New national collaborative will mine health data in search of gold
Instead of waiting for Washington, D.C., Dartmouth-Hitchcock and five other institutions are turning to each other to solve the cost and quality problems plaguing U.S. health care. The Collaborative, as the group of six institutions is being called, will pool data about outcomes, quality, and costs for the most common and expensive conditions and treatments.
Path: "If we know that the treatment path for diabetes at one institution results in better clinical outcomes, higher patient satisfaction, and lower overall costs," says Dr. Robert Nesse, CEO of the Mayo Clinic Health System, a Collaborative member, "then there is knowledge to be shared and replicated in other institutions."
In addition to Dartmouth-Hitchcock and Mayo, the members are the Cleveland Clinic, Denver Health, the Geisinger Health System, and Intermountain Healthcare. Together, they serve over 10 million people.
Initially, the group will focus on eight common conditions and treatments for which costs are ballooning and dramatic variations exist in quality and outcomes. The eight are asthma, depression, diabetes, heart failure, labor and delivery, spine surgery, weight-loss surgery, and knee replacements, which collectively represent hundreds of billions of dollars in direct medical costs each year.
Cost: The group is tackling knee replacements first. Over 500,000 are performed each year in the U.S., at a cost ranging from $16,000 to $24,000 per operation. Collaborative members have already begun pooling knee replacement data and expect to have identified best practices by mid-2011. Pilot testing of those practices will begin immediately afterward.
The fact that nothing like the Collaborative has ever been tried nationally is in part due to the logistical challenges of collecting data from so many different institutions. First, there are patient privacy and data security hurdles to be cleared. Then there is the fact that not all places collect the same data in the same way, and not all places are willing to have their practices scrutinized by their peers.
"While the lowest common denominator—billing data—is fairly standard across sites, the required clinical and patient-reported data is not at all standard," says Dr. James Weinstein, copresident of Dartmouth-Hitchcock, who's spearheading the Collaborative. Weinstein also directs the Dartmouth Institute for Health Policy and Clinical Practice (TDI), where the data will be housed and analyzed.
Over 30 additional institutions have expressed interest in joining.
"In the short term," he explains, "the data will be incomplete and accommodations in the statistical analysis will need to be made." But over time, Collaborative members will update their data collection systems so outcomes, quality, and costs can be easily compared.
Major: Such improvements will require the institutions to expend considerable time and money, and, so far, no outside funding for the Collaborative has been obtained. "It is a major commitment," says Weinstein, "especially in the uncertain and costly health-care environment that exists today."
Still, over 30 additional institutions have expressed interest in joining the Collaborative. But the group plans to add no more than 10 this year, says Weinstein, to keep the project "scalable."
While membership in the Collaborative may be limited, its findings will be disseminated broadly and rapidly. "If we can find the best knee replacement pathway," says Weinstein, "why shouldn't everyone have it?"
Weinstein and the leaders of Mayo, Intermountain, and Geisinger began working toward the Collaborative about three years ago. They wanted to find a way to "band together" around common goals and "act proactively" to solve the biggest challenges in U.S. health care, says Weinstein.
Solve: "My belief is that the federal government will not solve health care's problems," he recently told a group of DMS fourth-year students.
Whether the Collaborative will or not, one thing is clear: it won't be for lack of trying.
If you'd like to offer feedback about this article, we'd welcome getting your comments at DartMed@Dartmouth.edu.
This article may not be reproduced or reposted without permission. To inquire about permission, contact DartMed@Dartmouth.edu.