U.S. hospitals are improving the quality of the care they provide, but even the best fail too often to offer the right treatments, such as immediately giving aspirin to victims of heart attacks and properly administering antibiotics to pneumonia patients, according to the two most comprehensive analyses of the issue.

The studies of more than 3,000 hospitals nationwide found that despite overall improvement, care varies dramatically around the country, with those in the North and Midwest generally outperforming those in the South and West.

Even within regions, the quality of care is inconsistent, the studies show. Among the 40 most populous regions, for example, Washington area hospitals ranked 10th in treating a common condition in which the heart slowly fails but only 28th in caring for victims of sudden heart attacks and 29th in dealing with pneumonia.

"Wherever you live and whatever you're being treated for, you really should get the same quality of care," said Ashish K. Jha of the Harvard School of Public Health in Boston, who led one of the analyses published in today's issue of the New England Journal of Medicine. "Our study reinforces that that's really not happening."

There are also significant differences in quality between categories of hospitals: Nonprofit hospitals tend to deliver higher-quality care than for-profit ones, and those affiliated with medical schools score better than those that are not, the researchers found. There is also wide variation within hospitals, meaning patients have no guarantee they are going to get the best care by choosing, for example, big centers associated with well-known medical schools, Jha's analysis concluded.

"There is this notion that you can pick out something that will tell you, 'This is a very good hospital,' " Jha said. "We found you can't use any simple metric to say if they do well in one area, they do well across the board. Hospitals vary. They do well in one way but not necessarily in others."

Hospitals affiliated with medical schools, for example, tend to deliver better care to patients with heart attacks or heart failure but score more poorly when it comes to caring for those with pneumonia, the analysis found.

"It is a little scary that you can't look at one simple measure and know everything about a hospital's quality of care," Jha said.

Other researchers said that while the overall improvement was encouraging, the studies showed that many hospitals too often fail to deliver broadly accepted, easily administered therapies.

"What's going on here? These are treatments that are no-brainers. These are the easy things," said Lucian Leape of the Harvard School of Public Health, a medical quality expert who was not involved in the studies. "We're really looking at patients for whom 100 percent should be receiving these treatments. So why isn't it 100 percent?"

Many studies have examined the quality of care in hospitals, but most have been limited to individual institutions or geographic regions or specific types of hospitals or patients, such as Medicare recipients or veterans. The analyses are the first to use nationwide data collected by two new efforts to improve hospital quality.

In the first, Jerod M. Loeb and his colleagues at the Joint Commission on Accreditation of Healthcare Organizations analyzed data the group began soliciting in 2002 from every hospital it accredits. The researchers compared how 3,087 hospitals scored between the third quarter of 2002 and the second quarter of 2004 on 18 measures designed to assess the quality of care for three conditions: congestive heart failure, heart attacks and pneumonia.

Those conditions are among the most common reasons people are hospitalized. (Heart attacks involve a sudden blockage of arteries feeding the heart muscle, and heart failure is a progressive failure of the heart's pumping ability. For ratings of some local hospitals, see hospitalcompare.hhs.gov.) During the study period, hospitals improved overall on 15 measures, the researchers found. But even the best hospitals rarely scored 100 percent, and many often scored far lower.

"We are seeing significant improvements in care across the board," Loeb said. "But there are clearly areas where there is still room for improvement."

For the second study, Jha and his colleagues analyzed data from 3,558 hospitals collected by the Hospital Quality Alliance, which was created in 2003 by 12 organizations, including three hospital associations, the American Medical Association, the federal Centers for Medicare and Medicaid Services and the commission on accreditation.

While most hospitals scored highly on many of the measures, in 10 percent to 30 percent of cases patients failed to receive potentially life-saving treatments.

"Overall in many hospitals the care is still not nearly as good as it should be," Jha said. "These are important therapies, and these are easy to deliver. It really should be close to 100 percent."

Hospitals are making progress, but the variations show the need for further improvement, other experts said.

"Given how much we collectively spend on health care, we ought to be doing a lot better," said Patrick S. Romano of the University of California at Davis, who wrote a commentary on the research.