Psychotic disorders in DSM-5: A paradigm shift?

DSM Digest

Schizophrenia remains one of the most feared disorders by the lay public and attracts considerable media coverage – often because of inaccurate portrayals generated to entertain rather than inform (Owen, 2012). Remarkably, despite this interest and relentless research efforts spanning more than a century, we are yet to understand the mechanism of this illness. To progress meaningfully towards a better understanding and ultimately better treatments and hopefully the prevention of psychotic disorders, we require a system that can accommodate considerable heterogeneity across affected individuals (Heckers, 2008). Variations in symptom profile, cognitive function, treatment response and longer-term outcomes are particularly important in this context. The question now is - are we, researchers and clinicians alike, going to fare any better with the changes ushered in by DSM-5?

There are some changes, but none that are dramatic, concerning ‘schizophrenia spectrum and other psychotic disorders’ (Heckers et al., 2013). Schizophrenia is much less prominent in DSM-5 than it was in DSM-IV. Unlike DSM-IV, DSM-5 does not give the impression that all individuals displaying signs and symptoms of psychosis should first be evaluated for schizophrenia. Instead, it stresses that such individuals should be evaluated on all core domains (hallucinations, delusions, disorganised thought (speech), disorganised or abnormal motor behaviour (including catatonia) and negative symptoms, with particular emphasis on diminished emotional expression and avolition before reaching (or not) the diagnosis of a Delusional Disorder, Brief Psychotic Disorder, Schizophreniform Disorder, Schizophrenia, Schizoaffective Disorder, Substance/Medication-Induced Psychotic Disorder, Catatonic Disorder due to another medical condition, or Schizotypal Personality Disorder. There is clear guidance on how to assess the severity gradients of level, number and duration of symptoms to separate psychotic disorders from each other, which should make the diagnosis of particular psychotic disorders more consistent between clinicians. Importantly, Attenuated Psychosis Syndrome is included (appendix, section III) as a condition for further study, and not a diagnosis, at least in this iteration of DSM.

There are also a number of changes relating to schizophrenia. The subtypes of schizophrenia (paranoid, disorganized, catatonic, undifferentiated, and residual) have been deleted, justifiably so, given their poor diagnostic stability, low reliability, poor validity, no clear relationship with treatment response or longitudinal course, and lack of heritability. This change is unlikely to have any significant impact on either research or clinical practice. Another change is the elimination of special treatment of bizarre delusions and special hallucinations (e.g. two or more voices conversing): while DSM-IV required only one characteristic symptom if it happened to be a bizarre delusion or Schneiderian-first rank hallucination, DSM-5 requires two of the three criterion A symptoms (hallucinations, delusions, disorganized speech) even if one of them is a bizarre delusion or special hallucination. This change, also aimed primarily towards simplification of criterion A, should have negligible impact, given that less than 2% of patients received a DSM-IV diagnosis of ‘schizophrenia’ based on a single bizarre delusion or special hallucination (Tandon et al., 2013).

There is a change affecting the diagnosis of schizoaffective disorder, which now requires a major mood episode to be present for most of the disorder’s duration. This is intended to make the boundaries between schizophrenia and schizoaffective disorder clearer, and the diagnosis of schizoaffective disorder more stable over time. Criteria for Delusional Disorder are slightly changed (no longer separate from Shared Delusional Disorder), and Catatonia in all contexts now requires three of a total of 12 symptoms. With these changes, DSM-5 should increase its concordance with ICD-11. ICD-11 is not yet final but current proposals include deletion of schizophrenia subtypes, inclusion of dimensions, no special treatment of Schniedererian first-rank symptoms, and a common set of course specifiers to chart the current status and longitudinal course of psychotic disorders. Some differences, however, are likely to remain between DSM and ICD in the diagnostic criteria for schizophrenia, such as illness duration of six months in DSM-5 versus one month in ICD, and impairment being an illness criterion in DSM-5 but not in ICD-11.

DSM-5 with its categorical symptom-based approach is certainly not a paradigm shift that some had wished for. However, it acknowledges that the boundaries between nosological entities are unlikely to be categorical and proposes a dimensional assessment of symptoms and related phenomena (section III). It advocates that clinicians should assess the patient along the five domains that define schizophrenia spectrum disorders and, in addition, on cognitive function, depression and mania to capture the heterogeneity in symptoms and severity present across individuals. This, if adopted by the majority, should allow a clearer picture of the causes and treatments of schizophrenia spectrum disorders to emerge than afforded by DSM-IV. It may also enable a link between DSM-5 and the Research Domain Criteria (RDoC) initiative (Insel et al., 2010). Any such link, however, is likely to be weak because the domains proposed by DSM-5 differ considerably from the dimensions of RDoC. The dimensional assessment of cognitive function in DSM-5 is perhaps the one aspect most aligned to RDoC, but still unlike the single dimension of cognitive function in DSM-5, RDoC differentiates, quite rightly, among different aspects of cognitive functions (e.g. attention, working memory).