Medicine Is an Information Science: We Must Grow with the Pace of Our Data

By Harlan Krumholz,MD

Medicine is an information science. The volume, velocity, and variety of data and information that suffuses the medical profession are unprecedented and growing at a rapid pace. The medical landscape is changing from one in which medical knowledge is derived from laboratories and insights from the basic sciences to one that is supplemented by databases and insights from the experiences of patients and populations. In this environment, it is impossible for an individual to keep current with the ever-expanding knowledge base of medicine.

These changes have implications for the way we train doctors. The best doctors in the future will be able to manage streams of information, know how to access relevant information, be able to judge the validity of various sources of knowledge, and have the skills to communicate about such data and knowledge with patients. Training in how to manage this type of wide-scale data and use it effectively for each patient will be an essential skill.

Practicing and research physicians will work together, without boundaries, in this new framework of data sharing. Physicians and their interactions with patients will generate the data that will be the substrate for knowledge generation. . As such, practicing physicians and their patients will be an integral part of the knowledge generation enterprise, contributing data and questions in collaboration with data experts.

The clinical research enterprise will evolve as the utility of inductive reasoning emerges as a complement to hypothesis tests and deductive reasoning. The patient experience will be instrumental in revealing insights that can be applied at the bedside. Moreover, the integration of novel analytic strategies, already common in other industries, will potentiate our ability to personalize estimates, detect signal in the noise, and appreciate previously obscure patterns. As we refine our ability to understand the strengths and weaknesses of these evolving tools, practicing and research physicians will be better able to work together in generating answers to key questions.

The era of information will also be accompanied by an appreciation of the importance of open science. As we learn to see data as a social good, essential to making valid inferences about what works for whom in medicine, and move away from the current environment in which clinical scientists sequester their data and too often do not even report the results of their experiments, let alone publish them, there will be an increasing shift toward data sharing. We will collaboratively solve the challenges inherent in data sharing, which include ensuring that the identity of the subjects is protected, that the academic progenitors of the data receive appropriate credit, and that there is funding for the sharing.

Progress on this front has been substantial. Our group, the Yale University Open Data Access Project, and others are collaborating with industry partners to facilitate data sharing in a way that makes the data accessible for patients and physicians to navigate together. It is our goal to provide a means for rigorous and objective evaluation of clinical trial data to ensure that patients and physicians possess all necessary information about a drug or device when making treatment decisions. Part of the data sharing includes access to participant-level clinical research data and comprehensive reports of clinical research, such as full Clinical Study Reports (CSRs), with the aim of promoting scientific research that in order for us lead to improvements in individual and public health and health care delivery.

The Big Data revolution will not be so much about the size of data but rather, a change in the way the medical profession thinks about knowledge generation. In the next phase of medicine, we will learn from everyday experience and create opportunities for experiments that can be conducted within the bounds of clinical practice with efficiencies that are created by sophisticated information systems.

The next era, combining information from patients stemming from biology, social circumstances, and other contextual factors, will help us to inform care decisions in ways that have previously been beyond our grasp. Doctors who will be entering medicine during this period of exceptional change need to be prepared for understanding and applying the immense quantity of information that will be available to them. The next era can be another golden age of medicine – but only if we are ready to seize the moment.

KrumholzHarlan Krumholz is the Harold H. Hines Jr. Professor of Medicine (Cardiology) and Professor of Investigative Medicine and of Public Health (Health Policy). He additionally serves as Co-Director, Clinical Scholars Program; Director, Yale-New Haven Hospital Center for Outcomes Research and Evaluation. He can be reached at