Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Perspective
  • Published:

The perilous path from publication to practice

Molecular Psychiatry volume 23, pages 24–25 (2018)Cite this article

Subjects

Precision medicine has long been heralded as the future of clinical practice, discussed at least since the 1940s by the likes of Paul Meehl1 and Claude Bernard.2 Fundamentally, precision medicine seeks to shift focus away from the average effectiveness of a given treatment to identifying the optimal treatment for an individual patient, through understanding features of the phenotype that inform the individual’s response to particular intervention.3 However, even once we have evidence that a particular feature or group of features might guide treatment, we remain a disappointingly long way away from realizing this potential. In this Perspective, we would like to draw the readers’ attention to three kinds of obstacles between publication and practice.

Precision medicine has become increasingly synonymous with an analytic approach called predictive modeling, although using these techniques is not a formal requirement. The most fundamental concern in predictive modeling is the nature of the data—its source, amount and type. When a model is derived or validated using a small sample size, consumers of clinical research must be aware that estimates of model performance are inflated.4 This scenario occurs most often when the source of the data is expensive: either costly to acquire (as in neuroimaging or gene expression data); or from difficult-to-recruit groups of patients such as individuals in early or prodromal disease phases. Of course, if it is prohibitively expensive to acquire data, this cost will eventually, become an additional burden on a patient in need of treatment. Other recent viewpoints have offered desperately needed perspective on this issue, advocating that researchers adopt a more pragmatic approach from the outset and interact with multiple stakeholders to maximize the chance that an innovation would be used if it were to be successfully developed.5, 6, 7 In parallel, researchers could consider experimental design approaches that might be optimized or enriched for predictive analytics. For example, questionnaire items or scanner tasks could be chosen from the literature such that the most discriminative/predictive are collected first, allowing the collection of a smaller amount of more useful data.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on Springer Link
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

References

  1. Meehl PE . Clinical Versus Statistical Prediction: A Theoretical Analysis And A Review Of The Evidence. University of Minnesota Press: Minneapolis, MN, USA; 1954, pp 83–128.

  2. Bernard C . Principes de médecine expérimentale 1947.

  3. Simon GE, Perlis RH . Personalized medicine for depression: can we match patients with treatments? Am J Psychiatry 2010; 167: 1445–1455.

    Article  Google Scholar 

  4. Schnack HG, Kahn RS . Detecting neuroimaging biomarkers for psychiatric disorders: sample size matters. Front Psychiatry 2016; 7: 50.

    Article  Google Scholar 

  5. Paulus MP, Huys QJM, Maia TV . A roadmap for the development of applied computational psychiatry. Biol Psychiatry Cogn Neurosci Neuroimaging 2016; 1: 386–392.

    Google Scholar 

  6. Paulus MP . Evidenced-based pragmatic psychiatry—a call to action. JAMA Psychiatry 2017.

  7. Chekroud AM . Bigger Data, Harder Questions — Opportunities Throughout Mental Health Care. JAMA Psychiatry 2017.

  8. Ginsberg J, Mohebbi MH, Patel RS, Brammer L, Smolinski MS, Brilliant L . Detecting influenza epidemics using search engine query data. Nature 2009; 457: 1012–1014.

    Article  CAS  Google Scholar 

  9. Lazer D, Kennedy R, King G, Vespignani A . The parable of Google flu: traps in big data analysis. Science 2014; 343: 1203–1205.

    Article  CAS  Google Scholar 

  10. Ioannidis JPA . Why most clinical research is not useful. PLoS Med 2016; 13: 1–10.

    Article  Google Scholar 

  11. Chekroud AM, Zotti RJ, Shehzad Z, Gueorguieva R, Johnson MK, Trivedi MH et al. Cross-trial prediction of treatment outcome in depression: a machine learning approach. Lancet Psychiatry 2016; 3: 243–50.

    Article  Google Scholar 

  12. Chekroud AM, Gueorguieva R, Krumholz HM, Trivedi MH, Krystal JH, McCarthy G . Reevaluating the efficacy and predictability of antidepressant treatments. JAMA Psychiatry 2017; 74: 370–378.

    Article  Google Scholar 

  13. Koutsouleris N, Kahn RS, Chekroud AM, Leucht S, Falkai P, Wobrock T et al. Multisite prediction of 4-week and 52-week treatment outcomes in patients with first-episode psychosis: a machine learning approach. Lancet Psychiatry 2016; 25: 25.

    Google Scholar 

  14. Vanhaudenhuyse A, Noirhomme Q, Tshibanda LJF, Bruno MA, Boveroux P, Schnakers C et al. Default network connectivity reflects the level of consciousness in non-communicative brain-damaged patients. Brain 2010; 133: 161–171.

    Article  Google Scholar 

  15. Torous J, Baker JT . Why Psychiatry Needs Data Science and Data Science Needs Psychiatry. JAMA Psychiatry 2016; 73: 3–4.

    Article  Google Scholar 

  16. Billings J, Dixon J, Mijanovich T, Wennberg D . Case finding for patients at risk of readmission to hospital: development of algorithm to identify high risk patients. BMJ 2006; 333: 327.

    Article  Google Scholar 

  17. Wang PS, Simon GE, Avorn J, Azocar F, Ludman EJ, McCulloch J et al. Telephone screening, outreach, and care management for depressed workers and impact on clinical and work productivity outcomes: a randomized controlled trial. JAMA 2007; 298: 1401–1411.

    Article  CAS  Google Scholar 

Download references

Acknowledgements

The authors thank Drs Monica Rosenberg, Martin Paulus and John Krystal for thoughtful comments and discussions.

Author information

Authors and Affiliations

  1. Department of Psychology, Yale University, New Haven, CT, USA

    A M Chekroud

  2. Spring Health, New York City, NY, USA

    A M Chekroud

  3. Department of Psychiatry and Psychotherapy, Ludwig-Maximilian-University, Munich, Germany

    N Koutsouleris

Authors
  1. A M Chekroud
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. N Koutsouleris
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to A M Chekroud.

Ethics declarations

Competing interests

AMC holds equity in Spring Care Inc., a behavioral health startup. He received salary from Spring Care Inc. and Yale University. He is lead inventor on two patent submissions relating to treatment for major depressive disorder ((a) USPTO docket number Y0087.70116US00, and (b) USPTO. Provisional Appl. No. 62/491,660). He has consulted for Fortress Biotech on antidepressant drug development. NK has received paid speakerships from Otsuka/Lundbeck, and holds a patent relating to psychosis risk prediction (USPTO Appl. No. 14/910588).

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Chekroud, A., Koutsouleris, N. The perilous path from publication to practice. Mol Psychiatry 23, 24–25 (2018). https://doi.org/10.1038/mp.2017.227

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/mp.2017.227

This article is cited by

Search

Advanced search

Quick links