Evidence-Based Medicine and Psychiatry
An edited version of the following article has been selected for and is competing in JPMS International Medical Writing Contest 2012. You are welcome to visit it on the JPMS site, and read, share and comment, because that will contribute in part to the final result.
Evidence-Based Medicine and Psychiatry
M. Awais Aftab
Evidence-Based Medicine (EBM) is rapidly becoming the mantra of modern clinical practice. It is being taught in medical schools and discussed in conferences as the ‘gold standard’ of health care decisions. The paradigm is making its impact on psychiatry as well, and one can see practitioners and journals attempting to reframe their approaches in the light of EBM. While this methodology is well-argued in medicine and is showing great success, its application to psychiatry is on shakier grounds. I will argue that this is because of two factors: Firstly, the general and recognized limitations of EBM are more acutely felt when applied to psychiatry. Secondly, there are inherent nosological difficulties in psychiatry which make the results of research based on DSM diagnoses challenging to interpret for actual clinical practice.
Evidence-Based Medicine can be defined as the practice of applying the best available scientific evidence to clinical decision making. Triple-blind randomized placebo-controlled clinical trials and their meta-analyses are held in highest regard as scientific evidence. EBM is often allegedly seen as a ‘cookbook’ approach by critics due its insistence on scientifically proven treatment for defined diagnosis. However, this allegation that EBM recognizes only evidence from systematic research as the sole and exclusive criterion of clinical decisions is more of a straw-man derived from the manner of EBM hyper-enthusiasts, something that is counter to the actual spirit of EBM that its pioneers have espoused. Sackett et al insisted in their frequently cited 1996 paper that “The practice of evidence based medicine means integrating individual clinical expertise with the best available external clinical evidence from systematic research. By individual clinical expertise we mean the proficiency and judgment that individual clinicians acquire through clinical experience and clinical practice.” They rightfully foresaw that neither clinical expertise nor scientific evidence alone is enough, and warned that “Without clinical expertise, practice risks becoming tyrannised by evidence, for even excellent external evidence may be inapplicable to or inappropriate for an individual patient.” Ranga Krishnan voiced a similar view in an interview with Medscape, being a vocal proponent of applying EBM to psychiatry: “What we mean by EBM is often misconstrued as "cookbook" medicine; in other words, see somebody, do A, B, C, and then D. But that is not what EBM is. EBM is taking all the available data on a particular question, synthesizing it, reviewing it, putting it into the context of what it really means, and then taking it and applying it to the patient. It is really contextualizing available information in a systematic fashion.” EBM therefore argues for a synthetic approach in which clinical judgment and research evidence are balanced. This is all well and good. However, the balance that exists between clinical judgment and evidence in psychiatry differs significantly from the balance that exists in other medical specialties.
Let us first look at the general limitations of EBM that are applicable to all medical specialties but present with greater relevance in psychiatry.
1) Time lapse between study and publication
Articles submitted to scientific journals often have to undergo considerable waiting time before they see the light of publishing, which could be a year or even more. Additionally, the various committees which take these published studies into account and recommend treatment guidelines for various disorders take years to come up with a final document. In a rapidly developing field such as psychiatry, this means that any treatment guideline will be years behind the actual advancement of scientific knowledge.
Evidence-based guidelines are riddled with the problem of generalizing available evidence to populations at large or to treatment course and outcomes over a prolonged period of time. The extent of extrapolation will remain a troublesome question. This is all the more acute in psychiatry because individual factors affect illnesses and treatment response in ways that are poorly understood.
3) Publication bias
A strong bias exists in favor of publishing only trials with positive results. This means there will be a large amount of scientific evidence that has not been made accessible, and therefore cannot be taken into account by evidence-based medicine, and yet that unpublished data can be of significant impact. Turner et al investigated 74 studies registered with FDA on 12 antidepressant agents involving 12,564 patients. Their results were shocking: 31% of studies were not published, and this had to do whether the studies had a positive or a negative result. Only 1 study out of 37 with positive results was unpublished, but the studies whose results FDA saw as negative or questionable were either not published (22 studies) or published with a falsely positive outcome (11 studies) with only 3 exceptions. This led to a discrepancy, such that 94% of trials in published literature were positive, while 51% were positive in FDA analysis.
The problem with applying EBM to psychiatry goes deeper than these limitations. It has to do with the very manner in which diagnostic system in psychiatry functions. DSM classifies psychiatric diseases into discrete disorders based on clinical signs and symptoms. These disorders are basically ‘symptom-complexes’ and not specific illnesses whose pathophysiology is distinctly worked out. The diagnoses of DSM-IV are operational definitions, based on consultation and consensus to make sense of varying presentations, and to facilitate researchers in using a uniform set of criteria. This was acknowledged even by the authors of DSM-IV: “. . . there is no assumption that each category of mental disorder is a completely discrete entity with absolute boundaries dividing it from other mental disorders or from no mental disorder . . . [and] boundary cases will be difficult to diagnose in any but a probabilistic fashion.” This classification system is far from a classification system that is based on a proper understanding of cause and effect. We do not know the exact etiology of any psychiatric disorder. The integrity of this nosological method is challenged by several observations, such as the high rates of over-lap and comorbidity of psychiatric disorders, lack of homogeneity in treatment response and the fact that a broad range of different DSM disorders respond to the same medications. EBM employs these very DSM diagnoses in attempt to create specified algorithms for treatment, and the diagnostic issues at hand makes the whole venture uncertain at a fundamental level.
Research driven by EBM is focused on treatment efficacy for a particular diagnosis, but what is required in clinical psychiatric practice is the alleviation of specific symptoms. Furthermore, the diagnostic approach blinds us to particular features of a patient’s condition and circumstances, which are often playing a central role in patient’s psychopathology. Once this crucial aspect is recognized it is easy to see that the insistence that EBM should monopolize psychiatry can be very detrimental. Psychiatric practice remains as much art as it is science, and there should be no shame for psychiatrists to acknowledge that treatment rationales ought to be driven not just by diagnosis-based statistically driven protocols, but also by sensible and reasonable conjectures based on the knowledge of particulars of an individual patient.
1. Sackett DL, et al. Evidence based medicine: what it is and what it isn't. BMJ. 1996 Jan 13;312(7023):71-2. URL: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2349778/pdf/bmj00524-0009.pdf
2. Medscape. Evidence-Based Medicine in Psychiatry -- A New Perspective: An Expert Interview With Ranga Krishnan, MB, ChB. URL: http://www.medscape.org/viewarticle/475415
3. Turner EH, et al. Selective publication of antidepressant trials and its influence on apparent efficacy. N Engl J Med. 2008 Jan 17;358(3):252-60. URL:
4. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders DSM-IV. Introduction.