On mammograms

January 24, 2012 11:35 PM

"I believe the time has come to realise that breast cancer screening programmes can no longer be justified ... I recommend women to do nothing apart from attending a doctor if they notice anything themselves."

Peter Gøtzsche of the Cochrane Collaboration believes the evidence shows that routine breast cancer screening harms more women than it helps. This position is controversial, and making change in mammography guidelines has, historically, been difficult.

posted by latkes (52 comments total)

Previously, Gøtzsche presented evidence that the placebo effect probably does not exist.
posted by latkes at 11:36 PM on January 24 [1 favorite]

Luckily, for people without insurance this won't even be an issue!
posted by Malice at 11:43 PM on January 24 [31 favorites]

Is the screening itself harmful, or the treatment? Seems misleading, and I'm skeptical of his motives.
posted by Brocktoon at 12:05 AM on January 25 [1 favorite]

I'm 25 and "high risk," just had my first mammogram and in six months will have an MRI.

Ultimately, my mother's cancer WAS discovered by her but her doctor dismissed it—probably a "monthly" thing. Oh, and her constant fatigue was just depression.

She persisted, and she was diagnosed with Stage 4 breast cancer, bone metastasis. That's why she was so exhausted. It was in every part of her body—from her arms to her legs.

So, excuse me, but I will do anything recommended to me, to make sure I live longer than 38.
posted by autoclavicle at 12:10 AM on January 25 [35 favorites]

I am not "high risk." And for people in my [demographics, etc.], my outlook is that I'm not choosing screening for myself until age 50 (and subject to reconsideration). But this has got to be incredibly individualized. Different answers for different people. And insurance should cover.
posted by ClaudiaCenter at 12:18 AM on January 25 [3 favorites]

Is the screening itself harmful, or the treatment? Seems misleading, and I'm skeptical of his motives.

I've seen this argument before, and if I understand the claim correctly, the screenings (which do expose you to some radiation, but it's not dissimilar in scope to regular dental X-rays IIRC) catch tumors and cancers which will not necessarily be health problems, and because of the need to act on malignant tumors ASAP before they cause health problems and/or metastasize, a lot of things that don't actually require medical intervention end up being treated, often to the detriment of the patient (since anti-cancer treatments are generally pretty traumatic). What Gøtzsche and his companions in making this argument are saying is that the number of people who die and experience health problems when screening isn't being done is lower than the number of people who die and experience health problems when screening is done- that in order to reduce the harm that is being caused by cancer, we are hurting and killing even more people.

Assuming that Gøtzsche and co. are correct (and I will readily, cheerfully admit to not knowing, as I am not a doctor or scientist or statistician, and I am only trying to explain the argument as I understand it), this puts us in an incredibly unpleasant double-bind.
posted by Pope Guilty at 12:22 AM on January 25 [20 favorites]

Screening poses risks and benefits, that can be difficult to weigh big and small picture. Especially with incomplete knowledge.
posted by ClaudiaCenter at 12:24 AM on January 25 [1 favorite]

Assuming that Gøtzsche and co. are correct (and I will readily, cheerfully admit to not knowing, as I am not a doctor or scientist or statistician, and I am only trying to explain the argument as I understand it), this puts us in an incredibly unpleasant double-bind.

A suitable solution would be to let the general population know the risks and chances for over treating something that may not necessarily need to be treated (to the detriment of the patient) and let them make their own decisions.
posted by Malice at 12:25 AM on January 25 [2 favorites]

"Professor Julietta Patnick, director of the NHS cancer screening programmes, said: "We can't comment on screening programmes in other countries but here in England we do know that the best evidence available shows that women aged 50-69 who are regularly screened are less likely to die from breast cancer." She cited an estimate from the International Agency for Research on Cancer (IARC) of the World Health Organisation which said mortality was reduced by 35% through screening — a figure Gøtzsche disputes in his book."

"He compares screening advocates to religious believers and argues that their hostile attitudes are harmful to scientific progress. A lot of false evidence has been put forward to claim that the screening effect was large, he writes. Those who tried to expose the errors came under personal attack, as if they were blasphemers."

Religious believers? is this guy for real?
posted by marienbad at 12:30 AM on January 25 [1 favorite]

A suitable solution would be to let the general population know the risks and chances for over treating something that may not necessarily need to be treated (to the detriment of the patient) and let them make their own decisions.

I'd like to believe this, but a health system which relies on its patients being informed and rational about their healthcare and making decisions based on knowledge and reason strikes me as very poor planning. I mean, to make a reasonable decision about whether the costs of treatment do or don't outweigh the risks of harm from the disease itself is something doctors who've spent years in medical school and residency and practice often struggle with- to ask patients, most of whom are either uninformed or in many cases very badly misinformed about medicine and health, to make that decision strikes me as unreasonable.

I mean, ultimately, yeah, it's up to the patient, as you can't dictate that the patient receive medical care that they don't want. But at the same time, shifting the burden of decision-making onto the uninformed/misinformed public seems cruel to me.
posted by Pope Guilty at 12:30 AM on January 25 [26 favorites]

I do have some medical background, Billing and Codeing, they teach you a lot about different diseases. Breast cancer is NOT always genetic. I do not have any relatives who ever had breast cancer. I waited until 57 to be screened due to lack of coverage.
I however have a friend who got breast cancer in her early 30s. She has lots of relatives who got breast cancer pretty young. They as a family run a nursery,it's not organic so perhaps it's pesticide exposure. If so the whole family is going to be
affected. :(.
I think this guy is on shaky ground saying that testing does more harm than good. What if my friend had not been diagnosed or treated? She has two daughters. They need her.
Of zgrada a family history or symtoms, them get checked our sooner. I'd not The standard age or even age 50 is fine.
posted by Katjusa Roquette at 12:35 AM on January 25

Cancerous cells that will go away again or never progress to disease in the woman's lifetime are excised with surgery and sometimes (six times in 10) she will lose a breast.

I can't imagine that the tumors that lead surgeons to do mastectomies (rather than lumpectomies) were tumors that would have gone away or not progressed to disease. Wouldn't a tumor have to be pretty large or spread out for the surgeon to make that choice? If so, throwing those in the same sentence, as if sometimes women lose breasts for tumors that never would have caused them problems, seems dishonest.
posted by needs more cowbell at 12:41 AM on January 25

Katjusa, I suspect screening the close relatives of people who were diagnosed with breast cancer at an early age is much more likely to have net positive outcomes than screening the entire population for breast cancer.

The same arguments have been made about prostate screening: that the negative consequences of treating those who are given a false positive diagnose outweighs the benefits of the program to everyone else.

This would hardly be the first time that a major public health program turned out not to be beneficial.
posted by pharm at 12:44 AM on January 25 [2 favorites]

Oh good lord. This article is the epitome of the worst of cancer research. It is basically saying that the cost is too much because only a small percentage of the screened patients are actually at risk. That doesn't matter, we still pay our insurance premiums, taxes and medicare taxes. I don't care if I have to contribute to a thousand tests that prove false just to get one right, it is worth my money.

I may be biased, as I am a male that had skin cancer at age 20.
posted by Drumhellz at 1:05 AM on January 25 [2 favorites]

The take-away I've gotten is that for lots of women (maybe those age 40 to 50, maybe all women?), current breast cancer screenings have had a surprisingly high number not-helpful results, and an uncertain relationship to positive results (meaning a screening that materially helped). And then maybe there are sub-groups of women with higher risk or a more direct relationship to the benefits of screening. And all of the studies/science is murky and not reassuring. So it can't be totally recommended at this time for every/all women 40 to 50 to receive mammograms, but it can't be totally rejected for some women, and no one's totally sure who's in what group yet. So some people in the field are not exactly sure what to recommend other than to screen (40-50) if you're high risk or worried. But that's a super lay person summary.
posted by ClaudiaCenter at 1:08 AM on January 25 [2 favorites]

"Religious believers? is this guy for real?"

Absolutely for real. For a good explination of why read Welcome to Cancerland by Barbara Ehrenreich, it is a relatively short but powerful article based on her experience with breast cancer

This deuchebag's sensationalist framing aside, the basic premise of his arguments arn't his, and are sound.
posted by Blasdelb at 1:14 AM on January 25 [5 favorites]

Drumhellz. You're completely missing the point.

He is saying that the negative consequences of treating the false positives that arise from mass screening for breast cancer appear to outweigh the positive outcomes in terms of lives saved that would otherwise have died.

This is nothing to do with the expense of screening, which of course *should* always be an issue: if that money could be spent on some other program that would be more effective then the screening program would have cost lives.
posted by pharm at 1:17 AM on January 25 [7 favorites]

There's a lot of good info on this subject there, especially this. This guy is an oncologist specializing in breast cancer who lost his mother-in-law to it.
To sum his views in a sentence - there is certainly a lot of benefit from screening for women after 40 and those with breast cancer history in family (carrying high risk genes). Not so much for the rest of them.
posted by hat_eater at 1:59 AM on January 25 [3 favorites]

I'm in my 30s, with a kind of complicated family history with regards to breast cancer. I recently had a conversation about this with my doctor, and here's what she said.

She said that 40 and 50 are arbitrary numbers. The distinction is between pre- and post-menopausal women. Most women aren't at high risk for developing breast cancer before menopause, and changes in the breast tissue during menopause mean that mammograms are more effective post-menopause anyway. She's not convinced that routine mammograms before menopause make sense, and she thinks that routine mammograms after menopause probably do make sense.

For women who have a family history of pre-menopausal breast cancer, it probably makes sense to do more aggressive screening than just mammograms, like going straight to breast ultrasounds. But this would be a pretty small group of women.

So basically, it sounds like what we need is a more tailored approach to breast-cancer screening, but that requires educated consumers and doctors who have time to get to know their patients and really discuss this with them, and those aren't necessarily the conditions that actually prevail in various health care systems.
posted by craichead at 2:04 AM on January 25 [19 favorites]

Is this perhaps similar of the current rates of detection of prostate cancer in men and, as is maybe turns out, the "have to get rid of the cancer" treatment is worse than just treating the symptoms and otherwise leaving things be? Or maybe like tiny polyps in the gut?

I have good friends and family members who are breast cancer survivors. I'm happy those ladies are still here. But, the ones I know had genetic or chemically induced (fertility drugs, I'm looking at you) predispositions. Of the ones I, admittedly anecdotally, know, they found something suspicious themselves and had to insist on further screening. Then, they were able to get treatment early enough to be successful. And by successful, I mean double mastectomies, radiation and chemo, and gloriously done rebuilds in multiple stages (luckily the insurance happened to cover the rebuilds).

Please don't be offended at the use of "rebuild." I also have relatives who have had knees and shoulders replaced and retinas reatttached. I call them my cyborg family, and they have great senses of humor about it all. My personal opinion is that I don't know if I'm as strong as they are. Those are all amazingly hard things to do. My mom called me after getting her retina reattached each time ("Didn't wan't to worry you"), and my aunt nonchalantly mentioned her second upcoming knee replacement on Facebook a couple of days ago.

I also have a friend who is in hospice, a couple years older than me, who had a chronically painful "female" situation that was ignored by doctors for years, until it turned out she had a tumor the size of a freaking cantaloupe on one of her adrenal glands that was pressing on nerves related to her lady bits and had spread to bones and lungs. A number of experts across 3 states dismissed her complaints as med seeking for years and totally missed it.

I'm told that one of my grandmothers had a noticeable lump when she died at 80-something, but that is certainly not what caused her death, nor would have treating it improved her quality of life at that point. It was a localized thing.

I don't know. I can see cutting out a bad thing, but still that is traumatic to the body as a whole, especially at the age where it hit my gramma. I can't even go into all the chemo and radiation long-term effects. My aunt had to choose between waiting and seeing and almost dying from chemo. My friend who wanted to have a baby, and is a couple of years younger than me, will never be able to, so she also lost a husband. My friend who is slowly dying now, because people didn't listen to her seriously. Granted, I'm SO HAPPY these women are still here in my life, for whatever time. I could not imagine an amount of money I'd trade for them.

On the other hand, I don't think my much beloved Gramma with a lump when she died should've had to deal with the expense, pain, stress, and risk of extra treatments that would have done her no good.

I think it's amazing what we can do to save lives and quality of lives these days. I really, really do. But I also believe in quality versus quantity when it comes to the amount of time we're given on this rock flying through space. Fear does not equal quality.

I don't have a solution. Maybe the older you are, the less you want to be poked at. I know that's certainly true for my Dad (and me, too, as I get older). Maybe we should encourage the younger folks to get looked at more thoroughly.
posted by lilywing13 at 2:04 AM on January 25 [1 favorite]

Is this perhaps similar of the current rates of detection of prostate cancer in men and, as is maybe turns out, the "have to get rid of the cancer" treatment is worse than just treating the symptoms and otherwise leaving things be?

Yes and no. Differences:

Breast cancer is more lethal than prostate (130 women die of breast cancer for every 100 men)

The average years of life lost to breast cancer is 19; for prostate cancer, 9.

Breast cancer is the number one cancer killer of women age 15-54; prostate cancer is not the number one cancer killer of men at any age.

Also, it strikes women *much younger*. 1 in 228 women <39 will get breast cancer. Only 1 in 19,229 men will get prostate cancer at the same age. Age 40-59, breast cancer rates are still nearly twice as high (1 in 24, vs 1 in 45).

That ratio reverses from sixty, though, which brings up the crux of the issue about prostate cancers, namely: most - not all - are typically quite slow growing, and the odds are good that the patient will be dead before the cancer becomes deadly. In addition to invasive etc, the surgery has terrible lifestyle costs associated with it and can frequently result in incontinence etc.

So in this particular incidence - even though conversations about screening are taking place with similar arguments - I don't think a comparison to prostate cancer from a factual/statistical/epidemiological perspective is especially valid.
posted by smoke at 2:39 AM on January 25 [5 favorites]

Cancerous cells that will go away again or never progress to disease in the woman's lifetime are excised with surgery and sometimes (six times in 10) she will lose a breast.

This sounds very strange. I can understand a false positive from the mammography, but there still should be a biopsy to confirm exactly the type of cancer before the treatment. If doctors are prescribing unnecesary surgery for whatever reason, maybe the problem isn't the screening. I don't think they will be removing breasts and giving chemo for something that "will go away".
posted by radiobishop at 3:14 AM on January 25

If doctors are prescribing unnecesary surgery for whatever reason, maybe the problem isn't the screening.

Anecdotally, it is not unusual for the woman who has been diagnosed to be the one pushing for surgery: "I want it out of me/I want it gone" is the driving sentiment.
posted by Ritchie at 3:29 AM on January 25

The problem is, the oncologist often can't tell if given cancerous cells will be kept in check by the immune system indefinitely or if they evolve into a more aggressive and competent form.
posted by hat_eater at 3:34 AM on January 25

How big of a factor are malpractice concerns in all of this? I am under the impression that many unnecessary treatments are done in the US because of the threat of lawsuit. Is it a case of "we shouldn't look for cancer, because if we do, we will see ambiguous evidence that financial prudence will force us to act on, to the patient's likely detriment"?
posted by idiopath at 4:12 AM on January 25 [2 favorites]

this puts us in an incredibly unpleasant double-bind

I don't think so, but I do believe we need better methods for explaining the complexities to patients so they can make their own individualised, informed choices.

One of Gøtzsche's major charges is that national screening programmes rarely present any harms of the investigations they espouse. I guess that's also true of the highly incentivised US system. As some of the responses in this thread demonstrate it is extremely counter intuitive that any kind of medical test could be harmful. I don't believe we will ever get a clear-cut yes/no answer on this topic. The two sides are entrenched, and they each now have enormous intellectual conflict of interest. What we need are novel ways of explaining to people considering screening what the benefits are, what the harms are, and what choices they have.
posted by roofus at 4:16 AM on January 25 [1 favorite]

Well, I am not a doctor but I guess that if you have something big enough as to lose a breast from its removal, it's probably better to remove it anyway.
posted by radiobishop at 4:34 AM on January 25

Mammograms saved my mother's life, and not getting screened resulted in my aunt (who was the same age as Mom and was told, by all her doctors, that her cancer started probably about the same time as Mom's) dying a horribly painful death.

I think both would have considered the screening worth the risk.
posted by xingcat at 4:51 AM on January 25 [1 favorite]

My, that straw man is certainly burning brightly, Dr. Gøtzsche! Good thing you have found a way to sell tickets to watch it burn.

No credible organization recommends screening everyone in the way that he seems to think the NHS does. Most folks use the Gail Model to determine individual risk, but there are other heuristics for deciding who needs to be screened annually and who can get by with checks every 3 or 5 years. That way, we can screen high-risk 25 year-olds without slotting everyone into the same bucket. Perhaps this article is doing a poor job summarizing the book, but I don't see a single word in here about using patient risk as a determining factor in this process, which is do disingenuous it makes my teeth hurt.

But that doesn't sell books. Calling advocates for preventative care "religious zealots"--now THAT moves copies out the door. Meanwhile, the millions of women who die excruciating deaths every year (added bonus: poor and minority women suffer much higher mortality rates!) because they couldn't afford screenings or were encouraged not to do yearly screenings? Unfortunate but unavoidable collateral damage, eh, Doctor?

The vengeful cynic in me wants to see someone close to this asshole miss a diagnosis that would have shown up early on a routine mammogram, but having seen up-close what late-stage metastasized breast cancer does to its victims, I can't even bring myself to wish that on him.
posted by Mayor West at 4:57 AM on January 25 [2 favorites]

The first line of the Hippocratic Oath is "First, do no harm", so IF (big IF) Dr. G's research is correct, (I'm hoping it isn't), than his assertion that routine screening for every woman be stopped is ethically correct.

If I was in a high risk category, I'd get screened regardless of what he said.
posted by Renoroc at 5:13 AM on January 25

I don't see how this should change screening. It might have cause to affect how to treat very early cases of breast cancer. Just because the reaction to what is found isn't ideal, doesn't mean it doesn't need to be found. I'm also in the higher risk category and it's hard enough figuring out if I got screened early if my insurance would cover it, or if I'd have to pay out of pocket for one of the genetic tests.
posted by ejaned8 at 5:41 AM on January 25

Here's the thing: for someone who does develop breast cancer, regular, early screening is likely to have a large, positive effect on their health. But for someone who doesn't develop breast cancer, regular, early screening is likely to have a minor, negative effect on their health.

There are far, far more people in the latter category than the former.

This is an issue of risk management. Risk is evaluated in terms of both frequency, i.e. how likely a particular thing is to happen, and severity, i.e. how bad it will be if it does. Given those two factors, we get four quadrants:

- high frequency, high severity
- high frequency, low severity
- low frequency, high severity
- low frequency, low severity

Breast cancer is in the third category, i.e. it's relatively uncommon (the US has an incidence rate of about 125:100,000), but it's pretty severe, i.e. the people who get it tend to be pretty sick.

But regular screening is a tool that falls in the second and third categories, i.e. it commonly causes some minor negative health effects and occasionally causes major negative health effects. The minor effects are the effects of the screening itself, but the major effects are the results of false positives.

Thus, from a perspective of minimizing total risk, regular screening isn't obviously the thing to do. The question the author and others are wrestling with is if we have the numbers to tell us which way the thing tips. That isn't really a scientific or objective decision either, as risk management decisions have as much to do with personal risk tolerance and evaluation as they do with anything else. The author is suggesting that the way the numbers work out is that more women wind up getting a false positive and having negative outcomes as a result than actually wind up benefiting from regular screening. But because the health profession needs to recommend something, we have these debates.
posted by valkyryn at 5:42 AM on January 25 [10 favorites]

Meanwhile, the millions of women who die excruciating deaths every year (added bonus: poor and minority women suffer much higher mortality rates!) because they couldn't afford screenings

The number of breast cancer deaths in the US per year is about 40,000. In the UK, it's 12,000. Worldwide figures are harder, but the WHO estimates 460,000 deaths per year.

In 2001, the NHS statement on this was "We estimate that screening now is saving on average 1,250 lives a year." So the stakes in this decisoni are being measured in thousands, not millions.

There's no point in having an opinion about this subject without reading the actual studies, so I don't have an opinion. But I will say that the NHS response in the Guardian is somehow off-point:

"We can't comment on screening programmes in other countries but here in England we do know that the best evidence available shows that women aged 50-69 who are regularly screened are less likely to die from breast cancer."

Surely Gotzsche would agree with this. As I understand it, his argument is that in the current system, women who are regularly screened are less likely to die from breast cancer, but more likely to die from unnecessary surgery. And he would accuse Mayor West of callously dismissing those dead women as "unfortunate but unavoidable collatoral damage."

As I said, I have no idea who's right about this. But that's what's being argued.
posted by escabeche at 5:47 AM on January 25 [9 favorites]

I did not read this article, but I know from first-hand research experience that Gøtzsche seems to have made a name for himself for controversial stances, often pie-in-the-sky unrealistic proposals, and has a tendency to forgo any sense of diplomacy. He practically seems to revel in rabble-rousing, especially about particular pet issues. Which isn't to say he's not a smart, generally well-respected researcher.

He wrote a highly accusatory letter to the editor about one of my papers that essentially accused us of scientific misconduct, which was completely unfounded, as most of his points clearly described and/or refuted in the text of the article. It was as if he barely read the paper. We responded with a smackdown that was professional, but which nonetheless had pretty obvious subtext of a big STFU.

I never understood why the editor chose to publish his letter given easy response -- perhaps it was precisely to have him smacked down in a journal.
posted by ssmug at 5:48 AM on January 25 [6 favorites]

for someone who doesn't develop breast cancer, regular, early screening is likely to have a minor, negative effect on their health.

I think we need to clarify what your definition of minor is. Because surgery followed by months of chemotherapy will essentially take months and years away from ones life. (Those months spent doing nothing but walking from the bed to the bathroom and back, if I was unclear.)

This doesn't even include the psychological impact of having parts of your body removed. Parts, which for a large number of women, have a serious impact on ones social development and identity.
posted by Blue_Villain at 6:02 AM on January 25 [1 favorite]

Canada recently released new guidelines calling for fewer mammorgrams -- you can read about them in the Globe and Mail here.
posted by cider at 6:12 AM on January 25 [1 favorite]

Why is it that the screening produces so many false positives, and why are there so many cases where surgery, chemo, and radiotherapy go ahead even though it would in fact have been better to hold off? I may be naive but can anything be done about those issues, rather than dropping the idea of screening?
posted by Segundus at 6:42 AM on January 25

Why is it that the screening produces so many false positives, and why are there so many cases where surgery, chemo, and radiotherapy go ahead even though it would in fact have been better to hold off?

"False positive" seems to be tripping people up here. In this context, it doesn't mean getting a mammogram that shows you have cancer when you in fact have no cancer; what it means is that the mammogram shows a lump of abnormal, cancerous cells that would never had killed you if you left it alone. So the biopsy and surgery will likely be considered a success from the perspective of the surgeon and the patient, because at this point in time we are unable to differentiate between lumps of cancerous cells that will basically sit there and not do much (or even disappear) and those that will metastacize and kill. Most people don't even realize it's possible to have a cancerous tumor that won't cause problems if you don't do anything with it.

At a very high level, these studies work by looking at population data and seeing that (just making up numbers here) when nobody was screened or treated, 10K women a year died of breast cancer; after we start screening everyone, we discover 20K women per year with breast cancer and treat them all; and now 12K women per year are dying of breast cancer or complications associated with the treatment. (Note that in this situation, all 8K women who received treatment and did not die would consider themselves lucky to have received screening and "dodged a bullet," despite the fact that some large portion of them probably underwent treatment for something that would never have killed them.)
posted by iminurmefi at 6:55 AM on January 25 [12 favorites]

Mammograms saved my mother's life, and not getting screened resulted in my aunt (who was the same age as Mom and was told, by all her doctors, that her cancer started probably about the same time as Mom's) dying a horribly painful death.

This might be true, but isn't generally true for all women who do or do not attend screening. I can directly counter this anecdata with direct experience of two relatives, one of whom had a self-detected tumour just months after a negative mammogram, and another who had a mastectomy for a DCIS that she now believes might never have progressed.

The vengeful cynic in me wants to see someone close to this asshole miss a diagnosis that would have shown up early on a routine mammogram

This is the same blind-to-the-facts "logic", typical of US healthcare, that continues to see mass screening for prostate cancer and promotion of whole body CT screening, even though we know those cause harm.

I will do anything recommended to me, to make sure I live longer

Doctors who are incentivised to provide screening, are not in a good position to provide unbiased advice.
posted by roofus at 6:59 AM on January 25 [1 favorite]

I read a very interesting book on this subject last year--Should I Be Tested For Cancer? Maybe Not, and Here's Why--that discussed all these issues in much more depth. imihurmefi just talked about some of the findings I found most interesting--the idea that improved screening has led us to treat tiny, slow-growing cancers that probably don't need treatment because in many cases the person will die of something else before the cancer ever causes a symptom. Since reading the book, I keep seeing news articles, especially about prostate screening, that discuss this kind of problem.

Why is it that the screening produces so many false positives, and why are there so many cases where surgery, chemo, and radiotherapy go ahead even though it would in fact have been better to hold off?

In Should I Be Tested, the author talked about the factors that lead doctors to prescribe treatment rather than "wait and see" when screening shows something suspicious, or small cancers, including the desire to be doing something rather than nothing, to appear to be doing everything possible on behalf of the patient, and fear of the risk--even if relatively small--that taking a wait-and-see approach will lead to a bad outcome, in which case the doctor could be open to a malpractice claim.

I used to have a doctor who really pushed me to start having mammograms at 35, despite no family history of cancer. She undermined her argument when she told me that I needed to be screened because cancers could grow so fast--her father, also a physician, had discovered her mother's breast cancer just one month after a clean mammogram! I told her that that hardly seemed like an argument in favor of mammogram to me.
posted by not that girl at 7:10 AM on January 25 [2 favorites]

It seems totally clear to me that we over-treat prostate cancer and that a lot of men would be better off not getting their prostate cancer treated at all. But is the same true of breast cancer? My sense is that breast cancer is a lot more lethal than prostate cancer. Is there evidence that we're treating a lot of breast cancer that would probably not kill the patient if it were left alone?

I think part of the problem with breast cancer screening is that there's a high number of true false positives, and those require expensive and invasive additional screening to determine that they're not actually cancer. All that testing drives up the cost of health care, which makes it really difficult to fund an effective universal system. That's part of the reason that the US pays so much for health care that doesn't cover everyone, and it's one of the things that is causing challenges for universal systems in other countries.
posted by craichead at 7:21 AM on January 25

Part of the problem is that we don't have very good ways of determining how aggressive a tumor is. Right now it's mostly based on size, how encapsulated it is and if it has spread to lymph nodes. With breast cancer there are a couple of additional factors: whether it's a HER2 expressive tumor or a triple negative - factors which have significant bearing on what kind of chemo is appropriate. The more we learn about what genes and proteins a particular cancer expresses the more often it's possible to tailor treatment more appropriately. So you combine recommendations about possible less screening because from a public health perspective that might be better although not necessarily safer from an individual perspective; questions about family history, carcinogen exposure and other risk factors and it gets confusing fast for a lot of us. If you're not well versed in this stuff - and how many people who don't have a family history or a professional background are - it seems harder to make a good decision about how much screening is a minus.

Personally, given a lousy family history I'll keep up with the annual mammograms.
posted by leslies at 7:41 AM on January 25 [1 favorite]

I wish he wasn't framing this so confrontationally, because the basic argument that over-aggressive use of radiation treatment and surgery might cost lives when compared to a more targeted approach based on individual risk factors does not strike me as hugely controversial.
posted by Holy Zarquon's Singing Fish at 8:09 AM on January 25 [1 favorite]

One more difficulty is that everyone says we need to keep health care costs down but virtually any time you "give patients the choice" they will opt for more screenings and treatments which will be very expensive. So calls to simply present the facts and let patients decide aren't really sustainable in the long run; there have to be recommendations based on cost-benefit analysis. That's why patients don't start getting colonoscopies at 18 even though some lives really would be saved. Because it wouldn't be very many lives, it would be expensive, and there would be even more people who were negatively affected.

It's not beyond the realm of possibility that regular mammograms are in a similar boat; more expensive than can be justified in terms of lives saved or negatively affected.
posted by Justinian at 9:18 AM on January 25 [1 favorite]

Perhaps people wouldn't be so panicked about the valid risk assessment here if the medical establishment didn't have a long history, still ongoing, of dismissing women's health complaints as trivial and not worth treating.
posted by Electric Elf at 10:16 AM on January 25 [5 favorites]

One more difficulty is that everyone says we need to keep health care costs down but virtually any time you "give patients the choice" they will opt for more screenings and treatments which will be very expensive.

Actually, this is surprisingly not that true, at least when you look at some of the work that has been done around improving counseling for prostate cancer screening and treatment, usually called "shared decisionmaking." It actually doesn't take very much unbiased information at all (pamphlets, short video) to dramatically cut the proportion of men who want to receive PSA tests to screen for prostate cancer. In fact, these tools have been shown to reduce all sorts of costly medical procedures by something on the order of 25%, interesting article here.

I found that very surprising when I first read it, but on reflection it really mirrors what we already know about high-cost medical care at the end of life: most people say they don't want it and would prefer to die at home, don't want extraordinary measures taken to keep them alive when they don't have a chance of recovery... yet in many areas more than 90 percent of people die in the hospital. I don't think it's fair to say it's patient preference driving it, it's a more complicated set of factors that lead people to make a bunch of short-term decisions that are out-of-whack with what they want long-term, because they never get a chance to sit back and really consider the big picture.
posted by iminurmefi at 11:05 AM on January 25 [2 favorites]

Perhaps people wouldn't be so panicked about the valid risk assessment here if the medical establishment didn't have a long history, still ongoing, of dismissing women's health complaints as trivial and not worth treating.

Cite? A local hospital here, Hoag, recently spent millions on a massive wing completely dedicated to women's health. This does not include a large section of the hospital dedicated to pre and neo-natal care.
posted by Brocktoon at 11:06 AM on January 25

I think that there needs to be a balance, you cannot require everyone to have a test every year. But also allowing people to come when they please may have some slip through the cracks. Since we will all die eventually anyway, this isn't a complete travesty but when it is your own loved one then it is. So I believe something like testing every 2 - 3 years is a good benchmark. But you don't want to die from all the radiation if you go twice a year.
posted by BigK at 1:03 PM on January 25

As I read the comments in this thread, I was surprised no one mentioned the Norwegian study of mammography and breast cancer from 2009. I remember how amazed I was to learn that many cancers seem to go away by themselves. This study changed my own thinking about the value of yearly mammograms.

As Dartmouth Medicine says,

The study compared the number of invasive breast cancers in two nearly identical groups-each with about 100,000 Norwegian women aged 50 to 64. The first group was followed from 1992 to 1997; the second from 1996 to 2001. The key difference between the two periods is that before 1996, Norway had no national breast-cancer screening program, while after 1996, all women aged 50 to 69 were offered a mammogram every two years. So the women in the first group had only one mammogram, at the end of that six-year span. But the women in the second group (the screened group) had three mammograms during that six-year span.

Given the conventional thinking about invasive breast cancer-that it always progresses-one would expect that the total number of cancers detected in each group at the end of six years would be about the same. It wasn't. The incidence of invasive breast cancer was 22% higher in the screened group. (emphasis mine - Kristi)

So the personal cost - not the financial cost, but rather the health cost to individual women - of increased mammograms is a significantly higher rate of treatment for cancers that were not, in themselves, harmful.

Of course, as the Norwegian scientists point out, we don't yet have good tools to know which will become harmful and which will not.

While searching, I came across an excellent editorial, More Mammography Muddle: Emotions, Politics, Science, Costs, and Polarization" (or PDF version), published in the journal Radiology in May of 2010. It discusses the controversy in the wake of the U.S. Preventive Services Task Force (USPSTF) recommendations that regular mammogram screening happen every two years starting at age 50.

They quote the USPSTF:

... The USPSTF reasoned that the additional benefit gained by starting screening at age 40 years rather than at age 50 years is small, and that moderate harms from screening remain at any age. This leads to the ... recommendation against routine screening of women aged 40 to 49 years.... The decision to start regular, biennial screening mammography before the age of 50 years should be an individual one and take patient context into account, including the patient’s values regarding specific benefits and harms...

(The American Cancer Society still recommends yearly mammograms for women starting at 40, which is pretty different from every 2 years starting at age 50.)

As the Radiology editorial points out,

Recent appreciation of the frequency of premalignant, in situ, and even indolent low-grade cancers that will never progress to symptomatic or life-threatening disease has influenced everyone’s approach to cancer screening (43,44,62,63). Recent estimates are that up to “one in three breast cancers detected in a population offered screening is over- diagnosed” (62). The analogy between the overdiagnosis of breast cancer with mammographic screening and that of prostate cancer with prostate-specific antigen screening is well known. As one researcher observed, “There are many studies showing that mammograms find low-risk, well-behaved cancers preferentially and miss the bad actors preferentially, [with] paper after paper saying mammograms pick up cancers that don’t need to be found” (45).

Completely independent of the very real issue of dollars spent on mammograms, many scientists are genuinely concerned about the serious negative consequences for individual women of over-testing and misdiagnosis.

What is the best recommendation for routine mammogram screening? I think it's an important question - one we should examine as fully and impartially as we can.
posted by kristi at 2:15 PM on January 25 [3 favorites]

my mother's cancer WAS discovered by her but her doctor dismissed it ... So, excuse me, but I will do anything recommended to me

Isn't the lesson here that doctors can be wrong and know less about your situation than you do, and that you should NOT necessarily do anything they recommend if you want to be maximally healthy? Which is exactly what you mother did and exactly what you're doing and good for both of you and I'm sorry that you lost your mother (but your language is confusing and I have way too much time on my hands, so...). If your mother had followed every recommendation her doctors had given her, she would have done nothing. It's possible that they told her to do nothing because they didn't want to look carefully enough for some reason, but based on what you wrote I think it's likely that they told her to do nothing because, given the resources (mostly time) that they had available for diagnosis (not treatment) and all the information available to them, "do nothing" was the treatment with the best outcome for the population that they thought she belonged to.

"Do nothing" is a positive, active treatment recommendation. If you follow it, you might suffer and die or you might live and avoid suffering that you would have experienced if you had ignored the recommendation. If the question is "do X or do nothing" and "do nothing" has a better expected outcome for you than "do X," then you should do nothing. If "do nothing" has a better expected outcome for a population than "do X," then the population on average should do nothing. That is true even when X is a good idea for you, or when X sometimes saves lives, or when X would have a different outcome if it was presented differently, or if people reacted differently to it than they actually do (given a test with a 99% probability of an inconclusive result and a 99% probability that given an inconclusive result the patient will get a treatment that they would not have gotten if they had not taken the test and that will reduce their health for no gain even if you explain everything to them and refuse to do it yourself, do you give them the test? Depends on whether you prefer the patient's better health over covering your ass and letting the patient or another doctor or anyone else but you take responsibility for the patient's worse health), etc.

The choice in question is "recommend screening for everyone," or "do not recommend screening for everyone." "Do not recommend screening for everyone" is not the same as "do not recommend screening for anyone." The question is not "who should get screened?," or "how can we educate the public about the fact that not every cancer that shows up in a mammogram will kill you," or "how can we tell the difference between a breast cancer that will kill you and one that will not?" Those are all important questions, but they are questions about breast cancer and breast cancer treatment. The question of whether or not regularly screening a population for breast cancer improves that population's health outcomes is not a question about breast cancer treatment: is a question about public health. If you try to make it be about breast cancer or about your own personal situation, then you are working to impose a public health cost on that entire population for your own personal benefit or for the benefit of a sub-population. Not a money cost (although it may be that too). A public health cost. It is exactly the same as buying your medicine from a huge multinational that dumps chemicals in the ocean that poison the fish that feed millions of brown people (brown because on average my pale reader, you are probably white) in a country you've never heard of. You don't know about it because you don't care to find out because you want it to be the company's responsibility, not yours. Meanwhile the company doesn't care about the brown people or about you; they are busy maximizing shareholder value, because that is what they are legally (and perhaps ethically) required to do.

Of course this is also exactly why you should always be as informed and involved as possible on questions of your own personal health, and why you should never simply do whatever your doctor recommends because your doctor "knows more" than you do (or for any other reason). You doctor is not and ethically should not be trying to maximize your health outcome. Your doctor (assuming he or she is good doctor!) is attempting to maximize the health outcomes of a population, i.e. all of their patients. Taking care of you and taking care of the population as a whole are always in conflict because resources are always limited: given infinite money, perfect knowledge, and infinite time, a single doctor would still be limited in the amount of treatment they could give any one member of a population so long as their limbs were still limited by the speed of light and the population was reasonably large and had conditions that proceeded at the usual rates. Ergo, doctors do not follow you around and take down your medical history as you grow up, studying your body and learning how it works and keeping track of every little thing that happens to it, even though that would go a very long way towards helping them preserve your health. Instead they go to school and learn about the human body from textbooks and cadavers and random patients in random hospitals and they write research papers about large populations, because they are public health professionals. When they sit down with you, they learn a little bit about you and that gets sifted and sorted and contextualized and broken down and reformulated by the years and years of blood, sweat, and tears that they've put into learning about the general case and every single patient they've ever personally seen. It takes a huge amount of effort for a doctor to even see you as an individual and not just a member of some population that they try to slot you into as quickly as they possibly can so they can move on and deal with the next member of population XYZ. If you want a good outcome for the unique snowflake that is you, then you have to do your bit too.

There's another reason why you shouldn't always follow a doctor's recommendation, and that's because doctors don't always do the right thing, let alone give you the correct answer. That's because when they move on after slotting you, it's often to a member of population med seeker, who dominates their life by screaming and yelling about not getting their meds because the only socially "acceptable" way to get what they want is to harangue doctors and "raise awareness" until the doctor gives in to popular demand and "recommends" them a "treatment." There is a chance that your doctor will subsequently slot you into that group and consequently recommend you a treatment because it's popular, not because it improves the expected outcome of people in that group (independent of whether you should be in that group).

Should a med seeker who "raised awareness" (whether by organizing a million man march or by screaming in the lobby) take meds that have been subsequently prescribed to him because the doctor recommended it? Not if he wants to be healthy (not all med seekers are prioritizing health, but at least some of them think they are). Should the doctor be blamed for making a bad recommendation? Yes. Why? Because doctors should make recommendations based on their best opinion about what will improve health outcomes, not based on popularity. Should med seeker be blamed for harming his own health? Yes he should. Should patient zero do nothing because the doctor misdiagnosis and recommends that she do nothing? Not if she cares about her health and thinks the doctor may be wrong. Should she harangue the doctor for a specific treatment and "raise awareness" until she gets it? No, she should seek out as much information as she can from as many sources as she can so as to maximize the probability of finding information and treatment (if necessary), from a doctor or from any other source, that will resolve the situation to her satisfaction. Should she blame doctors for misdiagnosing her if she does not seek more information and she gets sick? If her goal is to change medical practice (for better or for worse, but most likely better of a sub-group that includes her and worse for the population as a whole), then yes. If her goal is to improve her own health, then no. Should she publicly shame med seekers? Maybe, maybe not (it might not do much for her health either, depending), but doing so would probably help the sick public (albeit indirectly, by making doctor's lives easier) more than raising public awareness about her own condition (without regard for its relevant importance as a public health issue) would. But then we don't expect the average individual to value the public interest above their own. That's what doctors are for.

(NB: I am not a doctor. These are my opinions and obviously not medical advice, etc.)
posted by the atomic kung fu panda bandit inquisition at 2:20 PM on January 25 [2 favorites]

The sad reality is that medical science is still severely lacking. The reason why over-treatment has bad health consequences, is because it raises mortality rates. It may surprise some to to learn that cancer cells arise in the human body constantly. There never is a time when you don't have some cancer cells circulating - 24/7. It's just that the body - usually - gets rid of them. Often the cells manage to form a tumor - not a benign tumor, a malignant one. But - and here's what trips people up - the immune system gets rid of the tumor. It's called spontaneous regression, and it happens with great frequency. So when people say "but medical science can't tell apart the tumors that will metastasize and the ones which are benign", this is only half the problem and only one variable. The other variable and the other half of the problem is that because exactly the same degree of aggressiveness in a tumor may have completely different outcome, in one case, the immune system manages to get rid of it, and in another it doesn't.

Here's how to think of it. The immune system is like the cops. The cancer is like the criminals. The degree of criminality is one issue - you always want to promote conditions that decrease crime, so eliminate poverty, wrong drug policy etc. (bad nutrition, lack of exercise, cancer-promoting habits like smoking etc.). But no matter how careful you are, you will never eliminate crime - or cancer in your body. The cancer can be of varying degrees of malignancy, criminals may be vicious or petty. But that's only half the issue, and the criminals by themselves don't rule the town. The other half is the cops - the immune system. It can handle both benign and malignant cancer. Knowing the degree of malignancy is not enough. Because if the cops can handle it - the immune system - then you are free and clear.

You have cancer cells in your body this very minute - and probably tumors as well - of varying degrees of malignancy. But your immune system is also working - and most of the time, it eliminates or controls the cancer cells and tumors. It's like cops and criminals. Crime is present all the time - it's always there, and always will be there, but most of the time the cops have a handle on it.

So what's bad about over-treatment? The discovery of tumors - which may be malignant, and histologically seemingly highly aggressive - but which the immune system would have eliminated or controlled. Now how do you respond, once you have discovered said tumor, since there is currently no way at all to tell if a small tumor will go on to kill you unless treated, or will go away on its own under the assault of the immune system?

Why not eliminate all tumors discovered, never mind the immune system? Because treatment is not merely costly, stressful and so on, but is frequently oncogenic (cancer-generating) itself. That may be obvious in the case of radiation therapy or chemo - both can cause subsequent cancer, but what's far less well known, is that surgery can also be oncogenic! This can be for a variety of reasons. Sometimes the surgeon's scalpel spreads the tumor cells outside of the encapsulated cancer. But it's often due to other factors - trauma is associated with tumor growth. And finally, healing from surgical incisions can generate aggressive cancer, because in order to heal, the body generates rapid growth of cells hormones and various growth factors, which sometimes gets out of control and causes a de novo cancer.

So you may have a tumor go away (regress) completely on its own. But the doctors found it, and by treating it, generated cancer down the road which kills you. That's why over-treatment is dangerous and can increase mortality compared to no treatment.
posted by VikingSword at 7:04 PM on January 25 [4 favorites]

People may be interested in this CBC Ideas program "You are Pre-diseased" which looks at the issue of cancer screening and mammograms in detail, and questions their benefit at a younger age.

(Full disclosure I know the program author, Alan Cassells, but it is directly on this topic).
posted by chapps at 6:00 PM on January 26