On February 21, 2002, U.S. Department of Health and Human Services Secretary Tommy G. Thompson announced an updated recommendation from the U.S. Preventive Services Task Force (USPSTF). This recommendation calls for screening mammography every 1 to 2 years for women 40 and older.
The transcript of the press conference is below. Select for the Web cast. For details about the recommendation, select Screening for Breast Cancer.
The USPSTF, the leading independent panel of private-sector experts in prevention and primary care, is sponsored by the Agency for Healthcare Research and Quality (AHRQ) and conducts rigorous, impartial assessments of scientific evidence for a broad range of preventive services.
The press conference featured:
Secretary Tommy Thompson: Good afternoon, ladies and gentlemen. Today we are going to discuss a very important subject for women and their families. It's good to see so many of you here today, and thank you so very much for coming.
Joining me today are Dr. Janet Allan. She's the Vice Chair of the United States Preventive Services Task Force. She's Dean of the School of Nursing at the University of Texas Health Science Center. And Dr. Peter Greenwald over here, who's the head of the Division of Cancer Prevention at NCI, the National Cancer Institute.
And also in the audience are people representing organizations that do such tremendous work on cancer and health issues, and I thank them so very much for all of their work. Thank you so very much for your contributions to keeping America healthy and fighting cancer.
Today we want to send a very important message, emboldened by a new recommendation about breast cancer and mammography. Breast cancer is a common cancer affecting women and early detection is so critical to fighting this disease, and especially if you're gonna have success.
Last year, more than 200,000 American women were diagnosed with breast cancer, according to the National Cancer Institute, and 40,600 women died from this terrible disease. It touches almost every family, either through relatives or friends or neighbors.
Seven years ago, my wife, Sue Ann, went to have her yearly mammogram and was diagnosed with breast cancer. Sue Ann likes to tell me—she likes to tell me a lot, likes to say that information leads to knowledge, knowledge leads to power, and that power leads to the ability to make informed choices and decisions, for both ourselves as well as for our families. Well, it was information and knowledge that led her to make an informed choice to go in for a mammography screening, and it may well have saved her life.
Over the course of the past few months, and particularly in the past few weeks, there has been renewed discussion, and controversy about mammography and its impact on saving lives.
The Department of Health and Human Services, with its agencies, is constantly reviewing the latest science and research in the most cutting-edge technology, the advances, so that we will be able to provide women with the best information, as well as the best advice on breast cancer, early detection, and treatment.
When it comes to mammography, the Federal Government's recommendation remains very clear. Women in their 40s and older should be screened every 1 to 2 years with mammography. The National Cancer Institute also advises women to consult closely with their clinician about the issue of breast cancer and what course is best for them.
It is very important for women to be fully aware of the recommendation, so that they can make informed choices for themselves and for their family.
Today we have a lot more information on mammography from an independent review of this early detection tool. In 1998, our Agency for Healthcare Research and Quality reconvened the highly respected United States Preventive Services Task Force and asked this independent panel of experts to review the value on a broad range of clinical preventative services, including mammography.
The result of that review, the U.S. Preventive Services Task Force recommends, and I want to underscore that, recommends routine mammography every 1 to 2 years for women 40 years and older. This is an update from the Task Force's 1996 recommendation on mammography, which recommended a routine screening for women ages 50 to 69.
The Task Force recommendation sends a very powerful and clear message about the value of mammography as an early detection tool that can help save lives. I'm going to ask Janet Allan to further discuss the Task Force recommendation in a moment.
Now, we certainly acknowledge that there are legitimate issues surrounding mammography. Mammography is not a perfect tool. Sometimes, for example, there can be false positives that cause a great deal of anxiety. But, mammography is an important and effective early detection tool that does help to save lives. We want women to understand this point very clearly.
We also are very encouraged by the exciting promise of the new developments in technology and treatment that will keep us moving forward in providing the best care possible for women with breast cancer. HHS, NIH [the National Institutes of Health] and NCI are deeply committed to further research on better methods of early detection and treatment for breast cancer, as well as evaluating the new data on mammography.
For example, the NCI, National Cancer Institute, is currently reviewing new digital mammography and its potential, and it does have some very good potential, in the earlier detection of breast cancer. Of course, new advancements and developing science will bring more debate. It will bring more reviews of our existing practices and recommendations, and we must keep these discussions in the right perspective.
The ongoing debate that comes with the progress of science and technology should not detract, or distract, from the existing science-based recommendation that women in their 40s and older should be screened with mammography every 1 to 2 years.
This is the National Cancer Institute recommendation, and it is now affirmed by an independent panel of medical experts. Please, all of you in this audience and those people watching, please take these recommendations to heart.
So again, as my wife and the scientists in this great department often remind me, women want to be informed consumers and partners in their care on the medical tests and practices. At the Department of Health and Human Services, we want to help them and assist them in this objective.
The best information we give women today, with the support of this new independent review, is to avail yourself of routine mammography if you're 40 or older. It's an important tool as part of our comprehensive effort to help save lives from the ravages of breast cancer.
Now, I'd like to introduce Janet Allan, who is Vice Chair of this task force. Janet holds a Doctorate Degree, is a Registered Nurse, a Fellow of the American Academy of Nursing, as well as being Dean of one of the finest schools of nursing in the country. And I want to thank her so much for her work and for her efforts on this task force, and for being here today.
Janet?
Dr. Janet Allan: Thank you, Secretary Thompson.
Good afternoon. I am very pleased to be here today as the Vice Chair of the U.S. Preventive Services Task Force, and to present our updated recommendation on mammography.
The Task Force, as Secretary Thompson mentioned, is an independent panel of 15 private-sector experts in prevention and primary care, and it was first convened in 1984 and is now sponsored by the Agency for Healthcare Research and Quality.
The Task Force published two earlier recommendations on mammography screening, first in 1989 and again in 1996. Both of those endorsed mammography for women 50 and older. Now, based on extensive review of the research, the Task Force has extended its recommendation to women 40 to 49. The task force now recommends screening mammography every 1 to 2 years for women age 40 and older.
To update its 1996 recommendation, the Task Force extensively analyzed the studies on breast cancer screening, including the critiques by Olsen and Gotzsche of the Nordic Cochrane Center in Copenhagen.
We acknowledge that the clinical trials are flawed and they are not perfect. These trials have been discussed in the literature for years. We found the same flaws found by others. But, the task force concluded that the studies were still valid and that mammography screening reduces deaths from breast cancer.
While mammography benefits women 40 and older, we found the strongest evidence of benefit and reduced mortality among women ages 50 to 69. Also, the risks—there are risks associated with mammography. False positive results that lead to additional mammograms, unnecessary biopsies and surgery. And these are less likely to occur in older women.
The evidence from the studies did not allow us to determine when women should have their first mammogram and how frequently to have a mammogram. For this reason, we urge women to discuss their personal preferences with their clinicians to determine when they should have their first mammogram and how often to have a mammogram.
The discussion of personal preferences should include talking about risk factors for developing breast cancer. The risk factors include age, women—having your first child after 30, a prior diagnosis of breast cancer, having a mother or a sister who has had breast cancer, or having had a previous breast biopsy showing atypical hyperplasia, which is an irregular pattern of cell growth in breast tissue.
We again found insufficient evidence to recommend for or against routine clinical breast examination by the clinician alone as a screening tool for breast cancer. We also found insufficient evidence for or against routinely teaching or having women perform breast self-exams.
While these techniques detect a few additional tumors, there were not enough data to determine whether they reduce deaths from breast cancer.
My colleagues and I on the U.S. Preventive Services Task Force rigorously assessed the available scientific evidence for a broad range of clinical preventive services. We based our recommendations on overall quality of the studies or the evidence.
Our bottom line is that mammography reduces deaths from breast cancer.
Thank you.
Secretary Thompson: Thank you very much, Janet. I appreciate your hard work and certainly thank you so very much for being here today.
Now I'd like to introduce Dr. Peter Greenwald, who's the head of the Division of Cancer Prevention at the National Institutes of Health. Peter and the scientists out there are doing an excellent job and are some of the best researchers in the world.
The National Cancer Institute, of course, is under the National Institutes of Health. They're doing a remarkable job in researching, developing a new wave of technology and treatment that's going to better care for women and breast cancer patients.
Dr. Peter Greenwald, please come up and say a few words.
Thank you, Peter.
Dr. Peter Greenwald: Early detection is one of the most important approaches to cancer control. The National Cancer Institute agrees with the analysis of the U.S. Preventive Services Task Force, and continues to recommend mammography every 1 to 2 years for women, beginning in their 40s.
Experts agree that mammography detects tumors when they are smaller, that is, earlier tumors; that it detects more tumors; and that mammography, because of the early detection, gives women more options for treatment.
The question is whether it saves lives in the long run. The answer to this question is determined by comparisons of breast cancer mortality in screened and unscreened groups, generally through randomized clinical trials.
Last October, a critique of seven clinical trials of mammography done in the 1960s through the 1980s was published in a British medical journal called Lancet. NCI reviewed and seriously considered the issues raised in this critique. This Olsen and Gotzsche critique—they were the authors in Lancet—this critique does not refute the evidence about mammography; thus, a change in NCI's recommendations is not warranted.
Breast cancer incidence rates by stage have shown a decline of later stage disease and larger size tumors, and an increase in smaller, earlier stage tumors and pre-invasive cancers.
Statistical modeling, this shifting of cases to earlier tumors with a better prognosis, predicted a decline in mortality during recent years, accounting for about one-quarter to one-third of the observed decline in breast cancer mortality since 1990. That is, roughly a quarter to a third of the decline in deaths were due to finding the tumors in an earlier stage.
NCI is collaborating with other investigators in an ongoing study of the contribution of treatment advances, screening, and changes in population risk to determine what proportion of the observed decline in mortality may be due to these factors.
Mammography is effective in reducing mortality from breast cancer, but it's not the answer to breast cancer. Like other medical tests, mammography is not perfect. We must maintain a strong research program on breast cancer prevention, early detection, and treatment.
NCI continues to support research to learn more about the value of mammography to supplement the data from clinical trials. We also are sponsoring trials of new technologies for the early detection of breast cancer, and research to assure that these are effective, acceptable, and available to all segments of our population.
Thank you.
Secretary Thompson: Thank you very much, Peter, for your great work and thank you for being here today.
Now, we'll open it up for questions. This gentleman first.
Unidentified Man: First of all, the specifics of the risk of mammography, and what about the Danish study finding that the trials that proving lower mortality rates were flawed? How do you react to that, and do you have any idea of the risk of—the specific risks of mammography?
Dr. Greenwald: Yes. If a woman has multiple mammograms over years, there is a possibility of some lesions being discovered. That has to be investigated. So, that is something that has to be taken into account. But, even given that, we believe the evidence is in favor of doing mammograms.
Now, in terms of this study in Lancet, what these reviewers did, or people that made the critique did, was they summed up the evidence from seven different clinical trials. Over 400,000 women took part in these trials. Unfortunately, those that summed up the evidence excluded 80 percent of the women who were in the trials, and they selectively excluded all of the trials, five of the seven, that showed benefits of mammography. So, we think that that in itself introduced a bias.
The end points of their trials were mortality reduction, and mortality reduction from breast cancer is the appropriate end point.
Dr. Allan: I think your question is critical, given all of the controversy in the media. The Task Force uses very rigorous methods to review the evidence and we looked at all of these seven trials, screening trials for breast cancer—some starting in 1963. And we looked at that evidence, we looked also at the issues that were raised by the Lancet article very specifically. And we found the same flaws. They interpreted the flaws as what we would call fatal flaws, where they just said the studies were not worthwhile and were not reliable.
We did not interpret the flaws as fatal flaws, and we continue to consider the studies of fair quality. The evidence supports a reduction in breast cancer deaths.
Unidentified Man: What about the risks of the mammography?
Dr. Allan: Well, I think as the Secretary said, and Dr. Greenwald said, all screening tests have risks and benefits. The benefits, we've talked about in terms of reducing breast cancer deaths. The risks involve false positives, which is the major issue of a woman having a mammogram that looks positive and it turns out to not be positive.
Over a 10-year period, and remember, women have multiple mammograms. Over a 10-year period, 23 percent of women on average will have one abnormal mammogram. But, most of the abnormal mammograms, 80 to 90 percent, turn out to be false positive.
If you look at that same group of women over 10 years, between 6 to 10 percent may go on to have a biopsy. So, there are risks. There are certainly more diagnostic tests, there are biopsies, there's some surgery. And that's why we on the Task Force very much urge women to discuss the benefits and the risks with their clinician, in terms of when they want to start their first mammogram and how often they want to have mammogram screening.
Secretary Thompson: Most of these studies were done in the 70s and the 80s and up to the 90s, and there are the false positives. The critical problem is that you may have a biopsy. But, the overall benefits for mammography outweigh it and this is what the researchers say from AHRQ, which is the research quality of the department, NCI, and the U.S. Preventive Services Task Force.
So, we think that, overall, these three groups that came together are much more viable, much more authentic, and much more informative than this study that was done based upon these protocols and studies done in the 80s and the 90s.
Mr. Todd Zwillich: Thank you, Mr. Secretary. Todd Zwillich with Reuters Health.
Can you give us an idea as to the timing of the U.S. Preventive Services Task Force recommendations? Have they come out sooner than they would have if the Danish study had never come out? Were they held back to give you time to consider them, since they were so important? And a followup for Dr. Greenwald.
Dr. Allan: Okay. The answer is a global no, and let me explain that. The Task Force was convened in the fall of 1998. The topic of breast cancer screening, which I worked on with two other task force members, we started to look at that topic in 1999, which was before the Lancet article.
It normally takes us often 2 years for the task force to collect the evidence...
[inaudible]
...So we actually started the review in 1999. It normally takes about 2 years for us to do a literature search and review the studies, which we look at as a Task Force. We vote on some drafts and make recommendations. We send those out to multiple groups, organizations that have interest in the topic for peer review, and then we finalize the recommendation.
So, we had a meeting in January and that was our time to vote on our final recommendations for breast cancer screening.
Mr. Todd Zwillich: Dr. Greenwald, is there any indication that you've seen at NCI that the studies that came out in October from Denmark and got a lot of publicity from the media have affected mammography rates among American women, or women internationally for that matter? Or have you gotten any sort of feedback at NCI that women are afraid to get screened or don't want to anymore?
Dr. Greenwald: Not a lot. First of all, we think women are smarter than they're given credit for and they understand that it's a complex issue and that none of the tests are perfect.
In 1993 and in 1997, there were some public issues, controversies about mammography. At those times, there were then surveys of women after the impact. About 75 percent kept doing what they were doing. If they were getting mammograms, they kept having mammograms. If they were not, they did not get them.
The other quarter said that they were interested in the issue and thinking about it. We don't know exactly what they did. But, most women followed through on what they had been doing before.
We do recommend if women have concerns that they talk with their physician about it and, obviously, we recommend mammograms.
Ms. Lauran Neergaard: Lauran Neergaard with AP. Could you go into a little more in depth with us about what it was that was different in your view about the studies then Olsen group found? The flaws that they cited seem to go to the very heart of whether the study is to be believed or not. The differences between all causes of mortality versus breast cancer mortality, things like that?
Dr. Allan: I'd like to mention the two major criticisms from the Lancet article. One was on how people were randomized into a screening group, screening mammography group, or a control group. And the other was the issue of what should the outcome be? Should it be breast cancer death or should it be all-cause mortality?
Again, we looked at, at the Task Force, through our methods that we use for all of our topics, we looked at the issue of randomization. And many of the studies did randomization by what they call cluster randomization. They took people's birth dates as a block and they would invite them to be—to come in to the screening, to be screened or to be in a control group.
And we looked at whether or not, again, that kind of randomization would bias the results. That's the critical question. Does the randomization create unequal groups and, therefore, the results are biased? And we did not find that—although that's not the ideal method of randomization, better to do it just individually, we did not find that sufficiently biased the results to change the outcome of benefit from screening.
The second issue is really the issue of outcomes. In all eight of the studies the outcome measure that they were designed to look at was death from breast cancer. They were not designed to look at outcome measures from all-cause mortality. And so that, in a way, it's an unfair criticism on the part of the Nordic Center to say that the studies are flawed because they did not look at all-cause mortality. They were not designed to do that.
Dr. Greenwald: Right, and many of the causes would be rather absurd to include and consider.
Let me say that the critique really showed the possibility of flaws. They did not provide evidence of the flaws. And even if they had evidence, you would then have to show that it was severe enough that it would change your interpretation of the results.
I can give you two more examples of this sort of thing. The first mammography trial was done in the United States in New York City, begun in 1963. It was called the HIP Study. Olsen and Gotzsche claimed, and showed I think, that after the women were randomized, more in the mammography group were excluded, about 400-500 more, and that that could have meant that more women with prior breast cancer were excluded from the trial, giving a bias in favor of mammography.
We talked to one of the people that worked on the trial and who is a world-class clinical trialist. He said that, in the 19 years of followup from the beginning of that trial, every breast cancer death was examined and, in both the mammography screened group and the non-screened group, all deaths where the breast cancer was diagnosed prior to onset were excluded. They were excluded from both groups, so there was an adjustment that took into account this potential bias.
A second thing that the critique said was in some of the Swedish studies, the women offered mammograms more often were called dying "with" breast cancer and, in the non-screened group, it was more often said they died "of" breast cancer. And that was described as a potential flaw.
The problem is that, when mammography does what it's supposed to do, that's exactly what happens. People with breast cancer don't die of breast cancer.