You probably have heard about the new Swedish study of mammograms that appeared in the journal Cancer last week. It seemed to suggest that an average-risk woman in her 40s would benefit from routine mammograms—contrary to the U.S. Preventive Service Task Force’s latest recommendation.
But over at the HealthNewsReviewBlog Gary Schwitzer points to some curious caveats in major news stories about the study. His team reviewed four stories and found that:
“Three of the four stories gave a pretty clear indication that there were methodological concerns about the Swedish research. (Only HealthDay – of the four we reviewed – offered no such hint.)
“• 4th paragraph of AP story: ‘The new study has major limitations and cannot account for possibly big differences in the groups of women it compares.’
“• 1st paragraph of LA Times blog story: ‘Critics charged that the study was poorly designed and potentially vastly misleading.’
“• 2nd sentence of NY Times story: ‘Results were greeted with skepticism by some experts who say they may have overestimated the benefit.’”
Unfortunately, “none of the stories did a very complete job of explaining those potential limitations,” Schwitzer writes. “Because of the confusion that must be occurring in the minds of women – especially those in their 40s – this is a time in which journalism must rise to the need and do a better job of evaluating evidence and helping readers make sense of what appear to be conflicting findings.
“I was in Chapel Hill, North Carolina when the study was published,” he continues “ and had the chance to talk about it with former US Preventive Services Task Force member, and a recognized thought leader on issues of prevention and especially of screening tests, Dr. Russell Harris, Professor and Director of the Health Care & Prevention Concentration of the University of North Carolina School of Public Health.
“He sent me an email with the following analysis of the study. Read this carefully. I'm confident you'll learn a great deal from his analysis.”
To read Harris’ analysis, go to Schwitzer’s blog.
What is most important, I think, is to recognize that mammograms carry risks as well as benefits—not just the risk of a false positive, but the danger that discovery of a tiny tumor that would have disappeared leads to treatment. Sometimes, this means that a woman undergoes an unnecessary mastectomy or lumpectomy.
Naomi and I have both written about this danger, most recently here where I note that the danger that a woman will lose all or part of a breast needlessly is growing. Advances in mammography and other advanced imaging technology over the past 30 years have meant that pathologists wind up rendering opinions on ever-smaller breast lesions, some the size of a few gains of salt.
Just how often are women misdiagnosed? According to the National Institute of Health (NIH) diagnosis of non-invasive breast cancer was rare, until the 1980s, when mammograms become popular. Now, 50,000 women are diagnosed with these small lesions (ductal carcinoma in situ, or DCIS) each year.
The question each woman has to ask herself is this: does the magnitude of the potential benefit equal the magnitude of the risk?
While analyzing the Swedish study Harris also underlines significant methodological problems.
Suffice to say, this new study does not end the debate.
Unfortunately, what this study highlights is what I have said earlier on this blog: the issue of mammography and breast cancer is clouded by the absence of good data about the effectiveness of breast cancer screening with mammography.
The Cochrane study on breast cancer and mammography which is the basis of recent thinking about mammography is, of course, actually a metastudy — a compilation of data from existing studies in an attempt to extract more conclusive information. Unfortunately, the studies Cochrane depends on are all deeply flawed. The Canadian study, which Cochrane puts the highest weight on, is well known to be poor quality due to poor supervision and qualifications of the interpreting doctors, use of inferior and inadequate mammography machines, and serious flaws in randomization, yet Cochrane finds it to be the BEST study available.
Mammography remains an important question, which the new Swedish study unfortunately only makes more confusing.
What is needed, obviously, is a good study. However, in the current climate it would be impossible to create a good study, since finding a large (over 200,000 women) cohert of subjects who would submit to randomization and would not cross over from one limb of the study to the other for the length of the study would be impossible short of force by a totalitarian state. Maggie’s own past comments have illustrated this: she has indicated that as a mammography agnostic she would be willing to volunteer to be in the control group for a study, but would be unwilling to submit to randomization that could assign her to the mammography group. That is a very understandable position, and women the world over feel the same way, unwilling to submit to randomization because they already have strong personal ideas about mammography, a problem that has contributed to the poor quality of many later studies.
The take away lesson here is that mammography and other interventions should not be adopted without good studies, and that the requirment that good studies exist probably mandates government involvement in order to overcome the careerism, personal grandstanding, and plain greed that damages so much medical research.
Unfortunately, in mammography the ship has already sailed, and it is probably impossible to call it back. This undoubtedly is causing women to suffer, either from potential damage from overuse of mammography and aggressive management following cancer detection or from loss of life due to failure to detect curable lethal cancers, a question that at this point remains unanswered. Since scientific information is so deficient, we are left with the fact that the choice about mammography has been left to the individual woman to decide without the benefit of more definitive scientific information , a situation more appropriate to botox therapy than to an issue like breast cancer.
Pat S wrote:
“Since scientific information is so deficient, we are left with the fact that the choice about mammography has been left to the individual woman to decide without the benefit of more definitive scientific information”
—————-
An astute post, but I do have an analogy question. We have free speech, but you cannot yell “fire” purposely without data in a crowded place so as to injure people unnecessarily and purposefully. When supposed medical authorities put out information that is false or unproven but related to life quality, is that not the same as yelling fire without proof in a crowded theatre. How many people can overlook such a warning at their own peril?
Ah, NG, you have hit on the essence of much of what is wrong with US and world medicine.
Once upon a time (say from 1820 to 1960) we lived in a world in which medical breakthroughs, from anesthesia to insulin to antibiotics, were so obviously worth while that standing around and debating or further testing them was not only not necessary but probably immoral — see for example the recent scandal about infecting people in Guatemala with syphilus in order to test the effectiveness of antibiotic treatment.
That day has passed, but unfortunately the mindset lives on.
We are awash in procedures, tests, and treatments that are at best unproven, often a waste, and at worst harmful. That includes things like coronary artery bypass and angioplasty, newer drugs for hypertension, widespread use of imaging such as CT and MR in settings in which its usefulness is questionable, diagnosis and management of back pain, etc. etc.
All of these have been agressively promoted based on questionable or inadequate research. All of them have helped the careers of researchers who were the innovators of the techniques. All provided interesting articles for journalists to write. All of them are very profitable for drug companies, medical equipment manufacturers, hospitals, and doctors. And all of them have created a huge and aggressive demand, often backed by political pressure on payers, by the general public.
Mammography came from the same pathway. First, it rests on an apparently flawless premise: early detection and treatment of breast cancer results in better outcomes. Second, it rested on impressive but flawed research. Third, it developed a groundswell of pressure from the press, women, and providers to demand that it be available, cheap for the patient, and promoted. All those pink ribbons, foundations, and politicians, as you can see. Unfortunately, few people (there was actually a metastudy from New Zealand that reached similar conclusions to the Cochrane study 20 years ago, as well as a few other people) were willing to stand against this gale force wind and point out that the emperor had no clothes.
All this is not to say that mammography is useless. WE DON’T KNOW. And, as I stated earlier, we are unlikely to know anytime soon. There is too much political and public feeling about this to do a decent study.
If you want to see the same thing in play today, involving a smaller number of patients (at least for now) but a huge amount of money, take a look at the ongoing adventure of left ventricular assist devices for endstage congestive heart failure patients (as in Dick Cheney.) Researchers, providers, and politicians have forced Medicare, and I guess Medicaid in many states, into providing a treatment which is poorly documented, dangerous to many patients, and costs over $250,000 a time. Watch for further developments.
Anyhow, I agree with your point that what is needed here is a good cop or an honest broker to assess this kind of thing before, not after, it gets out of hand. The only hope I could see at all is a federal board that is composed of qualified and competent people who are willing to back unpopular public positions and are safe, or nearly safe, from political pressure. If this sounds a lot like some of the proposals from reformers here, and like existing programs in Britain and elsewhere, that is because it is. Without this, we are doomed.
Pat S
A mammogram is a diagnostic test that uses a low-dose x-ray system to examine breasts. In digital mammography, the x-ray film is replaced by solid-state detectors that convert x-rays into electrical signals. A computer-aided detection (CAD) is a system that uses a digitized mammographic image that can be obtained from either a conventional film mammogram or a digitially acquire mammogram. Nonetheless, all are diagnostic tests.
What about the validity of diagnostic testing? The traditional criteria ever used to evaluate laboratory tests has been the predictive accuracy of the test. None of the available laboratory tests used in the selection of treatments for cancer patients have ever been tested for efficacy. This includes CT, MRI and Pet Scans to measure tumor response to treatment.
The only data supporting any of them relate to test accuracy. It is not essential to prove that the use of a diagnostic test improves clinical outcomes. All that is needed is to prove that the test has a useful degree of accuracy.
Now, mammograms are not used to measure tumor response to treatment, so how do mammograms fall on the spectrum of diagnostic testing based on accuracy vs. efficacy? If it is based on “accuracy” then it is not essential to prove that its use improves clinical outcomes. If it is based on “efficacy” then it is essential to prove that its use improves clinical outcomes.
Greg —
I may be misreading your comments, but looking at what you wrote you seem to suggest that the question of efficacy is not a criterion for evaluating radiologic exams and other diagnostic tests.
I disagree strongly. As a radiologist and mammographer, I believe that the ultimate value of a given test is how it effects patient outcomes. A test that is highly accurate but has very little effect on outcomes (for example, CT or MR for detection of pancreatic cancer) has limited value, since the goal is not really to simply identify the abnormality in question, but rather to improve life span and quality of life for the patient. This is especially true for a screening exam. If a screening exam discovers only conditions which cannot be successfully treated (as with pancreatic cancer) or leads to treatment of many lesions which do not require treatment (which is the question now being raised about mammography) the test is of little value, and may even have a negative impact on patient quality of life and survivorship.
This is a new trend in medical policy over the last 20 years. It has become important because in so many situations the radiologic test serves as an entry point for treatment and there are no other real alternatives available to impact the decision to enter the patient into management, since many radiologic exams (especially mammography, CT, and MR, where the sensitivity of the tests greatly exceeds the sensitivity of physical exam, history, and lab testing) act as the “trump card” in entry into management, determining that the patient needs treatment independent of clinical criteria. For these exams, the old adage that we “treat patients, not x-rays” is no longer true. A patient who has an abnormal mammogram, barring the existence of a condition which makes surgery impossible or unwise, will, by current practice standards, get a biopsy. If the biopsy is positive, the practice standards indicate that the patient be treated for cancer.
The confounding result in the Cochrane study is that the Nordic Cochrane group presents evidence that many patients who have lesions detected only by mammography (formerly considered the great strength of mammography) may have cancers that do not require treatment. Treatment in fact will make the patient’s quality of life worse, not better, and may even – due to potential complications of treatment – result in death or severe injury to the patient.
Consequently, it is important to evaluate mammography – as well as other studies – not just for its sensitivity (excellent,) specificity (not very good,) and accuracy (mediocre due to poor specificity,) but rather for its efficacy. There were many studies performed in the 70’s, 80’s, and 90’s in the wake of the Breast Cancer Detection Demonstration Project that attempted to prove that mammography could be life saving or life prolonging for the screened population. However, due to some studies which failed to show those effects (the Canadian study, for example) and other studies that appeared to show those effects (the Swedish Two County study, for example) but proved to be technically flawed, the studies have failed to show that effect. The new Swedish study, because of methodological flaws, also fails in that respect, joining ALL existing studies of mammography efficacy.
Anyhow, I am not completely sure if I am interpreting you correctly, but regardless I would maintain that the important question for adoption of radiologic exams on a widespread basis, and especially for adoption as a screening test, is not whether the exam is accurate or beautiful but rather what its effect on patient health and life is. The day is past when we can simply show that we have pretty pictures and find a lot of disease, and in the new day we need to know if the test works positively for patient health and leads to management that works positively for patient health. At this point, mammography is neither proven nor disproven to meet those criteria, since it turns out that the answer right now is we just don’t know.
This criterion for evaluating testing is now being applied to many other tests as well. Prostate screening with lab testing, MR for spine and knee evaluation, coronary angiography in patients not presenting with acute MI, and many other tests are now under attack since there is evidence that they are not efficacious. Many surgical and medical treatments are also being questioned aggressively. In my opinion this is a good trend, since it potentially provides for better care for patients by avoiding unnecessary surgery and treatment that does not lead to good outcomes. Incidentally — not a true medical goal but rather a happy accident — de-emphasis of these tests also could save hundreds of billions of dollars a year.
Pat S
The validation standard that private insurance companies accept from molecular profiling tests is “accuracy” and not “efficacy.” No longer will it be essential to prove that the use of a diagnostic test improves clinical outcomes, all they have to do for these molecular profiling tests is prove that the test has a useful degree of “accuracy.”
The traditional criteria ever used to evaluate “laboratory” tests has been the predictive “accuracy” of the test. None of the available laboratory tests used in the selection of treatments for cancer patients have ever been tested for “efficacy.”
This includes estrogen receptor, progesterone receptor, Her2/neu, Oncotype DX, MammaPrint, EGFR amplification/mutation, immunohistochemical (IHC) staining for tumor classification, Cell Culture Assays, Bacterial Culture and Sensitivity Testing, CT, MRI and Pet Scans to measure tumor response to treatment.
The only data supporting any of them relate to test “accuracy” and there is a total lack of information regarding test “efficacy” (randomized trials with outcome measurements for diagnostic tests).
The standards used to judge the utility of laboratory and radiographic tests have always been (1) acceptable “accuracy” of clinical correlations and (2) clinical utility, in the judgement of the physician ordering the test. Demanding proof of “efficacy” as opposed to proof of “accuracy” is completely unprecedented for any laboratory tests in cancer.
Greg —
I cannot comment on most lab tests, since they are outside my area of expertise, but the traditional scientific standard for radiologic exams is proof of efficacy. In fact, the many studies reviewed by Nordic Cochrane in their report on screening mammography were all evaluations of efficacy. Medicare has refused to pay for several radiologic exams in the past because of the lack of evidence of efficacy.
I do not doubt that what you say about the lab tests is true. For that matter, there are many surgical procedures, devices, and drug therapies that have poor validation of their efficacy but are paid for by insurers.
The problem that illustrates is that in the US payment by insurers, both public and private, is often as much related to politics, publicity campaigns, and public relations as to the utility of the effected procedures, tests, and drugs. The history of screening mammography has been heavily influenced by exactly those things, and the current attack on screening mammography by scientists and by various organizations is based on the evidence that efficacy is unproven despite mammography being aggressively sold to the public and the public largely eager to demand screening mammography.
However, the fight over efficacy is spilling over into the lab as well. Specifically, lab screening for prostate cancer is under attack, as Maggie’s immediately previous post notes. As with mammography, the fight is not over the accuracy of the tests, but rather over the question of efficacy of the test in many populations.
I will readily admit that radiology — and most other parts of medicine — is filled with examples of practices that create beautiful pictures, exciting procedures, and ingenious treatments that some researchers and clinicians love because of their elegance but that don’t do much good for patients.
To suggest that efficacy should not be the standard applied to all parts of health care puts us back in the world of the old Marx brothers joke: “The operation was a success but the patient died.”
Pat S (1st comment)
NG
Pat S. (2nd comment),
Gteg and Pat S. (your dialogue)
An excellent point. I totally agree that, when it comes to mammography, most woman have a pretty definite opinion and would not want to be in a randomized trial. As you say, that ship has sailed.
On the other hand, this would seem to me a case where observational data is enough for an informed woman to make a decision in keeping with her own temperament and values.
In this case, observational data seems to me valuable, It’s worth just observing (and counting) the number of younger woman who are treated after discovery of ductal carcinoma in situ.
Then compare that number to the number of younger average-risk women (in their 40s or younger)) who are treated–and suvive for 10 years or more–following discovery (via mammogram)of a larger tumor that most likely would have spread.
(For obvious reaons, it’s important to separate cases where the tumor was found via mammogram to cases where the tumor was discovered by the woman herself, or a doctor during an examination.)
It’s also worth comparing the experience, over time, of women who never had mammograms in their 40s to women who had regular mammograms in their 40s.
What percentage of those who never had mammograms died of breast cancer as a result of a tumor that could have been found by a mammogrom while they were in their 40s?
Everything I know about the statistics says that the odds that I, as an average-risk woman will die of cancer are very slim, whether I undergo mammograms or not.
The odds are slightly greater if I never have mammograms- but we still are talking about tiny, tiny numbers.
(Readers: for odds, see Gilbert Welch’s books: “Are you Sure You Want to Be Tested” the newest book by Lisa Schwarts & Steve Woloshin.)
It seems to me that this information is enough for a woman to make a decision.
Some average-risk women will want to guard against that tiny,tiny chance that not having regular mammograms will be the cuase of their death from breast ancer. (They would feel terribly guilty if they died because they didn’t bother to go for mammograms.
I might kick myself, but I wouldn’t feel as guilty because I believe that:
a) we all die of something and
b)we all contribute to our own deaths in some way–if only by worrying too much about everything.
One would feel guilty only if one feels–at some irrational level–that if you just did Everything Right, you would never die.
Given the information available, I wouldn’t go for mammograms because I wouldn’t want to deal the high odds that the mammograms would hurt me rather than help me. False positives take a pschological toll, and then there’s the danger that a small lesion will be discovered, and I’ll be faced with a horrible decision: do I want to go ahead with treatment (which most doctors would recommend) , or do I want to pay attention to the research that shows that some of the lesions just disappear?
(I still think we need a “watchful waiting” protocol for those small lesions. But I’m not an oncologist– don’t know if thsi is feasible. )
NG– Thank you for the comment–a provocative analogy.
I have to agree that much of the writing about breast cancer (and other types of cancer) creates unreasonable fear where the danger is minimal.
Like crying “fire” in a crowded theatre,it’s fear-mongering and it does real harm.
If the numbers are presented fairly, that’s one thing. But too often, we’re told that our chances of dying of cancer “double” unless we do X, Y or Z when “double” means that our odds go from 0.48% to 0.96%.
On the other hand, there are examples where we should cry “fire”.
For example, if people smoke, there is no question but what their chances of dying prematurely are greatly heightened.
Moreover–we actually can help people who smoke stop smoking. So shouting “fire” serves a real purpose. . .
(. But only if we make smoking cessation programs available to all.
The good news: Medicare is now going to pay for smoking cessation treatments. (It may be too late to save a senior’s life, but it sets an excellent precedent for companies that insure people under 65.)
Pat S.–Your second comment–
Yes, there was a time was things were much more clear-cut.
But now, when it comes to clearly effective treatments, most of the low-hanging fruit has been picked. (Decades from now, their could be a new revolution in medical discoveries, but I’m looking at what has been happening for the last 20-30 years and what is likely to happen over the next 20 year–30
Most of the new treatments being developed today are, at best, minimally more effective, at worst, much riskier, and no more effective.
Finally, you write: ” The only hope I could see at all is a federal board that is composed of qualified and competent people who are willing to back unpopular public positions and are safe, or nearly safe, from political pressure.”
It seems to me we have that board– the U.S. Preventive Services Task Froce. On Mammograms, and PSA tests, they really have stood up. (Insofar as they were surprised by the reaction, they may be more naive than courageous–but at least they seem insulated from public opinon –or at least were)
Now we need to make sure the HHS gives them political protection. I think that it will.
And other reform, we will have other disinterested experts looking at comparative effectiveness.
Greg & Pat S.
I have to agree with Pat–whatever the standard was in the past, today, we realize that it doesn’t matter if a particular type of diagnostic imaging gives us a better, finer, more accurate picture.
The only important question is this: does is save lives (or lengthen lives)? If not, the quality and accuracy of the picture just doesn’t matter.
Greg– I believe you that calling for proof of “efficacy” is unprecdented in many areas-but that is what health care reform is all about.
Breaking precedent. Looking at things in a very different way as we
move toward patient-centered, evidence-based medicine that focuses on outcomes.
Maggie
Thanks for pointing out about Medicare is now (finally) going to pay for smoking cessation treatments. I’ve written on cancer discussion boards in the past that when the Medicare Modernization Act (MMA) was passed, the specialists (medical oncologists) will be reimbursed for providing evaluation and management services, making referrels for diagnostic testing, radiation therapy, surgery and other procedures as necessary, and offer any other support needed to reduce patient morbidity and extend patient survival. At that time, I had hoped they would add smoking cessation guidance and support. They finally have???
Although it may be too late to save a senior’s life, since a substantial number of patients presenting with lung cancer either smoked in the recent past or continue to do so, it is important to make sure that the patient stops smoking as soon as possible to improve their treatment outcome. According to Dr. Carolyn M. Dresler, former Head, Tobacco and Cancer Group of the International Agency for Research on Cancer, the emphasis should be on improvement of treatment outcome and future health improvement.
She says that no pharmaceutical trial ever followed whether patients smoked during their clinical trials, despite dosing themselves daily with cigarettes with hundreds of chemicals in them. The addition of nicotine inhibits the ability of a chemo drug (like etoposide) to induce apoptosis by 61%. If a drug like nicotine, which occurs in the highest concentration of any drug in a cigarette, inhibits the ability of a major chemotherapy drug by 61%, a medical oncologist should care if it was being ingested during treatment. Perhaps the financial incentive will now encourage them to do so.
Maggie and Greg —
Two points:
First, we cannot derive meaningful data about management issues without randomization because people who end up in different health care paths on their own do so partly because of reasons which are often valid and that change the outcome. You, Maggie, for example, have chosen to forego mammographic screening partly due to your own assessment that you are a low risk patient for breast cancer. If that is correct, then including you in a non-mammographic group would bias the study because you did not have the same risk as other patients. When the comparitive pool contains large numbers of people who are significantly or even slightly different from the people in the treatment pool the data collected is much less meaningful. In fact, crossover of people from one limb to the other in attempts at controlled studies caused significant problems in failed mammography studies like the Canadian and Two County studies.
As to the idea that we can derive valid information from observation and common sense without using scientifically valid studies, that often does seem appealing, and as I said earlier actually often worked well in the early history of medicine. Unfortunately the history of medicine is littered with the wreckage of ideas that seemed sensible until submitted to more rigorous scientific evaluation. In fact, the problem we now face with the issue of screening mammography came about partly because of that type of approach.
Meanwhile, in regard to Greg’s observations about the use of accuracy as a standard for tests used in cancer management, I need to point out that — at least in the cases of CT, MR, and PET — for testing applications used in following results of cancer therapy, accuracy IS efficacy. Since the tests are used to measure size of tumors and development of new tumors in patients being treated for existing cancers, evidence that tumors are increasing in size versus remaining stable or regressing or that tumors are becoming more widespread is the critical information to evaluate whether treatments are working or failing.
As I said, in that setting, accuracy is efficacy. However in areas of testing where the question of efficacy is more complex, especially in screening and de novo diagnosis and problem solving, accuracy alone is not a sufficient, and more rigorous evaluation of efficacy, including long term evaluation of outcomes, is necessary.
Maggie
At least with regard to cancer medicine, were you to have reviewed studies showing that the use of estrogen receptor improved treatment outcomes, you would find no publications at all. Were you to have reviewed papers showing that the use of panels of immunohistochemical stains to subclassify tumors improved treatment outcomes, you would find no publications at all. Were you to have reviewed studies showing that treatment outcomes were improved through the use of MRI scans or PET scans or CT scans to monitor growth and shrinkage of tumors (for the purpose of influencing the decision to continue the same chemotherapy or to change chemotherapy), as opposed to simply following patients with history, physical, simple plain radiographs, and simple lab tests, you would find no publications at all.
The types of research that would be most effective in providing the needed evidence would be synthesis of existing evidence (e.g. qualitative review, meta-analysis), primary research using existing health care databases, primary research using prospective data collection without randomization (e.g. observational study, registry), and primary research through a prospective randomized trial. Clinical efficacy trials do not do “real world” studies under “real world” conditions. Patient outcomes need to be reported in real-time, so patients and physicians can learn immediately if and how patients are benefiting from new diagnostics and therapies.
Pat S
I agree, in the setting about the use of accuracy as a standard for tests used in cancer management, accuracy IS efficacy (as you call it). And I think you have answered my first inquiry. In areas of testing where the question of efficacy is more complex (mammograms), accuracy alone is not a sufficient evaluation of efficacy.
Pat S.
Agreed that randomized controlled are far better for the reasons you cited.
I’m self-selecting because I know the history of little/no cancer in my extended family, and so skew the results.
But I thought we had already established that, in this case, randomized controlled trials are not a possibility. That ship has already sailed.
Given that it seems to me that women have to use the available information to make a choice, factoring in their own tolerance for: the very small risk of dyring of breast cancer vs. the larger risk of false positive and being overtreated.
Most women are not aware of the statistics that Welch, Schwartz and Woloshin present: they don’t realize that for average-risk women, the risk of dying of breast cancer is very small to begin with, and regular mammograms trims that risk only slightly.
Women are under the impression that their risk of dying from breast cancer (eve if they do nothing) is much, much larger than it is.
My perspective, as an oncologist who’s had breast cancer detected upon an abnormal screening mammogram when I was 42 years old, is that it’s the press who’s yelling “fire” about mammography. I am deeply concerned that journalists are fear-mongering on this, so much so that middle-aged women who might benefit are now afraid to go for screening.
The Nordic Cochrane meta-study does not show that mammography is not useful as a screening tool. That is why in their plain language summary they start by saying that it is rational to not submit to screening but also rational to submit to screening.
What the Cochrane study shows is that there is no good evidence that screening mammography is useful. However, because the data that the study evaluated is ALL technically flawed, there is no good evidence that it is not. The Cochrane summary quoted by Welch et al is, in effect, a best guess, since in the absence of satisfactory data there is no good evidence that what they have concluded is true. There is no real data about the prognosis of ductal carcinoma in situ or other minimal breast cancer patients, rather there are educated guesses based on inferences from incomplete data. This is essentially the same thing that the supporters of screening mammography are doing in their articles.
In the absence of good evidence, it is not honest to give advice about screening as if it were based on science. This is equally true of the advice that Maggie cites and of the advice that Lazlo Tabar’s Swedish group offers.
It is true that in the absence of good evidence, the usual medically prudent position would dictate that we take the least aggressive path — in this case not advise screening for average women. However, this is not the same as advising patients that there is evidence they should not have screening.
We are stuck in a situation in which we must tell women that we do not know what to do, and that they must make their decisions based on their own thinking, biases, experience, and best guess. We are asking, in the words of Dirty Harry, “Do you feel lucky?” If that sounds like a bad situation to you, you are dead right.
Pat–
I agree there is no good evidence that screening is useful–and no good is evidence that it is not.
But there is also no good evidence that the anti-breast cancer diet now advertised on ivillage is not useful.
This doesn’t strike me as a good reason to go on the diet– or to have a mammogram.
In lieu of positive evidence that a particular medical treatment helps, I am virtually always in favor of doing nothing.
Finally –and I hate to keep harping on this point–but the chances that an average risk woman will die of breast cancer whether or Not She Has Regular Mammograms– are very small.
And Mammograms reduce her risk only slightly.
When I think of the amount of money that we put into breast cancer awareness, mammograms, treatments for late-stage breast cancer that do no good, etc. etc.–it strikes me that the hysteria over breast cancer greatly exaggerates the danger.
Moreover, as palliative care specialist Diane Meier points out, by investing hundreds of millions in trying to make sure that we don’t die of cancer, we are only sparing ourselves for Alzheimers’ (More than 50% of Americans who live past 85 will develop Alzheimers. A surprisingly large percentage (15% if memory serves) who live past 65 will develop Alzheimer’s.
Breast cancer is a terrible disease, but not the only terrible disease.
When I think of the resources focused on a few diseases (breast cancer, prostate cancer) and the lack of resources invested in other areas (a war against poverty comes to mind), I think we could do a better job of reducing human suffering if we didn’t let this irrational fixation on cancer overwhelm our thinking.
The same irrational fear of death by cancer also cuases us to prolong the suffering of people dying of cancer. I think of the poor young woman in Atul Gawande’s recent New Yorker story. It was tragic that she developed cancer at such a young age, but it also was tragic that her doctors abetted her in the fantasy that somehow, she was going to be able to beat the disease.
Elaine–
I wouldn’t worry. I don’t think I’ve ever heard a woman express fear of going for a mammogram–women remain afraid of not-going.
The fear-mongering and guilt-tripping surrounding mammograms has been very effective.
In the meantime, a 2010 study shows that women greatlyl overestimate the incidence of breast cancer and likelihood of dying of it. They also greatly overestimate the effectivenses of mammograms.
And the same study shows that while their doctors almost always recommend mammograms and talk about the benefits, very few talk about the risks–or the odds.
This is a case where share decision-making is needed.
And this is why I recommend reading work by doctors Gilbert, Schwartz and Woloshin to get the other side of the story. (You can Google them.)
They make it very clear that, as Pat S. says, there is no clear answer.–it’s a matter of temperament, what type of risk you’re willing to take, life situation, what you fear most, etc.
Maggie —
You are mixing up the risk of dying from breast cancer and the Cochrane calculation of increased risk of dying from breast cancer if you do not get screening mammography.
Slightly more than 40,000 American women die of breast cancer each year. Half of them have no known risk factors for breast cancer, including family history.
Granted, the rate of death for white women from breast cancer is only about 25 per 100,000, but that death rate is just slightly less than the death rate from auto accidents, slightly greater than the death rate from gunshot wounds, and greater than the death rate for women of any cancer except lung. If you are a non-smoker, breast is still the most common lethal cancer for women.
Just to be clear, the death rate from breast cancer means that in New York City alone, about 2500 women will die of breast cancer.
Unfortunately, although there is no good evidence that screening mammography is useful to lower the breast cancer death rate, and your personal decision is certainly rational in that context, that does not mean your risk of dying of breast cancer is not significant. Granted, it is less significant than your risk of death from heart disease and other vascular diseast but it is not correct to suggest that women in low risk groups have nothing to worry about.
And that is why this whole issue is not a minor medical side show.
Pat S.–
Thanks for continuing what has become a very interesting thread.
You write that my chance of dying of breast cancer is just “slightly greater than the death rate from gunshot wounds.”
I also don’t worry about dying as a result of gunshot wounds. (Even when I go up to the Bronx, I don’t wear a bullet-proof vest.)
Seriously– of course,I realize that I could be shot. And I could wind up dying of breast cancer.
But since there is no good evidence that I can do something about the breast cancer risk–why would I worry about it? (In the case of death via gunshot wounds, I don’t wander around the Bronx at night. Otherwise, not much I can do.)
I understand that many women die of breast cancer, and particularly if they die when they are young, this is tragic.
But older women will all die of something. Why is dying from breast cancer worse than dying in so many other ways?
About a year ago, I talked to a woman (who works for the Insitute for HealthCare Improvment)whose mother and sister both died of breast cancer. The mother died here–after prolonged suffering. The sister died in the Netherlands. She was working until about a month before she died. When her father and sister came to see her at that time, she was able to go out to dinner with them and enjoy the evening. She suffered far, far less than her mother, who underwent the much more aggressive treatment that we prescribe in the U.S.
I recognize that these were two different individuals, with differnt bodies. But I have to think that the sister’s experience wasably prob more typical of dying of breast cancer in the Netherlands, and the mother’s experience more typical of dying of breast cancer in the U.S. –where we have decided that Cancer is the enemy.
I really do think that the real enemy is (as Churchill suggested) Fear (fear of dying, fear of dying of cancer, etc.)
This is the enemy that undermines the pleasure we might take in growing older (less stress, less responsibility, much, much more perspective on what is important and what isn’t.)
Maggie —
Let me say again that we do not have proof that mammography doesn’t work. What we have is proof that claims that it does work are not valid. As it happens, claims that it doesn’t work are also not valid.
I am, as ever, hopeful about this. My hope is that the ongoing publicity about this uncertainty will eventually cause enough women to be doubtful about mammography and demand valid information that we can conduct a valid study to actually anwser this question and remove women from the situation of having to make a “best guess.”
Also — the fear quote is from FDR in his first inaugural address, not Churchill. However, if you really want to get the attribution correct, FDR was probably “inspired” by Thoreau, who said much the same thing in a book that FDR was reading at the time of the speech. Thoreau in turn was “borrowing” from Montaigne, a quote also borrowed earlier by Francis Bacon and by the Duke of Wellington.
Pat S quotes:
“if you really want to get the attribution correct, FDR was probably “inspired” by Thoreau, who said much the same thing in a book that FDR was reading at the time of the speech. Thoreau in turn was “borrowing” from Montaigne, a quote also borrowed earlier by Francis Bacon and by the Duke of Wellington.”
Sounds very much like medical papers. Somebody takes it from somebody else, who takes it from someone else, who takes it from somebody else, and so on and so on and so on (LOL!).
Pat S. & Gregory
Pat —
Indeed, FDR, not Churchill! And he was talking about the Great Depression, not the war. (Though I always associate the quote with the war–it seems to sum up the English attitude toward WWII, especially as London was bombed.)
I didn’t know that Montaigne (who I like very much) used the phrase so I looked it up. On fear, He wrote; ““The thing I fear most is fear.” Montaigne goes on to say that fear “ exceeds all other disorders in intensity . . . There is no other [passion] whatever, which carries our judgment away sooner from its proper seat.”
Meanwhile, Bacon wrote: ““Men fear Death, as children fear to go in the dark; and as that natural fear in children is increased with tales, so is the other.”
All of this is very much what I have in mind when I talk abour our national obsession with, and fear of, cancer. It is “increased with tales” (narratives in the media) and causes us to set aside good judgment. . .
On breast cancer, maybe women will demand those tests, and maybe we will solve the questions . . .
Though I still can’t help but think that if we took half of the money we spend on breast cancer, and invested it in a war on poverty (health, education and welfare) we could save many more lives–children’s lives.
Pat– doesn’t the money and hysteria that we invest in trying to cure cancer strike you as excessive?
I also wish we invested more in palliative care so that death by cancer was a tamer, much less horrifying death (as it is in some other countries.)
Gregory– Yes, all great writers are always rewriting and revising each other. That really is the history of literature.
And part of the pleasure of reading literature is seeing one author weaving another author into his or her work.
The cancer industry, like the heart disease industry and the back pain industry and the joint pain industry and any number of other parts of medicine, has indeed gotten seriously out of hand. The goal of breast cancer screening — early detection in order to treat with simple interventions — has been partly subverted by the use of more and more aggressive therapy to attain smaller and smaller increments of success. In fact, much of what the Cochrane people talk about as the personal and financial cost of intervention in minimal breast cancers is not due to detection per se, but rather to the use of full bore treatments, including radiation, chemo, and aggressive surgery. In some countries (Britain for example,) this cycle has been broken by use of much less aggressive surgical treatment (lumpectomy) without use of radiation, chemo, and mastectomy in early stage cancers, to be followed by careful yearly mammography and then by more aggessive treatment in the small number (11%) of patients who develope recurrences and later more aggressive tumors.
However, the whole world of breast cancer screening pales when compared with more aggressive use of expensive therapy in patients with tumors that have reached, or always were at, untreatable stages. The desparate attempt to add 3 to 9 months of added survival of questionable quality at the cost of hundreds of thousands of dollars in very expensive drugs is something that Greg Pawelski has written about on this blog, and that continues unabated despite any common sense thought on the subject.
As far as the culture of fear of death in the US (and the rest of the developed world,) that has also reached near psychotic proportions. The philosopher Ivan Illich devoted an entire book, “Medical Nemisis,” to the issue almost 40 years ago.
Personally, I think we should abandon most research on battles for extension of life at great cost and considerable sacrifice of quality in favor of working on the problem of being able to upload individual human consciousness into machine intelligence, creating the opportunity to live forever and have virtually unlimited experience and learning at very low cost. And I am only partly kidding.
Pat S.
I agree.
It’s not the screeening per se that does great harm (nor is it terribly expensive)–the problem is how we react when we find a small tumor. And the way we torture the patient while grasping for those extra few months.
I am 55 and have had several false positive screenings in the past, all based on the presence of calcifications that were on my very first mammogram at age 45. I was not at all worried about these calcifications because I have no family history of breast cancer and all my risk factors are low. At that time, I did a couple of six month diagnostic mammograms and then went back to yearly screening with no biopsy needed. Then we moved to another state, and I skipped a few years. Last year I had another mammogram screening and again the radiologist called me back to look at calcifications more closely. I was told that the earlier images did not have a high-enough resolution so they wanted to take a closer look and take new baseline images. So, I went back in for a diagnostic and again things looked OK, but they said to come back at 6 months again. My insurance pays for yearly screenings but not diagnostic mammograms. And calcifications don’t just disappear. No one has said there was every any CHANGE in the calcifications; they just want to look closer. So I didn’t go back at 6 months and decided to wait until 12 months so the insurance would pay for another screening. Well, I went in last week and was told that I really should have the diagnostic mammogram instead of a standard screening, and my insurance will pay $0 for that because I have a high deductible. My husband and I are both self-employed, and money is tight. Comments? Would you continue paying the extra $$$ for looking at the same calcifications that have been there for many years?
Angie – I think this is a classic example of the sort of defensive medicine that does not lend itself to precise quantification by academic researchers. The doctor’s thought process probably goes something like this: If I opt for the more conservative course of treatment and the patient develops breast cancer, there is a chance that I might be sued. If the case then goes to trial, I can already hear the plaintiff’s lawyer asking: What were you thinking, doctor, when you didn’t order the diagnostic mammogram? At the same time, there is no cost penalty for the doctor if he orders the diagnostic mammogram and, if anything, he will probably benefit financially. While he may be sympathetic to your financial circumstances, his first priority is to keep himself and his partners, if any, out of court.
Angie–
I really can’t give medical advice.
This is not because I am afraid of being sued, but because I am not an M.D., let alone an oncologist who could give specific advice on breast cancer.
I’m not capable of reading your medical records and giving you informed advice on what you should do.
I am capable of reading a great deal of medical research and coming away with an infomred opinion as to what might make sense for the average patient in a large population of patients –
But that is very, very differnt from examining an individual patient, listening to that patient, and coming up with a diagnosis and treatment plan.
But I can strongly urge you to go to the Foundation for Informed Decision Making (you can Google them), call them, and use my name as needed.
There are people at the Foudnatoin who can steer you to the information you need to answer your question.
If you have trouble getting through to them , e-mail me at Maggiemahar@yahoo.com with the headline– “Angie–need info on breast cancer from informed decision-making”
Barry,
Thanks for you comments.
Maggie,
Your blog is very helpful. Thanks for the suggestion to check with the Foundation for Informed Decision Making.
Hi there, I wish for to subscribe for this web site to take newest updates, thus where can i do it please assist.
Aurelio-
Go to the first page of this blog (www.healthbeatblog.com)
To the right of the first you will see “Subscribe Now”-
Click on “Sign up here to receive emails–
Then fill out the form.