Electronic Health Record Research in Critical Care: The End of the Randomized Controlled Trial?

  • S. Harris
  • N. MacCallum
  • D. Brealey
Part of the Annual Update in Intensive Care and Emergency Medicine book series (AUICEM)


We believe it is the duty of every hospital to establish a follow‐up system, so that as far as possible the result of every case will be available at all times for investigation by members of the staff, the trustees, or administration, or by other authorized investigators or statisticians (Ernest Amory Codman).

Codman was a surgeon from Boston, practicing at the beginning of the 20th century. He instituted the practice of morbidity and mortality conferences, and personally tracked all his patients using ‘End Result Cards’. Upon these he recorded demographics, diagnosis, treatment and outcomes. He argued that we should follow‐up all cases to improve the quality of care. We wish to argue here that we should capture all healthcare information to benefit the science of clinical medicine. This is indeed the promise of the electronic health record (EHR) – healthcare’s own ‘big data’1.

However, big data alone will not be a panacea. It has all the same fallibilities that we have...


  1. 1.
    Stigler SM (1986) The history of statistics. Harvard University Press, CambridgeGoogle Scholar
  2. 2.
    Galton F (1907) Vox populi. Nature 75:450–451CrossRefGoogle Scholar
  3. 3.
    Johnson AEW, Pollard TJ, Shen L et al (2016) MIMIC-III, a freely accessible critical care database. Sci Data 3:160035CrossRefGoogle Scholar
  4. 4.
    Powles J, Hodson H (2017) Google DeepMind and healthcare in an age of algorithms. Health Technol 29:1–17Google Scholar
  5. 5.
    Alaa AM, Yoon J, Hu S, van der Schaar M (2017) Personalized risk scoring for critical care prognosis using mixtures of Gaussian processes. Ieee Trans Biomed Eng 65:207–218CrossRefGoogle Scholar
  6. 6.
    Templ M, Kowarik A, Meindl B (2015) Statistical disclosure control for micro-data using the r package sdcMicro. J Stat Softw 67:1–36CrossRefGoogle Scholar
  7. 7.
    Guyatt GH, Oxman AD, Kunz R, Vist GE, Falck-Ytter Y, Schünemann HJ, Working Group (2008) What is “quality of evidence” and why is it important to clinicians? BMJ 336:995–998CrossRefGoogle Scholar
  8. 8.
    Vigen T. Spurious Correlations Project. Accessed 10 January 2018
  9. 9.
    Lewis EJ (2003) Ancient clinical trials. N Engl J Med 348:83–84CrossRefGoogle Scholar
  10. 10.
    Holland PW (1986) Statistics and causal inference. J Am Stat Assoc 81:945–960CrossRefGoogle Scholar
  11. 11.
    Angus DC (2015) Fusing randomized trials with big data: the key to self-learning health care systems? JAMA 314:767–768CrossRefGoogle Scholar
  12. 12.
    Fisher R (1935) The design of experiments. Macmillan, LondonGoogle Scholar
  13. 13.
    Baicker K, Taubman SL, Allen HL et al (2013) The Oregon experiment – effects of medicaid on clinical outcomes. N Engl J Med 368:1713–1722CrossRefGoogle Scholar
  14. 14.
    Harris S, Singer M, Rowan K, Sanderson C (2015) Delay to admission to critical care and mortality among deteriorating ward patients in UK hospitals: a multicentre, prospective, observational cohort study. Lancet 385:S40CrossRefGoogle Scholar
  15. 15.
    Hernan MA, Robins JM (2006) Instruments for causal inference: an epidemiologist’s dream? Epidemiology 17:360–372CrossRefGoogle Scholar
  16. 16.
    Rassen JA, Schneeweiss S, Glynn RJ, Mittleman MA, Brookhart MA (2009) Instrumental variable analysis for estimation of treatment effects with dichotomous outcomes. Am J Epidemiol 169:273–284CrossRefGoogle Scholar
  17. 17.
    Neuman MD, Rosenbaum PR, Ludwig JM, Zubizarreta JR, Silber JH (2014) Anesthesia technique, mortality, and length of stay after hip fracture surgery. JAMA 311:2508–2510CrossRefGoogle Scholar
  18. 18.
    Geneletti S, O’Keeffe AG, Sharples LD, Richardson S, Baio G (2014) Bayesian regression discontinuity designs: incorporating clinical knowledge in the causal analysis of primary care data. Stat Med 34:2334–2352CrossRefGoogle Scholar
  19. 19.
    Pearl J (1988) Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. Morgan Kaufmann Publishers, San FranciscoGoogle Scholar
  20. 20.
    Greenland S, Brumback B (2002) An overview of relations among causal modelling methods. Int J Epidemiol 31:1030–1037CrossRefGoogle Scholar
  21. 21.
    Rumsfeld DH (2002) DoD news briefing – secretary Rumsfeld and gen. Myers. Accessed 4 January 2018Google Scholar
  22. 22.
    Riedemann NC, Guo RF, Ward PA (2003) The enigma of sepsis. J Clin Invest 112:460–467CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2018

Authors and Affiliations

  1. 1.Critical Care DepartmentUniversity College London Hospitals NHS Foundation TrustLondonUK

Personalised recommendations