SAN DIEGO—“Big data” can change healthcare into a system that better focuses on early detection and prevention rather than acute care—if physicians make better use of it.
“Big data” refers to the massive amounts of information that can be interpreted by analytics to provide an overview of trends or patterns. Its impact on the future of cardiovascular medicine was the subject of a March 15 American College of Cardiology scientific session.
“Data is one of the big trends in health information technology,” said moderator James Tcheng, MD, of the Duke Clinical Research Institute in Durham, N.C. “But I would suggest that we’re really falling short of meaningful use of health information technology. We’re focusing on the wrong place.”
Physicians apply it as an administrative tool rather than figuring out how to use the data that flows through their systems to better care for individual patients, he said.
“Big data is squishy. It really is hard to put your hands around, but leveraging the appropriate data streams has the potential to change the paradigm of healthcare, by taking big data and focusing it down on the patient,” Tcheng said.
L. Kristin Newby, MD, of Duke University School of Medicine, illustrated with the BASELINE and other studies how integrating biological, clinical and environmental information can help researchers and clinicians develop predictive models to understand disease-state transitions and outcomes. On a smaller level, the data provide a way to multidimensionally characterize individuals that can yield testable hypotheses.
Big data should also provide physicians and healthcare systems with continuously updated estimates of individual risk and health of neighborhoods and populations they serve, enabling directed education, prevention and treatment programs targeted to the individuals and local neighborhoods, she said.
The variability and variety of big data lead to more complexity, difficulty and resource consumption in trying to manage that information, Tcheng said. In the era of big data, cardiologists have to change how they collect data. Now, it’s about aggregating many different data streams.
Tcheng pointed to a recent study from the University of Pennsylvania to show one way big data can be useful. Researchers found Twitter can be a better predictor of coronary heart disease rates than factors such as smoking, diabetes, income, education and obesity—combined.
The problem, said H. Vernon Anderson, MD, of the University of Texas Health Science Center at Houston, is that healthcare data is unstructured. Physicians need data standardization, an agreed-upon set of rules that allow information to be shared and processed in a uniform and consistent manner. It’s up to physicians to lead in the creation of that system.
And many clinicians are wary of big data because of the noise it brings, Anderson said. They know there’s a signal in the noise, but it’s hard to decipher.
It’s the noise, though, that allows a clinician to learn about the populations he or she serves by showing a big picture of the population, Anderson said.
More importantly, physicians need to take that general information and apply it to individual patients to shift the healthcare paradigm, said Jeroen Tas, CEO of Philips Healthcare. In a country that spends $3 trillion on healthcare, 83 percent goes to chronic disease, he said. Yet the system is organized around acute care. The information provided by big data can change that, allowing healthcare to be organized around a health continuum and the needs of each patient.
“I think if we do this, we cannot just marginally improve healthcare, we can truly exponentially improve healthcare,” Tas said.