Expert Panel Cites Doubts on Mammogram's Worth (NY Times article)

June 16, 2004

Expert Panel Cites Doubts on Mammogram's Worth

By GINA KOLATA

ROCKVILLE, Md., Jan. 23 - An independent panel of expertssaid today that there was insufficient evidence to showthat mammograms prevented breast cancer deaths.

The group, which meets regularly to assess evidence oncancer and provide that information to doctors and to thepublic, said that while it was possible that mammogramswere beneficial, it was also possible that they were not.

The group, called the P. D. Q. screening and preventioneditorial board, agreed here today that seven large studiesof mammography had serious flaws, weakening or castingdoubt on the studies' validity.

The group, which writes information for the National CancerInstitute's online database, said that it would rewriteprevious statements to reflect its new view.

Previously, the group, which does not make specificrecommendations for medical practices, had said that theevidence showed that mammograms prevented breast cancerdeaths starting at age 40.

The decision by this group clouds a continuing debate overwhether women should have regular mammograms. Since 1997,the National Cancer Institute has also recommended thatwomen have regular mammograms, starting in their 40's.

For now, the cancer institute, whose new director, Dr.Andrew C. von Eschenbach, was sworn in on Tuesday, has notannounced any plans to revisit the issue, raising thequestion of whether the institute will end up withconflicting information on its Web site, with one statementsaying that the screening is beneficial and another sayingthat those benefits have not been proved.

The institute's Web site, www. cancer. gov, says that womenin their 40's and older should have the test and that itreduces the breast cancer death rate by as much as 30percent.

Members of the P. D. Q. board said they knew that it was notgoing to be easy for women and doctors to decide what todo. At the meeting, the members wrestled with the questionof how much benefit, if any, mammograms confer. Some, likethe board chairman, Donald A. Berry, chairman of thedepartment of biostatistics at M. D. Anderson Cancer Centerin Houston, said they thought it was unlikely that therewere benefits that could make the test worthwhile.

But others said that they were uncertain and that they werehaving a hard time coming to terms with the profound flawsthey now saw in studies they had thought put mammograms onsolid ground.

"What this new material has contributed is a softening, atleast in my mind, of what I thought was true," said Dr.Isra Levy of the Canadian Medical Association. "I stillthink that for women aged 50 to 69 there might be somethingthere. But we've had our confidence shaken."

In the end, the group decided not to deal with age, sayingdoubts persisted for women of all ages.

The group focused on flaws in the studies that wererecently uncovered by two scientists in Denmark. Thereport, published in October by Dr. Peter C. Gotzsche andOle Olsen of the Nordic Cochrane Center in Copenhagen,concluded that the case for the screening was unproven. Thestudies were so poorly designed and carried out that theymight have found benefits when there were none orexaggerated what benefits there were, the researchers said.None of the studies found that mammography prolonged life,and even when the studies were analyzed as a group, womenwho had the test lived no longer, dying of diseases otherthan breast cancer.

Some experts have taken issue with Dr. Gotzsche and Mr.Olsen's analysis. But today, the P. D. Q. board said itbelieved that the scientists had found serious problems.The board's assessment of mammography is expected to beposted on the cancer institute's Web site in April.

The P. D. Q. board said it would like to see others take on adetailed and independent analysis of the mammographystudies and that some of the data that might settlequestions about the quality of the studies could beobtained with difficulty.

Some may be in medical records of women in a United Statesstudy decades ago. Others are held by Swedish scientistswho, the board and others said, have refused to divulgethem.

The issue of whether women should have mammograms has beencontroversial for some time. In January 1997, when anindependent panel convened by the National Institutes ofHealth took on the question of mammograms for women intheir 40's, Congress got involved. That panel said thenthat there was no evidence that the test prevented breastcancer deaths in women under 50 and suggested that thosewomen discuss mammograms with their doctors before decidingto have one.

Before the week was over, the panel's chairman, Dr. LeonGordis of the Johns Hopkins University, had been summonedby Senator Arlen Specter to testify before Congress. Later,going against the conclusions of the panel convened at itsbehest, the cancer institute said women in their 40'sshould have mammograms.

The P. D. Q. board, which issues its statementsindependently, has also said over the last several yearsthat mammograms prevent breast cancer deaths for women intheir 40's.

Today, the P. D. Q. board emphasized that mammograms havedrawbacks, leading at times to excessive treatments fortumors that would not have threatened a woman's life.

The group agreed that doctors should respect a woman'sdecision, adding that it was rational to decide to havemammograms and that it was rational to decide not to.

Whatever they decided, Dr. Berry said in a telephoneinterview on Tuesday that he was acutely aware of thedifficulty in questioning an enormous mammography business.

"Screening programs bring in patients," Dr. Berry said. "Itisn't just the mammography, but it's the biopsies, thesurgeries and the like. We know that screening isexquisitely fine at finding cancers. Therefore it brings inpatients and they demand treatment."

The problem, he said, is if the women who have mammogramsfare no better, or do even worse because of excessivetreatment, than women who are not screened.