Psychiatry Beyond the Current Paradigm, and DSM-5

Hugh Middleton, MD

Recently, two more waves of criticism have broken onto the beach of opinion concerning mental health services and practice. Allen Frances has mourned approval of DSM-5 in his Psychology Today blog (Frances 2012) and the British Journal of Psychiatry has published a paper by members of the UK Critical Psychiatry Network (Bracken, et al., 2012). What is notable about both of these is that they give further voice to criticism of conventional mental health services by those who have spent years providing and researching them.

Frances describes approval of DSM-5 by the American Psychiatric Association as “the saddest moment in my 45 year career of studying, practicing, and teaching psychiatry”. He asks clinicians, the press and the general public to “be skeptical and don’t follow DSM-5 blindly down a road likely to lead to massive over-diagnosis and harmful over-medication”. This is seminal given the fact that Allen Frances was chair of the Task Force that developed DSM-5’s predecessor. In recent years Frances has acknowledged many of the shortcomings and unintended harm that follow application of the descriptive, criterion-based definitions of “mental illnesses” that are DSM.

“Psychiatry Beyond the Current Paradigm” considers the conceptual and practical implications of accumulating real world mental health research findings. These are, broadly:

• There is no agreed and scientific understanding of how psychiatric medicines work … if and when they do.
• Claims of efficacy have to be tempered by respect for the idiosyncratic influences of expectancy and hope
• Repeated reviews stubbornly refuse to refute Rosenzweig’s 1936 Dodo Bird verdict: that it is common, non-specific features which make the greatest contribution to psychotherapy outcomes.

In other words psychopharmacology is not a therapeutic science, and a good psychotherapy outcome is primarily the result of a healing relationship.

Bracken et al emphasize, once again, that these now simple unassailable facts demand a paradigmatic change in the way mental health services are conceived and provided if practice is to be truly evidence based. Furthermore, they draw attention to a harsh reality. As mental health practitioners we are not equipped with tools that can be used instrumentally and in predictable and dependable ways, in the way a surgeon might wield their scalpel or the physician their fluids and antibiotics. We are not and cannot be tooled up specialists, with skills and facilities uniquely placed to “do something” helpful to our clientele. If we are to respect accumulated research findings then our role can be no more, or less, than that of providing support and advocacy as our clientele negotiate the complex web of difficulties and constraints that the confused, the despairing and the anxiety-provoking experience and can evoke.

That is not how mental health services are generally cast. Disturbing distress demands a response. Care is institutionalized, and therefore tends to be provided by paid professionals embedded in an organizational hierarchy and remunerated for providing a definable service. When residential facilities and office accommodation are necessary they have to be funded, and that generally means “provide a return on investment”. The historical social role fulfilled by mental health services is the provision of some form of institutional arrangement to provide for those who cause others distress, anxiety or embarrassment. The last century has seen medicine pick up that baton. We are beginning to find that medicine’s own standards of evidence based practice question the legitimacy of this, but understandably change is frustratingly slow and uncertain.

Allen Frances’ reflections upon DSM run into a similar quandary. DSM-5 has developed from DSM-IV, itself an extension of DSM-III which was the real game changer. During the 1970s a number of institutional forces combined to generate the lexicon of reliable but barely valid criterion-based descriptive definitions of “mental illnesses” that are the hallmark of DSM … whether it be III, IV or 5, and indeed Chapter V of ICD-9 & 10.

Despite the fact that forty years’ research and clinical experience have established that these “definitions” have little more than administrative value, they continue to exert considerable influence over psychiatric practice and policy. In his recent blog Frances describes how the (possibly premature) signing off on DSM-5 by the Board of Trustees of the American Psychiatric Association was driven by institutional pressures, in particular a reluctance to accept external review because of a “natural tendency of highly specialized experts to over value their pet ideas, to want to expand their own areas of research interest, and to be oblivious to the distortions that occur in translating DSM 5 to real life clinical practice”.

Bracken et al was drafted in response to an earlier contribution to this debate published in the same journal under the title “Wake-Up Call for British Psychiatry” and written by a number of British psychiatrists interested in promoting a bio-medical approach to mental health difficulties (Craddock et al 2008). The parallels are not difficult to see. Psychiatric diagnosis and classification is an institutionalized process that imposes administrative clarity upon an otherwise messy and uncertain field.

Bio-medical approaches and diagnostic schemes provide this process with a veneer of scientific respectability, but every now and again the inconvenient truths of real world human suffering and its vagaries break through. Allen Frances and many of the authors of Bracken et al are experienced practitioners and investigators who know this well. It is always good when truth pierces the gloom of prejudice and self-interest, but in this instance, how receptive can those interests be to evidence which threatens to undermine them?

Frances, A. (2012) accessed December 7th 2012.
Bracken, P., Thomas, P. et al (2012) Psychiatry Beyond the Current Paradigm. British Journal of Psychiatry. 201: 430-434.
Craddock, n., Artebi, D., et al (2008) Wake-up Call for British Psychiatry. British Journal of Psychiatry. 193: 6-9.


  1. I do want to push back a little bit about the Dodo Bird hypothesis. While I agree that the lack of validity of DSM diagnoses, the criteria used in diagnosis, and hence the difficulty of comparing treatment efficacy in helping people with particular diagnoses, different psychotherapeutic strategies do show different outcomes for particular kinds of diagnoses.

    A lot of the problems of of the meta-analyses used to supposedly confirm the Dodo Bird hypotheses is that they lump all sorts of outcome measures relating to all sorts of different kinds of outcomes on very different people. With that in mind, there is quite a lot of washout effect and the meaning of statistically significant differences become pretty meaningless as all sorts of measures get mixed together in unsystematic ways, really obscuring extent of effects of various interventions on various outcomes. It does not surprise me that such studies do not show meaningful differences between treatments, because the assessments themselves are not meaningful. Furthermore, these kinds of analyses do not then demonstrate that common factors are the key to psychotherapy (particularly for everyone with all issues).

    I would also add that some of lack of differences in efficacy in psychotherapies may be precisely because of the lack of validity of diagnoses in the first place, with a lot of variability in research participants that wash out effects. While this is often an argument now made by psychiatrists in regard to the very limited effects of antidepressants in related to the broadness MDD diagnosis, agreement with the Dodo Bird hypothesis based on the current studies might be actually be due to the poor validity of diagnoses to begin with.

    Mike Anestis at has written a lot about this.