'Network of Networks' - a step towards evidence based medicine

President Bush will propose a new initiative to develop a "network of networks" when he announces his fiscal year 2008 budget proposal on February 5. That was the most significant news coming out of a briefing put on in Washington last Friday by Health Affairs to coincide with publication of a special issue titled "A Rapid-Learning System."

The special issue is meant to focus attention on how a national system that knits together millions of electronic health records (EHRs) can provide almost instantaneous answers on the effectiveness of various aspects of medical care and medications. Such feedback is now available only after completion of a clinical trial, which can take 10 years or more. Moreover, the results of those clinical trials are not particularly relevant to most patients or their physicians because of the narrow selection criteria used.

The difference between clinical trial patients who are generally younger and have just one target condition and, say, poorer, multi-condition Medicare and Medicaid patients leads to an "inference gap" that forces physicians to surmise or guess about how to treat their patients. "We are straddling the inference gap," said Paul Wallace, M.D., medical director of health and productivity management programs at Kaiser Permanente, "and it is getting wider."

More real-time, realistic medical information is starting to be generated by isolated networks such as the Cancer Research Network funded by the National Cancer Institute and the Vaccine Safety Datalink funded by the Centers for Disease Control and Prevention. In the private sector, Kaiser Permanente has taken the lead, and now has EHRs for half of its 8.5 million patients, according to Wallace. Kaiser is focusing on an "early proof of concept" that makes use of its "Health Connect" system to better target care for its 40,000 patients who are diagnosed with cancer each year, said Wallace.

But Lynn Etheredge, consultant to the Health Insurance Reform Project at George Washington University, pointed to the 85 million enrollees in Medicare and Medicaid as a critical population "where we most need an evidence base" because those individuals do not resemble, for the most part, the population of participants in clinical trials, on whose results much of the nation's medical practice is based. But Etheredge, who was the driving force behind publication of the special issue, called rapid learning "a work in progress."

Etheredge pointed to the "network of networks" initiative that will be in Bush's 2008 budget proposal as evidence that good things are happening. Carolyn Clancy, M.D., director of the Agency for Healthcare Research and Quality (AHRC), disclosed the initiative, which she did not describe except to say that it would allow the AHRC to combine unspecified databases.

When contacted after the press conference, Karen Migdail, a spokeswoman for the AHRC, which will have responsibility for this "network of networks, explained it would spring from the AHRC's Effective Health Care Program announced in September 2005. That program has spawned a network of 13 Developing Evidence to Inform Decisions about Effectiveness research centers (referred to as DEcIDE) which have been working to fill knowledge gaps about treatment effectiveness through use of de-identified medical data available through insurers, health plans and other organizations.

Article from Health-IT World