Articles

“How do you feel about Evidence-Based Medicine?”

That was the question asked on a Medscape Connect discussion

I did a double-take. How do you feel? Could anybody object to the idea of basing treatments on evidence? The doctor who started the discussion asked:

Besides using EBM, a lot of my prescribing comes from anecdotal experience and intuition. How about you? Where do you get your information from that you use to treat your patients? Do you always ascribe to EBM, or do you deviate from it with certain medical conditions/patients?

I had naively thought that my profession uniformly embraced EBM. How could they not? The commenters broke my bubble big-time. Some of them summarily reject EBM… although it appears that what they are rejecting is not what I understand EBM to mean.

Note: I have excerpted the following quotations from the website, preserving the errors.

Some typical comments

  • EBM is touted by ivory-tower bozos
  • EBM is a pile of data-computer generated. You do not know which data are true or false.
  • The science that we push has money attached to it.
  • only 15% of what we do has any evidence base.
  • I would like EBM better if it actually worked
  • EBM: the biggest smoke screen in medicine
  • Much of the “evidence” today is fabricated and doctored by Big Pharma
  • hard working physicians are screwed from all directions even from statisticians who sit on rear end all day talking to computers and have all the time to pontificate without sweating looking after a dying patient or taking care of a bleeder that wont quit!
  • Find a paper that says one thing, and I will find another that refutes
  • Experience trumps EBM
  • I see EBM as a way of “dumbing down” medicine to allow NP’s to try to practice medicine
  • Allopathic medicine doesn’t have a monopoly on knowledge
  • EBM is code for rationing and cookbook medicine such that experience doesn’t matter.
  • it drives physicians to treat the diagnosis instead of the patient (the underlying flawed premise of EBM)
  • In this system, how are we any different than a really smart computer
  • It’s become the new religion.
  • EBMs come from Committees.
  • EBM is a “cookbook”. One size fits all. Experience and years of training does not matter
  • With regard to meningitis vacccine, now that we are vaccinating everyone, it is less likely that people will look for a better way to prevent full blown meningococcemia
  • EBM is statistics. Nothing else
  • I use EBM as a guideline, but sometimes they fly in the face of my own experiences. As the punchline in an old joke, “Are you going to believe me or your lying eyes?” There is validity in experience.
  •  THE INSURANCE COMPANIES PUSHED “EVIDENCE BASED MEDICINE” ON US!
  • there is actually NO EVIDENCE that a practice based on so called EMB is effective or even relevant to real patients.
  • I just love it when I get a note from an RN questioning my prescription and citing “Evidence Based Medicine”
  • Intuition based medicine, like art, is always about a century ahead of science.
  • “Evidence based medicine” is a form of doublespeak. It restricts recommendations to those proven in clinical trials, which don’t exist for most clinical situations
  • EVERY ten years the Literature changes!!! EBM today is the quackery of tomorrow
  • As long as drug companies own the experts and fund the vast majority of studies AND have the right to publish the findings or not as they see fit we will NEVER have fully reliable evidence… therefore evidence based medicine might well be WRONG medicine.

Keep in mind that these are not clueless celebrities, anti-vax mommies, or crank CAMsters. These are mainstream medical doctors. I weep!

Some of their less irrational objections (with my comments)

  • Evidence from a trial with selected patients doesn’t necessarily apply to individuals in your office. (True. That’s where judgment and common sense come in. The trick is to not blindly follow anything labeled “evidence” but to consider all applicable factors and apply available evidence rationally.)
  • Early adoption of EB treatments superseded by new data. (New data… isn’t that a good thing? Meanwhile, we can’t just do nothing while we are waiting for better data.)
  • EBM gives us a cookbook. No progress was ever made by following a cookbook. (EBM was never intended to be a cookbook. We don’t follow a recipe, but apply available knowledge to individual patients.)
  • I am not enthralled with EBM used by committee/hospital edict. (Me either. Who loves committees? That’s not the way it’s supposed to work).
  • population studies with correlations are helpful but can never substitute for scientific, logical reasoning applied to the individual patient in question. (And they were never intended to substitute; they are intended to inform.)
  • we’re supposed to turn in our fellow physicians who over-test or over-prescribe based upon “Evidence based Guidelines” (EBM is not Big Brother. Is this just paranoia, or have physicians actually been “turned in” or disciplined for not following these guidelines?)
  • what I fear is rigidly taking the conclusions of a research study, … and FORCING physicians to do things that way. (I fear that too. It shouldn’t happen.)

There were a few more rational comments:

  • I cannot comprehend what the alternative to EBM would be.
  • It’s really just a label for “good medicine”
  • there can be no doubt that our individual clinical experience is biased.
  • we are all practicing EMB all the time but are being fed a brand of EMB with which we are not always comfortable (SBM is not comfortable with the current brand of EBM either).
  • Are we reverting to the Dark Ages?
  • EBM should be used as a tool in the decision making process, but should not be deferred to in place of independent thinking nor considered a gold standard not to be deviated from.
  • there is NO better alternative.
  • EBM is a well-meaning and useful concept if tempered with judgment and judicious application. My challenge with it is that it has been turned into a religion, and we physicians are supposed to worship at its alter without the addition of any critical thinking.
  • Evidence medicine implies that the studies have been peer reviewed and meet rigid criteria regarding study design and conclusions. It should not mean that small, poorly designed studies with unsupported conclusions are cited as part of EBM recommendation.
  • simply saying “I know what I know” and practicing anecdotal medicine by it self is simply dangerous

One commenter provided a link to SBM and called us “the militant wing of the EBM movement.” Kind of ironic when we established the blog as a protest against the flaws in EBM.

Conclusion

Most people would agree that medicine should be based on evidence and on science whenever possible. Good medical practice is not an “art” but an applied science that takes all available knowledge into consideration and applies it to the individual patient using careful judgment, compassion, and common sense. EBM is a worthy goal; but as currently implemented, it has fallen short of its potential. We on SBM see it as neglecting basic science, prior plausibility, and common sense. Others apparently see it as a dogmatic institutionalized draconian interference with practitioner autonomy. Incredibly, some scientifically educated medical school graduates still see evidence as inferior to intuition and experience: those are the ones who give me chills.

Posted in: Science and Medicine

Leave a Comment (131) ↓

131 thoughts on ““How do you feel about Evidence-Based Medicine?”

  1. kathy says:

    “EBM gives us a cookbook. No progress was ever made by following a cookbook.” Cookbook = straw man.

    Physics is a cookbook. No progress was ever made by following physics.

    Errrr ….

    What I mean is, physics won’t tell you how to build an actual real-life bridge … every bridge is unique, individual, adapted to the local river, weather, traffic volumes, etc. So because this bridge we are about to build is unique, we can throw out that cookbook of physics-based formulae and calculations of stresses, etc.

    Errr …. I think I’ll drive a different route back home.

  2. tgobbi says:

    “Keep in mind that these are not clueless celebrities, anti-vax mommies, or crank CAMsters. These are mainstream medical doctors. I weep!”

    That’s what surprised me. By about the third or fourth “typical” comment I thought that you had tricked us by sneaking in excerpts from some chiropractic or naturopathic discussion group! You weep? I despair!

  3. Jan Willem Nienhuys says:

    Extremely interesting! Part of the ideas behind the scientific method is the realisation that the human mind is very fallible. That’s why the scientific method abounds in methods to check correctness.
    - writing down precisely what you have seen, so your memory can’t play tricks on you
    in a way is this what (for example) mathematicians do. When they have some insight, they write down the proof so as to be sure that they didn’t overlook a nasty little detail that could invalidate everything.
    - look for all kinds of contradictions: internal contradictions and contradictions with facts (experiments, results of others)
    - publish what you have found, so others can check it
    - try to quantify, so as to switch off fallible intuition
    - try to simplify. An important simplification effort is that you ask: ‘could it be merely chance?’ If someone would say ‘only chance and not your treatment determines whether people are cured’, what experiment should you do to refute that?
    - use if possible a machine to do observations (and check that machine for correct functioning)

    Many physicians enjoy a very high status especially with their patients. This high status makes them resist any suggestion that their judgment might be fallible. They are not the only ones. This overrating of intuition happens in courts of law, in the police force, among psychologists and probably in politics.

    Active scientific researchers (and science students) are almost on a daily basis confronted with the fact that these checks I mentioned are necessary. Mathematicians are often not considered scientists (because they don’t do ‘experiments’ to refute theories), but the daily practice of the research mathematician is a rapid cycle of getting ideas, trying to work them out, failing and then trying another approach, finding counterexamples and detecting errors; what apears in print is of course only “theorem – proof” and not the contents of their wastepaper baskets with all the dead ends and refutations. I think that medicine students are much less exposed to experiencing how wrong they can be. Not long ago (I have heard) there was in the medical world only one kind of person higher than a physician, namely the professor in medicine. The professor is always right. The only way to perfection is to imitate the professor in all repects. The professor shows how to do things and any deviation is unthinkable. This attitude was not conducive to the attitude that experiment must decide things.

    When Louis started in 1835 with the méthode numérique many physicians made the same objections as we now hear against EBM. A famous pupil of Louis was Semmelweis. But Semmelweis made another mistake. He didn’t bother to present his findings in an orderly fashion to the public or his colleagues and all they knew vaguely was that he recommended chlorine bleach. Duh. (Really: washing hands as recommended by Semmelweis meant washing in concentrated bleach until the hands were slippery; this prior to seeing each new patient.) The second mistake Semmelweis made was that he called all obsgyns murderers if they didn’t agree with him, i.e. accept his words without proof. Even when he came around to writing down his findings in a book, he kept using words like murderer.

    One can easily understand the feelings of all those doctors who thought themselves highly respected benefactors of mankind – without knowing a lot about bacteria or atoms or effective medicine – when they were called murderers by a colleague.

    Stressing why EBM is good, may in the short and the long run be more effective than making negative remarks about the opponents.

  4. BillyJoe says:

    “Experience trumps EBM”

    So what does a new medical graduate rely on when he has, as yet, no experience to guide him?
    Why would the faulty recall of 20 personal cases trump a systematic review involving 2000 carefully recorded cases?

  5. BillyJoe says:

    “Experience trumps EBM”

    I have been interested in the evidence base for PSA testing for prostate cancer ever since my father was diagnosed with that disease. I often come across comments made by medical practitioners which make it clear that they base their decision to promote the test amongst their patients as a result of a single experience they have had – for example a patient who asked for the test to be done, was diagnosed with prostate cancer, had a prostatectomy…and is still alive! This apparently trumps the systematic reviews based on tens of thousands of patients that conclude that PSA testing cannot be recommended.

  6. rork says:

    “physicians are screwed from all directions even from statisticians who sit on rear end all day”
    If I wanted to consider the evidence, I would have studied math harder. But I hated math. So I became a doctor.

  7. nybgrus says:

    First off, thank you JWN for being one of the few people to know the real story of Semmelweis. Even in medical school we are taught the romanticized version for some reason. It really limits the merits and teaching points of the story to “washing hands good. not washing hands bad.”

    In any event, I digress.

    As I have said before, one can make it through medical school without ever actually learning medicine. You can memorize a whole bunch of seemingly unrelated facts and algorithms and regurgitate them on exams. Then, when you get out to practice just keep doing the same. The incredible ability for humans to compartmentalize means that the same person disparaging EBM will read an article that doesn’t offer immediate affront to their biased sensibilities and adopt that new guideline, but rail against the rest of of it. And even worse, they can have their patients die time after time and rationalize it based on their “experience” by blaming the patient (he was too sick), the nurse (she screwed up my good work), the system (the hospital didn’t let me do exactly everything I wanted), and then a little bit of luck (this isn’t a science and sometimes things just go wrong). It is amazing how well this works – I actually know a neurosurgeon at my old hospital with this exact attitude. He typically takes twice as long to complete a surgery and claims he is just more “careful” and “meticulous” and consistently has a huge number of complications after the fact… which he then blames on everyone and everything but himself. No one is willing to challenge him and, since it is a private hospital not a teaching institution, there is nobody to push a review of his patients to demonstrate his incompetence. Of course, the methods to do so would be the exact same used by the EBM he so despises in the first place.

    The true problem with medical education as I see it – one among many, of course, but IMHO a crux of many others – is that the system is set up to train highly educated mechanics. The Jiffy Lube worker who can adjust any system in your car effortlessly, but fancies himself the Ford engineer who built it in the first place. You truly have to take an extra step – both mentally and in actual effort – to genuinely become a medical scientist instead of a medical monkey.

    So when you fancy yourself the engineer, but are really the mechanic, is it any surprise that there is backlash when EBM comes knocking at your door and tells you quite clearly you are wrong?

  8. nybgrus says:

    oh yes, and another though that just occurred to me – you can easily eschew the parts of medical education that would train you to actually be a scientist and critical thinker – at least in many if not most cases.

    For example, at my school, people tend to “hate” EBM, stats, ethics, and population health. It is the minor road bumb they throw at us that gets in the way of learning “real” medicine. And it was really (IMO appalingly) easy to literally not know a single stitch of it and be just fine. The reason being is that your overall grade is what determined whether you passed or not, and the EBM and ethics portions of the exams was limited to 20ish%. Which means you can learn enough random bits to get just some of that right, and with the rest of the test be just fine to pass – even with honors.

    Now, the curriculum has been re-vamped and you must pass each section in order to pass on to the subsequent year. And immediately cries of “Oh noes! We must pass teh stupidz EBM and ethicz??” began. And I was, sadly, not surprised. But I did say to those who would make such complaints to me or on a public forum “Why yes. You must actually know the basis of medicine to be a doctor. Fancy that!”

  9. Mojo says:

    “Experience trumps EBM”

    Good heavens! So all those doctors who knew from experience that bloodletting worked were right after all!

  10. ConspicuousCarl says:

    Early adoption of EB treatments superseded by new data.

    Where do they think that “new data” comes from? Some lone crank’s intuition? This is the same logical disconnect as is used by creationists who point to examples of “science” being corrected, as if those corrections came from prayer.

  11. windriven says:

    “Experience trumps EBM.” Name redacted to protect the clueless.

    “‘In my experience’ are the three most dangerous words in medicine.” Mark Crislip

    The horror is how difficult it is for a patient to determine where on the SBM curve a particular physician lies. How, as a consumer, is one to know whether a physician is a scientist or an anecdotalist or something in between?

  12. @Nybgrus: Re. the incompetent neurosurgeon, terrifying from a patient perspective. How would a patient possibly know to avoid this guy, and how would a referring doc know not to send patients there? Are there underground methods to get the word out to referring docs, or do patients have to just roll the dice and hope not to get a surgeon like this?

  13. Sadly, many of these objections have been answered before. Either we are not doinga great job of communicating with doctors, or those who make such comments have an intransigent resistance to the idea that goes beyond reason and logic and is rooted, like so much of alternative medicine, in fear and faith.

    Strauss SE, McAlister FA. Evidence-based medicine: A commentary on common criticisms. Canadian Med Assoc J. 2000;163(7):837-41.

    Sackett DL, Rosenberg WMC, Muir Gray JA, Haynes BR, Richardson WS. Evidence based medicine: what it is and what it isn’t. British Med J 1996;312:71.

  14. PJLandis says:

    “This overrating of intuition happens in courts of law, in the police force, among psychologists and probably in politics.”

    Probably in politics? That’s funny.

  15. Scott says:

    I wonder how the ages of those commenters varied. My HOPE would be that the older, more conservative doctors are the ones insisting on doing things the way they always have, i.e. by the seat of their pants. It would then be a matter of waiting for them to die/retire.

    It always seems to work that way; real acceptance of a new way of doing things often requires the prior generation of old fogies to go away. (And be replaced by the NEW generation of old fogies who will insist that the way things were done when THEY started is the One Right Way.)

  16. Mark Crislip says:

    Wow.

    When I rant on the “in my experience” as the three most dangerous words in medicine to residents, I am careful to distinguish the difference between diagnosis (I am infinitely better at diagnosing and following the course of disease as a result of almost 30 years of ID) and therapy (If I want to know the best therapy for an infection, I read the literature).

    Most docs, in my experience, do not separate the two.

    One of the many things that makes medicine hard is applying the population data/EBM to the individual, and experience is what we rely on. The two are not mutually exclusive and is certainly not cookbook.

  17. mousethatroared says:

    Speaking as a patient and the parent of a patient I will say, my first encounter with “EBM” was not a good one. In late 2008 our audiologist informed us that our son had profound hearing loss in one ear. We found that it was due to a congenital malformation, possibly related to his cleft lip and palate. At that time our son also had a severe expressive speech delay. At age 3 he was about 90% unintelligible and very seldom strung more than two words together. He received speech therapy three times a weeks, but was losing ground in comparison to his peers.

    Our audiologist at the time, who was our source for finding information about support for his hearing loss made very few recommendation on supports. She told us that there was no evidence that unilateral hearing lose increases the risk for speech problems. That his speech issues were cleft related not hearing related. When we became interested in a newer bone conduction hearing device used with adults with unilateral hearing and children with bilateral hearing loss, she recommended against it. One of her reasons was that at her latest conference, a children’s hospital who specializes in Evidence Based Medicine had recommended against their use due to the lack of evidence showing their benefit. When I asked about the risks of trying the device out, I was told that aside from the financial risk, there was none. Yet she still seemed very reluctant to pursue this support or other special accommodations (such as preschool supports or preschool FM systems). I was lead to believe that this was all based on the evidence.

    Looking online, I found that there had been very little research done with children with unilateral hearing lose. As far as I could decerns from a laymen’s view, one reason there was no evidence that UHL increased the risk of speech deficients could have been due to the lack of research looking at the issue.

    My attitude at the time was (and still is) that if you see a child struggling you look for evidence on what will help. If you try everything that’s recommended by the evidence (in our case the speech therapy based on his cleft lip and palate) and it’s not enough, you then look for plausible solutions with no or very low risks to try. You don’t just say “It’s not recommended by EBM, so don’t do it.”

    Ultimately, by asking around we found that our school system had access to a preschool, grade school program for children with hearing loss and speech deficients, they offered full day preschool that integrated speech, language and listening skills throughout the day. (Although our audiologist coresponded with the audiologist in this program, she had not told us of the program) They have been supportive of our son’s trial with the bone anchored device. In this program my son showed great progress in his speech, language and listening. The surgeries that he had for his cleft made a huge difference in his intelligibility, but I am pretty certain that he would not be speaking and reading at grade level today if it weren’t for me looking further than our original “EBM” audiologist’s recommendations. (n=1, I know)

    Now, maybe people will say, that’s not how EBM is supposed to work. I guess my response would be, if EBM is decidedly different in concept than it is in practice, then that is still a flaw with EBM.

  18. mousethatroared says:

    @ HH “Good medical practice is not an “art” but an applied science that takes all available knowledge into consideration and applies it to the individual patient using careful judgment, compassion, and common sense.”

    I’m afraid the minute you try to integrate subjective (varible) attributes like common sense and compassion the thing become an “art”. Meaning the character of the final product is at least partially determined by the personality and perspective of the individual practitioner. (At least that’s what this artist says.) :)

  19. mousethatroared says:

    Ackk, I’m a bad mom. My son’s hearing diagnoses was made at age 4 not age 3. Don’t know if that makes much difference in the point of my comment, but wanted to correct it regardless.

  20. mdstudent says:

    “For example, at my school, people tend to “hate” EBM, stats, ethics, and population health. It is the minor road bumb they throw at us that gets in the way of learning “real” medicine.”

    I come across the same attitude over and over again even towards organic chem among premeds. It disgusts me every time.

  21. Alia says:

    Over here we’ve been having a huge discussion about EBM. The thing is, our public health system is chronically short of money. So now patients need to pay full price for all medical products that have no EBM recommendation for their disease. And the problem is – some drugs were not tested on children so there are no EBM recommendation to use them for children – but some physicians prescribe them is some cases and see positive results. And now the chronically ill child that has used this particular drug for let’s say 3 years with great success will be left without it – or the parents will need to spend hundreds of $ to buy it at full price. In this way EBM is used in our country to lower costs and put some burden on the patients, even though medical care is theoretically free to all insured (which is the majority of population, including all people employed full time, all OAPs, all children and all students).

  22. Jan Willem Nienhuys says:

    BillyJoe writes:

    for example a patient who asked for the test to be done, was diagnosed with prostate cancer, had a prostatectomy…and is still alive! This apparently trumps the systematic reviews

    I was under the impression that there is a huge difference between testing a whole bunch of men irrespective of their personal situation and testing when the patient has symptoms. In the above story is lacking why the patient asked for the test. Did he think it was the right thing to do ‘When I’m sixty four’ or was he suspicious because he was using more and more time taking a leak? That makes a big difference.

  23. Jan Willem Nienhuys says:

    PJ Landis comments:

    Probably in politics? That’s funny.

    In medicine and several other branches of human activity one can use science methods to compare ‘intuition’ and ‘experience’ and so on with more rational ways of making decisions. But in politics? Is there any yardstick to compare? I couldn’t think of any.

  24. @BillyJoe, @JWN,

    It’s worth a read of the USPSTF recommendation statements and the evidence behind the recommendation:

    http://www.uspreventiveservicestaskforce.org/uspstf/uspsprca.htm

    The guidelines (for men under 75) absolutely do not say to not perform PSA screening. They state that the evidence is insufficient, it does not say to not recommend the PSA test. In fact, the ethical thing to do is have a conversation with the patient and discuss the pros and cons of pursuing PSA screening. The USPSTF even has a poorly produced video on the above linked website that states exactly this.

    Keep in mind that the positive predictive value of PSA increases as the PSA increases, meaning a value of 7 is more significant than a value of 4.1. 4.1 could be a transient elevation, but it could also mean a developing cancer, and this is where a trend of PSA becomes helpful. There is no doubt that PSA identifies cancer, the debate is whether or not detecting the cancer earlier saves lives in the long run.

    People who wish to proceed with PSA screening need to understand that if the PSA is elevated, it would mean that they are recommended to have a biopsy, and that the biopsy may have significant side effects, and that if positive they would possibly need a prostate resection, which has even greater side effects, all the while it may or may not diagnose a cancer that may or may not kill them in their life time. Beyond that, it is up to the patient if they wish to receive the screening. A doctor should not deter them from screening (are you really going to deny your patient the opportunity to detect and remove a potentially deadly cancer from the body?) or push them towards it if the patient is unconcerned about having this cancer.

    Some people are comfortable with the prognosis of cancer and are happy proceeding to live their life normally and hope that they die of some other cause first. Other’s aren’t. And we also need to keep in mind the general health of the person, or more accurately if the person is estimated to live at least another 10 years. A 65 year old male with no co-morbidities should and would probably prefer have a PSA test and eliminate any cancer early so they can live well into their 90s. Whereas a 65 year old male with COPD, CHF, HLD, HTN, and DM2 probably will die of his poor general health before prostate cancer kills him.

    … For me? I want to know, and I’d want it out. If my PSA is elevated, I’ll be first in line to get a biopsy, and if the biopsy has significant Gleason score, then I’ll be at MD Anderson having my prostate cut out with a robot. I couldn’t live knowing that I have prostate cancer waiting to kill me slowly and painfully. Other individuals are perfectly happy living this way.

    If anything, what we need is a better way to do the biopsy. We need something that can somehow give us a tissue diagnosis without being so invasive. Leave that one up to the medical engineers.

    Furthermore, it’s good to keep in mind that the USPSTF is just a recommendation, it’s not necessarily how things are done. The American Cancer Society often has different recommendations than the USPSTF (I believe, for example, they still differ in mammogram recommendations.) Google “american cancer society PSA” for their take on the test.

    Bottom line: it’s absolutely wrong to say that PSA screening is “not recommended” or is otherwise useless.

  25. PJLandis says:

    Hey Jan…about the Politics comment

    I see your point, in public and governmental policy you can’t get the same amount of certainty as you would in a scientific field, but I think there is ample amounts of evidence for politicians ignoring evidence, either evidence that justifies a certain position/policy or evidence that the outcomes of a certain position or policy aren’t as intuited.

    I think the “war on drugs” is a good example. The stated goal is to end, diminish, or decrease harm from drug use yet their is a staggering amount of evidence to show that it is having the opposite effect and creating a different type of harm by swelling our prison populations. You could argue drug users deserve prison because drug use is morally wrong, regardless of the ancillary costs, but that isn’t the most common explanation used for our current drug policies. Ditto for more and longer jail sentences.

    Same-sex marriage is another case where a large majority seem to be arguing that it harms heterosexual marriage, or it’s harmful to children in such families, but the evidence is pretty overwhelmingly in favor of homosexuals being just like you and me, having happy/healthy relationships and raising happy/healthy kids bout as often as heterosexual couples. Some people may argue that homosexuality is immoral regardless of the consequences, but again that is not a common position being taken, publicly at least, by most politicians.

    Or, to use a less controversial topic, how about Civil Rights for African-American’s? Has it wrought any of the terrible consequences people were once so sure would befall the country? That’s a pretty clear example of intuition being incorrect.

    To make it more on topic for this blog, how about the political decision to funnel so much money into the NCCAM? I’ve seen some pretty compelling arguments that decision was fueled by intuition more than evidence.

  26. PJLandis says:

    @Alia “patients need to pay full price for all medical products that have no EBM recommendation for their disease. And the problem is – some drugs were not tested on children so there are no EBM recommendation to use them for children – but some physicians prescribe them is some cases and see positive results.”

    I think Science-Based Medicine, so ably promoted by the humble blogging physician’s here at SBM, addresses some of that concern.

    There is good reason to expect that a drug used to treat an adult’s condition will also, in many cases, be effective at treating a child. Obviously kids are different, and my understanding is that pharmacokinetics is often the biggest difference but not the only one, which makes trials in children ideal and necessary at some point, but in the meantime I would assume you have to act on what you do know if the alternative is dire enough. I’m sure one of the medicos can confirm/deny that, but I think you could get some good ideas and arguments from this site to champion a more realistic, and scientific approach, to funding treatment within your countries healthcare system.

  27. lizditz says:

    #mdstudent

    minor road bumb

    I am imagining a new thing, a cross between a bump and a bomb.

  28. * Just to clarify, by “not recommended” I mean PSA screening is definitely not recommended against.

  29. BillyJoe says:

    Jan,

    I was talking about PSA as a screening test (ie patients are asymptomatic).

    Skeptical Health:

    I chose my words carefully: On the available evidence, PSA screening cannot be recommended.

    But you are correct about how a patient who comes in for a screening PSA test should be handled. But the doctor should not be ordering PSA routinely nor encouraging patients to have it done.

    Also I am surprised you would have PSA screening or a robotic prostatectomy because:
    - the evidence is that PSA screening is not useful in decreasing mortality sufficiently to justify the morbidity of investigations and treatments.
    - there is evidence that robotic surgery is no more effective than a standard prostatectomy. It jsut takes longer and is more expensive.

    It seems that, for some, SBM is fine untill you have to make a personal decision about your own health, then SBM goes out the window.

    The guidelines do not recommend PSA screening.

  30. BillyJoe says:

    ..oops, ignore the last sentence, it was supposed to be deleted.

  31. Chris says:

    Dr. Hall:

    EVERY ten years the Literature changes!!! EBM today is the quackery of tomorrow

    And that is bad?

    I spent last week at the medical mecca known as the Mayo Clinic. More than once the conversation between us and the specialists were that new research has changed certain things.

    One major case in point was that their literature said there were fourteen known genetic sequences that could cause our son’s problem, yet our local university medical system tested him for eighteen! But he had none of them, it is still an area where data is being collected (and if they can get curious parents to pay for a full genetic panel that would be golden!).

    Another thing that was changed was that this genetic anomaly is now known to manifest itself in persons up to at least age sixty. So instead of testing until mid-thirties, we the parents now need to be tested! (The patient next door to our son in the surgical recovery ward had a different genetic physical anomaly that required even more surgical remodeling, and it was only discovered when he was twice our son’s age! Human physiology is complicated, and it seems to work with many defects, for a while.)

    Now for the humorous bit: the specialist’s physician assistant made it a point to warn us against information on the Internet. Oh, I so much wanted to tell her I have met at least four of SBM’s editors in person!

  32. JMB says:

    EBM, or SBM is a method for analyzing scientific data and utilizing it in medical practice. USPSTF guidelines do not constitute the sole authority of EBM. The USPSTF does adhere to EBM methods. The scientific evidence on which we are supposed to base our practice of medicine are the scientific reports in the literature, not the dictates of a scientific authority (there really is no authority in science, just reproducibility of experimental data and accuracy of predictions). If a medical student thinks that quoting the USPSTF guidelines (or Cochrane reviews) are adequate to qualify as EBM, then the medical education system has failed. A properly trained medical student would read the evidence supporting articles published by the USPSTF, and a majority of the articles quoted as best evidence used in the meta analysis. Only after understanding the selection criteria used in the original randomized clinical trials, and reviewing both the discussion by the original authors and the meta analysis authors, and the statistics of relative risk (or risk ratio, or odds ratio used in case control studies), would the medical student have the scientific knowledge that would be used in a healthcare decision.

    The USPSTF does an excellent job of literature review and meta analysis. The published guidelines and supporting evidence are a good start for gathering the scientific information used in the decision process. However, the decision framework for population guidelines is different than the decision framework for an individual patient. Disagreeing with the USPSTF does not mean you are disagreeing with EBM or science, provided you have done your homework.

  33. Chris says:

    EBM does not equal SBM. From this blog’s first article:

    Within the practice of medicine there is already a recognition of the need to raise the standards of evidence and the availability of the best evidence to the practitioner and the consumer – formalized in the movement known as evidence-based medicine (EBM). EBM is a vital and positive influence on the practice of medicine, but it has its limitations. Most relevant to this blog is the focus on clinical trial results to the exclusion of scientific plausibility. The focus on trial results (which, in the EBM lexicon, is what is meant by “evidence”) has its utility, but fails to properly deal with medical modalities that lie outside the scientific paradigm, or for which the scientific plausibility ranges from very little to nonexistent.

  34. kathy says:

    Lizditz: “minor road bumb. I am imagining a new thing, a cross between a bump and a bomb.”

    I am imagining a road bump that gives one a boot on the bumb.

    Btw I love the English expression for this article of road furniture: a “sleeping policeman”. As in, “I didn’t see a sleeping policeman in the road, bumbed right over it at full speed.”

  35. Jan Willem Nienhuys says:

    JPLandis wrote:

    I think the “war on drugs” is a good example. [snip]

    Thank you for the examples. Actually this quoted example falls broadly under what police and judges do. The other examples are better in this respect. What politicians typically also do is compromising: ‘if you help me with A – which you don’t like too much, I’ll help you with B – even though I don’t care about B’. How to judge the rationality of that? Should one check whether this is done according to the rules of game theory?

    The politicians do many things that have serious economic consequences, but economy itself is riddled with scientific problems. What about fighting extremely costly and disruptive wars for ideological reasons? More related to medicine: why is it so difficult (in some countries) to organize an efficient public health system with universal health insurance? That’s politics isn’t it? The healthy and wealthy using politics to prevent that they should pay for the sick and poor. How could any scientific or rational consideration be applied to problems of allocation of funds when the parties involved are not considering the benefit of all but merely try to maximize the benefits of themselves and their political friends?

  36. mousethatroared says:

    MC “One of the many things that makes medicine hard is applying the population data/EBM to the individual, and experience is what we rely on. The two are not mutually exclusive and is certainly not cookbook.”

    That’s kinda how I use a cookbook. I like the basics of a recipe, but alter it based on individual needs, preferences, what I have available and past experience.

    if you check out the reviews at any recipe site you’ll find many people do the same. In fact it seems only people who claimed to have “followed the recipe exactly” have any complaints. :) The key is having the knowledge of what should be changed, (temperature for your altitude) what changes will cause a fail (substitute baking powder for baking soda) and what elements can be changed to suit (less sugar, cause I don’t like things too sweet).

    Maybe you should call EBM cookbook medicine instead. I don’t see alot of complaints on cooking sites that ‘Allopathic cookbooks don’t have a monolopy on knowledge.”

  37. nybgrus says:

    Now, maybe people will say, that’s not how EBM is supposed to work. I guess my response would be, if EBM is decidedly different in concept than it is in practice, then that is still a flaw with EBM.

    I think we here would agree. Evidence based doesn’t necessarily mean evidence restricted. Though it is SBM that specifically adds in the prior plausibility aspect and thus would allow someone the ability to translate care such as you described that lacks specific types of evidence but has others.

    The complaint made here at SBM is that EBM overvalues the top the evidence hierarchy and undervalues the bottom, especially in certain cases. It sounds like there was a distinct lack of evidence at the top for your son, but from an SBM perspective there still exists evidence at the bottom of the hierarchy in favor of the intervention for your son. To me, that constitutes evidence that I can use to treat him… though since it is indeed weak, it would be reserved for after other therapies had failed and be instituted only if there wasn’t significant risk of adverse event or outcome (which it seems would be the case).

  38. nybgrus says:

    @Mdstudent:

    I come across the same attitude over and over again even towards organic chem among premeds. It disgusts me every time.

    Couldn’t agree more. I find it quite dismaying when my colleagues brush over molecular interactions as irrelevant to them learning medicine.

  39. nybgrus says:

    @alia:

    The thing is, our public health system is chronically short of money. So now patients need to pay full price for all medical products that have no EBM recommendation for their disease. And the problem is – some drugs were not tested on children so there are no EBM recommendation to use them for children – but some physicians prescribe them is some cases and see positive results.

    I would agree that is indeed a serious issue. As I said to mousethatroared above, from an SBM framework there would still be some evidence to support those treatments in children, even though there are no specific controlled studies for children. It should be very much taken into consideration that trials are lacking in children since in most cases, but not all, it is just the pharmacokinetic profiles that need to be taken into account when administering an adult drug to children. It can get more complicated for some interventions, and we don’t really have an a priori way of predicting which ones it would be – so caution is advised. But to completely remove the ability to have it covered, especially after years of safe use (or seemingly safe, to be more accurate), is indeed terrible.

  40. nybgrus says:

    @lisditz:

    I am imagining a new thing, a cross between a bump and a bomb.

    LOL. That was actually my typo that mdstudent copypasta’d. But I like the visual

  41. nybgrus says:

    @pjlandis:

    I’m sure one of the medicos can confirm/deny that, but I think you could get some good ideas and arguments from this site to champion a more realistic, and scientific approach, to funding treatment within your countries healthcare system.

    Indeed I agree with you (as I commented above before reading yours). It does get more complicated in certain cases – the pharmacokinetic profile can change in a priori unpredictable ways for children, though that is generally uncommon. You can also have paradoxical reactions (i.e. a sedating drug amping you up) more frequently in kids. There can also be wholly unpredictable and unseen-in-adults reactions (like aspirin and Reyes syndrome). The therapeutic window could be different for children and we wouldn’t necessarily know. And lastly, you simply cannot predict what sort of long term effects will be experienced by children taking medications for extended periods of time. *

    But outside of those, especially with drugs that have a long established profile of being reasonable safe and a large therapeutic window in adults, it can be reasonably inferred from an SBM perspective that there is some (weaker) evidence to support the use in children. For certain indications the outcome of not using the drug is bad enough that the risk:benefit analysis certainly favors taking the risk with the drugs.

    *I’m sure there is probably something that I am missing. It is 5:45am and I am only halfway through my coffee.

  42. fledarmus1 says:

    The elephant in the room that everybody is ignoring is the effect of health insurance on the discussion. The even larger elephant in the room that seems to be completely invisible is what health insurance actually is. In most discussions involving EBM, SBM, or CAM, the entire discussion boils down to what insurance should be required to pay for. Health insurance itself is treated as some magic property that makes healthcare free, and all you have to do is convince enough people that the test, treatment, or drug is important for it to suddenly become free.

    I see nothing whatsoever wrong with a patient of any age worried about the possibility of prostate cancer requesting a PSA test. I see nothing wrong with a patient of any level of risk getting a mammogram once a year. I see nothing wrong with a person making an informed decision to visit a chiropracter once a month, a massage therapist once a week, and an acupuncturist any time he or she has a headache. The problem is the concept that “somebody else” should pay for it.

    This is part of what fuels the high-drug-prices argument as well. Ask yourself this question – if you were terminally ill and there was a drug available to save your life, how much would you pay for it? The answer seems to be, “Nothing at all. If the drug will save my life, it should be free, because I can’t live without it. Somebody else should pay.” And even saving a life isn’t the critical point anymore. Relieving suffering, making movement easier, working around a handicap, since these are all important, and more important to the sufferer than to a non-sufferer, then somebody else should pay. Now we have bottled water that might once have been within six degrees of separation to a molecule that once touched a flower being prescribed by people with less training and scientific understanding than a palm-reader, but if it might make somebody feel better, then somebody else should pay.

    There was a rallying cry around for a while that everybody is entitled to the same health care that a millionaire would get. Instead, we are achieving a point where nobody is entitled to better health care than the poorest of us could once afford.

  43. mousethatroared says:

    fledarmus1 “This is part of what fuels the high-drug-prices argument as well. Ask yourself this question – if you were terminally ill and there was a drug available to save your life, how much would you pay for it? The answer seems to be, “Nothing at all. If the drug will save my life, it should be free, because I can’t live without it. Somebody else should pay.”

    Free? Were do you live that you are getting health insurance for free? I’m moving there. Usually health insurance is paid for either through taxes, by individuals or by employers as part of the payment for labor received.

    Ask yourself this. If you pay for home insurance and your house burns down, how much should you pay to have it rebuilt? The answer usually is, not a divesting amount because you paid for the bloody insurance.

    Of course the whole concept of insurance is that you are paying a little at a time into a pool, to insure against risk of a large expense. When you buy insurance, you are accepting the fact that, if you are lucky, you will mostly be paying for other people’s misfortune, but you are receiving a certain amount of finically stability in return.

    Items that are paid for by the insurance are not free, they are part of the negotiated contracts between the sellers and buyers of insurance.

  44. mousethatroared says:

    That should have been financial stability, I don’t know what finically stability is.

  45. nybgrus says:

    There is nothing wrong with anyone requesting a PSA or mammogram at any age or at any risk profile. There Is something wrong with a physician ordering those tests when not indicated by the guidelines. The same way it its wrong to prescribe antibiotics for viral illness or even a stronger/broader spectrum antibiotic than indicated for a bacterial illness… Regardless of who is paying.

  46. nybgrus says:

    Sorry. Screwed up the html tags. Should be obvious though

  47. mousethatroared says:

    nybrgus-”Evidence based doesn’t necessarily mean evidence restricted”

    That is a good succinct way to put it. But a question arises when we consider the reality of scarcity of resources. EBM may be a good tool in diagnosing and treating patients, but it seems that some insurance companies, governments or hospital organizations are using ERM to direct or ration resources.

    The reality is that medical resources will be rationed, but is EBM (or ERM) the best tool for the job. What are the benefits and pitfalls of using that tool for that job?

  48. mousethatroared says:

    Also nybgrus – thanks for confirmation of my basic approach. One reason I became facinated with this blog was that I could see that the writers were trying to develop a methodology that would overcome the shortcomings that I experienced with my initial impression of EBM (my anecdote above).

    While I may be vocal in my disagreements with some the arguments on this blog, I genuinely appreciate and find comfort in the fact that there are medical folks who are taking time to consider these issues and disseminate information on the topic.

  49. Alia says:

    @PJLandis, @nybgrus – What you are saying is exactly the point that our physicians and parents of ill children were making to the government. In some cases with success. Sadly enough, first they had to go to the media, which naturally started crying out “The government is killing our children!” (or compatible) and then the financing guidelines were changed.

    BTW – my insurance covers one pap test every three years. However, I’m in a risk group, my gyn/ob tells me I should have it every year, so I pay for it.

  50. fledarmus1 says:

    @mousethatroared – that is exactly my point. Health insurance is not free, but the argument for any health care cost at all is, “health insurance should be required to cover that!” Without, of course, affecting the rates for health insurance. Just pass a law, and suddenly it’s automatically covered. And isn’t it terrible how much our health insurance rates keep going up – that must be because not enough people who don’t use it are required to purchase it. There seems to be a huge disconnect between what consumers think health insurance should pay for and what they think it should cost. Health insurance doesn’t resonate as spreading out your health care costs over times that you are healthy so you don’t get hit so bad when you are ill, but as membership in a discount market – just pay a small entry fee and all your health care is free!

    If there really was some negotiation between buyer and seller, that would be a different story. That would mean people with chronic illnesses and genetic predispositions to disease would be paying more for insurance at a rate equal to their risk, unmarried males would not be paying for insurance which covered pregnancy and birth control pills, females would not be paying for insurance which covered prostate cancer, childless people would not be paying for pediatric care, and only people who believed in chiropracty or homeopathy would be paying for insurance which covers it. Instead, we have a four-way fight between customers, regulatory agencies, health care providers, and insurance companies over what “health insurance” should mean, and a one-size-fits-all insurance plan with the only differences being the cost of coverage (which depends not on your risk but on the size and power of your demographic) and how much of the cost comes out of your pocket.

  51. evilrobotxoxo says:

    I think EBM is the only game in town because evidence really has to be the primary thing that a physician should bases their decision-making on. However, in practice, there are a lot of problems with EBM. The one that is discussed most often on SBM is the lack of attention paid to prior plausibility for CAM treatments. However, I think there are more important flaws in EBM as practiced, the primary ones being 1) physicians as a group are still not well-trained enough to analyze evidence critically, and 2) most clinical evidence is of low quality and generalizability.

    An example of a common mistake made by physicians is to think that a failure to demonstrate X is the same as demonstrating that X is false. For example, there are studies that compare two treatments and show no statistically significant difference between them. Most physicians interpret that as meaning that the treatments are equal when, in reality, a huge number of those studies have too small of a sample size to detect a clinically significant difference. Another example would be studies of the type that demonstrate that a drug lowers cholesterol, for example, but doesn’t prevent heart attacks, so physicians think it’s useless. They don’t realize that it’s statistically easier to demonstrate a change in a continuous variable that can be measured for every patient than it is to demonstrate a decreased rate of rare events, so a study can be simultaneously powered to detect very small changes in cholesterol levels but underpowered to detect even large changes in the rates of heart attacks.

    Another example is when physicians get very concrete about EBM and state that a treatment is not “evidence based” because the particular combination of treatment, indication, and comorbidities has never been tested in a large double-blind RCT. An example is the PORT group, which is an NIH-funded group that’s supposed to review the literature and issue treatment guidelines for schizophrenia. In their guidelines, they state that benzodiazepines have no evidence base for treatment of anxiety in schizophrenia. Of course, benzos have a huge evidence base for treatment of anxiety in humans, and schizophrenics are humans, so it stands to reason that maybe it could work for them too. One could make similar arguments, stating that antibiotics have no evidence base for treating cellulitis in Parkinsons patients, or that aspirin has no evidence base for treatment of headaches in diabetics. A less trivial example of this is when one medication has a larger evidence base than another medication in the same class, and there it can get a bit murky.

    Finally, there is the issue of low quality of clinical studies. For practical reasons, many studies are performed on a patient population that is not representative of an actual clinical population. Actual patients have many medical problems at the same time, but for many research studies they cherry-pick the rare patients with a single isolated disease state that are easiest to study. And they have to do this, or else it would be impossible to get anything done. However, this makes things difficult for physicians because we’re often stuck trying to generalize nonrepresentative data to our real world patient population, which can get very tricky.

    Anyway, those are just a few comments off the top of my head.

  52. mousethatroared says:

    fledarmus1- insurance really is a negotiated contract between buyer and seller. In individual plans, the individual negotiates directly with the insurance company and usual pays a higher rate and has more restrictions, due to low volume.

    In a group plan, the employer negotiates with the insurance company, they often get a better rate due to volume, they pay for all or part of the insurance for the employees in lieu of the salary that an employee would expect for a job without benefits. There is also a negotiation between the employer and employee. An employer offering insurance that requires more co-pays or premiums from all employees or only ones with asthma, a genetic defect or congenital condition is going to have a hard time competing for experience, competent workers against a company offering standard insurance.

    The other problem with an employee offering variable benefits based on congenital, genetic conditions is that it would probably be considered discrimation based disability or genetic information, which voters decided should be prohibited by federal law. (crazy voters).

    With government paid insurance, the goverment representatives of the people decide what they think should be covered and negotiate with medical providers…Still negotiations taking place, just slightly different parties involved, and like most government functions, one has to put up with the flaws of democracy.

    You seem to think it would be better if people who were at higher risk for medical expenses paid more. I suspect that you consider yourself low risk. Of course, in your system, you may be just one blood test away from insurance rates that you could never afford. It seems that a lot of voters are not okay with your system. They want to spread the expense out between the lucky folks and the unlucky one. It’s our chimpanzee instincts, I guess. But it’s democracy at work.

  53. evilrobotxoxo says:

    A couple of other problems I have with EBM as practiced:

    - the statistical paradigm used in most RCTs is often set up to determine whether a treatment works or not, or whether treatment A is better than treatment B. The problem is that MDs fall into the trap of that sort of binary thinking, when really we want to know HOW WELL a treatment works, or HOW MUCH better treatment A is than treatment B. But when a P-value is the primary thing that gets reported, it doesn’t help us understand that because p-value is dependent on sample size. I remember a case where someone was telling me about how there were two drugs of the same class, and drug #1 was effective for some indication (e.g. p 0.05). I looked at the studies, and the sample size in the study for drug #1 was twice as large as in the study for drug #2, so the difference in efficacy is probably just an artifact (the drugs have been shown to be equivalent for other indications in other studies). I think someone needs to figure out a different way to report RCT results that is easier for clinicians to interpret. Some sort of confidence vs. effect size plot that says “We’re 99% confident that this treatment is effective. There’s a 50% chance that this treatment will improve outcomes by a factor of 3, and a 70% chance it will improve outcomes by a factor of 2, etc.”

    - in my field, psychiatry, I think too many research dollars are being allocated to RCTs designed to validate and optimize existing treatments, and too few dollars are allocated to developing new treatments. I can’t speak for other fields, but in psychiatry it is a real problem. There are huge NIH-funded RCTs that cost tens of millions of dollars that end up strengthening the evidence base of treatments that already had a decent evidence base, without improving the actual treatments at all. To some extent, mindless worship at the altar of EBM contributes to this.

  54. rork says:

    I’ve been studying Avastin (bevacizumab, I call it anti-VEGFA, call me crazy) for ovarian cancers. Looks like data is not so great, though some papers try to make it seem as good as possible. I saw FDA removed approval for metastatic Her2- breast cancers.
    What I wonder is how many docs prescribe it off-label, fancying they know something. Perhaps, like Wagner, it is really better than it sounds.

    And here’s a complaint: If I wanna test if we’ve actually been killing the older lower-stage women (I actually think that is possible) using the data-sets of the papers I’m reading I can’t do it. If the authors don’t want to ask those questions, or did, but don’t want to tell me the results, that is what they do and I can’t stop it. But not letting me see the data seems wrong. It lets them dodge some tough questions, if reviewers are asleep. And they are.

  55. PJLandis says:

    Hey Jan, one last comment because I recognize this isn’t all that relevant to the post. I guess I’m referring more to legislators but in a democratic country they are one and the same thing.

    Judges and police enforce laws passed by legislators, they don’t decide what is criminal or what to do with people who’ve committed crimes. If a law is passed to achieve some goal then we can evaluate whether it achieved that goal based on the outcomes. If my intuition is that putting more people in jail will decrease crime or drug use, and I put more people in jail for longer but crime or drug use doesn’t decrease, can’t we reasonably say that intuition was wrong?

    I get that they politic to get bills passed, but they also write those bills with the intent to accomplish specific goals for their constituents or themselves. If they fail to accomplish these goals then I think you can judge the law a failure or success, or mediocre, on evidentiary grounds. If the goal is to pass a healthcare bill that maximizes quality and coverage while minimizing costs/waste/fraud, then I think we can make a rational determination as to whether a specific law has effectively accomplished those goals with the least unintended effects.

  56. BillyJoe says:

    fledarmus,

    “I see nothing whatsoever wrong with a patient of any age worried about the possibility of prostate cancer requesting a PSA test.”

    I’m not sure what point you are trying to make here. Of course there is nothing wrong with a patient requesting a PSA test. But what IS wrong is the doctor not explaining to that patient why PSA tests are not recommended. He probably should refuse to order the test but, if the patient insists, there is probably no way to avoid it. If the patient did turn out to have prostate cancer, the doctor would be “in deep shit” if he outright refused to order the test. In Australia, it is the media that is driving the push for PSA testing using testimonials of celebrities who have developed prostate cancer to urge men to attend their GPs to have it done.

  57. nybgrus says:

    an excellent discussion :-D I rather enjoy commentaries that don’t consist of trolls, flamers, and sCAMsters trying to squawk their lack of knowledge.

    in any event….

    The reality is that medical resources will be rationed, but is EBM (or ERM) the best tool for the job. What are the benefits and pitfalls of using that tool for that job?

    as evilroboto stated, what better option does one have? If you can’t/won’t use evidence to guide rationing (which, as Dr. Gorski has argued does exist and must always exist – there is no such thing as unlimited resources) and restricting what treatment are available to what people then what can you use? The adjunct to that is societal desires – there is such a thing as culture shaping what we value.

    For example, a friend of mine believes that health care should be completely opened up to the free market and all care should be paid for up front in cash. Those who can afford it get care. Those who can’t are SOL. Ron Paul seems to think along those same lines (which is why it is no surprise my friend is a Paul supporter).

    This friend has argued with me that this would lead to better care and more accesibility. I believe I have sucessfully and continuously refuted those claims. However, it always ends up that I have to admit that his system would “work” if the mortality and morbidity of a significant portion of the population was his desire and was culturally acceptable. I believe I would leave that culture for a different one.

  58. nybgrus says:

    @evilroboto:

    Anyway, those are just a few comments off the top of my head.

    Those are some very good thoughts. I’ve had many similar ones myself. I believe the answer to the questions posed can be summed up neatly and succinctly (if not broadly) by the concept of “intellectual honesty.” And that is the problem. Creating a set of rules and algorthims that would force intellectual honesty is, IMHO, tantamount to impossible.

    For a completely unrelated, but entirely analagous discussion of the practilaties of this problem you can read the Sam Harris / Bruce Schneier debate on profiling for airport security. The crux of Schneier’s point is interesting – creating vast systems where anything other than simple rules that monkeys can follow is simply impractical to implement in the real world. Harris’ point is that such systems are inefficient and make gross errors of both the false positive and negative type and can miss true positives. I think the points you raise are quite analagous to that discussion.

  59. @BillyJoe, you absolutely have to stop talking out of your rectum. You have no idea what you are talking about with respect to PSA screening and how doctors “should” act. You stated that doctors “probably should refuse to order the test but, if the patient insists, there is probably no way to avoid it.” You are absolutely incorrect. And you are incorrect in the way that you represent the PSA test as “not recommended”, in that you imply that it is something that should be avoided, but only done if the patient absolutely insists. (Your words, not mine.)

    Please re-refer to my long post discussing the PSA above. What I wrote is 100% in accordance with both the standard of care and current guidelines. Your statements are absolutely wrong. Please do not bother to argue with me until you have read the full recommendations that I linked to previously.

  60. nybgrus says:

    Sadly enough, first they had to go to the media, which naturally started crying out “The government is killing our children!” (or compatible) and then the financing guidelines were changed.

    That is the problem. Bombastic rhetoric and polarizing swings are the way people like to function and what they respond to. It also boosts ratings and can get things done. When used for a good purpose the outcome is good. But, once again, that requires intellectual honesty. And the problem is most people lack it.

    It is a bad tactic, and in this case the ends do not justify the means (they rarely, if ever, do IMO).

    BTW – my insurance covers one pap test every three years. However, I’m in a risk group, my gyn/ob tells me I should have it every year, so I pay for it.

    Depends on exactly what you mean by high risk. If you have had abnormal Paps then you should be getting more frequent screens and/or colposcopy. If it is a family history and positive HPV test, but you have had 3 consecutive normal Paps then every three years is the current acceptable guideline. This is based on recent data that demonstrates abnormal Paps frequently regress spontaneously and they take a very long time (can’t remember exactly off the top of my head – on the order of ~5 years) to progress to cancer if they were to do that. So a 3 year interval will still catch it more often than not in a precancerous state.

    If you truly are high risk (you’ve had an abnormal Pap in the last year) and they still won’t pay then I would reckon you have a suit on your hands.

  61. nybgrus says:

    @evilroboto:

    I think someone needs to figure out a different way to report RCT results that is easier for clinicians to interpret. Some sort of confidence vs. effect size plot that says “We’re 99% confident that this treatment is effective. There’s a 50% chance that this treatment will improve outcomes by a factor of 3, and a 70% chance it will improve outcomes by a factor of 2, etc.”

    Ben Goldacre specifically addresses this sort of issue – the example that comes to mind is pharma companies testing the max dose of typical anti-psychotics against a therapeutic (but not maximal) dose of atypicals to demonstrate equial efficacy but less side effects (i.e. tardive dyskinesia). This is, of course, underhanded but is a minor point that can often be hard to glean from studies.

    Once again that falls into “intellectual honesty”… well, and just plain old “honesty” too.

    But yes, the other equally valid point is that people in our profession are not adequately trained to really understand studies. Hell, I’m not adequately trained (at least, I certainly don’t feel like it). But I am significantly better at it than the vast majority of my peers and at least I know these issues exist and my own shortcomings so I can apply more scrutiny and seek more expert opinion on important and/or controversial topics. I can’t say the same for my colleagues.

  62. nybgrus says:

    I’ll close out my string of comments with an anecdote from today (BTW, I am doing it this way since not everyone will care to read all of them nor try and find what they may want to read in a giant wall of text. Plus, I reckon it is not much harder to skip multiple comments entirely over one giant one for those who don’t care to read anything I have to write).

    Today I was early for M&M (mortality and morbidity) conference. I ended up chatting with a real old timer of a doc. I’d guess somewhere in his late 80′s easily.

    He proceeded to tell me (and another student) about what is wrong with medicine these days. Basically, it all boiled down to “that darned evil gubmint.” He hates the USPTF and thinks they are all idiots with their only goal in mind to save money, regardless of what that may mean in terms of human life. He thinks that all the new guidelines changing the screening recs to less frequent and older are purely to save money because Washington (da gubmint) doesn’t care if people die. And he specifically said that the new recs against PSA screening are total crap, because if it wasn’t for a PSA screen he’d be dead himself. He also commented that “in all his years” he’d never had a problem with [X] which is now recommended against, also clearly because “they” want to save money at the expense of people’s lives and health.

    I nodded politely and let the old man rant while I drank my coffee and munched on my fruit plate. The other student was from a different school and this was her first clinical rotation ever – 4th week of clinical experience. She was like a deer in headlights and clearly had no idea what to think. From my limited interaction with her, I doubt she had a clue what might be the basis for anything he talked about, and would likely go on thinking that the evil gubmint was trying to limit our autonomy as physicians to practice medicine. Hopefully I’m wrong or if not, she’ll get disabused of these notions at some point.

    But that was also the running undercurrent of his rant – that “they” are meddling in what doctors do. It was clear that he didn’t like the idea of anyone telling him (or any doctor for that matter) what they can and cannot do. If he wanted to prescribe an antibiotic to treat multiple sclerosis because he’d seen it help a patient or two, why the heck not??

    In any event, the point of this anecdote was to add that IMO much of the pushback against EBM as outlined in Dr. Hall’s piece is likely initiated or at least fueled by this old school cowboy (though even cowboys don’t act like cowboys these days – a greadt TED talk by Atul Gawande worth watching) attitude which is hard to extinguish because this old guard can and does still influence the way my colleagues think and act.

  63. BillyJoe says:

    Skeptical Health,

    “Bottom line: it’s absolutely wrong to say that PSA screening is “not recommended”….”

    I live in Australia, so let me first give the view of the Australian college of GPs.
    The RACGP (Australia) does not recommend PSA screening. They recommend that PSA screening not be done as a routine and that patients should not be encouraged to have PSA screening. They base this on the lack of evidence that the treatment of prostate cancer saves sufficient lives to justify the high rate of morbidity from the biopsy and prostatectomy.

    As for how to handle the patient who comes requesting a PSA, their recommnedation is more as less what you have stated in your long post: Supply the patient with the pros and cons of PSA testing and then let them decide. The obvious reasoning behind this is to avoid alienating patients who have been convinced by media and testimonials from celebrities who have been diagnosed with prostate cancer after having had a PSA test, that all males over a certain age should have the test done. The other obvious reason is to avoid litigation in a case where a patient is declined a PSA test and then is found to have probably had prostate cancer at the time of the test.

    Similary you reference, the USPSTF, also does not recommend PSA testing. However, as you say, they do not recommend against it either. In other words they are being more circumspect.

    So, as a sceptic, you tell me: Given the lack of evidence, why would you recommend it? And why wouldn’t you recommend against doing it unless and until that evidence become available? Why, as a sceptic would you be led to make the following statement:

    “… For me? I want to know, and I’d want it out. If my PSA is elevated, I’ll be first in line to get a biopsy, and if the biopsy has significant Gleason score, then I’ll be at MD Anderson having my prostate cut out with a robot. I couldn’t live knowing that I have prostate cancer waiting to kill me slowly and painfully.”

    Your last sentence is also disengenuous in that you know, or should know, that most men die with prostate cancer rather than of prostate cancer; and that you are far more likely to suffer life-long morbidity from the prostatectomy than be saved by the procedure.

  64. evilrobotxoxo says:

    @nybgrus: I think your point about first- vs. second-generation antipsychotics raises an important question about how who pays for research determines what research gets done. In that case, I know there was some sketchy stuff in the trials, but it is true that second-generation antipsychotics do cause lower rates of tardive dyskinesia (with the possible exception of risperdal). But nobody has been able to quantitate the TD risk of each antipsychotic in a head-to-head trial because no company would pay for that and (probably) no academic would want to do a study like that, despite its clinical importance.

  65. @BillyJoe,

    I don’t know anything about the RACGP, but this one professional organization of Australia does not dictate how the rest of the world practices medicine. In fact, a survey of Australian GPs found that more than half (55%) of them do *not* agree with the RACGP’s belief to avoid PSA testing:

    http://www.medicalobserver.com.au/news/half-of-doctors-reject-racgp-advice-on-psa-tests

    Further Google-ing finds that other Australian professional medical associations disagree with the RACGP, for instance, the Royal College of Pathologists of Australia.

    As for your question “So, as a sceptic, you tell me: Given the lack of evidence, why would you recommend it?” This implies that you simply do not understand the purpose of the PSA, nor did you fully comprehend my somewhat lengthy post about it. I never wrote that I would recommend it, as in “Good morning, thank you for coming in today. You’re 50 years old, so we need to screen for prostate cancer with a PSA.”

    What I wrote, and what is ETHICAL, is to have a conversation about PSA. “Good morning. Well, I’m glad that you’re interested in remaining healthy. Oh you’re curious about prostate cancer? Do you have any family history of it? What other medical conditions do you have? Have you had any urinary symptoms? (…) What is your understanding of the PSA? Well, the PSA can and does detect prostate cancer, but before we go down that road, there are things about the test you need to understand, such as the possibility of false positives leading to unnecessary procedures, etc.”

    My writing of “I would, or would, not screen” to me, really means bring up a conversation about PSA. I would never refuse to order a PSA on a patient who was concerned about prostate cancer, regardless of what I believe their risk is. (I am not psychic. I do not intuitively know if they have malignant prostate cancer.)

    So, to answer your question, I have conversations with my patients about it. Please re-read my previous post for specifics. I think the point that you aren’t getting is that not everyone has equal risk for prostate cancer, and not everyone has equal risk of dying from prostate cancer. Again, if you had a 50 year old male that had 2 brothers, an uncle, and a father with prostate cancer then I would screen early. If you had a 90 year old male who is otherwise healthy but hasn’t seen a doctor in 30 years, then I would not screen. So, while we shouldn’t push PSA screening on patients, we need to give them the information and the choice to pursue PSA screening, based on their beliefs and attitudes towards cancer.

    As to your question of me saying that I would happily remove my own prostate cancer: my statement is absolutely in line with what is ethical and in line with recommendations. My personal attitude towards cancer is that I do not want it, and I could not live knowing that I have prostate cancer actively growing inside of me. I am in my 30s, and I am absolutely healthy. I have no sudden or early death in my family, but I do have a family history of prostate cancer and multiple myeloma (one in each parent, both “in remission” by state of the art medical treatment.) I am not riddled with other diseases that lowers my life expectancy. I will undergo regular PSA testing and if mine is elevated, I will undergo biopsy, and if a significant Gleason score is found, I will have my prostate removed. I obviously know that prostate cancer usually doesn’t kill males (please re-read my previous post), but do not forget that prostate cancer is the second leading cause of cancer death in men (at least in the USA.)

    My statement is not disingenuous and I do not appreciate the suggestion of that, when in reality it is simply that you are attempting to generalize and speak in absolutes on a topic that you simply do not fully understand. I’m sure your beliefs about prostate cancer are somewhat troubled by your father being diagnosed and treated for it, but you simply cannot do a little bit of internet research and start making blanket statements on how doctors should practice. You are not a doctor and do not understand how medicine is practiced.

  66. weing says:

    ” most men die with prostate cancer rather than of prostate cancer; and that you are far more likely to suffer life-long morbidity from the prostatectomy than be saved by the procedure.”

    But don’t forget the tip of the iceberg. The second leading cause of cancer deaths in men is prostate cancer, with a rate that is double that of pancreatic cancer. And it isn’t painless. Medicine is hard, isn’t it? What we need is a better test. One that would be positive for the aggressive tumors and not the indolent ones.

  67. mousethatroared says:

    nybgrus “as evilroboto stated, what better option does one have? If you can’t/won’t use evidence to guide rationing (which, as Dr. Gorski has argued does exist and must always exist – there is no such thing as unlimited resources) and restricting what treatment are available to what people then what can you use? The adjunct to that is societal desires – there is such a thing as culture shaping what we value”

    Dr Gorski argues there’s rationing? Does he also argue that water is wet? :)

    Kidding aside, nybgrus – i think we are on different tracks and you missed my station (oh wait, that was a little more kidding) I wasn’t against using evidence to guide resource allocation, I was suggesting that we look at the best and worst ways that evidence can be used so as to avoid some potential flaws in an evolving system.

    evilrobotxoxo has some great observations in that area.

    @evilrobotxoxo – I know that gushing is unfashionable, but, can I just say how much I love your comments. They are so insightful and thought provoking, I get excited whenever I see your name in the comment’s section, I wish you had a blog.

  68. nybgrus says:

    Skeptical, BJ:

    I think you two are talking past each other a bit.

    SH – BJ and I have a long history on this forum and he is consistently rational, skeptical, and evidence based. He makes some mistakes but they are not terribly common and is always amenable to having his point of view changed.

    As for the discussion regarding PSA I think you are both on the same page. Especially in Australia (where I did two years of my own medical education) “screening” is used very much to indicate the testing of anyone meeting minumum criteria. You are clearly talking about different criteria for high risk patients and talking about PSA velocity and trends. I believe BJ was envisioning a single spot test of a random Joe Blow and then him having a PSA of 4.5

    Just as was the case with the USPTF chaning the screening guidelines from women aged 40 to women aged 50, the PSA test has been continually criticized for the excessive morbidity associated with false positive which are common. The rate of false positives decreases as your population becomes more high risk and as you test PSA velocity rather than spot tests. In those cases the PPV is often such to make it worthwhile. However, also just like the mammagrophy changes the critiques against the PSA have been vehemently opposed by many docs – typically like those in my anecdote above where they “just know it saves lives” and have been “using it forver and it works great!” and “how can you tell me, a doctor what tools I can and can’t use to treat my patients the best!” So, SH, citing a popular media article about a bunch of docs being angry at the RACGP regarding the PSA testing policies doesn’t bolster your case, IMO.

    As for the robotic surgery – the evidence also shows that it does not have any better (or worse) outcomes than traditional prostatectomy. So opting for that specifically is not the most “skeptical” thing to do since you are paying more for essentially the same thing. Having said that, I am of the opinion that the reason the outcomes are the same is because we have not yet had a generation of young doctors who grew up playing video games trained from the beginning in its use. In other words, a physician that is quite comfortable in that sort of virtual environment and has been training on it from the very beginning of their medical career would have better outcomes than traditional prostatectomy. Once again this is my opinion (further solidified after having actually used one of the robots on a prostatectomy about 3 weeks ago) and I cannot predict how much better the outcomes would be and if they would really be worth it.

    So hopefully I am right in my assessment of y’all talking past each other since the both of you are actually the kind of intelligent skeptical contributors I like seeing around here (despite SH’s tendency to hurl insults rather loosely :-p)

    BJ – SH is talking about carefully applied PSA to higher risk populations and doing it in a manner that will engender a good physician-patient rapport taking into account the significant media hype surrounding it.

    SH – BJ is talking about not routinely recommending every male over some age get PSA testing and actively recommending against it in low-risk patients.

    And of course, weing is right – but we need that for all cancers, not just prostate.

  69. nybgrus says:

    I wasn’t against using evidence to guide resource allocation, I was suggesting that we look at the best and worst ways that evidence can be used so as to avoid some potential flaws in an evolving system.

    Ah. Mea culpa then. It was the end of my day and I was tired. Must have just misread things. That does indeed seem reasonable and a very good thing to do. And once again comes down to intellectual honesty IMO.

    And yes, evilroboto does have very good posts. I guess I’ll have to take back all the nasty things I’ve said about psychiatrists…. (j/k!!)

  70. mousethatroared says:

    nybgrus – I guess I’m not understanding the intellectual honesty thing. You think EBM as it is today would work out perfectly if everyone just decided to be intellectually honest?

  71. nybgrus says:

    @mouse:

    Perfectly? No. But pretty darned close.

    I am of the opinion (shared here by the authorship as well, I believe) that the framers of the EBM hierarchy did so assuming intellectual honesty. Or at the very least that the way the evidence hierarchy is structured demands intellectual honesty to function properly. The thesis of SBM (as in why the “S” instead of the “E”) is that adding a Bayesian function – a priori likelihood based on basic scientific understanding – will mitigate a large part of the problem EBM has with accepting, utilizing, and wasting resources on highly improbable ideas (like Reiki or homeopathy).

    But even that codification – a requirement for all EBM to be viewed from a Bayesian rather than frequentist perspective – won’t mitigate all the problems which will inevitably arise. No system, no matter how complex or well thought out – is perfect after all. So there will always be a role for intellectual honesty.

    Evilroboto raised exactly the kind of issues I am speaking of. Nobody with intellectual honesty would reasonably say that benzodiazepenes can’t be used to treat anxiety in schizophrenics. The best they could (and should) say is that it is untested and we know the neurochemistry of schizophrenics is different than a normal population so there might be potential for unforseen adverse effects. If it seems significant enough, a study could be done. Otherwise, we should act in an intellectually honest fashion and administer the medication as necessary, but with extra scrutiny (which schizophrenics should be receiving anyways).

    Also, the trick is that people, at least in many cases, can’t just “decide” to be intellectually honest. They can decide to try and actively pursue it (as I have) and then it takes effort and a willingness to learn in order to achieve it. It is a process – not an end state.

  72. mousethatroared says:

    nybgrus – Nope, sorry, still not getting it. They set up a framework based on the idea that people would be honest? That’s like a software engineer building an e-commerce site based on idea that people would respect other’s property. Or a nation setting up it’s state department based on the idea that people would be peaceful.

    Maybe I’m misunderstanding…

  73. evilrobotxoxo says:

    @mousethatroared: thanks!

    @nybgrus: as far as “intellectual honesty” goes, I agree with your overall point, but I would call it something more like “intellectual rigor.” For example, the people who think benzos aren’t an evidence-based treatment for anxiety in schizophrenics are not being dishonest, they’re being overly concrete in their thinking, which is a type of intellectual sloppiness. Dishonesty implies some sort of malicious intent, but I think most of the people in question mean well, they just don’t always draw the correct conclusions from the data.

  74. nybgrus says:

    They set up a framework based on the idea that people would be honest? That’s like a software engineer building an e-commerce site based on idea that people would respect other’s property.

    Imagine living in a country where no steals. Ever. You’ve heard of this concept of “stealing” but it is very rare and only happens in other places that never have an influence on your own country. So you design a market system that doesn’t have safeguards against theft not because you explicitly expect everyone in the world not to steal, but because the notion is so far removed from your experience that you don’t think much about it.

    In academia there was a time when it was easy to marginalize quacks and charlatains and when writing out the rules of EBM it would have been easy to think, “I know this “reiki” and “energy healing” stuff exists but it is weirdos and quacks who practice this” and then have no impulse to include a specific clause that while lowest on the hierarchy basic sciences should inform the validity of higher forms of evidence such that “reiki” would never even have a shot at an RCT because it makes no sense. That is simply no longer the case with the re-branding of quackery to be “alternative medicine” then “complimentary and alternative” and now “integrative” medicine and “functional” medicine. Many decades ago can you imagine a serious scientist thinking a Yale medical professor would be lauded for claiming that a different standard of evidence is necessary to accept “CAM” (whatever that means) so that failures to live up to typical standards still equates to parity in validity and recommendation by medical professionals? I think not (and were I Descartes I would disappear now).

    Perhaps evilroboto is right – intellectual “honesty” might not be the most correct and fully encompassing phrase and “rigor” may be more apt. I would say both need to be taken into account. Though from my understanding at least, the “honesty” in intellecual honesty does not necessarily imply an actively malicious agenda on the part of those who don’t engage in it (though that certainly does exist). I see it at mental shortcuts that we give ourselves a by on because we aren’t being honest about the import of them or even the fact that we are doing them. That could be called a lack of “rigor” I suppose. But I’ll suffice it to say that semantics aside, that is indeed the largest fatal flaw of EBM and one that leads to exactly what mousethatroared has encountered – methodolatry. The reverence of the RTC above and beyond anything else. Intellectual honesty and/or rigor would also preclude the pendulum from swinging in that direction as well.

  75. mousethatroared says:

    Okay, I’m beginning to get what you are saying, nybgrus, thanks for elaborating.

    But if what you are saying is true and EBM is assuming intellectual honesty or rigor to be successful, then considering the fact that at least two major players in the medical field, drug companies and insurance companies, have a financial interest in presenting evidence to their advantage, possibly without intellectual honestly or rigor, is it surprising that a system based on the players pursuing those virtues may be distrusted by some doctors (quoted in HH’s article) and some patients (often commenting here)?

  76. BillyJoe says:

    nybgrus,

    I think you hit the nail on the head.
    Sort of saves me from replying except to say to SH (hey that’s almost as bad shorthand as BJ)…

    Skeptical Health,

    I trust the RACGP above the urologist college and the pathologist college for two main reasons: The RACGP look at all the evidence, and continue to look at any new evidence that becomes available. And they don’t have a vested interest. The other two colleges do not look at all the evidence, they both reacted to a bit of new unconfirmed evidence. And, of course, they have a vested interest.
    The third reason is that, because of my family history, I have looked at the evidence myself over the years since my father’s diagnosis – so I can’t provide links easily – and the RACGP guidelines are consistent with that evidence and a sceptical attitude towards that evidence.

  77. JMB says:

    The primary decision of the USPSTF to to assign a grade to a strategy used in preventive medicine. Under Obamacare, the grade becomes a directive to health insurance companies of what the minimum coverage is required to be offered in the health insurance plans. Grade A or B must be covered. Grade C or D need not be covered. They did not intend their grading to be used for individual patient care decisions. After the uproar over the change in mammogram recommendations, they are careful to include statements that for individual decisions, a patient should have a discussion with their doctor to assess the patient’s individual risk, and consider the individual patient’s assessment of the value of the benefits and risk.

    The USPSTF methods changed in 2009 to incorporate new ideas. The traditional method was to review clinical trials to determine the expected mortality rate reduction, and then estimate the cost per year of life saved (although some changed cost of year of life saved to cost per year of life saved multiplied by a fraction representing quality of life after intervention). The new method incorporated computer simulations to estimate mortality from different strategies, and use the Efficient Frontier analysis to pick the most efficient strategy from the computer generated data.

    I believe Oregon state had a government designed rationing program in which the medical science was used to rank the effectiveness of a treatment or test. Those interventions that reduced mortality were ranked highest, followed by those interventions whose benefits involved improvement of quality of life, but not increased longevity. With that ranking set by the medical science review committees, the elected politicians would then vote on the funding for the medical interventions. The available funding was applied first to the most effective interventions. Consequently, the state legislators would have to face the consequences of their votes for funding with the constituents being fully aware of where the politicians were drawing the line on funding. I am not sure that is an accurate description of the process, but the effort resulted in a sophisticated apparatus for assessing medical effectiveness. The Oregon Evidence-based Practice Center is routinely assigned the task of evidence synthesis by the USPSTF.

    In Obamacare, the politicians are spared the wrath of the constituents for rationing resources because such decisions are made by committees of appointees. There are convoluted effects from the process of grading strategies for health insurance companies and meaningful use criteria for electronic health records that ultimately means that not only will the insurance companies not pay for grade C or D interventions, but the healthcare provider will receive less money for their work when they vary frequently from guidelines (also known under the euphemism of paying for quality… when the government committees devise the grading system for quality).

    I would have preferred the Oregon state approach. Hold elected politicians accountable to the constituency for those decisions. Let medical scientists only debate scientific data, not resource allocation.

    The American brand of socialized medicine (medicare/medicaid/VA/military) was unique from other countries’ socialized medicine because of the open ended funding for interventions that cost no more than $50,000 USD per year of life saved. Now it will be unique because politicians won’t have to suffer the consequences for government funding decisions. However, there certainly has been a track record in Washington of dealing with such issues poorly in the past (there have been plenty of interventions funded that failed the criteria of $50000 per year of life saved).

    Here are links to the evidence synthesis articles for the USPSTF guidelines on PSA screening and mammography screening.

    mammography
    http://www.ncbi.nlm.nih.gov/books/NBK36392/

    PSA
    http://www.ncbi.nlm.nih.gov/books/NBK36392/

    8 RCTs on screening mammography graded fair gave an average .85 RR for breast cancer mortality in the age group 40 – 50 with screening mammography. For PSA, the two RCTs graded fair yielded an average of no discernible benefit (RR 1.11 and 0.80 ).

    Consequently, there is a fundamental difference in the USPSTF decisions on the grading of screening mammography and PSA screening. In the mammogram screening guidelines, the evidence synthesis concluded that, “Mammography screening reduces breast cancer mortality for women age 39–69 years; data are insufficient for women age 70 years and older. False-positive mammography results and additional imaging are common.” The evidence synthesis concluded that PSA screening “After about 10 years, PSA-based screening results in the detection of more cases of prostate cancer, but small to no reduction in prostate cancer-specific mortality. ” The USPSTF didn’t bother with the Efficient Frontier analysis in the case of PSA because the evidence was insufficient to confirm a mortality reduction from screening for prostate CA. The Efficient Frontier analysis was utilized int the decisions for grading mammogram screening strategies because the evidence from RCTs supported a decrease in breast cancer mortality.

  78. BillyJoe says:

    “” The evidence synthesis concluded that PSA screening “After about 10 years, PSA-based screening results in the detection of more cases of prostate cancer, but small to no reduction in prostate cancer-specific mortality. ” The USPSTF didn’t bother with the Efficient Frontier analysis in the case of PSA because the evidence was insufficient to confirm a mortality reduction from screening for prostate CA.”

    Seems the USPSTF are in agreement with our own RACGP(Australia).

  79. BillyJoe says:

    …the link is rong though.

  80. @BillyJoe, you “trust” the RACGP because their position is directly in line with your own personal beliefs.

  81. mousethatroared says:

    @SkepticalHealth – perhaps you have some sort of qualifier filter on your keyboard or are you using your ESP to channel BillyJoe’s brain waves?

    Normally one might say “I’m speculating that, I suspect that or you may possibly be” (trusting the RACGP position)

    Personally, I find these kinds of qualifiers important. They show the reader that you can tell the difference between an established fact and a speculation. If too many of your speculations are stated as fact, well… that makes me sad. :(

  82. @mousethatroared, that was so utterly douchey that it gave me chills.

  83. mousethatroared says:

    @SkepticalHealth – oh, I guess it was impolite for me to notice that you are making stuff up and calling it fact.

  84. BillyJoe says:

    In fact, given my family history, what I have had to do is to guard against the natural tendency to look for evidence that would suggest what I might do to lessen my chances of dying of prostate cancer.

  85. mousethatroared says:

    also@ SkepticalHealth – Actually, I’m sorry that was the wrong tact on my part.

    I do think that your credibility would be greatly improved if you qualified your statements a lot more based on what you know vs what you are speculating (even in social interactions). It actually really bugs me, because otherwise you seem knowledgable and like your intentions are good. Clearly, you are more intelligent than me. It seems a waste to undermine that for lack of a little finesse. i felt I should say something. but I don’t think it was my intention to send an over the top douchey comment your way.

    I think was going for a critique that includes a joke to take the bite out of the criticism. But to be honest, I’m sometimes not very good at taking the bite out criticism, sometimes a little snark for snark’s sake is hard to resist. Of course, Either way, it was poorly conceived and/or executed if the end result was to give you chills.

    In fact, clearly I shouldn’t be telling anyone about finesse.

    I don’t know, maybe I shouldn’t have said anything at all. Sometimes I’m ultimately happy that someone pointed out something that they think is a flaw in my approach. Other times, not so much.

    My second comment, just knee jerk annoyance at being called douchey. Pointless, that.

    Regardless, I will leave you alone on this qualifier thing and mind my own business, now.

  86. mousethatroared says:

    @BillyJoe, also, my apologies for butting in on your discussion with SkeptH – with a tangent.

  87. pmoran says:

    If I may butt in — I have read what you say, Skeptical health, and I am not sure that you fully grasp the philosophy of whole population screening. We doctors tend to have a pateint focus, not a population focus.

    The argument against PSA screening is not that an occasional life will not be saved by it but that the global cost of those is excessive.

    I think you know this but haven’t yet grasped the implications for public policy . The simple fact is that no health care system of the modern era can support a policy where it costs X millions of dollars to save one life, on top of which there are more costs in lives lost unnecessarily and loss of quality of life in many others.

    If a patient asks about screening (probably unfortunately for them in all but rare cases), a different set of questions arise.

  88. Please do not presume to know what I am and am not aware of. I find that highly offensive and condescensing especially when you obviously have not followed this discussion at all.

  89. Harriet Hall says:

    “The argument against PSA screening is not that an occasional life will not be saved by it but that the global cost of those is excessive.”

    It’s not what universal PSA screening costs in dollars to the whole population, but what it costs in morbidity to individuals whose outcome is not improved and who suffer unnecessarily. On average, the harm is greater than the benefit for individuals at average risk. Testing is justified in individuals at higher risk.

    You could compare this to annual chest x-ray screening for lung cancer. Occasionally a cancer would be detected early with improved outcome, but many other patients would have invasive diagnostic procedures to determine that an abnormal x-ray finding was benign and would be exposed to radiation with no benefit. So we don’t do annual chest x-rays even on smokers; and the reason is not that x-raying everyone would be too expensive.

  90. I mean seriously, nobody here advocated pan-screening males for prostate cancer. The “differences of opinion” are actually rather quite small. BillyJoe seems to believe that nobody should be screened for prostate cancer, except to prevent legal troubles, and I wrote that the ethical thing to do is have a discussion with patients about prostate cancer screening if they are interested in it, based upon their personal preferences, risk factors, and co-morbidities. My statements are absolutely correct, ethical, and in-line with current recommendations in the USA (where I am a practicing physician.)

    Here, let me paste in some information from the USPSTF website:

    Most major U.S. medical organizations recommend that clinicians discuss the potential benefits and known harms of PSA screening with their patients, consider their patients’ preferences, and individualize screening decisions. They generally agree that the most appropriate candidates for screening include men age 50 years or older who have a life expectancy of at least 10 years. These organizations include the American Academy of Family Physicians, American College of Physicians,16 American College of Preventive Medicine,17 and American Medical Association. The American Cancer Society18 and American Urological Association19 recommend offering PSA measurement and digital rectal examination to men annually beginning at age 50 years.

    Gee. Sounds familiar. Spot on with what I have been writing.

    We never even discussed costs. Nobody made any claims regarding costs in any way whatsoever. In looking through the USPSTF statement, supporting article, and evidence synthesis, I could only find the word “cost” twice, in a diagram. No matches for “expense” or “price.” While nobody denies the costs associated with, well, everything in health care, we simply were not discussing costs whatsoever, so your condescending “interruption” was completely unwarranted and unappreciated.

  91. David Gorski says:

    You could compare this to annual chest x-ray screening for lung cancer. Occasionally a cancer would be detected early with improved outcome, but many other patients would have invasive diagnostic procedures to determine that an abnormal x-ray finding was benign and would be exposed to radiation with no benefit. So we don’t do annual chest x-rays even on smokers; and the reason is not that x-raying everyone would be too expensive.

    The same has been generally true of doing chest CT scans for lung cancer even in high risk populations, such as smokers, where the expected incidence of lung cancer would be much higher than in the general population.

    People, even sometimes doctors, tend to scoff at overdiagnosis, but overdiagnosis leads to overtreatment, which leads to morbidity with no possible benefit. It’s only been recently that overdiagnosis and overtreatment have become major concerns. I’ve written about these issues fairly extensively right here on this very blog.

  92. As always, Dr. Hall worded the issue about costs and morbidity perfectly. I do wish I was as good with words as she is.

  93. weing says:

    It’s interesting that you mentioned CT for lung cancer. The recommendations have just come out for low-dose CT for lung cancer screening. Prior to this all the studies showed no benefits from screening. These new recommendations will probably drive up costs. http://abcnews.go.com/blogs/health/2012/05/20/major-cancer-groups-recommend-ct-scans-for-lung-cancer/

  94. pmoran says:

    Harriet: It’s not what universal PSA screening costs in dollars to the whole population, but what it costs in morbidity to individuals whose outcome is not improved and who suffer unnecessarily.

    I think I made it clear that I was including those costs. However, repellent as it may seem to some minds, there IS also a limit to the monetary cost that any health care system (other than an entirely self-funded one) can afford to pay to save one life. I can’t speak for America, but the sums were certainly done before the UK and Australia introduced screening mammography. It was worked out that each life saved should cost about 40,000 pounds Sterling if my memory serves me correctly, and that was regarded as quite reasonable. This was back in 1970 or thereabouts.

    There has to be some interest in costs per life saved if there is going to be a rational and effective distribution of health care resources, with affordable insurance and taxes, and tolerable levels of rationing of other life-saving resources.

  95. pmoran says:

    We never even discussed costs. Nobody made any claims regarding costs in any way whatsoever

    All medicine is a matter of cost vs risk vs benefit, and screening with PSA is especially so because of the potentially massive costs in unhelpful medical services that can be generated, along with unnecessary patient anxiety, suffering and deaths and all this extending over many decades.

    If you have indeed not allowed for the costs you are generating then I say again that you don’t quite grasp the philosophy behind such screening programs.

  96. Harriet Hall says:

    “There has to be some interest in costs per life saved”

    Of course, but the role of science is to determine what will save lives, to provide information. Then society can use that information to make value judgments about how resources are to be used. Bodies like the USPSTF should be in the business of informing about optimum measures, not of judging whether society can afford specific measures.

  97. @pmoran, I have absolutely zero respect for you.

    You made a colossal mistake in that you wrote an extraordinarily condescending post to me implying that I’m missing the whole point by not accounting for monetary costs. Nobody was discussing costs, and in fact we were talking about the increased morbidity and mortality associated with PSA screening. Everybody here acknowledges it. I was writing, correctly, that PSA screening should be discussed with and offered to patients based on their personal beliefs, risk factors, and co-morbidities. Nobody recommended pan-screening.

    You jumped in and wrote a ridiculous post that shows you either have no clue what the actual issue with PSA screening is, or that you simply didn’t read the thread and attacked me for no reason. Either way, it’s your mistake. God forbid you admit it instead of trying to divert the discussion sideways by focusing on monetary costs.

    Of course, *everybody* is aware that EVERYTHING in medicine costs money, and that unnecessary tests and procedures strain an already strained system. If you think that information is something unique or unusual that needs to be pointed out then I feel terrible for you because you’re stuck at square one. Hey look at that, the sky is blue, and trees have leaves. Who would have thought?

  98. BillyJoe says:

    Skeptical Health,

    “Dr. Hall worded the issue about costs and morbidity perfectly. I do wish I was as good with words as she is.”

    I think you should take a deep breath and count to ten.
    Nobody is being rude, condescending, presumptious, or offending.
    They are simply replying in their own way to what you are saying or what they think you are saying.

    I agree that the original discussion was not about cost.
    It was about benefits v risks.

    The discusion began by your statement, which could not possibly be based on a sceptical appraisal of the evidence for PSA screening:

    “… For me? I want to know, and I’d want it out. If my PSA is elevated, I’ll be first in line to get a biopsy, and if the biopsy has significant Gleason score, then I’ll be at MD Anderson having my prostate cut out with a robot. I couldn’t live knowing that I have prostate cancer waiting to kill me slowly and painfully. ”

    That’s an unqualified decision to have a PSA test (the evidence does not support that), followed by a biopsy if elevated (fine) followed by robotic prostatectomy (the evidence is that the extra time and cost of the robotic procedure does not produce a better outcome) in order to avoid prostate cancer killing you slowly and painfully (not c/w the facts; and an appeal to emotion)

    If you are over 50, or if you are over 40 and have a strong family history of prostate cancer, you have not said so. Even then, you would need to consider the evidence that treatment of prostate cancer offers a “small to no reduction in prostate cancer-specific mortality” in exchange for a very high risk of ongoing morbidity (impotence and incontinence).
    This is why the RACGP does not recommend PSA testing.

    (The RACGP agree with you on how you should handle a patient who presents to you for a PSA test – as have I – so that is not the argument here)

  99. I believe in “Experience-Based Medicine”… I hear about something, (non toxic and safe) I try it..if it works, I go with it. So far, pretty good! :)

Comments are closed.