The media and some of the psychiatry profession have been critical of the recent DSM-5, questioning the practice of psychiatric diagnosis because the diagnostic categories are not based on objective laboratory tests.

Psychiatry is not the only medical field where diagnoses are not based on biological tests and indeed biomarkers or cognitive markers for psychiatric illnesses are not yet available in a clinical setting. In medicine, the patient’s history typically accounts for 75% of the diagnostic yield when evaluating common symptoms whereas physical examination only accounts for approximately 15% and diagnostic tests less than 10%. Psychiatry is also not the only disease area where there is controversy surrounding the appropriate threshold for diagnosis but he did point out that many mental disorders are actually on a continuum of normality.

One important query raised from the audience was that of whether DSM and ICD could be more integrated. Prof. Maj noted that improvements are being made in this regard in that steps are being taken towards more harmonisation of the DSM and ICD; for example they are sharing nomenclature and a metastructure.

Alternatives to the current classification systems were suggested, such as diagnostic prototypes, which are more user-friendly but are also associated with problems such as difficulty memorising the description of the prototypes and over- or under- diagnosis of disorders. Indeed prototype systems should only be used if they are applicable to clinical practice and have a clinical utility.

Another proposed criteria reform was that of the RdOC (Research Domain Criteria) project that uses criteria based on neuroscience and behavioural science and focuses on a variety of domains such as cognitive systems and positive valence systems. However, he acknowledged that one flaw of the system is that meaning, subjective experience and mental representations are downplayed. He reminded us of the importance of prioritising clinical utility, and as we know that is only partly related to pathophysiology.

Will the RdOC transform psychiatric diagnosis? It appears not, but it may help develop neurobiological measures which help in the subtyping of diagnostic entities rather than replacing them completely. It will also complement existing classification to improve prediction of outcome and treatment response.

It’s clear that diagnostic systems are not currently perfect but simply giving unconstructive criticism of the current systems will impede rather than improve the clinical state and will also harm the image of psychiatry. Indeed the priority must be causing minimal disruption to the field and ensuring that patients and their relatives are not confused by any new classification. Clinical developments in this area must result in a system that is of more clinical use, both to psychiatrists and patients. However, it is important that the stigma associated with new psychiatric diagnoses does not outweigh improved interventions and clinical outcomes.

I hope our coverage has given you some insights into the awareness, diagnosis and treatment of psychiatric disorders. It’s been a fascinating few days, showcasing some of the most progressive research in psychiatry and I’ve been highly impressed by the calibre of the speakers and topics.

Tuesday, and the final day of EPA was ushered in by the subject of memory. The beginning of today’s symposium, ‘Update on treatment options in Bipolar Disorders’, recalled the debate over pharmacogenetics in schizophrenia, reported in yesterday’s post. This time it was the turn of Professor Thomas G Schulze to offer an overview of this emerging field when applied to bipolar disorder.

While his tone of excitement mixed with pragmatism was similar to yesterday’s consensus, Professor Schulze had one exciting glimpse of the future to offer us. In response to the lack of robust evidence of pharmacogenetics in bipolar disorder, he showcased a new information-sharing initiative, ConLiGen. This network, created in 2008 to identify markers for lithium response, has collated data from a range of study groups across 18 countries, building a patient base of 3000 and allowing for powerful, post-hoc analyses.

ConLiGen is yet to yield any definite outcomes but seems like a cost-effective model to build upon. The development of pharmacogenetics is, says Professor Schulze, a process which will take time to deliver but it is a process which has definitely begun.

So my final day at EPA ends on an upbeat note of new insights shared and new collaborations highlighted. It seems clear that with bipolar disorder, as with many of the other mental disorders showcased here, the state of treatment is still in flux. I’ve seen few definitive endpoints offered, but that seems no bad thing as it leaves the conversation open and keeps psychiatric medicine moving forward.

That’s all from your correspondents at EPA. Next time we look forward to bringing you coverage from ISAD.