Boston-based Brigham and Women’s Hospital and its Center for Patient Safety Research and Practice, led by David Bates, are announcing today a collaboration with GNS Healthcare to use supercomputing technology to improve patient care. Cambridge, MA-based GNS Healthcare’s computer-simulation models will be used to predict the likelihood of adverse drug events and hospital readmission in patients with congestive heart failure.
“That’s really just a starting point,” says Tom Neyarapally, senior vice president of corporate development for GNS Healthcare, of the congestive heart failure application. “In the long run the idea is that we will be to do the same type of work across many diseases.”
It’s the first partnership GNS Healthcare has announced regarding work it’s done with healthcare payers, providers, or prescription benefit managers, says Neyarapally. In 2010 Cambridge-based Gene Network Sciences formed the GNS Healthcare subsidiary to focus on applying its computer models in the healthcare space—by helping doctors determine which drugs or combination of drugs will be most successful in patients, and conversely, which will lead to negative reactions. Gene Network Sciences now operates as a subsidiary of Via Science under the name GNS Healthcare.
In the Brigham partnership, GNS Healthcare will take data from the electronic health records of thousands of congestive heart failure patients admitted into the Partners HealthCare system. The aim is to track the relationships between such variables as diagnoses and combinations of drugs patients are taking, and then determine how they might cause an adverse reaction. The technology can also take into consideration the order in which certain diagnoses occurred or drugs were taken—factors that add a new level of complexity to predicting patient outcomes, Neyarapally says. Brigham can then use the hypotheses created to look at a given patient and determine how he or she matches up against certain variables tracked in the system, to better understand the likelihood of a negative drug reaction or hospital readmission.
“Something like this is not a research project; it gets deployed to be able to support decision making,” says Neyarapally. Doctors can then avoid administering the drugs likely to cause harm and find alternate treatments, or pay closer attention to patients in which the adverse events cannot be prevented—rather than treating a patient after the fact and backtracking to determine the cause of reactions.
“Experts can hone in on a small number of trillions of possibilities,” Neyarapally says.
GNS Healthcare, whose scientific advisory board is co-chaired by Xconomist Eric Schadt, is not revealing the financial terms surrounding the Brigham and Women’s partnership, Neyarapally says. When former Xconomy correspondent Ryan McBride profiled GNS Healthcare in April 2010, its CEO Colin Hill saw big revenue potential in the healthcare unit. He suggested it would be easier to track the success of the technology in reducing unsuccessful treatments and lowering healthcare costs for payers than it would be to trace the success of the computer models in drug development—the company’s original focus. Adverse events that could have been prevented are estimated to run the U.S. healthcare system $50 billion per year, GNS said in its announcement today.
Neyarapally says GNS is still toying with revenue models for the healthcare business, but options include sharing in the savings that result from deploying the technology, licensing the technology to groups who want to use them, or working with disease management firms who manage care on a per member, per month basis.