HICAGO, Dec. 25 — A common virus may increase the risk of developing multiple sclerosis, new research suggests. The study bolsters evidence that links multiple sclerosis with the Epstein-Barr virus.

Researchers at Harvard University found that women whose blood had significant levels of antibodies to Epstein-Barr were four times more likely to develop multiple sclerosis than women without high levels.

The virus, from the herpes family, is best known as a cause of mononucleosis, the "kissing disease." It also has been linked to other ailments, including other nerve disorders and cancers. The virus is so common that some experts estimate that it has infected 95 percent of Americans by age 40.

Most exposure probably occurs in childhood, when there may be no symptoms, but antibodies would remain into adulthood, said Dr. Alberto Ascherio of the Harvard School of Public Health.

Coupled with previous research that showed that people without Epstein-Barr antibodies rarely developed multiple sclerosis, the findings are "strong evidence in favor of a link," said Dr. Ascherio, lead author of the new study, which The Journal of the American Medical Association is publishing on Wednesday.

But Dr. Donald Gilden, a neurologist at the University of Colorado who wrote an editorial in the journal, said that although multiple sclerosis was probably caused by a virus, Epstein-Barr was probably not the culprit. Dr. Gilden said blood tests like those in the study were not useful in determining cause and effect.

Nicholas LaRocca of the National Multiple Sclerosis Society said the study showed "a statistical association, but not anything that you could really pin down as being causal or anything that really identifies a mechanism."

About 350,000 Americans have multiple sclerosis, a nerve disease that causes symptoms like numbness, muscle weakness and paralysis. It is generally thought to be an autoimmune disorder, meaning that the body's immune system mistakes normal tissue as foreign. In multiple sclerosis, a protective coating for nerve fibers, myelin, is attacked. Previous research has suggested that viruses may set off the autoimmune response in diseases like multiple sclerosis.

One problem with such studies is that it is unclear which came first, exposure to the virus or multiple sclerosis. Determining when multiple sclerosis begins is difficult, because symptoms may appear after the disease has begun.

Dr. Ascherio said he sought to avoid that problem by studying women who had blood tests up to six and a half years before developing multiple sclerosis. The authors examined samples from 62,439 women in a study of nurses that began in 1989. The women gave samples in 1989 and 1990 and from 1996 to 1999 and were followed up through 1999.

Among them, 144 had multiple sclerosis, although most were thought to have developed the disease before their tests.

For each woman with the disease, two women without the disease were randomly selected, matched by birth year.

Of 16 women whose blood was taken before they developed multiple sclerosis, 15 had significantly elevated levels of Epstein-Barr antibodies, Dr. Ascherio said.

The study of more than 200,000 nurses, in two groups starting in 1989 and 1976, has led to new findings in fields like exercise, stroke and cancer.