The National Institutes of Health has made little progress in carrying out a four-year-old policy to include more women in government-funded studies of diseases and their treatments, according to a study by the General Accounting Office described at a congressional hearing yesterday.

Although an NIH memorandum acknowledged last year that underrepresentation of women in such research has caused "significant gaps" in medical knowledge, the agency has been slow and inconsistent in its efforts to get researchers to include more women in studies, said Mark V. Nadel, the GAO's associate director for national and public health issues.

"This is not a question of legal access or affirmative action. It is a question of health and well-being," said Rep. Henry A. Waxman (D-Calif.), chairman of the House Energy and Commerce subcommittee on health and the environment.

For example, a 1988 study of 22,000 male physicians, funded by the National Heart, Lung and Blood Institute (NHLBI), found that aspirin could prevent heart attacks in men. Doctors subsequently recommended that older men at increased risk for heart disease take an aspirin every other day but specifically said they could not offer women the same advice. Heart disease is the leading cause of death in both sexes.

"If heart disease studies use data solely on men, should women take an aspirin a day?" Waxman asked. The question cannot be answered authoritatively yet.

William F. Raub, acting director of the NIH, said in an interview the "vast majority" of NIH studies include adequate numbers of women and are in compliance with the policy. He said the GAO study had uncovered procedural problems within the agency, and that the NIH would follow the GAO's recommendations to correct those problems.

"I don't believe the NIH system is grossly out of focus," he said. "We do need to do some fine tuning."

The NIH, which pays for most U.S. medical research, first stated in October 1986 that it would encourage greater inclusion of women in studies, but it did not issue guidelines on the new policy for its officials until last July.

The GAO found variations among the NIH's various research institutes in how the policy is interpreted. A booklet sent to grant applicants makes no mention of the policy. The NIH has no central mechanism for counting the number of women or minority members included in its studies, and NIH officials have not encouraged researchers to analyze findings to reveal gender-related differences, according to the GAO study.

Raub said that in studies of diseases that occur more frequently in men, such as heart disease, researchers must often decide between doing an all-male study in order to get a quicker answer or doing a larger, slower study that includes both sexes in order to get results that can be applied to both.