Women confused about whether to start having mammograms at age 40 are being advised by cancer experts to ask their doctors to help determine their risk of breast cancer.

In a move that outraged the American Cancer Society, advisers to the National Cancer Institute concluded Thursday that "each woman should decide for herself" whether to start having mammograms during her 40s."We are disappointed," said the society's Robert Smith. "We should-n't underestimate the influence this report may have."

The experts agreed that mammograms starting at age 50 are vital, cutting breast cancer deaths by about 30 percent.

But NCI Director Richard Klaus-ner convened the panel to decide whether to reverse a 1993 government decision that there was insufficient scientific evidence to justify testing younger women.

The two days of debate convinced Klausner that mammograms in the 40s can cut breast cancer deaths - and in an unprecedented move, he announced that he disagreed with and was surprised by the conclusions of his own advisers.

He will take the debate to the presidentially appointed National Cancer Advisory Board next month."It is a difficult problem," Klausner said. But "my own view is that there is a benefit in terms of mortality."

The American Cancer Society recommends mammograms every year or two starting at age 40. But because the NCI does not recommend annual mammograms until age 50, thousands of women hear conflicting opinions from doctors and struggle to get insurance payments for earlier testing.

Klausner sought to settle the controversy in the wake of new research, including a study of women in Sweden suggesting that mammograms during women's 40s cut breast cancer deaths by 44 percent.

But the NCI panel debated dozens of studies and concluded mammograms showed no mortality bene-fit until women in their 40s had been followed for 10 years - raising questions of whether testing early in the decade or later was responsible.

The panel wound up advising doctors to simply help patients weigh whether the expense and the possibility of being frightened by benign tumors or lulled into a false sense of security are worth the possible benefit.

But it did unanimously recommend that insurance and managed-care companies pay for mammo-grams for any 40-year-old.

"We are not saying there's no benefit; we're just saying the benefit might be small and might not occur until late," said panelist Dr. Leslie Laufman, a Columbus, Ohio, oncologist.

Breast cancer strikes about 180,000 American women each year and is expected to kill 44,000 this year - about 10 percent of them under age 50.

Why not err on the side of caution? In addition to mammograms' cost - $40 to $150 - as many as 90 percent of the abnormalities they uncover are benign. To be sure, women often undergo stressful, somewhat painful further testing.

Still, federal surveys show 63 percent of women in their 40s have had a mammogram in the past two years.

For a confused 40-year-old, the report "is going to give her misinformation," said Dr. Stephen Feig of Thomas Jefferson University. He reported in December that annual mammograms could cut breast cancer deaths by some 35 percent among women in their 40s.