Cardiologists Weigh in on Big Data’s Role in Health Care

Big data analytics has emerged as a trend across nearly every industry – from transportation to education and health care – and experts around the country and across the world are seeking ways to leverage insights gleaned from these large data sets to advance their fields. While there are unlimited opportunities for big data analytics to transform health care, there are also many implications to consider and serious complexities to navigate.

A recent Special Issue of Circulation: Cardiovascular Quality and Outcomes focuses on big data research methods and their impact on cardiovascular care. The issue includes several studies that used big data analysis and features perspective pieces from several cardiologists on this hot topic.

“Be assured, research and clinical care are about to join the digital, mobile, mathematical, personalized revolution. We all need to ensure that the changes produce progress for people and society,” said Harlan M. Krumholz, MD, SM, FACC, in the Editor’s Perspective. Krumholz offered insights and recommendations on training, application, replication, dissemination, interoperability, funding and collaboration, all of which must be addressed in order to successfully transition to a big data analytics strategy.

This era of digital health care opens doors for applying big data analytics to the emerging area of precision medicine; however, health care lags behind other industries in this space for a variety of reasons. There was consensus among the authors of the perspectives that numerous and significant changes must be made in order to evolve the research paradigm that has driven medicine for decades. Several authors stressed that big data analysis must be held to the same standards as traditional research methods and ensuring reproductibility was a common concern.

“The rise of big data analytics in health care settings presents an exciting opportunity to leverage the power of increasingly voluminous health care data in ways that were simply impossible as recently as 10 years ago,” said John S. Rumsfeld, MD, PhD, FACC, ACC’s Chief Innovation Officer, and Peter W. Groeneveld, MD, MS, in a Cardiovascular Perspective. “However, it is critical to recognize that the fundamental pitfalls of observational data analysis cannot be ignored, and in fact, the risks of such pitfalls demand rigorous scientific testing and novel methods for peer review of big data analytic models.”

In another Cardiovascular Perspective, Sanjeev P. Bhavnani, MD, and coauthors provide insight into implications of big data science for early career investigators. They stress that retraining and collaboration are key, and clinical and research teams must be restructured. The authors point to several examples of new research models that signal the future, such as PatientsLikeMe, the United Kingdom’s BioBank and Apple’s Research Kit. “Similar to the objectives of established data sources such as census and public health data sets, or standardized patient registries such as the National Cardiovascular Data Registry where data are structured and aggregated to monitor population trends, develop guideline-based care, and infer changes to health care policy, new citizen science and crowdsourcing initiatives aim to leverage public and patient participation to collect health data and vital statistics through new, massive, open, and online data repositories,” write the authors.

“One of the issues with big data is that if it comes from clinical databases, the accuracy and completeness of coding can be an issue,” commented Kim A. Eagle, MD, MACC, editor-in-chief of “Some processes for educating providers about coding have focused on maximizing revenue, not necessarily on creating a proper list of diagnoses according to severity and timeframe. Furthermore, the process of caring for patients in the EMR is further complicated by the time required for documentation. It has been estimated that a physician may spend an extra hour or two after a day-long clinic on this aspect of care, often without either adequate training or compensation. Needless to say the coded diagnoses may be far from optimal. These nuances of the interface between providers and the ‘big data’ they are creating presents the potential for both inaccuracies and also the drawing of relationships between codes and outcomes that may not be as precise as we may wish to think.”

Exploring and leveraging big data analytics is a core component of ACC’s Innovation Agenda and this was a hot topic of discussion during the College’s recent Cardiovascular Innovation Summit. Stay tuned to for updates on how the ACC is moving cardiovascular care forward through its innovation efforts. 

Keywords: Crowdsourcing, Public Health, Registries, Research

< Back to Listings