It is an embarrassing but no longer well-kept secret that despite health care spending of about $1.3 trillion a year -- including about $25 billion in federally funded research -- many Americans receive medical care that is not terribly good.

Only 57 percent of people with the heart problem atrial fibrillation are prescribed warfarin, even though it dramatically cuts the risk of stroke. Most experts advise that women over 50 have a mammogram at least every two years, but only 60 percent do. Even when researchers at a hospital participate in a study that demonstrates 20 percent lower mortality in heart attack patients given ACE inhibitor drugs, their colleagues in the same hospital are no more likely to adopt the practice than doctors elsewhere.

Why is this? And how can it be changed?

Many organizations are trying to answer those two questions. In the federal government, the office always pondering those questions is the Agency for Healthcare Research and Quality (AHRQ).

AHRQ enters a new phase of its occasionally rocky journey with the appointment expected today of Carolyn M. Clancy, 49, as the fourth director since its founding in 1989. She has been acting director since the death last year of the much-beloved John M. Eisenberg.

Clancy, an internist and health services researcher, has been with the agency for 12 years. Over that period, AHRQ -- originally named the Agency for Health Care Policy and Research -- has at times found itself the object of overweening hope and undeserved criticism.

Despite the agency's modest size -- its budget is about $300 million a year -- politicians have periodically held it out as the place that will find the magic key to painlessly stopping the growth of health care costs. However, when it made its boldest and most prescriptive steps -- sponsoring a series of clinical practice guidelines for common medical conditions in the 1990s -- it found itself facing the buzz saw of medical special interests and congressional inquiry.

As Clancy sees it, AHRQ has two big challenges. One is to help develop the "evidence base" of medicine -- the raw material of better care. The second is to identify the best ways to get doctors and nurses to use optimal, up-to-date treatments.

The first effort acknowledges that the body of medical research on just about any important subject is vast -- too big for the average practitioner to grasp. Over the past 15 years, however, there has emerged a set of rules and methods by which a team of experts examines all the studies on a given question, evaluates their validity (combining numerous ones, on occasion, to increase the statistical power of the results) and ultimately extracts conclusions -- the "best evidence" -- from the mass of information.

This work is expensive and laborious. Health care organizations and professional societies cannot easily do it. Consequently, AHRQ has helped establish "evidence-based practice centers" at 13 universities, and is paying researchers there to create "systematic reviews" of many topics.

The results are lengthy -- the one that analyzed practices designed to enhance patient safety was 668 pages -- and invaluable. And they are free for anyone to use.

This work has proved much less controversial than drawing up practice guidelines for health care practitioners -- a task that required making judgment calls on issues where there was not research pointing unambiguously to the best treatment.

"The difference is this is just the facts," Clancy said.

She describes the second challenge as the effort to figure out "what systems or strategies make the right thing to do the easy thing to do."

AHRQ is paying for dozens of studies of how clinics and hospital systems improve care, vaccination rates and patient satisfaction, and reduce waiting time, overlooked lab results and medication errors. One recent study, for example, looked at the relationship between workload in primary care practices and "risky prescribing events." Another is examining the link, if any, between financial incentives and quality of care in a large group practice.

Clancy keeps her hand in practice, supervising residents at a George Washington University clinic one evening a week. Among other things, it helps her remember the frustration practitioners face on a daily basis -- namely, that questions will arise whose answers the doctor cannot remember and does not have time to find while the patient waits.

Hand-held computer devices and Web-based programs are helping bridge this gap. So are more passive systems that check on performance as it happens. Nevertheless, the solution to this problem remains something close to the holy grail of practice -- and a priority of Clancy's and AHRQ.

"The key is information," Clancy said. "Making information available at the point of care when you need it."

Carolyn M. Clancy, acting director of the Agency for Healthcare Research and Quality, is expected to be named its director today.