Persistent Depression: Should Such a DSM-5 Diagnostic Category Persist?

If you’re not confused, you’re not paying attention. —Anon

DSM-5 has introduced yet another type of depressive diagnosis—persistent depressive disorder (dysthymia) (PDD)—which it describes as “a consolidation of DSM-IV-defined chronic major depressive disorder and dysthymic disorder” (p. 168). ¹ The condition and its definition have attracted little consideration and thus encouraged this critique. In light of its problematic status (as we review), we will argue for its potential inclusion, albeit modified as a duration specifier for major depression.

Its criteria A and C require a depressed mood for at least 2 years (or 1 year in children and adolescents), for most of the day, for more days than not, and for the individual to not be without symptoms for more than 2 months over that period. Thus, while titled a “persistent” mood disorder with implications of constancy, its definition allows its absence across the day, over days, and for a period of up to 2 months. Thus, the first concern is that persistence is not mandated.

Criterion B requires 2 or more of the following 6 symptoms: poor appetite or overeating, insomnia or hypersomnia, low energy or fatigue, low self-esteem, poor concentration or difficulty making decisions, and feelings of hopelessness. Other criteria allow a coterminous major depressive episode but disallow making the diagnosis in bipolar patients, in those with psychotic conditions, or when symptoms can be attributed to substance abuse or a medical condition. Criterion H requires that the symptoms cause “clinically significant distress or impairment in social, occupational or other important areas of functioning.” ¹

As noted earlier, PDD is positioned as a “consolidation” of DSM-IV chronic major depressive disorder (MDD) and dysthymic disorder. ² DSM-IV criteria for a chronic major depressive episode (MDE) include meeting MDE criteria “continuously for the past 2 years” (p. 382). ² However, as noted, persistence over 2 years is not mandatory for a diagnosis of PDD to be made. Criterion D for PDD states that “criteria for a major depressive disorder may be continuously present for 2 years,” in which case a diagnosis of both PDD and MDD should be assigned, while PDD specifiers allow MDD to be absent or, alternately, intermittently or persistently present over the preceding 2 years—and thus allow their independence rather than interdependence as suggested by the supposed coalescence of chronic major depression and dysthymia. In allowing that PDD can exist in the absence of MDD, it is unclear as to what is meant by “consolidation” of the 2 conditions.

The term double depression emerged following the introduction of both MDD and dysthymia in the DSM-III manual. ³ It is generally viewed as capturing a scenario whereby an individual experiencing dysthymia (i.e., a milder state of depression lasting for 2 years) has his or her condition worsen from time to time and leading in such instances to the development of MDD being “superimposed,” so that the term double depression presupposes 2 extant conditions. In clinical practice, what is most commonly observed is depressed individuals experiencing variation in depressive severity and with any MDD condition building on a less severe depressive base (including dysthymia) rather than there being 2 concurrent (and by inference distinguishable) conditions. The attempted “consolidation” of the 2 conditions increases heterogeneity along the severity axis while reducing heterogeneity along the chronicity dimensions, an issue of relevance when severity is more likely than chronicity to dictate treatment choice. The concept of “double depression” has long merited challenge. Its potential now to be applied to those who meet criteria for PDD and then have MDD emerge or be concurrent with PDD is also worthy of clarification.

In the subsidiary description (“Diagnostic Features”) of PDD, it is stated that individuals experiencing the condition “describe their mood as sad or ‘down in the dumps’” (p. 169). Combined with the few criteria needed for such a diagnosis and some (e.g., fatigue) being common to multiple conditions or some possibly more related to personality style (e.g., low self-esteem), the case for this condition being a clinical disorder as against a prolonged normative low mood or sadness remains unclear. Indeed, the only feature arguing for its clinical status is the requirement for distress or functioning to be “clinically significant,” a criterion that is necessarily subjective but also a stated or unstated criterion for all DSM disorders.

In the descriptive section, it is observed that symptoms can “become part of an individual’s day-to-day experience” and that this is especially likely in those with an early onset who may observe that “I’ve always been this way.” Judging persistence validly is often difficult in depressed individuals but even more difficult when depressive “states” may merge with depressive personality traits. Such observations (together with the “symptom” of low self-esteem) suggest that the criteria set is likely to be affirmed as much by those with a primary personality style of low self-esteem and concomitant feelings of hopelessness as for those with a mild continuing depressed mood. Such a reframing (i.e., the condition is more a personality style with associated depressive symptoms than a primary mood disorder) is further supported by DSM-5 observing that the condition “often has an early and insidious onset” (beginning as early as childhood) and a “chronic course” (p. 170) and that early onset PDD (before age 21 years) “is strongly associated with DSM-IV Cluster B and C personality disorders” (p. 171). ¹

Further, it is stated that when “symptoms rise to the level of a major depressive disorder, they are likely to subsequently revert to a lower level” (p. 170). ¹ Such an observation again allows that the symptom complex is equally likely to reflect a personality style rather than a depressive mood disorder of necessity. While DSM-5 positions “higher levels of neuroticism (negative affectivity)” as a temperament risk factor (p. 170), it fails to concede that “persistent depressive disorder” may commonly simply reflect the manifestation of such a temperament style. ¹ Beck et al. ⁴ detailed nuances of a “cognitive triad” held by those who developed depression, in that the individual has 1) a negative view of himself or herself (e.g., defective, inadequate, deprived, worthless), 2) a tendency to interpret ongoing experiences in a negative way, and 3) a negative view of the future. Thus, for those meeting criteria for a “persistent depressive disorder,” and particularly if it is very persistent if not chronic, the appropriate formulation may be that the individual meets diagnostic criteria on the basis of his or her primary personality style alone or that such a personality style predisposes to ongoing mild depressive symptoms.

The 12-month prevalence in the United States for PDD is provided by the DSM-5 as approximately 0.5% for the condition (as against 1.5% for “chronic major depressive disorder,” p. 170). ¹ At face value, this unreferenced prevalence estimate appears very low whether it is capturing a “depressive personality” style alone and/or a smouldering depressive condition. There appear to be no published community studies of PDD alone, while it was not included in the DSM-5 field trials ⁵ examining the test-retest reliability of multiple conditions, including several mood disorders. Ildirli et al. ⁶ have seemingly reported the only clinical prevalence study, involving assessment of 140 patients being treated for depression at adult psychiatric clinics in Turkey, and showed that PDD was common, with more than 54.2% meeting DSM-5 criteria. A much larger study conducted some years earlier by Murphy and Byrne ⁷ analysed data from the 2007 Australian community survey that questioned 8841 households. It quantified a lifetime prevalence of chronic depression (lasting at least 2 years) of only 4.6% (but this group comprised those meeting criteria for both dysthymic disorder and MDD). The significant variation in these figures cannot be explained completely by differences in samples and populations and most likely reflects the variability in application of diagnostic criteria.

In considering risk factors, DSM-5 nominates parental loss or separation in addition to negative affective temperament. It is surprising that wider psychological and social factors are not nominated. DSM-5 also suggests a hereditary contribution and notes that several “brain regions (e.g., prefrontal cortex, anterior cingulate, amygdala, hippocampus) have been implicated in persistent depressive disorder,” but findings weighting biological determinants and mechanisms are few—for example, Vilgis et al. ⁸ reporting less activation in left prefrontal regions—and thus as yet not providing any clinically meaningful findings.

In summary, in reframing “dysthymia” as “persistent depression” (albeit retaining dysthymia in parenthesis), DSM-5 has further weakened an already unstable and conceptually flawed foundation diagnosis and has provided virtually no substantive scientific base for its status as a psychiatric disorder worthy of categorisation. The validity, reliability, and prevalence of PDD remain unknown, as do its determinants and underlying mechanisms. Its foundation status resides in the DSM-5 model that it consolidates DSM-IV–defined major depressive disorder and dysthymic disorder, but as detailed in relation to its criteria, each can exist without the other, so negating the “consolidation” concept. Despite the 2-word diagnosis having “persistent” in its naming, “illness” symptoms are not necessarily persistent. Additionally, its status as a mood disorder rather than as a personality style remains unclear. Its symptom set requires only 2 (out of a set of 6) relatively minor symptoms that might be expected to be experienced by a substantial percentage of the population as a consequence of temperament style and/or normal life vicissitudes, with Pepper et al. ⁹ reporting that early onset dysthymia is associated with greater axis II comorbidity than episodic major depression. Distinguishing chronic mood disorder states from personality-style nuances clinically can be problematic when they may be independent or interdependent and, with personality an antecedent factor or a consequence, an issue of some relevance in relation to PDD. Furthermore, DSM-5’s risk factors weight biological causes without there being any substantive evidence base. Finally, DSM-5 preserves the evocative model of “double depression,” which imputes 2 differing but coterminous depressive states and with one imposed on the other, rather than a dimensional model where those experiencing dysthymia are understandably at risk of simply having their depressive state worsen in severity and symptom numbers. Given the lack of definition for this diagnosis, it is perhaps not surprising that literature on PDD remains scarce even though DSM-5 was published 5 years ago and the diagnosis was drafted many years earlier.

It is important that we do not simply criticize the immense effort that is required to develop criteria for diagnostic categories that allow clinician communication and epidemiological and treatment studies to be advanced as well as assist insurance coding but that we also offer some solutions for consideration while suggesting that any DSM changes should not be made in the absence of data improving valid definition or clinical utility. The simple option would be to delete the PDD diagnosis. A clearer conceptual and operational diagnosis is difficult to conceive of as a diagnosis weighting “persistence” is not consistent with most classificatory and diagnostic systems in medicine. Perhaps a better alternative would be to begin with a definition of the principal depressive subtypes subsumed by major depression and then use duration (persistence) as a specifier rather than a definer of the listed depressive conditions. PDD could then be so subsumed or, alternately, defined as a persistent “minor” depressive disorder, although the latter option risks creating a new problem (e.g., unnecessary or inappropriate drug prescription) while seeking to solve another problem. The specifier option would also allow insight into the impact of the duration of depressive disorders and the various patterns that may exist and so facilitate studies of causes and management options for contrasting groups.