et al., 2025). Some of its cases rested on self-report rather than a formal diagnosis.

The older candidate-gene programme has not held up. Preregistered analyses of 18 candidate genes in samples of 62,138 to 443,264 found no support for any of them as depression risk loci, and none for polymorphism-by-environment interactions involving childhood abuse or socioeconomic adversity. The authors concluded that earlier positive findings were likely false positives (Border et al., 2019). A collaborative meta-analysis of 31 data sets (38,802 participants of European ancestry) likewise found no interaction between stress and the 5-HTTLPR serotonin transporter polymorphism in any subgroup, and no main effect of genotype. Life stressors themselves raised the risk of depression. The authors judged that an interaction, if real, would be modest and context-limited (Culverhouse et al., 2018).

The standing is therefore as follows. Depression is moderately heritable, the heritability is spread across a very large number of common variants of small effect, and the early candidate genes and candidate gene-by-environment interactions are unsupported.

Developmental Pathways

Kendler and colleagues modelled how risk accumulates from childhood to adulthood. In 1,942 adult female twins followed over nine years, structural equation modelling of 18 risk factors in five developmental tiers (six correlations and 64 paths) explained 52% of the variance in liability to depressive episodes. Childhood tiers held genetic risk, family environment, abuse and parental loss. Adolescence added neuroticism, self-esteem, anxiety and conduct disorder. Adult and recent tiers held divorce, prior depression, marital problems and life events. The model resolved into three risk pathways, internalizing, externalizing and psychosocial adversity (Kendler et al., 2002). Among the study's limitations, the authors pointed to difficulties with causal inference, possible bias in retrospective recall, and the fact that their statistical model was purely additive.

A parallel analysis of 2,935 adult male twins, with the same 18 factors and five tiers (six correlations and 76 paths), explained 49% of the variance in liability and recovered the same three pathways. The overall picture was much the same for both sexes, and where differences appeared they were modest: parental loss and low self-esteem counted for more in men, genetic risk had a broader spectrum of action in men, and the externalizing pathway was not more prominent in men than in women (Kendler et al., 2006).

Figure 1

Three Risk Pathways to Major Depression in Kendler's Developmental Model

Schematic. Three boxes on the left, labelled internalizing pathway, externalizing pathway and psychosocial adversity pathway, each connect by an arrow to a box on the right labelled major depression. Internalizing pathway Externalizing pathway Psychosocial adversity pathway Major depression
Note. After Kendler et al. (2002, 2006). Original schematic; the full models contain many measured risk factors within each pathway.

Early Adversity

Adverse childhood experiences are associated with depressive disorders in a graded way. In a retrospective survey of 9,460 adults in a San Diego health plan, a higher count of adverse childhood experiences went with more lifetime and more recent depression. Childhood emotional abuse was associated with depression too, with odds ratios of 2.7 in women and 2.5 in men. The dose-response pattern attenuated slightly after adjustment for a mentally ill household member, and the design relied on recall (Chapman et al., 2004).

Whether such associations are causal is a separate question, taken up in Testing Causal Claims About Cognition. A meta-analysis of 34 quasi-experimental studies (54,646 participants in 29 cohorts) found that the association between maltreatment and mental health problems fell from a Cohen's d of 0.56 unadjusted to 0.31 after adjustment, 45% smaller. The authors read this as consistent with a small causal contribution. The outcome there was mental health problems broadly, not depression alone (Baldwin et al., 2023).

Personality and Shared Vulnerability

Personality traits link depression to the other common disorders. Pooling 175 studies (66 analyses, 851 effect sizes), a meta-analysis found that every diagnosis it examined showed higher neuroticism (mean d = 1.65) and lower conscientiousness (mean d = -1.01) than controls, with dysthymic disorder also standing out for low extraversion (d = -1.47) (Kotov et al., 2010). The authors concluded that the common disorders are strongly personality-linked and share similar profiles (Kotov et al., 2010).

What that link means causally is unsettled. A narrative review comparing five models (vulnerability, spectrum, common cause, scar and state) judged neuroticism's causal role unresolved and no one model sufficient. It concluded that neuroticism is etiologically uninformative for now but a useful marker of general risk, and that adjusting for baseline symptoms attenuated its prospective association with common mental disorders (Ormel et al., 2013).

Stress, Stress Generation and Kindling

Stress is both a cause and a consequence. A systematic review of 57 articles found that depression predicted later dependent stress, meaning events partly shaped by the person's own behaviour. This stress generation finding was replicated, and the authors saw substantial support while calling for more research. Most studies measured episodic rather than chronic stress, and self-report measures were common (Liu & Alloy, 2010).

The kindling hypothesis proposes that stress matters most for early episodes. A theoretical review described first episodes as more strongly associated with life stress than recurrences, and offered a preliminary framework for competing models given inconsistencies in the literature (Monroe & Harkness, 2005). The hypothesis also holds that episodes become closer together as their number grows, and a longitudinal community cohort tested that tenet. Among 304 participants with two or more episodes, the first interval was about three years shorter in highly recurrent people (mean 3.24 versus 6.26 years). Within persons, however, the shortening was small and not significant (b = -0.15 years, p = .309), so the pooled effect (b = -0.90 years) reflected between-person differences (b = -1.39 years). The authors read the pooled shortening as an artifact, so the cycle-acceleration tenet remains contested; their questioning of it applies to a young, white, middle-class community sample with few severe cases (Anderson et al., 2016).

Individuals also differ in how much any environment affects them. A narrative review proposed differential susceptibility, in which plasticity varies across people beyond simple vulnerability to harm, as an alternative to diathesis-stress (Belsky & Pluess, 2009). Rival Theories and Accounts weighs that proposal.

Shared and Specific Risks

Social determinants, including poverty, shape exposure to many of the risks above and are covered in Culture and Society. Most of the strands in this section describe risk common to the depressive disorders as a group: polygenic liability, adversity, neuroticism and stress. Some risks are specific to a subtype, such as those of premenstrual, postpartum, seasonal, cerebrovascular and substance-related depression, and each is set out in its entry in Types of Depressive Disorder.

Biology of Depressive Disorders

No single mechanism accounts for major depressive disorder. Reviews of the field describe it as multifactorial, with genetic, stress-related and psychosocial factors each linked to higher risk (Belmaker & Agam, 2008). A disease primer reaches the same verdict and adds that no one mechanism explains everything, so each biological account below is a partial one (Otte et al., 2016). That primer names hippocampal volume alterations, a cognitive control network, an affective-salience network and the hypothalamic-pituitary-adrenal axis among the candidate systems, and it lists childhood abuse among the environmental risks. Brain-level findings are developed in the Brain Networks section, and sleep-related systems in Sleep and Circadian Rhythms. Most of the evidence below compares people with and without depression at one point in time, and designs of that kind cannot show which factor came first or whether it caused the disorder.

The Monoamine Account

The monoamine hypothesis is the best-known account, and its standing is contested. In the monoamine oxidase theory, the actions of antidepressant drugs are explained through monoamine systems, and one review treats the theory as explaining those drug actions rather than as a full account of depression (Belmaker & Agam, 2008). A narrative review judged simple theories that tie low serotonin to low mood untenable, while noting that serotonin still informs the neuropsychology of emotional processing. It also held that serotonin impairment is neither necessary nor sufficient for depression (Cowen & Browning, 2015). A systematic umbrella review of 17 studies across six research areas found no consistent association between serotonin and depression. It also noted that prior antidepressant use was not excluded in the underlying work, that evidence on the 5-HT 1A receptor and the serotonin transporter was weak, and that antidepressants may themselves lower serotonin (Moncrieff et al., 2023). A critical commentary countered that the umbrella review's null conclusion was overstated, pointing to PET imaging and tryptophan depletion evidence. Its authors disclosed pharmaceutical industry ties, and wide confidence intervals limited the depletion findings (Jauhar et al., 2023). The exchange is set out in full in Rival Theories and Accounts; the point here is that serotonin has not been established as the cause of depression, and the question remains open.

The Stress-Response System

The stress-response system has the largest body of evidence. A meta-analysis of 361 studies and 18,454 individuals linked depression to hyperactivity of the hypothalamic-pituitary-adrenal axis, with the size of the effect varying by group. Cortisol and adrenocorticotropic hormone were higher in depressed than in nondepressed people, corticotropin-releasing hormone did not differ significantly, and men and women did not differ significantly. The cortisol difference was larger in older hospitalised samples and in melancholic or psychotic depression, and smaller (d = 0.33 against d = 0.60) in the more rigorous studies. Methods differed between studies (Stetler & Miller, 2011).

Inflammation

Inflammation is a second proposed pathway. A meta-analysis of 24 studies found tumor necrosis factor-alpha and interleukin-6 higher in depressed people than in controls, by 3.97 pg/mL (13 studies) and 1.78 pg/mL (16 studies). Other cytokines did not differ significantly, individual studies gave mixed results, and only unstimulated measures from English-language literature were included (Dowlati et al., 2010). A later systematic review and meta-analysis of 3212 people with major depressive disorder and 2798 healthy controls also found raised interleukin-6 and tumor necrosis factor-alpha, with lower interferon-gamma. Heterogeneity was large, several other cytokines were not significant, and only peripheral levels were measured (Köhler et al., 2017). These are group differences, so they do not show that inflammation produces depression. In the social signal transduction theory of depression, social threat such as rejection raises proinflammatory cytokines, and that proinflammatory phenotype may drive depression (Slavich & Irwin, 2014). The authors presented the account as a biologically plausible hypothesis that might offer explanatory and treatment insights, not as a settled mechanism. The same review links early life stress to depression, which connects this pathway to the stress findings above.

Neurotrophic Factors

A neurotrophic account holds that low brain-derived neurotrophic factor (BDNF) contributes to depression and that antidepressants raise it. A critical review put this case and acknowledged some limitations, and it said other growth factors were only possibly involved (Duman & Monteggia, 2006). A meta-analysis found serum BDNF lower in 2384 antidepressant-free depressed patients than in 2982 healthy controls. The authors cautioned that publication bias shrinks the effects, that newer and larger studies show smaller differences, and that BDNF was not consistently linked to symptom severity, so the evidence is noisier than first thought (Molendijk et al., 2014).

Genetics

Genetics contributes risk across many loci and not through any single gene. A genome-wide association meta-analysis of 807,553 adults, 246,363 of them cases, identified 102 independent variants, and 87 of the 102 replicated in a further 1,306,354 people (Howard et al., 2019). The associated genes were enriched in prefrontal brain regions and in synaptic structure and neurotransmission, and single-nucleotide polymorphism heritability was 0.089. The phenotypes were broad and often self-declared, and 15 of the 102 variants were not replicated. How inherited risk combines with experience is taken up in What Causes Depressive Disorders.

Where a biological system matters mainly for one presentation, the Types of Depressive Disorder section points to that type's own entry instead of repeating the account here.

Brain Networks

Resting-state functional connectivity (rsFC) measures how strongly activity in different brain regions rises and falls together while a person is not performing a task. The closest thing to a direct synthesis of this literature is a meta-analysis that pooled 27 seed-based resting-state datasets from 25 publications, covering 556 people with major depressive disorder and 518 controls, with 91 seeds and 346 effects (Kaiser et al., 2015). The authors analysed these with multilevel kernel density analysis.

The pooled result was an imbalance across large-scale networks, with some connections weaker than in controls (hypoconnectivity) and others stronger (hyperconnectivity). The networks involved were the frontoparietal, default, dorsal attention, affective and ventral attention networks. The pattern for the ventral attention network was mixed (Kaiser et al., 2015). The default mode network is the same system the meta-analysis calls the default network.

Several caveats limit how far this can be read. Only seed-based studies were included, and the methods varied from study to study. Whether participants were scanned at rest or during a task was sometimes unclear. A trend in unmedicated patients did not reach significance (p = 0.09 and p = 0.12), so the analysis does not show that medication explains the findings or that it can be ruled out. Because few seed regions recurred across studies, the authors also warned against reading much into the lack of anatomical specificity (Kaiser et al., 2015).

Earlier single studies had pointed to increases within the default mode network. In a case-control resting-state scan using independent component analysis, 28 depressed participants showed raised default mode connectivity in the subgenual cingulate and thalamus relative to 20 healthy controls. The authors could not fully exclude a medication effect, since the patients had medication exposure and the controls were unmedicated (Greicius et al., 2007). A later study of 35 first-episode, untreated patients and 35 matched controls found a split along the front-to-back axis: connectivity in anterior regions (medial prefrontal cortex and anterior cingulate) was higher than in controls, while posterior connectivity (posterior cingulate and precuneus) was lower. Within the depressed group only, anterior connectivity correlated positively with depressive rumination and posterior connectivity correlated negatively with overgeneral autobiographical memory. These links are correlational, and the sample was young adults at a first episode (Zhu et al., 2012).

A large multisite resting-state consortium reported the opposite direction. Default mode connectivity was decreased in depression, in an analysis of 848 patients and 794 controls drawn from 17 sites. The reduction held only in recurrent depression and not in first-episode, drug-naive patients, and it was linked to medication rather than to illness duration (Yan et al., 2019). Set against the earlier reports of increases, this is a contrary finding, and the direction of default mode changes in depression remains disputed.

Whether subgroups of patients differ in their network patterns is a separate question, taken up in the section on heterogeneity and the dimensional debate.

Sleep and Circadian Rhythms

Disturbed sleep is among the most familiar companions of depression, and polysomnography, the laboratory recording of sleep, has shown that the disturbance has a recognisable shape. A narrative review describes depression as going with a disturbance of sleep continuity and with REM disinhibition (Riemann et al., 2020). Relative to controls, people with depression show higher REM density and more total REM time, together with a shorter REM latency and less slow wave sleep (Riemann et al., 2020). Those same authors found the clinical use of these REM measures only partly realised. REM suppression appeared to be linked to treatment response, but initial hopes that REM abnormalities would serve as markers of subtypes were not fulfilled (Riemann et al., 2020).

Insomnia is a symptom that cuts across diagnoses, which raises the practical question of direction. A meta-analysis of twenty-one longitudinal epidemiological studies, covering work from 1980 to 2010, found that baseline insomnia predicted later depression, with an odds ratio of 2.60 (CI 1.98 to 3.42) across all studies and 2.10 (CI 1.86 to 2.38) once outliers were adjusted for (Baglioni et al., 2011). Two cautions attach to that result. The studies were heterogeneous until the outlier adjustment made them homogeneous, and confounders were not always considered (Baglioni et al., 2011). Whether treating insomnia would prevent depression is, in the assessment of the narrative review above, still open (Riemann et al., 2020). Readers interested in the sleep problem itself can turn to the page on insomnia.

The body clock adds a second line of evidence. A narrative review of roughly 24-hour rhythms describes the link between circadian disruption and mood disorders as bidirectional, with exposures such as jet lag, night-shift work and artificial light at night in view (Walker et al., 2020). Its authors noted that direct interactions have been studied only recently, that the effects appear in susceptible individuals, and that the benefits of lifestyle changes are presumed rather than shown (Walker et al., 2020).

The sleep findings in depression are also not specific to it. A meta-analysis of controlled polysomnography studies, restricted to parameters examined in at least three studies, found sleep alterations in most of the psychiatric disorders it covered, spanning sleep continuity, sleep depth and REM sleep pressure, and no single parameter belonged to one disorder alone (Baglioni et al., 2016). The authors read this as consistent with a transdiagnostic imbalance of arousal. Two conditions showed no alterations: ADHD and seasonal affective disorder (Baglioni et al., 2016). That exception bears on the seasonal pattern taken up under Types of Depressive Disorder, since a seasonal presentation was not marked by the laboratory sleep changes seen elsewhere. The meta-analytic source was an abstract only, so the number of participants behind each comparison is not available here.

Measurement Validity and Everyday Life

Whether performance on laboratory tests says anything about how a person with a depressive disorder lives is a separate question from whether the performance is poor. A systematic narrative review addressed it for major depressive disorder, searching 488 articles published between 1980 and 15 October 2013 and retaining 10 studies of adults younger than 60 years (Evans et al., 2014). Those studies related deficits on neuropsychological tests to psychosocial functioning, and a broad association between neurocognitive deficits and functional impairment appeared. The authors rated the evidence as weak. The studies differed so much that a meta-analysis was not possible, each had methodological limitations, and the evidence base was small (Evans et al., 2014). The review also sampled only nongeriatric adults, so it leaves older adults unaddressed. The link between test scores and daily functioning is therefore plausible but not firmly established.

A different method measures the disorder as it unfolds in daily life rather than in a testing session. A systematic review of 19 studies of clinically depressed individuals, spanning 25 years, concluded that Experience Sampling Methodology is informative for depression research (Telford et al., 2012). Across the studies, six facets of major depressive disorder were examined, ranging from positive and negative mood and cortisol output to the response to antidepressants, functioning at work and genetic risk. The review noted methodological limitations in the existing studies (Telford et al., 2012).

Laboratory bias tasks raise a further problem, which is whether they measure anything stably. A theoretical review with two empirical studies found that standard attentional bias scores were unreliable, whereas a trial-level measure, the Trial-Level Bias Signal, was reliable (Rodebaugh et al., 2016). The demonstration used a small clinical sample of 24 patients with anxiety disorders, and the authors' own data served for demonstration only. The same review noted that most prior studies skip psychometrics, and its abstract urged that the reliability of all measures be reported. The practical consequence is that a null or inconsistent bias finding may reflect a noisy score rather than an absent bias, so findings from standard scores should be read with that in mind. The peak trial-level scores, however, showed only modest stability across sessions (Rodebaugh et al., 2016).

Co-occurring Conditions

Depression rarely arrives alone, and anxiety is its most frequent companion. In a Netherlands psychiatric cohort of 1,783 participants, 67% of those with a current depressive disorder also had a current anxiety disorder, and 75% had one at some point in life. Seen from the other side, 63% of those with a current anxiety disorder had a current depressive disorder, and 81% had one across their lifetime (Lamers et al., 2011). The same cohort showed a temporal pattern: anxiety came first in 57% of cases, depression first in 18% (Lamers et al., 2011). Participants with both conditions, compared with those with a pure disorder, reported more childhood trauma, higher neuroticism, an earlier age at onset, and greater symptom severity and duration (Lamers et al., 2011). The authors concluded that assessing both conditions routinely is advisable (Lamers et al., 2011).

The pattern is not confined to clinic samples. A national survey of 36,309 US adults (NESARC-III) linked anxiety disorders and substance use disorders to higher odds of major depressive disorder. The adjusted odds ratio was 5.7 for generalized anxiety disorder, 2.1 for specific phobia, 3.0 for any drug use disorder and 1.8 for alcohol use disorder (Hasin et al., 2018).

Why Anxiety and Depression Overlap

One account holds that the overlap reflects a shared core rather than a single condition. Clark and Watson reviewed psychometric evidence from patient and normal samples and argued for a tripartite structure. General distress is common to both, with no difference between anxiety and depression. Physiological hyperarousal is higher in anxiety, and anhedonia, the loss of pleasure, is higher in depression, tied to low positive affect. On this reading the two syndromes are separable despite their shared distress, and the authors proposed a mixed anxiety-depression diagnosis for the overlap (Clark & Watson, 1991).

A structural account goes further. Watson argued that mood and anxiety disorders form roughly one emotional class, with three subclasses (bipolar, distress and fear disorders), and that an empirically derived structure should replace the DSM-IV groupings. The placement of obsessive-compulsive disorder remained unclear, and the proposal called for further research (Watson, 2005). Both accounts are weighed against rivals in Rival Theories and Accounts.

Personality as a Shared Correlate

A meta-analysis of 175 studies (66 analyses, 851 effect sizes) found that every diagnostic group scored higher than controls on neuroticism, with a mean d of 1.65, and lower on conscientiousness, with a mean d of -1.01 (Kotov et al., 2010). Dysthymic disorder and social phobia stood out for lower extraversion than the other disorders. Effects were smaller in epidemiologic than in patient samples, and Eysenck's inventories gave weaker associations than the NEO scales (Kotov et al., 2010). The overlapping profiles suggest why disorders cluster, and What Causes Depressive Disorders returns to personality as a risk factor.

Physical Illness

Depression also travels with chronic physical disease. A cross-sectional World Health Survey of 245,404 adults in 60 countries used regression modelling to compare health scores. Depression produced a larger health decrement than angina, arthritis, asthma or diabetes, and people with depression plus a chronic disease fared worse than those with either alone (Moussavi et al., 2007). Because the data are cross-sectional, they describe how health states co-occur and do not show which condition came first.

Diabetes is the best-studied pairing. A meta-analysis of 42 studies of adults with type 1 or type 2 diabetes found that diabetes doubled the odds of comorbid depression (OR = 2.0, 95% CI 1.8-2.2). Prevalence varied with the method used, but the odds ratio stayed stable across those differences (Anderson et al., 2001). The direction of the link matters. In a meta-analysis of 20 prospective studies, baseline depression predicted incident type 2 diabetes strongly (RR = 1.60, 95% CI 1.37-1.88), while baseline diabetes predicted incident depression only modestly (RR = 1.15, 95% CI 1.02-1.30). The authors noted heterogeneity among the depression-to-diabetes studies, that only one database was searched, and that confounder adjustment varied (Mezuk et al., 2008).

In heart disease the evidence points to outcomes. An American Heart Association scientific statement, built on a systematic review of 53 studies and 4 meta-analyses, linked depression to worse outcomes after acute coronary syndrome. It noted heterogeneity across studies, along with varied depression measurement, follow-up length and covariates (Lichtman et al., 2014). The association is not shown to be causal.

Posttraumatic Stress Disorder

Major depression is common in people with current posttraumatic stress disorder. A meta-analysis of 57 studies covering 6,670 participants with PTSD put the rate at 52% (95% CI 48-56). Subgroup differences, with higher rates in military than civilian samples and after interpersonal trauma than natural disaster, emerged only after outlier removal, and the authors regarded the magnitude of the problem as uncertain (Rytwinski et al., 2013).

Suicide and Self-Harm

Death by suicide is the gravest outcome linked to depressive disorders, and the literature on it is as much about the limits of prediction as about risk. The strongest single association comes from psychological autopsy studies, which reconstruct a person's history after death. A systematic review of 76 of them, 54 case series and 22 case-control studies, found mental disorder to be the factor most strongly tied to suicide. Among the case-control studies, 90% of those who died had a mental disorder, against 27% of controls (Cavanagh et al., 2003). The same review found particular disorders and sociological variables insufficiently studied, and it concluded that prevention may work best through treatment.

Risk within affective disorders varies with where a person is treated. A meta-analysis of suicide-risk studies in three groups of patients with affective disorders found risk graded from suicidal inpatients, highest, through inpatients and a mixed group, to nonaffective patients, lowest. Affective disorders carried more risk than the general population. The authors judged that hospitalization itself may signal higher risk, that no diagnostic subtype or other factor was a reliable predictor of suicide, and that the often-quoted 15% figure had been overgeneralized (Bostwick & Pankratz, 2000).

Within depression itself, a systematic review of nineteen studies (28 publications) identified several correlates of suicide: male gender, family psychiatric history, previous attempt, severe depression, hopelessness, anxiety comorbidity, and alcohol or drug misuse. The review had fewer eligible studies than expected and did not examine how these factors depend on one another (Hawton et al., 2013). Prospective data temper the picture. A meta-analysis of 166 longitudinal papers (357 prediction cases) found that hopelessness and depression predicted later suicidality only weakly to moderately. Weighted odds ratios were 1.96 for ideation, 1.63 for attempt and 1.33 for death overall. Between-study heterogeneity was high, publication bias was evident, and estimates shrank after bias adjustment (Ribeiro et al., 2018).

A meta-analysis of 365 studies covering 50 years found that prediction of suicidal thoughts and behaviors stayed only slightly above chance across five broad categories of risk factor. Neither longer studies nor more recent ones predicted better, and the authors called for machine-learning risk algorithms (Franklin et al., 2017). Sorting patients into high-risk and lower-risk categories fares no better. Across 37 cohort studies of 315,309 psychiatric patients, 3,114 of whom died by suicide, the high-risk group had odds of suicide 4.84 times those of the lower-risk group (95% CI 3.79 to 6.20). Sensitivity was 56% and specificity 79%, and 5.5% of the high-risk group died by suicide, against 0.9% of the lower-risk group. Heterogeneity was very high and publication bias probably inflated the pooled odds, so reliable, strong risk stratification remains elusive (Large et al., 2016).

Evidence on suicide prevention points to several measures with no clear leader. A systematic review of 1,797 studies found that restricting means, school awareness programmes, and treatments such as clozapine and lithium helped. School programmes were linked to fewer attempts (OR 0.45) and less ideation. Evidence on primary-care screening was insufficient, and the review concluded that no single strategy stands out (Zalsman et al., 2016). Help is available: in the United States the 988 Suicide & Crisis Lifeline can be reached by call or text at 988, and the International Association for Suicide Prevention lists crisis centres in other countries.

Culture and Society

Depression is described, reported and treated within particular societies, and the evidence on it is not evenly spread across them. This section covers how its expression and measured frequency differ across cultures, what social circumstances are linked to it, and who is less likely to be treated.

Expression and Prevalence Across Cultures

Kleinman (2004) argued, in a commentary on Chinese society and on Chinese immigrants in the United States, that depression there is often experienced physically, and that somatic complaints outnumber reports of sadness. He also warned that the pattern may confound diagnosis, that it changes over time, and that a psychiatric diagnosis is morally unacceptable to many in these communities (Kleinman, 2004).

A clinical comparison put the claim to a test. Ryder et al. studied 175 Chinese and 107 Euro-Canadian outpatients with three methods: a spontaneous problem report, a structured clinical interview and a symptom questionnaire. The Chinese patients were more somatic and the Euro-Canadians more psychological. The somatic difference appeared in only two of the three methods, whereas the psychological difference was larger and more consistent. Externally oriented thinking mediated the somatic effect. A difference in emotion identification was not significant (Ryder et al., 2008). The finding therefore concerns what patients emphasise, and it is clearer for psychological expression than for somatic expression.

Reported frequency also varies between countries. Lifetime prevalence of major depressive episode differs widely across the World Mental Health surveys, and Kessler and Bromet (2013) reported that methodological studies found no evidence that this variation is due to methodological factors, and noted that direct data are missing for most countries (Kessler & Bromet, 2013). Across 18 countries, lifetime episodes were more common in high-income than in low- and middle-income countries, whereas 12-month rates were similar (Bromet et al., 2011). The figures themselves are given in the Prevalence section.

A wider critique bears on all of this. Henrich et al. (2010) reviewed cross-population comparisons and concluded that Western, educated, industrialised, rich and democratic (WEIRD) samples are unrepresentative and are frequent outliers. They found no a priori grounds for assuming universality, although their evidence covers only selected domains (Henrich et al., 2010). Applied to depression, the critique implies that symptom profiles and risk estimates from one kind of society cannot be assumed to hold in another, a caution that the thin direct data noted above reinforce.

Social Determinants

Poverty and mental illness appear to act on each other in both directions. A review of natural and controlled experiments found that cash transfers and low-cost therapy helped (Ridley et al., 2020). The review's constructs included depression and anxiety, and its design allows more than a correlational reading.

Social conditions reach beyond income. Drawing on 289 articles, a review of reviews sorted the social influences on mental disorders into five domains: demographic and economic factors, neighbourhood conditions, environmental events, and social and cultural circumstances. It linked them to the Sustainable Development Goals and suggested that optimising progress on those goals could aid prevention (Lund et al., 2018).

Discrimination is one such determinant. A meta-analysis of 293 studies (333 articles, 1983 to 2013, mostly from the United States) found that racism was associated with poorer mental health (r = -.23 for negative mental health) and poorer general health (r = -.13), physical health (r = -.09) and positive mental health (r = -.13). The associations were larger in cross-sectional than in longitudinal studies and larger in non-representative samples than in representative ones. The size of the association did not vary significantly with age, sex, birthplace or education level, but it did vary with ethnicity (Paradies et al., 2015). These are associations, and the predominantly American evidence limits how far they travel.

Minority Groups and Access to Care

Sexual minorities carry a higher burden. A systematic review of 199 studies found more depression, anxiety and suicide attempts among sexual minority people than among heterosexual people, with bisexual people at highest risk. The elevation was larger in more recent and in higher-quality studies. Alcohol problems were the exception, and in sexual minority men they were often equal or lower; the authors judged the risk proposition supported overall and in subgroups, though gender and subgroup differences were complex (Plöderl & Tremblay, 2015).

Minority stress is linked to depression in gender minority samples too. A meta-analysis of 85 cross-sectional studies of transgender and gender diverse individuals found that minority stress was positively associated with depression, suicidal ideation and suicide attempts. Distal stress, expectations of rejection, internalised transphobia and concealment each showed this pattern, and expectations of rejection and internalised transphobia had larger effect sizes than distal stress or concealment. Because every study was cross-sectional and the samples were neither diverse nor representative, the authors called for longitudinal designs and more representative samples (Pellicane & Ciesla, 2022).

Treatment differs by ethnicity as well. In pooled US household surveys of 8,762 adults from four ethnic groups, 1,082 were currently depressed, and Latino, Asian and African American respondents were each less likely than white respondents to have had treatment in the past year (p < .001). Of the adults with current depression, the share who had no treatment in the past year was 63.7% for Latinos, 68.7% for Asians and 58.8% for African Americans, compared with 40.2% for non-Latino whites. Adequate care was reached by 33.0% of whites, 22.3% of Latinos, 13.1% of Asians and 12.1% of African Americans. Among those who did receive treatment, the adequacy of Latino and Asian patients did not differ significantly from that of whites, so the gap lay in access, but African Americans were less likely to receive adequate care. The authors concluded that health care systems alone are unlikely to shift these disparities. The Latino difference was only marginally significant (Alegría et al., 2008).

Treatment Evidence

Major depression is common and severely disabling, and its presentations, course and response to treatment all vary (Malhi & Mann, 2018). The treatment literature therefore has to be read against that variability: pooled effects describe averages, and most of the trials behind them concern acute treatment of major depressive disorder. The subsections below set out the evidence for medication, for psychological therapy and for exercise, the dispute over drug-placebo differences, the debate over stopping antidepressants, what happens when a first treatment fails, what two national guideline bodies report, and how the evidence divides by type of depressive disorder.

Medication

The largest comparison of antidepressants is a network meta-analysis that drew on 522 trials and 116 477 participants, covering 21 antidepressants identified from 28 552 citations (Cipriani et al., 2018). Every drug outperformed placebo for acute treatment of major depressive disorder, and the drugs differed from one another only modestly. For efficacy, odds ratios against placebo ranged from 2·13 for amitriptyline to 1·37 for reboxetine; for acceptability, only agomelatine (OR 0·84) and fluoxetine (OR 0·88) had fewer dropouts than placebo, whereas clomipramine (OR 1·30) was worse than placebo (Cipriani et al., 2018).

The authors attached heavy qualifiers to this result. Credible intervals were wide for most comparisons, head-to-head results were more variable than drug-versus-placebo results, and certainty in the evidence ran from moderate to very low. Risk of bias was rated high in 46 trials (9%), moderate in 380 (73%) and low in 96 (18%), so most of the evidence base carried at least some concern (Cipriani et al., 2018). Treatment in children and adolescents is taken up in the Depressive Disorders Across the Lifespan section, and nothing in this section extends the adult comparison to youth.

Psychological Therapies

A network meta-analysis of 331 randomized trials with 34,285 depressed patients compared the main psychotherapies. All of them outperformed care-as-usual and a waiting list, and most outperformed pill placebo; the exceptions were supportive counseling and psychodynamic therapy, which did not differ from placebo (Cuijpers et al., 2021a). The authors concluded that the therapies are efficacious and acceptable, with few differences between types. Effect sizes ranged from SMD = –0.81 for life-review against care-as-usual to SMD = –0.32 for non-directive supportive counseling, and non-directive supportive counseling did worse than every other therapy. Differences among the remaining therapies were not significant, although problem-solving therapy held an edge over some others at long-term follow-up, which the authors regarded as somewhat higher than the other contrasts (Cuijpers et al., 2021a). No acceptability difference between therapies was consistent. The authors added that patient preference and the availability of a therapy may matter when choosing among them, and that better characterization of patients could refine the matching. Psychotherapy estimates come with a caution about publication bias: an earlier meta-analysis pooling 117 randomized trials (175 comparisons, 9537 adults) judged that this bias had inflated its effects, since adjustment with 51 imputed studies lowered the overall effect from d = 0.67 to 0.42, and adjustment with 26 missing studies lowered the cognitive-behavioural therapy effect from d = 0.69 to 0.49 (Cuijpers et al., 2010).

Whether to add a psychological therapy to medication was examined in a separate meta-analysis of 52 randomized trials with 3,623 patients, 32 of the studies concerning depressive disorders and 21 anxiety disorders (Cuijpers et al., 2014). Combined treatment beat pharmacotherapy alone (Hedges' g = 0.43, 95% CI 0.31-0.56, NNT 4.20), and the advantage held in major depression as well as in panic disorder and obsessive-compulsive disorder. Against placebo, the combination produced roughly twice the effect of pharmacotherapy. The effects of the drug and of the psychotherapy were independent of one another, and each contributed about equally to the combined effect. The authors' conclusion was hedged: combined treatment is probably better than drugs alone, and monotherapy may not be optimal (Cuijpers et al., 2014).

Exercise

Physical activity has its own network meta-analysis, pooling 218 studies and 14 170 participants. Exercise lowered depression with moderate effects and outperformed active controls; walking or jogging (g −0.62), yoga (g −0.55) and strength training (g −0.49) did better than other forms, and effects grew with the intensity prescribed (Noetel et al., 2024). The authors placed exercise alongside psychotherapy and medication, but the certainty behind that conclusion is weak: only one study was rated at low risk of bias, confidence in the estimates was low or very low, and expectancy effects mean better-blinded trials are needed.

Severity and the Placebo Debate

A meta-analysis of FDA trial data, covering 35 trials and 5,133 patients (3,292 randomized to medication and 1,841 to placebo), found that the gap between drug and placebo was tied to baseline severity (Kirsch et al., 2008). The mean gap was 1.80 points on the Hamilton Rating Scale of Depression (d = 0.32), below the NICE criterion for clinical significance, which corresponds to d = 0.50. The difference reached that criterion only at the extremes of severity, with baseline scores above 28, and even there the size-by-severity interaction was only marginal (p = 0.0586).

The same paper supplies its own limits. The growing gap with severity reflected a decline in placebo response, not a gain in drug response. The curvilinear pattern rested on one trial, no trials in the severe range existed, standard deviations were imputed for some trials, and post-approval trials were left out (Kirsch et al., 2008).

Set against the later network meta-analysis, the two sources address different questions. Kirsch and colleagues examined how large the drug-placebo gap is and how it varies with baseline severity in 35 trials (Kirsch et al., 2008); Cipriani and colleagues, with 522 trials, found every antidepressant more effective than placebo for acute treatment, while rating the certainty of much of that evidence as moderate to very low (Cipriani et al., 2018). The two do not contradict each other on direction. They leave open how much the average difference matters clinically, and the earlier dataset's omission of post-approval trials is one reason to read its size estimate with care.

Stopping Antidepressants: Discontinuation and Withdrawal

How often stopping an antidepressant produces symptoms, and how bad they are, is an open dispute. One side rests on a meta-analysis of 79 studies and 21 002 patients: discontinuation symptoms, including severe ones, occurred more often after an antidepressant than after placebo, and the authors put the net figure at approximately 15%, about one in six to seven patients, which they presented as the probable extent of discontinuation symptoms (Henssler et al., 2024). Heterogeneity was substantial, and some reported symptoms may reflect residual or re-emerging psychopathology rather than discontinuation itself.

The other side comes from a systematic review of 24 studies, which described withdrawal as common, severe and often prolonged, with incidence figures ranging from 27% to 86% (Davies & Read, 2019). Its authors argued that guidelines calling withdrawal self-limiting and lasting about two weeks probably misdiagnose it and need urgent revision, although the studies differed widely in method, sample size and population, and only four gave a mean duration.

Neither review settles the matter. The Canadian guideline discussed below reports no consensus on how to stop, rates the evidence on stopping as low certainty and finds the evidence for hyperbolic tapering insufficient (Lam et al., 2024). What remains open is how frequent and how lasting withdrawal is once relapse and non-specific effects are separated out, and which way of stopping is safest.

Sequenced Treatment

In STAR*D, 3,671 outpatients with nonpsychotic major depression passed through successive treatment steps, with symptoms rated on the self-report QIDS-SR(16) and on the 17-item Hamilton scale, HRSD(17) (Rush et al., 2006). Remission rates fell as patients needed more steps: 36.8% remitted at the first step and 30.6% at the second, against 13.7% and 13.0% at later steps. Relapse over 12 months rose in step, so that patients who needed more steps were more likely to relapse. Among those who remitted, the relapse advantage appeared only across steps one to three. The authors noted that the best treatment sequences remain unknown (Rush et al., 2006).

Depression that persists through repeated treatment is treated as its own category; it is described under Types of Depressive Disorder (treatment-resistant depression).

What Clinical Guidelines Recommend

In Canada, the 2023 update of the CANMAT guideline for adults with major depressive disorder describes care that is personalized and stepwise, with treatment decisions shared between clinician and patient (Lam et al., 2024). It lets severity guide the first choice of treatment, weighs patient preference, availability and past response, and reports that shared decision-making improved knowledge and satisfaction over usual care without changing adherence or symptoms. The authors present the document as guidance for clinicians rather than an established standard of care, and they note that many of the underlying meta-analyses rest on weak original studies.

In the United States, the US Preventive Services Task Force, drawing on a systematic review it commissioned, issued a recommendation statement on screening asymptomatic adults aged 19 years or older, including pregnant and postpartum people (US Preventive Services Task Force, 2023). It found a moderate net benefit from screening for major depressive disorder, gave it a B recommendation at moderate certainty, and favoured screening, while issuing an I statement for screening for suicide risk because the evidence there was insufficient.

Treatments by Type

Table 3 lists only those types for which a cited work names an approach. The evidence is thinnest outside major depressive disorder. For persistent depressive disorder, which is often more disabling than episodic major depression and whose dysthymic form can be difficult to detect until a major episode is superimposed (Schramm et al., 2020), a guidance paper built on a systematic search and expert consensus found combined medication and psychotherapy best and treated it as first-line, while noting that chronic depression responds less well to psychotherapy than episodic depression and that it is unclear which patients benefit (Jobst et al., 2016). Pooling randomized trials in a network meta-analysis (45 of drugs, 15 of psychotherapy), one review found several medications better than placebo but a mixed record for psychotherapy: interpersonal therapy alone did worse than medication and worse than CBASP, adding it to medication helped in chronic major depression but not in dysthymia, and several treatments rested on a single trial (Kriston et al., 2014).

For premenstrual dysphoric disorder, a scoping review of 113 articles found the best evidence for selective serotonin reuptake inhibitors and combined oral contraceptives, with hormonal suppression and surgery reserved for refractory cases; study quality varied and placebo responses reached up to 50% (Carlini et al., 2022). For perinatal depression, a meta-analysis of 43 randomized trials and 6270 participants found that psychotherapy beat control conditions with a moderate-to-large effect (g = 0.67, NNT = 4.39), probably lasting 6 to 12 months, alike for antenatal and postnatal depression, though heterogeneity was high and some publication bias was present (Cuijpers et al., 2023).

Table 3

Treatment Evidence by Type

TypeEvidence-based approach named in the articleSection
Major depressive disorderAntidepressants (Cipriani et al., 2018); psychotherapies (Cuijpers et al., 2021a)Medication; Psychological Therapies
Persistent depressive disorderCombined medication and psychotherapy (Jobst et al., 2016); several drugs above placebo, psychotherapy results mixed (Kriston et al., 2014)Treatments by Type
Premenstrual dysphoric disorderSelective serotonin reuptake inhibitors and combined oral contraceptives (Carlini et al., 2022)Treatments by Type
Perinatal depressionPsychotherapy, above control conditions (Cuijpers et al., 2023)Treatments by Type
Seasonal affective disorderCognitive behavioral therapy adapted for SAD, more durable than light therapy by the second winter (Rohan et al., 2016)Types of Depressive Disorder
Treatment-resistant depressionKetamine and esketamine, with rapid-onset efficacy; safety and place in treatment algorithms uncertain (McIntyre et al., 2021)Types of Depressive Disorder

Note. Entries reflect only the works cited in this article, and a type not listed is a gap in that evidence, not a statement that no treatment exists. The seasonal row comes from a randomized trial of 177 community adults with recurrent winter seasonal affective disorder; the two groups did not differ in the first winter (Rohan et al., 2016). The treatment-resistant row comes from an expert consensus review that also notes comparative effectiveness is uncertain (McIntyre et al., 2021). The premenstrual and perinatal rows carry the quality caveats given in the text above.

How Treatments Are Thought to Work

Whether a treatment helps is a separate question from why it helps, and the second has proved harder to answer. For antidepressant drugs, a narrative review combining behavioural and functional magnetic resonance imaging work reported that the drugs change the processing of emotional information earlier than they change mood, shifting a negative emotional bias in limbic and prefrontal circuitry. The authors judged this possibly consistent with cognitive theories, while noting that challenges remain in establishing it (Harmer et al., 2009).

For psychotherapy, the question is which psychological mediators carry the benefit. A systematic review of 35 mediator studies concluded that demonstrating causal mediation is probably the field's biggest challenge. It described the research as heterogeneous and methodologically unsatisfactory, and noted that the causal link between a mediator and symptoms is hard to show. It also suggested that single-mechanism models are likely too simple and that several mechanisms, at several levels, may operate together (Lemmens et al., 2016).

Behavioural activation illustrates the difficulty at the level of one therapy. A systematic review with narrative synthesis covered 14 prospective treatment studies and 21 mediators. Activation and environmental reward, the processes most often proposed, each showed only weak evidence as mediators of outcome. The authors drew no firm conclusions on any mediator, and they cited non-significant results, poor methodological quality in some studies and differing questionnaires as limits (Janssen et al., 2021).

The standing of the field is therefore open. The evidence for an early emotional-processing effect of antidepressants is suggestive, and for psychotherapy no mediator has been firmly established.

Brain Stimulation and Other Biological Treatments

Biological treatments beyond conventional medication act on the brain by electrical, magnetic or pharmacological routes. Evidence for each differs in maturity, and the three strands below are best read separately.

Electroconvulsive therapy has the longest evidence base. A systematic review and meta-analysis by the UK ECT Review Group pooled randomized trials and found it effective for depressive illness in the short term, with an advantage over both simulated ECT and drug treatment (UK ECT Review Group, 2003). The pooled comparisons ran from six trials with 256 patients to 22 trials with 1408 participants. The review also qualified its own conclusions. Only short-term efficacy was pooled, superiority over drugs was judged only probable, and the advantage of bilateral over unilateral electrode placement was moderate (UK ECT Review Group, 2003). How the treatment works is still not completely understood, and a narrative review asks whether changes in the applied electricity, in the resulting seizure, or both are responsible for the improved safety of modified ECT techniques (Deng et al., 2024).

Cognitive side effects are the main concern attached to the treatment. A meta-analysis of 84 studies and 2981 depressed patients examined 24 cognitive variables, including processing speed, working memory, anterograde memory and executive function. It concluded that the cognitive abnormalities were mainly limited to the first three days after treatment (Semkovska & McLoughlin, 2010). Moderators had no effect once that window had passed. The authors flagged three limits: no standardized tests of retrograde amnesia were available, only within-subjects studies were included, and the literature search ended in January 2009 (Semkovska & McLoughlin, 2010).

Magnetic stimulation is the less invasive alternative. A network meta-analysis of 81 studies and 4233 patients, with a mean age of 46 years and 59.1% women, compared protocols of transcranial magnetic stimulation delivered as repetitive pulses (rTMS) with sham stimulation (Brunoni et al., 2017). Active rTMS outperformed sham, and the odds ratios ranged from 2.37 for low-frequency stimulation to 4.66 for priming low-frequency stimulation. Bilateral stimulation (3.96), high-frequency stimulation (3.07) and θ-burst stimulation (2.54) fell between those values (Brunoni et al., 2017). The newer protocols did not show the same benefit. Accelerated, synchronized and deep rTMS were not effective, and the θ-burst remission result was not significant (Brunoni et al., 2017). The authors described their results as imprecise, because some interventions had few trials, and called for larger randomized trials (Brunoni et al., 2017).

Ketamine and esketamine form the third strand. An expert consensus review found rapid-onset efficacy for both in treatment-resistant depression (McIntyre et al., 2021). The same review noted areas of consensus alongside safety and tolerability concerns, an uncertain place for these drugs in treatment algorithms and unsettled comparative effectiveness, so research gaps remain (McIntyre et al., 2021). Several mechanisms have been proposed for ketamine's action, and they are not mutually exclusive, though the mechanistic work reviewed is preclinical (Zanos & Gould, 2018). Separate scrutiny has addressed esketamine specifically, with one article setting out seven concerns about its efficacy and about its FDA approval (Turner, 2019).

Emerging Treatments

Two newer approaches have been tested in randomised trials: the psychedelic compound psilocybin, given with psychological support, and smartphone applications. The evidence so far is early, and the trials themselves call for larger and longer studies.

Psilocybin

The double-blind phase 2 trial lasted six weeks and enrolled 59 patients whose major depressive disorder was of moderate to severe intensity and had lasted a long time. It set two 25 mg doses of psilocybin, given three weeks apart, against two 1 mg doses of psilocybin alongside a daily course of escitalopram (Carhart-Harris et al., 2021). The primary outcome was the change in the 16-item self-rated QIDS score (QIDS-SR-16). It fell by 8.0 points in the psilocybin group and 6.0 in the escitalopram group, a difference of 2.0 points that was not significant (95% CI, -5.0 to 0.9; P = 0.17) (Carhart-Harris et al., 2021).

The secondary outcomes leaned toward psilocybin. Response reached 70% against 48%, although the 22-point gap had a confidence interval that included zero. Remission reached 57% against 28% (Carhart-Harris et al., 2021). The authors did not correct the 16 secondary outcomes for multiple comparisons, and the patients were a selected group, so these differences are best read as hypotheses. Adverse events were similar across the two arms (Carhart-Harris et al., 2021).

A separate phase 2 double-blind trial gave a single dose to 233 adults with treatment-resistant depression. By week 3, the 25 mg group (n = 79) had improved by 12.0 points on the Montgomery-Åsberg Depression Rating Scale, against 5.4 points for the 1 mg group (n = 79). The 10 mg group (n = 75) improved by 7.9 points, and its difference from 1 mg was not significant (P = 0.18) (Goodwin et al., 2022). The authors concluded that 25 mg helped, but adverse effects occurred, suicidal ideation appeared in all dose groups, and a sustained response at 12 weeks was not supported (Goodwin et al., 2022). They also noted that larger and longer trials, and comparison with existing treatments, were needed (Goodwin et al., 2022).

Smartphone Applications

A meta-analysis pooled 18 randomised trials of 22 smartphone apps, with 3,414 participants in all. Depressive symptoms fell more with an app than with a control, a small-to-moderate effect (g = 0.383, 95% CI 0.24 to 0.52, p < 0.001), and heterogeneity was substantial (I² = 74.0%) (Firth et al., 2017). The size of the effect depended on the comparison group. Against inactive controls it was g = 0.558, and against active controls it was smaller (g = 0.216) (Firth et al., 2017).

Several limits temper the result. Only apps that had been tested in randomised trials were included, most studies blinded participants inadequately, and the subgroup analyses were post hoc. The effect was significant in mild-to-moderate depression, whereas the analyses of major depressive disorder and bipolar disorder were underpowered and not significant. Which components of an app do the work remained unclear (Firth et al., 2017). The authors presented the findings as an early indication of efficacy (Firth et al., 2017).

Prevention

Preventing Onset

Preventive psychological interventions reduce the occurrence of new depressive disorders in people at high risk. A meta-analysis of 50 randomized trials, covering 14,665 participants of all ages, found a relative risk of 0.81 (95% CI: 0.72-0.91) for incident depression at one year, a 19% lower chance than in control conditions. With a control event rate of 30%, that effect translates to a benefit for roughly one person in twenty-one. The trials were of relatively high quality, but they enrolled high-risk groups only, so the result has not been shown for the general population (Cuijpers et al., 2021b). The authors read this as promising alongside treatment rather than as a replacement for it.

Targeting is easier when a visible warning sign exists. A meta-analysis of 16 longitudinal cohort studies (n = 67,318) found that subthreshold depression carried a higher incidence of later major depression than no depression, with an incidence rate ratio of 1.95 (95% CI 1.28-2.97). The ratio did not differ significantly by age group or by sample type, which favours indicated prevention at any age and in any setting (Lee et al., 2019).

Universal programmes have a weaker record. A cluster randomized trial placed 84 schools and 8,376 students aged 11 to 14 years in either school-based mindfulness training, delivered as 10 lessons, or teaching as usual. At one year the two arms did not differ on the co-primary outcomes: the standardized mean difference was 0.005 (95% CI -0.05 to 0.06) for depression and 0.02 (-0.03 to 0.07) for well-being. Superiority over usual teaching was therefore not supported. Students and staff were not blind to allocation, and home practice was low (Kuyken et al., 2022).

Preventing Relapse

Stopping an antidepressant after recovery raises the chance of a return of illness. In a double-blind placebo trial in 150 UK general practices, 478 adults in primary care were randomized either to continue their medication (238) or to discontinue it (240). Over 52 weeks, 39% of the maintenance group relapsed (92 of 238), against 56% of the discontinuation group (135 of 240), a hazard ratio of 2.06 (95% CI 1.56 to 2.70; P<0.001). Adherence was lower in the discontinuation group, at 52% against 70%, which qualifies the comparison (Lewis et al., 2021).

Psychological treatment offers a second route. Mindfulness-based cognitive therapy (see mindfulness-based cognitive therapy) was examined in an individual participant data meta-analysis of 9 studies and 1,258 patients with recurrent depression. Over 60 weeks, relapse risk was lower than in comparators that were not mindfulness-based (hazard ratio 0.69, 95% CI 0.58-0.82) and lower than with active treatments (hazard ratio 0.79, 95% CI 0.64-0.97). Age, sex, education, relationship status, age of onset and number of prior episodes did not moderate the effect. Higher baseline severity was associated with a larger benefit, though the authors described that evidence as limited, and full patient data were missing for one study. They judged the therapy probably efficacious (Kuyken et al., 2016).

The wider class of psychological interventions shows a similar pattern. An individual participant data meta-analysis of 14 randomized controlled trials (N = 1,720) found a lower relapse risk at 12 months than with treatment as usual, medication or evaluation only (hazard ratio, 0.60; 95% CI, 0.48-0.74). Benefit was larger for patients with three or more previous episodes, and the authors noted that lower-risk patients may need less intensive approaches. They recommended adding psychological interventions to treatment (Breedvelt et al., 2024).

Suicide Prevention

Strategies aimed at suicide, such as means restriction and school programmes, are reviewed under Suicide and Self-Harm, where the evidence for each is set out (Zalsman et al., 2016).

Living with a Depressive Disorder

Depressive disorders reduce quality of life, and the size of the reduction depends on the diagnosis. Pooled baseline data from 11 treatment trials, measured with the Quality of Life Enjoyment Satisfaction Questionnaire, put clinically severe impairment at 85% among patients with chronic or double depression, against 63% in major depressive disorder and 56% in dysthymic disorder (Rapaport et al., 2005). The depressive disorders were more impaired than panic disorder (20%), and the same comparison placed them above obsessive-compulsive disorder and social phobia. Two limits apply: the samples were people who entered clinical trials, and illness-specific symptom scales accounted for only a small to modest share of the variance in baseline quality of life. Perceived quality of life therefore adds information that symptom counts do not capture (Rapaport et al., 2005).

Work is a second area of loss. A review of studies published since 2002 found that depressed people had more unemployment, more absences and more deficits while at work than non-depressed people, and that greater symptom severity went with greater work impact. Symptom relief improved work outcomes only partly, and differences in method between studies obscured how large the effects were (Lerner & Henke, 2008). A secondary analysis of an online survey of about 1,000 employees in each of eight countries found productivity costs of depression in every one. Costs of presenteeism exceeded those of absenteeism by 5 to 10 times. The diagnoses were self-reported, the panel had low response rates, and the costs for China were probably underestimated (Evans-Lacko & Knapp, 2016).

Stigma can be turned inward. In a mail survey of 1,182 members of bipolar or depression organisations in 13 countries, 21.7% reported moderate or high self-stigma, roughly 1 in 5. Empowerment, perceived discrimination, areas of social contact, education and employment each predicted self-stigma, and 71.6% reported moderate or high perceived discrimination. The sample was unweighted and drawn from charity members, so it may not represent people with depression generally (Brohan et al., 2011). The authors suggested that interventions tailored to self-stigma may help. Stigma also bears on whether people seek help: a systematic review of 144 studies with 90,189 participants found a small negative association between stigma and help-seeking (d = -0.27) (Clement et al., 2015). The review noted that the association shrank to d = -0.18 when low-quality studies were excluded, and that the studies were mostly cross-sectional and drawn largely from the USA and Canada (Clement et al., 2015). The wider public side of the problem is treated under social stigma.

The burden extends to relatives. A meta-analysis of 16 studies, covering 18 interventions and 2,178 caregivers, found that non-pharmacological interventions reduced caregiver depression, and anxiety followed a similar pattern. The benefit was mainly short term and dissipated over time, methodological quality was moderate, and the moderator analysis was not significant. Caregiving outside dementia was under-researched (Lambert et al., 2021). The strain on such carers is discussed further under caregiver burden.

Three neighbouring conditions press on the boundary of depressive disorder: the low mood that follows a death, the depressive pole of bipolar illness, and the anxiety disorders. Each raises the same question of whether a distinct condition or a variant of depression is being described.

Grief and Depression

The diagnostic manual once told clinicians not to diagnose a major depressive episode when the symptoms followed a recent death. A review that added studies published after 2006 to earlier reviews concluded that bereavement-related depression resembles other major depressive episodes and recommended that the exclusion be dropped from DSM-5 (Zisook et al., 2012). A critical literature review reached the opposite verdict, holding that the exclusion was valid and should be retained in DSM-5 (Wakefield & First, 2012). Its authors noted that the two 2007 reviews drew mostly on samples that had not been excluded, and that three later studies were inconclusive. They nonetheless proposed changes meant to reduce false negatives (Wakefield & First, 2012).

The dispute concerns where depression ends and an expectable response to loss begins. In the event, the bereavement exclusion was replaced in DSM-5 by clinical judgment. A review of the DSM-5 criteria changes observed that this makes the diagnosis less objective and complicates research on adversity and depression (Uher et al., 2014).

A separate line of work treats persistent grief as its own condition, and prolonged grief disorder was added in DSM-5-TR. Its criteria were tested with the PG-13-Revised scale in three bereaved samples (Yale, Utrecht and Oxford), and the criteria and scale were judged valid (Prigerson et al., 2021). The condition centres on attachment disturbance and yearning for the deceased. The diagnosis was distinct from major depressive disorder (phi = 0.25) (Prigerson et al., 2021). The article on grief covers the wider bereavement literature.

Unipolar and Bipolar Depression

A narrative review of bipolar and unipolar depression found broad parallels between them. Severity did not differ, and dopamine and serotonin findings were similar in the two groups (Cuellar et al., 2005). Differences appeared too. Bipolar depression began at a younger age, with onset about 6 years earlier, and it brought more frequent episodes and greater mood variability. Unipolar depression showed more anxiety, activity and somatization, whereas anhedonia was higher in the bipolar group (Cuellar et al., 2005). The review took its direct comparisons to be sparse and advised that its conclusions be treated with caution. It raised the possibility that mania and depression are separate comorbid syndromes (Cuellar et al., 2005).

Mixed features mark the border between the two conditions in practice (see With Mixed Features). An editorial argued that the DSM-5 mixed-features criteria are scientifically invalid and that agitation-based criteria would be preferable. By its account, the DSM-5 definition identifies mixed presentations in 0 to 12% of cases, against 33 to 47% for a definition that includes agitation (Koukopoulos et al., 2013). That paper is an opinion piece and reports no new data. The distinction from bipolar disorder therefore rests partly on how mixed presentations are defined.

Anxiety Disorders

Anxiety disorders are the commonest companions of depression and are treated under Co-occurring Conditions. As a point of difference, the narrative review above found more anxiety in unipolar than in bipolar depression (Cuellar et al., 2005).

Worked Example: Two People, One Diagnosis

The two people below are invented for illustration. They come from no study, and the symptom lists are constructed arithmetic, not data.

Person A reports depressed mood, a change in weight or appetite, insomnia, psychomotor change and fatigue. Person B reports diminished interest or pleasure, a change in weight or appetite, feelings of worthlessness or guilt, poor concentration and thoughts of death. Both report five of the nine listed symptoms and both meet the remaining requirements, so each receives a diagnosis of major depressive disorder. The only symptom they share is the weight or appetite change.

The example is extreme but not impossible, because the definition is polythetic: many different sets of symptoms satisfy it. Counting the ways to meet the criteria gives 227 possible combinations. In interviews with more than 1,500 patients, 170 of those combinations occurred and 57 did not, so about a quarter were absent. The most frequent was the full set of all nine criteria (10.1%, n = 157), and nine combinations together accounted for over 40% of diagnoses (Zimmerman et al., 2015). Variation was wide, though some combinations co-occurred more than chance would predict.

Variation of this kind is common in practice. A secondary analysis of trial data on 3,703 outpatients found 1,030 unique symptom profiles, and 864 of them (83.9%) were each shown by five or fewer people. Another 501 profiles belonged to one individual apiece. Even with a subsample of median symptom count, heterogeneity did not shrink. The authors questioned whether the diagnosis is a consistent syndrome (Fried & Nesse, 2015).

A specifier might seem to separate the two people. In a secondary analysis of baseline data from 3,717 outpatients, however, the melancholic specifier did not significantly reduce the number of distinct symptom profiles, and the atypical specifier did not either. The authors found no evidence that specifiers make groups more homogeneous (Lorenzo-Luaces et al., 2021). Whichever label Person A or Person B received, each would sit in a group that was still varied.

A dimensional description would not ask whether each person has the disorder. It would locate each along symptom dimensions, so the two profiles would register as different positions on the same spectrum. This is the approach of the Hierarchical Taxonomy of Psychopathology, which a narrative review judged better than traditional categories on boundaries and comorbidity. The review called it promising but still a work in progress, with only some domains ready for clinical use (Kotov et al., 2017).

Discussion

Some conclusions about depressive disorders are secure. The disorders are common and heavily burdensome, as the survey and burden figures in Prevalence, Course and Burden show. Explicit criteria made diagnosis reproducible enough for most clinical purposes: in the DSM-5 field trials, reliability across the tested diagnoses was mostly good to very good in clinical populations assessed with usual interview methods, although six diagnoses fell in a questionable band and three in an unacceptable one (Regier et al., 2013). Inherited risk is real and diffuse (Adams et al., 2025), and the early candidate genes did not hold up when preregistered analyses tested 18 of them in samples of up to 443,264 people (Border et al., 2019). Effective treatment exists in each major class. Every one of 21 antidepressants outperformed placebo in acute treatment of major depressive disorder, with certainty that ranged from moderate to very low (Cipriani et al., 2018), and the main psychotherapies outperformed care-as-usual and waiting lists, with few differences between them (Cuijpers et al., 2021a).

Whether depression is one disorder or many is the largest open question. Across psychopathology, a quantitative review of taxometric studies found most latent variables dimensional and judged early taxonic findings likely spurious (Haslam et al., 2012). Symptom-based subtyping produced no conclusive subtypes, with latent classes tracking severity more than distinct symptom clusters (van Loo et al., 2012). The best-known biological subtyping, four fMRI biotypes, was not reproduced in a methodological replication, although that study changed the clinical instrument and the preprocessing pipeline and so was not a direct test (Dinga et al., 2019). The present position is a dimension-leaning evidence base read through categorical manuals. Direct replications of each subtype or biotype solution in independent samples, using the original methods, together with prospective tests of whether a subgroup predicts course or treatment response beyond severity, would settle it. The same kind of longitudinal follow-up bears on the contested types and specifiers described in Types of Depressive Disorder, such as disruptive mood dysregulation disorder and the mixed-features criteria.

The monoamine question has narrowed rather than closed. An umbrella review found no consistent association between serotonin and depression (Moncrieff et al., 2023), and a critical commentary answered that the null conclusion was overstated, citing receptor imaging and tryptophan depletion work while noting wide confidence intervals in the latter (Jauhar et al., 2023). What survives both papers is that no simple serotonin deficit is supported, while the part serotonin plays is still argued. Studies of drug-naive patients that separate illness from treatment effects would bear most directly on it.

Whether cognitive biases cause depression is similarly unsettled. The experimental test, cognitive bias modification, produced small effects on depression that were no longer significant after adjustment for publication bias (Cristea et al., 2015), and the reanalysis set out in Testing Causal Claims About Cognition argues that the average hides procedures that failed to move the bias. Causal roles remain plausible but unproven; trials that confirm the bias has changed before measuring later mood, in clinical samples, are the design the question needs. Suicide prediction is the most sobering case. Across 365 studies spanning 50 years, prediction of suicidal thoughts and behaviours stayed only slightly above chance and did not improve over time (Franklin et al., 2017). Prospective studies that combine many risk factors, which the existing literature rarely did, are the step its authors called for.

Two limits run through every section. Most findings come from samples with major depressive disorder, so their reach to the other depressive disorders is an inference, and much of the evidence comes from high-income, Western populations. A plain-language guide to recognising the condition is on the depression symptoms page.

Current Directions

Six lines of work are moving fastest, and each is still short of settled. The first is genetic. A trans-ancestry GWAS meta-analysis pooled 109 cohorts from 29 countries, with 688,808 people with major depression and 4,364,225 controls, and found 697 associations at 635 loci, of which 293 were novel (Adams, 2025). The associated genes were enriched in neuronal cell types and among antidepressant targets, which led the authors to suggest that genetics may guide drug targets and repurposing. The caveats are the sample, which was 76.6% European ancestry, and polygenic scores that transferred poorly to African-ancestry samples, so it remains unclear whether genetic architecture differs by ancestry. What Causes Depressive Disorders and Biology of Depressive Disorders place these results among the other biological findings.

A second line asks whether depression can be divided by brain circuit rather than by symptom count. Clustering fMRI measures of circuit function, among them task-free connectivity in the default mode, salience and frontoparietal attention circuits, in 801 patients not receiving treatment produced six biotypes, and these groups differed in symptoms, cognitive test performance and response to treatment (Tozzi et al., 2024). Treatment-response differences were tested in a trial that randomised 250 of the patients. The authors call this a promising route to precision psychiatry. It rests on one derivation-plus-trial report, so it is best read as a hypothesis to be replicated, and the sections on brain networks and heterogeneity take up what circuit findings can and cannot say about heterogeneity.

Third, some researchers propose replacing categories with a hierarchy of dimensions. The Hierarchical Taxonomy of Psychopathology (HiTOP) arranges signs and symptoms into syndromes, spectra and subtypes, and a narrative review argued that its dimensions address the unclear boundaries and heavy comorbidity of traditional categories (Kotov et al., 2017). The same review called HiTOP a work in progress, with only some domains ready for clinical use. A later commentary answering critics (Haeffel et al., 2022) judged many of their concerns unwarranted while conceding some real current limits, which the HiTOP consortium is working to address (DeYoung et al., 2022). The dispute is therefore open; How Depressive Disorders Are Classified and One Disorder or Many? give its background.

Fourth, prevention is receiving more attention, for first onset and for relapse. A meta-analysis of 50 randomised trials with 14,665 participants from high-risk groups found that preventive psychological interventions cut incident depression by 19% (RR = 0.81, 95% CI 0.72-0.91), judged by diagnostic interview (Cuijpers et al., 2021b). Its evidence concerns high-risk groups only. For relapse, an individual participant data meta-analysis of 14 trials (N = 1,720) found that a psychological intervention lowered relapse over 12 months compared with treatment as usual, medication or evaluation only (hazard ratio 0.60, 95% CI 0.48-0.74) (Breedvelt et al., 2024). The benefit was larger for people with three or more previous episodes, and the authors noted that lower-risk patients may need less intensive approaches. Prevention covers this evidence in full.

Fifth, digital delivery is being tested. A meta-analysis of 18 randomised trials of 22 smartphone apps (N = 3,414) found a small-to-moderate reduction in depressive symptoms relative to controls (g = 0.383, 95% CI 0.24-0.52), with substantial heterogeneity (I² = 74.0%) (Firth et al., 2017). The effect was larger against inactive controls (g = 0.558) than against active controls (g = 0.216), and the analyses of major depressive disorder were underpowered (g = 0.085, not significant). Most studies blinded participants inadequately, and which app components work is unclear. Treatment Evidence and Emerging Treatments set these findings beside other therapies.

Sixth, psychedelic trials have reached phase 2. In a double-blind randomised trial of a single dose of psilocybin in 233 adults with treatment-resistant depression, the 25 mg dose lowered Montgomery-Åsberg scores more than the 1 mg dose at week 3 (change of -12.0 against -5.4; difference -6.6, 95% CI -10.2 to -2.9) (Goodwin et al., 2022). The 10 mg dose did not differ significantly from 1 mg (P = 0.18). Sustained response at 12 weeks was not supported, and suicidal ideation and other adverse effects occurred in all dose groups. The authors called for larger, longer trials and for comparison with existing treatments, a point developed in Emerging Treatments.

Common Misconceptions

Depression is simply a chemical imbalance of serotonin.
An umbrella review of 17 reviews found no consistent association between serotonin and depression (Moncrieff et al., 2023). A commentary disputed how strongly that null was stated, so the role of serotonin is still argued (Jauhar et al., 2023), but no single mechanism explains major depressive disorder (Otte et al., 2016). The monoamine account can explain many of the actions of antidepressant drugs, a narrower claim than an account of depression, in which genetic, stress and psychosocial factors also play a part (Belmaker & Agam, 2008).
Everyone with a depressive disorder has the same symptoms.
Among 3,703 STAR*D outpatients there were 1,030 unique symptom profiles, and the most common covered only 1.8% of the sample (Fried & Nesse, 2015). Because the criteria are polythetic, 227 combinations qualify, and 170 of them occurred in more than 1,500 patients (Zimmerman et al., 2015). A single diagnostic label suggests a single picture; the rule behind it does not require one.
Scientists have found the gene for depression.
Preregistered analyses of 18 candidate genes in samples of 62,138 to 443,264 people found no support for any of them, and the authors judged earlier positive reports likely false positives (Border et al., 2019). A meta-analysis of 31 data sets likewise found no interaction between life stress and the 5-HTTLPR polymorphism, though life stress itself raised risk (Culverhouse et al., 2018). The idea survives from those early single-gene reports; inherited risk is in fact spread across many variants of small effect.
Depression is a problem of wealthy Western countries.
Surveys of 89,037 adults in 18 countries found lifetime major depressive episodes in 11.1% of adults in low- and middle-income countries, and 12-month prevalence there (5.9%) was similar to that in high-income countries (5.5%) (Bromet et al., 2011). Lifetime rates are higher in high-income countries, and although methodological studies did not trace that spread to methodological factors, the review judged greater recall error in low- and middle-income surveys a plausible and more parsimonious explanation; direct data are missing for most countries (Kessler & Bromet, 2013).

Glossary

Anhedonia.
Loss of interest or pleasure, one of the two core symptoms of a major depressive episode; in the tripartite structure it is the feature that is higher in depression than in anxiety.
Anxious distress specifier.
A specifier added in DSM-5 for a depressive episode with prominent anxiety; in one interview study of 260 patients with major depressive disorder it applied to about three-quarters and went with poorer functioning and coping.
Bereavement exclusion.
A former DSM rule against diagnosing a major depressive episode that followed a recent death; DSM-5 replaced it with clinical judgment, after reviews disagreed over whether it was valid.
Biotype.
A proposed subgroup of patients defined by a brain measure, such as resting functional connectivity, rather than by symptoms; a methodological replication in a different sample found its own clusters not statistically significant.
Candidate gene.
A gene selected for study because of a hypothesised role in a disorder; for depression, preregistered tests in large samples supported none of 18 such genes or their interactions with adversity.
Catatonia.
A syndrome that can be recorded in depression with its own specifier; one meta-analysis put its pooled prevalence across conditions at 9.0%, with highly heterogeneous results.
Cognitive bias modification.
Training that tries to change an attentional or interpretive bias so that its effect on mood can be tested; pooled effects on depression were small and non-significant after adjustment for publication bias.
Diathesis-stress model.
The view that depression arises when a vulnerability combines with life stress; challenged by life-stress research and contested, but not replaced.
Differential susceptibility.
A proposed alternative to diathesis-stress in which people differ in plasticity to their environment, not only in vulnerability; proposed rather than established.
Disruptive mood dysregulation disorder.
A childhood depressive category introduced in DSM-5, centred on chronic irritability and temper outbursts; uncommon in community samples, highly comorbid, and of unresolved distinctness from oppositional defiant and conduct disorders.
Double depression.
A major depressive episode superimposed on dysthymia, now within persistent depressive disorder; in pooled data from treatment-trial entrants, 85% had clinically severe quality-of-life impairment.
Dysthymia.
A chronic depressive diagnosis of earlier manuals, merged in DSM-5 with chronic major depression; a narrative review judged its validity doubtful because it fails to predict illness outcome.
Endogenous depression.
A historical label for depression thought to arise from within the person, set against reactive or neurotic depression; factor analyses found the two sets of features independent without establishing two aetiological types, and the term survives as a MeSH entry term.
Hierarchical Taxonomy of Psychopathology.
A dimensional system (HiTOP) that arranges symptoms into components, syndromes and spectra in place of categories; its developers call it promising but a work in progress, and it is actively contested.
Kindling hypothesis.
The proposal that life stress matters more for first episodes than for recurrences and that episodes come closer together as they accumulate; one community-cohort test found the shortening reflected differences between people, not acceleration within them, so its standing is contested.
Melancholia.
A historical term that moved in nineteenth-century psychiatry from a disorder of the intellect to a disorder of mood; it survives as a MeSH entry term and in the melancholic specifier.
Mixed features specifier.
A DSM-5 specifier for mixed presentations at the border of depressive and manic episodes; one meta-analysis put it at 11.6% of major depressive episodes, and its criteria remain contested.
Network model.
An account that treats a disorder as a set of symptoms in causal interplay rather than the expression of one underlying factor; whether its networks replicate is disputed, so it remains unsettled.
Neuroticism.
A personality trait on which every common diagnostic group scored higher than controls in one meta-analysis (mean d = 1.65); its causal role is unresolved, but it is a useful marker of general risk.
Persistent depressive disorder.
The DSM-5 chronic category that brought dysthymia and chronic major depression together; its validity is debated, and it is expected to be heterogeneous.
Premenstrual dysphoric disorder.
A DSM-5 depressive category tied to the menstrual cycle; an expert review argued for the full category, while a bioethical critique holds that calling it a mental disorder is unethical, so its status is still argued.
Response styles theory.
The account that a habit of ruminating rather than distracting contributes to depression and to its higher rate in women; the link from rumination to new onsets is supported, while its explanation of the sex difference rests on a proposal.
Rumination.
A habitual way of responding to low mood that the response-styles account contrasts with distraction; in longitudinal analyses it predicted new onsets of depressive disorders, but not chronicity once baseline symptoms were controlled.
Specifier.
A descriptor that records features of an episode, such as melancholic or atypical features, seasonal pattern, course or severity, without creating a separate disorder.
Stress generation.
The replicated finding that depression predicts later dependent stress, meaning events partly shaped by the person's own behaviour; most studies measured episodic rather than chronic stress.
Subthreshold depression.
Depressive symptoms below the full diagnostic threshold; a meta-analysis of 16 cohorts linked it to a higher incidence of later major depression (incidence rate ratio 1.95).
Taxometric analysis.
Statistical procedures that test whether a construct is a category with a natural boundary or a continuum; for depression they mostly favour a continuum, though the items analysed can change the answer.
Tripartite model.
A structure in which anxiety and depression share general distress, physiological hyperarousal is higher in anxiety, and anhedonia with low positive affect is higher in depression; a live structural proposal rather than a settled classification.

Key Researchers

Aaron T. Beck

(1921-2021). Psychiatrist at the University of Pennsylvania who developed the cognitive model of depression and its later integrative and evolutionary revision. Faculty Page

Pim Cuijpers

Clinical psychologist, emeritus at Vrije Universiteit Amsterdam, who led network meta-analyses of psychotherapies for depression and meta-analyses of preventing its onset. ORCID - Faculty Page

Eiko I. Fried

Psychologist at Leiden University who studies the heterogeneity of depression symptoms and the content overlap of depression scales. ORCID - Faculty Page

Ian H. Gotlib

Psychologist at Stanford University who studies cognitive and neural processes in depression. ORCID - Faculty Page

Kenneth S. Kendler

Psychiatrist and psychiatric geneticist at Virginia Commonwealth University who built developmental twin models of major depression and wrote histories of melancholia and the Feighner criteria. ORCID - Faculty Page

Ronald C. Kessler

Sociologist and psychiatric epidemiologist at Harvard Medical School who led national and cross-national surveys of depression prevalence. ORCID - Faculty Page

Daniel N. Klein

Psychologist at Stony Brook University who proposed replacing the depressive categories with dimensions of severity and chronicity. ORCID

Roman Kotov

Psychologist at Stony Brook University who meta-analysed personality across common mental disorders and helped develop the Hierarchical Taxonomy of Psychopathology. ORCID - Faculty Page

Susan Nolen-Hoeksema

(1959-2013). Psychologist at Yale University who proposed the response-styles account of rumination and the sex difference in depression. Faculty Page

Gordon Parker

Psychiatrist at the University of New South Wales who argued for hierarchical models of depression and for melancholia as a distinct disorder. ORCID - Faculty Page

Frequently Asked Questions

What is a depressive disorder?

A depressive disorder is one of a family of mood disorder diagnoses that share a persistent disturbance of mood or of interest and pleasure, as distinct from a passing low mood. Its best-studied member, major depression, is common and severely disabling, and its course and response to treatment vary from person to person (Malhi & Mann, 2018).

How do the types of depressive disorder differ?

Some types are diagnostic categories, such as persistent depressive disorder or premenstrual dysphoric disorder, while others, such as melancholic or atypical features, are specifiers that describe an episode rather than separate disorders. In STAR*D baseline data, the melancholic and atypical specifiers did not make the symptom profiles of their groups any less varied (Lorenzo-Luaces et al., 2021).

Is depression one disorder or many?

The question is unresolved. A systematic review of 34 latent variable analyses found no conclusive evidence for data-driven symptom subtypes, and the classes that did emerge mostly reflected differences in severity (van Loo et al., 2012).

Is depression caused by a chemical imbalance?

No simple serotonin deficit has been shown. An umbrella review found no consistent association between serotonin and depression, while noting that the quality of the reviews was variable and that receptor and transporter studies had not reliably excluded prior antidepressant use (Moncrieff et al., 2023). Researchers still dispute how large a part serotonin plays.

How common are depressive disorders?

Surveys of 89,037 adults in 18 countries found that 14.6% of adults in high-income countries and 11.1% in low- and middle-income countries had had a major depressive episode, with women affected about twice as often as men (Bromet et al., 2011).

What is the risk of suicide in depressive disorders, and where is help available?

Within depression, a systematic review of nineteen studies identified male gender, family psychiatric history, a previous attempt, severe depression, hopelessness, anxiety comorbidity and alcohol or drug misuse as correlates of suicide, although it did not examine how these factors depend on one another (Hawton et al., 2013). In the United States the 988 Suicide & Crisis Lifeline can be reached by call or text at 988, and the International Association for Suicide Prevention lists crisis centres in other countries.

How are depressive disorders treated?

The main classes are antidepressant medication, psychological therapies and brain stimulation. A network meta-analysis of 522 trials found all 21 antidepressants it compared more effective than placebo in acute treatment of major depressive disorder, with certainty in the evidence ranging from moderate to very low (Cipriani et al., 2018). A plain-language overview is on the depression treatment page.

Can depression be prevented?

In people at high risk, preventive psychological interventions lowered the chance of a new depressive disorder by 19% at one year across 50 randomized trials, a result not yet shown for the general population (Cuijpers et al., 2021b).

How This Article Was Prepared

This is a narrative reference article, not a systematic review. Its sources were identified through a survey of the 25 most-cited works on the subject and the most-cited research of eight of its key researchers, on 22 November 2026.

- Considered and excluded: eight candidate works, each with a recorded reason (for example, superseded by a later review, a single case report, or outside the article's scope). - Included: 215 sources, each confirmed against the publisher's or indexer's own record: 214 by DOI and one by ISBN. 59 read in full text, 151 from their abstracts and five cited without a retrievable abstract. Abstracts came from PubMed, Crossref, OpenAlex and Scite. - Identifiers: 206 references carry a PubMed ID and 89 a PubMed Central ID, each confirmed by the National Library of Medicine against the cited DOI. - Retractions: 214 sources were checked for retractions and editorial notices; none was retracted. - Publication dates: one from the 1960s, five from the 1970s, two from the 1980s, four from the 1990s, 42 from the 2000s, 117 from the 2010s and 44 from 2020 or later. - Checked against sources: each of the 423 cited sentences was matched to a passage in its source (401 supported by that passage, 10 resting on a source without a retrievable abstract), and an independent reviewer checked the article against those sources over five passes before publication. - Limits: a narrative review reflects choices about which studies to include, so it can over-represent some lines of work more than a systematic review would. To limit this, each of the field's 25 most-cited works was considered, and the reviewer searched each cited source for contrary evidence before accepting a sentence.

See our Editorial Policy for the full method.

References

Abramson, L. Y., Metalsky, G. I., & Alloy, L. B. (1989). Hopelessness depression: A theory-based subtype of depression. Psychological Review, 96(2), 358-372. https://doi.org/10.1037/0033-295X.96.2.358

Abramson, L. Y., Seligman, M. E. P., & Teasdale, J. D. (1978). Learned helplessness in humans: Critique and reformulation. Journal of Abnormal Psychology, 87(1), 49-74. https://doi.org/10.1037/0021-843X.87.1.49

Adams, M. J., et al. (Major Depressive Disorder Working Group of the Psychiatric Genomics Consortium). (2025). Trans-ancestry genome-wide study of depression identifies 697 associations implicating cell types and pharmacotherapies. Cell, 188(3), 640-652.e9. https://doi.org/10.1016/j.cell.2024.12.002 PMID: 39814019 · PMCID: PMC11829167

Alegría, M., Chatterji, P., Wells, K., Cao, Z., Chen, C.-n., Takeuchi, D., Jackson, J., & Meng, X.-L. (2008). Disparity in depression treatment among racial and ethnic minority populations in the United States. Psychiatric Services, 59(11), 1264-1272. https://doi.org/10.1176/ps.2008.59.11.1264 PMID: 18971402 · PMCID: PMC2668139

Alexopoulos, G. S., Meyers, B. S., Young, R. C., Campbell, S., Silbersweig, D., & Charlson, M. (1997). 'Vascular depression' hypothesis. Archives of General Psychiatry, 54(10), 915-922. https://doi.org/10.1001/archpsyc.1997.01830220033006 PMID: 9337771

Alloy, L. B., Abramson, L. Y., Whitehouse, W. G., Hogan, M. E., Panzarella, C., & Rose, D. T. (2006). Prospective incidence of first onsets and recurrences of depression in individuals at high and low cognitive risk for depression. Journal of Abnormal Psychology, 115(1), 145-156. https://doi.org/10.1037/0021-843X.115.1.145 PMID: 16492105

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). https://doi.org/10.1176/appi.books.9780890425596

American Psychiatric Association. (2022). Diagnostic and statistical manual of mental disorders (5th ed., text rev.; DSM-5-TR). American Psychiatric Association Publishing. https://doi.org/10.1176/appi.books.9780890425787

Anderson, R. J., Freedland, K. E., Clouse, R. E., & Lustman, P. J. (2001). The prevalence of comorbid depression in adults with diabetes: A meta-analysis. Diabetes Care, 24(6), 1069-1078. https://doi.org/10.2337/diacare.24.6.1069 PMID: 11375373

Anderson, S. F., Monroe, S. M., Rohde, P., & Lewinsohn, P. M. (2016). Questioning kindling: An analysis of cycle acceleration in unipolar depression. Clinical Psychological Science, 4(2), 229-238. https://doi.org/10.1177/2167702615591951 PMID: 27034912 · PMCID: PMC4809541

Axelson, D., Findling, R. L., Fristad, M. A., et al. (2012). Examining the proposed disruptive mood dysregulation disorder diagnosis in children in the Longitudinal Assessment of Manic Symptoms study. Journal of Clinical Psychiatry, 73(10), 1342-1350. https://doi.org/10.4088/JCP.12m07674 PMID: 23140653 · PMCID: PMC3581334

Baglioni, C., Battagliese, G., Feige, B., Spiegelhalder, K., Nissen, C., Voderholzer, U., Lombardo, C., & Riemann, D. (2011). Insomnia as a predictor of depression: A meta-analytic evaluation of longitudinal epidemiological studies. Journal of Affective Disorders, 135(1-3), 10-19. https://doi.org/10.1016/j.jad.2011.01.011 PMID: 21300408

Baglioni, C., Nanovska, S., Regen, W., Spiegelhalder, K., Feige, B., Nissen, C., Reynolds, C. F., & Riemann, D. (2016). Sleep and mental disorders: A meta-analysis of polysomnographic research. Psychological Bulletin, 142(9), 969-990. https://doi.org/10.1037/bul0000053 PMID: 27416139 · PMCID: PMC5110386

Baldwin, J. R., Wang, B., Karwatowska, L., Schoeler, T., Tsaligopoulou, A., Munafò, M. R., & Pingault, J.-B. (2023). Childhood maltreatment and mental health problems: A systematic review and meta-analysis of quasi-experimental studies. American Journal of Psychiatry, 180(2), 117-126. https://doi.org/10.1176/appi.ajp.20220174 PMID: 36628513 · PMCID: PMC7614155

Beach, S. R. H., & Amir, N. (2003). Is depression taxonic, dimensional, or both? Journal of Abnormal Psychology, 112(2), 228-236. https://doi.org/10.1037/0021-843X.112.2.228 PMID: 12784832

Beck, A. T., & Bredemeier, K. (2016). A unified model of depression: Integrating clinical, cognitive, biological, and evolutionary perspectives. Clinical Psychological Science, 4(4), 596-619. https://doi.org/10.1177/2167702616628523

Beck, A. T. (2008). The evolution of the cognitive model of depression and its neurobiological correlates. American Journal of Psychiatry, 165(8), 969-977. https://doi.org/10.1176/appi.ajp.2008.08050721 PMID: 18628348

Beijers, L., Wardenaar, K. J., van Loo, H. M., & Schoevers, R. A. (2019). Data-driven biological subtypes of depression: Systematic review of biological approaches to depression subtyping. Molecular Psychiatry, 24(6), 888-900. https://doi.org/10.1038/s41380-019-0385-5 PMID: 30824865

Belmaker, R. H., & Agam, G. (2008). Major depressive disorder. New England Journal of Medicine, 358(1), 55-68. https://doi.org/10.1056/nejmra073096 PMID: 18172175

Belsky, J., & Pluess, M. (2009). Beyond diathesis stress: Differential susceptibility to environmental influences. Psychological Bulletin, 135(6), 885-908. https://doi.org/10.1037/a0017376 PMID: 19883141

Benarous, X., Lahaye, H., Consoli, A., Cohen, D., Labelle, R., & Guilé, J.-M. (2025). Prevalence and comorbidity rates of disruptive mood dysregulation disorder in epidemiological and clinical samples: A systematic review and meta-analysis. European Psychiatry, 68(1), e11. https://doi.org/10.1192/j.eurpsy.2024.1813 PMID: 39809705 · PMCID: PMC11795450

Border, R., Johnson, E. C., Evans, L. M., Smolen, A., Berley, N., Sullivan, P. F., & Keller, M. C. (2019). No support for historical candidate gene or candidate gene-by-interaction hypotheses for major depression across multiple large samples. The American Journal of Psychiatry, 176(5), 376-387. https://doi.org/10.1176/appi.ajp.2018.18070881 PMID: 30845820 · PMCID: PMC6548317

Borsboom, D., & Cramer, A. O. J. (2013). Network analysis: An integrative approach to the structure of psychopathology. Annual Review of Clinical Psychology, 9, 91-121. https://doi.org/10.1146/annurev-clinpsy-050212-185608 PMID: 23537483

Borsboom, D., Fried, E. I., Epskamp, S., Waldorp, L. J., van Borkulo, C. D., van der Maas, H. L. J., & Cramer, A. O. J. (2017). False alarm? A comprehensive reanalysis of "Evidence that psychopathology symptom networks have limited replicability". Journal of Abnormal Psychology, 126(7), 989-999. https://doi.org/10.1037/abn0000306 PMID: 29106282

Bostwick, J. M., & Pankratz, V. S. (2000). Affective disorders and suicide risk: A reexamination. American Journal of Psychiatry, 157(12), 1925-1932. https://doi.org/10.1176/appi.ajp.157.12.1925 PMID: 11097952

Breedvelt, J. J. F., Karyotaki, E., Warren, F. C., Brouwer, M. E., Jermann, F., Hollandare, F., Klein, N., de Jonge, M., Klein, D. N., Farb, N., Segal, Z., Biesheuvel Leliefeld, K. E. M., Jarrett, R., Vittengl, J., Thase, M., Ma, H., Kuyken, W., Shallcross, A. J., van Heeringen, C., Hoorelbeke, K., Koster, E., Williams, M., Huijbers, M. J., Speckens, A., Cuijpers, P., van Oppen, P., Gilbody, S., & Bockting, C. L. (2024). An individual participant data meta-analysis of psychological interventions for preventing depression relapse. Nature Mental Health, 2(2), 154-163. https://doi.org/10.1038/s44220-023-00178-x

Brohan, E., Gauci, D., Sartorius, N., Thornicroft, G., & GAMIAN-Europe Study Group. (2011). Self-stigma, empowerment and perceived discrimination among people with bipolar disorder or depression in 13 European countries: The GAMIAN-Europe study. Journal of Affective Disorders, 129(1-3), 56-63. https://doi.org/10.1016/j.jad.2010.09.001 PMID: 20888050

Bromet, E., Andrade, L. H., Hwang, I., Sampson, N. A., Alonso, J., de Girolamo, G., de Graaf, R., Demyttenaere, K., Hu, C., Iwata, N., Karam, A. N., Kaur, J., Kostyuchenko, S., Lépine, J.-P., Levinson, D., Matschinger, H., Medina Mora, M. E., Oakley Browne, M., Posada-Villa, J., Viana, M. C., Williams, D. R., & Kessler, R. C. (2011). Cross-national epidemiology of DSM-IV major depressive episode. BMC Medicine, 9, 90. https://doi.org/10.1186/1741-7015-9-90 PMID: 21791035 · PMCID: PMC3163615

Browne, T. K. (2015). Is premenstrual dysphoric disorder really a disorder? Journal of Bioethical Inquiry, 12(2), 313-330. https://doi.org/10.1007/s11673-014-9567-7 PMID: 25164305

Brunoni, A. R., Chaimani, A., Moffa, A. H., Razza, L. B., Gattaz, W. F., Daskalakis, Z. J., & Carvalho, A. F. (2017). Repetitive transcranial magnetic stimulation for the acute treatment of major depressive episodes: A systematic review with network meta-analysis. JAMA Psychiatry, 74(2), 143-152. https://doi.org/10.1001/jamapsychiatry.2016.3644 PMID: 28030740

Carhart-Harris, R., Giribaldi, B., Watts, R., Baker-Jones, M., Murphy-Beiner, A., Murphy, R., Martell, J., Blemings, A., Erritzoe, D., & Nutt, D. J. (2021). Trial of psilocybin versus escitalopram for depression. New England Journal of Medicine, 384(15), 1402-1411. https://doi.org/10.1056/NEJMoa2032994 PMID: 33852780

Carlini, S. V., Lanza di Scalea, T., McNally, S. T., Lester, J., & Deligiannidis, K. M. (2022). Management of premenstrual dysphoric disorder: A scoping review. International Journal of Women's Health, 14, 1783-1801. https://doi.org/10.2147/IJWH.S297062 PMID: 36575726 · PMCID: PMC9790166

Cavanagh, J. T. O., Carson, A. J., Sharpe, M., & Lawrie, S. M. (2003). Psychological autopsy studies of suicide: A systematic review. Psychological Medicine, 33(3), 395-405. https://doi.org/10.1017/S0033291702006943 PMID: 12701661

Chapman, D. P., Whitfield, C. L., Felitti, V. J., Dube, S. R., Edwards, V. J., & Anda, R. F. (2004). Adverse childhood experiences and the risk of depressive disorders in adulthood. Journal of Affective Disorders, 82(2), 217-225. https://doi.org/10.1016/j.jad.2003.12.013 PMID: 15488250

Cipriani, A., Furukawa, T. A., Salanti, G., Chaimani, A., Atkinson, L. Z., Ogawa, Y., Leucht, S., Ruhe, H. G., Turner, E. H., Higgins, J. P. T., Egger, M., Takeshima, N., Hayasaka, Y., Imai, H., Shinohara, K., Tajika, A., Ioannidis, J. P. A., & Geddes, J. R. (2018). 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. The Lancet, 391(10128), 1357-1366. https://doi.org/10.1016/S0140-6736(17)32802-7 PMID: 29477251 · PMCID: PMC5889788

Cipriani, A., Zhou, X., Del Giovane, C., Hetrick, S. E., Qin, B., Whittington, C., Coghill, D., Zhang, Y., Hazell, P., Leucht, S., Cuijpers, P., Pu, J., Cohen, D., Ravindran, A. V., Liu, Y., Michael, K. D., Yang, L., Liu, L., & Xie, P. (2016). Comparative efficacy and tolerability of antidepressants for major depressive disorder in children and adolescents: A network meta-analysis. The Lancet, 388(10047), 881-890. https://doi.org/10.1016/S0140-6736(16)30385-3 PMID: 27289172

Clark, L. A., & Watson, D. (1991). Tripartite model of anxiety and depression: Psychometric evidence and taxonomic implications. Journal of Abnormal Psychology, 100(3), 316-336. https://doi.org/10.1037/0021-843X.100.3.316 PMID: 1918611

Clement, S., Schauman, O., Graham, T., Maggioni, F., Evans-Lacko, S., Bezborodovs, N., Morgan, C., Rüsch, N., Brown, J. S. L., & Thornicroft, G. (2015). What is the impact of mental health-related stigma on help-seeking? A systematic review of quantitative and qualitative studies. Psychological Medicine, 45(1), 11-27. https://doi.org/10.1017/S0033291714000129 PMID: 24569086

Copeland, W. E., Angold, A., Costello, E. J., & Egger, H. (2013). Prevalence, comorbidity, and correlates of DSM-5 proposed disruptive mood dysregulation disorder. American Journal of Psychiatry, 170(2), 173-179. https://doi.org/10.1176/appi.ajp.2012.12010132 PMID: 23377638 · PMCID: PMC3573525

Cowen, P. J., & Browning, M. (2015). What has serotonin to do with depression? World Psychiatry, 14(2), 158-160. https://doi.org/10.1002/wps.20229 PMID: 26043325 · PMCID: PMC4471964

Cristea, I. A., Kok, R. N., & Cuijpers, P. (2015). Efficacy of cognitive bias modification interventions in anxiety and depression: meta-analysis. British Journal of Psychiatry, 206(1), 7-16. https://doi.org/10.1192/bjp.bp.114.146761 PMID: 25561486

Cuellar, A. K., Johnson, S. L., & Winters, R. (2005). Distinctions between bipolar and unipolar depression. Clinical Psychology Review, 25(3), 307-339. https://doi.org/10.1016/j.cpr.2004.12.002 PMID: 15792852 · PMCID: PMC2850601

Cuijpers, P., Franco, P., Ciharova, M., Miguel, C., Segre, L., Quero, S., & Karyotaki, E. (2023). Psychological treatment of perinatal depression: a meta-analysis. Psychological Medicine, 53(6), 2596-2608. https://doi.org/10.1017/S0033291721004529 PMID: 37310303 · PMCID: PMC10123831

Cuijpers, P., Karyotaki, E., Eckshtain, D., Ng, M. Y., Corteselli, K. A., Noma, H., Quero, S., & Weisz, J. R. (2020). Psychotherapy for depression across different age groups: A systematic review and meta-analysis. JAMA Psychiatry, 77(7), 694-702. https://doi.org/10.1001/jamapsychiatry.2020.0164 PMID: 32186668 · PMCID: PMC7081149

Cuijpers, P., Pineda, B. S., Quero, S., Karyotaki, E., Struijs, S. Y., Figueroa, C. A., Llamas, J. A., Furukawa, T. A., & Muñoz, R. F. (2021). Psychological interventions to prevent the onset of depressive disorders: A meta-analysis of randomized controlled trials. Clinical Psychology Review, 83, 101955. https://doi.org/10.1016/j.cpr.2020.101955 PMID: 33333441

Cuijpers, P., Quero, S., Noma, H., Ciharova, M., Miguel, C., Karyotaki, E., Cipriani, A., Cristea, I. A., & Furukawa, T. A. (2021). Psychotherapies for depression: a network meta-analysis covering efficacy, acceptability and long-term outcomes of all main treatment types. World Psychiatry, 20(2), 283-293. https://doi.org/10.1002/wps.20860 PMID: 34002502 · PMCID: PMC8129869

Cuijpers, P., Sijbrandij, M., Koole, S. L., Andersson, G., Beekman, A. T., & Reynolds, C. F. (2014). Adding psychotherapy to antidepressant medication in depression and anxiety disorders: a meta-analysis. World Psychiatry, 13(1), 56-67. https://doi.org/10.1002/wps.20089 PMID: 24497254 · PMCID: PMC3918025

Cuijpers, P., Smit, F., Bohlmeijer, E., Hollon, S. D., & Andersson, G. (2010). Efficacy of cognitive-behavioural therapy and other psychological treatments for adult depression: Meta-analytic study of publication bias. British Journal of Psychiatry, 196(3), 173-178. https://doi.org/10.1192/bjp.bp.109.066001 PMID: 20194536

Culverhouse, R. C., Saccone, N. L., Horton, A. C., Ma, Y., Anstey, K. J., et al. (2018). Collaborative meta-analysis finds no evidence of a strong interaction between stress and 5-HTTLPR genotype contributing to the development of depression. Molecular Psychiatry, 23(1), 133-142. https://doi.org/10.1038/mp.2017.44 PMID: 28373689 · PMCID: PMC5628077

Davies, J., & Read, J. (2019). A systematic review into the incidence, severity and duration of antidepressant withdrawal effects: Are guidelines evidence-based? Addictive Behaviors, 97, 111-121. https://doi.org/10.1016/j.addbeh.2018.08.027 PMID: 30292574

Deng, Z.-D., Robins, P. L., Regenold, W., Rohde, P., Dannhauer, M., & Lisanby, S. H. (2024). How electroconvulsive therapy works in the treatment of depression: Is it the seizure, the electricity, or both? Neuropsychopharmacology, 49(1), 150-162. https://doi.org/10.1038/s41386-023-01677-2 PMID: 37488281 · PMCID: PMC10700353

Desmet, M. (2013). Some preliminary notes on an empirical test of Freud's theory on depression. Frontiers in Psychology, 4, 158. https://doi.org/10.3389/fpsyg.2013.00158 PMID: 23675357 · PMCID: PMC3646245

DeYoung, C. G., Kotov, R., Krueger, R. F., Cicero, D. C., Conway, C. C., Eaton, N. R., Forbes, M. K., Hallquist, M. N., Jonas, K., Latzman, R. D., Rodriguez-Seijas, C., Ruggero, C. J., Simms, L. J., Waldman, I. D., Waszczuk, M. A., Widiger, T., & Wright, A. G. C. (2022). Answering questions about the Hierarchical Taxonomy of Psychopathology (HiTOP): Analogies to whales and sharks miss the boat. Clinical Psychological Science, 10(2), 279-284. https://doi.org/10.1177/21677026211049390 PMID: 35444863 · PMCID: PMC9017579

Dinga, R., Schmaal, L., Penninx, B. W. J. H., van Tol, M. J., Veltman, D. J., van Velzen, L., Mennes, M., van der Wee, N. J. A., & Marquand, A. F. (2019). Evaluating the evidence for biotypes of depression: Methodological replication and extension of Drysdale et al. (2017). NeuroImage: Clinical, 22, 101796. https://doi.org/10.1016/j.nicl.2019.101796 PMID: 30935858 · PMCID: PMC6543446

Diniz, B. S., Butters, M. A., Albert, S. M., Dew, M. A., & Reynolds, C. F., III. (2013). Late-life depression and risk of vascular dementia and Alzheimer's disease: Systematic review and meta-analysis of community-based cohort studies. The British Journal of Psychiatry, 202(5), 329-335. https://doi.org/10.1192/bjp.bp.112.118307 PMID: 23637108 · PMCID: PMC3640214

Doblyte, S., & Jiménez-Mejías, E. (2017). Understanding help-seeking behavior in depression: A qualitative synthesis of patients' experiences. Qualitative Health Research, 27(1), 100-113. https://doi.org/10.1177/1049732316681282 PMID: 27956660

Dowlati, Y., Herrmann, N., Swardfager, W., Liu, H., Sham, L., Reim, E. K., & Lanctôt, K. L. (2010). A meta-analysis of cytokines in major depression. Biological Psychiatry, 67(5), 446-457. https://doi.org/10.1016/j.biopsych.2009.09.033 PMID: 20015486

Drysdale, A. T., Grosenick, L., Downar, J., Dunlop, K., Mansouri, F., Meng, Y., Fetcho, R. N., Zebley, B., Oathes, D. J., Etkin, A., Schatzberg, A. F., Sudheimer, K., Keller, J., Mayberg, H. S., Gunning, F. M., Alexopoulos, G. S., Fox, M. D., Pascual-Leone, A., Voss, H. U., Casey, B. J., Dubin, M. J., & Liston, C. (2017). Resting-state connectivity biomarkers define neurophysiological subtypes of depression. Nature Medicine, 23(1), 28-38. https://doi.org/10.1038/nm.4246 PMID: 27918562 · PMCID: PMC5624035

Duman, R. S., & Monteggia, L. M. (2006). A neurotrophic model for stress-related mood disorders. Biological Psychiatry, 59(12), 1116-1127. https://doi.org/10.1016/j.biopsych.2006.02.013 PMID: 16631126

Epperson, C. N., Steiner, M., Hartlage, S. A., Eriksson, E., Schmidt, P. J., Jones, I., & Yonkers, K. A. (2012). Premenstrual dysphoric disorder: Evidence for a new category for DSM-5. American Journal of Psychiatry, 169(5), 465-475. https://doi.org/10.1176/appi.ajp.2012.11081302 PMID: 22764360 · PMCID: PMC3462360

Evans, V. C., Iverson, G. L., Yatham, L. N., & Lam, R. W. (2014). The relationship between neurocognitive and psychosocial functioning in major depressive disorder: A systematic review. The Journal of Clinical Psychiatry, 75(12), 1359-1370. https://doi.org/10.4088/JCP.13r08939 PMID: 25551235

Evans-Lacko, S., & Knapp, M. (2016). Global patterns of workplace productivity for people with depression: Absenteeism and presenteeism costs across eight diverse countries. Social Psychiatry and Psychiatric Epidemiology, 51(11), 1525-1537. https://doi.org/10.1007/s00127-016-1278-4 PMID: 27667656 · PMCID: PMC5101346

Everaert, J., Podina, I. R., & Koster, E. H. W. (2017). A comprehensive meta-analysis of interpretation biases in depression. Clinical Psychology Review, 58, 33-48. https://doi.org/10.1016/j.cpr.2017.09.005 PMID: 28974339

Feighner, J. P., Robins, E., Guze, S. B., Woodruff, R. A., Jr., Winokur, G., & Munoz, R. (1972). Diagnostic criteria for use in psychiatric research. Archives of General Psychiatry, 26(1), 57-63. https://doi.org/10.1001/archpsyc.1972.01750190059011 PMID: 5009428

Ferrari, A. J., Charlson, F. J., Norman, R. E., Patten, S. B., Freedman, G., Murray, C. J. L., Vos, T., & Whiteford, H. A. (2013). Burden of depressive disorders by country, sex, age, and year: Findings from the Global Burden of Disease Study 2010. PLoS Medicine, 10(11), e1001547. https://doi.org/10.1371/journal.pmed.1001547 PMID: 24223526 · PMCID: PMC3818162

First, M. B., Gaebel, W., Maj, M., Stein, D. J., Kogan, C. S., Saunders, J. B., Poznyak, V. B., Gureje, O., Lewis-Fernández, R., Maercker, A., Brewin, C. R., Cloitre, M., Claudino, A., Pike, K. M., Baird, G., Skuse, D., Krueger, R. B., Briken, P., Burke, J. D., Lochman, J. E., Evans, S. C., Woods, D. W., & Reed, G. M. (2021). An organization- and category-level comparison of diagnostic requirements for mental disorders in ICD-11 and DSM-5. World Psychiatry, 20(1), 34-51. https://doi.org/10.1002/wps.20825 PMID: 33432742 · PMCID: PMC7801846

Firth, J., Torous, J., Nicholas, J., Carney, R., Pratap, A., Rosenbaum, S., & Sarris, J. (2017). The efficacy of smartphone-based mental health interventions for depressive symptoms: A meta-analysis of randomized controlled trials. World Psychiatry, 16(3), 287-298. https://doi.org/10.1002/wps.20472 PMID: 28941113 · PMCID: PMC5608852

Fleming, S. K., Blasey, C., & Schatzberg, A. F. (2004). Neuropsychological correlates of psychotic features in major depressive disorders: A review and meta-analysis. Journal of Psychiatric Research, 38(1), 27-35. https://doi.org/10.1016/S0022-3956(03)00100-6 PMID: 14690768

Forbes, M. K., Wright, A. G. C., Markon, K. E., & Krueger, R. F. (2017). Evidence that psychopathology symptom networks have limited replicability. Journal of Abnormal Psychology, 126(7), 969-988. https://doi.org/10.1037/abn0000276 PMID: 29106281 · PMCID: PMC5749927

Frank, E., Prien, R. F., Jarrett, R. B., Keller, M. B., Kupfer, D. J., Lavori, P. W., Rush, A. J., & Weissman, M. M. (1991). Conceptualization and rationale for consensus definitions of terms in major depressive disorder: Remission, recovery, relapse, and recurrence. Archives of General Psychiatry, 48(9), 851-855. https://doi.org/10.1001/archpsyc.1991.01810330075011 PMID: 1929776

Franklin, J. C., Ribeiro, J. D., Fox, K. R., Bentley, K. H., Kleiman, E. M., Huang, X., Musacchio, K. M., Jaroszewski, A. C., Chang, B. P., & Nock, M. K. (2017). Risk factors for suicidal thoughts and behaviors: A meta-analysis of 50 years of research. Psychological Bulletin, 143(2), 187-232. https://doi.org/10.1037/bul0000084 PMID: 27841450

Fried, E. I., & Nesse, R. M. (2015). Depression is not a consistent syndrome: An investigation of unique symptom patterns in the STARD study. Journal of Affective Disorders, 172, 96-102. https://doi.org/10.1016/j.jad.2014.10.010 PMID: 25451401 · PMCID: PMC4397113

Fried, E. I. (2017). The 52 symptoms of major depression: Lack of content overlap among seven common depression scales. Journal of Affective Disorders, 208, 191-197. https://doi.org/10.1016/j.jad.2016.10.019 PMID: 27792962

GBD 2019 Mental Disorders Collaborators. (2022). 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. The Lancet Psychiatry, 9(2), 137-150. https://doi.org/10.1016/S2215-0366(21)00395-3 PMID: 35026139 · PMCID: PMC8776563

Ghaemi, S. N. (2013). Bipolar spectrum: A review of the concept and a vision for the future. Psychiatry Investigation, 10(3), 218-224. https://doi.org/10.4306/pi.2013.10.3.218 PMID: 24302943 · PMCID: PMC3843012

Goodwin, G. M., Aaronson, S. T., Alvarez, O., Arden, P. C., Baker, A., Bennett, J. C., Bird, C., Blom, R. E., Brennan, C., Brusch, D., et al. (2022). Single-dose psilocybin for a treatment-resistant episode of major depression. New England Journal of Medicine, 387(18), 1637-1648. https://doi.org/10.1056/NEJMoa2206443 PMID: 36322843

Gotlib, I. H., & Joormann, J. (2010). Cognition and depression: Current status and future directions. Annual Review of Clinical Psychology, 6, 285-312. https://doi.org/10.1146/annurev.clinpsy.121208.131305 PMID: 20192795 · PMCID: PMC2845726

Grafton, B., MacLeod, C., Rudaizky, D., Holmes, E. A., Salemink, E., Fox, E., & Notebaert, L. (2017). Confusing procedures with process when appraising the impact of cognitive bias modification on emotional vulnerability. British Journal of Psychiatry, 211(5), 266-271. https://doi.org/10.1192/bjp.bp.115.176123 PMID: 29092835

Greenberg, P. E., Fournier, A.-A., Sisitsky, T., Pike, C. T., & Kessler, R. C. (2015). The economic burden of adults with major depressive disorder in the United States (2005 and 2010). Journal of Clinical Psychiatry, 76(2), 155-162. https://doi.org/10.4088/jcp.14m09298 PMID: 25742202

Greicius, M. D., Flores, B. H., Menon, V., Glover, G. H., Solvason, H. B., Kenna, H., Reiss, A. L., & Schatzberg, A. F. (2007). Resting-state functional connectivity in major depression: Abnormally increased contributions from subgenual cingulate cortex and thalamus. Biological Psychiatry, 62(5), 429-437. https://doi.org/10.1016/j.biopsych.2006.09.020 PMID: 17210143 · PMCID: PMC2001244

Haeffel, G. J., Jeronimus, B. F., Kaiser, B. N., Weaver, L. J., Soyster, P. D., Fisher, A. J., Vargas, I., Goodson, J. T., & Lu, W. (2022). Folk classification and factor rotations: Whales, sharks, and the problems with the Hierarchical Taxonomy of Psychopathology (HiTOP). Clinical Psychological Science, 10(2), 259-278. https://doi.org/10.1177/21677026211002500 PMID: 35425668 · PMCID: PMC9004619

Hardeveld, F., Spijker, J., De Graaf, R., Nolen, W. A., & Beekman, A. T. F. (2010). Prevalence and predictors of recurrence of major depressive disorder in the adult population. Acta Psychiatrica Scandinavica, 122(3), 184-191. https://doi.org/10.1111/j.1600-0447.2009.01519.x PMID: 20003092

Harmer, C. J., Goodwin, G. M., & Cowen, P. J. (2009). Why do antidepressants take so long to work? A cognitive neuropsychological model of antidepressant drug action. British Journal of Psychiatry, 195(2), 102-108. https://doi.org/10.1192/bjp.bp.108.051193 PMID: 19648538

Hasin, D. S., Sarvet, A. L., Meyers, J. L., Saha, T. D., Ruan, W. J., Stohl, M., & Grant, B. F. (2018). Epidemiology of adult DSM-5 major depressive disorder and its specifiers in the United States. JAMA Psychiatry, 75(4), 336-346. https://doi.org/10.1001/jamapsychiatry.2017.4602 PMID: 29450462 · PMCID: PMC5875313

Haslam, N., Holland, E., & Kuppens, P. (2012). Categories versus dimensions in personality and psychopathology: A quantitative review of taxometric research. Psychological Medicine, 42(5), 903-920. https://doi.org/10.1017/S0033291711001966 PMID: 21939592

Hawton, K., Casañas i Comabella, C., Haw, C., & Saunders, K. (2013). Risk factors for suicide in individuals with depression: A systematic review. Journal of Affective Disorders, 147(1-3), 17-28. https://doi.org/10.1016/j.jad.2013.01.004 PMID: 23411024

Henrich, J., Heine, S. J., & Norenzayan, A. (2010). The weirdest people in the world? Behavioral and Brain Sciences, 33(2-3), 61-83. https://doi.org/10.1017/S0140525X0999152X PMID: 20550733

Henssler, J., Schmidt, Y., Schmidt, U., Schwarzer, G., Bschor, T., & Baethge, C. (2024). Incidence of antidepressant discontinuation symptoms: A systematic review and meta-analysis. The Lancet Psychiatry, 11(7), 526-535. https://doi.org/10.1016/S2215-0366(24)00133-0 PMID: 38851198

Howard, D. M., Adams, M. J., Clarke, T.-K., Hafferty, J. D., Gibson, J., Shirali, M., Coleman, J. R. I., Hagenaars, S. P., Ward, J., Wigmore, E. M., Alloza, C., Shen, X., Barbu, M. C., Xu, E. Y., Whalley, H. C., Marioni, R. E., Porteous, D. J., Davies, G., Deary, I. J., . . . McIntosh, A. M. (2019). Genome-wide meta-analysis of depression identifies 102 independent variants and highlights the importance of the prefrontal brain regions. Nature Neuroscience, 22(3), 343-352. https://doi.org/10.1038/s41593-018-0326-7 PMID: 30718901 · PMCID: PMC6522363

Janssen, N. P., Hendriks, G.-J., Baranelli, C. T., Lucassen, P., Oude Voshaar, R., Spijker, J., & Huibers, M. J. H. (2021). How does behavioural activation work? A systematic review of the evidence on potential mediators. Psychotherapy and Psychosomatics, 90(2), 85-93. https://doi.org/10.1159/000509820 PMID: 32898847

Jauhar, S., Arnone, D., Baldwin, D. S., Bloomfield, M., Browning, M., Cleare, A. J., Corlett, P., Deakin, J. F. W., Erritzoe, D., Fu, C., Fusar-Poli, P., Goodwin, G. M., Hayes, J., Howard, R., Howes, O. D., Juruena, M. F., Lam, R. W., Lawrie, S. M., McAllister-Williams, H., . . . Cowen, P. J. (2023). A leaky umbrella has little value: Evidence clearly indicates the serotonin system is implicated in depression. Molecular Psychiatry, 28(8), 3149-3152. https://doi.org/10.1038/s41380-023-02095-y PMID: 37322065 · PMCID: PMC10618084

Jobst, A., Brakemeier, E.-L., Buchheim, A., Caspar, F., Cuijpers, P., Ebmeier, K. P., Falkai, P., Jan van der Gaag, R., Gaebel, W., Herpertz, S., Kurimay, T., Sabaß, L., Schnell, K., Schramm, E., Torrent, C., Wasserman, D., Wiersma, J., & Padberg, F. (2016). European Psychiatric Association Guidance on psychotherapy in chronic depression across Europe. European Psychiatry, 33, 18-36. https://doi.org/10.1016/j.eurpsy.2015.12.003 PMID: 26854984

Kaiser, R. H., Andrews-Hanna, J. R., Wager, T. D., & Pizzagalli, D. A. (2015). Large-scale network dysfunction in major depressive disorder: A meta-analysis of resting-state functional connectivity. JAMA Psychiatry, 72(6), 603-611. https://doi.org/10.1001/jamapsychiatry.2015.0071 PMID: 25785575 · PMCID: PMC4456260

Kendell, R. E. (1976). The classification of depressions: A review of contemporary confusion. British Journal of Psychiatry, 129(1), 15-28. https://doi.org/10.1192/bjp.129.1.15 PMID: 938800

Kendler, K. S. (2016). The phenomenology of major depression and the representativeness and nature of DSM criteria. American Journal of Psychiatry, 173(8), 771-780. https://doi.org/10.1176/appi.ajp.2016.15121509 PMID: 27138588

Kendler, K. S. (2020). The origin of our modern concept of depression—The history of melancholia from 1780-1880: A review. JAMA Psychiatry, 77(8), 863-868. https://doi.org/10.1001/jamapsychiatry.2019.4709 PMID: 31995137

Kendler, K. S., Gardner, C. O., & Prescott, C. A. (2002). Toward a comprehensive developmental model for major depression in women. American Journal of Psychiatry, 159(7), 1133-1145. https://doi.org/10.1176/appi.ajp.159.7.1133 PMID: 12091191

Kendler, K. S., Gardner, C. O., & Prescott, C. A. (2006). Toward a comprehensive developmental model for major depression in men. American Journal of Psychiatry, 163(1), 115-124. https://doi.org/10.1176/appi.ajp.163.1.115 PMID: 16390898

Kendler, K. S., Muñoz, R. A., & Murphy, G. (2010). The development of the Feighner criteria: A historical perspective. American Journal of Psychiatry, 167(2), 134-142. https://doi.org/10.1176/appi.ajp.2009.09081155 PMID: 20008944

Kessler, R. C., & Bromet, E. J. (2013). The epidemiology of depression across cultures. Annual Review of Public Health, 34, 119-138. https://doi.org/10.1146/annurev-publhealth-031912-114409 PMID: 23514317 · PMCID: PMC4100461

Kessler, R. C., Berglund, P., Demler, O., Jin, R., Koretz, D., Merikangas, K. R., Rush, A. J., Walters, E. E., & Wang, P. S. (2003). The epidemiology of major depressive disorder: Results from the National Comorbidity Survey Replication (NCS-R). JAMA, 289(23), 3095-3105. https://doi.org/10.1001/jama.289.23.3095 PMID: 12813115

Kirsch, I., Deacon, B. J., Huedo-Medina, T. B., Scoboria, A., Moore, T. J., & Johnson, B. T. (2008). Initial severity and antidepressant benefits: A meta-analysis of data submitted to the Food and Drug Administration. PLoS Medicine, 5(2), e45. https://doi.org/10.1371/journal.pmed.0050045 PMID: 18303940 · PMCID: PMC2253608

Klein, D. N. (2008). Classification of depressive disorders in the DSM-V: Proposal for a two-dimension system. Journal of Abnormal Psychology, 117(3), 552-560. https://doi.org/10.1037/0021-843x.117.3.552 PMID: 18729608 · PMCID: PMC3057920

Kleinman, A. (2004). Culture and depression. New England Journal of Medicine, 351(10), 951-953. https://doi.org/10.1056/nejmp048078 PMID: 15342799

Köhler, C. A., Freitas, T. H., Maes, M., de Andrade, N. Q., Liu, C. S., Fernandes, B. S., Stubbs, B., Solmi, M., Veronese, N., Herrmann, N., Raison, C. L., Miller, B. J., Lanctôt, K. L., & Carvalho, A. F. (2017). Peripheral cytokine and chemokine alterations in depression: A meta-analysis of 82 studies. Acta Psychiatrica Scandinavica, 135(5), 373-387. https://doi.org/10.1111/acps.12698 PMID: 28122130

Kotov, R., Gamez, W., Schmidt, F., & Watson, D. (2010). Linking "big" personality traits to anxiety, depressive, and substance use disorders: A meta-analysis. Psychological Bulletin, 136(5), 768-821. https://doi.org/10.1037/a0020327 PMID: 20804236

Kotov, R., Krueger, R. F., Watson, D., Achenbach, T. M., Althoff, R. R., Bagby, R. M., ... Zimmerman, M. (2017). The Hierarchical Taxonomy of Psychopathology (HiTOP): A dimensional alternative to traditional nosologies. Journal of Abnormal Psychology, 126(4), 454-477. https://doi.org/10.1037/abn0000258 PMID: 28333488

Koukopoulos, A., & Sani, G. (2014). DSM-5 criteria for depression with mixed features: A farewell to mixed depression. Acta Psychiatrica Scandinavica, 129(1), 4-16. https://doi.org/10.1111/acps.12140 PMID: 23600771

Koukopoulos, A., Sani, G., & Ghaemi, S. N. (2013). Mixed features of depression: why DSM-5 is wrong (and so was DSM-IV). British Journal of Psychiatry, 203(1), 3-5. https://doi.org/10.1192/bjp.bp.112.124404 PMID: 23818531

Kriston, L., von Wolff, A., Westphal, A., Hölzel, L. P., & Härter, M. (2014). Efficacy and acceptability of acute treatments for persistent depressive disorder: A network meta-analysis. Depression and Anxiety, 31(8), 621-630. https://doi.org/10.1002/da.22236 PMID: 24448972

Kroenke, K., Spitzer, R. L., & Williams, J. B. W. (2001). The PHQ-9: Validity of a brief depression severity measure. Journal of General Internal Medicine, 16(9), 606-613. https://doi.org/10.1046/j.1525-1497.2001.016009606.x PMID: 11556941 · PMCID: PMC1495268

Kuyken, W., Ball, S., Crane, C., Ganguli, P., Jones, B., Montero-Marin, J., ... Williams, J. M. G. (2022). Effectiveness and cost-effectiveness of universal school-based mindfulness training compared with normal school provision in reducing risk of mental health problems and promoting well-being in adolescence: the MYRIAD cluster randomised controlled trial. Evidence-Based Mental Health, 25(3), 99-109. https://doi.org/10.1136/ebmental-2021-300396 PMID: 35820992 · PMCID: PMC9340028

Kuyken, W., Warren, F. C., Taylor, R. S., Whalley, B., Crane, C., Bondolfi, G., Hayes, R., Huijbers, M., Ma, H., Schweizer, S., Segal, Z., Speckens, A., Teasdale, J. D., Van Heeringen, K., Williams, M., Byford, S., Byng, R., & Dalgleish, T. (2016). Efficacy of mindfulness-based cognitive therapy in prevention of depressive relapse: An individual patient data meta-analysis from randomized trials. JAMA Psychiatry, 73(6), 565-574. https://doi.org/10.1001/jamapsychiatry.2016.0076 PMID: 27119968 · PMCID: PMC6640038

Lam, R. W., Kennedy, S. H., Adams, C., Bahji, A., Beaulieu, S., Bhat, V., Blier, P., Blumberger, D. M., Brietzke, E., Chakrabarty, T., Do, A., Frey, B. N., Giacobbe, P., Gratzer, D., Grigoriadis, S., Habert, J., Husain, M. I., Ismail, Z., McGirr, A., . . . Milev, R. V. (2024). Canadian Network for Mood and Anxiety Treatments (CANMAT) 2023 update on clinical guidelines for management of major depressive disorder in adults. The Canadian Journal of Psychiatry, 69(9), 641-687. https://doi.org/10.1177/07067437241245384 PMID: 38711351 · PMCID: PMC11351064

Lambert, S., Brahim, L. O., McCusker, J., Coumoundouros, C., Audet, L.-A., Yaffe, M., Kayser, J., Magalhaes, M., Belzile, E., & Turner, N. (2021). Non-pharmacological interventions for caregivers with depression and caregivers of care recipients with co-morbid depression: Systematic review and meta-analysis. Journal of General Internal Medicine, 36(10), 3159-3178. https://doi.org/10.1007/s11606-021-06891-6 PMID: 34159546 · PMCID: PMC8481392

Lamers, F., van Oppen, P., Comijs, H. C., Smit, J. H., Spinhoven, P., van Balkom, A. J. L. M., Nolen, W. A., Zitman, F. G., Beekman, A. T. F., & Penninx, B. W. J. H. (2011). Comorbidity patterns of anxiety and depressive disorders in a large cohort study: The Netherlands Study of Depression and Anxiety (NESDA). The Journal of Clinical Psychiatry, 72(3), 341-348. https://doi.org/10.4088/JCP.10m06176blu PMID: 21294994

Large, M., Kaneson, M., Myles, N., Myles, H., Gunaratne, P., & Ryan, C. (2016). Meta-analysis of longitudinal cohort studies of suicide risk assessment among psychiatric patients: Heterogeneity in results and lack of improvement over time. PLoS ONE, 11(6), e0156322. https://doi.org/10.1371/journal.pone.0156322 PMID: 27285387 · PMCID: PMC4902221

Lee, Y. Y., Stockings, E. A., Harris, M. G., Doi, S. A. R., Page, I. S., Davidson, S. K., & Barendregt, J. J. (2019). The risk of developing major depression among individuals with subthreshold depression: A systematic review and meta-analysis of longitudinal cohort studies. Psychological Medicine, 49(1), 92-102. https://doi.org/10.1017/S0033291718000557 PMID: 29530112

Leibenluft, E. (2011). Severe mood dysregulation, irritability, and the diagnostic boundaries of bipolar disorder in youths. American Journal of Psychiatry, 168(2), 129-142. https://doi.org/10.1176/appi.ajp.2010.10050766 PMID: 21123313 · PMCID: PMC3396206

Lemmens, L. H. J. M., Müller, V. N. L. S., Arntz, A., & Huibers, M. J. H. (2016). Mechanisms of change in psychotherapy for depression: An empirical update and evaluation of research aimed at identifying psychological mediators. Clinical Psychology Review, 50, 95-107. https://doi.org/10.1016/j.cpr.2016.09.004 PMID: 27770716

LeMoult, J., & Gotlib, I. H. (2019). Depression: A cognitive perspective. Clinical Psychology Review, 69, 51-66. https://doi.org/10.1016/j.cpr.2018.06.008 PMID: 29961601 · PMCID: PMC11884012

Lerner, D., & Henke, R. M. (2008). What does research tell us about depression, job performance, and work productivity? Journal of Occupational and Environmental Medicine, 50(4), 401-410. https://doi.org/10.1097/jom.0b013e31816bae50 PMID: 18404013

Levis, B., Benedetti, A., Thombs, B. D., & DEPRESsion Screening Data (DEPRESSD) Collaboration. (2019). Accuracy of Patient Health Questionnaire-9 (PHQ-9) for screening to detect major depression: Individual participant data meta-analysis. BMJ, 365, l1476. https://doi.org/10.1136/bmj.l1476 PMID: 30967483 · PMCID: PMC6454318

Lewis, G., Marston, L., Duffy, L., Freemantle, N., Gilbody, S., Hunter, R., Kendrick, T., Kessler, D., Mangin, D., King, M., Lanham, P., Moore, M., Nazareth, I., Wiles, N., Bacon, F., Bird, M., Brabyn, S., Burns, A., Clarke, C. S., Hunt, A., Pervin, J., & Lewis, G. (2021). Maintenance or discontinuation of antidepressants in primary care. New England Journal of Medicine, 385(14), 1257-1267. https://doi.org/10.1056/NEJMoa2106356 PMID: 34587384

Lichtman, J. H., Froelicher, E. S., Blumenthal, J. A., Carney, R. M., Doering, L. V., Frasure-Smith, N., Freedland, K. E., Jaffe, A. S., Leifheit-Limson, E. C., Sheps, D. S., Vaccarino, V., & Wulsin, L. (2014). Depression as a risk factor for poor prognosis among patients with acute coronary syndrome: Systematic review and recommendations: A scientific statement from the American Heart Association. Circulation, 129(12), 1350-1369. https://doi.org/10.1161/CIR.0000000000000019 PMID: 24566200

Lin, K., Xu, G., Lu, W., Ouyang, H., Dang, Y., Lorenzo-Seva, U., Guo, Y., Bessonov, D., Akiskal, H. S., So, K.-F., & Lee, T. M. C. (2014). Neuropsychological performance in melancholic, atypical and undifferentiated major depression during depressed and remitted states: A prospective longitudinal study. Journal of Affective Disorders, 168, 184-191. https://doi.org/10.1016/j.jad.2014.06.032 PMID: 25061955

Liu, R. T., & Alloy, L. B. (2010). Stress generation in depression: A systematic review of the empirical literature and recommendations for future study. Clinical Psychology Review, 30(5), 582-593. https://doi.org/10.1016/j.cpr.2010.04.010 PMID: 20478648 · PMCID: PMC3049314

Liu, R. T., Kleiman, E. M., Nestor, B. A., & Cheek, S. M. (2015). The hopelessness theory of depression: A quarter-century in review. Clinical Psychology: Science and Practice, 22(4), 345-365. https://doi.org/10.1111/cpsp.12125 PMID: 26709338 · PMCID: PMC4689589

Lorenzo-Luaces, L., Buss, J. F., & Fried, E. I. (2021). Heterogeneity in major depression and its melancholic and atypical specifiers: a secondary analysis of STARD. BMC Psychiatry, 21(1), 454. https://doi.org/10.1186/s12888-021-03444-3 PMID: 34530785 · PMCID: PMC8447832

Lund, C., Brooke-Sumner, C., Baingana, F., Baron, E. C., Breuer, E., Chandra, P., Haushofer, J., Herrman, H., Jordans, M., Kieling, C., Medina-Mora, M. E., Morgan, E., Omigbodun, O., Tol, W., Patel, V., & Saxena, S. (2018). Social determinants of mental disorders and the Sustainable Development Goals: A systematic review of reviews. The Lancet Psychiatry, 5(4), 357-369. https://doi.org/10.1016/S2215-0366(18)30060-9 PMID: 29580610

Maier, S. F., & Seligman, M. E. P. (2016). Learned helplessness at fifty: Insights from neuroscience. Psychological Review, 123(4), 349-367. https://doi.org/10.1037/rev0000033 PMID: 27337390 · PMCID: PMC4920136

Maj, M., Pirozzi, R., Magliano, L., Fiorillo, A., & Bartoli, L. (2007). Phenomenology and prognostic significance of delusions in major depressive disorder: A 10-year prospective follow-up study. The Journal of Clinical Psychiatry, 68(9), 1411-1417. https://doi.org/10.4088/JCP.v68n0913 PMID: 17915981

Malhi, G. S., & Mann, J. J. (2018). Depression. The Lancet, 392(10161), 2299-2312. https://doi.org/10.1016/S0140-6736(18)31948-2 PMID: 30396512

McIntyre, R. S., Rosenblat, J. D., Nemeroff, C. B., Sanacora, G., Murrough, J. W., Berk, M., Brietzke, E., Dodd, S., Gorwood, P., Ho, R., Iosifescu, D. V., Lopez Jaramillo, C., Kasper, S., Kratiuk, K., Lee, J. G., Lee, Y., Lui, L. M. W., Mansur, R. B., Papakostas, G. I., Subramaniapillai, M., Thase, M., Vieta, E., Young, A. H., Zarate, C. A., Jr., & Stahl, S. (2021). Synthesizing the evidence for ketamine and esketamine in treatment-resistant depression: An international expert opinion on the available evidence and implementation. American Journal of Psychiatry, 178(5), 383-399. https://doi.org/10.1176/appi.ajp.2020.20081251 PMID: 33726522 · PMCID: PMC9635017

Mendels, J., & Cochrane, C. (1968). The nosology of depression: The endogenous-reactive concept. American Journal of Psychiatry, 124(11, Suppl.), 1-11. https://doi.org/10.1176/ajp.124.11s.1 PMID: 4872231

Mezuk, B., Eaton, W. W., Albrecht, S., & Golden, S. H. (2008). Depression and type 2 diabetes over the lifespan: A meta-analysis. Diabetes Care, 31(12), 2383-2390. https://doi.org/10.2337/dc08-0985 PMID: 19033418 · PMCID: PMC2584200

Molendijk, M. L., Spinhoven, P., Polak, M., Bus, B. A. A., Penninx, B. W. J. H., & Elzinga, B. M. (2014). Serum BDNF concentrations as peripheral manifestations of depression: Evidence from a systematic review and meta-analyses on 179 associations (N=9484). Molecular Psychiatry, 19(7), 791-800. https://doi.org/10.1038/mp.2013.105 PMID: 23958957

Möller, H.-J., Bandelow, B., Volz, H.-P., Barnikol, U. B., Seifritz, E., & Kasper, S. (2016). The relevance of 'mixed anxiety and depression' as a diagnostic category in clinical practice. European Archives of Psychiatry and Clinical Neuroscience, 266(8), 725-736. https://doi.org/10.1007/s00406-016-0684-7 PMID: 27002521 · PMCID: PMC5097109

Moncrieff, J., Cooper, R. E., Stockmann, T., Amendola, S., Hengartner, M. P., & Horowitz, M. A. (2023). The serotonin theory of depression: A systematic umbrella review of the evidence. Molecular Psychiatry, 28(8), 3243-3256. https://doi.org/10.1038/s41380-022-01661-0 PMID: 35854107 · PMCID: PMC10618090

Monroe, S. M., & Harkness, K. L. (2005). Life stress, the kindling hypothesis, and the recurrence of depression: Considerations from a life stress perspective. Psychological Review, 112(2), 417-445. https://doi.org/10.1037/0033-295X.112.2.417 PMID: 15783292

Monroe, S. M., & Simons, A. D. (1991). Diathesis-stress theories in the context of life stress research: Implications for the depressive disorders. Psychological Bulletin, 110(3), 406-425. https://doi.org/10.1037/0033-2909.110.3.406 PMID: 1758917

Montgomery, S. A., & Åsberg, M. (1979). A new depression scale designed to be sensitive to change. The British Journal of Psychiatry, 134(4), 382-389. https://doi.org/10.1192/bjp.134.4.382 PMID: 444788

Moussavi, S., Chatterji, S., Verdes, E., Tandon, A., Patel, V., & Ustun, B. (2007). Depression, chronic diseases, and decrements in health: Results from the World Health Surveys. The Lancet, 370(9590), 851-858. https://doi.org/10.1016/S0140-6736(07)61415-9 PMID: 17826170

Na, K.-S., Kang, J. M., & Cho, S.-E. (2021). Prevalence of DSM-5 mixed features: A meta-analysis and systematic review. Journal of Affective Disorders, 282, 203-210. https://doi.org/10.1016/j.jad.2020.12.149 PMID: 33418368

Noetel, M., Sanders, T., Gallardo-Gómez, D., Taylor, P., del Pozo Cruz, B., van den Hoek, D., Smith, J. J., Mahoney, J., Spathis, J., Moresi, M., Pagano, R., Pagano, L., Vasconcellos, R., Arnott, H., Varley, B., Parker, P., Biddle, S., & Lonsdale, C. (2024). Effect of exercise for depression: Systematic review and network meta-analysis of randomised controlled trials. BMJ, 384, e075847. https://doi.org/10.1136/bmj-2023-075847 PMID: 38355154 · PMCID: PMC10870815

Nolen-Hoeksema, S. (1987). Sex differences in unipolar depression: evidence and theory. Psychological Bulletin, 101(2), 259-282. https://doi.org/10.1037/0033-2909.101.2.259

Nolen-Hoeksema, S. (2000). The role of rumination in depressive disorders and mixed anxiety/depressive symptoms. Journal of Abnormal Psychology, 109(3), 504-511. https://doi.org/10.1037/0021-843X.109.3.504

Łojko, D., & Rybakowski, J. K. (2017). Atypical depression: Current perspectives. Neuropsychiatric Disease and Treatment, 13, 2447-2456. https://doi.org/10.2147/NDT.S147317 PMID: 29033570 · PMCID: PMC5614762

Ormel, J., Jeronimus, B. F., Kotov, R., Riese, H., Bos, E. H., Hankin, B., Rosmalen, J. G. M., & Oldehinkel, A. J. (2013). Neuroticism and common mental disorders: meaning and utility of a complex relationship. Clinical Psychology Review, 33(5), 686-697. https://doi.org/10.1016/j.cpr.2013.04.003 PMID: 23702592 · PMCID: PMC4382368

Otte, C., Gold, S. M., Penninx, B. W., Pariante, C. M., Etkin, A., Fava, M., Mohr, D. C., & Schatzberg, A. F. (2016). Major depressive disorder. Nature Reviews Disease Primers, 2, 16065. https://doi.org/10.1038/nrdp.2016.65 PMID: 27629598

Paradies, Y., Ben, J., Denson, N., Elias, A., Priest, N., Pieterse, A., Gupta, A., Kelaher, M., & Gee, G. (2015). Racism as a determinant of health: A systematic review and meta-analysis. PLOS ONE, 10(9), e0138511. https://doi.org/10.1371/journal.pone.0138511 PMID: 26398658 · PMCID: PMC4580597

Parker, G., & Malhi, G. S. (2019). Persistent depression: Should such a DSM-5 diagnostic category persist? The Canadian Journal of Psychiatry, 64(3), 177-179. https://doi.org/10.1177/0706743718814429 PMID: 30453771 · PMCID: PMC6405812

Parker, G. (2000). Classifying depression: should paradigms lost be regained? American Journal of Psychiatry, 157(8), 1195-1203. https://doi.org/10.1176/appi.ajp.157.8.1195 PMID: 10910777

Parker, G., Fink, M., Shorter, E., Taylor, M. A., Akiskal, H., Berrios, G., Bolwig, T., Brown, W. A., Carroll, B., Healy, D., Klein, D. F., Koukopoulos, A., Michels, R., Paris, J., Rubin, R. T., Spitzer, R., & Swartz, C. (2010). Issues for DSM-5: Whither melancholia? The case for its classification as a distinct mood disorder. American Journal of Psychiatry, 167(7), 745-747. https://doi.org/10.1176/appi.ajp.2010.09101525 PMID: 20595426 · PMCID: PMC3733615

Pellicane, M. J., & Ciesla, J. A. (2022). Associations between minority stress, depression, and suicidal ideation and attempts in transgender and gender diverse (TGD) individuals: Systematic review and meta-analysis. Clinical Psychology Review, 91, 102113. https://doi.org/10.1016/j.cpr.2021.102113 PMID: 34973649

Pezawas, L., Angst, J., & Kasper, S. (2005). Recurrent brief depression revisited. International Review of Psychiatry, 17(1), 63-70. https://doi.org/10.1080/00207390500064650 PMID: 16194772

Pizzagalli, D. A. (2014). Depression, stress, and anhedonia: toward a synthesis and integrated model. Annual Review of Clinical Psychology, 10, 393-423. https://doi.org/10.1146/annurev-clinpsy-050212-185606 PMID: 24471371 · PMCID: PMC3972338

Pizzagalli, D. A., Holmes, A. J., Dillon, D. G., Goetz, E. L., Birk, J. L., Bogdan, R., Dougherty, D. D., Iosifescu, D. V., Rauch, S. L., & Fava, M. (2009). Reduced caudate and nucleus accumbens response to rewards in unmedicated individuals with major depressive disorder. American Journal of Psychiatry, 166(6), 702-710. https://doi.org/10.1176/appi.ajp.2008.08081201 PMID: 19411368 · PMCID: PMC2735451

Plöderl, M., & Tremblay, P. (2015). Mental health of sexual minorities. A systematic review. International Review of Psychiatry, 27(5), 367-385. https://doi.org/10.3109/09540261.2015.1083949 PMID: 26552495

Prigerson, H. G., Boelen, P. A., Xu, J., Smith, K. V., & Maciejewski, P. K. (2021). Validation of the new DSM-5-TR criteria for prolonged grief disorder and the PG-13-Revised (PG-13-R) scale. World Psychiatry, 20(1), 96-106. https://doi.org/10.1002/wps.20823 PMID: 33432758 · PMCID: PMC7801836

Prisciandaro, J. J., & Roberts, J. E. (2005). A taxometric investigation of unipolar depression in the National Comorbidity Survey. Journal of Abnormal Psychology, 114(4), 718-728. https://doi.org/10.1037/0021-843x.114.4.718 PMID: 16351392

Radoš, S. N., Akik, B. K., Žutić, M., Rodriguez-Muñoz, M. F., Uriko, K., Motrico, E., Moreno-Peral, P., Apter, G., & Lambregtse-van den Berg, M. (2024). Diagnosis of peripartum depression disorder: A state-of-the-art approach from the COST Action Riseup-PPD. Comprehensive Psychiatry, 130, 152456. https://doi.org/10.1016/j.comppsych.2024.152456 PMID: 38306851

Rapaport, M. H., Clary, C., Fayyad, R., & Endicott, J. (2005). Quality-of-life impairment in depressive and anxiety disorders. American Journal of Psychiatry, 162(6), 1171-1178. https://doi.org/10.1176/appi.ajp.162.6.1171 PMID: 15930066

Regier, D. A., Narrow, W. E., Clarke, D. E., Kraemer, H. C., Kuramoto, S. J., Kuhl, E. A., & Kupfer, D. J. (2013). DSM-5 field trials in the United States and Canada, Part II: Test-retest reliability of selected categorical diagnoses. American Journal of Psychiatry, 170(1), 59-70. https://doi.org/10.1176/appi.ajp.2012.12070999 PMID: 23111466

Rhebergen, D., & Graham, R. (2014). The re-labelling of dysthymic disorder to persistent depressive disorder in DSM-5: Old wine in new bottles? Current Opinion in Psychiatry, 27(1), 27-31. https://doi.org/10.1097/yco.0000000000000022 PMID: 24270481

Ribeiro, J. D., Huang, X., Fox, K. R., & Franklin, J. C. (2018). Depression and hopelessness as risk factors for suicide ideation, attempts and death: Meta-analysis of longitudinal studies. British Journal of Psychiatry, 212(5), 279-286. https://doi.org/10.1192/bjp.2018.27 PMID: 29587888

Ridley, M., Rao, G., Schilbach, F., & Patel, V. (2020). Poverty, depression, and anxiety: Causal evidence and mechanisms. Science, 370(6522), eaay0214. https://doi.org/10.1126/science.aay0214 PMID: 33303583

Riemann, D., Krone, L. B., Wulff, K., & Nissen, C. (2020). Sleep, insomnia, and depression. Neuropsychopharmacology, 45(1), 74-89. https://doi.org/10.1038/s41386-019-0411-y PMID: 31071719 · PMCID: PMC6879516

Robinson, R. G., & Jorge, R. E. (2016). Post-stroke depression: A review. American Journal of Psychiatry, 173(3), 221-231. https://doi.org/10.1176/appi.ajp.2015.15030363 PMID: 26684921

Rock, P. L., Roiser, J. P., Riedel, W. J., & Blackwell, A. D. (2014). Cognitive impairment in depression: A systematic review and meta-analysis. Psychological Medicine, 44(10), 2029-2040. https://doi.org/10.1017/s0033291713002535 PMID: 24168753

Rodebaugh, T. L., Scullin, R. B., Langer, J. K., Dixon, D. J., Huppert, J. D., Bernstein, A., Zvielli, A., & Lenze, E. J. (2016). Unreliability as a threat to understanding psychopathology: The cautionary tale of attentional bias. Journal of Abnormal Psychology, 125(6), 840–851. https://doi.org/10.1037/abn0000184 PMID: 27322741 · PMCID: PMC4980228

Rohan, K. J., Meyerhoff, J., Ho, S.-Y., Evans, M., Postolache, T. T., & Vacek, P. M. (2016). Outcomes one and two winters following cognitive-behavioral therapy or light therapy for seasonal affective disorder. American Journal of Psychiatry, 173(3), 244-251. https://doi.org/10.1176/appi.ajp.2015.15060773 PMID: 26539881 · PMCID: PMC4939843

Rush, A. J., Trivedi, M. H., Ibrahim, H. M., Carmody, T. J., Arnow, B., Klein, D. N., Markowitz, J. C., Ninan, P. T., Kornstein, S., Manber, R., Thase, M. E., Kocsis, J. H., & Keller, M. B. (2003). The 16-Item Quick Inventory of Depressive Symptomatology (QIDS), clinician rating (QIDS-C), and self-report (QIDS-SR): A psychometric evaluation in patients with chronic major depression. Biological Psychiatry, 54(5), 573-583. https://doi.org/10.1016/s0006-3223(02)01866-8 PMID: 12946886

Rush, A. J., Trivedi, M. H., Wisniewski, S. R., Nierenberg, A. A., Stewart, J. W., Warden, D., Niederehe, G., Thase, M. E., Lavori, P. W., Lebowitz, B. D., McGrath, P. J., Rosenbaum, J. F., Sackeim, H. A., Kupfer, D. J., Luther, J., & Fava, M. (2006). Acute and longer-term outcomes in depressed outpatients requiring one or several treatment steps: A STARD report. American Journal of Psychiatry, 163(11), 1905-1917. https://doi.org/10.1176/ajp.2006.163.11.1905 PMID: 17074942

Ryder, A. G., Yang, J., Zhu, X., Yao, S., Yi, J., Heine, S. J., & Bagby, R. M. (2008). The cultural shaping of depression: Somatic symptoms in China, psychological symptoms in North America? Journal of Abnormal Psychology, 117(2), 300-313. https://doi.org/10.1037/0021-843X.117.2.300 PMID: 18489206

Rytwinski, N. K., Scur, M. D., Feeny, N. C., & Youngstrom, E. A. (2013). The co-occurrence of major depressive disorder among individuals with posttraumatic stress disorder: A meta-analysis. Journal of Traumatic Stress, 26(3), 299-309. https://doi.org/10.1002/jts.21814 PMID: 23696449

Salk, R. H., Hyde, J. S., & Abramson, L. Y. (2017). Gender differences in depression in representative national samples: Meta-analyses of diagnoses and symptoms. Psychological Bulletin, 143(8), 783-822. https://doi.org/10.1037/bul0000102 PMID: 28447828 · PMCID: PMC5532074

Santomauro, D. F., Mantilla Herrera, A. M., Shadid, J., et al. (COVID-19 Mental Disorders Collaborators). (2021). Global prevalence and burden of depressive and anxiety disorders in 204 countries and territories in 2020 due to the COVID-19 pandemic. The Lancet, 398(10312), 1700-1712. https://doi.org/10.1016/S0140-6736(21)02143-7 PMID: 34634250 · PMCID: PMC8500697

Schramm, E., Klein, D. N., Elsaesser, M., Furukawa, T. A., & Domschke, K. (2020). Review of dysthymia and persistent depressive disorder: History, correlates, and clinical implications. The Lancet Psychiatry, 7(9), 801-812. https://doi.org/10.1016/S2215-0366(20)30099-7 PMID: 32828168

Schuckit, M. A., Smith, T. L., Danko, G. P., Pierson, J., Trim, R., Nurnberger, J. I., Kramer, J., Kuperman, S., Bierut, L. J., & Hesselbrock, V. (2007). A comparison of factors associated with substance-induced versus independent depressions. Journal of Studies on Alcohol and Drugs, 68(6), 805-812. https://doi.org/10.15288/jsad.2007.68.805 PMID: 17960298

Semkovska, M., & McLoughlin, D. M. (2010). Objective cognitive performance associated with electroconvulsive therapy for depression: A systematic review and meta-analysis. Biological Psychiatry, 68(6), 568-577. https://doi.org/10.1016/j.biopsych.2010.06.009 PMID: 20673880

Slavich, G. M., & Irwin, M. R. (2014). From stress to inflammation and major depressive disorder: A social signal transduction theory of depression. Psychological Bulletin, 140(3), 774-815. https://doi.org/10.1037/a0035302 PMID: 24417575 · PMCID: PMC4006295

Snyder, H. R. (2013). Major depressive disorder is associated with broad impairments on neuropsychological measures of executive function: A meta-analysis and review. Psychological Bulletin, 139(1), 81-132. https://doi.org/10.1037/a0028727 PMID: 22642228 · PMCID: PMC3436964

Sobieraj, D. M., Martinez, B. K., Hernandez, A. V., Coleman, C. I., Ross, J. S., Berg, K. M., Steffens, D. C., & Baker, W. L. (2019). Adverse effects of pharmacologic treatments of major depression in older adults. Journal of the American Geriatrics Society, 67(8), 1571-1581. https://doi.org/10.1111/jgs.15966 PMID: 31140587

Solmi, M., Pigato, G. G., Roiter, B., Guaglianone, A., Martini, L., Fornaro, M., Monaco, F., Carvalho, A. F., Stubbs, B., Veronese, N., & Correll, C. U. (2018). Prevalence of catatonia and its moderators in clinical samples: Results from a meta-analysis and meta-regression analysis. Schizophrenia Bulletin, 44(5), 1133-1150. https://doi.org/10.1093/schbul/sbx157 PMID: 29140521 · PMCID: PMC6101628

Soumerai, S. B., Koppel, R., Naci, H., Madden, J. M., Fry, A., Halbisen, A., Angeles, J., Koppel, J., Rubin, R., & Lu, C. Y. (2024). Intended and unintended outcomes after FDA pediatric antidepressant warnings: A systematic review. Health Affairs, 43(10), 1360-1369. https://doi.org/10.1377/hlthaff.2023.00263 PMID: 39374452

Spielmans, G. I., Spence-Sing, T., & Parry, P. (2020). Duty to warn: Antidepressant black box suicidality warning is empirically justified. Frontiers in Psychiatry, 11, Article 18. https://doi.org/10.3389/fpsyt.2020.00018 PMID: 32116839 · PMCID: PMC7031767

Spitzer, R. L., Endicott, J., & Robins, E. (1978). Research diagnostic criteria: Rationale and reliability. Archives of General Psychiatry, 35(6), 773-782. https://doi.org/10.1001/archpsyc.1978.01770300115013 PMID: 655775

Stetler, C., & Miller, G. E. (2011). Depression and hypothalamic-pituitary-adrenal activation: A quantitative summary of four decades of research. Psychosomatic Medicine, 73(2), 114-126. https://doi.org/10.1097/PSY.0b013e31820ad12b PMID: 21257974

Stewart, D. E., & Vigod, S. (2016). Postpartum depression. New England Journal of Medicine, 375(22), 2177-2186. https://doi.org/10.1056/NEJMcp1607649 PMID: 27959754

Stone, M., Laughren, T., Jones, M. L., Levenson, M., Holland, P. C., Hughes, A., Hammad, T. A., Temple, R., & Rochester, G. (2009). Risk of suicidality in clinical trials of antidepressants in adults: analysis of proprietary data submitted to US Food and Drug Administration. BMJ, 339, b2880. https://doi.org/10.1136/bmj.b2880 PMID: 19671933 · PMCID: PMC2725270

Sullivan, P. F., Neale, M. C., & Kendler, K. S. (2000). Genetic epidemiology of major depression: Review and meta-analysis. American Journal of Psychiatry, 157(10), 1552-1562. https://doi.org/10.1176/appi.ajp.157.10.1552 PMID: 11007705

Telford, C., McCarthy-Jones, S., Corcoran, R., & Rowse, G. (2012). Experience sampling methodology studies of depression: The state of the art. Psychological Medicine, 42(6), 1119-1129. https://doi.org/10.1017/S0033291711002200 PMID: 22008511

Tozzi, L., Zhang, X., Pines, A., Olmsted, A. M., Zhai, E. S., Anene, E. T., Chesnut, M., Holt-Gosselin, B., Chang, S., Stetz, P. C., Ramirez, C. A., Hack, L. M., Korgaonkar, M. S., Wintermark, M., Gotlib, I. H., Ma, J., & Williams, L. M. (2024). Personalized brain circuit scores identify clinically distinct biotypes in depression and anxiety. Nature Medicine, 30(7), 2076-2087. https://doi.org/10.1038/s41591-024-03057-9 PMID: 38886626 · PMCID: PMC11271415

Treadway, M. T., & Zald, D. H. (2011). Reconsidering anhedonia in depression: lessons from translational neuroscience. Neuroscience & Biobehavioral Reviews, 35(3), 537-555. https://doi.org/10.1016/j.neubiorev.2010.06.006 PMID: 20603146 · PMCID: PMC3005986

Turner, E. H. (2019). Esketamine for treatment-resistant depression: Seven concerns about efficacy and FDA approval. The Lancet Psychiatry, 6(12), 977-979. https://doi.org/10.1016/S2215-0366(19)30394-3 PMID: 31680014

Uher, R., Payne, J. L., Pavlova, B., & Perlis, R. H. (2014). Major depressive disorder in DSM-5: Implications for clinical practice and research of changes from DSM-IV. Depression and Anxiety, 31(6), 459-471. https://doi.org/10.1002/da.22217 PMID: 24272961

UK ECT Review Group. (2003). Efficacy and safety of electroconvulsive therapy in depressive disorders: A systematic review and meta-analysis. The Lancet, 361(9360), 799-808. https://doi.org/10.1016/S0140-6736(03)12705-5 PMID: 12642045

US Preventive Services Task Force. (2023). Screening for depression and suicide risk in adults: US Preventive Services Task Force recommendation statement. JAMA, 329(23), 2057-2067. https://doi.org/10.1001/jama.2023.9297 PMID: 37338872

van Loo, H. M., de Jonge, P., Romeijn, J.-W., Kessler, R. C., & Schoevers, R. A. (2012). Data-driven subtypes of major depressive disorder: A systematic review. BMC Medicine, 10, 156. https://doi.org/10.1186/1741-7015-10-156 PMID: 23210727 · PMCID: PMC3566979

Vidal-Ribas, P., Brotman, M. A., Valdivieso, I., Leibenluft, E., & Stringaris, A. (2016). The status of irritability in psychiatry: A conceptual and quantitative review. Journal of the American Academy of Child & Adolescent Psychiatry, 55(7), 556-570. https://doi.org/10.1016/j.jaac.2016.04.014 PMID: 27343883 · PMCID: PMC4927461

Viduani, A., Arenas, D. L., Benetti, S., Wahid, S. S., Kohrt, B. A., & Kieling, C. (2024). Systematic review and meta-synthesis: How is depression experienced by adolescents? A synthesis of the qualitative literature. Journal of the American Academy of Child and Adolescent Psychiatry, 63(10), 970-990. https://doi.org/10.1016/j.jaac.2023.11.013 PMID: 38340896

Wakefield, J. C., & First, M. B. (2012). Validity of the bereavement exclusion to major depression: Does the empirical evidence support the proposal to eliminate the exclusion in DSM-5? World Psychiatry, 11(1), 3-10. https://doi.org/10.1016/j.wpsyc.2012.01.002 PMID: 22294996 · PMCID: PMC3266762

Walker, W. H., II, Walton, J. C., DeVries, A. C., & Nelson, R. J. (2020). Circadian rhythm disruption and mental health. Translational Psychiatry, 10, Article 28. https://doi.org/10.1038/s41398-020-0694-0 PMID: 32066704 · PMCID: PMC7026420

Watson, D. (2005). Rethinking the mood and anxiety disorders: A quantitative hierarchical model for DSM-V. Journal of Abnormal Psychology, 114(4), 522-536. https://doi.org/10.1037/0021-843x.114.4.522 PMID: 16351375

Williams, J. M. G., Barnhofer, T., Crane, C., Herman, D., Raes, F., Watkins, E., & Dalgleish, T. (2007). Autobiographical memory specificity and emotional disorder. Psychological Bulletin, 133(1), 122-148. https://doi.org/10.1037/0033-2909.133.1.122 PMID: 17201573 · PMCID: PMC2834574

World Health Organization. (2024). Clinical descriptions and diagnostic requirements for ICD-11 mental, behavioural and neurodevelopmental disorders (CDDR). World Health Organization.

Yan, C.-G., Chen, X., Li, L., Castellanos, F. X., Bai, T.-J., Bo, Q.-J., Cao, J., Chen, G.-M., Chen, N.-X., Chen, W., Cheng, C., Cheng, Y.-Q., Cui, X.-L., Duan, J., Fang, Y.-R., Gong, Q.-Y., Guo, W.-B., Hou, Z.-H., Hu, L., ... Zang, Y.-F. (2019). Reduced default mode network functional connectivity in patients with recurrent major depressive disorder. Proceedings of the National Academy of Sciences, 116(18), 9078-9083. https://doi.org/10.1073/pnas.1900390116 PMID: 30979801 · PMCID: PMC6500168

Zalsman, G., Hawton, K., Wasserman, D., van Heeringen, K., Arensman, E., Sarchiapone, M., Carli, V., Höschl, C., Barzilay, R., Balazs, J., Purebl, G., Kahn, J. P., Sáiz, P. A., Lipsicas, C. B., Bobes, J., Cozman, D., Hegerl, U., & Zohar, J. (2016). Suicide prevention strategies revisited: 10-year systematic review. The Lancet Psychiatry, 3(7), 646-659. https://doi.org/10.1016/S2215-0366(16)30030-X PMID: 27289303

Zaninotto, L., Guglielmo, R., Calati, R., Ioime, L., Camardese, G., Janiri, L., Bria, P., & Serretti, A. (2015). Cognitive markers of psychotic unipolar depression: A meta-analytic study. Journal of Affective Disorders, 174, 580-588. https://doi.org/10.1016/j.jad.2014.11.027 PMID: 25560194

Zanos, P., & Gould, T. D. (2018). Mechanisms of ketamine action as an antidepressant. Molecular Psychiatry, 23(4), 801-811. https://doi.org/10.1038/mp.2017.255 PMID: 29532791 · PMCID: PMC5999402

Zhu, X., Wang, X., Xiao, J., Liao, J., Zhong, M., Wang, W., & Yao, S. (2012). Evidence of a dissociation pattern in resting-state default mode network connectivity in first-episode, treatment-naive major depression patients. Biological Psychiatry, 71(7), 611-617. https://doi.org/10.1016/j.biopsych.2011.10.035 PMID: 22177602

Zimmerman, M., Ellison, W., Young, D., Chelminski, I., & Dalrymple, K. (2015). How many different ways do patients meet the diagnostic criteria for major depressive disorder? Comprehensive Psychiatry, 56, 29-34. https://doi.org/10.1016/j.comppsych.2014.09.007 PMID: 25266848

Zimmerman, M., Martin, J., McGonigal, P., Harris, L., Kerr, S., Balling, C., Kiefer, R., Stanton, K., & Dalrymple, K. (2019). Validity of the DSM-5 anxious distress specifier for major depressive disorder. Depression and Anxiety, 36(1), 31-38. https://doi.org/10.1002/da.22837 PMID: 30311733

Zisook, S., Corruble, E., Duan, N., Iglewicz, A., Karam, E. G., Lanouette, N., Lebowitz, B., Pies, R., Reynolds, C., Seay, K., Shear, M. K., Simon, N., & Young, I. T. (2012). The bereavement exclusion and DSM-5. Depression and Anxiety, 29(5), 425-443. https://doi.org/10.1002/da.21927 PMID: 22495967