WASHINGTON -- Doctor Ross Fletcher wheeled a laptop computer on a metal cart into an elderly patient's hospital room here and clicked on the "remote data" button. Seconds later, the patient's complete medical record from a hospital in West Virginia flashed on the screen.

Down the hall, a nurse using a bar-code device scanned a patient's wristband and a syringe. A nearby computer, linked to the scanner and the hospital pharmacy, confirmed that she was giving the right drug to the right patient. When she touched the screen again, the time of the injection was entered into the patient's record.

For years, this has been one of the great fantasies of health care: a network of all the farflung data about a patient, linked together to improve safety and the quality of care. In most hospitals, it is still a long way off, but it is coming to life at an unlikely place: the U.S. Veterans Health Administration.

Over much of its history, the federal system that provides medical care to four million military veterans and their families has been known for bloated bureaucracy and poor patient care rather than innovation or visionary leadership. The first thing that springs to mind for many people is the shoddy treatment of Vietnam vets in the 1989 film "Born on the Fourth of July," notes Thomas L. Garthwaite, the Veterans Affairs undersecretary who runs the agency. After federal auditors uncovered serious management problems in 1995, some critics called for disbanding the VHA and funneling its annual budget to the private health care system.

Six years later, the VHA is leading a nascent movement to unlock the data lurking in hospitals to help doctors improve patient care and reduce errors. A new patient safety center is training doctors and others to analyze errors and avoid repeating them. The VHA is bringing researchers and physicians together to discuss the latest medical literature and identify the best treatments for various diseases, and the agency is tracking doctors' performance and using that information to show them how to do better.

"What's needed in the U.S. is nothing short of a medical revolution and the VHA has gone further than most any other organization to revamp its culture and systems," says Robert Brook, an expert in quality assessment at Rand Corp., a Santa Monica, California, think tank.

The VHA isn't alone in these efforts, but it is doing it on a unique scale, leveraging its big budget and 1,300 hospitals and outpatient clinics in a way that no other health care system can. Most private hospitals and doctors, by contrast, are independent contractors in a fragmented system. There is no standardized information technology that would allow everyone to swap data. Fears about patient privacy are another obstacle. And medical institutions outside of prepaid health-maintenance organizations typically make more money treating disease than preventing it, there is little incentive to spend heavily on automated systems that could improve care but don't promise clear financial rewards.

The changes at the VHA began with its former chief, Kenneth W. Kizer. When the emergency-medicine doctor took over the VHA as Undersecretary of Health in October, 1994, he found an organization that was nearly paralyzed by its centralized structure - "sclerotic," he called it. Early on, a request landed on his desk to approve a $10 reimbursement for a computer cable. "It was the epitome of what was wrong," he says.

Efforts to modernize the agency had been thwarted by veterans, researchers and politicians. Each with their own competing agendas. Established by Congress in 1930 to consolidate services for war veterans, the Veterans Administration grew quickly after the end of World War II with the return of some 16 million soldiers. In recent times the population of veterans had been shrinking, but more veterans were enrolling in the VHA system as they grew older and faced health problems. The agency was under pressure to streamline itself and boost services while living on a tight budget.

What's more, hospitals are some of the worst bureaucracies to overhaul. Doctors resist change and are notoriously independent. Almost immediately, Dr. Kizer flew in managers from across the VHA for a two-week brainstorming session in Washington. Their mandate: to create a completely new blueprint for the agency.

They cooked up a master plan with 22 geographic networks, each with its own budget, managers and performance goals. The new structure was designed to move resources into the treatment of chronic diseases affecting many veterans, such as diabetes and congestive heart failure. Hospitals, which provide relatively expensive, episodic care, would be de-emphasized. By 1999, 300 new clinics had opened, but 55% of the agency's hospital beds were eliminated.

The wrenching change was enough to make lasting enemies of some veterans groups and powerful forces in Congress -- and ultimately cost Dr. Kizer his job. While applauding Dr. Kizer's quality initiatives, some veterans with serious medical conditions protested the shift of resources away from hospitals. Dr. Kizer's "philosophy of care just didn't meet the patients' needs," says Jim Peters, Executive Director of the Eastern Paralyzed Veterans Association, which represents 2,200 catastrophically disabled veterans in the New York area. "They need exotic modern medicine."

In 1999, Dr. Kizer's reappointment was blocked, and he decided to quit rather than fight to keep his job. He was succeeded by his deputy, Dr. Garthwaite, who is also preparing to leave once a Bush administration appointee is confirmed in the post.

But the cost savings achieved by eliminating all those hospital beds helped Dr. Kizer create a paperless information system. The VHA had a rich store of digital data stretching back to the late 1970s. Doctors rarely consulted the information because there weren't many computers, and the ones that the agency did have weren't easy to use.

'Dumb Terminals'

By the mid 1990s, the VHA was slowly adding "dumb terminals" -- ones that couldn't talk to each other -- but expansion efforts were scattered and haphazard. All that changed after Dr. Kizer took the helm. The VHA spent hundreds of millions of dollars over three years wiring its hospitals and outpatient centers. Personal computers, with features that allowed doctors to customize their notes and minimize their need for typing skills, were purchased for nearly every office and exam room.

Patient records could now be retrieved easily, along with medical histories, test results and drug prescriptions. Blood pressure readings and other information could be graphed to show patterns over time. Reports from cardiologists, urologists and other medical specialists were included, helping doctors coordinate care. "We like having all these things so we don't make mistakes," says Dr. Pletcher, Chief of Staff at the VHA hospital in Washington.

During a 1998 visit to the VHA's medical center in Topeka, Kansas, Dr. Kizer saw another project that captured his imagination. The hospital had developed a prototype that used bar-code scanners to match patients, drugs and doctors' orders. Sue Kinnick, a longtime nurse at the hospital, had conceived the idea after watching an Avis car-rental agent wield a wireless bar-code scanner and portable computer to check in her car and print out an invoice. At a patient safety conference shortly after his Topeka visit, Dr. Kizer declared that the VHA was going to install the technology in all of its hospitals.

By September, 2000, the system was operating in 170 hospitals around the country. An in-house study in the Topeka hospital found that the system had reduced medication errors there by 70%.

From the start, Dr. Kizer struggled with the broader problem of how to improve patient safety. Studies showed that most medical mistakes could be prevented but that conventional safety training for nurses and doctors didn't help much. To come up with a new approach, Dr. Kizer in 1997 recruited James P. Bagian, an anesthesiologist, engineer and former astronaut who had investigated the Challenger space-shuttle disaster.

Dr. Bagian set out to persuade doctors, nurses and risk managers to think like airline pilots and plane crash investigators. "When a plane crashes, everyone asks: 'What happened?' while in medicine they ask 'Whose fault was it?' " he says. By focusing on blame and punishment, investigators cut their odds of hearing about close calls that could warn of the next mistake. To assure the anonymity of individual health care workers who report errors, Dr. Bagian brought on the National Aeronautics and Space Administration to develop a system modeled on a confidential reporting setup used in the aviation industry.

The VHA's National Center for Patient Safety, established in Ann Arbor, Michigan, in 1998, aimed to create what Dr. Bagian called a "culture of safety" at the agency, Nurses and other health care professionals were encouraged to query doctors when they saw something fishy, "Do you really want a No. 10 blade?" a technician should ask the surgeon in the operating room If there is some doubt, Dr. Bagian advised during a recent training session in Schaumburg, Illinois.

At the heart of the VHA's safety mission is something that has long been anathema in most of the medical world getting doctors and nurses to fess up to mistakes, rather than bury them, so that they can be quickly investigated and won't be repeated, "You can never train a person to be perfect," says Dr. Garthwaite, the current VHA chief. "But if an error happens, you can find out why." In 1999, the agency issued a public report admitting that it had committed more than 3,000 errors that killed 710 people over the previous 19 months. Medical experts said those numbers were in line with other hospitals outside the VHA, but most of them observe an unspoken code of silence.

The open-book approach is one that has been favored for many years by the VHA's medical center in Lexington, Kentucky, and pushed more recently by Dr. Garthwaite. The policy evolved out of a 1986 case involving a patient who died at the Lexington hospital because of a medication error. Rather than cover up the mistake, the hospital administration decided to tell the truth. Not only is such "extreme honesty" the right thing to do -- it also makes victims and their families more forgiving and less eager to punish, says Steve S. Kraman, the hospital's Chief of Staff.

That was the way Sandra Dee Reynolds felt after she and other family members received a visit from a hospital attorney several weeks after the 1997 death of her father. The message, Ms. Reynolds remembers, was blunt: "We killed your dad."

The hospital pharmacy bad given the wrong medicine to Claudie Holbrook, a 67-year-old Korean War veteran. When she learned the truth, Ms. Reynolds cried but was no longer angry. The pharmacist involved was "a good man who made a mistake,'' she says. Later, she toured the pharmacy to see changes designed to avoid another mix-up. Her family settled with the hospital for compensation of $50,000, ''It wasn't about the money with us at all. It is about the apology and knowing what happened," she said.

Measuring the Costs

Sonic lawyers caution that the VHA approach could bankrupt a hospital that makes a bad mistake. But the Lexington VHA's data show that its average malpractice payments over the past 14 years have been slightly below those of many other VHA hospitals. And there are signs that the approach is taking hold. The nonprofit organization that accredits US hospitals, the Chicago-based Joint Commission on Accreditation of Healthcare Organizations, recently adopted new rules that require hospitals to tell patients when they have been harmed.

Another big problem that Dr. Kizer tackled was the failure of the agency's health care and research sides to talk to each other. The two groups distrusted each other, like a couple "hell-bent on divorce," says John R. Feusser, a Duke University professor and Director of the VHA Health Services Research Center in Durham, NC, who Dr. Kizer brought in to work on the problem. The researchers viewed the health care system as backward and bureaucratic, while health system managers thought research was a luxury the agency couldn't afford.

After a couple of false starts, Dr. Feussner in 1998 launched a drive to systematically translate research findings into better health care practices. The initiative had clinical and research experts jointly identify the best medical evidence and treatments for various diseases. It urged researchers to target their efforts in ways that would have the greatest potential benefit for Veterans, for example, cardiovascular disease is a particular problem among the aging veteran population. With that in mind, Physicians treating stroke patients came to Dr. Feussner in 1999 with an idea to study the possibility of monitoring a blood thinner called Coumadin right in patients' homes. The researchers are now planning to start human testing of the devices that do this. If they work, veterans won't have to come to a hospital or clinic to have their blood checked every two weeks. Fewer patient visits would also save money, where these sorts of studies were once viewed as an extravagance, "Now the people who are taking care of the patients are asking us to do the research," Dr. Feussner says.

The VHA's computerized information system now helps measure whether doctors are following recommended guidelines. For instance, the pneumonia vaccination rate at VHA facilities has jumped to about 84%, compared with an average of only 50% nationwide, after studies that show these shots prevent death or serious illness in elderly Americans and some others who are at risk.

Measurement ... changes behavior," says Jonathan B. Perlin, the VHA's Chief Quality and Performance Officer who was brought in by Dr. Kizer in 1999. "Physicians don't want to be laggards."