For immediate release:
Friday, January 26, 2007
9:30 a.m. EDT


Christopher Fleming

Researchers Call For Rapid Learning Through Electronic Health Records To Advance Medical Knowledge, Improve Care

Studies Show How Health IT Is Improving Care For Cancer, Diabetes

Bethesda, MD -- Data gathered in electronic health records on the experience of millions of patients have the potential to dramatically accelerate clinical research and provide the nation with timely, urgently needed knowledge about the value of new medical technologies, researchers report in a special edition of Health Affairs on “rapid learning” published today.

There are major gaps today in clinical research and evidence-based medicine. But in organizations such as the Veterans Health Administration (VHA), Kaiser Permanente, and the Geisinger Health System, clinical data captured in EHRs are already being used to answer practical questions about the safety, effectiveness, and costs of new treatments much faster and more efficiently than the traditional process of randomized clinical trials alone possibly could. Rapid learning through EHR databases and other health information technology (IT) tools could dramatically expand the health care system’s research capacity, making it possible to combine information from millions of patients each year to advance medical understanding of such diseases as cancer and diabetes, the studies say.

“New clinical products and applications are coming at an overwhelming rate. Neither regulatory nor market solutions have been effective in allowing us to make the best, most cost-effective use of these technologies,” said Lynn Etheredge, a consultant with the Health Insurance Reform Project at the George Washington University and the author of the lead paper in the Health Affairs special edition, titled “A Rapid-Learning Health System.” “Rapid-learning systems, which offer the capacity for real-time learning from the experiences of millions of patients in actual clinical settings, can help physicians decide when to use which clinical tool, and for which individuals.”

The wealth of information in EHRs can be used to enable rapid learning in various ways, David Eddy, the founder and medical director of Archimedes Inc., a large-scale physiology-based simulation model, explains in another article. For some questions, such as the percentage of diabetics with poorly controlled glucose, researchers might simply be able to “look up” the answer in a database. But for many important questions, such as the likely effect of treating this subgroup of diabetics with a new medication, computerized predictive models, based on information in the database, will greatly advance our ability to choose the best therapies, just as aircraft manufacturers construct models to avoid having to build prototypes of every possible design variation. These predictive models can also be used to create “virtual clinical trials” that will speed up the research and development process for new drugs.

Today, physicians often face a large “inferential gap” between clinical trials studies that use carefully selected patients under ideal conditions, and most real-world patients, who are typically older and have multiple chronic conditions, according to Walter Stewart of the Geisinger Health System. EHR databases will help fill in these knowledge gaps and make more personalized health care possible.

Rapid learning can help answer questions such as why costs are increasing, the risks and benefits of prescription drugs, and environmental effects on disease patterns. Rapid-learning capabilities may help reduce health disparities affecting minorities and special-needs groups, give patients the ability to make better treatment decisions, and allow physicians to practice “personalized medicine” by revealing the effects of genetic variations on responses to treatments. But to make rapid-learning initiatives successful, it will take a national investment, leadership from both the public and private sectors, and an increased focus on government research, Etheredge writes.

The Health Affairs special edition was supported by the Robert Wood Johnson Foundation’s Pioneer Portfolio, Kaiser Permanente, and the Agency for Healthcare Research and Quality. Strategies for advancing rapid learning in health care was the topic of a Health Affairs-sponsored conference in Washington, D.C., today that included an appearance by AHRQ Director Carolyn Clancy, as well as several authors from the January 26 issue. A webcast of the briefing will be available on Tuesday, January 30, at 12 p.m. EST at:

Federal Initiatives Can Support Rapid Learning, More Collaboration Needed

The federal government is in a unique position to use rapid learning to provide information on the quality of care and help fill major gaps in medical knowledge, says Sean Tunis, director of the Center for Medical Technology Policy in San Francisco, who has an article in the issue.

The Centers for Medicare and Medicaid Services (CMS) plans to integrate new drug claims data from Medicare Part D with medical data to improve understanding about the safety and effectiveness of medications. Similarly, AHRQ is exploring the comparative effectiveness of health care technologies in identifying gaps in medical research. The federal government is also linking payment for services to the collection of patients’ clinical, demographic, and economic data.

But for these efforts to succeed, stronger collaboration in the public and private sector, additional funding, and more clarity about the projects’ goals are required, Tunis and colleagues write. “The private sector can also contribute greatly on its own to rapid learning in health care; however, no private-sector entities have the size, visibility, and resources potentially available to the CMS and other federal health programs,” they say.

Health IT Holds Potential To Improve Care For Cancer, Diabetes

Another article showcases Kaiser Permanente’s experience using “KP HealthConnect” as a tool for rapid learning and cancer care. Kaiser’s EHR system makes it possible to generate detailed information on all Kaiser cancer patients, enabling researchers and clinicians to better monitor care and expanding options for detection, diagnosis, and treatment.

Although widespread interoperability among IT systems in health care does not yet exist, this should not stop the acceleration and expansion of rapid learning in cancer care today, writes Paul Wallace, medical director of Health and Productivity Management Programs at Kaiser Permanente.

The VHA has used its EHR system to provide better care, monitor outcomes, and advance evidence-based care for veterans with diabetes, writes Joel Kupersmith, the chief research and development office at the Department of Veterans Affairs. For example, the VHA has used clinical data from its system to identify chronic kidney disease in one-third of veterans with diabetes, fewer than half of whom have renal impairment indicated on their individual medical records. The system also has helped identify patients at high risk for amputation.

In addition, the VHA has used in-home monitoring devices to collect important data on high-risk patients and transmit the information to care coordinators. The VHA is also using IT to test the use of digital retinal imaging to screen for diabetic retinopathy, a preventable complication for diabetic patients. Future plans include a system, called My HealthEVet, that lets veterans manage their own personal health records, identify clinical studies for which they are eligible, and track their health outcomes online.

Other authors in the special issue include John Lumpkin, senior vice president at the Robert Wood Johnson Foundation; Louise Liang, senior vice president of quality and systems support at Kaiser Permanente; Jonathan Perlin, chief medical officer of HCA Healthcare; Peter Neumann, director of the Center for the Evaluation of Value and Risk in Health at Tufts-New England Medical Center; Greg Pawlson, executive vice president of the National Committee for Quality Assurance; Jean Slutsky, director of the Center for Outcomes and Evidence at AHRQ; and Richard Platt, chair of the Department of Ambulatory Care and Prevention at Harvard Medical School and Harvard Pilgrim Health Care.

A listing of papers in the January 26 special edition on rapid learning can be found at


Health Affairs, published by Project HOPE, is the leading journal of health policy. The peer-reviewed journal appears bimonthly in print with additional online-only papers published weekly as Health Affairs Web Exclusives at


©2007 Project HOPE–The People-to-People Health Foundation, Inc.