You are on page 1of 20

Unipolar depression in adults: Course of illness

William Coryell, MD
Section Editor
Peter P Roy-Byrne, MD
Deputy Editor
David Solomon, MD
All topics are updated as new evidence becomes available and our peer review process is
Literature review current through: Mar 2017. | This topic last updated: Dec 21,

INTRODUCTION — Major depressive disorder (unipolar major depression) and
persistent depressive disorder (dysthymia) represent depressive syndromes that are
distinguished by the type and number of symptoms that occur as well as their duration.
Depressive symptoms can include depressed mood, loss of interest or pleasure in most or
all activities, insomnia or hypersomnia, change in appetite or weight, psychomotor
retardation or agitation, low energy, poor concentration, thoughts of worthlessness or
guilt, and recurrent thoughts about death or suicide [1]. The World Health Organization
estimated that major depressive disorder was the 11th greatest cause of disability and
mortality in the world among 291 diseases and causes of injuries [2].

Preliminary studies suggest that course of illness may be associated with structural brain
changes. As an example, a one-year prospective magnetic imaging study followed
patients with treatment resistant unipolar major depression (n = 26) who were treated
with pharmacotherapy, and found that remission was associated with subtle increases in
hippocampal volume and cortical thickness, whereas nonremission was associated with
decreased volume and thickness [3].

This topic reviews the course of illness in patients with major depressive disorder and
persistent depressive disorder (dysthymia). The course of illness in psychotic depression
and minor depression is discussed separately, as is the epidemiology, clinical features,
diagnosis, and treatment of depression:

●(See "Unipolar major depression with psychotic features: Maintenance treatment and
course of illness", section on 'Course of illness'.)
●(See "Unipolar minor depression in adults: Epidemiology, clinical presentation, and
diagnosis", section on 'Course of illness'.)
●(See "Unipolar depression in adults: Epidemiology, pathogenesis, and
●(See "Unipolar depression in adults: Assessment and diagnosis".)
●(See "Unipolar major depression in adults: Choosing initial treatment".)
●(See "Unipolar depression in adults: Treatment of resistant depression".)
●(See "Diagnosis and management of late-life unipolar depression".)

DEFINITIONS — Major depressive disorder (unipolar major depression) and persistent
depressive disorder (dysthymia) are defined in the American Psychiatric Association's
Diagnostic and Statistical Manual, Fifth Edition (DSM-5) [1]:

●Major depressive disorder is diagnosed in patients who have suffered at least one
major depressive episode (table 1). An episode is a period lasting at least two weeks,
with five or more of the following nine symptoms: depressed mood, loss of interest or
pleasure in most or all activities, insomnia or hypersomnia, change in appetite or
weight, psychomotor retardation or agitation, low energy, poor concentration, thoughts
of worthlessness or guilt, and recurrent thoughts about death or suicide.
●Persistent depressive disorder (dysthymia) is diagnosed in patients with depressed
mood for at least two years that is accompanied by at least two of the following
symptoms: decreased or increased appetite, insomnia or hypersomnia, low energy,
poor self-esteem, poor concentration, and hopelessness (table 2).

Additional information about the clinical presentation and diagnosis of major depressive
disorder and persistent depressive disorder (dysthymia) is discussed separately. (See
"Unipolar depression in adults: Assessment and diagnosis".)

The term recovery is used to indicate the resolution of a depressive episode [4]. Although
different definitions of recovery exist, many long-term observational studies require at
least two consecutive months with no more than one or two mild symptoms of depression
and no impairment of psychosocial functioning.

STUDY SETTING — The setting of a study can affect the observed course of illness for
depressed patients. Individuals who are identified in the community surveys may have a
more benign course than patients at tertiary care facilities; in either case, the individuals
or patients may not be receiving treatment [5,6]. In addition, patients seen in routine
clinical practice often differ from patients who are followed after they complete
randomized trials. Patients who participate in trials are usually recruited through
advertisements, are willing to risk assignment to placebo, and meet extensive exclusion
criteria that are typically used in industry-sponsored trials; thus, these patients may have
less suicidality, psychosis, comorbidity, and functional impairment [7-9].

MAJOR DEPRESSIVE DISORDER — The course of illness for major depressive
disorder (unipolar major depression) is heterogeneous, which may reflect that the
disorder represents a number of different illnesses that differ in their pathogenesis,
clinical presentation, and treatment response [10-12]. Patients may experience a single
major depressive episode, follow a highly recurrent course with full resolution of
symptoms between episodes, or spend much of their lives struggling with persistent,
fluctuating symptoms. Depressive episodes can range in intensity from states that
produce limited impairment and are little noticed by others, to catatonic or psychotic
conditions that render the patient incapable of self-care.

Although the large majority of major depressive episodes eventually end [5,13-15], some
patients are ill for much of their lives due to recurrences and/or lengthy episodes. A

which differ in treatment.) Recovery — The median time to recovery from a major depressive episode in prospective observational studies is approximately 20 weeks [13. section on 'Unipolar major depression'. Patients with bipolar disorder may suffer one or more episodes of major depression prior to their first manic or hypomanic syndrome. (See "Bipolar disorder in adults: Assessment and diagnosis". delusions and/or hallucinations) •Decreased need for sleep •Unusually high energy •Increased goal directed activity In addition. flat affect. In a 10-year follow-up study of 77 patients diagnosed with psychotic major depression at baseline. . and initially receive a diagnosis of major depressive disorder.prospective observational study of 431 patients with major depressive disorder who sought treatment at study intake and were then followed for up to 12 years found that they were depressed for 59 percent (mean average) of the follow-up time [6]. apathy. and for individuals in the community who have not sought treatment. poverty of speech. (See "Depression in schizophrenia". This finding is consistent in treated patients with multiple recurrent episodes (up to five).) At least 5 to 15 percent of patients initially diagnosed with major depressive disorder eventually change diagnosis to a schizophrenia-spectrum disorder [19-23]. The probability of switching diagnoses was higher in patients with the following: ●Younger age of onset of their first lifetime major depressive episode ●Family history of bipolar disorder ●During major depressive episodes: •Psychosis (eg. a prospective observational study assessed 550 patients with major depressive disorder at least annually for up to 31 years. the diagnosis changed to schizophrenia in 30 percent. Diagnostic stability — Patients who are initially and correctly diagnosed with major depressive disorder may eventually change diagnoses. As an example.23.24]. and abulia) and psychosocial impairment [24]. predictors included negative symptoms (eg. treatment resistance in patients with major depressive disorder and the presence of comorbid attention deficit hyperactivity disorder or substance use disorders are associated with diagnostic conversion to bipolar disorder [17. and found that 20 percent eventually changed diagnoses to bipolar disorder [16].14]. Additional information about distinguishing unipolar depression and bipolar depression.18]. Repeated diagnostic examinations find that the change from major depressive disorder to schizophrenia occurs more frequently than the reverse [20. is discussed separately.

26]. In addition. Thus.) There are several risk factors for a longer time to recovery from major depression.31]. reassurance and watchful waiting with short-term follow-up (eg.46-50] ●Childhood maltreatment [39. delusions and/or hallucinations) [33-35] ●Higher levels of anxiety [30. in a reproducible manner across different patient populations (figure 2 and figure 1) [13.49. This decreasing rate of recovery is graphically illustrated (figure 1 and figure 2). ≤8 weeks).51. for mild to moderate depression with recent onset and no suicidality or psychosis.36-39] ●Pre-existing comorbid disorders [29. section on 'Reluctance to start treatment'.40- 42] ●High levels of neuroticism [42-45] ●Poorer psychosocial functioning [30.14]. Prospective observational studies have found that episodes (some of which were untreated) often remitted soon after onset (eg.52] Recurrence — Major depressive disorder is highly recurrent [5.39. the cumulative rate of recurrence at [56]: •5 years was 13 percent •10 years was 23 percent •20 years was 42 percent ●In a prospective study of 318 patients who recovered from a major depressive episode and were assessed semi-annually or annually for up to 10 years. (See "Unipolar depression in adults and initial treatment: General principles and prognosis". which have been identified in at least two separate prospective observational studies that included at least two years of follow-up and were not confined to a particular age or diagnostic subgroup: ●Longer episode duration at baseline [27-29] ●Greater baseline symptom severity [30-32] ●Psychotic features (ie.30. 64 percent suffered at least one subsequent episode [57]. including personality disorders [27. Major depressive episodes may remit quickly.Prospective observational studies have found that the probability of recovery from major depression progressively decreases as the duration of the episode increases [13.53-55]: ●In a prospective study of individuals in the Dutch general population who had recovered from an episode of major depression (n = 687). remission of depressive episodes that occurs during treatment that is started soon after onset of the episode often appears to be the result of a placebo effect [25.5 years. The median time to recurrence for the first the recurrent episode was approximately 3 years and for subsequent episodes was 1 to 1. . two weeks) may be a reasonable alternative to antidepressant treatment.14].

27.191 depressed individuals) found that recurrent depressive episodes were twice as likely among individuals who were mistreated during childhood (physical or sexual abuse. As an example.65.56. As an example. low mastery) [66] •Emotional dysregulation and repeated exposure to adversity [67. ●Other factors such as: •Greater severity (intensity) of the preceding depressive episode [18. approximately [57]: ●20 percent suffered a recurrence in months 1 through 6 after recovery ●19 percent in months 7 through 12 after recovery ●15 percent in months 13 through 18 ●13 percent in months 19 through 24 ●11 percent in months 25 through 30 ●9 percent in months 31 through 36 Clinical factors that may be associated with recurrence of major depression include: ●Prior history of recurrence – This is the most consistently identified risk factor [40. a prospective study of 318 patients found that each recurrence increased the risk of a subsequent recurrence by 16 percent [57]. A subsequent study of 687 euthymic individuals with a lifetime history of unipolar major depression found that time to recurrence was shorter among individuals with negative youth experiences [56].72]. neglect. which prospectively administered up to four sequential trials of pharmacotherapy to patients who presented with unipolar major depression. In the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study. The rate of relapse after each treatment step was as follows: .42] •Poorer psychosocial functioning [50.56-67].56] •Feeling that life circumstances are beyond one’s control (ie. Thereafter.64.75] •Comorbid personality disorder [40. the probability of recurrence progressively decreases as the duration of recovery (wellness) increases.76] Treatment resistant depression — Relapse appears to be greater in patients with major depression who require more than one course of treatment to remit. As an example. ●Residual depressive symptoms – This is another potent risk factor [59.74] •Younger age of onset of major depressive disorder [60.The risk of recurrence appears to be greatest in the first few months after recovery from a major depressive episode. a prospective study of 322 patients found that the median time to recurrence was four times shorter for patients with one or two mild symptoms during the recovery period than for patients with no symptoms (32 versus 135 weeks) [71. or family conflict or violence) compared with individuals who were not [51].73.67- 70].73] •Younger age at time of assessment [56.59. more than 1500 patients remitted and were followed for up to one year [78]. a prospective study found that among 318 patients who recovered from a major depressive episode. ●Childhood maltreatment – A meta-analysis of seven epidemiologic studies (10.62. compared with patients who remit after the initial course of treatment [77].59.

and examined the proportion of weeks ill with major depression [83]. As an example. and 55 years). community-based study of individuals (n >8000) found that early age of onset. and anxiety during the first lifetime episode of unipolar major depression were associated with a greater persistence and severity of subsequent depressive symptoms [84]. long-term prospective studies of patients with major depression. Clinicians may conclude that the course of illness tends to worsen in major depression (eg. have found that course of illness remains the same for the cohort over time. (See 'Mortality' below. The amount of time ill with depression did not change as patients moved from their third decade of life to their fifth decade.82]: ●Mean time to recovery from an episode of major depression ●Probability of recurrence ●Amount of time ill with major depression As an example. As an example. functional recovery often takes longer than syndromal recovery [87]. time to recovery and time to recurrence are inconsistent across multiple episodes [14. suicidal ideation and behavior. ●Remission occurred with initial treatment – 34 percent relapsed ●Remission occurred with step two – 47 percent ●Remission occurred with step three – 43 percent ●Remission occurred with step four – 50 percent The difference in relapse following remission after initial treatment and after step two was statistically significant. Functioning — Although psychosocial functioning typically improves in patients who recover from a major depressive episode [86]. Clinical features that are present during the initial episode of unipolar major depression may be associated with a greater lifetime burden of depressive symptoms. Mortality in depression is discussed elsewhere in this topic. The 20 years of follow-up were grouped into 5-year periods to determine whether the proportion of weeks ill with depression changed over time. episodes appear to progressively become longer or more frequent) when in fact it does not. a prospective study of patients with recurrent .81. It is not clear if all-cause mortality or suicide is greater in treatment resistant depression. because patients who experience more symptoms are more likely to continue visiting clinicians than patients who remain euthymic for long periods [85]. including [13. and patients were divided into three groups according to their age at study intake (mean age 25. multinational. However.) Long-term course of illness — For any individual patient who suffers more than one episode of major depression.79]. 36. from their fourth to their sixth decade. receiving various levels of treatment or none at all. severe dysphoria.14. or their sixth to their eighth decade of life (figure 3).80].57. a retrospective. a study prospectively followed 220 patients with major depressive disorder for 20 years. compared with the general population of patients with depression [77.

symptom patterns. much of the literature is based upon the diagnoses used prior to DSM-5. functional recovery lagged behind syndromal recovery by approximately one year [88]. patients with dysthymic disorder recovered more slowly than patients with non- chronic major depression [95].) Quality of life — Depression is associated with impaired quality of life. section on 'Functional impairment'. Fifth Edition (DSM-5) [1] are nearly identical to the criteria for dysthymic disorder in DSM-IV-TR [90]. Among patients who remitted (n >800). and family history [91-94]. The primary difference is that persistent depressive disorder also subsumes patients who were labelled as chronic unipolar major depression in DSM-IV- TR (which categorized chronic major depression as a subset of unipolar major depression). DSM-5 consolidated dysthymic disorder and chronic major depression into persistent depressive disorder (dysthymia) because there was little difference between dysthymic disorder and chronic major depression with regard to demographics. and was severely impaired in 9 percent. Quality of life was impaired in nearly all patients at study intake. In an observational study of 49 patients followed . psychological. Additional information about functional impairment in unipolar depression is discussed separately. The delay in functional recovery was attributed primarily to subsyndromal symptoms. However.) Dysthymic disorder — In follow-up studies of dysthymic disorder. (See "Unipolar depression in adults: Clinical features".unipolar depression (n >1000) found that on average. As an example. most episodes resolved but recurrence was common [95]: ●A prospective study of 82 patients with dysthymic disorder followed for up to 10 years found that approximately 74 percent recovered. and social functioning. PERSISTENT DEPRESSIVE DISORDER (DYSTHYMIA) — The diagnostic criteria for persistent depressive disorder (dysthymia) in the Diagnostic and Statistical Manual. treatment response. Quality of life may remain impaired despite resolution of the depressive syndrome. the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study prospectively administered citalopram to patients with unipolar major depression who were followed for up to 12 months [89]. Additional information about the clinical features and diagnosis of depressive disorders is discussed separately. and the persistence of these symptoms speaks to the need for continuation treatment. the median time from study entry to recovery was approximately four years [96]. which refers to subjective satisfaction with one’s physical. ●A prospective study of 53 patients who recovered from dysthymic disorder and were followed for a median of eight years found that 55 percent suffered a recurrence [96]. Also. quality of life remained impaired in more than 30 percent. (See 'Definitions' above and "Unipolar depression in adults: Assessment and diagnosis".

clinicians should encourage patients to continue treatment and efforts to manage their illness despite persistent symptoms [107]. studies found that among patients who recovered from major depressive episodes superimposed upon dysthymic disorder. recovery occurred in fewer patients with dysthymic disorder compared with unipolar major depression (25 versus 63 percent) [97].99]. less severe depressive symptoms. these patients are given the diagnosis persistent depressive disorder with persistent major depressive episode [1]. the formal diagnosis for double depression is persistent depressive disorder with intermittent major depressive episodes [1].105]: ●A prospective observational study. only 12 percent continued to meet criteria for major depression [52]. recovery occurred in 38 percent [106]. patients who met full criteria for an episode of major depression continuously for two years were diagnosed with chronic major depression [90]. In DSM-5. Underlying dysthymia also seemed to worsen the course of major depression.) In studies of dysthymic disorder.100]. Observational studies of dysthymic disorder found that patients usually had exacerbations that met criteria for a major depressive episode. Clinical factors at baseline that were associated with a shorter time to recovery from chronic major depression in prospective studies included a high level of psychosocial functioning.103] Chronic major depression — In DSM-IV-TR. at a tertiary medical center. Thus. ●A nationally representative community survey in the United States identified individuals with chronic major depression (n = 504) and found that after three years of follow-up. Nevertheless. patients with chronic major depression recovered [104. time to relapse of another major depressive episode was shorter than that experienced by patients who recovered from major depression not superimposed upon dysthymic disorder [98. two studies of dysthymic disorder (190 and 87 patients) found that most patients also met criteria for major depression (62 and 59 percent of patients) [96. Prospective observational studies found that the probability of recovery from an episode of chronic major depression was less than the rate of recovery from shorter episodes [104]. of 35 patients with major depression who were continuously ill for five years found that during the following five years.101] ●High neuroticism scores [101] ●Comorbid anxiety disorder [102. absence of psychiatric comorbidity. a phenomenon labelled “double depression” [98]. predictors of poor outcome included: ●Family history of dysthymic disorder or chronic major depression [96. . As an example. (In DSM-5. and absence of psychosis [104].prospectively for six months.

For some patients with depression plus a general medical disease.000 with a lifetime diagnosis of unipolar depression [126]. Assuming onset of depression at age 30. depression in later life may be an independent risk for subsequent cognitive decline [116-118]. As an example: ●A meta-analysis of 293 studies from 35 countries (n >100. it appears that the mortality risk of depression is not fully explained by the comorbid medical illness [131- 133]: .600.) The excess mortality that is observed in depression is greater for men than women. and diagnosis". compared with nondepressed individuals [119-123].000) found that the mean age of death was 71 versus 76 years [127].000. and diagnosis". clinical presentation. Major depression and less severe (minor) depression are each associated with excess mortality. heterogeneity across studies was high [125] ●A subsequent national registry study identified more than 900.000 depressed individuals and >1. and with worse outcomes in comorbid general medical illnesses [109- 115]. and clinical assessment". In addition. Depression is associated with coronary heart disease.) MORTALITY All cause — All cause mortality is approximately 50 to 100 percent greater in depressed individuals.000) or without depression (n >4. (See "Unipolar minor depression in adults: Epidemiology. and for depressed females was 7 and 13 years shorter [129. A meta-analysis of 13 prospective observational studies (n >5500 depressed men and women) found that the risk of death was twice as great in depressed men than in depressed women [128]. pathology. All-cause mortality was two times greater in people with depression. clinical presentation.000 individuals who died. section on 'Neuropsychiatric symptoms'. In addition.000 nondepressed individuals) found that the risk of mortality was 52 percent greater in depressed individuals. compared with the general population.130]. diabetes mellitus. life expectancy for depressed males was 11 and 15 years shorter compared to the general population. stroke.) MORBIDITY — Depression may adversely affect the overall health of patients [108]. (See "Unipolar minor depression in adults: Epidemiology. Parkinson disease. including more than 78. (See "Mild cognitive impairment: Epidemiology. a retrospective study of military veterans with a history of depression (n >700. life expectancy in depressed individuals was roughly 11 years shorter. section on 'Course of illness'. In two community registry studies. section on 'Mortality'. heterogeneity across studies was high [124] ●A subsequent meta-analysis of 43 studies (sample size not provided) found that the risk of mortality was 71 percent greater in depressed individuals.MINOR DEPRESSION — Course of illness in minor depression is discussed separately.

●A meta-analysis of 25 observational studies (9417 patients) showed that mortality rates were 25 percent higher in cancer patients with depressive symptoms compared with nondepressed cancer patients (risk ratio 1.5-1. In a prospective study that followed 186 patients with major depressive disorder for up to 38 years.000 suicides.10-1. 3. falls. mortality was higher for women with depression plus diabetes mellitus (relative risk 3. ●A community survey and mortality database of 61. One aspect that is typically not addressed in studies of depression and mortality is whether the increased risk for death varies according to lifetime extent of depressive morbidity. 95% CI 2-4). the incidence of suicide was 27 times greater than that for the general population [120]. accidental deaths were two fold greater among depressed patients [141].7.000.6-fold higher (hazard ratio 3. hopelessness. Continuity of care with the same psychiatrist for major depression is associated with decreased all-cause mortality [137]. the absolute risk was small. 95% CI 2. 95% CI 1. which compared depressed patients (n >200.5. motor vehicle accidents.25. and comorbidity.4-1. Accidental death — Depressed patients may be at increased risk of dying from accidents (eg.7-3. ●In a prospective study of 4873 women. to the risk among individuals with no history of mental disorders [140]. (See "Suicidal ideation and behavior in adults".000 controls).7) [134]. After controlling for sociodemographic factors. family history of psychiatric disorder. Suicide — Many depressed patients kill themselves [138].6) than diabetes alone (relative risk 1. the risk factor that was most strongly associated with suicide was prior history of suicide attempt (odds ratio 5. Other risk factors included male sex. 95% CI 1. However. .12-1.) Homicide — Mortality due to homicide is elevated in patients with depressive syndromes. In a Swedish national registry study.000) with the general population (n >6. 95% CI 1. A retrospective study of Swedish national registries compared the risk of homicidal death among individuals with a lifetime history of unipolar depression and no history of substance use. 95% CI 1. the risk of death due to homicide among depressed patients was 2.9) [136].1. 95% CI 3-7) [139]. or accidental poisoning).39. Many risk factors for suicide have been identified. Additional information about suicide is discussed separately. and 39 percent higher in cancer patients diagnosed with major or minor depression (risk ratio 1.9 million) and controlled for sociodemographic factors and substance use disorders.89) [135]. more severe depression. In a meta-analysis of results from 28 articles (n >12.349 individuals found that depressive disorders were associated with an increased mortality that was only partly explained by somatic symptoms or illnesses (hazard ratio 1.40).

pre-existing comorbid disorders. Factors that may be associated with recurrence include prior history of recurrence. younger age of onset of major depressive disorder.) ●The median time to recovery from a major depressive episode is approximately 20 weeks. and amount of time ill with major depression. completed suicide. However. and with worse outcomes in comorbid general medical illnesses. residual depressive symptoms. two weeks) may be a reasonable alternative to antidepressant treatment. However. for groups of patients. (See 'Recurrence' above. (See 'Recovery' above. homicidal death. including mean time to recovery from an episode of major depression.) ●Depression is associated with increased all-cause mortality. recurrences are common.) ●Although most episodes of dysthymia resolve. (See 'Diagnostic stability' above. episodes may remit quickly.) ●Depression is associated with poor psychosocial and physical functioning. The risk of recurrence appears to be greatest in the first few months after recovery from a major depressive episode and thereafter progressively decreases as the duration of recovery (wellness) increases. psychotic features.) ●For any individual patient who suffers more than one episode of major depression. poorer psychosocial functioning.SUMMARY ●Major depressive disorder (unipolar major depression) (table 1) and persistent depressive disorder (dysthymia) (table 2) represent depressive syndromes that are distinguished by the type and number of symptoms that occur as well as their duration. course of illness remains the same for the cohort over time. reassurance and watchful waiting with short-term follow-up (eg. younger age at time of assessment. and comorbid personality disorder. and accidental death. childhood maltreatment. symptom severity of the preceding depressive episode. diabetes mellitus. and a history of childhood maltreatment.) Use of UpToDate is subject to the Subscription and License Agreement. recurrent episodes occur in approximately 50 percent. and stroke. Most patients with dysthymia have exacerbations that meet criteria for a major depressive episode. time to recovery and time to recurrence are inconsistent across multiple episodes.) ●Depression is associated with coronary heart disease. (See 'Persistent depressive disorder (dysthymia)' above.) ●Patients who are initially and correctly diagnosed with major depressive disorder may eventually change diagnoses to bipolar disorder or schizophrenia. . (See 'Morbidity' above. for mild to moderate depression with recent onset and no suicidality or psychosis. (See 'Mortality' above. high levels of neuroticism. Thus. (See 'Definitions' above and "Unipolar depression in adults: Assessment and diagnosis". Possible risk factors for a longer time to recovery include: longer episode duration at baseline. (See 'Long- term course of illness' above. probability of recurrence. greater baseline symptom severity.) ●Among patients with major depressive disorder. (See 'Functioning' above. higher levels of anxiety.

et al. The time course of nonchronic major depressive disorder. Disability-adjusted life years (DALYs) for 291 2 diseases and injuries in 21 regions. Lozano R. Arch Gen Psychiatry 1998. J Affect Disord 2000. 59:1. REFERENCES American Psychiatric Association. Nestadt G. J Affect Disord 2011. Co-occurring manic symptomatology as 12 a dimension which may help explaining heterogeneity of depression. Akiskal HS. Solomon DA. Uniformity across episodes and samples. Ryan CE. Posternak MA. Understanding the heterogeneity of 11 depression through the triad of symptoms. Lancet 2012. Lightfoot T. VA 2013. J Epidemiol Community Health 2007. Zimmerman M. 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010. et al. Vos T. Shao H. 55:694. Judd LL. Chen L. Batten LA. Arch Gen Psychiatry 1997. 159:469. Fifth Edition (DSM-5). Are subjects in pharmacological treatment 9 trials of depression representative of patients in routine clinical practice? Am J Psychiatry 2002. Nestadt G. Keller MB. 65:513. Mattia JI. Diagnostic and Statistical Manual of Mental 1 Disorders. Posternak MA. 380:2197. Generalizability of antidepressant 8 efficacy trials: differences between depressed psychiatric outpatients who would or would not qualify for an efficacy trial. Gilbody S. Int J Neuropsychopharmacol 2015. Leon AC. Arch Gen Psychiatry 1994. 131:224. et al. Gallo JJ. National Institute of 13 Mental Health Collaborative Program on the Psychobiology of Depression--Clinical Studies. Zimmerman M. Population-based study of first onset and 5 chronicity in major depressive disorder. A Prospective. Arch Gen Psychiatry 2008. et al. Beekman AT. Neuropsychopharmacology 2006. 31:1841. A prospective 12-year study of subsyndromal 6 and syndromal depressive symptoms in unipolar major depressive disorders. Murray CJ. Penninx BW. 51:405. Is low folate a risk factor for depression? A meta- 10 analysis and exploration of heterogeneity. Recovery from major depression. How many subjects with major depressive 7 disorder meet eligibility requirements of an antidepressant efficacy trial? J Clin Psychiatry 2003. Akiskal HS. 162:1370. course and risk factors: a longitudinal. American Psychiatric Association. Maser JD. et al. Rush AJ. Leon AC. population-based study. Eaton WW. Sheldon T. Keitner GI. Arlington. . Coryell W. Jabben N. Posternak MA. Report by the ACNP Task Force on 4 response and remission in major depressive disorder. 64:1091. Tremblay P. et al. 18. Chelminski I. 61:631. Eaton WW. A 10- 14 year prospective follow-up across multiple episodes. et al. Am J Psychiatry 2005. Sackeim HA. Kraemer HC. et al. Longitudinal Study of the 3 Effect of Remission on Cortical Thickness and Hippocampal Volume in Patients with Treatment-Resistant Depression. Phillips JL.

Diagnostic shifts during the decade 24 following first admission for psychosis. Szádóczky E. Baldessarini RJ. Zámbori J. bipolar and 28 schizoaffective patients. Tohen M. 57:593. Demler O. JAMA 2003. McLean-Harvard International First- 23 Episode Project: two-year stability of ICD-10 diagnoses in 500 first-episode psychotic disorder patients. Am J Psychiatry 2011. 4:125. Crowe RR. 69:196. Bukh JD. Risk factors for sustained nonremission of 31 depressive symptoms: a 4-year follow-up. et al. Salvatore P. 289:3095. Angst J. Swindle RW Jr. Hsu JW. et al. Diagnostic stability in depressive disorder as according to ICD-10 in 20 clinical practice. Kessing LV. Psychol Med 2016. 41:782. Neuropsychopharmacology 1991. 38:32. Results of a prospective study from 1959 to 1985. Melartin TK. Dager SR. Holma IA. Arch Gen Psychiatry 2000. Ross D. et al. J Affect Disord 2004. Psychopathology 2005. Kessler RC. Comorbidity of ADHD and subsequent bipolar 17 disorder among adolescents and young adults with major depression: a nationwide longitudinal study. 17:315. Rózsa S. Woolson RF. Identification of true drug response to 26 antidepressants. 46:1151. Clinical and public health significance. et al. et al. Lavori PW. et al. Bipolar Disord 2015. Bromet EJ. Cloninger CR. Andersen PK. Klerman GL. Unipolar depression: diagnostic 19 inconsistency and its implications. Khan A. Martin RL. Clayton PJ. Use of pattern analysis. . Course of a clinical cohort of unipolar. J Clin Psychiatry 2011. Predictors for 2-year outcome of major 30 depressive episode. Rates and predictors of remission. Congruence of diagnoses 2 years 21 after a first-admission diagnosis of psychosis. J Clin Psychiatry 2008. Am J Psychiatry 2011. et al. J Affect Disord 1992. Kotov R. Winokur G. 26:111. 83:49. et al. Subthreshold hypomanic symptoms in 16 progression from unipolar major depression to bipolar disorder. Keller MB. 40-year period. Preisig M. Guze SB. J Nerv Ment Dis 1998. Fennig S. 252:788. Quitkin FM. Holma KM. 168:40. 72:183. Schwartz JE. Arch Gen Psychiatry 1981. Tsuang MT. Stewart JW. Arch Gen Psychiatry 1984. recurrence 18 and conversion to bipolar disorder after the first lifetime episode of depression--a prospective 5-year follow-up study. Berglund P. Endicott J. Stability of psychiatric diagnosis. Chronicity of depressive episode in relation to 25 antidepressant-placebo response. et al. Leon AC. 168:1186. Fiedorowicz JG. Chen MH. 22 Schizophrenia and affective disorders followed up over a 30. Rabkin JG. 38:535. Cronkite RC. Long-term outcome of episodes of major 29 depression. Cohen S. Long-term outcome of major depressive 27 disorder in psychiatric patients is variable. Chen YS. Moos RH. JAMA 1984. Schweiz Arch Neurol Psychiatr (1985) 1995. 146:5. Tanenberg-Karant M. Füredi J. Kessing LV. Fochtmann LJ. 186:462. The epidemiology of major depressive 15 disorder: results from the National Comorbidity Survey Replication (NCS-R).

Grove WM. Andersen PK. J Affect Disord 2008. Importance of psychotic features to long-term 34 course in major depressive disorder. Spijker J. Hirschfeld RM. Deeg DJ. Markowitz JC. Can we predict the persistence of depression? Br J 43 Psychiatry 1992. Hawton K. Steunenberg B. et al. Am J Psychiatry 2011. Am J Psychiatry 1991. 71:1629. Am J Psychiatry 1992. Six-year longitudinal course and 32 outcomes of subtypes of depression. de Graaf R. 208:467. Klerman GL. Results from the Netherlands Mental Health Survey and Incidence Study (NEMESIS). Leon AC. 161:633. et al. J Affect Disord 2010. Coryell W. Psychosocial stressors and the prognosis of 49 major depression: a test of Axis IV. J Affect Disord 2004. 148:648. 168:257. Predictors of remission from generalized anxiety disorder and 39 major depressive disorder. Kelly KM. Follow-up and family study of anxious 36 depression. Beekman AT. Am J Psychiatry 1996. 31:115. 208:62. Br J Psychiatry 2016. Determinants of persistence of major depressive 48 episodes in the general population. Sherrington JM. et al. Relationship of personality disorders to the 41 course of major depressive disorder in a nationally representative sample. Lamers F. Fiedorowicz JG. et al. 148:1512. 81:123. 123:164. Boys R. Winokur G. Psycho-social predictors of 44 chronicity in depressed patients. Br J Psychiatry 2012. 200:210. 33 Angst J. 19 Suppl 2:47. Winokur G. Coryell W. Kerkhof AJ. Gilman SE. et al. 77:e704. J Clin Psychiatry 2010. Consistency of remission and outcome in bipolar and 35 unipolar mood disorders: a 10-year prospective follow-up. Personality disorders predict relapse after 40 remission from an episode of major depressive disorder: a 6-year prospective study. Grilo CM. J Affect Disord 2017. Beekman AT. van Hemert AM. et al. Predicting recovery from episodes of major 47 depression. Solomon D. Outcome of women admitted to hospital for 46 depressive illness: factors in the prognosis of severe depression. Psychopathology 1986. J Clin Psychiatry 2016. Clayton PJ. 43:303. et al. Endicott J. Bukh JD. Smoller JW. Leon A. Keyes KM. Scott J. Personality predicts recurrence 45 of late-life depression. Personality and the Long-Term Outcome of 42 First-Episode Depression: A Prospective 5-Year Follow-Up Study. Mezuk B. Bijl RV. Coryell W. Kessing LV. 81:231. Psychol Med 2001. Effects of anxiety on the long-term 38 course of depressive disorders. Psychol Med 2013. Grilo CM. Solomon DA. et al. et al. Stout RL. Trinh NH. et al. Anxiety syndromes as epiphenomena of primary 37 major depression: outcome and familial psychopathology. 149:100. Andreasen NC. J Affect Disord 2004. et al. Coryell W. 153:483. Harrow M. Fagg J. Goldberg JF. The course of affective disorders. Coryell W. . Eccleston D. Skodol AE. Br J Psychiatry 1986. 107:285.

Multiple recurrences of major depressive 57 disorder. Gili M. Recurrence after recovery from major 63 depressive disorder during 15 years of observational follow-up. National 54 Clinical Practice Guideline 90. et al. et al. J Affect Disord 2013. Depress Anxiety 2014. et al. Risk factors for onset of 50 multiple or long major depressive episodes versus single and short episodes. Keller MB. Am J Psychiatry 1999.5-year prospective study. Stegenga Bulloch A. 2010 http://guidance. 43:39. Rubio JM. Is the type of remission after a major depressive 59 episode an important risk factor to relapses in a 4-year follow up? J Affect Disord 2004. Williams J. Pirozzi R. Torres-González F. 31:72. Psychiatric history and subthreshold 65 symptoms as predictors of the occurrence of depressive or anxiety disorder within 2 years. Lavori PW. Garcia-Toro M. Hartman CA. Predictors of relapse in major 58 depressive disorder. Navarro V. 157:229. Jarrett RB. Biggs MM. Giles DE. Recurrence of major depressive 55 episodes is strongly dependent on the number of previous episodes. Soc Psychiatry Psychiatr Epidemiol 2013. Br J Psychiatry 2004. 82:291. Karsten J. Leon AC. Am J Psychiatry 2012. 151:306. Pintor L. Patten S. Br J Psychiatry 2011. . Spijker J. 53 2010. et al. Zeng Y. 2011). Am J Psychiatry 1992.asp x (Accessed on May 29. Third Edition. Keller MB. Maj et al. et al. Childhood maltreatment predicts unfavorable course of 51 illness and treatment outcome in depression: a meta-analysis. 2011). Geerlings MI. 198:206. Keller MB. Prediction of recurrence in 64 recurrent depression: a 5. Bockting CL. Smit JH. J Clin Psychiatry 2010. Recurrence of major depressive disorder 56 and its predictors in the general population: results from the Netherlands Mental Health Survey and Incidence Study (NEMESIS). Danese A. JAMA 1983. De Graaf R. 71:984. 250:3299. 169:141. 156:1000. Koeter MW. Psychol Med 2013. 184:153. Depression: The Treatment and Management of Depression in Adults. et al. Mueller TI. http://www. Solomon DA. Severity of depressive episodes according to ICD-10: prediction of risk 61 of relapse and suicide. Lewis CE. Kessing LV. Veltro F. Predictors of long-term prognosis of depression. Am J Psychiatry 1989. Naicker K.nice. Hardeveld F. 66 Colman I. Persistence of chronic major depression: a 52 national prospective study. Clinical predictors of recurrence in 60 depression. Uher R. Practice Guideline for the Treatment of Patients with Major Depressive Disorder. Am J Psychiatry 2000. Pattern of recurrence of illness after recovery from 62 an episode of major depression: a prospective study. et al. ten Doesschate MC.psych. 149:795. et al. Lavorato D. (Accessed on May 29. 146:764. Torres X. Leon AC. Klerman GL. et al. et al. Nanni V.

77:1065. Longitudinal syndromal and sub-syndromal 81 symptoms after severe depression: 10-year follow-up study. et al. de Jonge P. Kiosses DN. Emotion dysregulation mediates the relationship between 76 lifetime cumulative adversity and depressive symptomatology. Multiple risk factors predict 67 recurrence of major depressive disorder in women. Schettler PJ. 43:341. J Affect Disord 2010. Akiskal HS. Effect of age at onset on the course of major 75 depressive disorder. et al. et al. The prognostic significance of subsyndromal 70 symptoms emerging after remission of late-life depression. Keller MB. Gardner CO. A New Empirical Definition of Major 72 Depressive Episode Recovery and Its Positive Impact on Future Course of Illness. Wooderson SC. Zisook S. Husain MM. Arch Gen Psychiatry 1997. Maser JD. Jayewardene AK. Am J Psychiatry 2006. Lesser I. et al. Kendler KS. Olin B. 7:e48002. Psychol Med 2013. 163:1905. Abravanel BT. Stewart JW. Kaptein KI. Rush AJ. Am J Psychiatry 2007. Judd LL. Does incomplete recovery from first lifetime 69 major depressive episode herald a chronic course of illness? Am J Psychiatry 2000. Endicott J. Major depressive disorder: a prospective study 71 of residual subthreshold depressive symptoms as predictor of rapid relapse. Coryell W. Eaton WW. et al. PLoS One 2012. The Baltimore Epidemiologic Catchment Area follow-up. Residual symptoms after remission 68 of major depressive disorder with citalopram and risk of relapse: a STAR*D report. Acute and longer-term outcomes in 78 depressed outpatients requiring one or several treatment steps: a STAR*D report. Rush AJ. J Psychiatr Res 2015. Judd LL. Judd LL. Moreno F. Gallo J. Bunker M. Gamma A. Paulus MJ. Anthony JC. Markopoulo K. van Loo HM. 148:1353. 116:4. et al. Natural history of Diagnostic Interview 74 Schedule/DSM-IV major depression. 157:1501. Br J Psychiatry 2004. et al. CMAJ 2011. Sellaro R. 125:141. 50:97. Predictors of relapse into major depressive 73 disorder in a nonclinical population. Kennedy N. Schettler PJ. Abbott R. 40:41. J Affect Disord 2009. Fekadu A. Alexopoulos GS. Psychol Med 2010. Trivedi MH. 180:52. et al. Wisniewski SR. Aggen SH. et al. et al. Duration of subsequent episodes and 80 periods of recovery in recurrent major depression. Sinha R. J Affect Disord 1998. 184:330. Conradi HJ. Recurrence of bipolar disorders and major . 61:89. 82 Angst J. What happens to patients with 77 treatment-resistant depression? A systematic review of medium to long term outcome studies. Trivedi MH. J Clin Psychiatry 2016. Paykel ES. Am J Psychiatry 1991. Mortality and suicide risk in 79 treatment-resistant depression: an observational study of the long-term impact of intervention. 183:1969. 164:1539. J Affect Disord 2015. 54:993. Nierenberg AA.

"Double depression": two-year follow-up. Primary early-onset dysthymia: comparison with primary nonbipolar nonchronic major depression on demographic. 39:1689. et al. Group comparisons of DSM-IV 92 subtypes of chronic depression: validity of the distinctions. Shankman SA. Am J Psychiatry 2004. Borian FE. Trivedi MH. Endicott J. McCullough JP Jr. Morris DW. J Abnorm Psychol 1988. et al. 152:1124. personality. Mueller TI. Functional versus syndromal 88 recovery in patients with major depressive disorder and bipolar disorder. Cai T. familial. Lavori PW. 76:e809. 94 13:11. McCullough JP Jr. Rose S. Psychol Med 1999. Clarke GN. Mirocha J. Endicott J. Klein DN. Lewinsohn PM. Family study of chronic depression 93 in a community sample of young adults. IsHak WW. 161:646. Taylor EB. Coryell W. Shankman SA. A life-long perspective. Characteristics and significance of untreated 85 major depressive disorder. Fourth Edition. J Abnorm Psychol 2003. Wisniewski SR. Quality of life in major depressive disorder 89 before/after multiple steps of treatment and one-year follow-up. Winokur G. Am J Psychiatry 2013. 170:633. Solomon DA. Seldenrijk A. Am J Psychiatry 2006. Washington DC 2000. et al. 29:869. et al. James D. Depress Anxiety 2001. Dickstein S. van Meijel B. Differential subtyping of depression. van Loo HM. Ten-year prospective follow-up study of the 96 naturalistic course of dysthymic disorder and double depression. Major depressive disorder subtypes to predict 84 long-term course. depression. et al. Dunner DL. Comparison of DSM-III-R chronic 91 major depression and major depression superimposed on dysthymia (double depression): validity of the distinction. part 2. Coryell W. 98 Keller MB. van der Voort TY. 109:419. Eur Arch Psychiatry Clin Neurosci 2003. Increase in work productivity of 87 depressed individuals with improvement in depressive symptom severity. et al. 253:236. 163:872. American Psychiatric Association. Keller MB. American Psychiatric Association. Longitudinal follow-up of unipolar 95 depressives: an investigation of predictors of relapse. 131:51. 112:614. Does major depressive disorder change with 83 age? Psychol Med 2009. et al. Leon AC. J Consult Clin Psychol 1985. Text Revision. and socioenvironmental characteristics and short-term outcome. Solomon D. J Abnorm Psychol 2000. Yang T. 31:765. Lewinsohn PM. Gonzales LR. Am J Psychiatry 1995. et al. Klein DN. et al. Leon A. 97 clinical. Acta Psychiatr Scand 2015. Harding K. . et al. The Range of Impaired Functioning Tool 86 (LIFE-RIFT): a brief measure of functional impairment. Klein DN. Gruber MJ. Klein DN. J Clin Psychiatry 2015. 53:461. Depress Anxiety 2014. 97:387. Klein DN. et al. Diagnostic and Statistical Manual of Mental 90 Disorders.

7 67:1412. Longitudinal association between 11 depressive symptoms and incident type 2 diabetes mellitus in older adults: the 3 cardiovascular health study. Keller MB. 10 Coryell W. 140:1054. The societal costs of chronic major depression. 2 54:317. 53:794. Berbano E. Lazo M. Endicott J. Ryan CE. 10 Mueller TI. Am 0 Heart J 2000. 62 Suppl 6:5. Biol Psychiatry 2003. The impact of comorbid anxiety disorders on the course of 10 dysthymic disorder: a 5-year prospective longitudinal study. Barzilay JI. 42:408. Risk of Parkinson disease after depression: a 5 nationwide population-based study. Tsai SJ. Hirschfeld RM. Mood disorders in the medically ill: scientific 8 review and recommendations. early adversity. Realistic expectations and a disease management 10 model for depressed patients with persistent symptoms. et al. DeZee K. 158:1864. J Clin Psychiatry 2006. Hayden EP. 167:802. Neurology 2013. et al. Five-year course and outcome of 10 dysthymic disorder: A prospective. Biol Psychiatry 2003. 10 Jackson JL. Schwartz JE. Biol Psychiatry 2005. 11 Krishnan KR. and treatment. J Psychiatr Res 2008. Outcome of dysthymic disorder at 5-year follow-up: the effect 10 of familial psychopathology. 11 Cui X. Rose S. Klein DN. Recovery after 5 years of unremitting major 6 depressive disorder. Solomon DA. Am J Psychiatry 1983. Tu X. 152:843. Examining a bidirectional association 4 between depressive symptoms and diabetes. Keitner GI. 11 Golden SH. 54:227. Dysthymic disorder and double depression: 2 prediction of 10-year course trajectories and outcomes. and chronic 1 stress. 3 70:211. Am J Psychiatry 1995. 58:175. Phillips LS. 140:70. Biggs ML. Leon AC. Does depression precede or follow executive . 140:689. Lewis L. Betan E. Shankman SA. 11 Rudisch B. et al. 81:1538. Arch Intern Med 2007. Arch Gen Psychiatry 1996. Keller M. Pettit AR. Larsen H. 299:2751. naturalistic follow-up study. biology. Perng CL. 10 Pincus HA. et al. Klein DN. Outcome of patients with chronic affective disorder: 4 a five-year follow-up. Epidemiology of comorbid coronary artery disease and 1 depression. Shankman SA. Leader JB. Depression as a contributing factor in cerebrovascular disease. Carnethon MR. et al. personality. et al. Nemeroff CB. Am J Psychiatry 0 2000. JAMA 2008. Carnethon M. et al. Charney DS. Lyness JM. 157:931. 11 Shen CC. Klein DN. 147:1627. comorbidity. Can treating depression improve disease 9 outcomes? Ann Intern Med 2004. J Clin 5 Psychiatry 2001. Klein DN. 10 Klein DN. Am J Psychiatry 2001. J Affect Disord 2002. Relationship of depression to 11 diabetes types 1 and 2: epidemiology. Am J Psychiatry 1990. Musselman DL. Rose S. 10 Evans DL. Results of the DSM-IV mood disorders 99 field trial. Keller MB.

Association of mental disorders in early 12 adulthood and later psychiatric hospital admissions and mortality in a cohort study of 3 more than 1 million men. et al. Druss BG. Batty GD. J Affect Disord 2002. Perera G. Akbaraly TN. 13:748. et al. Freedland KE. Depression. Chang CK. Am J Psychiatry 2007. Walker ER. et al. Morales KH. Vogelzangs N. Mortality of patients with mood disorders: 0 follow-up over 34-38 years. Vogelzangs N. The gap in life expectancy from preventable 13 physical illness in psychiatric patients in Western Australia: retrospective analysis of 0 population based registers. 12 Markkula N. Early mortality and years of potential life lost 7 among Veterans Affairs patients with depression. Comprehensive meta-analysis of excess 12 mortality in depression in the general community versus patients with specific 4 illnesses. Kisely S. Br J Psychiatry 1 2013. Psychiatr Serv 2012. Am J Geriatr 9 Psychiatry 2005. Hancock KJ. Twisk J. Persistent depressive symptoms 11 and cognitive function in late midlife: the Whitehall II study. 13 2 . 200:143. 193:203. 12 Laursen TM. 44:472. Am J Psychiatry 2014. and cognitive decline. J Clin Psychiatry 2010. A meta-analytic comparison. Mortality in mental disorders and global disease 12 burden implications: a systematic review and meta-analysis. Osborn DP. Singh-Manoux A. Mortality in people with depressive. 346:f2539. Clayton PJ. Musliner KL. Alexopoulos GS. et al. and two-year mortality among older. Otey E. 6:e19590. et al. Life expectancy at birth for people with 12 serious mental illness and other major disorders from a secondary mental health care 9 case register in London. Arch Gen Psychiatry 2006. 12 Zivin K. Depression as a risk factor for post-MI 1 mortality. 13 Carney RM. Angst J. mild 7 cognitive impairment.6 dysfunction? Outcomes in older primary care patients. Bogner HR. 11 Steffens DC. 8 71:1379. 63:823. Stassen HH. Is excess mortality higher in depressed men 8 than depressed women?. et al. 5 72:334. J Am Coll Cardiol 2004. 69:823. McGee RE. Br J Psychiatry 2012. et al. Härkänen T. cardiovascular disease. Jaffe AS. Twisk J. author reply 473. 63:130. 68:167. J Affect Disord 2016. Cuijpers P. 164:1221. et al. 202:22. et al. Benros ME. Differential mortality rates in major and 12 subthreshold depression: meta-analysis of studies that measured both. Pfeiffer PN. et al. Gale CR. Mortality and life expectancy in 6 persons with severe unipolar depression. primary-care patients. et al. anxiety 2 and alcohol use disorders in Finland. 12 Angst F. Vogelzangs N. PLoS One 2011. Ilgen MA. Perspectives on depression. Arch Gen Psychiatry 2012. J Affect Disord 2014. Lawrence D. 171:453. 12 Cuijpers P. Gallo JJ. JAMA Psychiatry 2015. et al. Perälä J. 11 diabetes. Cuijpers P. Hayes RD. Twisk J. Marmot M. BMJ 2013.