A new study from Sweden is stirring fresh debate over whether women in their 40s should get mammograms. It suggests that the breast cancer screening test can lower the risk of dying of the disease by 26 per cent or more in this age group.

That's a bigger benefit than was found by earlier studies, which a year ago led an influential panel of U.S. science advisers to recommend against routine screening before age 50. The panel said the benefits were so small and potential problems from screening so great that the decision should be left to each woman and her doctor.

That advice set off cries of outrage from many cancer screening advocates and even from members of Congress. But since then, there have been other studies supporting the task force's stance. And just last week, another study in Norway found that the benefit of mammograms even for women 50 and older is less than has been believed.

The new study has major limitations, and cannot account for possibly big differences in the groups of women it compares. Nor does it consider the harm -- such as unnecessary stress, unwarranted biopsies and overtreatment -- of screening women in their 40s. Breast cancer is less common in that age group, so mammograms can raise many false alarms.

However, the new Swedish study appears to be the largest of this age group -- about one million women. And "it captures the real-world experience" they have from regular mammograms, said Dr. Jennifer Obel, a spokeswoman for the American Society of Clinical Oncology.

The Swedish researchers estimate that 1,252 women in their 40s would have to be offered screening every other year for 10 years to save one life.

Results of the new study were reported in a teleconference on Wednesday and will be presented at a cancer conference in Washington later this week. They also were published online by the journal Cancer.

Advocates of mammograms for younger women praised the study.

"It's just one piece of evidence supporting the fact that screening women in their 40s does save lives," said the American Cancer Society's chief medical officer, Dr. Otis Brawley. "We believe that women in their 40s should be screened, but we also believe that women should be informed of the limitations of mammography."

About 5,100 of the roughly 40,000 breast cancer deaths each year in the U.S. are in women under age 50, but many women who die at older ages were diagnosed at younger ages, according to the cancer society.

"If we screened all 22 million American women who are aged 40 to 49 this year, this study suggests it would translate to 2,000 to 2,200 lives saved," Brawley said.

Dr. Daniel Kopans, a radiology professor at Harvard Medical School and an American College of Radiology spokesman, said the study "should end any debate and end the use of age 50 as a threshold for screening."

But others were unconvinced and stood by the U.S. Preventive Services Task Force. That panel last fall recommended against routine screening for women in their 40s.

"Everybody is confused, but in my opinion, the strength of evidence is robust" for following the panel's advice, said Dr. Ranit Mishori, a family medicine specialist at Georgetown University.

Dr. Jeanne Mandelblatt, a Georgetown doctor who headed six research teams for the government panel, said the new study "does not balance the benefits against the harms," as the panel sought to do.

The new study looked at Sweden's mammography program. Since 1986, the country has required that screening be offered to women over 50 but left it up to each county to decide whether to offer it to younger women. About half of counties did, and researchers compared breast cancer death rates in areas where it was and wasn't offered.

They counted breast cancer deaths of women who had been diagnosed in their 40s and died within 16 years of followup.

There were 803 such deaths among women in counties that offered screening versus 1,238 in counties that didn't offer it, although the number of women in each group and the amount of time they were followed differed. Researchers did not express the results in terms of death rates, which would have made comparing these groups much easier.

The results translated to a 26 per cent lower risk of dying of breast cancer for those offered screening. There was a 29 per cent lower risk for women who actually had mammograms, said lead researcher Hakan Jonsson of Umea University in Umea, Sweden.

Researchers did not have information on, and the study therefore could not account for, any differences in general health and other factors that could have affected the number of deaths.

"It very well may be that blue-collar counties may have higher rates," Brawley said.

A comparison of breast cancer death rates in the same counties in the 15 years before screening began found 6 per cent fewer deaths in the areas offering screening, suggesting this group was healthier, Jonsson said. But this difference was not so large that it would significantly affect the study's findings, he said.