Medical writer Dawn O’Shea wrote a compelling piece in last month’s Irish Medical Times calling out the psychiatry profession. Prof Brendan Kelly responds but says the human brain is difficult to measure and understand
Dawn O’Shea’s discussion of psychiatry in last month’s Irish Medical Times is compelling, valuable, and timely. She identifies critical issues and urges change. After exploring childhood traumas, psychological symptoms, PTSD, and encounters with psychiatry, O’Shea writes that her ‘clinical diagnosis is depression and anxiety disorder’, but, ‘for me, these are just symptoms of the real illness, which is a pervasive, debilitating and unrelenting self-loathing’.
O’Shea ‘tried everything – medication, ECT, and decades of counselling. Nothing has even made a dent in it’. She is ‘eternally grateful to psychiatry for saving my life, but at the same time, I feel the profession has let me down’.
Problems with psychiatric diagnoses emerge repeatedly in her account. I agree that this area needs to be explored, addressed, and improved.
There are no blood tests or brain scans to diagnose common conditions such as depression, bipolar disorder, PTSD, or schizophrenia. Diagnoses are based on clusters of symptoms that commonly occur together. Every few years, diagnostic categories are revised in new editions of the two classification systems: ICD in Europe and DSM in the US.
This is a highly unsatisfactory situation, but scientific knowledge of the brain has not (yet) advanced to a point that allows diagnosis based on biological tests. So, symptom-based categories are the best that anyone can do for now.
Dawn O’Shea.
Photo: Conor McCabe Photography.
O’Shea writes that ‘suicide behaviour disorder still hasn’t been accepted by the profession’, and ‘it’s unlikely that self-loathing will be classified as a clinical disorder in my lifetime’. O’Shea is correct on both counts: it seems improbable that either of these will be classified as a specific ‘disorder’ anytime soon. Each revision of DSM involves field-trials of the occurrence and reliability of revised diagnoses, so these conditions would need to be identified and characterised in such studies before they would be included.
The over-arching problem here is balancing the need for generality (i.e., identifying commonly occurring syndromes) with the need for specificity (i.e., everyone is different). Nobody fits perfectly into any diagnostic category, but there is still a need to identify common patterns of suffering across populations in a reliable way, if research is to advance and treatments can be tested.
The key lies in being circumspect about the usefulness of ‘diagnoses’ in the first place. Psychiatric diagnoses are reasonable guides for research and treatment at a general level, but some people fit into more than one category. Others fit into none and yet suffer deeply.
I am always interested when the media describe DSM as the ‘psychiatrists’ bible’, despite the fact that DSM itself warns sternly against ‘tick-box’ diagnosis. As O’Shea demonstrates, every individual is vastly more complicated than any list of symptoms.
Nevertheless, comparing DSM to a bible is not without merit: the majority of people engage with their ‘bible’ or scripture in a highly nuanced fashion, taking certain sections literally, interpreting other sections metaphorically, and completely ignoring vast swathes.
It is sensible and necessary to look at ICD and DSM in a similar fashion. Diagnostic systems are useful guides in very uncertain landscapes, but the person always comes first.
O’Shea adds that, ‘compared to other specialties, psychiatry has made very little progress in elucidating the pathophysiology of psychiatric illnesses and, consequently, driving the search for novel treatments’. Again, O’Shea and I find common ground here. Over past decades, neuroscience has contributed very little to our treatment of common mental illnesses such as depression, anxiety, and schizophrenia.
This is not due to any lack of effort by psychiatrists. The world is laden with neuroscientific studies by psychiatrists. Millions of dollars and decades of research have gone into psychiatric studies of brain imaging, genetics, inflammation, and other biological parameters.
The problem is that the yield has been minimal, at least in terms of treatments for common illnesses. There have been incremental advances in certain areas, such as dementia and intellectual disability, but, as O’Shea points out, ‘there’s a lot of ground to be made up’.
I share O’Shea’s concern and I discuss this in my new book, In Search of Madness: A Psychiatrist’s Travels Through the History of Mental Illness. The problem is that, too often, small sample sizes undermine the reliability of much-touted neuroscientific findings. Selective outcome reporting and selective analysis of some neuroimaging studies are particular issues, limiting the usefulness of research.
To compound matters, mindless forms of neuroscience commonly command a seductive appeal that they do not merit, especially when coloured images of ‘brain scans’ accompany press releases and media reports about under-powered studies. The allure of research technologies has blinded us to their limitations. Adding ‘neuro’ to the front of a word does not make it science.
What is the solution? Virtually everyone agrees that current diagnostic systems are imperfect. While ICD and DSM remain necessary to guide clinical care in a general sense, most psychiatric diagnoses do not reflect distinct biological entities and will, therefore, elude purely biological explanation. New approaches often focus on symptom dimensions rather than diagnoses, which might improve results in the future and might well incorporate suicide behaviour and self-loathing. Also, larger, better powered studies are needed, rather than more small ones.
In the meantime, we can do a great deal with treatments we already have. For problems of living, the best solutions are rooted in communities rather than psychiatric services. For more concerning issues, GPs require rapid access to low-threshold psychological programmes. For serious mental illness, specialist psychiatric services are needed, in conjunction with social interventions.
O’Shea is right: we need an improved neuroscientific basis for psychiatry. This is slow to come, chiefly because the human brain is almost infinitely intricate.
But, as O’Shea demonstrates, this complexity should inspire more focused research, rather than deterring us. The brain might be complex and formidable, but so are we. 
Information
Brendan Kelly is Professor of Psychiatry at Trinity College Dublin and author of In Search of Madness: A Psychiatrist’s Travels Through the History of Mental Illness (Gill Bools, 2022).
