Researchers have found evidence that Jewish deaths decline just before Passover and increase afterward, supporting the folk wisdom that some people near death can cling to life to reach an important event.

In analyzing deaths over 19 years in California, researchers found evidence for that ability in people with cancer or diseases of the heart or brain blood vessels.Results suggest that some people can "hang on until occasions that are important to them, which is something that has been speculated on for thousands of years," said researcher David Phillips, a sociology professor at the University of California at San Diego.

The work suggests death can be postponed about a week, said Phillips, who reports the study with graduate student Elliot King in Saturday's issue of The Lancet, a British medical journal.

In interviews, experts said similar effects have appeared in other studies and in treatment of dying patients, but one suggested the new finding may reflect anticipation before the holiday and emotional letdown after-ward.

The study included deaths three months before and three months after the large family meal that occurs on Passover eve. Deaths among Jews and non-Jews were compared for 1966 through 1984.

Researchers said the strategy let them compare death rates of one group for whom the event held great meaning with other people for whom it did not. And because the date of Passover varies annually, its impact would stand out from unrelated chronological trends, researchers wrote.

Death records did not list religion, so researchers worked with a list of surnames common among Jewish immigrants from central and Eastern Europe, separating out names also common among non-Jews.

Analysis found the "Passover effect" only among white males. Researchers said that because of intermarriage in California, many women with Jewish surnames are not Jewish.

For white men with Jewish surnames that are not common among non-Jews, the study found 275 deaths in the week before the Passover meal and 346 in the week following, compared to the 311 deaths expected for an average week.

The effect was stronger when the first day of Passover fell on a weekend, with 23 percent fewer than expected deaths in the week before the meal and 19 percent more than expected the following week.

Researchers said a weekend Passover allows for more travel, creating larger family gatherings, which may increase the social im

portance of the holiday for the person near death.

In contrast, no difference in death rates was found within a comparison group of Chinese and Japanese, for whom Passover presumeably has no particular significance.

Phillips said many variables affect the timing of death, and if a person dies just before a major event it does not mean he did not care about it.

The finding "does not surprise me in the least," said Therese Rondo, a Rhode Island psychologist who has worked with dying patients.

"We often see individuals who will hang on until after a grandchild is born or until somebody gets there to say goodbye," and previous studies have suggested such abilities, said Rondo, author of a new book on grieving.