An Example Of The Application Of Big Data From AHA 2016
Among a variety of personal and professional goals I’d set in coming to my first American Heart Association conference, I wanted to gain an appreciation for what’s “in the pipeline” that will impact how I care for patients with cardiovascular diseases.
So much of what we do in Medicine relies on the acquisition of good, reliable data. We rely on it and so do our patients. We have seen—and will continue to see—all kinds of changes in the types of data that become available. One area of huge potential is with regard to “big data,” which brings me back to this year’s AHA conference.
One of the themes of this year’s conference is “big data.” While I don’t see that a standard definition of big data exists, I would think that most people have a general understanding of what it is: when extremely large data sets are employed to reveal patterns or associations of disease. Among the biggest/newest contributions to big data includes that from smart devices and human genomics1.
The main purpose of big data is to provide the knowledge base that informs personalized medicine. Personalized medicine seeks to inform patient care by understanding an individual patient’s specific co-morbidities and genetics. To some degree, this replaces the standard model of relying on large population studies or clinical trials, when providers are asked to ensure that their patient’s characteristics “fit” that of the study population1.
Yesterday, Kelly Myers, on behalf of the Familial Hypercholesterolemia (FH) Foundation, presented data concerning an intriguing project. The database they maintain includes characteristics and parameters on a huge number of patients reporting any cardiovascular problem. From this pool of data, they’re able to assign a probability that a patient suffers from FH. The foundation then collaborates with physicians in key areas to identify and treat this illness, which can be quite difficult to recognize before patients experience negative outcomes that herald the disease.
I think this example does a lot to illustrate the potential of big data, even as we struggle as a community of physicians to understand how to interpret and—ultimately—use this information to better care for our patients.
Because of its timeliness, I will make one final point. It wouldn’t be a fair comparison to pair data from other fields to one within healthcare (it’s comparing apples versus oranges, right?), but I think the reminder of caution underlying the comparison remains the same. We just witnessed the inaccurate prediction of big data—regardless of one’s political leanings—on an enormous national stage: http://www.forbes.com/sites/johncarpenter1/2016/11/10/trump-win-shows-limits-of-big-data-power-of-emotional-intelligence/#2f74e2845539
Certainly, understanding big data and the role it may play for our patients needs to be more completely understood. There will be limitations to this type of data as well. For more information about the FH Foundation and/or FIND FH, please refer to: https://thefhfoundation.org/ http://www.cardiometabolichealth.org/fh-foundation-initiative-helps-find-fh.html
Mark Kaeppler, MD
Cardiology Fellow, Medical College of Wisconsin, Milwaukee, Wis.
Mark Kaeppler is a Cardiology Fellow at the Medical College of Wisconsin. He’s in the process of focusing his research interests. The opinions expressed are solely his own.
1.Empowering Personalized Medicine with Big Data and Semantic Web Technology: Promises, Challenges, and Use Cases. Panahiazar M, Taslimitehrani V, Jadhav A, Pathak J.