Public health expert and breast cancer survivor puts new mammography guidelines in perspective

Monday, November 23, 2009; 12:05 PM

Last week, the U.S. Preventive Services Task Force published new mammography guidelines which recommended against routine mammography screening for women in their 40s and less-frequent screening for older women at average risk of developing breast cancer. The guidelines also said there was no evidence to support doctors' teaching women to do breast self-exams. The changes prompted confusion among many patients as well as accusations that the changes amounted to rationing health care. The Post posed some questions to Kay Dickersin, director of the U.S. Cochrane Center and the Center for Clinical Trials at the Johns Hopkins Bloom berg School of Public Health. In 1997, as a member of the National Cancer Advisory Board, Dickerson voted in favor of a similar recommendation for less screening. She also received a diagnosis of breast cancer in 1986.

1. Are the new recommendations well thought out?

Yes. The recommendations are made by a group of doctors who see patients and are also skilled in how to interpret research data. The data from relevant clinical trials were summarized by a team skilled in doing systematic reviews as well as doctors who see patients. This approach has been endorsed by the Institute of Medicine. The recommendations also considered how screening schedules would affect the harms and benefits of screening. The investigators in this case used six different statistical models to see whether having mammograms once a year was better, worse or the same as having a mammogram every other year. All the models showed about the same thing even though they used different assumptions -- screening every other year saves just as many lives but does less harm.

2. Then why are they so controversial?

First, mammogrpahy's benefit in terms of saving lives is very small for women 40-49. That is well known and has been for some time. Some people who do not like the new recommendations say that even if one life is saved, then it is worth whatever a mammogram costs. But cost is not just money, it is also harm to the woman. In a low risk population, such as women 40-49, a mammogram results in many false positives and this is an important potential harm.

A false positive means that a woman is told that the mammogram shows something that makes the doctor concerned that there could be a problem. Because she or he is not sure, the doctor tells the woman she needs to have additional mammograms or an MRI. And some women go on to have a biopsy because the additional images still show a possible problem. Being told that she needs additional images often results in worry for the woman.

The high financial cost of mammograms for younger women is important in that if the mammograms are only marginally helpful, and are also harmful, maybe we should use the money in a different way. For example, many people feel that we would be better off spending money on encouraging older women to get screening mammograms than paying for younger women to have them, because more lives would be saved in older women, with fewer false positives.

In addition, people are concerned that with the change in recommendations, something is being taken away from them. This is compounded by the fact that because of the confusing terminology doctors and researchers use, many people incorrectly believe that mammography prevents breast cancer. I can imagine that the public would wonder why a procedure that prevents a disease would be taken away.

But mammography doesn't prevent breast cancer, it merely detects it. And detecting it earlier doesn't necessarily mean a life is saved or even extended. The idea of early detection is that by detecting a cancer early, we are catching it before it does damage. But we know now that this is not always the case and sometimes we merely detect something that wouldn't have harmed a women anyway. In other cases we are detecting a cancer earlier, but can't change the course of the cancer. In this case women and their families live longer knowing they have cancer, but they don't actually live longer than they would have if the cancer were detected later.

Often I hear a woman say something like "my life was saved by because of a mammogram I got when I was 39 and breast cancer was detected". But we don't know that her life was saved by that mammogram. She might have found the lump herself the next day, in the shower, or the cancer might have been an "in situ" cancer that would not have become invasive and might never have harmed her. It seems to her as if the mammogram "saved her life" but we cannot know that, and if one looks across many women in her age group , we don't see that on average this would be true.

Finally, some communities feel that the available data are not relevant to all populations. For example, the randomized trials involved mainly white women and black women are wondering whether the recommendations also apply to them. While the recommendations are aimed at all women, it is also probably underappreciated that breast cancer appears to be more aggressive in young black women compared to white women. Sooner rather than later we need to do new research to confirm the appropriate screening schedule for black women 40-49.

3. Is mammography the best way to screen for breast cancer? Or is there a better method?

In the trials examined, mammography is compared to "control" (no two trials are exactly the same in the comarison group used; some trials compared mammography to clinical breast exam). Compared to "control" mammography was associated with a 15% mortality benefit (about 29 deaths per 10,000 in the mammography group vs. 31 deaths per 10,000 in the control group the way I read their table). So when compared to "control" it is better.

4. Is the risk of exposure to radiation from a mammogram significant?

While the radiation risk from a single mammogram is low, it would be better to limit the number of exposures one has over a lifetime, if there is no clear advantage of additional mammograms.

5. Some women get breast cancer in their 20s. Why don't we screen everyone?

The lower the risk of breast cancer in a population (and in 20-29 year olds, for example, this is very low), the higher the likelihood of false positives (this is true for all screening tests and diseases, not just breast cancer and mammogrpahy). So the potential harms are high with no real associated benefit.

6. In deciding whether to get a mammogram, how should a woman consider factors like family history and lifestyle?

Family history, especially a mother or sister getting breast cancer before menopause, is an important consideration. Women with a family history may be advised to have mammograms before age 50, depending on which relative had breast cancer and the age at which the relative was diagnosed.

"Lifestyle" isn't typically thought of as a risk factor for breast cancer. One of the most important risk factors is age -- the median age of getting breast cancer in the US is around 61. Other risk factors include early age for starting menstruation and late age at menopause.

7. What's the problem with doing self breast exams?

Breast cancer is often detected by the woman herself or by her partner. Readers shouldn't interpret breast self exam to be the same as touching one's own breasts or a partner's breasts. Keep doing that! Breast self exam, in contrast, is when one uses special breast touching techniques (such as going all around the breast in concentric circles). Breast self exam is encouraged by reminder cards some groups hand out, that show how to do self exams in the shower, and via media messages about how breast self exam saves lives. The trouble with these special techniques and messages is that there is no evidence that breast self exams saves lives. We do know, though, that a breast self exam is associated with increased biopsies and imaging, when women find something they think is suspicious. This is not good.

8. In thinking about one's own health, it's hard to separate fact from opinion, and data from anecdote. What experience do you bring to your understanding of screening?

I have been involved in assessment of recommendations such as these. In 1997, an NIH Consensus Panel made recommendations not so different from those of the USPSTF. Expressing concern with the Consensus Panel's recommendation, the Director of the National Cancer Institute at that time, Rick Klausner, asked the National Cancer Advisory Board (NCAB) to make its own recommendation. As a member of the NCAB, I voted to support the Consensus Panel's recommendation but was outvoted 17-1.

I have also seen treatment from a patient's perspective. In 1986, I found a breast lump that turned out to be cancer. I was 34. Because of my age, I had never had a mammogram. But my cancer did have implications for my sisters and mother, who immediately became "at increased risk" . The current USPSTF recommendations do not apply to those at increased risk of breast cancer, they apply only to those women at average risk.

When I hear some women say, "a mammogram found my breast cancer when I was 39 and that mammogram is the reason I am still alive," I sometimes wonder whether it would make just as compelling a soundbite if I said "I found my own breast cancer without breast self exam or mammography and that's why I am still alive." While it is true that across populations taking early action against a breast cancer diagnosis saves lives, it is not always true that the method of detection can be credited. That is what the review is saying: Except in a few cases, we cannot credit mammography with saving women's lives in the 40-49 age group.

© 2009 The Washington Post Company