Back in 1977, the first mammography screening study showed that women who had been screened had a lower mortality than those that had not.  This led us to conclude that early detection was the key to curing all breast cancers.  It was a reasonable hypothesis at a time when it was thought that all breast cancers were the same and that they slowly grew to a certain size, then “got out” and metastasized elsewhere in the body, leading to death.  Since then, there have been several studies of mammography on different age groups with  different screening schedules.  Even the most encouraging results showed  only a 26% decrease in mortality regardless of whether the screening was with the old style mammograms of the 80’s or the fancy new digital mammograms of today. 

This week, we heard about the 25-year follow-up of a study done in Canada where women aged 40-59 were randomized to mammogram and physical breast exams or strictly physical breast exams for five years. The data showed no difference in mortality between the two groups. This suggests that a clinical breast exam is the same as mammography in asymptomatic Canadian women 40-59 years old.  The women who had mammograms had smaller tumors detected, but that did not shift the mortality.

What this tells us is not that mammography is worthless but rather, that the hypothesis of early detection may need reevaluation. Over the last 37 years, we have learned that not all breast cancers are the same. There are at least five different kinds based on their molecular biology!  Some of them are very slow-growing and others are very aggressive.  Some are sensitive to treatment and others are not.  Some can be found at an early stage on mammography and some have spread by the time they are visible.  The good news is that we now have targeted treatments for some of these tumors (HER2/neu positive and hormone positive) that significantly impact mortality.  We need more effective treatments for triple negative breast cancers and metastatic disease.

And what is the cost of mammography screening?  There are false negatives, which result in normal mammograms when women actually have cancer, and false positives in which mammograms thought to be abnormal result in biopsies to double check.  There is also the over-diagnosis of lesions (22% in this study) that never would have become clinical cancers in the first place, not to mention the cumulative risk of yearly radiation.

So, should we throw out mammography?  No!  It is a great diagnostic tool for women with lumps and an important diagnostic option for high-risk women as well. However, it is time that we recognize that the premise that early detection is the only factor between life and death may be flawed.


Think about it.  Prior to the 16th century, people thought the earth was flat because that is how it looked. It was counterintuitive to think that it was round and public opinion didn’t change until it was proven. In the same way, we previously hypothesized that all breast cancers could be detected “early” through screening and that early detection was the answer to curing the disease.  Like TSA screening in the airport, having an annual mammogram makes us feel like we are doing something to prevent catastrophe. But just as TSA screening does not guarantee there will be no terrorist attack, this study is another piece of evidence that mammographic screening is no guarantee that you won’t die from breast cancer. In both situations, refocusing investments from early detection to prevention would save more lives in the long run.

Tagged with →  
Share →

20 Responses to How is Mammographic Screening Like the TSA?

  1. Carmen says:

    Since the 70s the value of mammograms have been questioned based on flawed studies showing a benefit and knowledge of their severe risks (source: “The Mammogram Myth” by Rolf Hefti). Nothing has changed on what the real facts are. Nothing has changed about the dismissal and denial of anti-mammogram data. And nothing has changed on the endless pro-mammogram propaganda and shoddy, misleading studies in support of it. Women have been misguided for decades about the value of mammography.

  2. Jeannie Dowd says:

    With fibrocystic breasts, self-exams were difficult. I had two exams by docs who felt nothing suspicious. It was only the mammogram that found my lobular cancer which was stage IIB in three nodes. That mammogram saved my life.

  3. Interesting take on the study. I appreciate how you stress the varieties of these cancers.

  4. Bev Wills says:

    It’s so refreshing to read Dr Love’s accurate, non-jargon summaries of the status quo. Thanks.
    A question re the Canadian study: Was it possible to associate mortality with apparently more aggressive tumors, as guessed from clinical or mammographic characteristics? If the older 26% stud(ies) have any statistical (unbiased?) significance, I’d rather be among the 26% — or even 10%. I know this has to be balanced with overdiagnoses; how do you measure the psychological pain of an overdiagnosis!
    Many thanks.

  5. kathy says:

    I started having mammograms at age 30 due to fibrocystic breasts. All negative, then at age 67 another negative mammogram, two sonograms showing a tiny “benign lesion”, followed, at my request, by an MRI showing the lesion and a 3.5 centimeter stage 11B tumor. Obviously I no longer trust mammograms or sonograms. Lesson learned, we must be our own health advocates. If something feels not quite right ask for further testing. All the radiologists at a major women’s hospital reviewed the mammograms and sonograms as did my breast specialist, and the tumor was not visible. It’s been three years and I am doing great. My doctor said I saved my own life.

  6. Dana says:

    Without a mammogram, my Stage III breast cancer would not have been detected 6 years ago at the age of 70. So what if there are false positives. I can live with that. It is doubtful if I could have lived without that particular mammogram. The same conversation is now ongoing with PSA tests for men which is an easier and cheaper test. I have a friend in his early sixties who is dying of prostate cancer. His doctor told him he didn’t need PSA tests. I fear that this avoidance of screening tests claiming that some people will be inconvenienced by false positives will result in yet more deaths. Note that the breast cancer study only looks at the mortality data. Quality of life is not even considered.

  7. Dana says:

    Without a mammogram, my Stage IIb breast cancer would not have been detected 6 years ago at the age of 70. So what if there are false positives. I can live with that. It is doubtful if I could have lived without that particular mammogram. The same conversation is now ongoing with PSA tests for men which is an easier and cheaper test. I have a friend in his early sixties who is dying of prostate cancer. His doctor told him he didn’t need PSA tests. I fear that this avoidance of screening tests claiming that some people will be inconvenienced by false positives will result in yet more deaths. Note that the breast cancer study only looks at the mortality data. Quality of life is not even considered.

  8. Mary says:

    I saved my own life, as well. Neither lump showed up on regular mammograms or clinical or self exam due to dense tissue. It was a nipple, slightly more inverted than usual that, finally, told the tale of what turned out to be an aggressive Her 2 + tumor. Also, if the lump is small,and the tissue dense mammography won’t, necessarily pick it up, even if it is palpable. This from personal experience. Ah yes the old analogy of the polar bear in a snow storm! I personally question the benefit of mammogram, in my case and when I questioned my radiologist, he says it would pick up DCIS and that is why I needed to continue to have them.. Doesn’t all ductal cancer start out as DCIS? Well DCIS is not a diagnosis I have had and I’ve gone around with BC twice. Actually, I am forced to now rely on high risk MRI and that was denied by insurance,this last time. I, however, did get that reversed on appeal. This without the help of my PCP who did not want to get involved. Boy, do I sound jaded. Yea, I am.

  9. Linda says:

    My first cancer at age 37 did not show up on a mammogram, but I knew there was a lump not previously there, so I had a biopsy at my request. My second cancer did show up on a mammogram, even though I did not feel much of a lump at all. So my advice is to pay close attention to your body and have the mammogram. Do everything you can to remain diligent, informed, and proactive about your health.

  10. Patricia says:

    It is very confusing. I was diagnosed with Breast Cancer in 2000. Had small lump that was NOT invaded but the cancer was described as like tiny grains of sand. Anyway, Mastectomy, Chemo. Radiation and Reconstruction via Tram Flap and reduction of other breast and many years later. I am being told via Mammogram and ultrasound that there is a “suspicious” mass in the good breast so an MRI was recommended and done. THE ABSOLUTE WORST TEST I HAVE EVER UNDERGONE AND WILL NEVER DO AGAIN. Inconclusive. So they are now saying Mammogram every 6 months if I am unwilling to undergo another MRI? At 71 I wonder if any of it is worth it?

  11. Teresa Masters says:

    My annual mammograms were clear, at times I was backed up by hand held ultrasound, and sent home for another year. Early un the year of my 80th birthday, after both mammogram and hand held ultrasound, I was told they were “watching” something they felt was nothing, but would I return in 6 months. I did and by that time I was aware of Automated Whole Breast Ultrasound. I asked and was told “I’m not sure it is necessary.” Both mammogram and ultrasound were done with ultimately the head of the department and two other radiologists disagreeing with the technician. I was told they wanted to do a fine needle aspiration right then. It could not be scheduled and was put off until the middle of the next week and I was sent home.Told a friend who arranged a Monday, (this was Friday) appointment for a SonoCine 100 miles North. I brought along the CD of my recent tests. The areas under suspicion were simply fatty tissue as the fine needle aspiration would have shown, but far deeper and below there was an invasive ductal carcinoma. Subsequently I learned I had Dense Breast Tissue, had never known that, and the failure rate of mammography with this. Had I simply had the fine needle aspiration, I would have had to wait another year for the next mammogram and the cancer , well who knows what the cancer would have done until it was detected. I selected my surgeon after researching three, because of her focus on Breast Sparing as well as her research and teaching. A lumpectomy and sentinel node removal were preformed, followed by again my choice of Brachytherapy Radiation. 2.5 years later I am NED. I will never go back to that University based Imaging Center, my choice.

  12. Laurie says:

    Well, without mammogram, I wouldn’t have found my first cancer either. I’m kinda figuring that taking out a 5 mm cancer is a lot better than waiting til you can feel the obvious lump and may have spread. Mine was an aggressive type. Had lumpectomy and radiation. 7 yrs later, cancer #2, only found because had mammogram done. ‘Nother lumpectomy. 3 yrs later, cancer #3 (and 4—two spots at once). So, it kept recurring (hence double mastectomy), but because they were always small, they didn’t have much of a chance to spread, so I escaped without chemo. PS—I had dense breasts and lots of cysts, but they still managed to see the microcalcification that clued them in. PPS—I was 41 first time. So…………………….I’ll take those false positives!

  13. Ginger says:

    My aggressive invasive ductal carcinoma was discovered by my annual mammogram. After a lumpectomy and radiation I was so happy to be here five years later for the birth of my first grandchild!

  14. Diane says:

    I was diagnosed via mammogram with Stage 2A high-grade her2/neu positive invasive ductal carcinoma in 1999 at the age of 46. I had no lumps. I had a mastectomy and 6 months of chemo. I have not had a recurrence. At the time of my diagnosis, I had a 4-year old daughter and a daughter just about to start college and wondered how they would manage without me. I’m blessed to say that I have been able to see the older daughter get her master’s degree and the younger graduate high school. I have 2 beautiful grandsons. Where would I be without early detection? Because my cancer was aggressive, I’m sure I would not be here now if I had waited until the cancer was big enough to detect via breast exam.Thank goodness for mammography!

  15. Marguerite Gunn says:

    I have dense breasts. I had fibrocystic breasts. Annual mammograms for 10 years. Then the government said that after 60 I only needed one every other year. I was going to cancel one that was already scheduled but somehow I had it anyway. It showed DCIS. I had a lumpectomy. I asked the surgeon to take out a large chunk to make sure the margins were clear-at my age that would mean no radiation necessary. But some invasive ductal was found in the margins, fairly distant from the biopsy, and also 2.3 cm of LCIS which was pleomorphic (aggressive) and signaled a 40% chance of having it in the other breast. I had bilateral partial mastectomies with a sentinel node biopsy (neg.) and a lift- I have adorable revised C cup breasts and a lot of tissue went to pathology. A little more invasive was found. Margins were clean. Smaller breasts made 33 radiation treatments more precise. I am so grateful that I had that mammogram and didn’t wait a year. The cancer was caught at stage 1a. If I had waited a year, due to the aggressive nature of the abnormal cells, I would have had a higher stage necessitating chemotherapy. I am a fan of annual mammograms and early diagnosis. So what if 22% are false pos. or neg. The other 78% of us are diagnosed and treated at the earliest possible stage. Should we throw out screenings or surgery for DCIS because of a few, or keep them and help the many with the earliest diagnosis possible? I just gave you my answer!!

  16. Luz says:

    Reading all these comments only tells me one thing: we still don’t have a handle on it! Just as breast cancer treatment is individualized, so should screening be. I also had dense breast tissue but no other screening was ever recommended. Too bad I didn’t know then what I know now. If I had, I would have insisted on further screening. In my case, I had no family history of breast cancer (or any other cancer for that matter). I had mammograms every two years. So in May of 2010, I had a regularly scheduled mammogram, and guess what??? There was a large tumor detected. Then an ultrasound revealed another mass deep in the chest wall. Then an MRI revealed more tumors, all on the right breast. I had three tumors biopsied, which showed two were cancerous and one was benign. So my cancer was already metastasized on initial diagnosis! After mastectomy and the CEA blood levels still high, a PET scan was ordered, where it showed that the cancer had spread to the pelvic bones. I did not respond to hormonal therapies. I have been on chemotherapy (Xeloda, then Navelbine, and now Doxil). I did have radiation to my left hip and then months later to my right hip. I am thankful that I am currently stable; however, I will always be on some type of chemo. When the chemo stops working, then on to the next one. Eventually, I know I will run out of options.Try living with that over your head! In the meantime, I have made many dietary changes – like no dairy, eating more plant-based and raw foods, to name a few – and incorporated regular exercise. So I repeat: We really don’t have a handle on this!!!

  17. Christine says:

    I think most, not all, commenters missed the points that Dr. Love was making (and that many similar writers have made in recent years) and that is that regular annual mammograms for the vast majority of women do not cut death rates from breast cancer (isn’t that supposed to be the purpose?) and instead have many drawbacks. Dr. Love notes that there are at least five different kinds of breast cancer based on their molecular biology. These either respond to current treatments based on their biology or they don’t—irregardless of size or spread. A particular cancer either responds to currently available treatments or it doesn’t. A cancer can and will be deadly or will not be deadly based not on its size when it’s found—say whether it’s found in January of one year or January of the next year, but rather based on its biology. If a particular breast cancer proves deadly, the bottom line is how long the patient knew she had the cancer. If it was found “early” by way of mammogram, maybe she will think she lived longer than if it had not been found until it was a large palpable mass. Either way it proved deadly, because it did not respond to currently available treatments and it spread and killed her. Some commenters have flippantly dismissed (“So what?”} the numerous false-positives that regular annual mammograms result in. This is callous and ignorant, and Dr. Love addresses this issue by mentioning the “cumulative risk of yearly radiation”—in other words, we females are increasing our chances of cancer by bombarding our bodies with yearly screening mammograms and, for many of us, being called back every time for “comprehensive” (even higher-dose) mammograms. And we are faithfully submitting to yearly mammograms (and all that radiation) when there is NO evidence that mammograms saves lives.

  18. Connie Selby says:

    This professor of radiology clearly details why the Canadian study is scewed and unreliable.
    It’s Just Wrong
    Dr. Paula Gordon
    It’s just wrong that a poorly designed and poorly executed study done decades
    ago in Canada could potentially cost thousands of American and Canadian
    women’s lives. It’s a national embarrassment, and particularly to Canadian breast
    radiologists, to see the coverage of this work. It will dissuade women from having
    mammograms, and women will die unnecessarily.
    This is not new research. It has been known for decades that this trial was fatally
    flawed. It was because of problems with the Canadian National Breast Screening
    Study (NBSS), that the World Health Organization’s International Agency for
    Research on Cancer conference in 2002 excluded the NBSS from analysis of the
    impact of early detection using screening mammography, on mortality from
    breast cancer. And yet here we go again, twelve years later: reading about this
    discredited trial on the front page.
    Women who read the New York Times, the Vancouver Sun, and any other
    newspaper that gave front-page coverage to the update, need to know the truth
    about the NBSS, and why it should not influence decisions about screening
    mammography, either for an individual or on a policy level. Keep in mind that
    some of the technology used in the trial was in use before women in their forties
    today were born.
    Journalists referred to the study as a randomized trial. Even if it were, and the
    randomization process was blinded, and even if a significant percentage of the
    control group hadn’t had mammograms outside the trial, it would still not have
    shown the true capability of mammography to save lives, because of the poor
    quality of the mammography.
    There were several serious problems with the NBSS, and experts from around
    the world were quick to point these out right from the start. But the trial
    investigators forged ahead. In my view, the biggest flaw of the NBSS was
    corruption of the randomization process. A randomized trial is supposed to have
    2 groups: one having the test and one not. And the 2 groups are supposed to be
    equal in every other respect. The other trials of screening mammography
    elsewhere in the world were done that way, but not the Canadian trial. In other
    trials, women were assigned to either the mammogram or control group on a
    population basis. Women in the control group did not know there was a trial
    going on, and did not know that they were in the control group. They just lived
    their lives.
    In Canada, the NBSS advertised, and asked women to volunteer to be in the trial.
    When they came forward as volunteers, but before they were assigned to the
    control or study group, they had a thorough breast exam by an experienced
    nurse or a doctor. Now, imagine that you are the nurse working in the clinic, and
    you examine a woman who has come to volunteer to participate in the study. You
    feel a lump in her breast (a possible cancer), and lumps in her armpit (possible
    spread of cancer to the lymph nodes). Are you going to put this woman in the
    control group? The authors of the study deny that this took place, but in fact, one
    of the center coordinators was fired for doing just that. The randomization
    process was supposed to be done by a method where the result of the physical
    exam was not known, but the names were then written on open lists, making it
    possible for the staff to leave blank lines, into which they could then write the
    names of the women with lumps. So it’s not surprising that there were more
    advanced cancers, including many with feelable lymph nodes in the armpits, in
    the mammogram group. No wonder more women died in the mammogram group
    than in the control group. Not because mammograms don’t work; but because
    more women with cancer were PUT into the mammogram group. Of women who
    died within seven years because of advanced cancer, eight were in the
    mammogram group, and only one in the control group. This wouldn’t have
    happened if the randomization process had been blinded.
    When the randomization process was challenged, a review was conducted by
    Drs. John Bailar and Brian MacMahon. But the review was “essentially worthless,
    because the reviewers did not interview the women who performed the
    randomization to determine whether the process had been compromised.” They
    said, “we did not interview any of the field staff because any who participated in
    active subversion would have been unlikely to admit it to us.”
    And now imagine that you are a woman who has volunteered, hoping to be put
    into the mammogram group. But you are put into the control group, and you are
    disappointed. Seventeen percent of women aged 50-59, and 26% of the women
    aged 40-49 in the control group had mammograms, outside the trial! This is
    called “contamination.” And some of them would have had cancer detected and
    treated, so they did not die. But they were still counted in the control group, so
    when the comparison was made later, of how many women had died of breast
    cancer in the mammogram and control groups, this contamination added to the
    appearance that the death rate was similar between the 2 groups.
    Do you remember the exposé by the TV program 20/20, where they showed that
    a high proportion of mammograms from a clinic were of poor quality? That
    program was a wake-up call and lead to the American College of Radiology and
    Canadian Association of Radiologists starting mammography accreditation
    programs and subsequently to the Mammography Quality Standards Act (MQSA)
    in the United States. The mammography done in the NBSS was not dissimilar to
    those poor-quality mammograms. As a breast radiologist, I know that the
    mammography unit used in the Vancouver Centre of the NBSS was NOT stateof-
    the-art, even for the 1980’s; it was already 10 years old, at the beginning of the
    trial, and that machine was used for the first 4 years! More than half the
    mammograms done in years 1-4 were judged to be poor or completely
    unacceptable. Two successive advisors resigned over the quality of the
    When a woman was diagnosed with cancer on a mammogram in the NBSS, that
    cancer was visible in retrospect, 1-5 years earlier in 25% of the cases. When a
    woman was diagnosed with cancer because she had a lump, some time after a
    normal mammogram, the cancer was actually visible, but missed on the
    mammogram (false negative) 42% of the time. This false negative rate, which is
    a measure of the quality of the mammography and the interpretations by the
    radiologists, was worse than in studies done in the 1960’s and 1970’s. And this
    poor-quality was seen in more than ¾ of the centres. The physicist for the NBSS
    acknowledged that the mammography quality was far below the current state of
    the art.
    In the NBSS, the average size of the cancers detected was 19mm. Poor quality
    mammography is the explanation, and finding larger cancers does not lead to a
    decrease in the death rate. Compare that to our screening program in British
    Columbia, where 65% of cancers detected at screening are ? 15mm, and 76%
    are node negative. A publication from our program in 2007 showed mortality
    reduction of 40% in women actually being screened. Our interval cancer rate is
    between 5 and 7%.
    Some cancers in the NBSS were missed because the woman’s breast was not
    positioned properly in the mammogram machine, so the cancer was not included.
    It was standard at the time, to compress the breast at an angle that allowed
    inclusion of the breast tissue near the armpit. But for the first several years of the
    NBSS, many technologists were not trained to do this, and it was not done.
    And even among the women with suspicious findings on the mammogram, when
    a biopsy was recommended, it was NOT done 25% of the time. So even some of
    the early cancers detected didn’t get appropriate treatment: another reason this
    trial didn’t show decreased mortality among women having mammograms.
    The American (November 2009) and Canadian (November 2011) Task Forces
    on Preventative Health were unduly influenced by this trial, by including it in their
    meta-analysis. They then balanced what they considered the benefits and harms
    of mammography, and concluded that it should be done less frequently. But they
    understated the benefits: They considered only mortality reduction as a benefit,
    and even that was underestimated by using the NBSS results. They gave no
    weight to the benefit to women, of being able to have a lumpectomy instead of a
    mastectomy. And they gave no weight to the benefit of being able to avoid big
    armpit surgery, and have a less invasive “sentinel node biopsy.”
    And the Task Forces overstated the harms: They considered it a significant harm
    to make women nervous after being called back from screening to have
    sometimes just one more picture to ensure there was no cancer. I can attest to
    how hard it is on women to wait for the necessary tests to be done. For some
    women, it’s torture. But most women would choose that anxiety over being
    denied access to screening.
    The task forces cited unnecessary biopsies done to rule out cancer, as harm. Yet
    the vast majority of these biopsies are needle biopsies done with local anesthetic.
    These are really not bad. I ask every patient after I’ve done their needle biopsy,
    whether it was as bad as they expected. The answer is always: NO. The best
    answer I’ve had was from a patient who told me, “I have shoes that are more
    uncomfortable than this test.” So to save some women the anxiety of being called
    back, or a short period of discomfort, the Task Forces recommend denying
    access to mammography, and thereby let other women die unnecessarily.
    The concept of over-diagnosis (the possibility that mammography will discover a
    cancer that would never kill the patient, but lead to unnecessary treatment) is real,
    but the rate of over-diagnosis is unclear. The NBSS says it is 22%, but most
    experts agree that it is between 1 and 10%, probably in the lower part of that
    range. And it’s an unavoidable component of screening. But isn’t it reasonable
    that a few women have more therapy than they need, than for many more
    women to die unnecessarily? These women do have cancer, after all.
    Researchers are working to find a way to discriminate which cancers need less
    aggressive therapy, but we do not yet know, and so until we do, some women
    will have more therapy than they might need, to save many more lives.
    The authors of this trial, and the journalists who report on it uncritically will have
    blood on their hands. Not right away; because all women diagnosed with breast
    cancer will receive state-of-the-art treatment, even when it is large at the time of
    diagnosis. It will take many years to see a change in the death rate. But if women
    heed the bad advice of these authors, we will see, in the next 5 years, the
    average size of breast cancers, and the rate of spread to the lymph nodes
    I will continue to have annual mammograms.
    Dr. Paula Gordon is a Clinical Professor at the University of British Columbia,
    and a member of the Order of British Columbia

    AND here is an article just published in the NYT:

    Published in New York Times
    To the Editor:
    Congratulations on your latest cont
    ribution to the premature deaths and
    unnecessary suffering of untold
    numbers of young women who will
    contract breast cancer in the next fe
    w years. Your bi
    ased and one-sided
    coverage of the 25 year results of
    the Canadian breast screening trial
    gives undeserved credence
    to one of the only randomized controlled trials
    (of at least 10 others) to show no be
    nefit in decreasing deaths from breast
    cancer. The Canadian trial is ou
    r very best example of how bad
    mammography is worse than no mammo
    graphy at all, and has been
    criticized by scientists
    around the world for its poor design and highly
    questionable results. It is truly “gar
    bage in, garbage out”. Why you have
    not given equal coverage to the multit
    ude of even larger studies by Tabar
    and others showing as mu
    ch as a 63% decrease in
    breast cancer deaths
    in women screene
    d with mammography
    is very disturbing to me. You will
    clearly convince many unsuspecting
    and less knowledgeable women to
    forego mammography, and lose their only
    chance of surviv
    ing this terrible
    disease. The consequent
    blood on your
    hands will likely never wash out.

  19. Luz says:

    That’s why I support Dr. Love in her work. She sees that there is absolutely no consistency in this crazy “epidemic” of breast cancer – starting from screening. And, even if caught early and treated, there is still a good percentage who develop metastatic disease many years later. I was already metastatic on initial diagnosis, but if I had been diagnosed and treated earlier, I still could have been in this situation. Too many variables here, and Dr. Love seems to be the one of the few (or maybe the only one) that is really trying to figure it out!

Leave a Reply

Your email address will not be published. Required fields are marked *