Racial and socioeconomic disparities that mar the quality of medical care for people infected with the AIDS virus actually decreased after the arrival of $10,000-a-year "triple therapy" antiviral drugs three years ago.

Nevertheless, substantial differences in treatment for the human immunodeficiency virus (HIV) still exist, with white men infected through homosexual intercourse likely to get the best care, as gauged by several key measures.

Those are among the findings of a massive, government-financed survey of HIV treatment in the United States published today in the Journal of the American Medical Association.

"Clearly, there has been some improvement over time in access to care for people with HIV," said Martin F. Shapiro, a physician at the University of California at Los Angeles and at the Rand Corp., a research firm in Santa Monica, Calif., who was one of the leaders of the study. "The gaps between some groups did narrow over time, but they're still there."

The study found virtually no difference in quality of care between people with private insurance and those covered by health maintenance organizations (HMOs). However, the stability and quality of care of patients covered by Medicaid--the state and federal insurance program for the poor and disabled--more closely resembled that of the uninsured than the privately insured, the researchers found. Sex and education were among the strongest predictors of adequate HIV treatment, with men and college graduates faring better than women and people who didn't graduate from high school.

Between 650,000 and 900,000 Americans are infected with HIV, though about 275,000 are unaware of their infection. The HIV Cost and Services Utilization Study questioned nearly 3,000 people, randomly selected from 28 urban and 24 rural areas, who went to the doctor in the first two months of 1996. They were interviewed three times over three years.

The study began just as drugs in the "protease inhibitor" family of pharmaceuticals were being introduced. The drugs have cut AIDS mortality in half and have restored thousands of chronically ill people to vigorous life. The researchers used six measures of quality and stability in medical care. They included going to the doctor regularly, staying out of the hospital, taking triple-therapy drugs and getting preventive medicine for a kind of pneumonia, called PCP, that often kills people with AIDS.

In the first year of the study, 59 percent of people had tried triple therapy. Three years later, 85 percent had. For blacks, the percentage increased from 44 to 80; for Hispanics, 56 to 84; and for whites, 68 to 88. For men, the percentage rose from 61 to 87; for women, 49 to 78. About 72 percent of people with private insurance were on the life-extending drugs sometime during the first year. By the third, this had risen to 91 percent.

In one of the more startling findings, the percentage of people with no insurance who were on triple therapy rose from 46 to 79 in three years. Presumably, the drugs, which often cost $10,000 a year, are paid for by the patients, charities, or state-run AIDS Drug Assistance Programs. Triple therapy by Medicaid patients rose from 53 percent to 81 percent.

Patients who got care through HMOs were more likely to get preventive treatment for PCP than patients treated privately, by Medicaid or by Medicare, or the uninsured. The main finding on that quality measure, however, was that the treatment was used only about 75 percent of the time it should be, regardless of where a person was treated.

The percentage of people whose care was good by all six indicators rose from 29 to 47 in three years. The percentage whose care was substandard or unstable in two or more indicators fell from 34 to 17.

The HIV Cost and Services Utilization Study is the first national, random-sample study evaluating quality of care for patients with a chronic illness. It cost $25 million, about $15 million of which was provided by the federal Agency for Health Care Policy and Research. The researchers hope to follow the surviving patients to see how their care, attitudes and health evolve.

The Spread of AIDS

Here is the total number of AIDS cases reported by the end of 1998.

Race or ethnicityCases

White, non-Hispanic 304,094

Black, non-Hispanic 251,408

Hispanic 124,841

Asian/Pacific islander 4,974

Native American 1,940

Unknown 943

The number of women, blacks and heterosexuals contracting HIV is growing rapidly.

Percentage of HIV diagnoses 1994-97

Women 28%

Blacks 57%

Hetero-sexuals 18%

SOURCE: Centers for Disease Control and Prevention

CAPTION: DISPARATE TREATMENT (This chart was not available)