William M. Briggs

Statistician to the Stars!

Breast Cancer Screening Ruins Lives?

Peter GotzscheToday’s title is from the Daily Mail: “The expert branded a woman hater for saying breast cancer screening ruins lives.”

What could be more sensible than having a mammogram?

If a tiny tumour is growing in your breast, you want to find it as soon as possible and treat it before it has a chance to spread and become life-threatening

That is the reasoning used by most women. But it doesn’t satisfy Peter Gotzsche, a Danish physician with statistical sympathies, and author of the “controversial” book Mammography Screening: Truth, Lies and Controversy.

The standard reasons given for young, healthy, asymptomatic women to receive breast cancer screening don’t satisfy yours truly, either. I wrote about this in the Decision Calculator and
in the peer-reviewed (which guarantees its accuracy, right?) paper “Assessing the skill of Yes/No Predictions” in Biometrics (pdf).

From Gotzsche’s book introduction:

The most effective way to decrease women’s risk of becoming a breast cancer patient is to avoid attending screening.

Mammography screening is one of the greatest controversies in healthcare, and the extent to which some scientists have sacrificed sound scientific principles in order to arrive at politically acceptable results in their research is extraordinary. In contrast, neutral observers increasingly find that the benefit has been much oversold and that the harms are much greater than previously believed.

When Gotzsche “looked at the figures produced to mark the 20th anniversary of the UK screening programme, Professor Gotzsche’s analysis suggested that for every 2,000 women screened regularly for ten years, just one will benefit from the screening.” Even worse, “At the same time, ten healthy women will, as a consequence, become cancer patients and be treated unnecessarily.”

We’d thus expect that since Gotzsche claims and has documented inefficiencies, errors, misconceptions, and the harm that can result from screenings, he has been lauded far and wide, particularly by women’s groups. Right?

Instead, as the DM reports, “for almost ten years there has been a concerted campaign to discredit him, while scanning authorities in the UK, U.S. and Europe have done little to address his criticisms.” When his first paper on the subject was published, “Experts in the screening industry expressed ‘dismay’ that it would erode public confidence in screening and urged women to ignore it.” It was said that those who recommended against screening “don’t like women.”

Another researcher, thinking of Gotzsche’s swimming against the tide of Consensus, wrote, “What is remarkable to me is that this man calls himself a scientist since he obviously, knowingly ignores the scientific method to further his own agenda. I cannot believe his [sic] is so intellectually deficient.” Now where else have we seen this kind of calm, rational discussion of scientific matters? Ah, skip it.

The DM does a good job summarizing one of the ways screening can be harmful:

But what can be wrong with checking to catch breast cancer as early as possible?

If every tumour that showed up on a mammogram eventually spread around the body, no one could object. But cancer isn’t like that.

‘It is a biological fact of life that we cannot avoid getting cancer as we get older’ says Professor Gotzsche.

‘It’s so common nearly all middle-aged people will have some sign of it and most of them will die without having had any symptoms as a result.’

In other words, scanning finds cancerous changes that would otherwise never have caused a problem in your lifetime.

But once a mammogram picks up something that might be a tumour, you’re on your way to becoming a cancer patient because there are no reliable ways of telling if you’ve got the slow-growing or disappearing type, or if it is going to become dangerously invasive.

You will be sent for a biopsy and, if it’s cancerous, you get the full cancer works — surgery, chemotherapy and radiation, and possibly have your breast removed.

Thousands who would otherwise have remained perfectly healthy — because their cancers would never have caused a problem — become cancer patients.

And there is the chance of a “false positive”, when the screening incorrectly says a woman has cancer when she does not. This leads to angst, worry, stress, possibly even tears and fretting. And to further, possibly more intrusive, screenings to verify the first.

Screenings has costs and not just benefits. Each woman needs to weigh those costs and benefits for herself and her family—and she must realize that those costs and benefits are not the same as the doctors or the women groups advocating screening.

14 Comments

  1. It’s a tough call. People give up eating salty foods when there isn’t a history of heart problems in themselves or their families. The cost is learning to live with bland food unnecessarily (assuming it is ever necessary). But the cost of not attending to what looks like cancer is more than a tad higher. To some, their is no higher.

    I ran an informal poll today. I didn’t keep track of the numbers but nearly all shrugged and said “so what?” — effectively saying that the cost of being treated for cancer was far better than dying from it. Partly, I suspect, because they don’t really know how bad the treatment is.

    People really freak out over health. It comes with increasing longevity. The Roman had their health fads, too. They were just different. Over-reaction comes with the territory. In the month following some reported “interesting” disease in any given health article and you will suddenly find a lot of people who have “had” it. Having received a positive result could propel one onto the Cancer Survivor list — a kind of Purple Heart.

    Is it surprising that the people who stand to gain monetarily from increased breast cancer awareness are the most vociferous here?

    BTW: the reasoning behind the reaction I received likely also accounts for the necessity to remove one’s shoes before boarding and airplane.

  2. Interestingly, my mother was treated for stage zero breast cancer — stage zero! Beyond the angst, there were the radiation treatments that are still not without possible effect. Of course, the test and treatment resulted in a real monetary payoff for certain individuals in the healthcase industry (industry is an apt term in this instance). So, most certainly, each individual (patient?) must evaluate her own payoff.

  3. The comparison with salt is rather lame: all but a tiny proportion of the population just pisses away any excess salt. Any animal that hadn’t evolved such abilities would be extinct.

  4. As we should know all too well – recent rant by Ashley Judd, anyone? – this particular target malady has a strong, emotional pull on self-worth and body image issues facing more or less 122% of today’s adult women. Thus for the preventative health care industry (yes, a tellingly accurate term) moguls it is a golden egg producer. Of course they won’t stand for Dr. Gotzsche demeaning their goose. Facts on this issue are unimportant. It’s “feelings” that keep the cash registers churning out tape. Lets be sensitive out there, everybody.

  5. ‘It is a biological fact of life that we cannot avoid getting cancer as we get older’ says Professor Gotzsche.

    ‘It’s so common nearly all middle-aged people will have some sign of it and most of them will die without having had any symptoms as a result.’

    ———

    I guess the doctor hasn’t gotten a memo that cancers are likely caused by chronic vitamin D deficiency… It won’t be for years before it becomes official though…

    http://www.biochemj.org/bj/441/0061/bj4410061.htm

    http://www.vitamindcouncil.org/health-conditions/cancer/an-introduction-to-cancer/prevention/

    “There has been one successful RCT of vitamin D supplementation with respect to cancer incidence. In a study involving post menopausal women living in Nebraska, those taking 1100 IU/day of vitamin D3 and 1450 mg/day of calcium had a 77% reduction in all-cancer incidence between the ends of the first and fourth years, while those taking only the calcium had a 40% risk reduction18. Calcium seems to reduce the risk of cancer independently from vitamin D19. Those taking vitamin D supplements in the Lappe et al. study increased their serum 25(OH)D levels from 29 ng/mL to 39 ng/mL.”

  6. And yet, overexposure to sunlight, hence more vitamin D, results in skin cancer. Of course, it’s only the supplements that prevent cancer occurrence.

  7. Calcitriol, don’t worry, before those few years are up, I am certain we can produce some peer reviewing that shows that Vitamin D is causing just as many cancers.

  8. There is the problem of overdiagnosis and overtreatment, but prostate cancer would be more fitting for this post, I think.

    It is a biological fact of life that we cannot avoid getting cancer as we get older’ says Professor Gotzsche.

    I guess this statement can’t be wrong, can it? We simply can’t avoid getting cancer as if there is a vaccine for it. And they say that men would sooner or later have prostate cancer IF they live to be more than 100 years old.

    Cancer screening saves and ruins lives, though I am not sure what “ruins life” means. I have a tight-knit group of Chinese women friends who meet regularly for an entire afternoon. Two of them had breast cancer in their mid-40s. One had regional recurrence after primary treatment. The other one always has a smile on her face. I doubt that their lives have been ruined. Well, I guess if they didn’t have insurance, the cancer could incur financial burden.

  9. My mother (In her 80s), had her 5-yearly cervical smear test. She was called back due to “abnormailities” being found for a retest & some laser treatment.
    She was worried sick for a couple of weeks, between original results & treatment.
    The whole area of cancer screaning, is highly politicised.

  10. The screening fallacy was my first introduction to Bayesian statistics, to whit:

    You have a test with a 1% false positive rate.

    You have a population with 0.1% incidence rate.

    You need to test 1,000 people to find one real positive.

    In testing 1,000 people, you will find 10 false positives.

    You have to consider the fact that every person taking the test is an ‘unknown’ prior to the test to see this result – (which is where the Bayesian stuff comes in).

    When you tell/show people this, even the epidemiologists are stunned – and they are supposed to be professionals!

    Unless false positive rates are equivalent or lower than incidence rates they are actively BAD for you as at the very least they waste resources in follow-up and cause serious distress. If there is, further, little potential for the disease to affect your life, then a positive test result simply makes you a “sick person” for no reason. Breast cancer and prostate cancer are the two biggest problem areas for this reason (the treatment is sometimes a worse outcome than the no-treatment option). Screening for other cancers may be less problematic, but implementation of screening should take these factors into consideration.

  11. I’m a female MD and, although in my practice, I do not do primary care, and thus do not make recommendations regarding screening for others, I did myself begin getting annual screening mammography at age 40. This is because I believe it may possibly save my life and is very unlikely to hurt me. Most female docs, I believe, also share this view, and don’t recommend mammography for others merely because they are self-serving.

    Some “cancerous changes” (to use Dr. Gotzshce’s term, it’s kind of vague) detected by mammography are going to be insignificant, and there is a lot of uncertainty at this end of the spectrum. But mammography also often is the first means of detection of frankly invasive tumors. By definition, these are differentiated from non-invasive tumors by histologic examination, and this differentiation has predictive value, i.e. it can be told, with some certainty, that many tumors found on mammography are “dangerously invasive.”

    The benefit of mammography is potentially not simply a matter of survival alone, as early detection almost always means less treatment, which means better quality of life for survivors. The statement that “if it’s cancerous, you get the full cancer works” is false. Stage really matters in determining how much and what kind of treatment in recommended.

    Also, mammography is not a yes/no test, and does not purport to be one. The final assessment of the report will be in one of 7 categories, ranging from “negative” to “highly suggestive of malignancy,” and the full report will include much more information than just this.

  12. “But once a mammogram picks up something that might be a tumour, you’re on your way to becoming a cancer patient because there are no reliable ways of telling if you’ve got the slow-growing or disappearing type, or if it is going to become dangerously invasive.”

    According to my pathology, I had the fastest growing type tumor (grade 3). My tumor was 4 cm and had spread to lymph nodes. Isn’t that reliable enough to know that I had the dangerously invasive type?

    But then again, I found the lump myself last year when I was 31 years old. I’d never had a mammo.

  13. Gerd Gigerenzer coivered this area in his book Reckoning with Risk. That came out 10 years ago: it seems that people do not learn.

  14. The latest excuse for ignoring the better cancer survival rates in the U.S. is that it’s an artifact caused by increased screening. Look for more criticisms of cancer screening as the Obamacare debate continues.

Comments are closed.

© 2014 William M. Briggs

Theme by Anders NorenUp ↑