Review

Review of dysthymia and persistent depressive disorder:

history, correlates, and clinical implications
Elisabeth Schramm*, Daniel N Klein*, Moritz Elsaesser, Toshi A Furukawa, Katharina Domschke

Persistent depressive disorder is a chronic mood disorder that is common and often more disabling than Lancet Psychiatry 2020;
episodic major depression. In DSM-5, the term subsumes several chronic depressive presentations, including 7: 801–12

dysthymia with or without superimposed major depressive episodes, chronic major depression, and recurrent *Joint first authors
major depression without recovery between episodes. Dysthymia can be difficult to detect in psychiatric and Department of Psychiatry and
Psychotherapy, University
primary care settings until it intensifies in the form of a superimposed major depressive episode. Although
Medical Center
information is scarce concerning the cause of persistent depressive disorder including dysthymia, the causation (Prof E Schramm PhD,
is likely to be multifactorial. In this narrative Review, we discuss current knowledge about the nosology and M Elsaesser MSc,
neurobiological basis of dysthymia and persistent depressive disorder, emphasising a dimensional perspective Prof K Domschke PhD) and
Center for Basics in
based on course for further research. We also review new developments in psychotherapy and pharmacotherapy for
Neuromodulation
persistent depressive disorder, and propose a tailored, modular approach to accommodate its multifaceted nature. (Prof K Domschke), Faculty
of Medicine, University
Introduction depression have a chronic course.6,7 Dysthymic disorder of Freiburg, Freiburg, Germany;
Department of Psychology,
About 20–30% of depressive disorders have a chronic was introduced in DSM-III and subsequently ICD-10 as a
Stony Brook University,
course,1,2 and are associated with more severe health diagnosis for unipolar depressions that are chronic and Stony Brook, NY, USA
consequences and poorer outcomes than non-chronic low grade. (Prof D N Klein PhD); and
depression.3 Although previously not distinguished from Dysthymia represents the confluence of several older Department of Health
Promotion and Human
episodic forms of depression, persistent depressive clinical constructs, including neurotic depression and
Behavior, Kyoto University
disorder is now defined as a unique clinical entity in depressive personality, which overlap but have somewhat Graduate School of Medicine
DSM-5.4 different meanings.8 The introduction of dysthymia into and School of Public Health,
In this narrative Review, we explore the previous the mood disorders section of DSM-III was controversial Kyoto University, Kyoto, Japan
(Prof T A Furukawa MD)
controversies surrounding the concept of persistent as many people believed that low-grade persistent symp­
Correspondence to:
depressive disorder. Given the differences between toms, often dating from childhood or adolescence,
Prof Elisabeth Schramm,
chronic and non-chronic depressions on numerous indicated a personality style rather than a mood disorder.8,9 Department of Psychiatry and
variables relevant to cause, this distinction, which is Numerous studies show that dysthymia is associated with Psychotherapy, University
often overlooked, has the potential to provide more higher rates of personality disorders and more extreme Medical Center, Faculty of
Medicine, University of Freiburg,
homo­geneous phenotypes for research focused on cause frequencies of normative personality traits, such as higher
DE–79104 Freiburg, Germany
and treatment. Of the different forms of chronic neuroticism and lower extraversion, than non-chronic elisabeth.schramm@uniklinik-
depression subsumed under the term persistent major depressive disorder.10–16 Thus, critics continue to freiburg.de
depressive disorder, dysthymia is especially likely to be argue that persistent depressive disorder, particularly in
overlooked by clinicians and by individuals with its milder form (ie, pure dysthymia), reflects a personality
persistent depressive disorder themselves since it style rather than a clinically significant mood disorder
appears milder than other forms of chronic depression (Klein provides a commentary on this).17–19
from a cross-sectional perspective and could be difficult However, evidence from family, follow-up, and treat­
to distinguish from the patient’s premorbid personality. ment studies show a close relationship between
However, we challenge the view that so-called pure dysthymia and non-chronic major depressive disorder.
dysthymia is a mild condition from a longitudinal Family studies have reported that the rate of major
viewpoint, while recognising that differences in severity depressive disorder is significantly higher in the first-
have important treatment implications. The purpose of degree relatives of people with dysthymia than in the
this Review is to provide a scoping summary of the relatives of healthy controls.20–22 Additionally, almost all
controversies and difficulties surrounding the subject individuals with dysthymia eventually develop major
(panel 1) and set the tone for further research by depressive episodes,23 and patients with dysthymia have a
addressing future directions in the classification, under­ significantly better response to antidepressant medi­
standing of cause, and treatment of persistent depressive cations than to placebo.24,25 A difference in response to
disorder including dysthymia. drug versus placebo might be even greater in people with
dysthymia than in people with non-chronic major
Nosology, epidemiology, and assessment depressive disorder.24 Hence, despite links with person­
of persistent depressive disorder ality disorders, strong evidence exists for viewing
Evolution of the construct dysthymia as a mood disorder.
Mood disorders have traditionally been viewed as DSM-III-R added a chronic specifier for major
episodic, remitting conditions.5 Beginning in the late depressive episodes to capture more severe forms
1970s, it was recognised that many patients with of chronic depression than included in previous criteria.

www.thelancet.com/psychiatry Vol 7 September 2020 801

 Review

DSM-IV continued to expand coverage of persistent

Panel 1: Challenging issues in dysthymia and persistent depressions by adding specifiers for major depressive
depressive disorder episodes with partial remission and recurrent major
• Controversies surrounding the definition and depressive episodes without full recovery between
conceptualisation of persistent depressive disorder over episodes.
the years (from DSM-III to DSM-5 and ICD-11) have not
been fully resolved. DSM-5: consolidating chronic depression
• Varying terms are used for chronic depression in different Although initially conceptualised as mild, most
classification systems, research, and textbooks, which individuals with dysthymia far exceed the minimum
poses a challenge for synthesising and understanding the number of symptoms required for the diagnosis, and
literature, and comparing studies. they have as much functional impairment and a poorer
• So-called pure dysthymia continues to be a controversial long-term course than patients with non-chronic major
construct, with some arguing that it should be depressive disorder.15,20,26–28 DSM-5 consolidated dysthymia
distinguished from more severe forms of chronic and severe forms of chronic depression, such as major
depression. depressive episodes superimposed on pre-existing
• Pure dysthymia is easily overlooked clinically because dysthymia (ie, double depression)29 and chronic major
of its milder presentation than other forms of chronic depressive episodes, into the broader diagnostic category
See Online for appendix depression and the difficulty in distinguishing it from of persistent depressive disorder (panel 2; appendix p 1).
premorbid personality. However, because of its By definition, chronic major depressive episodes and
persistence, the effect of pure dysthymia on functioning double depression have greater symptom severity than
and long-term outcome (eg, risk for major depressive dysthymia.26,35 However, these three conditions show few
episodes) can be substantial. differences on comorbid anxiety, substance, and per­
• Generally, chronic and non-chronic depressions are sonality disorders; personality traits; depressotypic
frequently not distinguished and the differences between cognitions; coping style; history of childhood adversity;
them are often unrecognised by researchers and clinicians, psychopathology in first-degree relatives; and out­
contributing to the heterogeneity of depression as comes.10,12,15,22,36–40 Moreover, a naturalistic study followed
a diagnostic category. up over 10 years noted that the various forms of chronic
• Assessment of persistent depressive disorder must be depression often shifted to other types of chronic
approached somewhat differently from non-chronic depression over the course of the illness.23 Together, this
depression, and current depression rating scales and work suggested that the forms of chronic depression
self-report inventories have substantial limitations in delineated in DSM-IV reflect distinctions without
assessing persistent depressive disorder. differences, and simplification was warranted. Hence,
• Genetic and neurobiological research on persistent DSM-5 brought these multiform presentations together
depressive disorder have lagged behind research on under a single rubric. However, to be conservative and
major depressive disorder, and should be a high priority. preserve the detailed information about severity in
• Psychotherapy is less effective for persistent depressive DSM-IV, which can be relevant for treatment planning,
disorder than for episodic major depressive disorder, DSM-5 allows for coding four persistent depressive
perhaps because of treatment delay, low motivation, disorder specifiers (panel 2). ICD-11 continues to use the
specific features complicating treatment, insufficient narrower, milder category of dysthymia than DSM-5
treatment duration, and the absence of any tailored, does, without including the more severe forms of chronic
modular approaches to accommodate its multifaceted depression.
nature. Most studies have combined pure dysthymia with
• Psychotherapy for pure dysthymia might be less effective other forms of chronic depression because the prevalence
than pharmacotherapy. However, existing studies are of of pure dysthymia is low. As much of the criticism of
low quality, brief duration, and do not use psychotherapies persis­tent depressive disorder focuses on grouping pure
tailored specifically for chronic depression. Hence, these dysthymia with more severe forms of chronic depression,
therapies should be a high priority for research. further studies of this group are needed.18,19,35 However,
• Greater focus on quality of life related to health and the the fact that most patients with pure dysthymia
effects of treatment in the long term are needed for subsequently experience a major depressive episode
patients with persistent depressive disorder. highlights how closely intertwined the forms of chronic
• Models that conceptualise depression along the depression are and indicates that, from a longitudinal
dimensions of severity and longitudinal course are worthy perspective, pure dysthymia is usually not a mild
of exploration as a unifying and more informative condition.23
approach to classifying depressive disorders than currently Chronic depression and persistent depressive disorder
available. are often confused with treatment-resistant depression,
which is discussed later in this Review.41 Some patients
with chronic depression, and some with non-chronic

802 www.thelancet.com/psychiatry Vol 7 September 2020

 Review

major depressive disorder, do not respond to multiple

trials of established antidepressant treatments. However, Panel 2: DSM-5 criteria for persistent depressive disorder
as discussed later, many patients with chronic depression (code 300.4)
respond well to pharmacotherapy and psychotherapy. Criteria
• Depressed mood most of the day, more days than not,
Epidemiology for at least 2 years
Various epidemiological studies of dysthymia and several • At least two of the following six symptoms:
of chronic major depressive disorder show lifetime • Poor appetite or overeating
prevalence ranging from 1–6%.10,42–46 Only two studies • Insomnia or hypersomnia
have estimated the lifetime prevalence of persistent • Low energy or fatigue
depressive disorder. Consistent with previous studies of • Low self-esteem
dysthymia and chronic major depressive disorder, in a • Poor concentration or difficulty making decisions
nationally representative Australian sample, Murphy • Hopelessness
and Byrne1 reported that the prevalence of persistent • Never without these symptoms for more than 2 months
depressive disorder was 4·6%. However, in a Swiss study at a time
of adults aged 35–66 years, Vandeleur and coll­eagues35 • Criteria for a major depressive disorder can be continuously
reported that the prevalence was 18%. Study design and present for 2 years
differences in assessment (discussed later in this Review) • No history of mania or hypomania
most likely contribute to variations in prevalence. • Not better explained by a non-mood psychotic disorder
The prevalence of chronic depression is considerably • Not due to physiological effects of a substance or another
higher in clinical settings, comprising 33–50% of patients medical condition
with depressive disorders.47–49 The rate of chronic depres­ • Symptoms cause pronounced distress or impairment
sion is almost twice as great in women than in men.10,35,42,46
Chronic depression is also associated with lower incomes Specify if:
but associations with race, ethnicity, and education are • Pure dysthymic syndrome
inconsistent.10,35,45,46 • Persistent major depressive disorder
• Intermittent major depressive disorder, with current
Chronic and non-chronic depression: related but episode
different • Intermittent major depressive disorder, without current
Consolidating multiple forms of chronic depression episode
under the persistent depressive disorder rubric implies Specify current severity:
that although this disorder belongs in the depressive • Mild
disorder section of the DSM, chronic depression and • Moderate
non-chronic major depressive disorder are sufficiently • Severe
different to distinguish them as separate diagnostic
constructs. In contrast to the few differences between Specify if:
various forms of chronic depression, many differences • Early onset (before age 21 years)*
between chronic depressions and non-chronic major • Late onset (at age 21 years or older)†
depressive disorder have been consistently documented *Greater familial loading for mood disorders; more childhood adversity and
(panel 3). childhood maltreatment; increased anxiety, substance, and personality comorbidity;
greater neuroendocrine dysregulation than late onset.30–33 †Greater association with
These differences include evidence, from both retro­
stressful life events, particularly chronic stressors related to general medical disorders
spective and prospective studies, that chronic depression or illness or loss of loved ones than early onset.30,34
is associated with greater childhood adversity (eg, poorer
relationships with parents)2,15,37,50–55 and higher rates of
mood disorders in first-degree relatives than non-chronic depressive dis­ order are typically episodic, whereas
major depressive disorder is.13,20–22,56 Moreover, this pattern recurrences of chronic depression are generally persis­
of familial aggregation is highly specific, with increased tent (although, as noted earlier, the precise form of
rates of chronic depression in the relatives of individuals chronic depression varies).23 Finally, evidence suggests
with chronic depression compared with relatives of that the distinction between chronic depression and non-
individuals with non-chronic major depressive disorder chronic major depressive disorder is qualitative, rather
and healthy controls. Prevalence of chronic depression in than quantitative, in nature. A naturalistic study lasting
the relatives of individuals with major depressive 10 years showed a clear discontinuity in that higher
disorder and in the relatives of healthy controls did not amounts of chronicity during the first 30 months of
differ from one another.20,22,56 follow-up were linearly related to symptom and
Furthermore, follow-up data indicate that the dis­ functional outcomes; by contrast, lower amounts of
tinction between chronic and episodic depression is chronicity were unrelated to subsequent outcomes.57 The
stable over time; recurrences of non-chronic major findings were strikingly consistent across four waves

www.thelancet.com/psychiatry Vol 7 September 2020 803

 Review

course of depression, while ruling out a history of mania

Panel 3: Features of chronic depression as opposed to or hypomania. If the patient has a history of major
non-chronic major depressive disorder depressive episodes, it is crucial to explore the presence
Antecedent of low-grade depression before and after the acute
• Greater childhood adversity and maltreatment episodes. Constructing a life chart or timeline with the
• Higher rates of depression, particularly chronic patient is often helpful to track their level of depression
depression, in relatives over time.58
• Earlier onset of depressive symptoms Most fully structured and semi-structured diagnostic
interviews include a section to assess dysthymia.
Concurrent However, as some interviews assess only current
• Greater comorbidity with psychiatric disorders dysthymia or instruct interviewers to skip dysthymia if
(particularly anxiety and personality disorders) there are recent or recurrent major depressive episodes,
• Greater comorbidity with somatic disorders diagnoses of dysthymia and double depression are often
• Greater neuroticism and lower extraversion overlooked.59
• Higher frequency of depressotypic cognitions, such Most rating scales and self-report inventories for
as dysfunctional attitudes and rumination depression have limited usefulness when applied to
• More avoidant, submissive, and hostile interpersonal styles persistent depressive disorder. For example, the widely
• Higher frequency of alexithymia used 17 item and 21 item versions of the Hamilton Rating
• Smaller social networks Scale for Depression do not assess many of the most
• Less social support common dysthymia symptoms (eg, cognitive symptoms).60
• Greater functional impairment Hence, use of modified versions of this scale is advisable,
Predictive such as use of the 24 item Hamilton Rating Scale for
• Symptomatic for a greater proportion of time Depression that includes helplessness, hopelessness,
• Recurrence of episodes more likely to be chronic worthlessness, and reversed vegetative features. Addi­
• Greater functional impairment tionally, most rating scales assess patients’ symptoms with
• Higher suicide rate reference to differences from their usual state. However,
• Higher frequency of treatment approaches in the history for many patients with persistent depressive disorder,
their usual state is depressed. Hence, symptoms could be
overlooked. To address these problems, Mason and
of follow-up evaluations and with multiple outcome colleagues61 developed the Cornell Dysthymia Rating
measures and model-fit indices. Scale, which has greater content validity, a broader range
Taken together, this evidence supports the distinction of symptom severity, and greater sensitivity to change
between the chronic forms of depression that comprise than the Hamilton Rating Scale for Depression.62
persistent depressive disorder in DSM-5 and non-chronic The General Behaviour Inventory is the only self-report
major depressive disorder. Unfortunately, most clinical measure explicitly developed to screen for chronic mood
and causal research on depressive disorders has not disorders.63 This measure assesses symptoms on a trait
distinguished between chronic and non-chronic forms of basis rather than during a recent discrete time period.
depression. The General Behaviour Inventory has good psychometric
properties and concordance with diagnoses in
Heterogeneity: age of onset community and clinical samples of adolescents and
On the basis of Akiskal’s work,6 DSM-III and subsequent adults.63–65
editions divided dysthymia and persistent depressive
disorder into early-onset (younger than 21 years at onset) Biology
and late-onset (21 years or older at onset) subtypes. Few, and mostly dated, studies have attempted to unravel
Although it is unclear whether age 21 is the optimal the biological underpinnings of dysthymia. Although
cutoff point, considerable support exists for this this section will use the term dysthymia to reflect the
distinction (panel 2). terminology used by the authors of most of these studies,
in most cases the samples include more severe forms
Assessment of chronic depression along with pure dysthymia.
Persistent depression is often overlooked, as clinicians Early research efforts have implicated serotonergic
tend to focus on more acute conditions, such as abnormalities in the biology of dysthymia, including
a superimposed major depressive episodes. Additionally, lower maximum rate of platelet serotonin uptake,
many patients with early-onset persistent depressive decreased urinary serotonin metabolite 5-hydroxy­
disorder consider dysphoria to be normal or a part of indoleacetic acid concentration, and decreased platelet
their usual self, and do not report it to clinicians.6 monoamine oxidase concentration in patients with
To assess and diagnose persistent depressive disorder, dysthymia compared with people who are healthy.
clinicians must take a careful history of the patient’s past Dysregulation of the immune system has also been

804 www.thelancet.com/psychiatry Vol 7 September 2020

 Review

implicated, for example increased CD16 (FCGR3) or which was particularly evident when processing negative
CD56 (NCAM1), or both, increased CD4/CD8 ratio, images.77 In a sample of male children and adolescents
elevated basal IL1B concentration, and increased concen­ with dysthymia, less activation was observed in left and
tration of CXCL10 in patients with dysthymia compared medial regions during a mental rotation task and higher
with people who are healthy. Some studies66–68 point to activation was observed in the right precuneus and
neuroendocrine abnormalities, such as increased base­ posterior cingulate cortex during a delay-match to sample
line concentrations of plasma corti­ cotropin-releasing task compared with in boys with healthy development.78
hormone and cortisol concen­trations in patients with However, because of the unrelated and unreplicated
dysthymia compared with patients who are healthy; findings as mentioned earlier, no firm conclusions can be
however, these could not be unequivocally replicated or drawn regarding brain structure or function of the neural
supported by other studies, although some evidence network in dysthymia.
suggested that neuroendo­crinological altera­tions might Genetic and neurobiological research on dysthymia
be limited to specific subgroups, such as those with an and chronic depression has lagged far behind research
early onset. Finally, some reports show psycho­physio­ on major depressive disorder. So far, no clear-cut
logical alterations (eg, in N200-wave, P300-wave, O-wave, biological cause and pathophysiology have been
and N1-P2 response) and altered sleep-related physio­ confirmed for dysthymia, which could be due to various
logical markers in dysthymia (eg, increased percentage reasons. These reasons include the clinical heterogeneity
in rapid eye move­ ment sleep, shortened rapid eye of the disease phenotype; the high frequency of
movement latency, higher percen­tage of stage 1 sleep, concomitant major depressive episodes; the high com­
and reduction of slow wave sleep compared with patients orbidity of dysthymia with anxiety disorders, somato­
who are healthy). For infor­mation regarding these early form disorders, substance misuse disorders, and
studies see Griffiths et al,69 Howland and Thase,70 and personality disorders; small sample sizes; few attempts
Miller and Yee.71 at replication; and most studies not using what would
In a study investigating the all-male Vietnam Era now be considered adequate methods or not focusing on
Twin Registry, dysthymia was not affected by genetic relevant variables. The identification of specific and
factors but rather by family environmental and non- possibly causal biomarkers of dysthymia is a high priority
shared environmental factors.72 This observation for future research, preferably considering clinically
supports previous evidence from a smaller Norwegian distinct phenotypes such as early-onset dys­thymia79 as
twin study including both sexes.73 Very few candidate well as applying longitudinal approaches including gene-
gene studies are available (appendix p 2). In summary, environment interaction studies and epigenetic analyses.
the scarcity and low quality of behavioural, genetic, and
molecular genetic literature on dysthymia and chronic Clinical management
depression, particularly the grossly underpowered Antidepressants and psychotherapies, used alone or
sample sizes; the absence of replication attempts in in combination, are the primary treatments for persis­
independent samples; and the absence of genome-wide tent depressive disorder. Research has suggested that
approaches, do not permit any conclusions about the pharmacotherapeutic and psychotherapeutic inter­
genetics of dysthymia at the present stage and warrant ventions might be less effective in chronic forms of
further research on this topic. depression, particularly dysthymia, than in non-chronic
Brain imaging studies showed that female patients with depressive disorders.80–82
early-onset dysthymia or depressive personality disorder
displayed a significantly smaller genu and posterior Scarcity of treatment success
midbody of the corpus callosum compared with healthy There could be several reasons why treatment success is
controls.74 Male children and adolescents with dysthymia limited in persistent depressive disorder (panel 4).
were characterised by clusters of lower fractional Individuals with persistent depressive disorder, and
anisotropy within the left uncinate, inferior fronto- clinicians too, can feel helpless when faced with the
occipital, and cerebrospinal white matter tracts in a study challenge of treating persistent or so-called treatment-
using diffusion tensor imaging.75 Similar to findings in resistant depression. The depressive symptoms of
major depressive disorder, resting-state functional approximately 40% of patients with persistent depressive
connectivity MRI showed a greater coherence of neural disorder are considered treatment resistant. However,
activity within the default mode network in patients with there is no consensus regarding the definition of
dysthymia as compared with healthy controls, which treatment resistance of depressive symptoms in research
normalised after treatment with duloxetine for 10 weeks.76 and clinical practice and in many cases this is probably
A functional MRI study using the international affective pseudo resistance.41 About only 33% (262 of 801) of
pictures system iden­tified significantly reduced activation patients with persistent depressive disorder have received
in the dorsolateral prefrontal cortex and increased medication of adequate dose and dura­tion,83 and patients
activation in the amygdala, anterior cingulate, and insula show low compliance.84 Additionally, response to
in patients with dysthymia as compared with controls, psychotherapy is not included in the definition of

www.thelancet.com/psychiatry Vol 7 September 2020 805

 Review

Psychotherapy
Panel 4: Reasons why treatment success is often low in In a meta-analysis of chronic major depression and
persistent depressive disorder dysthymia, Cuijpers and colleagues80 noted that psycho­
Treatment delay therapy has a small but significant effect (d=0·23) on
• Individuals frequently do not seek treatment for several persistent depression when compared with control
years from the onset of the disorder groups (care as usual, non-specific control, placebo, or
• Delay in starting treatment can result in a less positive waiting list). The authors also noted that the effect size
outcome than in patients who do not have a delay, even was associated with the number of treatment sessions
when adequate treatment is applied and that at least 18 sessions are needed for psychotherapy
to have an optimal effect. However, the highest number
Low motivation of sessions given in the studies of dysthymia included in
• Most patients with chronic depression see dysphoria as the meta-analysis was 17 sessions. Providing fewer than
part of their personality and untreatable 18 sessions could explain why psychotherapy was less
• Pessimistic thinking leads individuals to passivity and efficacious than medication in patients with dysthymia.
reduced motivation (eg, “Nothing I do makes a difference”) Other studies also suggest that a longer duration of
Label of treatment resistance therapy and a higher number of sessions are associated
• Can lead to pessimism dominating the therapeutic with better outcomes in chronic forms of depression.90–92
relationship Regarding the benefit of a targeted approach, a study
• Specific features complicating treatment showed that CBASP was more effective than a common
• Insufficient treatment duration factors approach (supportive therapy) in patients with
• Absence of any tailored, modular approaches to early-onset persistent depressive disorder who were
accommodate multifaceted nature unmedicated,90 particularly for those with childhood
maltreatment87 and those with comorbid social anxiety
disorder.93
treatment resistance of depressive symptoms. Therefore,
therapists should resist the temptation to disregard these Pharmacotherapy
patients, but rather specifically address issues, such as A network meta-analysis of pharmacological inter­ventions
reduced motivation, at the beginning and offer psycho­ for persistent depressive disorder noted that SSRIs,
education about the nature of persistent depressive moclobemide, imipramine, ritanserin, and amisulpride
disorder as a mood disorder and not a personal weakness. were more efficacious than, and at least as acceptable as,
placebo. However, fluoxetine proved to be less effective,
Other features complicating treatment and imipramine less acceptable than some other drugs.94
Other specific features of patients with persistent Episode length has been repeatedly noted to predict
depressive disorder potentially complicate therapy poorer response to pharmacotherapy.95–97 However, some
(panel 3). These characteristics might be a consequence of patients with persistent depression do respond and remit
child­hood maltreatment, especially emotional abuse and after acute phase medication treatment.98,99
neglect, which are risk factors for early-onset depression In 2019, a Cochrane review concluded that whether
with a chronic course.85 These risk factors could require continued or maintenance pharmacotherapy is a robust
specific therapeutic strategies aimed at tackling avoidance treatment for preventing relapse and recurrence in
and healing trauma. Among patients who are persistently patients with persistent depressive disorder is unclear,
depressed, those reporting early childhood adversity because of moderate or high risk of bias and clinical
improved significantly more with a psychotherapeutic heterogeneity in the analysed studies.100
approach specific to the disorder, the cognitive behavioural
analysis system of psychotherapy (CBASP), than with Comparison between psychotherapy,
medication86 or with a non-specific approach.87 pharmacotherapy, and combination therapy
Cuijpers and colleagues80 showed that psychotherapy was
Low rate of spontaneous remission, placebo response, significantly less effective than pharmacotherapy in
and floor effects direct comparisons (d=–0·31), especially when compared
Another reason for poor treatment outcome is that, not with SSRIs. However, this result was completely
surprisingly, the rate of spontaneous remission in patients attributable to patients with pure dysthymia because the
with chronic depressive symptoms is low.88 Furthermore, studies examining patients with dysthymia were the
placebo response is low compared with non-chronic major same studies that examined SSRIs. In 2014, a network
depressive disorder.89 Another factor for dysthymia is floor meta-analysis on acute treatments for persistent
effects. Despite the chronicity and impairment associated depressive disorder and dysthymia concluded that
with dysthymia, the severity of symptoms is mild. Hence, medication might be the most preferable option in
showing a significant reduction in symptoms is more dysthymia,94 but that CBASP could be effective for
difficult than it is for more severe forms of depression. dysthymia as well. The authors noted an advantage of

806 www.thelancet.com/psychiatry Vol 7 September 2020

 Review

pharmacotherapy over interpersonal psycho­therapy and patients with persistent depression.113–116 In an analysis117
supportive therapy in persistent depressive disorder, but of a study comparing CBASP and supportive psycho­
not over CBASP and cognitive behavioural therapy. therapy,90 two distinct and clinically meaningful sub­
However, most of the eight studies of dysthymia that groups were identified: patients responding better to
these meta-analyses were based on were limited by low CBASP were more severely depressed and more likely
methodological quality, small sample sizes, approaches affected by childhood trauma than patients responding
not tailored specifically for chronic depression, and a brief more favourably to a non-specific supportive approach,
duration of psychotherapy, thus limiting the conclusions who seemed to be more mildly impaired. Patients
that can be drawn. More research is needed to examine responding more favourably to a non-specific approach
whether dysthymia requires specific treatment. had a higher social functioning level before treatment, a
When evaluating combination treatment against higher quality of life, and more often a recurrent rather
medication or psychotherapy alone, findings consistently than a pervasive illness pattern compared with patients
suggest that patients with chronic forms of major responding better to CBASP.
depressive disorder are most likely to experience increased The duration of treatment and whether to switch
acute benefits from combined therapy.80,94,101,102 A network treatments are also important factors to consider. The
meta-analysis using individual participant data concluded longstanding nature of persistent depressive disorder
that, on average, the combination of CBASP and implicates the need for longer therapy courses, including
medication showed significant superiority over both continuation and maintenance treatment, to address the
monotherapies in terms of efficacy and acceptability in particularly high risk of recurrence.118,119 Even with a
chronic major depressive disorder, whereas the respective psychotherapy specific to the disorder applied over 1 year,
monotherapies displayed essentially similar results.103 The more than half (61 of 91) of the patients were not in
efficacy of the monotherapies was dependent on severity remission 2 years after acute treatment.120 CBASP did not
of baseline depression and anxiety, previous history of result in more sustainable effects on depression
pharmacotherapy, age, and depression subtypes. This symptoms than a non-specific approach at follow-up
indicates that for some subgroups of patients either drug after 2·0 years or compared with medication 4·5 years
therapy or CBASP alone is a reasonable treatment option. after termination of treatment.121 Klein and colleagues122
Additionally, another meta-analysis examining patients reported the superiority of maintenance CBASP over
with chronic types of depression has noted that combi­ 1 year compared with assessment-only appointments,
nation treatment significantly improved quality of life with significantly fewer patients having a relapse. For
over medication monotherapy.102,104 Most patients preferred patients who did not respond to psychotherapy, a switch
combination therapy over monotherapy in two extensive to medication proved to be effective, and vice versa.123
clinical trials of treatments for chronic major depressive The persistent nature of this disorder might require not
disorder.105,106 In dysthymia, however, several studies of only longer but also more structured treatments than are
combination therapy versus anti­ depressant medication currently used, focusing on the disturbed mechanisms
alone did not show a clear benefit for combination (panel 3). Although little evidence exists on the general
treatment.107–110 Nevertheless, use of combined therapy characteristics of interventions that are most likely to
resulted in higher response rates, better improvement of be beneficial in persistent depressive disorder, therapy
functioning,110 and higher cost-effectiveness107 than did without a clear structure or rationale appears unlikely to
medication. effectively address the demoralisation and helplessness
that are main features of persistent depression. Some
A treatment model for persistent depressive disorder evidence suggests that both specific (eg, learning CBASP
Consideration of tailored treatment strategies is impor­ skills; appendix p 4) and non-specific (eg, therapeutic
tant for this disease. The different presentations of alliance) factors make independent contributions to
persistent depression appear to be more alike than treatment outcome.124
different, but persistent depression differs in important Patients with persistent depressive disorder show
ways from non-chronic forms of depression (panel 3). a poor therapeutic response to traditional models of
These differences provide special challenges for the psycho­therapy originally proposed to treat acute-episodic
clinician (appendix p 3) and point to the need for major depressive disorder—eg, interpersonal therapy.125
treatment strategies tailored to the risk factor constel­ This poor response might be partly caused by the diffi­
lation of the individual patient. For instance, a patient’s culty of establishing a cooperative therapeutic relation­
history of early developmental abuse and recent negative ship with­out specific strategies, and the dysfunction in
interpersonal experiences produces a pervasive avoidance cognitive-emotional maturation in which thought is
pattern that is maintained by interpersonal fears.111 often egocentric and less capable of doing operational
Several studies support McCullough’s interpersonal mental tasks. The treatment of patients with persistent
theory112 and provide evidence that change in hostile- depressive disorder therefore needs to move beyond
submissiveness over the course of a specific therapy attempts to modify symptoms while taking distinct
approach (CBASP) is related to depression reduction in targets into consideration. Personalised treatment

www.thelancet.com/psychiatry Vol 7 September 2020 807

 Review

Search strategy and selection criteria

High severity
We searched Medline via PubMed, PsycINFO, and Cochrane
Library for all references published up until Jan 6, 2020 with
Chronic major the terms [(“dysthym*” OR “minor*” OR “mild*” OR
Episodic major depression or
depression double depression “chronic*” OR “persist*” OR “long-stand*” OR “long-term” OR
“dysphor*” OR “double*” OR “recur*” OR “resist*” OR
“subthreshold”) AND “depress*”] OR MeSH: Dysthymic
Disorder OR MeSh: Neurasthenia without any language or
Low chronicity High chronicity time restrictions. All articles resulting from these searches
and key references cited in those articles were reviewed. Key
papers from the identified publications on the basis of topic
covered and quality of research were selected.
Episodic minor Dysthymic
depression disorder
research and the development of treatments for persis­
tent depressive disorder have been hindered by the slow
Low severity process of finding a consensus regarding definitions and
diagnostic criteria, and by underestimating how severe
and functionally impairing persistent depressive disorder
Figure: Two-dimensional classification of depression with separate, is.
orthogonal axes for severity and chronicity
Additionally, it is now well established that dimensional
Higher saturation indicates greater impairment.
measurement is generally superior to categories with
respect to reliability and statistical power.127 This super­
models, such as modular psychotherapy, are a promising iority has spurred interest in dimensional classification
novel approach to address the multifaceted problems and systems that are empirically based.128 Unfortunately,
the high rate of comorbid disorders in patients with current dimensional classification systems focus
persistent depres­sive disorder. Specific modules are exclusively on cross-sectional symptomatology, and do not
selected by use of universally applied therapeutic princi­ incorporate longitudinal aspects, such as the development
ples, depending on the patients’ presenting problems.126 and course of symptoms. To address this, Klein129 proposed
The treatment modules for patients with persistent a two-dimensional classification of depression with
depressive disorder should be aimed at: the patient’s separate, orthogonal axes for severity and chronicity. Such
mentalising capabilities, recog­nising the conse­quences a system would give equal priority to both factors and
of one’s own behaviour, developing authentic empathy, crossing the two dimensions accounts for most forms of
increasing motivation, and overcoming avoid­ ance; depression that are currently recognised (figure).
improving active social problem solving, including social As we described, many factors appear to contribute to
skills; the healing of interpersonal trauma (including persistent depressive disorder, including interpersonal,
being able to connect with and trust other people); behavioural, cognitive, emotional, environmental, and
improving depressive symptoms, such as hopelessness biological factors. Thus, a single treatment approach is
and helplessness; improving quality of life; and reducing unlikely to be universally successful. Corresponding
other comorbidities. with its multifactorial nature, treatment of persistent
depressive disorder will need to adopt a modular
Future directions in the classification, approach, for instance by use of structured elements
understanding of cause, and treatment taken from CBASP as free-standing modules. Pharma­
Several priorities exist for future research on persistent cotherapy modules will need to be integrated into this
depressive disorder including dysthymia. Most research approach. Modular treatments are assumed to be more
on depression does not distinguish between chronic and efficient than traditional psychotherapy approached as
episodic depression, contributing to the clinical and they allow therapists to tailor a unique treatment for
causal heterogeneity that burdens the field of depression. each patient by selecting evidence-based therapeutic
However, chronicity, early onset, and comorbidity often strategies.130 More high-quality research into outcomes
cluster together, and each can contribute to the differ­ and processes and more dismantling studies are
ences between chronic and non-chronic depres­sions. It necessary to identify which patients benefit from which
would be informative to tease apart the variance treatment strategies. This research will result in an
contributed by each of these factors. Chronicity is algorithm for the optimal sequencing and combination
considered the most important factor in the complex of psychological and pharmacological interventions. In
clinical presentation of persistent depressive disorder; summary, a modular treatment tailored to the risk factors
it is important to test this empirically. Unfortunately, of the individual patient, exemplifying a so-called

808 www.thelancet.com/psychiatry Vol 7 September 2020

 Review

precision psychiatry approach, might aid in lowering the 19 Rhebergen D, Graham R. The re-labelling of dysthymic disorder
individual and societal burden conferred by persistent to persistent depressive disorder in DSM-5: old wine in new bottles?
Curr Opin Psychiatry 2014; 27: 27–31.
depressive disorder, including dysthymia. 20 Klein DN, Shankman SA, Lewinsohn PM, Rohde P, Seeley JR.
Contributors Family study of chronic depression in a community sample
All authors planned, conceptualised, wrote, and edited this Review and of young adults. Am J Psychiatry 2004; 161: 646–53.
take joint responsibility for its contents. 21 Sang W, Li Y, Su L, et al. A comparison of the clinical characteristics
of Chinese patients with recurrent major depressive disorder with
Declaration of interests and without dysthymia. J Affect Disord 2011; 135: 106–10.
ES received book royalties and honoraria for workshops and 22 Klein DN, Riso LP, Donaldson SK, et al. Family study of early
presentations relating to Interpersonal Psychotherapy and CBASP. onset dysthymia: mood and personality disorders in relatives
TAF reports grants and personal fees from Mitsubishi-Tanabe and of outpatients with dysthymia and episodic major depression and
personal fees from Merck Sharp & Dohme and Shionogi, outside of the normal controls. Arch Gen Psychiatry 1995; 52: 487–96.
submitted work; and he has a patent (2018–177688) pending concerning 23 Klein DN, Shankman SA, Rose S. Ten-year prospective follow-up
the use of artificial intelligence in smartphone cognitive behavioural study of the naturalistic course of dysthymic disorder and double
therapy applications. DNK, KD, and ME declare no competing interests. depression. Am J Psychiatry 2006; 163: 872–80.
24 Levkovitz Y, Tedeschini E, Papakostas GI. Efficacy of
References antidepressants for dysthymia: a meta-analysis of placebo-controlled
1 Murphy JA, Byrne GJ. Prevalence and correlates of the proposed randomized trials. J Clin Psychiatry 2011; 72: 509–14.
DSM-5 diagnosis of chronic depressive disorder. J Affect Disord
2012; 139: 172–180. 25 von Wolff A, Hölzel LP, Westphal A, Härter M, Kriston L. Selective
serotonin reuptake inhibitors and tricyclic antidepressants in the
2 Angst J, Gamma A, Rössler W, Ajdacic V, Klein DN. Long-term acute treatment of chronic depression and dysthymia: a systematic
depression versus episodic major depression: results from the review and meta-analysis. J Affect Disord 2013; 144: 7–15.
prospective Zurich study of a community sample. J Affect Disord
2009; 115: 112–121. 26 Leader JB, Klein DN. Social adjustment in dysthymia, double
depression and episodic major depression. J Affect Disord 1996;
3 Satyanarayana S, Enns MW, Cox BJ, Sareen J. Prevalence and 37: 91–101.
correlates of chronic depression in the Canadian community health
survey: mental health and well-being. Can J Psychiatry 2009; 27 Klein DN, Kocsis JH, McCullough JP, Holzer CE, Hirschfeld RM,
54: 389–98. Keller MB. Symptomatology in dysthymic and major depressive
disorder. Psychiatr Clin North Am 1996; 19: 41–53.
4 Klein DN, Black SR. Persistent depressive disorder.
In: DeRubeis RJ, Strunk DR, eds. Oxford Handbook of Mood 28 Hung CI, Liu CY, Yang CH. Persistent depressive disorder has
Disorders. Oxford: Oxford University Press, 2017: 227–37. long-term negative impacts on depression, anxiety, and somatic
symptoms at 10-year follow-up among patients with major
5 Goodwin FK, Jamison KR. Manic-depressive illness: bipolar depressive disorder. J Affect Disord 2019; 243: 255–61.
disorders and recurrent depression. Oxford: Oxford University
Press, 2007. 29 Keller MB, Shapiro RW. ‘Double depression’: superimposition
of acute depressive episodes on chronic depressive disorders.
6 Akiskal HS. Dysthymic disorder: psychopathology of proposed Am J Psychiatry 1982; 139: 438–42.
chronic depressive subtypes. Am J Psychiatry 1983; 140: 11–20.
30 Devanand DP, Adorno E, Cheng J, et al. Late onset dysthymic
7 Weissman MM, Klerman GL. The chronic depressive in the disorder and major depression differ from early onset dysthymic
community: unrecognized and poorly treated. Compr Psychiatry disorder and major depression in elderly outpatients. J Affect Disord
1977; 18: 523–32. 2004; 78: 259–67.
8 Klein DN, Riso LP, Anderson RL. DSM-III-R dysthymia: 31 Klein DN, Schatzberg AF, McCullough JP, et al. Age of onset
antecedents and underlying assumptions. in chronic major depression: relation to demographic and clinical
Prog Exp Pers Psychopathol Res 1993; 16: 222–53. variables, family history, and treatment response. J Affect Disord
9 Kocsis JH, Frances AJ. A critical discussion of DSM-III dysthymic 1999; 55: 149–57.
disorder. Am J Psychiatry 1987; 144: 1534–42. 32 Klein DN, Schatzberg AF, McCullough JP, et al. Early versus
10 Blanco C, Okuda M, Markowitz JC, Liu S-M, Grant BF, Hasin DS. late onset dysthymic disorder: comparison in out-patients with
The epidemiology of chronic major depressive disorder and superimposed major depressive episodes. J Affect Disord 1999;
dysthymic disorder: results from the national epidemiologic survey on 52: 187–96.
alcohol and related conditions. J Clin Psychiatry 2010; 71: 1645–56. 33 Klein DN, Taylor EB, Dickstein S, Harding K. The early-late onset
11 Garyfallos G, Adarnopoulou A, Karastergiou A, et al. Personality distinction in DSM-III-R dysthymia. J Affect Disord 1988;
disorders in dysthymia and major depression. Acta Psychiatr Scand 14: 25–33.
1999; 99: 332–40. 34 Barzega G, Maina G, Venturello S, Bogetto F. Dysthymic disorder:
12 Pepper CM, Klein DN, Anderson RL, Riso LP, Ouimette PC, clinical characteristics in relation to age at onset. J Affect Disord
Lizardi H. DSM-III-R axis II comorbidity in dysthymia and major 2001; 66: 39–46.
depression. Am J Psychiatry 1995; 152: 239–47. 35 Vandeleur CL, Fassassi S, Castelao E, et al. Prevalence and
13 Klein DN, Taylor EB, Dickstein S, Harding K. Primary early-onset correlates of DSM-5 major depressive and related disorders in the
dysthymia: comparison with primary nonbipolar nonchronic major community. Psychiatry Res 2017; 250: 50–58.
depression on demographic, clinical, familial, personality, and 36 Jonsson U, Bohman H, von Knorring L, Olsson G, Paaren A,
socioenvironmental characteristics and short-term outcome. von Knorring AL. Mental health outcome of long-term and episodic
J Abnorm Psychol 1988; 97: 387–98. adolescent depression: 15-year follow-up of a community sample.
14 Kotov R, Gamez W, Schmidt F, Watson D. Linking ‘big’ personality J Affect Disord 2011; 130: 395–404.
traits to anxiety, depressive, and substance use disorders: 37 Lizardi H, Klein DN, Ouimette PC, Riso LP, Anderson RL,
a meta-analysis. Psychol Bull 2010; 136: 768–821. Donaldson SK. Reports of the childhood home environment
15 Rhebergen D, Beekman AT, Graaf R de, et al. The three-year in early onset dysthymia and episodic major depression.
naturalistic course of major depressive disorder, dysthymic disorder J Abnorm Psychol 1995; 104: 132–39.
and double depression. J Affect Disord 2009; 115: 450–59. 38 McCullough JP Jr, Klein DN, Keller MB, et al. Comparison
16 Wiersma JE, van Oppen P, van Schaik DJF, van der Does AJW, of DSM-III-R chronic major depression and major depression
Beekman ATF, Penninx BWJH. Psychological characteristics of superimposed on dysthymia (double depression): validity of the
chronic depression: a longitudinal cohort study. J Clin Psychiatry distinction. J Abnorm Psychol 2000; 109: 419–27.
2011; 72: 288–94. 39 Yang T, Dunner DL. Differential subtyping of depression.
17 Klein DN. Persistent depressive disorder: commentary on Depress Anxiety 2001; 13: 11–17.
Parker and Malhi. Can J Psychiatry 2020; 65: 16–18. 40 McCullough JP Jr, Klein DN, Borian FE, et al. Group comparisons
18 Parker G, Malhi GS. Persistent depression: should such a DSM-5 of DSM-IV subtypes of chronic depression: validity of the
diagnostic category persist? Can J Psychiatry 2019; 64: 177–79. distinctions, part 2. J Abnorm Psychol 2003; 112: 614–22.

www.thelancet.com/psychiatry Vol 7 September 2020 809

 Review

41 Brown S, Rittenbach K, Cheung S, McKean G, MacMaster FP, 63 Depue RA, Krauss S, Spoont MR, Arbisi P. General behavior
Clement F. Current and common definitions of treatment-resistant inventory identification of unipolar and bipolar affective conditions
depression: findings from a systematic review and qualitative in a nonclinical university population. J Abnorm Psychol 1989;
interviews. Can J Psychiatry 2019; 64: 380–87. 98: 117–26.
42 Kessler RC, McGonagle KA, Zhao S, et al. Lifetime and 12-month 64 Danielson CK, Youngstrom EA, Findling RL, Calabrese JR.
prevalence of DSM-III-R psychiatric disorders in the United States: Discriminative validity of the general behavior inventory using
results from the national comorbidity survey. Arch Gen Psychiatry youth report. J Abnorm Child Psychol 2003; 31: 29–39.
1994; 51: 8–19. 65 Klein DN, Dickstein S, Taylor EB, Harding K. Identifying chronic
43 Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, affective disorders in outpatients: validation of the general behavior
Walters EE. Lifetime prevalence and age-of-onset distributions inventory. J Consult Clin Psychol 1989; 57: 106–11.
of DSM-IV disorders in the national comorbidity survey replication. 66 Catalán R, Gallart JM, Castellanos JM, Galard R. Plasma
Arch Gen Psychiatry 2005; 62: 593–602. corticotropin-releasing factor in depressive disorders. Biol Psychiatry
44 Markkula N, Suvisaari J, Saarni SI, et al. Prevalence and correlates 1998; 44: 15–20.
of major depressive disorder and dysthymia in an eleven-year 67 Leake A, Griffiths HW, Ferrier IN. Plasma N-POMC, ACTH and
follow-up—results from the Finnish health 2011 survey. cortisol following hCRH administration in major depression and
J Affect Disord 2015; 173: 73–80. dysthymia. J Affect Disord 1989; 17: 57–64.
45 Riolo SA, Nguyen TA, Greden JF, King CA. Prevalence of depression 68 Szádóczky E, Fazekas I, Rihmer Z, Arató M. The role of
by race/ethnicity: findings from the national health and nutrition psychosocial and biological variable in separating chronic and
examination survey III. Am J Public Health 2005; 95: 998–1000. non-chronic major depression and early-late-onset dysthymia.
46 Rubio JM, Markowitz JC, Alegría A, et al. Epidemiology of chronic J Affect Disord 1994; 32: 1–11.
and nonchronic major depressive disorder: results from the 69 Griffiths J, Ravindran AV, Merali Z, Anisman H. Dysthymia: a
national epidemiologic survey on alcohol and related conditions. review of pharmacological and behavioral factors. Mol Psychiatry
Depress Anxiety 2011; 28: 622–31. 2000; 5: 242–61.
47 Benazzi F. Chronic depression: a case series of 203 outpatients 70 Howland RH, Thase ME. Biological studies of dysthymia.
treated at a private practice. J Psychiatry Neurosci 1998; 23: 51–55. Biol Psychiatry 1991; 30: 283–304.
48 Ildirli S, Şair YB, Dereboy F. Persistent depression as a novel 71 Miller GA, Yee CM. Risk for severe psychopathology: psychometric
diagnostic category: results from the Menderes depression study. screening and psychophysiological assessment. In: Ackles PK,
Noro Psikiyatr Ars 2015; 52: 359–66. Jennings JR, Coles MGH, eds. Advances in psychophysiology, vol 5.
49 Markowitz JC, Moran ME, Kocsis JH, Frances AJ. Prevalence and London: Jessica Kingsley, 1994: 1–54.
comorbidity of dysthymic disorder among psychiatric outpatients. 72 Lyons MJ, Eisen SA, Goldberg J, et al. A registry-based twin study
J Affect Disord 1992; 24: 63–71. of depression in men. Arch Gen Psychiatry 1998; 55: 468–72.
50 Barnhofer T, Brennan K, Crane C, Duggan D, Williams JMG. 73 Torgersen S. Genetic factors in moderately severe and mild affective
A comparison of vulnerability factors in patients with persistent disorders. Arch Gen Psychiatry 1986; 43: 222–26.
and remitting lifetime symptom course of depression. 74 Lyoo IK, Kwon JS, Lee SJ, et al. Decrease in genu of the corpus
J Affect Disord 2014; 152: 155–61. callosum in medication-naïve, early-onset dysthymia and depressive
51 van Randenborgh A, Hüffmeier J, Victor D, Klocke K, personality disorder. Biol Psychiatry 2002; 52: 1134–43.
Borlinghaus J, Pawelzik M. Contrasting chronic with episodic 75 Vilgis V, Vance A, Cunnington R, Silk TJ. White matter
depression: an analysis of distorted socio-emotional information microstructure in boys with persistent depressive disorder.
processing in chronic depression. J Affect Disord 2012; 141: 177–84. J Affect Disord 2017; 221: 11–16.
52 Wiersma JE, Hovens JGFM, van Oppen P, et al. The importance 76 Posner J, Hellerstein DJ, Gat I, et al. Antidepressants normalize the
of childhood trauma and childhood life events for chronicity default mode network in patients with dysthymia. JAMA Psychiatry
of depression in adults. J Clin Psychiatry 2009; 70: 983–89. 2013; 70: 373–82.
53 Wu W, Wang Z, Wei Y, et al. Clinical features of patients with 77 Ravindran AV, Smith A, Cameron C, et al. Toward a functional
dysthymia in a large cohort of Han Chinese women with recurrent neuroanatomy of dysthymia: a functional magnetic resonance
major depression. PLoS One 2013; 8: e83490. imaging study. J Affect Disord 2009; 119: 9–15.
54 Brown GW, Craig TKJ, Harris TO. Parental maltreatment and 78 Vilgis V, Chen J, Silk TJ, Cunnington R, Vance A. Frontoparietal
proximal risk factors using the childhood experience of care & function in young people with dysthymic disorder
abuse (CECA) instrument: a life-course study of adult chronic (DSM-5: persistent depressive disorder) during spatial working
depression — 5. J Affect Disord 2008; 110: 222–33. memory. J Affect Disord 2014; 160: 34–42.
55 Horwitz AV, Widom CS, McLaughlin J, White HR. The impact 79 Niculescu AB III, Akiskal HS. Proposed endophenotypes
of childhood abuse and neglect on adult mental health: of dysthymia: evolutionary, clinical and pharmacogenomic
a prospective study. J Health Soc Behav 2001; 42: 184–201. considerations. Mol Psychiatry 2001; 6: 363–66.
56 Mondimore FM, Zandi PP, MacKinnon DF, et al. Familial 80 Cuijpers P, van Straten A, Schuurmans J, van Oppen P, Hollon SD,
aggregation of illness chronicity in recurrent, early-onset major Andersson G. Psychotherapy for chronic major depression and
depression pedigrees. Am J Psychiatry 2006; 163: 1554–60. dysthymia: a meta-analysis. Clin Psychol Rev 2010; 30: 51–62.
57 Klein DN, Kotov R. Course of depression in a 10-year prospective 81 Cuijpers P, Andersson G, Donker T, Straten A van. Psychological
study: evidence for qualitatively distinct subgroups. treatment of depression: results of a series of meta-analyses.
J Abnorm Psychol 2016; 125: 337–48. Nord J Psychiatry 2011; 65: 354–64.
58 McCullough JP Jr, Clark SW, Klein DN, First MB. Introducing 82 Thase ME. Preventing relapse and recurrence of depression: a brief
a clinical course-graphing scale for DSM-5 mood disorders. review of therapeutic options. CNS Spectr 2006; 11: 12–21.
Adv Psychiatr Treat 2016; 70: 383–92.
83 Kocsis JH, Gelenberg AJ, Rothbaum B, et al. Chronic forms
59 Klein DN, Black SR. Persistent depressive disorder: dysthymia of major depression are still undertreated in the 21st century:
and chronic depression. In: Craighead WE, Miklowitz DJ, systematic assessment of 801 patients presenting for treatment.
Craighead LW, eds. Psychopathology: history, theory, and diagnosis, J Affect Disord 2008; 110: 55–61.
2nd edn. Hoboken, NJ: John Wiley & Sons, 2013: 334–63.
84 Gopinath S, Katon WJ, Russo JE, Ludman EJ. Clinical factors
60 Keller MB, Klein DN, Hirschfeld RM, et al. Results of the DSM-IV associated with relapse in primary care patients with chronic
mood disorders field trial. Am J Psychiatry 1995; 152: 843–49. or recurrent depression. J Affect Disord 2007; 101: 57–63.
61 Mason BJ, Kocsis JH, Leon AC, et al. Assessment of symptoms 85 Nelson J, Klumparendt A, Doebler P, Ehring T. Childhood
and change in dysthymic disorder. In: Kocsis JH, Klein DN, eds. maltreatment and characteristics of adult depression: meta-analysis.
Diagnosis and treatment of chronic depression. New York, NY: Br J Psychiatry 2017; 210: 96–104.
Guilford Press, 1995: 73–88.
86 Nemeroff CB, Heim CM, Thase ME, et al. Differential responses
62 Hellerstein DJ, Batchelder ST, Lee A, Borisovskaya M. Rating to psychotherapy versus pharmacotherapy in patients with chronic
dysthymia: an assessment of the construct and content validity of the forms of major depression and childhood trauma.
Cornell Dysthymia Rating Scale. J Affect Disord 2002; 71: 85–96. Proc Natl Acad Sci USA 2003; 100: 14293–96.

810 www.thelancet.com/psychiatry Vol 7 September 2020

 Review

87 Klein JP, Erkens N, Schweiger U, et al. Does childhood maltreatment 106 Steidtmann D, Manber R, Arnow BA, et al. Patient treatment
moderate the effect of the cognitive behavioral analysis system of preference as a predictor of response and attrition in treatment
psychotherapy versus supportive psychotherapy in persistent for chronic depression. Depress Anxiety 2012; 29: 896–905.
depressive disorder? Psychother Psychosom 2018; 87: 46–48. 107 Browne G, Steiner M, Roberts J, et al. Sertraline and/or
88 McCullough JP, Kasnetz MD, Braith JA, et al. A longitudinal study interpersonal psychotherapy for patients with dysthymic disorder
of an untreated sample of predominantly late onset in primary care: 6-month comparison with longitudinal 2-year
characterological dysthymia. J Nerv Ment Dis 1988; 176: 658–67. follow-up of effectiveness and costs. J Affect Disord 2002; 68: 317–30.
89 Kocsis JH, Frances AJ, Voss C, Mann JJ, Mason BJ, Sweeney J. 108 de Mello MF, Myczcowisk LM, Menezes PR. A randomized
Imipramine treatment for chronic depression. Arch Gen Psychiatry controlled trial comparing moclobemide and moclobemide plus
1988; 45: 253. interpersonal psychotherapy in the treatment of dysthymic disorder.
90 Schramm E, Kriston L, Zobel I, et al. Effect of disorder-specific J Psychother Pract Res 2001; 10: 117–23.
vs nonspecific psychotherapy for chronic depression: a randomized 109 Markowitz JC, Kocsis JH, Bleiberg KL, Christos PJ, Sacks M.
clinical trial. JAMA Psychiatry 2017; 74: 233–42. A comparative trial of psychotherapy and pharmacotherapy for
91 Schramm E, Zobel I, Schoepf D, et al. Cognitive behavioral analysis “pure” dysthymic patients. J Affect Disord 2005; 89: 167–75.
system of psychotherapy versus escitalopram in chronic major 110 Ravindran AV, Anisman H, Merali Z, et al. Treatment of primary
depression. Psychother Psychosom 2015; 84: 227–40. dysthymia with group cognitive therapy and pharmacotherapy:
92 Wiersma JE, Schaik DJFV, Hoogendorn AW, et al. The effectiveness clinical symptoms and functional impairments. Am J Psychiatry
of the cognitive behavioral analysis system of psychotherapy for 1999; 156: 1608–17.
chronic depression: a randomized controlled trial. 111 McCullough JP. Treatment for chronic depression: cognitive
Psychother Psychosom 2014; 83: 263–69. behavioral analysis system of psychotherapy. New York, NY:
93 Assmann N, Schramm E, Kriston L, et al. Moderating effect of Guilford Press, 2000.
comorbid anxiety disorders on treatment outcome in a randomized 112 McCullough JP Jr, Schramm E. Cognitive behavioural analysis
controlled psychotherapy trial in early-onset persistently depressed system of psychotherapy for persistent depressive disorder. In:
outpatients. Depress Anxiety 2018; 35: 1001–08. Norcross JC, Goldfried MR, eds. Handbook of psychotherapy
94 Kriston L, von Wolff A, Westphal A, Hölzel LP, Härter M. Efficacy integration, 3rd edn. Oxford: Oxford University Press, 2018: 303–21.
and acceptability of acute treatments for persistent depressive 113 Constantino MJ, Manber R, DeGeorge J, et al. Interpersonal styles
disorder: a network meta-analysis. Depress Anxiety 2014; 31: 621–30. of chronically depressed outpatients: profiles and therapeutic
95 Fournier JC, DeRubeis RJ, Shelton RC, Hollon SD, Amsterdam JD, change. Psychotherapy (Chic) 2008; 45: 491–506.
Gallop R. Prediction of response to medication and cognitive 114 Constantino MJ, Laws HB, Arnow BA, Klein DN, Rothbaum BO,
therapy in the treatment of moderate to severe depression. Manber R. The relation between changes in patients’ interpersonal
J Consult Clin Psychol 2009; 77: 775–87. impact messages and outcome in treatment for chronic depression.
96 Löwe B, Schenkel I, Bair MJ, Göbel C. Efficacy, predictors of therapy J Consult Clin Psychol 2012; 80: 354–64.
response, and safety of sertraline in routine clinical practice: 115 Constantino MJ, Laws HB, Coyne AE, et al. Change in patients’
prospective, open-label, non-interventional postmarketing interpersonal impacts as a mediator of the alliance-outcome
surveillance study in 1878 patients. J Affect Disord 2005; 87: 271–79. association in treatment for chronic depression.
97 Furukawa TA, Kato T, Shinagawa Y, et al. Prediction of remission J Consult Clin Psychol 2016; 84: 1135–44.
in pharmacotherapy of untreated major depression: development 116 Arnow BA, Steidtmann D, Blasey C, et al. The relationship between
and validation of multivariable prediction models. Psychol Med 2019; the therapeutic alliance and treatment outcome in two distinct
49: 2405–13. psychotherapies for chronic depression. J Consult Clin Psychol 2013;
98 Trivedi MH, Rush AJ, Wisniewski SR, et al. Evaluation of outcomes 81: 627–38.
with citalopram for depression using measurement-based care in 117 Serbanescu I, Walter H, Schnell K, et al. Combining baseline
STAR*D: implications for clinical practice. Am J Psychiatry 2006; characteristics to disentangle response differences to
163: 28–40. disorder-specific versus supportive psychotherapy in patients with
99 Sung SC, Haley CL, Wisniewski SR, et al. The impact of chronic persistent depressive disorder. Behav Res Ther 2019; 124: 103512.
depression on acute and long-term outcomes in a randomized trial 118 Keller MB, Kocsis JH, Thase ME, et al. Maintenance phase efficacy
comparing selective serotonin reuptake inhibitor monotherapy of sertraline for chronic depression: a randomized controlled trial.
versus each of 2 different antidepressant medication combinations. JAMA 1998; 280: 1665–72.
J Clin Psychiatry 2012; 73: 967–76. 119 Keller MB, Trivedi MH, Thase ME, et al. The prevention of
100 Machmutow K, Meister R, Jansen A, et al. Comparative recurrent episodes of depression with venlafaxine for two years
effectiveness of continuation and maintenance treatments for (PREVENT) study: outcomes from the 2-year and combined
persistent depressive disorder in adults. Cochrane Database Syst Rev maintenance phases. J Clin Psychiatry 2007; 68: 1246–56.
2019; 5: CD012855. 120 Schramm E, Kriston L, Elsaesser M, et al. Two-year follow-up after
101 Spijker J, van Straten A, Bockting CLH, Meeuwissen JAC, treatment with the cognitive behavioral analysis system of
van Balkom AJLM. Psychotherapy, antidepressants, and their psychotherapy versus supportive psychotherapy for early-onset
combination for chronic major depressive disorder: a systematic chronic depression. Psychother Psychosom 2019; 88: 154–64.
review. Can J Psychiatry 2013; 58: 386–92. 121 Bausch P, Fangmeier T, Schramm E, et al. Cognitive behavioral
102 von Wolff A, Hölzel LP, Westphal A, Härter M, Kriston L. analysis system of psychotherapy versus escitalopram in patients
Combination of pharmacotherapy and psychotherapy in the with chronic depression: results from a naturalistic long-term
treatment of chronic depression: a systematic review and follow-up. Psychother Psychosom 2017; 86: 308–10.
meta-analysis. BMC Psychiatry 2012; 12: 61. 122 Klein DN, Santiago NJ, Vivian D, et al. Cognitive-behavioral analysis
103 Furukawa TA, Efthimiou O, Weitz ES, et al. Cognitive-behavioral system of psychotherapy as a maintenance treatment for chronic
analysis system of psychotherapy, drug, or their combination for depression. J Consult Clin Psychol 2004; 72: 681–88.
persistent depressive disorder: personalizing the treatment choice 123 Schatzberg AF, Rush AJ, Arnow BA, et al. Chronic depression:
using individual participant data network metaregression. medication (nefazodone) or psychotherapy (CBASP) is effective
Psychother Psychosom 2018; 87: 140–53. when the other is not. Arch Gen Psychiatry 2005; 62: 513–20.
104 Molenaar PJ, Dekker J, Van R, Hendriksen M, Vink A, 124 Santiago NJ, Klein DN, Vivian D, et al. The therapeutic alliance
Schoevers RA. Does adding psychotherapy to pharmacotherapy and CBASP-specific skill acquisition in the treatment of chronic
improve social functioning in the treatment of outpatient depression. Cognit Ther Res 2005; 29: 803–17.
depression? Depress Anxiety 2007; 24: 553–62. 125 Jobst A, Brakemeier EL, Buchheim A, et al. European Psychiatric
105 Kocsis JH, Leon AC, Markowitz JC, et al. Patient preference Association guidance on psychotherapy in chronic depression
as a moderator of outcome for chronic forms of major depressive across Europe. Eur Psychiatry 2016; 33: 18–36.
disorder treated with nefazodone, cognitive behavioral analysis 126 Chorpita BF, Daleiden EL, Weisz JR. Modularity in the design and
system of psychotherapy, or their combination. J Clin Psychiatry application of therapeutic interventions. Appl Prev Psychol 2005;
2009; 70: 354–61. 11: 141–56.

www.thelancet.com/psychiatry Vol 7 September 2020 811

 Review

127 Markon KE, Chmielewski M, Miller CJ. The reliability and validity 130 Sauer-Zavala S, Gutner CA, Farchione TJ, Boettcher HT, Bullis JR,
of discrete and continuous measures of psychopathology: Barlow DH. Current definitions of “transdiagnostic” in treatment
a quantitative review. Psychol Bull 2011; 137: 856–79. development: a search for consensus. Behav Ther 2017; 48: 128–38.
128 Kotov R, Krueger RF, Watson D, et al. The hierarchical taxonomy
© 2020 Elsevier Ltd. All rights reserved.
of psychopathology (HiTOP): a dimensional alternative to
traditional nosologies. J Abnorm Psychol 2017; 126: 454–77.
129 Klein DN. Classification of depressive disorders in the DSM-V:
proposal for a two-dimension system. J Abnorm Psychol 2008;
117: 552–60.

812 www.thelancet.com/psychiatry Vol 7 September 2020