Author + information
- Spencer B. King III, MD, MACC, Editor-in-Chief, JACC: Cardiovascular Interventions∗ ()
- ↵∗Address correspondence to:
Dr. Spencer B. King III, Saint Joseph’s Heart and Vascular Institute, 5665 Peachtree Dunwoody Road NE, Atlanta, Georgia 30342.
The interventional cardiology course that we began after Andreas Gruentzig joined us 35 years ago continues to evolve. Initially this was the primary exposure for invasive cardiologists desiring to take up the new method of coronary angioplasty. Later, the course evolved to a much more intimate and highly popular course led by John Douglas and focused on former fellows and early practitioners in the near region. For the last couple of years, the course has expanded with 15 academic institutions across the south collaborating to stage the EPIC-SEC (Emory Practical Interventional Course—Southeastern Consortium). The name may be awkward but the collaboration between dynamic young investigators and faculty from institutions from Virginia to Texas is impressive. Habib Samady and the other organizers should be congratulated. Among the many excellent presentations on best practice methods and cutting-edge technologies, there was a discussion of different clinical research methods. As I listened to Gregg Stone present the value and necessity of randomized controlled trials following by James Tcheng’s dissertation of the potential for big data coming from all kinds of observational sources, I sensed some strains of competition between these types of scientific inquiry. It seems inevitable that the massive amounts of data that are available to drive decisions in commerce will be put to use to help understand disease processes and to influence treatment decisions. Big data is coming. It seems equally obvious that regardless of the amount of data we have, that without randomized trials we can never rule out unidentified confounders. What is driving the interest in large observational data sets to answer questions that may in the past have been approached with prospective randomized trials? Although the ability to try to obtain answers quickly is a driver, the massive costs of mega randomized multicenter trials remains a dominant reason to look to other methods of answering important questions.
In discussing these approaches to clinical research with the course participants, it occurred to me that these methods should not be competitive approaches but rather complementary. With randomized controlled trials, we often find ourselves dissecting various subsets in whom we feel the treatment was helpful and others in the same trial that were not benefited. Subset analyses, especially of neutral trials, are fraught with error. How could research methods become more effective? Could the information coming from “big data” sources identify populations likely to benefit from diagnostic or therapeutic interventions? Armed with such information, could necessary randomized controlled trials be targeted to the population of interest, thereby increasing the chance of obtaining reliable answers to the questions? These more targeted trials, some wonder, might not be attractive to industry sponsors who are looking for “blockbuster” therapies with wide application. Balancing this attitude is what I heard from an industry representative who said that they were tired of so many negative trials of things that may have value if applied to the appropriate population and not diluted by sampling a very heterogeneous population. Of course the results of targeted trials will not be generalizable, but for things that are appropriate for a specific condition or patient group, personalization, not generalizability, is what is needed. A consortium of regional institutions who have trained thousands of cardiologists, such as this Southeastern Consortium group of academic centers, has the potential to identify targeted questions and generate hypotheses that can be tested in appropriately sized trials. The protocols designed to create such studies could be informed by observational data sources such as the NCDR (National Cardiovascular Data Registry) and many others including administrative data. I was very impressed by the collaborative spirit shown by these dynamic investigators and leaders in interventional cardiology coming from institutions with highly competitive football programs but cooperative attitudes toward their educational and research opportunities. I hope this and other interinstitutional efforts will be able to attract support beyond their individual institutions’ abilities to do so. Groups such as this one have the power to produce research that is not only effective but cost-effective.
An obvious concern regarding collaborative research is how resources are distributed and how participation is recognized. All participants must have ownership, but, as a major benefactor to Emory University, Mr. Robert Woodruff, often said, “There is no limit to what can be accomplished if you don’t care who gets the credit.”
- American College of Cardiology Foundation