Classification and the clinician

DSM-5 was bound to disappoint the clinician. Psychiatric illness may never “carve at the joints” but DSM-5 could have been a little revolutionary, could have had some “visionary changes” (Malhi, 2013: p.8). DSM-5 provides a very blunt instrument with which to pick apart complex disorders. It continues the count-the-symptoms approach - although I suspect most psychiatrists name conditions according to DSM but very rarely look up criteria. Other clinicians, not psychiatrists, speak of a patient who “meets criteria for” a DSM diagnosis.

Major Depression remains. This category has distorted the idea of a depressive illness so now “Major Depression” is in common use, has diminished our skill in obtaining exact phenomenology, has made us forget complicated clinical language and has contributed to the vast over-prescription of SSRIs and SNRIs. It means the literature is full of evidence that anti-depressants don’t work and leaves clinicians, faced with very different clinical pictures, unable to find evidence bases for factors predicting response to different treatments.

Dysthymia is gone, replaced with “Persistent Depressive Disorder”. This includes chronic major depression disorder (MDD) and may again lead to over use of anti-depressants. There was a “bereavement exclusion” in DSM-IV for severe, prolonged or psychotic depression. The bereavement change in DSM-5 further widens the concept of MDD and, has been extensively commented on in the general press, pathologises normal processes. In general, diagnostic thresholds are lowered (Malhi, 2013) and while “it is appropriate for citizens of advanced societies to identify and deal with milder forms…” (Sachdev, 2013: p.11) this lowering can make all psychiatric illness appear mild and of little consequence.

I welcome the changes in the way hypo/mania is included in the diagnosis of Bipolar Disorder although I do not think any clinicians really followed the exclusion. More recognition of Mixed States is also an advance although mixed specifiers appear to be everywhere – even manic symptoms in MDD. The inclusion of the poor insight/delusional type of Obsessive Compulsive Disorder (OCD) and Body Dysmorphic Disorder (BDD) is also welcome (although I am sorry Anorexia Nervosa didn’t make it).

Anorexia Nervosa is included in a strange new category of “Feeding and Eating Disorders” – the only common element being food. Using DSM-IV, most patients had “Eating Disorder Not Otherwise Specified” and “EDNOS” became a diagnosis. “Binge Eating Disorder” may account for some “EDNOS” but there will continue to be many patients who fall outside any of these diagnoses.

In Schizophrenia, it was time for Schneidarian first rank symptoms to go, but DSM-5 also eliminates sub-types, There are obviously many patterns of illness within what we call schizophrenia: in failing to describe sub-types it must not be assumed schizophrenia is a monolith (we should care: ABF distinguishes between sub-types with quite different rates of reimbursement: “Paranoid Schizophrenia” attracts a much larger sum than “Schizophrenia”). It is a pity Schizoaffective Disorder remains but there is some narrowing of the concept and this may reduce the wild over-diagnosis of the condition.

ICD-11 is due in 2017. ICD-11 necessarily takes a more public health view and takes into account cultural differences. It will attempt to match the diagnostic process rather than set out operational diagnostic criteria. ICD-11 has to fit many applications and be usable by psychiatrically trained and untrained staff. However, the development of a Primary Care version of ICD-11 - similar to that of ICD-10 – may undermine an assertion that diagnosis in psychiatric illness is as important as that in other illness.

For some, ICD-11 should shadow DSM-5 and there are similarities. Sub-types and “first rank symptoms” are removed, Bipolar II is recognized and elevated mood/overactivity in the context of anti-depressants can be diagnosed as mania/hypomania. There are different descriptions of mixed states and schizoaffective disorder. The introduction of a complex Post-traumatic stress disorder (PTSD) with disturbance in the sense of self in relationships and in affect regulation reflects our better understanding of the effect of early trauma. A dimensional view of Personality Disorders (or at least to some acknowledgement of the difficulty of drawing a diagnostic threshold) might be possible.

The NIMH research classification system, the Research Domain Criteria (RDoC) seeks to generate classification from basic behavioural neuroscience (Cuthbert and Insel, 2013). It begins with understandings of behaviour-brain relationships and links them to clinical phenomena – rather than starting with the illness and looking at the underlying processes. It is not directly applicable to clinical work. It may be we need groupings of clinical phenotypes to be abandoned, but for clinicians the promises of neurosciences are yet to be delivered to the bedside.

Clinicians want to be scientific but want the complexity of clinical experience reflected in systems of taxonomy that also guide research and thus treatment (Reed et al., 2013). We needed more reliable diagnoses, more standardised terms. We need to be able to give patients terms that acknowledge illness and recognize disability (Malhi, 2013). But I miss cycloid psychoses and neurosis and Jaspers. I miss having other frameworks in which to think and different language to capture ideas. I too miss discussions about phenomena and how they might fit into different taxonomies (Andreasen, 2007). Classifications have to be accessible and useful. DSM-5 is seductively transparent and will retain its local influence: ICD-11 needs to be explored and considered as an alternative to DSM in everyday practice.