Conventional Canadian wisdom suggests that all women should have a screening mammogram starting at age 50 (age 40 in most of the USA). The reason given is that mammograms detect cancers earlier, while it is still possible to treat them. Instead of dying of cancer, you can be a “survivor”. It’s a powerful message preying on the fears of women and their families. October is “breast cancer awareness” month and the promoters of screening mammography were out in full force when I got the telephone call to come in and get my first mammogram. I left their message on my machine to think about it.
As I was thinking about it, I even heard one of these mammography cheerleaders talking on television. As I flipped through the official provincial literature on the topic, she said that anyone who publishes studies that do not support mammograms must be working for an insurance company that doesn’t want to pay for mammograms. She has concluded this because everyone “just knows” mammograms work.
I have long been highly skeptical of what the medical community tells me must be done to me for my own good. It began when I had a doctor tell me I had to consent to a routine episiotomy with the delivery of my first child in 1981. The doctor told me that having him cut open my vagina was much better than letting me tear a jagged mess on my own. Furthermore, by consenting to having him slice me open, I would heal faster, would spare my baby’s head, have less chance of damage to my anus, and result better sex in future for my husband. (He made no mention of my pleasure, that apparently being irrelevant.) As a bonus, the procedure would prevent my uterus from falling down into my vagina as an old woman. Routine episiotomy was the standard of care in North America in those days. I am certain that my doctor sincerely and honestly believed he was doing the right thing for his patients. Unfortunately he was wrong.
Subsequent (and long overdue) studies have since shown that routine episiotomies were worse than just an unfortunate fad. Routine episiotomies actually cause or make worse the problems they are supposed to prevent during delivery. There have been no studies at all to justify statements like preventing a woman’s uterus from falling into her vagina. All those women who had their vaginas sliced open in the name of prevention were basically subjected to a well-meant form of ritual genital mutilation. I took a lot of flack from my doctor after telling him I was unlikely to consent. It was soon obvious he was planning on cutting me anyway, consent or no consent. He was the doctor, I was the mere patient. He just knew what was best for me and my baby. He was not going to put up with any foolishness from me.
At 36 weeks I found a non-North American trained doctor who actually knew how to attend a normal delivery without slicing open the mother's perineum. I had a very nice birth with an intact perineum as a bonus. The other women in the maternity ward all shuffled awkwardly sideways down the hall and spent their time with perineal sprays, and painkillers sitting on inflatable pillow rings. I got to concentrate my energy on fighting with the staff about stupid hospital regulations that were interfering with my choice to breastfeed. I was the only woman in the hospital who had this crazy idea of actually using my breasts to feed my baby.
The whole routine episiotomy debate later broke wide open and it turned out I was right1. Since then, whenever I hear I should submit to something for my own good, especially when my intuition tells me otherwise, my hackles rise. When I am also told my doctors know better than me just because they are doctors, I get spitting angry. I then run to the medical literature to look for myself. I have a Ph.D. in Human Genetics and my specialty is genetic epidemiology. I am not a medical doctor, but I know how to read the medical literature for myself, especially those big epidemiological studies. And my training in statistics and epidemiology far exceeds that of the average physician. I know because I often tutored them in these topics.
Usually, after I read carefully for myself, I find my doctors do indeed know best. For example, I thought I should have a routine screening colonoscopy at age 50 but my doctor said I should instead do the occult blood test and avoid the risks of a colonoscopy2. I read up on it and decided she was right. Nonetheless, my personal experience suggests the more emphatic the medical rhetoric in favor of a routine procedure or a routine screening test, the more likely it is that the evidence is not there to support it. Because the evidence is not there, patients ask a lot of difficult questions.
The medical profession as a whole has found that the best way to silence uncomfortable questions is to bully patients until they stop asking them. I hear a whole lot of rhetoric, hysteria and name calling about routine mammograms directed at patients who question them and at professionals who wonder what we are doing. I have not heard a lot of substantive debate. This is a red flag warning to me.
The whole breast cancer issue first came into my consciousness at the age of 8 when a family member found a small lump. The treatment in those days was a double radical mastectomy. I recall my mother saying there was a surgeon in Montreal, educated in Europe, who was doing lumpectomies instead of double radical mastectomies. My relative decided she would have the biopsy in Saskatchewan and go to Montreal for treatment if the biopsy came back positive. Her surgeon reluctantly agreed. She went under anesthetic having only consented to a biopsy. She woke up with no breasts. The surgeon made the decision for her, against her specific instructions, while she was unconscious. A vibrant beautiful woman in her thirties was left disfigured. She could no longer raise her arms above a 450angle. She was horribly scarred. She suffered the rest of her life with pain and swelling from poor lymph drainage. The surgeon who disfigured her was absolutely convinced he was saving her life and she was being foolish and hysterical. He felt he had an obligation to save her in spite of herself and so he ignored her specific directives and cut off her breasts and most of the tissue under her arms. Yet today, the lumpectomy she wanted is now the procedure of choice with no worse outcome than the double radical mastectomy.
My Ph.D. in Human Genetics included a lot of training in the pros and cons of screening tests. The value of screening tests is measured by four important criteria. These are the risks of taking the test itself, false positive and false negative rates, and treatment options for those who screen positive. A false positive means the screen test finds something might be wrong when in fact everything is fine. A false negative means the test assures us all is well when in fact there is a problem. The risks of taking the test should not be higher than the risks involved in getting the condition the test is designed to find. Finally, if the test does find something, there should be something you can do for the patient.
A good example of a test that is low risk screen test that is considered to have a high false positive rate by many in the field is maternal serum prenatal screening tests for birth defects like anencephaly (where the baby is born without a brain) or genetic disorders like Down Syndrome3. Taking the screening test involves a simple blood test for the mother to measure certain substances in her blood. The levels indicate the baby's health. The test itself is almost no risk because it is just a blood test. The risks are maybe some bruising and a almost zero possibility of an infection at the vein where the blood is drawn. The other risk is the emotional upset that goes with having a positive result and the fear and worry that something is wrong with the baby until follow up tests can be done. The false positive rate on these prenatal screen tests is high, about 5% of all women who take the test. Most women who get a positive result end up going in for an ultrasound and most of these women find out their baby is perfectly normal. The calendar delivery date turns out to be wrong or they discover they are expecting twins. A few women with positive screen tests will then be offered riskier tests, such as amniocentesis, which has a 1/200 chance of causing the death of the baby. Many women who would have been considered high risk because of their age, will have a normal blood screen test and decide to skip the riskier test. Rarely, a result of serum screening will be they discover their baby is not normal. Most women have the prenatal test. If they have the rare misfortune of discovering, on ultrasound and amniocentesis, their baby has, for example, no brain, they will choose to terminate the pregnancy.
However, I know a woman who is an Orthodox Jew who would never have an abortion no matter what was wrong with the baby. She had a false positive result in her fourth pregnancy. Even after the ultrasound was normal she worried and fretted all the way through the pregnancy until the baby was born. The baby was fine. She told her doctor not to do the standard screening blood tests in her fifth pregnancy because she would not have an abortion and she didn’t want to spend half her pregnancy worrying again. She would prefer to not know until the baby was born. She wanted to be able to enjoy the pregnancy while assuming all was well. Her choice was not the same choice as most women. Nonetheless, her choice was an equally valid choice. It was therefore the right choice for her.
Prenatal serum screening is a good screen test with low risk to the mother and baby. Can we say the same about risks and false positives for mammography? Not really. Serum screening in pregnancy is a blood test. When screening mammography is done, it involves a dose of radiation. Granted, it is a small dose, but it is still a dose. How many cancers are we accidentally causing withall these doses of radiation? We simply don’t know. We do know that somewhere in the range of about 5% of the general population is carrying some defect in the repair enzymes that fix radiation damage to the DNA. There is no one real good answer about how many cases of cancer we cause by the radiation required for a mammogram and if it only in those with variant repair enzymes. Proponents of mammography assure us the number of cancers caused by mammograms is exceedingly small and far less than the number of cancers caught early enough to save lives. The best figure is 48 lives saved from breast cancer by mammograms for every life lost due to cancer caused by mammograms4.
Those against mammograms quote figures far lower than that. The only thing everyone agrees on is we are not at zero. There is even some concern about the compression of breasts for mammography being not just uncomfortable to painful, but maybe even dangerous. The evidence is theoretical and proper studies have not been done to prove or disprove this but, in theory, compressing the breasts may burst small blood vessels around small cancers and cause them to grow and spread5. If we had a clear benefit and we knew exactly what the risks were, we could decide the test to be worth all these risks. Right now, we just don’t know. The studies have simply not been done.
We do know the false positive rate for screening mammographyis very high. A recent study from the New England Journal of Medicine6 concluded that one out of ten screening mammograms gives a false positive. This means if you start having an annual screening mammogram at age 40, you will have a 65% chance that you will to go through the whole mess of being told they found something that isn’t there by the time you are 50. What about the follow up required for a false positive? In serum screening in pregnancy the next test is ultrasound which almost everyone agrees is harmless to mother and baby. What about mammography? Well first there are more x-rays creating more risk of radiation induced cancer. Needle biopsy may be spreading cancers that otherwise would not have spread. Surgery is a common result of a false positive and that is also not harmless and low risk, especially if the woman has to have a general anesthetic. I have not even yet mentioned the enormous psychological toll a positive screen can cost a woman and her family. When was the last time someone talking in breast cancer awareness month on television discussed the false positive rate and the risks associated with it?8
The second measure of a good screening test is the rate for “false negative”. For a cancer screening test, a “false negative” means the test says there is no cancer when, in fact, cancer is present. There is a blood test called prostate specific antigen (PSA). In many forms of prostate cancer, this antigen is elevated and measuring the level can be used to see how well a treatment is working to stop the cancer or test if the cancer has come back. Because of this, PSA was once suggested as a possible screening test for prostate cancer. However, this test has a high false negative rate meaning it often suggests no cancer is present when cancer actually is. If a man skips the unpleasant digital exam (which is a much better screening test) because the blood test is negative, the blood test has made things worse for the man not better. This blood test is no longer recommended as a routine screen for prostate cancer except when used with other screening tools, especially the finger-you-know-where prostate exam7.
For breast cancer, there are many studies, some good, some not good, on false negatives and the figures are astonishingly bad in light of all the hype for mammography. Overall, up to 10% of women who have confirmed breast cancer had a previous screening mammogram and were told everything was fine. The really good studies using large groups of women and following these women over a long time period of time have shown that the false negative rate for mammograms for women between the age of 40 to 50 is so high (>30%) it’s just not worth the test.
That’s why many countries, like Canada, will simply not pay for screening mammograms for women with no risk factors until they are 50 years old. After age 50, screening mammograms do show benefits in the form of reduction of death rates for a specified number of years after age 50. The false negative rate for mammograms is tied to how x-ray opaque the breasts are and the older one is the more translucent the breasts are. Most professionals agree on offering screening mammography for all women 50 and over. There’s nothing really magical about 50. It’s just a nice round number when most women have started menopause and their breasts have changed to translucent enough to actually get a mammogram a radiologist can see something in, if it's there to be seen.
The third and most important measure of the value of a screening test is what kind of treatment can be offered. I have my doubts about our cancer treatments for breast cancer. There was a study in Denmark done on women who have died of other things like car accidents. It was a small study but they found 19% of women had small breast tumors. Yet 19% of women in Denmark don’t die of breast cancer9. Obviously, some of these cancers simply lie quietly or the body cures itself. These are called ductal in situ carcinoma or DCIS. Since we don’t know which ones will spread and kill, all of them are treated. I really wonder about our “cure rates”.
How many of the women “cured” of their early cancer would have never developed the disease if no one had discovered they had cancer with a routine screening mammogram? How many women have gone through surgery, radiation, and chemotherapy but didn’t really need it? How many women went through all that and it did them no good and they then died of breast cancer anyway? It seems to me that even if mammograms find early cancer, the odds are better than 50:50 it would have just gone away by itself no matter what women let the doctors do to their bodies. The Danish study is a small one and needs to be confirmed by a very large proper study but I can’t just ignore that data. If we are catching breast tumors earlier, when they are treatable, using screening mammography, then logic dictates we should also be seeing more small earlier tumors and less bigger later tumors. Unfortunately we aren’t. Why not?
Sometimes the treatment for the condition is so horrific it’s just not worth knowing. When a prostate exam comes back as positive and cancer is found, a man then has a series of choices to make. A lot of prostate cancers will simply lie dormant for decades and never leave the prostate and never cause death. The treatment of removing the prostate often leaves a man impotent and incontinent. Should a 60 year old man, for example, choose to have his prostate removed and have a high risk of ending up unable to enjoy sex and having to wear diapers the rest of his life when the cancer might just have sat there for twenty years? Many men would prefer to do nothing and just have their doctors monitor the prostate. They would only choose to have the prostate removed if the cancer seems to be spreading. Other men might find they worry so much about the cancer they prefer knowing the prostate is gone even if means being incontinent and impotent. Both choices are considered valid choices for men to make.
And this brings me to what I find most disturbing about the entire mammography debate. I never hear talk of choices for women. The promoters of the routine screening mammogram only discuss those cases of women who had a cancer found early, had it treated and survived for a certain amount of time afterward, five to ten years being the common time period looked at. A woman may suffer the rest of her life from the treatment of her cancer but she is nevertheless counted as a success. She may have undergone treatment for a cancer that wouldn’t have killed her anyway but she is still counted as a success. The women who had her chemo, radiation, surgery and lived five years only to have the cancer reoccur and kill her in year six also gets counted as a survivor for the five year studies. She survived. No one asked her if her quality of life made survival worth it. She made the five year mark even if the cancer finally killed her. Was she really better off for having gone through everything but dying of cancer anyway? Maybe the treatment did give her five extra years. But do we really know for sure if she would not lived for five years in blissful ignorance and then died at year six if she hadn’t gone through all that treatment?
We talk about breast cancer as a war and those who deal with it as heroes fighting the good fight. Cancer is the only disease I know of where we are expected to “fight”. Well I don’t want to fight if they find it in me. I honestly know that I personally would prefer to live in cowardly ignorance until the end. I don’t want to spend my time worrying about whether or not I have breast cancer. I have seen enough women go through long, complex and difficult surgeries, chemotherapies, radiation treatments and been left horribly disfigured, have dramatically reduced quality of life and still die of cancer.
I think it is a valid choice for me to say I would prefer to have them find the cancer very late so it kills me fast in terms of when I find out, instead of a long slow death dragged out over many years living as a "fighting" cancer "survivor". Yet, when I say this, people react with horror. Why? Is it not my personal choice to make? Come back and talk to me when the equation has changed because some new screen test is on the scene or some new break through treatment is available and I might change my mind. I did undergo the occult blood screening for colon cancer so that shows I am not just a hysterical female with my head in the sand about cancer.
When I turned fifty, my doctor assumed I would go for a routine screening mammogram. I declined. She was very upset with me although she did respect my choice. My reasons are complex. I am at very low risk for cancer. I started menstruating late. I had early pregnancies, prolonged breastfeeding and early menopause. I have no history of radiation therapy and I have not taken birth control pills or estrogen replacement therapy. In short, I have no risk factors. Weighed against the false positive and false negative rate of the test, I figure I’m better off not taking it. While there are many women who do not have risk factors who then go on to develop breast cancer, I take comfort in my lack of risk factors. My friend, armed with the same knowledge, decided instead to have a mammogram. Her mother had breast cancer at a young age and she had several risk factors. I think we both made the right decision because we both made the decision in an open reasoned way weighing all the positives and negatives. No one is calling her crazy so why do I get that treatment?
Let’s face it. Mammography is a poor test. The false positive and false negative rates are too high. The treatment when cancer is found is drastic and the effectiveness of the treatment is uncertain for a lot of women. The risk of mammography itself is not known but it is not zero. So why aren't we working on a better screen test? Instead of pushing so hard to have a poor test performed on everyone, at huge cost to society, I would like to see some of that money and energy going to more research on improving the screening test for breast cancer. We need to think outside the box and we aren’t doing it.
Suppose we could develop a mammogram that could include the watchful waiting component? We find a small tumor (DCIS) and do nothing but retest in a few months, or a few years, and only go in and do biopsies and surgery if the tumour is actually growing. Standard mammograms can’t do this. We could also start prescreening women for the genetic variants that make them less able to repair their DNA after radiation damage from a mammogram. In theory that might greatly reduce breast cancers caused by the test we use to find it.
Maybe we could develop entirely new types of mammograms that are sensitive enough that we could treat breast cancer like we currently do cervical cancers. With a pap smear, you screen test women for cervical cancer. (And you don’t irradiate them doing it.) If you have a suspicious result, the doctor removes the suspicious cells and only those cells and a few around them. Why aren't we working on techniques to find abnormalities in the breast at a very early stage of only one or two millimeters before they metatasize? We could destroy just these suspicious cells using ablation techniques inside the breast where the suspicious cells are. This would not change the breast itself nor would we require chemotherapy and radiation but we could, in theory, stop the cancer. Okay we might ablate a lot of tiny harmless lumps but who cares? As far as I know, no one is doing this kind of research even though the technology already exists to kill tiny clusters of suspicious cells in the breast if we can find them.
It’s very scary for women to hear that a mammogram likely won’t find anything even if it's there and if it does, it is likely not to be breast cancer but we will subject her to all kinds of nasty risky stuff anyway. And if it is cancer, we will subject her to all kinds of treatment she might not need. The reason why it is so scary is we simply don’t have any other test to offer women instead. We all know women who died of breast cancer between ages 40-50 and surely a bad test is better than no test, right? We have to be able to DO something about breast cancer.
This seems to be why in the USA the consensus is, we ignore all the evidence and recommend the screening mammogram for 40-50 year olds. Otherwise we feel so helpless and vulnerable. Better to stick our fingers in our ears and chant “Pink! Pink! Pink!” loudly enough to drown out annoying scientific facts than to face our own fear of being helpless against a terrible disease.
Maybe it’s even more pernicious than that. I sometimes wonder if this whole mammography debate, or lack thereof, is just one more example of the callous, paternalistic attitude medicine has had towards women’s bodies from the days when doctors refused to wash their hands and spread childbed fever. I don’t know for certain. I think doctors mean well. I just really have a lot of doubts about mammograms. The only thing I do know is no one seems to be asking these very important questions loud and clear. And if I, as someone steeped in screen testing, epidemiology and genetic counseling, am generally assumed to be incapable of making an informed choice to not have a mammogram, then is any woman capable of doing so? Is this whole professional horror at my personal and well-informed choice possibly also tinged with a patriarchal need to dominate poor little female me and protect me from my own ridiculous hysteria? Medicine may now have 50% female doctors but it’s still deeply patriarchal and highly authoritarian. This is, after all, the very same system that brought us women routine episiotomies only three decades ago.
I think it is time to carefully assess our attitudes about this disease and our promotion of routine screening mammography, without informed choice, for all women. If a woman chooses to have a mammogram, knowing full well the risks and benefits of the test, I fully support her choice. By the same token, an informed choice to not have the test is an equally valid choice I think we should also be supporting. The solution to breast cancer is not to wear pink and go on a run through parks while shouting “All woman should have mammograms!” and accusing those who don’t agree with that of being uncaring, cheap or stupid. The solution is to gather in the parks, and laboratories, and demand we all tackle these hard questions about mammography and breast cancer treatment head on and truthfully.
We need answers without fearful coercion or prejudice against those who ask the questions, or only answering the questioners with preconceived notions and old prejudices. The cure is out there and it is only waiting for us to stop shouting at each other about getting routine mammograms and go and find a real solution to this terrible scourge. Maybe then the dying from breast cancer will finally begin to stop.
- Hartmann et al, “Outcomes of Routine Episiotomy, A Systematic Review”, JAMA, 305:2037, 2005
- Mandel JS et al, NEJM. Reducing mortality from colorectal cancer by screening for fecal occult blood. Minnesota Colon Cancer Control Study. 993 May 13;328(19):1365-71
- Chodirker, BN, Canadian Journal of Diagnosis, July 2001, pg 58
- Watmough, Quan&Aspden, “Breast compression: a preliminary study.” J Biomed Engin 15:121, 1993
- Elmore JG, et al, “Ten-year risk of false positive screening mammograms and clinical breast examinations..” NEJM. 1998;33:1089-1146.
- Thomas IM et al, “Prevalence of prostate cancer among men with a prostate-specific antigen level < or =4.0 ng per milliliter.” N Engl J Med. 2004 May 27;350(22):2239-46.
- Lidbrink et al, “Neglected aspects of false positive findings of mammography in breast cancer screening: analysis of false positive cases from the Stockholm trial” BMJ, 312:273, 1996
- Nielsen et al, (1987) Breast cancer and atypia among young and middle aged women: a study of 110 medicolegal autopsies. Br J Cancer 56:814–819