Mammogram Recommendations Draw Widespread Anger

[bytor2112]

Is this beginning of health care rationing? Ladies, how do you feel about this?

New guidelines saying women between the ages of 40 and 50 should not receive mammograms to screen for breast cancer have met a groundswell of rejection from many medical centers, breast cancer survivors and numerous doctors -- some of whom have advised their patients to ignore the recommendation

[pam]

The USPSTF recommends against teaching breast self-examination (BSE).

I was flabbergasted to see this on their recommendation list. What???? How many women have discovered lumps by doing this that could have been a leading cause of saving their lives by having it checked out?

I personally would ignore any and all of their recommendations. You start limiting as to when women have mammograms and the cost of cancer treatment goes up as fewer women will have them done. It seems like we are going backwards on this instead of forwards. At some point there needs to be a baseline mammogram done to have something to compare to on future ones. Is the answer to wait until women are menopausal and more prone to breast cancer? I don't think so.

[Guest]

It really gets me peeved when somebody else tries to tell me how to take care of my own body. So, give me scientific studies, give me facts, educate me, but don't dare decide for me!

[pam]

Bravo anatess. I totally agree.

[ADoyle90815]

I don't think it's "rationing" at all. As for the recommendations, I don't have any family history of breast cancer, but I'm still going to do self-exams since those have saved lives. By recommending that mamorgrams start later, the cost of treating cancer is going to rise, even if there is no reform of health care.

[bytor2112]

I don't think it's "rationing" at all. As for the recommendations, I don't have any family history of breast cancer, but I'm still going to do self-exams since those have saved lives. By recommending that mamorgrams start later, the cost of treating cancer is going to rise, even if there is no reform of health care.

Ok....not rationing YET. Imagine government health care and this recommendation......that is rationing.

Edited November 18, 2009 by bytor2112

[MarginOfError]

I have been knocking heads with people about this since the announcement. So let me just set a few things straight. In doing so, I remind you that I am a biostatistician. My primary field of research is Women's Health, including breast services ranging from screening, diagnostics, surgical and medical treatment, and reconstruction. I work for an institution that does 5,000 breast cancer screenings every year; 1,500 biopsies; and 700 mastectomies.

We have known for a few years now that breast self exams do not improve early detection or outcomes in breast cancer. They actually generate an inordinate number of false positives. Many institutions have discontinued the practice of encouraging the self exam because there is absolutely no difference in detection of breast cancer between women who perform the self exams and those who don't.

With regards to the screening recommendations: first of all, the recommendation to start screening at age 50 is only for women with no genetic factors or family history. That is, the low risk group of patients need not start screening until age 50. Those with increased risk are still recommended to start screening by age 40.

Also, the new recommendations indicate that on average, one additional death from breast cancer will happen for every 1,000 screenings performed. In contrast, we expect 420 fewer false positives for every 1,000 screenings under the new recommendations. Let's look at what those numbers mean for an institution like mine.

We perform 5,000 screenings every year, and service a population of about a quarter million people. Under the new guidelines, we would expect five more deaths per year due to breast cancer. But, we will stop 2,100 false positives, meaning we can expect to do 2,100 fewer biopsies every year.

Now, you may notice that these projections don't make sense. We only do 1,500 biopsies a year, and we expect to do 2,100 fewer!? How does that work!? As it turns out, my institution already has a relatively low false positive rate--among the best in the country. There are several factors that have helped us achieve this low false positive rate; primarily we have a team of technicians that specializes in breast mammography. They are very, very good at what they do, and being able to do 5,000 a year helps a lot. But most institutions don't do that many screenings and don't have specialized technicians devoted to mammography. As a result, they tend to be more cautious and call more false alarms.

But now look at this fact. We perform 1,500 biopsies each year, and only 700 mastectomies. That means 800 of our positive mammograms are false alarms. We have one of the best false positive rates in the country, and over half of our biopsies are false alarms. That should give you an idea of how many false alarms we set in order to catch one true case of breast cancer.

There has actually been a lot of discussion about the value of early breast cancer screening among breast professionals, and it certainly isn't a new idea that we might be overscreening. I'm not entirely sold on these recommendations yet, but they don't surprise me. I would still like to see another study corroborate this, but my gut tells me this is pretty accurate.

But calling this a rationing of health care is just plain absurd. The evidence indicates that a lot of these procedures are being done needlessly, and biopsies aren't a comfortable thing to go through. Why shouldn't we cut these out?

[pam]

We have known for a few years now that breast self exams do not improve early detection or outcomes in breast cancer. They actually generate an inordinate number of false positives. Many institutions have discontinued the practice of encouraging the self exam because there is absolutely no difference in detection of breast cancer between women who perform the self exams and those who don't.

I hear what you are saying MOE..but I'm looking at this in a totally personal way. I know of two women personally, that discovered lumps while doing a breast exam. They immediately sought medical advice and it was discovered they did in fact have breast cancer. Luckily they caught both of them early, were able to be treated and have had no further treatment in quite some time now. All because of doing a breast exam.

I realize it's two women out of how many? You are the numbers guy. But to me..saving two women because they were taught to do breast exams out weigh more than telling women to stop doing them and that it's unnecessary.

[bytor2112]

MOE....as always I appreciate your insight, BUT, it seems that some would disagree with the new recommendations like: The American Cancer Society and the American College of Obstetrics and Gynecology are among the many groups that supported the old guidelines and have stood firmly by them and M.D. Anderson, the Mayo Clinic, Baylor College of Medicine, Beth Israel Deaconess Medical Center and Fox Chase Cancer Center were among many hospitals that said they are sticking with the current guidelines, recommended by the American Cancer Society.

Maybe your right, but early detection has been ingrained in to our way of thinking and this occurring during the health care debates probably doesn't help.

[Wingnut]

There are a number of problems with self exams that I can think of off the top of my head, and I'm not even a healthcare professional.

• The average woman doesn't always know what she's looking for.
• If she does them on different days each month, she's going to get different results, based on different hormone levels.
• Progesterone can cause lumps to form that are perfectly normal and go away after a few days, but if a woman self examines in the later half of her cycle, she's likely to feel these.
• The average woman may just not do it right.
• The average woman doesn't do it regularly.
• The average woman is more likely to worry is she feels the slightest anomaly, whereas a trained professional (read: doctor or nurse practitioner) has the experience to know the difference between normal and abnormal.

Obviously, no one is going to prevent any woman from performing self exams, but I can completely understand why the studies seem to show that they aren't effective.

[talisyn]

I don't remember an outcry when it was recommended women under 40 get a pap smear every 5 years instead of every year. Of course that could be due to relief, while a mammogram isn't nearly so personal

But to be absolutely completely controversial do the number of women found with breast cancer even come close to equaling the number of women who get mammograms?

Since when was consumers saving money 'rationing health care'?

Edited November 18, 2009 by talisyn

[beefche]

From the article:

Lyman said his primary criticism is that in between the last set of screening guidelines in 2002 and the current ones, only one study has come out in the area, and it did nothing to change what doctors know about mammograms.

"I'm puzzled why, when the evidence hasn't really changed, when the estimate in benefit and risk hasn't really changed, why they reversed their position," he said.

I found that interesting. BTW, Lyman is Dr. Lyman an oncologist at Duke University.

Here's the scary part for me. If I find a lump in self examination, (actually isn't it more likely that a partner finds lumps in breasts?) does that mean I can't have a mammogram to find out what it is? It's the personal stories of women who have no family history of breast cancer who on either a routine exam or self exam found a lump and by early detection was able to live.

I'm also really nervous that so many of the major groups we consider experts disagree with this finding of less examinations. That's really disconcerting.

[Just_A_Guy]

I think part of the issue, Talisyn, is that if the President gets his way it won't be the consumer's money. It'll be the government's money. I'd be totally fine with this IF the President's expressed wishes of a single-payer system weren't lurking in the background.

I appreciate MOE's thoughtful analysis; but at the end of the day that cold calculus means that some people are going to die in order to save money. The "death panel" hyperbole was unfortunate; but the issue remains that a group of (no offense, MOE) anonymous statisticians is going to be making decisions about the value of a human life that could affect the type of medical care each of us winds up getting (even though those decisions aren't aimed at any one patient in particular).

It's one thing to have a medical or consumer advocate group "suggest" that I might not wish to purchase a particular product. It's quite another thing when the same group basically says government need not purchase it for me, when the President's end game is a single-payer system where (as I understand it) if you don't get it from the government, you don't get it at all.

Edited November 18, 2009 by Just_A_Guy

[MarginOfError]

I hear what you are saying MOE..but I'm looking at this in a totally personal way. I know of two women personally, that discovered lumps while doing a breast exam. They immediately sought medical advice and it was discovered they did in fact have breast cancer. Luckily they caught both of them early, were able to be treated and have had no further treatment in quite some time now. All because of doing a breast exam.

I realize it's two women out of how many? You are the numbers guy. But to me..saving two women because they were taught to do breast exams out weigh more than telling women to stop doing them and that it's unnecessary.

The numbers of women who successfully identify a breast cancer are ludicrously small when compared to the number of women who misdiagnose a cancer and then insist on treatment, not to mention the number of women who have breast cancer and never detected in in their self exams. Remember the false positive rate among the technicians at my institution? And they're the professionals! I can appreciate the sentiment of saving every life, but false positives impose a very real strain on finances, organizations, resources, and emotions. It makes no sense at all to give population-based recommendations on the experiences of a very few special cases.

It might also be interesting to know that breast cancer is typically very slow to develop. It's fairly common for technicians to notice something during a screening and mark it for follow-up. In some cases, it can be as many as five years later that the technician decides that it has become a cause for concern. Frequently, the abnormality never develops into cancer. If it's that tricky to determine with imaging, I can only imagine how difficult it must be by touch.

MOE....as always I appreciate your insight, BUT, it seems that some would disagree with the new recommendations like: The American Cancer Society and the American College of Obstetrics and Gynecology are among the many groups that supported the old guidelines and have stood firmly by them and M.D. Anderson, the Mayo Clinic, Baylor College of Medicine, Beth Israel Deaconess Medical Center and Fox Chase Cancer Center were among many hospitals that said they are sticking with the current guidelines, recommended by the American Cancer Society.

Maybe your right, but early detection has been ingrained in to our way of thinking and this occurring during the health care debates probably doesn't help.

And this is where I point you back to where I said, "I'm not entirely sold on these recommendations yet, but they don't surprise me. I would still like to see another study corroborate this, but my gut tells me this is pretty accurate." I don't know that switching to these recommendations right now would be prudent. Very rarely do we change standards of practice based on a single study. Instead, we should duplicate the study, and perhaps pilot the recommendation in a couple areas and see what happens during the next five years. But the fact remains, that ratio of 1 to 420 is huge. Based on my knowledge of statistics, I kind of doubt the direction of that ratio would change to dramatically.

[talisyn]

From the article:

I found that interesting. BTW, Lyman is Dr. Lyman an oncologist at Duke University.

Here's the scary part for me. If I find a lump in self examination, (actually isn't it more likely that a partner finds lumps in breasts?) does that mean I can't have a mammogram to find out what it is? It's the personal stories of women who have no family history of breast cancer who on either a routine exam or self exam found a lump and by early detection was able to live.

I'm also really nervous that so many of the major groups we consider experts disagree with this finding of less examinations. That's really disconcerting.

I don't know of a doctor that will tell you no as long as you have the means of paying for the test. Especially when it comes to cancer. But then that depends on what insurance coverage you have. Will your HMO take this story and run with it?

JAG, I understand your point. But I fail to see how universal health care can possibly be worse than what health care insurance providers are legally allowed to do now.

Edited November 18, 2009 by talisyn

[Just_A_Guy]

JAG, I understand your point. However, the fact is this is happening now. Right now. Before the bill has even been read on the floor of the Senate. Maybe you can say 'It will get worse with universal health care', but I can say 'Knowing human nature it will probably get worse without universal health care'.

Will not! Pbbbbbbbbbbbbbbbbbbbbbbt!

[talisyn]

Will not! Pbbbbbbbbbbbbbbbbbbbbbbt!

Nuts! You caught me before I could change my post I looked back at it and realized I could have put it much much better, sorry!

Oh, and will too!

[beefche]

I don't know of a doctor that will tell you no as long as you have the means of paying for the test. Especially when it comes to cancer. But then that depends on what insurance coverage you have. Will your HMO take this story and run with it?

So, now we are saying that doctors are money hungry sharks? I have never felt that my doctor has made a recommendation for me based on money issues. In fact, all of my doctors have been very helpful in assisting me with less expensive choices. And I've had HMOs before and didn't feel that the doctors were like car salesmen.

Perhaps I just lack life experience, but I think we do a disservice to physicians who take their jobs seriously and make recommendations based on actual needs rather than the pocketbook.

[talisyn]

So, now we are saying that doctors are money hungry sharks? I have never felt that my doctor has made a recommendation for me based on money issues. In fact, all of my doctors have been very helpful in assisting me with less expensive choices. And I've had HMOs before and didn't feel that the doctors were like car salesmen.

Perhaps I just lack life experience, but I think we do a disservice to physicians who take their jobs seriously and make recommendations based on actual needs rather than the pocketbook.

Lol I'm not saying dr.s are money-hungry sharks. I'm saying a lot of dr.s are contracted with various insurance companies and have to follow the guidelines set for in the contract. I think anyone who spends decades in school with the desire to help their fellow beings are well deserving of respect. But in order to pay for basic necessities and insurance it's beneficial to enter these contacts, to become a 'preferred provider' as my insurance calls them.

I'm not even saying insurance companies are evil, either. They simply are as they were made. They have rules and guidelines based on statistical evidence that will create more profit than loss, like any business. But they are a business.

[beefche]

Yes, they are a business which means they need to make a profit in order to stay alive. However, they aren't going to say, "well, the experts say to get a mammo once a year, but you are required to get them quarterly." They base their decisions (at least partly) on recommendations by experts.

And the experts are saying regular mammos. Now, the gov't task force is making recommendations which are in opposition to the experts. Sorry, even if the task force is made up of doctors and statisticians (however you spell that), they are going against MANY other experts. Why?

[MarginOfError]

From the article:

I found that interesting. BTW, Lyman is Dr. Lyman an oncologist at Duke University.

Here's the scary part for me. If I find a lump in self examination, (actually isn't it more likely that a partner finds lumps in breasts?) does that mean I can't have a mammogram to find out what it is? It's the personal stories of women who have no family history of breast cancer who on either a routine exam or self exam found a lump and by early detection was able to live.

I'm also really nervous that so many of the major groups we consider experts disagree with this finding of less examinations. That's really disconcerting.

If you find a lump, my understanding is that you would consult with your physician, and if your physician felt there was need for a mammogram, then it would be ordered. The idea is not to restrict access to health care, to but make more efficient use of it.

[MarginOfError]

I don't remember an outcry when it was recommended women under 40 get a pap smear every 5 years instead of every year. Of course that could be due to relief, while a mammogram isn't nearly so personal

But to be absolutely completely controversial do the number of women found with breast cancer even come close to equaling the number of women who get mammograms?

Since when was consumers saving money 'rationing health care'?

Like I said earlier, we screen 5,000 women each year (through mammograms) and perform 700 mastectomies. In other words, 14% of women who are screened are found to have breast cancer.

[MarginOfError]

I think part of the issue, Talisyn, is that if the President gets his way it won't be the consumer's money. It'll be the government's money. I'd be totally fine with this IF the President's expressed wishes of a single-payer system weren't lurking in the background.

I appreciate MOE's thoughtful analysis; but at the end of the day that cold calculus means that some people are going to die in order to save money. The "death panel" hyperbole was unfortunate; but the issue remains that a group of (no offense, MOE) anonymous statisticians is going to be making decisions about the value of a human life that could affect the type of medical care each of us winds up getting (even though those decisions aren't aimed at any one patient in particular).

It's one thing to have a medical or consumer advocate group "suggest" that I might not wish to purchase a particular product. It's quite another thing when the same group basically says government need not purchase it for me, when the President's end game is a single-payer system where (as I understand it) if you don't get it from the government, you don't get it at all.

No offense taken. Statisticians always get the short end of the stick. But in our defense, we're asked to do some pretty impossible things.

Let's explain something about how statistics in diagnostic medicine. When you are trying to diagnose a disease, the ideal is to have a perfect test...one that always correctly diagnoses disease (Sensitivity) while never falsely indicating disease (False Positive). In reality, these tests rarely exist. Some of the best diagnostic tests have 90% sensitivity with a 10% false positive rate, but tests that perform this well are pretty rare or are for diseases that are pretty obvious anyway.

The goal when designing any diagnostic screening is to simultaneously maximize sensitivity and minimize the false positive rate. Unfortunately, the relationship between the two is such that one usually comes at the expense of the other. That is, if you increase your sensitivity, you increase your false positive rate. If you decrease your false positive rate, you increase your sensitivity. It's actually a pretty frustrating conundrum.

Fortunately, for breast cancer we have biopsies, which are extremely accurate and are the ideal test for identifying cancer. Unfortunately, the procedure is painful, expensive, and requires time that is in high demand among laboratory technicians. It isn't reasonable to do this for every woman because the time could be better spent on people who are actually sick, the money could be better spent on people who are actually sick, and it is considered poor health care practice to subject healthy people to painful procedures.

Enter screening. Screening is intended to narrow the field of candidates for biopsy, hopefully by identifying the patients most likely to actually have a cancer.

But now we have to ask ourselves this philosophical question: Is it better to catch every case of the disease, or to maximize the diagnostic capability of the screening. I know those sound like they are the same, but they really aren't. If we use the screening to catch every case of cancer, our experience indicates that we drastically increase the false positive rate. In this case, we are still subjecting a huge number of healthy women to a test that really isn't necessary.

So now if we maximize the diagnostic capability of the test, meaning it makes the correct diagnosis regardless of disease status, we will miss some of the cancer cases (false negatives).

What the task force report found was that under the new screening recommendations, there would be one more death from false negatives for every 1,000 screenings and 420 fewer false positives from those screenings. This represents a huge increase in the diagnostic accuracy of the test.

So now you kind of see where we statisticians stand in this battle. If we suggest the new recommendations, we're told that we're insensitive and can't put a value on a human life (which we agree with). But under the old regulations, we're submitting thousands of healthy women to a painful procedure they don't need--also an unethical decision.

I might also add that statisticians never ever ever apply their population based findings to individuals. Statisticians may be able to develop guidelines for screening, but they're only guidelines and cannot possibly make a substitute for informed discussion between patients and physicians. Nothing we do can replace the intuition of a good physician. Recommendations such as these will, on average, work for about 95% of the population. It's up to individual physicians and patients to identify the other 5%.

[MarginOfError]

Yes, they are a business which means they need to make a profit in order to stay alive. However, they aren't going to say, "well, the experts say to get a mammo once a year, but you are required to get them quarterly." They base their decisions (at least partly) on recommendations by experts.

And the experts are saying regular mammos. Now, the gov't task force is making recommendations which are in opposition to the experts. Sorry, even if the task force is made up of doctors and statisticians (however you spell that), they are going against MANY other experts. Why?

Because you have to go against the experts to make vital changes to the system. This is exactly how the system is supposed to work, with the exception that they seem to have expected their recommendations to be implemented immediately (that part was foolish). But it's part of the scientific process: you do your study, publish your results, and the someone else replicates your study to either corroborate or discredit your. If the process is done objectively, the truth eventually floats to the surface. That's why I said I want to see another study in this. If you were to gather two or three studies with such convincing results, it would probably be grounds for changing the system.

[Just_A_Guy]

So now you kind of see where we statisticians stand in this battle. If we suggest the new recommendations, we're told that we're insensitive and can't put a value on a human life (which we agree with). But under the old regulations, we're submitting thousands of healthy women to a painful procedure they don't need--also an unethical decision.

Absolutely. I don't mind so much if "recommendations" is all these are. I just get worried when "recommendations" become "policy" to which the individual may be arbitrarily and inescapably subjected.

I might also add that statisticians never ever ever apply their population based findings to individuals. Statisticians may be able to develop guidelines for screening, but they're only guidelines and cannot possibly make a substitute for informed discussion between patients and physicians. Nothing we do can replace the intuition of a good physician. Recommendations such as these will, on average, work for about 95% of the population. It's up to individual physicians and patients to identify the other 5%.

Certainly. I guess my beef isn't so much with the statisticians, as it is with those who would use statistical findings in order to effect bad policy.