Advertisement
Quantifying clinical relevance in treatments for psychiatric disorders Editorial| Volume 33, ISSUE 12, PB1-B2, December 01, 2011

Quantifying Clinical Relevance in Treatments for Psychiatric Disorders

      Imagine the following dialogue between a journal editor and a potential author:
      • Author: Mr. Journal Editor, I have a paper I want to publish. My result has a P value less than 0.05.
      • Editor: Tell me more. What is the effect size?
      • Author: It does not matter. My result has a P value less than 0.05.
      • Editor: Do you even know what an effect size is?
      • Author: It does not matter. My result has a P value less than 0.05.
      • Editor: An effect size tells us how clinically important the result may be. It is a measure of clinical significance. A statistically significant result can be clinically irrelevant if the effect size is small.
      • Author: It does not matter. My result has a P value less than 0.05.
      • Editor: Perhaps you have heard about number needed to treat.
      • Author: It does not matter. My result has a P value less than 0.05.
      • Editor: Oh my! Number needed to treat or NNT is a way of quantifying clinical significance. It can tell us how many patients need to be treated with one intervention instead of another before we can expect to encounter one additional outcome of interest. It is easy to calculate from dichotomous or binary outcomes. It is easy to understand. Clinicians like it.
      • Author: It does not matter. My result has a P value less than 0.05.
      • Editor: Oh my! Tell me, are the subjects in your study even remotely similar to the patients we see in clinical practice?
      • Author: It does not matter. My result has a P value less than 0.05.
      • Editor: Our journal seeks papers that will inform clinical decision making.
      • Author: I will take my paper to the journal of irreproducible results. It will get many citations. My result has a P value less than 0.05.
      This fictional conversation has been immortalized on the Internet
      • Citrome L.
      Does it work, will it work, and is it worth it? A call for papers (and reviewers) [video].
      and accompanied a call for papers that made a plea for reports that would help to answer the question of whether an intervention is actually worth using.
      • Citrome L.
      Does it work, will it work, and is it worth it? A call for papers (and reviewers) regarding effective treatments for psychiatric disorders.
      This is not a simple task. Over the years researchers and their students have been so tyrannized by P values that obtaining a P < 0.05 has become a goal in itself, all too often without taking a step back and asking the question of whether what is being accomplished is actually clinically important.
      • Citrome L.
      The tyranny of the P-value: effect size matters.
      For the practitioner who is applying data toward the treatment of an individual patient, assessing clinical relevance is a critical step in medical decision making.
      • Sackett D.L.
      • Rosenberg W.M.
      • Gray J.A.
      • et al.
      Evidence based medicine: what it is and what it isn't.
      This issue of Clinical Therapeutics contains 6 articles that expand on this theme, starting with a perspective from Jamie Karagianis,
      • Karagianis J.
      a clinician-researcher who has spent his professional career in a variety of different roles, most recently as an employee of a large pharmaceutical manufacturer. Dr. Karagianis shares with us his enthusiasm and mindset regarding the teaching and application of the philosophy and principles of evidence-based care. The publisher's perspective is then presented by Julia Ballard and colleagues from Wiley-Blackwell.
      • Ballard J.
      • Graf C.
      • Young C.
      Three articles review clinical relevance regarding the treatments for 3 different disease states, starting with schizophrenia by Christoph Correll et al,
      • Correll C.U.
      • Kishimoto T.
      • Nielsen J.
      • Kane J.M.
      followed by bipolar disorder by Shefali Srivastava and Terence Ketter
      • Srivastava S.
      • Ketter T.A.
      and major depressive disorder by S. Nassir Ghaemi and Paul Vöhringer.
      • Ghaemi S.N.
      • Vöhringer P.A.
      Themes include measurement-based care and the appropriate interpretation of treatment effects and the subsequent individualization of care. Trade-offs among different outcomes for these perplexing disorders can be understood by quantifying effect sizes using the clinically intuitive metrics of number needed to treat (NNT) and number needed to harm (NNH). Likelihood to be helped or harmed (LHH) is introduced within the context of examining the tension between benefit and risk. In the matter of treatments for major depressive disorder, close examination of effect sizes can also help ascertain the utility of antidepressants for different degrees of disease severity.
      • Ghaemi S.N.
      • Vöhringer P.A.
      The concluding report by Renrong Wu et al
      • Wu R.
      • Kemp D.E.
      • Sajatovic M.
      • et al.
      summarizes ways to accurately communicate benefits and harms to patients and payors.
      It is worth restating that external clinical evidence can inform but never replace individual clinical expertise. It is this expertise that makes possible the decision of whether the external evidence applies to the individual patient at all, and, if so, how it should be integrated into a clinical decision.
      • Sackett D.L.
      • Rosenberg W.M.
      • Gray J.A.
      • et al.
      Evidence based medicine: what it is and what it isn't.
      • Citrome L.
      Evidence-based medicine: it's not just about the evidence.
      It is hoped that this confluence of material about the quantification of clinical relevance will spur additional interest in the application of the philosophy and tools of evidence-based medicine among clinicians, researchers, authors, editors, publishers, and students.

      References

        • Citrome L.
        Does it work, will it work, and is it worth it?.
        (Accessed October 7, 2011)
        • Citrome L.
        Does it work, will it work, and is it worth it?.
        Int J Clin Pract. 2011; 65: 232-233
        • Citrome L.
        The tyranny of the P-value: effect size matters.
        Bull Clin Psychopharmacol. 2011; 21: 91-92
        • Sackett D.L.
        • Rosenberg W.M.
        • Gray J.A.
        • et al.
        Evidence based medicine: what it is and what it isn't.
        BMJ. 1996; 312: 71-72
        • Karagianis J.
        Clin Ther. 2011; 33: B3-B10
        • Ballard J.
        • Graf C.
        • Young C.
        Clin Ther. 2011; 33: B11-B15
        • Correll C.U.
        • Kishimoto T.
        • Nielsen J.
        • Kane J.M.
        Clin Ther. 2011; 33: B16-B39
        • Srivastava S.
        • Ketter T.A.
        Clin Ther. 2011; 33: B40-B48
        • Ghaemi S.N.
        • Vöhringer P.A.
        Clin Ther. 2011; 33: B49-B61
        • Wu R.
        • Kemp D.E.
        • Sajatovic M.
        • et al.
        Clin Ther. 2011; 33: B62-B76
        • Citrome L.
        Evidence-based medicine: it's not just about the evidence.
        Int J Clin Pract. 2011; 65: 634-635