SEARCH

SEARCH BY CITATION

Science is simply common sense at best; that is, rigidly accurate in observation and merciless to fallacy in logic. – Thomas Huxley

As psychiatry lacks benchmark tests, both classification and diagnosis rely extensively on clinical description whether attempted by prototypic ‘thick description’ or, in the case of recent DSM and ICD manuals, by symptom criteria sets. The recent publication of the DSM-5 manual allows its utility as an operationalized symptom-based criteria model to be explored and preferably at multiple levels.

The development of DSM-5 occurred over a lengthy period and clearly required considerable effort to produce a manual of nearly 1000 pages. Criticisms during its development and following its publication have been distinctive in their intensity and breadth. Meta-concerns include the risk of psychiatric expansionism in diagnosis and the pathologizing of normative human states.

This editorial examines the published manual and offers a more fine-focussed critique of DSM-5 logic – at least in relation to its approach to mood disorders. Exemplars of fallacious reasoning across architectural, clinical and editorial levels are provided for closer consideration and possible redressing. Ten illustrative examples are offered for consideration.

Major depression – positioned as a diagnostic entity rather than a ‘domain diagnosis’

  1. Top of page
  2. Major depression – positioned as a diagnostic entity rather than a ‘domain diagnosis’
  3. ‘Major depression’ is disallowed by bipolar status
  4. ‘Major depression’ and grief
  5. Specifiers positioned as diagnostic subtypes and subtypes as specifiers
  6. Limited definitional ‘cleavage’ between manic and hypomanic episodes in differentiating bipolar I and II disorders
  7. How useful is the specifier ‘anxious distress’?
  8. Questions of logic in relation to the ‘Mixed features’ specifier
  9. Is dysruptive mood dsyregulation disorder a mood disorder?
  10. The specifier ‘With atypical features’
  11. Distinct quality of mood
  12. A summation
  13. Acknowledgements
  14. References

A major depressive episode was initially defined and operationalized in DSM-III, while its DSM-5 criteria preserve the same nine DSM-IV defining symptom criteria. A key concern is that, since inception, ‘major depression’ was largely positioned (and has been increasingly reified) as a clinical entity or subtype rather than as simply defining a diagnostic ‘domain’ – akin to (say) a ‘major cancer’ domain. Thus, and as noted elsewhere [1], opening statements to journal articles and research grant applications commonly position major depression as an ‘it’, opine that ‘it’ is a ‘disease’ and provide other singular descriptors such as ‘severe’, ‘chronic’ and ‘impairing.’

DSM-5 further advances its entity status in an accompanying descriptive section, informing us, for example, that major depression is ‘associated with high mortality’, neuroticism is a ‘well-established risk factor’ and that adverse childhood experiences are ‘potent risk factors’.

As detailed previously [2], in reality, major depression comprises multiple differing constituent conditions and with each subset likely to have differing biological, psychological and social causal weightings, varying intrinsic illness trajectories and differential responses to contrasting treatment modalities. Research conducted at the ‘major depression domain’ level effectively swamps such subset differences and therefore predictably has failed to identify consistent neurobiological changes, while response rates to principal (drug and non-drug) treatment modalities examined in meta-analyses are essentially comparable [3]. Such a treatment ‘equipotency’ finding promotes therapeutic eclecticism, with practitioners ‘fitting’ patients with ‘major depression’ to their preferred treatment modality (a Procrustean model) rather than therapy being ‘fitted’ to the specific determinants of the subset condition. Returning to the analogy, while the overall domain of ‘major cancers’ might generate equal response rates to radiotherapy and to chemotherapy, differing individual cancers would be expected to show quite varying responses to chemotherapy and/or to radiotherapy. Such logic argues for a diagnostic model respecting nuances of the diagnostic subtype rather than weighting the higher-order domain in aetiological and treatment considerations.

The intrinsic subset heterogeneity constrained (or homogenized) by a diagnosis of major depression is further attested to by a DSM-5 field trial study [4] involving assessment of patients by two independent clinicians. The diagnosis of major depressive disorder generated an intraclass correlation coefficient of 0.25 (the lowest of all the tabulated conditions) and judged by the authors as indicating ‘questionable’ reliability and reflecting ‘the marked heterogeneity of patients’ meeting such a diagnosis. Quantification was strikingly consistent with a kappa coefficient of 0.25 quantified by Anthony et al. [5] for DSM-III-defined major depression three decades earlier and underlining that its unreliability is at least consistent. As the prioritized ‘diagnosis’ for research endeavours and clinical management guidelines in those with ‘clinical’ depression, this species-ignoring monoculture has proliferated from a fallow terrain with limited critical evaluation.

‘Major depression’ is disallowed by bipolar status

  1. Top of page
  2. Major depression – positioned as a diagnostic entity rather than a ‘domain diagnosis’
  3. ‘Major depression’ is disallowed by bipolar status
  4. ‘Major depression’ and grief
  5. Specifiers positioned as diagnostic subtypes and subtypes as specifiers
  6. Limited definitional ‘cleavage’ between manic and hypomanic episodes in differentiating bipolar I and II disorders
  7. How useful is the specifier ‘anxious distress’?
  8. Questions of logic in relation to the ‘Mixed features’ specifier
  9. Is dysruptive mood dsyregulation disorder a mood disorder?
  10. The specifier ‘With atypical features’
  11. Distinct quality of mood
  12. A summation
  13. Acknowledgements
  14. References

A DSM-5 criterion not present in the DSM-IV diagnostic criteria set for a major depressive disorder is that there ‘has never been a manic or a hypomanic episode’ (Criterion D, page 161), raising some unanswered questions. Firstly, why? Secondly, how realistic is that exclusion criterion to clinical practice? Thirdly, what diagnosis should then be given to those with a bipolar disorder who, during depressive episodes, meet formalized major depressive symptom criteria? The criterion is also at categorical variance with the DSM-5 chapter considering bipolar disorders, and where it is stated (page 123) that the ‘vast majority of individuals’ meeting bipolar I manic episodes ‘experience major depressive episodes during the course of their lives’, and, in the diagnostic criteria set, that ‘manic episodes may have been preceded by and may be followed by…. major depressive episodes’; and that bipolar II disorder requires ‘the lifetime experience of at least one episode of major depression’.

‘Major depression’ and grief

  1. Top of page
  2. Major depression – positioned as a diagnostic entity rather than a ‘domain diagnosis’
  3. ‘Major depression’ is disallowed by bipolar status
  4. ‘Major depression’ and grief
  5. Specifiers positioned as diagnostic subtypes and subtypes as specifiers
  6. Limited definitional ‘cleavage’ between manic and hypomanic episodes in differentiating bipolar I and II disorders
  7. How useful is the specifier ‘anxious distress’?
  8. Questions of logic in relation to the ‘Mixed features’ specifier
  9. Is dysruptive mood dsyregulation disorder a mood disorder?
  10. The specifier ‘With atypical features’
  11. Distinct quality of mood
  12. A summation
  13. Acknowledgements
  14. References

The DSM-IV definition of major depression was commonly interpreted as excluding grief states although that is not actually true when the rather obtuse text is studied (and described by Wakefield [6] as ‘semantic chicanery’). Thus, DSM-IV Criterion E (page 327) obliquely stated that a diagnosis of major depression would not apply if symptoms could be better accounted for by bereavement following the ‘loss of a loved one’ (unless) such symptoms persisted for longer than 2 months or were characterized by marked functional impairment, morbid preoccupations, suicidal ideation, psychotic symptoms or psychomotor disturbance (page 327, pages 684–685).

Some earlier arguments [7] are now reprised. Firstly, pre-publication DSM-5 deliberations (as posted on the DSM-5 website) appeared to argue for positioning grief states within the major depressive category, a proposition eliciting considerable concern from both health professionals and the community about ‘pathologizing’ a normative state. Secondly, such DSM-5 deliberations more emphasized similarities between grief and major depression, rather than detailing historically derived phenomenological distinctions, or differences in natural history, staging and treatment response (especially to antidepressants). Thirdly, on the DSM-5 website, Kendler argued that, as there was little difference between major depression in response to bereavement compared to other stressors such as being physically assaulted or raped, and that the ‘DSM-IV position is not logically defensible. Either the grief exclusion criterion needs to be eliminated or extended so that no depression that arises in the setting of adversity would be diagnosable.’ This led me to suggest [7] that such logic risked opening a Pandora's box – in that rather than drawing bereavement within the domain of the clinical depressive disorders, many clinical depressive disorders (especially reactive ones) might more readily be positioned within a grief paradigm.

The now published DSM-5 manual appears to have opened that box. The manual has a Note (page 161) stating that symptom criteria for major depression may ‘resemble’ responses to ‘a significant loss (e.g. bereavement, financial ruin, losses from a national disaster, a serious medical illness or disability)’, so joining bereavement with a broader set of ‘loss’ scenarios (qua conditions). If grief is quintessentially a response to a break in an affectional or attachment bond, then its candidate loss scenarios (e.g. death or separation from a partner) should be relatively few. The non-bereavement loss ‘exemplars’ nominated in DSM-5 are difficult to so position. For example, while financial ruin is undeniably a loss, unless the individual had a Scrooge McDuck affectional bond to money, it is more likely to lead to a phenomenologically weighted reactive depressive condition rather than to a grief reaction.

The risk generated by the DSM Note is to position all ‘loss-induced depressive states’ as separate from residual major depressive states, but DSM-5 fails to offer any points of differentiation. As it would be a rare individual experiencing a new episode of clinical or major depression (meeting Criterion A symptom numbers) not to nominate losses or stressors, how logical is it for the presence or absence of a ‘significant loss’ to be the determining feature for assignment as a grief state or as a major depression episode? While DSM-5 does provide a quite useful footnote (page161) addressing the phenomenological distinction of grief and major depression, it is marred by some simple confounding inclusions. Thus, the footnote informs us that the ‘predominant affect’ in grief is feelings of ‘emptiness and loss’ as against a predominant affect of depressed mood, lowered self-esteem and anhedonia in major depression. However, Criterion A for a major depressive episode states that the required depressive mood is marked by feeling sad, empty or hopeless. Thus, both states weight an affect of ‘emptiness’. Logic would have suggested avoiding a shared descriptor.

The earlier preoccupying issue (i.e. would DSM-5 entomb grief in the depressive disorders) is not addressed other than the Note considering their co-occurrence. While DSM-5 generally adopts an authoritative tone, any prescriptive tone disappears in this section and the reader is left perplexed as to how their co-occurrence is defined or conceptualized. Inconclusiveness is evident in the Note makings a rather vague suggestion that decisions as to whether major depression is present in conjunction with ‘the normal response to a significant loss’ should be ‘carefully considered’ and that such decisions require ‘exercise of clinical judgment’ and consideration of ‘cultural norms’ in relation to loss-generated distress.

Specifiers positioned as diagnostic subtypes and subtypes as specifiers

  1. Top of page
  2. Major depression – positioned as a diagnostic entity rather than a ‘domain diagnosis’
  3. ‘Major depression’ is disallowed by bipolar status
  4. ‘Major depression’ and grief
  5. Specifiers positioned as diagnostic subtypes and subtypes as specifiers
  6. Limited definitional ‘cleavage’ between manic and hypomanic episodes in differentiating bipolar I and II disorders
  7. How useful is the specifier ‘anxious distress’?
  8. Questions of logic in relation to the ‘Mixed features’ specifier
  9. Is dysruptive mood dsyregulation disorder a mood disorder?
  10. The specifier ‘With atypical features’
  11. Distinct quality of mood
  12. A summation
  13. Acknowledgements
  14. References

The DSM-5 introduction differentiates between ‘subtypes’ and ‘specifiers’ (pages 21–22). A ‘subtype’ is positioned as defining ‘mutually exclusive and jointly exhaustive phenomenological subgroups within a diagnosis’, while a ‘specifier’ is defined by the lack of such features and capturing parameters such as course, severity and/or descriptive features. The phrase ‘jointly exhaustive’ is unclear in and of itself; if synonymous with the preceding ‘mutually exclusive’ phrase it is redundant.

A new DSM-5 depressive diagnostic subtype is ‘persistent depressive disorder,’ and described as a ‘consolidation of DSM-IV-defined dysthymia and chronic major depression’ (page 169), and as for dysthymia, requires a minimum of two symptom criteria and a duration of two or more years. While positioned as a ‘subtype’, there are three specifiers linking it to a major depressive episode, so arguing against their ‘mutual exclusivity.’ Thus, it appears to more meet DSM-5 definition of a ‘specifier’, differing from major depression by course (i.e. longer duration) and severity (i.e. two or more Criterion B symptoms) parameters.

The converse – of subtypes being positioned as specifiers – is also suggested. Major depression has several subset specifiers including ‘with melancholic features’ and ‘with (mood-congruent or mood-incongruent) psychotic features. The binary view of the depressive disorders positioned melancholic (qua endogenous) depression as a separate depressive type and with summary arguments [8] including several over-represented symptoms and signs, primary biological underpinnings, a minimal placebo response and a superior response to physical treatments (i.e. antidepressant medication and electroconvulsive therapy or ECT) than to psychotherapy. Psychotic depression is distinguished from other depressive disorders by its defining categorical feature (i.e. psychosis), while a meta-analysis [9] quantified a marked treatment gradient – in the order of one-quarter responding to antidepressant medication alone, one-third to antipsychotic medication alone and four-fifths to combination antidepressant and antipsychotic medication or to ECT. Such phenomenological and treatment differential data argue strongly for melancholic and psychotic depression as depressive subtypes (and mutually exclusive from the non-melancholic depressive conditions) rather than being positioned as major depression specifiers.

Limited definitional ‘cleavage’ between manic and hypomanic episodes in differentiating bipolar I and II disorders

  1. Top of page
  2. Major depression – positioned as a diagnostic entity rather than a ‘domain diagnosis’
  3. ‘Major depression’ is disallowed by bipolar status
  4. ‘Major depression’ and grief
  5. Specifiers positioned as diagnostic subtypes and subtypes as specifiers
  6. Limited definitional ‘cleavage’ between manic and hypomanic episodes in differentiating bipolar I and II disorders
  7. How useful is the specifier ‘anxious distress’?
  8. Questions of logic in relation to the ‘Mixed features’ specifier
  9. Is dysruptive mood dsyregulation disorder a mood disorder?
  10. The specifier ‘With atypical features’
  11. Distinct quality of mood
  12. A summation
  13. Acknowledgements
  14. References

If bipolar I disorder and its integral manic states are more severe than bipolar II and its hypomanic episodes – as generally conceptualized clinically – it might be expected that DSM criteria would capture that severity component by differing symptoms or differing symptom numbers. However, DSM-5 Criterion A defining the mood state for manic and hypomanic episodes is identical apart from imposing a minimum duration of 7 and 4 consecutive days respectively. Further, the seven Criterion B symptoms for mania and hypomania are absolutely identical. Further again, the cut-off score (of 3 or more, or 4 or more symptoms if irritable) is identical for mania and hypomania. The only differentiating criteria are impairment (marked vs. not marked), hospitalization and presence of psychotic features for mania. Thus, if an individual has a non-psychotic high that lasts more than a week and is not hospitalized, the only differentiating feature is level of impairment, a highly subjective parameter generally but even more so in those with an elevated mood state, when functioning may be unimpaired and, occasionally, actually enhanced.

How useful is the specifier ‘anxious distress’?

  1. Top of page
  2. Major depression – positioned as a diagnostic entity rather than a ‘domain diagnosis’
  3. ‘Major depression’ is disallowed by bipolar status
  4. ‘Major depression’ and grief
  5. Specifiers positioned as diagnostic subtypes and subtypes as specifiers
  6. Limited definitional ‘cleavage’ between manic and hypomanic episodes in differentiating bipolar I and II disorders
  7. How useful is the specifier ‘anxious distress’?
  8. Questions of logic in relation to the ‘Mixed features’ specifier
  9. Is dysruptive mood dsyregulation disorder a mood disorder?
  10. The specifier ‘With atypical features’
  11. Distinct quality of mood
  12. A summation
  13. Acknowledgements
  14. References

DSM-5 includes an ‘anxious distress’ specifier for a number of mood disorders (e.g. major depression, persistent depressive disorder, bipolar I and II, and cyclothymia), with a text statement (page 149) observing that anxious distress ‘has been noted as a prominent feature of both bipolar disorder and major depressive disorder’. However, as many of the DSM-5 mood disorders (e.g. major depressive and persistent depressive disorder) have a mandatory criterion that their defining symptoms cause ‘clinically significant distress,’ the specifier risks being tautological. Of greater importance, however, is its utility. Clinicians might reasonably view ‘anxious distress’ as a common concomitant (if not a defining component) of most psychiatric and medical conditions. Conversely, it can be criticized for its clinical relevance to bipolar disorder, when many such individuals generally describe during hypomanic and manic states that their anxiety attenuates or disappears, while they feel care free and invulnerable. This seeming lack of awareness of clinical nuances is displayed in several DSM-5 decision rules for the mood disorders.

Questions of logic in relation to the ‘Mixed features’ specifier

  1. Top of page
  2. Major depression – positioned as a diagnostic entity rather than a ‘domain diagnosis’
  3. ‘Major depression’ is disallowed by bipolar status
  4. ‘Major depression’ and grief
  5. Specifiers positioned as diagnostic subtypes and subtypes as specifiers
  6. Limited definitional ‘cleavage’ between manic and hypomanic episodes in differentiating bipolar I and II disorders
  7. How useful is the specifier ‘anxious distress’?
  8. Questions of logic in relation to the ‘Mixed features’ specifier
  9. Is dysruptive mood dsyregulation disorder a mood disorder?
  10. The specifier ‘With atypical features’
  11. Distinct quality of mood
  12. A summation
  13. Acknowledgements
  14. References

This specifier is allowed for manic, hypomanic and depressive episodes in those with bipolar I and bipolar II disorders. For those who meet criteria for a manic or hypomanic state, allocation to the ‘mixed features’ specifier is achieved by the presence of three of six listed depressive symptoms. The latter list (pages 149–150) corresponds somewhat to major depression criteria but does not include psychomotor agitation, sleep or appetite/weight changes or impaired concentration. The exclusion of agitation is worth challenging as, clinically, it is one of the most distinctive features of a mixed bipolar state. Patients describe feeling ‘scratchy,’ irritable, unable to sit still and often profound mental perturbation, a scenario similar to a serotonergic reaction and one that many so experiencing describe as promoting suicidal ideation (which is uncharacteristic of pure manic or hypomanic states). Koukopoulos et al. [10] also noted the exclusion of ‘characteristic’ psychomotor agitation, irritability and distractibility, and observed that the DSM-5 logic in relation to ‘mixed features’ resembled a definition of migraine that excluded symptoms of pain in the head. Thus, the ‘mixed features’ specifier excludes the most likely pathognomonic and clinically salient features of such states.

For those who meet criteria for major depression or persistent depressive disorder, three of seven manic/hypomanic features (and virtually the same DSM-5 set defining manic/hypomanic status) are required to meet the ‘mixed features’ specifier. Again, however, the key clinically salient features or motor agitation and mental perturbation are absent. Further, DSM-5 requires that mixed symptoms must be observable by others and not attributable to medication. In clinical practice, mixed states are commoner than generally conceded, often not observable by family members or even by managing clinicians as they are rarely spontaneously reported by patients. One of the most common determinants of a mixed state is the introduction of an antidepressant medication – and with noradrenergic medications likely having a higher propensity to initiate mixed states. Thus, the exclusion criteria for mixed states – neither observable by others nor a consequence of medication – are not clinically sagacious, risk compromising management and will lower prevalence estimates of such states.

Is dysruptive mood dsyregulation disorder a mood disorder?

  1. Top of page
  2. Major depression – positioned as a diagnostic entity rather than a ‘domain diagnosis’
  3. ‘Major depression’ is disallowed by bipolar status
  4. ‘Major depression’ and grief
  5. Specifiers positioned as diagnostic subtypes and subtypes as specifiers
  6. Limited definitional ‘cleavage’ between manic and hypomanic episodes in differentiating bipolar I and II disorders
  7. How useful is the specifier ‘anxious distress’?
  8. Questions of logic in relation to the ‘Mixed features’ specifier
  9. Is dysruptive mood dsyregulation disorder a mood disorder?
  10. The specifier ‘With atypical features’
  11. Distinct quality of mood
  12. A summation
  13. Acknowledgements
  14. References

In the DSM-5 set of ‘depressive disorders’, the first condition detailed is disruptive mood dysregulation disorder (DMDD). All defining criteria (A–D) refer to ‘temper outbursts’ as the central feature, while the explanatory text (page 156) observes that the ‘core feature’ is ‘chronic, severe persistent irritability’ and that it was essentially added to DSM-5 to avert concerns about the overdiagnosis of bipolar disorder in children simply showing ‘severe, non-episodic irritability’ (page 157). Logic might suggest that if childhood bipolar disorder exists, it would be defined by salient and precise criteria ensuring against false positive diagnoses rather than by creating a default ‘distinct category’ (page 157) which is more likely to be indistinct in capturing quite heterogeneous states such as attention deficit hyperactivity, oppositional defiance and several other personality-weighted states. Further, as neither the criteria set nor the explanatory text include any ‘depressive’ feature, why is this ostensibly behavioural state a formalized (and the first listed) DSM-5 depressive disorder?

The specifier ‘With atypical features’

  1. Top of page
  2. Major depression – positioned as a diagnostic entity rather than a ‘domain diagnosis’
  3. ‘Major depression’ is disallowed by bipolar status
  4. ‘Major depression’ and grief
  5. Specifiers positioned as diagnostic subtypes and subtypes as specifiers
  6. Limited definitional ‘cleavage’ between manic and hypomanic episodes in differentiating bipolar I and II disorders
  7. How useful is the specifier ‘anxious distress’?
  8. Questions of logic in relation to the ‘Mixed features’ specifier
  9. Is dysruptive mood dsyregulation disorder a mood disorder?
  10. The specifier ‘With atypical features’
  11. Distinct quality of mood
  12. A summation
  13. Acknowledgements
  14. References

In DSM-IV, this specifier was applied only to those with a unipolar or bipolar I or II course and whose major depressive or dysthymic episode was associated with mood reactivity (Criterion A) and two or more criterion B features (i.e. appetite/weight gain, hypersomnia, leaden paralysis, interpersonal rejection sensitivity). In DSM-5, it is listed as a specifier (if relevant to the ‘current or most recent episode’) for those meeting criteria for bipolar I disorder, major depressive disorder and persistent depressive disorder, but is not a listed specifier for bipolar II disorder – with that modification from DSM-IV unexplained.

The term ‘atypical depression’ has had many ascriptions [11]. Initially (and as noted on page 152 in DSM-5) it referred to the non-endogenous depressive disorders (as endogenous depression was viewed as the ‘typical’ depressive condition) and positioned by British psychiatrists as representing a primary anxiety disorder with secondary depression. Later the Colombia school weighted a personality foundation (and so-termed it ‘hysteroid dysphoria’) and even later positioned it as a primary depressive subtype selectively responsive to monoamine oxidase inhibitor antidepressants. Our research [11] favoured its primary feature being a personality style of interpersonal sensitivity (both to praise and to rejection) and leading to the adoption of homeostatic features such as hypersomnia and food cravings. This argues for its positioning as a non-melancholic ‘spectrum’ depressive condition’ (in having a predisposing personality style as well as over-represented symptoms). However, it is important to differentiate such an ‘atypical depression’ syndrome or subtype from ‘atypical symptoms’ – which historically comprise hypersomnia and hyperphagia only. The latter are common in younger patients with unipolar and bipolar melancholic depressive episodes – although a bipolar II depressive episode is one of the few mood conditions DSM-5 excludes from this specifier. In melancholia, mood reactivity is rare (and mood non-reactivity often a defining feature), and it is unclear whether any personality style (including interpersonal rejection sensitivity) is over-represented – so, while atypical symptoms are common, an atypical depressive subtype would not be expected. The key concern then is that DSM-5 seems to confuse an ‘atypical depressive’ subtype with ‘atypical symptoms’ and positions the former as the specifier rather than the more relevant latter.

Distinct quality of mood

  1. Top of page
  2. Major depression – positioned as a diagnostic entity rather than a ‘domain diagnosis’
  3. ‘Major depression’ is disallowed by bipolar status
  4. ‘Major depression’ and grief
  5. Specifiers positioned as diagnostic subtypes and subtypes as specifiers
  6. Limited definitional ‘cleavage’ between manic and hypomanic episodes in differentiating bipolar I and II disorders
  7. How useful is the specifier ‘anxious distress’?
  8. Questions of logic in relation to the ‘Mixed features’ specifier
  9. Is dysruptive mood dsyregulation disorder a mood disorder?
  10. The specifier ‘With atypical features’
  11. Distinct quality of mood
  12. A summation
  13. Acknowledgements
  14. References

A criterion for melancholia in DSM-III and DSM-IV was ‘distinct quality’ – being defined in both manuals as a ‘depressed mood perceived as distinctly different from the kind of feeling following the death of a loved one’. We [12] suggested that such a definition indicated what it was not, rather than providing positive defining characteristics – akin to defining baseball as ‘not cricket’. We further noted quite differing and protean definitions indicating the indistinct status of ‘distinct quality,’ but that there were some weightings, especially to the loss of ‘vitality’ (melancholia once being termed ‘vital depression’). DSM-5 has, admirably, now provided a definition – stating (page 185) that ‘distinct quality’ is ‘characterized by profound despondency, despair, and/or moroseness or by so-called empty mood’. This is, however, a concatenated description, containing synonyms of depression as well as some broader constructs, with the ‘or’ allowing a weighting to depressive symptoms or to a broader ‘empty mood’ construct. As noted earlier, ‘emptiness’ is positioned in DSM-5 as a defining feature of both major depression and grief. It is also included as a defining feature for melancholia, but its inclusion here is even more concerning when the objective was to define melancholia's distinct quality. If melancholia has a ‘distinct quality’ distinguishing it from other clinical depressive states, then defining it using synonyms of depression does not address that objective, with the descriptors preceding the ‘or’ providing no cleavage. As the architects of DSM-5 have argued for differentiating clinical depression from grief, and that melancholia has a distinct quality, then the ubiquitous and confounding presence of ‘emptiness’ across all three conditions is perplexing.

A summation

  1. Top of page
  2. Major depression – positioned as a diagnostic entity rather than a ‘domain diagnosis’
  3. ‘Major depression’ is disallowed by bipolar status
  4. ‘Major depression’ and grief
  5. Specifiers positioned as diagnostic subtypes and subtypes as specifiers
  6. Limited definitional ‘cleavage’ between manic and hypomanic episodes in differentiating bipolar I and II disorders
  7. How useful is the specifier ‘anxious distress’?
  8. Questions of logic in relation to the ‘Mixed features’ specifier
  9. Is dysruptive mood dsyregulation disorder a mood disorder?
  10. The specifier ‘With atypical features’
  11. Distinct quality of mood
  12. A summation
  13. Acknowledgements
  14. References

A number of individual logical fallacies within the mood disorder section of DSM-5 have been detailed. Some consideration is now given to their constituent groupings as well as some suggestions for their potential redressing.

First, the DSM-5 specification and description of ‘subtypes’ and ‘specifiers’ is somewhat fuzzy and, in application, often contrary to their definition, leading to some unexplained inconsistencies of application. Redress would require clearer definitions, a review of the actual ‘architectural’ formwork of the classificatory model and consideration as to the cogent and relevant option. Second, and a long-standing concern from its inception in DSM-III, is the positioning of ‘major depression’ as if the concept is a meaningful homogeneous entity. This concern merits greater awareness and urgently needs redressing to advance research endeavours and clinical management of its ‘meaningful’ subset conditions, rather than reifying ‘it’ as an actual diagnostic condition. In addition to examining historical candidates (e.g. melancholia), the opening of Pandora's box argues for close examination of DSM-5′s identification of loss-weighted (qua reactive depressive) disorders. How might they differ from bereavement and from non-loss induced and perhaps more autonomous depressive states – both definitionally and in relation to cause and management modalities? A third substantive consideration is to address the current poor operational cleavage between bipolar I and II disorders and their respective over-represented manic and hypomanic states. Options include generating disorder-differing symptoms, imposing differing cut-off symptom numbers or dictating that a categorical feature such as psychosis be present or absent, respectively, in mania and hypomania. The current inclusion of hospitalization as a criterion for assignment to mania should be challenged as it is an illness consequence – not an intrinsic illness-defining characteristic.

Fourth, issues of clinical salience need greater consideration. True hypomanic (and some true manic) states are not always observable to others. The specifier ‘anxious distress’ is ubiquitous to most psychiatric conditions and thus raises questions about its utility. Conversely, and of some diagnostic utility, it is often absent in elevated bipolar states – but here DSM-5 does not offer it as a specifier.

Fifth, inconsistencies and obscurities need attention. For example, why is a diagnosis of major depression disallowed by bipolar status? Definition of a subtype includes, in part, that it is ‘mutually exclusive and jointly exhaustive’. How do those two phrases differ, and what does ‘jointly exhaustive’ mean? The exclusion of agitation from the definition of ‘mixed states’, and the requirement that it be observable and not due to medication, strains credibility. As does the logic of creating ‘dysruptive mood dysregulation disorder’ as a mood disorder without any mood state criteria and as an indirect and convoluted strategy for averting the overdiagnosis of bipolar disorder in children. Confusion about defining an atypical depressive subtype as against simply defining atypical depressive features and the diagnostic implications of each require resolution.

Sixth, criteria seeking to define and differentiate conditions should have specificity rather than evidence commonality. An affect of ‘emptiness’ is common to DSM-5′s definition of major depression, grief and to defining the so-called ‘distinct quality’ of melancholia. Distinction between these three conditions is of the highest importance and should not be compromised by lack of simple logic.

De Montaigne observed that ‘No one is exempt from talking nonsense; the misfortune is to do it solemnly.’ The DSM manual is imbued with gravitas and commonly described as a bible. This carries the risk that it will be taken on faith; but it is a guide, no more, no less. Nevertheless, its defining text should be as exact and unambiguous as possible. At least in relation to the classification of mood disorders, this new testament appears more a road map to terra incognita than to terra firma. We need to question our guides as to whether we have lost our way.

References

  1. Top of page
  2. Major depression – positioned as a diagnostic entity rather than a ‘domain diagnosis’
  3. ‘Major depression’ is disallowed by bipolar status
  4. ‘Major depression’ and grief
  5. Specifiers positioned as diagnostic subtypes and subtypes as specifiers
  6. Limited definitional ‘cleavage’ between manic and hypomanic episodes in differentiating bipolar I and II disorders
  7. How useful is the specifier ‘anxious distress’?
  8. Questions of logic in relation to the ‘Mixed features’ specifier
  9. Is dysruptive mood dsyregulation disorder a mood disorder?
  10. The specifier ‘With atypical features’
  11. Distinct quality of mood
  12. A summation
  13. Acknowledgements
  14. References