Effective Patient Care in a Data-Driven World

Register or Login to View PDF Permissions
Permissions× For commercial reprint enquiries please contact Springer Healthcare:

For permissions and non-commercial reprint enquiries, please visit to start a request.

For author reprints, please email
Average (ratings)
No ratings
Your rating



Correspondence Details:Andrew Grace, Department of Biochemistry, University of Cambridge, Tennis Court Road, Cambridge CB2 1QW, United Kingdom, E:

Copyright Statement:

The copyright in this work belongs to Radcliffe Medical Media. Only articles clearly marked with the CC BY-NC logo are published with the Creative Commons by Attribution Licence. The CC BY-NC option was not available for Radcliffe journals before 1 January 2019. Articles marked ‘Open Access’ but not marked ‘CC BY-NC’ are made freely accessible at the time of publication but are subject to standard copyright law regarding reproduction and distribution. Permission is required for reuse of this content.

Andrew GraceOur society acquires and stores data at unprecedented rates. Making sense of it has become a subject of academic interest,1,2 it is widely discussed and has provided a focus of interest to professional advisors,3 in each case with the objective of considering practical applications. Sophisticated interpretation of these data can provide powerful insights. Just recently, in the field of public economics, a massive scale examination that combined data from cross sections of the US Census and Current Population Survey and de-identified tax records categorically identified declining absolute income mobility, a finding with substantial policy implications.4 Using such large-scale (‘big’) data though is not by any means straightforward and requires special skills far removed from simple maths.2 Uninformed analysis can have catastrophic consequences; it was significantly implicated in the financial crisis of 10 years ago.5 Essentially, no matter how much data is out there, it needs diligent collection, informed analysis and effective implementation to provide positive impact.1,2

The application of data analytics to healthcare has gained traction from the sheer scale of the recent unanticipated demands on clinical services,6, 7 not least in electrophysiology. In both biology and medicine data is being generated across a broader range of parameters derived from ever-increasing numbers of model systems and patients.6–8 One driving hope is that data science will allow sense to prevail with an increased efficiency along the biomedical continuum, enhancing discovery, development and implementation and assessment of treatment responses.7, 9–11

Electrophysiology remains a nascent sub-speciality, based always in quantitation (not least through data available from wearable and implanted devices), populated by a workforce with curious minds, some especially well grounded in mathematics and physical sciences so we have a starting advantage. And there are many tractable questions that are unanswered, particularly in predicting risk, much of it determined by a complex genetic architecture.12,13 One informal observation is that we tend as group members towards independence and are not always immediately willing to follow the guidance of others. This practical complexity has the potential to constrain community-wide efforts to achieve key goals of improving care while containing costs. We therefore have to put some aspects of our natural tendency to one side and engage closely in gathering data, informing each stage of analysis and facilitating integration so ensuring this is a physician-led enterprise.

The substantial underestimates in the lead times to implementation of the findings from recent biomedical innovations, such as gene therapy, genomics and stem cells,14 has influenced recent commentators to offer appropriately reserved opinions regarding the speed of integration into practice of data science.8,15,7 In aggregate they also provide some principles to help physicians work through how they as individuals might best contribute: we need to articulate a clear sense of purpose regarding data gathering and analytics;6 we need to inform our teams this is going to take time and that approaches will evolve; we need to help refine electronic health records, that many of us now use daily, to reinforce the structural underpinnings of data collection (and no, we will not be going back to paper records);16 we should continue to refine our clinical skills as accurate taxonomy is our start point and disease misclassification reduces statistical power and impedes research; we also need to continue to be enthusiastic contributors to clinical trial recruitment, even though the work needed to maintain confidentiality and security with enhanced data de-identification has increased.

It is likely that to be effective all of us will need specialist instruction in aspects of data science17 and the integration of training modules into fellowship programs, as has also been argued for in genomics.6,18 More generally, our professional societies have been playing an important part including working internationally. ‘BigData@heart’, the most recent public–private and European Society of Cardiology-backed consortium program, will, over 5 years, have among key objectives further standardising disease definitions and outcomes, providing reliable sub-phenotyping and documenting the burden of disease across time and geographies.19

The editors of the New England Journal of Medicine during its 200th anniversary celebrations articulated a future vision of medicine that employed clinical support algorithms derived entirely from data.11 A major concern among electrophysiologists (and of course, physicians in general) is that such machine learning/structured algorithms will come to dominate and erode individual decision-making and maybe through default undermine relationships with patients.20 The key, as the authors explicitly stated,11 is that the data supports critical clinical choices.13 Physicians will still have to make decisions in close discussion with patients based on limited certainties regarding outcomes.

We are in an era of momentous change in medicine but have no doubt that big data-based evidence will be central to much that we will do, whether that be a translational interpretation of biology or determining financial resource allocation. Arrhythmia and Electrophysiology Review will endeavour to keep our readers at the vanguard of developments.


  1. The Alan Turing Institute. 2017. Available at: (accesses 16 August 2017)
  2. Spiegelhalter DJ. Statistics. The future lies in uncertainty. Science 2014;345:264–265.
    Crossref | PubMed
  3. Mayhew H, Saleh T, Williams S. Making data analytics work for you. McKinsey Quaterly 2016. Available at: business-functions/digital-mckinsey/our-insights/makingdata- analytics-work-for-you-instead-of-the-other-way-around (accessed 16 August 2016).
  4. Chetty R, Grusky D, Hell M, Hendren N, Manduca R, Narang J. The fading american dream: Trends in absolute income mobility since 1940. Science 2017;356:398–406.
    Crossref | PubMed
  5. Broughton PD. Big data hasn’t changed everything. Wall Street Journal 2 July 2013.
  6. Kim J. Big data, health informatics, and the future of cardiovascular medicine. J Am Coll Cardiol 2017;69:899–901.
    Crossref | PubMed
  7. Krumholz HM. The promise of big data: Opportunities and challenges. Circ Cardiovasc Qual Outcomes 2016;9:616–617.
    Crossref | PubMed
  8. Groeneveld PW, Rumsfeld JS. Can big data fulfill its promise? Circ Cardiovasc Qual Outcomes 2016;9:679–682.
    Crossref | PubMed
  9. Savage N. The measure of a man. Cell. 2017;169:1159–1161.
    Crossref | PubMed
  10. Topol E. The smart-medicine solution to the health-care crisis. Wall Street Journal 7 July 2017.
  11. Kohane IS, Drazen JM, Campion EW. A glimpse of the next 100 years in medicine. N Engl J Med 2012;367:2538–2539.
    Crossref | PubMed
  12. Grace AA, Roden DM. Systems biology and cardiac arrhythmias. Lancet 2012;380:1498–1508.
    Crossref | PubMed
  13. Yeh RW, Kramer DB. Decision tools to improve personalized care in cardiovascular disease: Moving the art of medicine toward science. Circulation 2017;135:1097–1100.
    Crossref | PubMed
  14. Joyner MJ, Paneth N, Ioannidis JP. What happens when underperforming big ideas in research become entrenched? JAMA 2016;316:1355–1356.
    Crossref | PubMed
  15. Groeneveld PW. Moving beyond big data to causal inference and clinical implementation. J Am Coll Cardiol 2017;69:901–902.
  16. Rudin RS, Bates DW, MacRae C. Accelerating innovation in health IT. N Engl J Med 2016;375:815–817.
    Crossref | PubMed
  17. Munevar S. Unlocking big data for better health. Nature Biotechnology 2017;35:684–686.
    Crossref | PubMed
  18. Aday AW, MacRae CA. Genomic medicine in cardiovascular fellowship training. Circulation 2017;136:345–346.
    Crossref | PubMed
  19. IMI BigData@Heart. 2017. Available at: www.Bigdata-heart.Eu (accessed 16 August 2017).
  20. Sniderman AD, D’Agostino RB, Sr., Pencina MJ. The role of physicians in the era of predictive analytics. JAMA 2015;314:25–26.
    Crossref | PubMed