The government's controversial attempt to trim Medicare costs by changing its system for paying hospitals has not adversely affected the quality of care offered to the nation's elderly, according to the most comprehensive study of the new Medicare payment system ever conducted.

But the four-year, government-commissioned survey cautioned that because of the new, stricter guidelines used by the government to pay for health care of the elderly, hospitals are increasingly discharging Medicare patients before their condition has completely stabilized. That trend has increased the burdens on patients' families and other members of society.

"This study reassures us about some things," said Katherine L. Kahn, one of the leaders of the Rand Corp. team that conducted the multimillion-dollar study. ". . . The introduction of the new payment system did not interrupt an ongoing trend in improvement in medical care. But it also says we have to keep looking at the situation. It also says that there are problems that need to be addressed."

The subject of the study was Medicare's Prospective Payment System (PPS), introduced by the federal government in 1984 in an attempt to limit spiraling health care inflation. Rather than simply reimbursing hospitals for the bills they submitted for caring for Medicare patients, the government set clear and specific guidelines for how much it would pay for specific medical problems and procedures. For example, for a common operation such as pinning a fractured hip, the government stated that it would pay a flat fee, and hospitals would make money only if they managed to perform the operation for less money than the fee the government was paying.

The idea of paying a fixed fee for a specific package of services -- which since has been adopted by many private insurance companies -- has been controversial because critics said that it would encourage doctors and hospitals to cut corners in providing care and to discharge patients before they were well enough to go home.

In a series of articles that take up today's entire issue of the Journal of the American Medical Association, the Rand Corp. researchers lay many of those concerns to rest.

The study compared the performance of hospitals in treating Medicare patients in 1981-82 -- before PPS was introduced -- with their record in 1985-86 -- after the new system were in place. It did not look at the treatment of privately insured patients, who make up about half of all medical expenditures.

After analyzing the medical records of 17,000 people treated in 297 hospitals around the country, the group concluded that the steady increase in the quality of hospital care that had characterized the free-spending years prior to the introduction of prospective payment continued even after spending restraints were imposed.

PPS, in other words, did not seem to hamper hospitals' ability to improve their services, efficiency and medical technology. The percentage of patients receiving poor or very poor care, for example, declined from 25 percent to 12 percent after PPS was introduced. Mortality rates within the first month after a patient was admitted to the hospital also declined.

The researchers did find -- as had been predicted -- that the percentage of Medicare patients discharged from hospitals before they had fully recovered increased under PPS, jumping from 10 to approximately 15 percent of all patients.

This tendency to discharge patients earlier, which the study attributed at least in part to attempts by hospitals to cut costs, caused a modest increase in mortality among Medicare patients. But that increase was offset by the improvements achieved by hospitals in overall care. In addition, the study found no evidence that the trend toward discharging patients earlier was causing increasing numbers of patients to be readmitted to hospitals, or to nursing homes.

"The forces that push up health care costs will probably not abate in the near future, but one possible lesson from the PPS experience is that we can simultaneously constrain costs and improve quality," concluded Gail Wilensky, the administrator of the Medicare program, in an accompanying editorial.