Filed at 8:48 p.m. ET

LONDON (AP) -- A team from Cornell Medical Center has taken a fresh look at data that caused some investigators to doubt that mammography saves lives, and has concluded that there is a significant benefit over the long term.

``Screening does not have an immediate effect; the deaths that get to be prevented by screening are in the future, years away,'' said Claudia Henschke of Weill Cornell Medical Center, whose team published its findings in the British medical journal, The Lancet.

The report is the latest contribution to a heated debate on the value of mammography in preventing breast cancer deaths -- and the efficacy of cancer detection in general.

Ten U.S. medical groups, including the American Cancer Society and the American Medical Association, placed a full-page ad in The New York Times on Thursday, supporting mammography while conceding the limitations ``and even some flaws'' in existing studies.

``Early breast cancer detection means a greater chance for successful treatment and a greater range of treatment options,'' the advertisement said.

``We have grave concerns that these public debates have already begun to erode the confidence in mammography that has been built up over the past two decades.''

Henschke's team reviewed data from a study published in the British Medical Journal in 1988, based on data from Malmoe, Sweden.

The authors of that study concluded that mammography led to a significant reduction in deaths from breast cancer after six years.

Those conclusions have been attacked by Ole Olsen and Peter C. Goetzsche of the Nordic Cochrane Center in Copenhagen, who reported their findings in The Lancet two years ago.

They concluded that ``screening for breast cancer with mammography is unjustified,'' and in a follow-up report declared ``there is no reliable evidence that screening for breast cancer reduces mortality.''

Olsen and Goetzsche looked at seven studies but rejected five as flawed. They accepted two as valid, the Malmoe study and another from Canada, published in 1992 in the Canadian Medical Association Journal. Henschke's team looked only at the Swedish data.

``The Canadian follow-up stopped at the point at which the Malmoe follow-up started to show fewer breast-cancer deaths among the screened, the Canadian screening having been continued for only three to four years after study entry,'' Henschke's team said.

They criticized Olsen and Goetzsche for initially looking only at total breast cancer deaths over 11 years -- 63 among those screened and 66 among those not screened -- without analyzing when the deaths occurred. In a subsequent review, Henschke's team said, Olsen and Goetzsche based their conclusions on deaths from all causes rather than isolating deaths from breast cancer.

The Henschke team's report added that early detection and treatment may be associated with ``somewhat increased mortality'' in earlier years because of possible complications from surgery and other treatments.

Dr. Robert Smith, head of cancer screening at the American Cancer Society, commented that the latest report in The Lancet would not end the debate. ``But I think it should highlight what most experts who have a track record in the field have said, that the Goetzsche and Olsen analysis was severely flawed.''

Smith, who was not involved in the Henschke analysis, said it was expected that the benefits would appear after some years.

``Usually between year three and year six to seven, the difference in deaths should begin to separate as the advantages of treating tumors when they are smaller ... becomes evident. If your screening has been very good, that will come a little bit earlier,'' he said.

The Lancet also published an exchange of letters between Goetzsche and members of the Cochrane Breast Cancer group debating technical issues of methodology and publication.

Goetzsche said he had published his earlier reports ``because we believe it is important for women to know that screening increases their risk of losing a breast.''

``This finding contrasts with the information they get from screening advocates who generally say the opposite, but without reliable data to support their claims.''

He also argued that data on deaths from breast cancer ``are biased in favor of screening.''

Lancet editor Richard Horton added a comment that the exchange highlighted the intensity of a debate which has now drawn wide public attention.

``I cannot imagine anybody wishing that screening mammography does not succeed in reducing both breast cancer and overall mortality among women,'' Horton wrote.