• Priceless: Chapter 4

    If you haven’t read the prior posts on John Goodman’s book Priceless, catch up using the Priceless tag. This post pertains to Chapter 4. I will warn you up front, to me this was the oddest chapter yet. Many arguments either didn’t relate to the main thrust of the book or didn’t make sense to me. Sometimes I felt like a punchline was left out. Examples follow.

    The chapter begins with one of the strangest attacks on and rejection of science I’ve ever seen. I think what John might be trying to say is that a dogmatic imposition of how to use science in the exam room or operating theater is a threat to innovation. Putting aside whether I would agree with that, the fact is, he doesn’t clearly say this, and only this. He goes much, much further, and, in my view, unnecessarily and misguidedly so.

    At heart, and by formal education, I’m a scientist. Evidence is very important to me. But I’m not a fool. I know there are all manner of problems in the process of gathering, reporting, and interpreting evidence. There is bias. There is corruption. There is lack of repeatability. There is lack of generalizability. There is bad science. There is non-science. There is motivated thinking. In the face of these, we should strive hard to improve the process and use of science, not reject it.

    For, what is the alternative? Maybe, in time, John will tell us. While we wait, should we banish science and shun all evidence? Should we reject out-of-hand all the evidence he cites in his book? Should we discard evidence that the health system is as flawed as he (and I) think it is? Look, if neither doctor nor patient have access to evidence-based guidance in some form — not necessarily hard-and-fast rules, just something to light the way — they can only stumble in the dark on faith. Does your doctor talk to God? About your health? Are you sure? Do they teach that in medical school before or after the courses based on medical science?

    Discussion of how to better acquire and use evidence are worthwhile. For instance, on the subject of heterogeneity with respect to comparative effectiveness research, see the recent Health Affairs post by Dana Goodman and colleagues. But John’s treatment of evidence didn’t invite that discussion. Why not?

    Oddly, toward the end of his attack on science and evidence-based standards, John admits that

    Evidence-based guidelines could be a boon to medical practice, helping doctors do their jobs. One place where standardization seems to be working remarkably well, for example, is at Geisinger Health Systems in central Pennsylvania.

    If there is a doctor at Geisinger who thinks patient care would be better without evidence-based guidelines, I’d like to meet her. In all honesty, it is after reading this section that I was strongly tempted to read no further.

    But I did. Next, John covered ways in which employers and insurers try to select healthier workers and consumers. I have no significant disagreements with him here. Along the way, he praises the bureaucrats in the Office of Personnel Management (OPM) for their ability to prevent the worst insurance company abuses in the Federal Employees Health Benefits Program (FEHBP), as well as those who run managed competition-like environments for state employees, universities, and large corporations. Why does he expect managers of state exchanges under Obamacare cannot do the same? I don’t know. Still, I agree with him that risk selection among competing plans is a valid concern.

    John critiques the Medicare drug benefit’s donut hole and notes it will close under the ACA. Which is better policy? He doesn’t say. Was something cut from the book in the editing process?

    The chapter closes with a brief section in which John notes that Medicaid providers receive relatively low reimbursements. To the extent that results in access or quality problems, no argument that this is problematic. But John didn’t go into this in the chapter. Again, it felt like something was missing.

    The chapter is titled “What Being Trapped Means to You.” John never explained.

    Expect my Chapter 5 post later this week.


    • I agree – the attack on evidence guidelines is astounding. What Goodman doesn’t seem to realise when he writes:

      “When these tools developed by outside entities substitute for the doctors’ judgments, however, patients are likely to be the losers.”

      is that there is substantial harm the world over precisely because too many doctors are _not_ following guidelines, often because they are working in specialisms divorced from multidisciplinary working and are also often incentivised not to use them.

      The overall problem with the book is becoming clear – Goodman is acting very much like a specialist who thinks he knows best despite the evidence…

    • Austin writes: “Oddly towards the end of his attack on science and evidence-based standards”

      Austin, it almost sounds as if we are reading two different chapters. I didn’t think John attacked science at all, but he can explain for himself. Earlier he brought up self-interest and I think one of the points is that self interest can affect the science of government agencies just like it can affect doctors and all  individuals. We saw this in my discussion with Aaron where he claimed his science was better than mine even though his way of looking at outcomes was littered with the problems of genetics, societal problems, data collection problems etc. Conflicting data was summarily dismissed. These answers are not so clear. To date our system has used a lot of data that is mixed with these problems and others. That is not science rather self interest where one manipulates data so the data creates the desired conclusion. That IMO is not science and that is where I believe the problem of evidence-based medicine begins because all too frequently what is touted as evidenced-based is not evidenced-based. I won’t even delve into the fact that the evidenced based guidelines might have had testing on populations that were disimilar even with regard to age and that patients with multiple diseases are seldom if ever taken into account by the studies performed. Guidelines hava a place, but we have to be very careful about them and make sure they are not misused and interfere with innovation.

    • Austin,

      You are completely mischaracterizing John’s writing here. He isn’t attacking science at all. He is critical of the bureaucratic imposition of “guidelines.” He makes six specific points about the problems with imposing guidelines and you fail to answer even one of them. He says:
      1. In most of medicine there are no guidelines.
      2. Guidelines are written for the average patient, but few are average.
      3. Guidelines are often written by parties with a conflict of interest.
      4. The studies underlying the guidelines often target a limited population.
      5. Randomized controlled trials are extremely difficult and expensive.
      6. Many treatments have to be customized to individual patients and guidelines are of limited help here.

      He supports each of these points with current research, but this should all be self-evident. In fact, rather than standardizing medicine (as guidelines seek to do), we are entering an era of customized medicine that recognizes every patient is unique based on variations in DNA, comorbidities, health histories, family support resources, and emotional status. A drug that works well for one patient might kill another.

      This is a key topic in current thinking. I am stunned that you dismissed it so blithely.

      • Greg (and Al),

        I acknowledge that you have valid views of the chapter and John’s message. I did address and acknowledge some of his specific comments (e.g., about heterogeneity and all manner of bias and bad science). I have no debate with him on many of the other points. (I can’t list every single area of agreement or disagreement in each post. I am picking a few examples from each chapter.)

        To give a clearer impression that he embraces good science, it would have been nice to see more affirmative statements to that effect. I am afraid that most readers will interpret his lengthy list of the problems as reason to discard evidence and support of it. We need more, not less.

        More broadly, in the first half of the book at least, John often provides far more destructive dialog than constructive. That gives a distinct impression. I am sorry if you or he does not like my impression. If he cared enough about it, he’d adjust his presentation in future work accordingly. (Personally, I do care about the impression I give to those who react strongly to my words. And I do adjust my presentation when I think those impressions are not of benefit to me or the world. In short, reactions as feedback are useful to me. I put them to use. I adjust as warranted. It’d be nice to see others do the same, not always, just sometimes. It is the hallmark of an active, open mind, and of an individual who wishes to work with, not just fight against, others.)

        My post was neither dismissive nor blithe.

      • No, no and no. What is your specialty? Where do you practice? Yes, every patient is unique. We should always treat them as such. That said, most patients, usually by a very large percentage, react in predictable ways to most meds and therapies. Guidelines have proved helpful over and over when applied. Your drugs that kill one person while helping another are, setting aside allergies and improper dosing, are pretty rare when you are using the drug for the right patient.

        I have often said that medicine is the art of applying large scale statistics to a single person. We are fortunate that people are enough alike that what we do for most people works for most people.

        As an aside, let me note again that guidelines are not straight jackets. A good doc knows when guidelines do not apply. He knows when to go outside of them. He can explain why. You could probably learn to be a doc for 90% of patients in a very short time. We spent most of our time in training learning how to care for that difficult 10% of patients. When John pushes for increased usage of non-MD practitioners, remember that. One of the problems you will see over and over with extenders is that they often do not know what to do with that 10%. The good ones recognize those 10% and know when it is something they can still handle or when to get help.


    • No – Austin is right about this. The critical paragraph is where Goodman writes:

      “Remember the words evidence-based care. They are likely to be very much a part of your future. To its advocates, evidence-based care follows medical treatment guidelines and protocols developed by experts, based on the scientific literature—as reported in medical journals and scholarly reports. If all doctors follow the same protocols, they argue, patients with similar conditions will be treated the same way. Wide variations in the practice of medicine would be replaced by uniform, standardized treatments.”

      But it’s a straw man in itself as he profoundly mis-states what evidence-based medicine is – it isn’t about following the same protocols or standardised treatments. It’s about integrating evidence – including latest evidence – with clinical expertise, for individual patients.

      This excellent article from 1996 in the BMJ explains it well:

      Evidence based medicine: what it is and what it isn’t

      ‘Evidence based medicine is not “cookbook” medicine. Because it requires a bottom up approach that integrates the best external evidence with individual clinical expertise and patients’ choice, it cannot result in slavish, cookbook approaches to individual patient care. External clinical evidence can inform, but can never replace, individual clinical expertise, and it is this expertise that decides whether the external evidence applies to the individual patient at all and, if so, how it should be integrated into a clinical decision.’

      And to Greg:

      ‘…we are entering an era of customized medicine that recognizes every patient is unique based on variations in DNA, comorbidities, health histories, family support resources, and emotional status. ‘

      This is why we badly need consensus and guidelines – because of complexity and the very position by some doctors that you cannot apply them. Take advanced breast cancer – the first international consensus meeting was held last year to meet a pressing need for a set of recommendations for a large patient group who have been denied optimal treatment and care. Yes there are lots of factors – but I can assure you that women are dying sooner and suffering without holistic care because evidence is not being applied.

      See http://www.abc-lisbon.org

      • @marc Brown: “But it’s a straw man in itself as he profoundly mis-states what evidence-based medicine is – it isn’t about following the same protocols or standardised treatments. It’s about integrating evidence – including latest evidence – with clinical expertise, for individual patients.”

        Marc, what you and I might think evidence based medicine is might be completely different from the way our government has used evidence based medicine or whatever they called it in the past. Evidence based medicine is really the flavor of the month. They say you can have your choice of chocolate, strawberry , mango, vanilla etc. ice cream, but no matter what your choice all the ice cream tastes like vanilla.

        • No. Marc nails this pretty well. What I teach med students is to not fear the guidelines. They are a compilation of current best practices, but do not apply to every patient. If you know your pathophsyiology and pharmacology and your patient, you will know when you need to deviate from them, and be able to explain why you did so.

          I also make sure people know the impact of guidelines. My specialty, anesthesiology, saw its malpractice rates drop when others were rising. This was largely due to practice standards (guidelines) developed in our specialty, including standard monitoring. As we progress beyond preventing morbidities and mortality, we find that developing more uniform practices, guidelines, are improving outcomes. We simply dont tolerate a physician who wants to practice some way because he “feels” it is better. They must show actual results. We find that most patients respond best to a common practice regimen. If you have ever managed support personnel or dealt with night time issues in a hospital this will make lots of sense. Again, we always allow for deviation to allow for unusual patients.


          • @Steve “What I teach med students is to not fear the guidelines. They are a compilation of current best practices, but do not apply to every patient.” With emphasis on do not apply to every patient.

            Steve, we have no difference with regard to what you teach your students, but first let me quote something you said a few comments above. “What I teach med students is to not fear the guidelines. They are a compilation of current best practices, but do not apply to every patient. ” That is Excellent.

            There-in lies the problem. Marc is talking about the evidenced-based guidelines which is the new schtick of CMS. Guidelines have been around forever, but have been compiled from bottom up and have not been mandated which negatively affects that 10% you mention and can negatively affect anyone that can benefit from a recent innovation. In fact different groups have different sets of guidelines for the same problem. Close, but not the same. HMO’s have had different guidelines. Many years ago (I don’t know about today) Kaiser’s guidelines for A1C was substantially higher than the ADA, to save money? Would you follow that guideline? Likely not. Things were different at Kaiser as they were financially compelled to follow guidelines.

            Thus, it is my contention that guidelines are very useful, but in the end they are destroyed by the care vs compel argument I discussed with Austin earlier. I think all have to remember that when things head south physicians are sued. If they didn’t follow a reasonable guideline that will be put into evidence and might be the strongest evidence to turn the jury’s head against the doctor being sued. Therefore there is tremendous pressure on physicians to treat appropriately. On the other hand guidelines can blind physicians when they follow them like the gospel. That leaves the 10% in a lot of trouble. Then again how can one have an effective (compelled) guideline to cover multiple major diagnosis with many major drugs? All too frequently they don’t pertain and those back in Washington doing the chart review have no experience to guide them. Dangerous! 

            Government evidence-based guidelines which is why we have the new words (the old words become known for what they are) are devastating to innovation and the appropriate practice of medicine. Government has a place, but it should separate financial concerns from science so that intellectual honesty is maintained. Additionally except for a few exceptions guidelines should not be compelled from the federal level.

            • Al – it’s clear you’re not keen on ‘government’ but government is just another way of saying we need a system of centralising and reviewing evidence – it can’t be left only to professional organisations. And you cannot divorce economics from the process – a key reason that evidence is assembled and reviewed is to determine whether there are more cost-effective ways of doing the same thing, or whether it’s worth doing at all.

            • Marc, I love government, but I love people too and want to see a balance. Government can do certain things well and other things better than without government, but healthcare outside of public health is personal so I would like the individual to control his most personal life along with his privacy. That is not saying that government cannot help its citizens in need. It can and I pointed that out awhile ago. 

              Why do we need a centralized bureaucratic system to sort through scientific data? I thought that is something the scientific community did? Go to a scientific journal and take a look at how many different points of view there are. Then look at how these points of views get organized. The government can provide their input as well. In fact I welcome that input, but only on the same level as all other scientific input.

              You say we can’t divorce economics from the process. Sure we can. We can state the best method and the best cost saving method. We need not combine science with cost except where cost is part of the science. 

              You have just stated the truth about evidenced-based medicine coming from many government agencies that have a stake in lowering cost. ” a key reason that evidence is assembled and reviewed is to determine whether there are more cost-effective ways of doing the same thing” Thus that is NO longer evidence based with regard to the best science. Physicians need to know the science and perhaps the costs so they can adequately inform the patient of the choices that exist.

        • If you aren’t basing medical treatment on scientific evidence, then exactly what are you basing treatment on? If you say clinical experience, then you really are just basing treatment on the accumulated bias of the particular doctor you are seeing. Sure, it may be good. It also may be bad. That is the whole point of EBM (or even better, science based medicine). It’s to weed out the things that appear to work but don’t. Or to attempt to quantify the risks.

          A great example is in emergency medicine. Many doctors, EMT’s and paramedics think backboards and cervical collars for alleged fractures and epi for heart attack are wonderful. Unfortunately the few studies that exist show no or negative evidence. But we’ve always done this they say or “their” patients are always different. Why, exactly?

          • MV, I think a distinction ought to be made between government compelled evidence based medicine that is based upon government need and guidelines that have been around forever.

            • @Al- Not many guidelines are around for a very long time. They change as we learn more. Most guidelines are developed by specialties or by groups. As a result of our internal follow up on outcomes, we have established protocols, guidelines if you will, for some procedures since we found certain techniques have much better outcomes. Those in the group who were doing something different have grumbled about independent thinking, forcing them to conform, blah, blah, blah, but their methods were clearly shown to be inferior.

              Just so you know, there are very few govt imposed guidelines on day to day practice. Those that exist are pretty well accepted. Things like the central line initiative and antibiotics before surgery. (I should caveat this by noting I am not a PCP, so I know less about what they face.) They are developed by specialties, then adopted as a reward or punishment in pay by CMS. JCAHO also looks at these for accreditation. Have they been perfect? Of course not, but when you look at the alternative of no guidelines, they have been wildly successful at improving care. (Off the top of my head, I remember one, JCAHO I believe, mandated target for glucose control post-CABG that was abandoned when it was found to cause problems. It was in effect for less than a year IIRC, so we sorted it out pretty quickly.)


            • Steve, yes, of course guidelines change and the literature changes them as well long before the guideline is updated. That is the beauty of it. Guidelines that can be referred to but are not compelled so that innovation can take place along with change that doesn’t have to wait until the next guideline. As the guidelines become more solid as they do with certain practices in anesthesiology more and more people agree and almost all accept them. Not to do so would open one to a major suit if bad results occurred. Not doing so could also lead to removal from a hospital staff. We are in solid agreement as long as government-C (Compel) is left out of it because government has been known to make too many poor decisions for political reasons.

              We understand that government compelled guidelines are few in practice today (let us keep it that way), but that is one of the problems with the ACA. It is creating government evidence-based guidelines not guidelines. The good guidelines created by government mostly came from private practitioners or perhaps private practitioners adopted something discovered by government researchers. Historically rules, laws etc. follow those that have already been adopted by most of society. When on a daily basis they don’t there is generally a dictatorship.   

            • Al, I’m really struggling to understand what seems obvious to you. Can you write a clear sentence that illustrates when a guideline is bad and when it is good? For example, you wrote,

              We understand that government compelled guidelines are few in practice today (let us keep it that way), but that is one of the problems with the ACA. It is creating government evidence-based guidelines not guidelines. The good guidelines created by government mostly came from private practitioners or perhaps private practitioners adopted something discovered by government researchers. Historically rules, laws etc. follow those that have already been adopted by most of society. When on a daily basis they don’t there is generally a dictatorship.

              Who is “government”? Can you point to some “government guidelines” that did not come from the work of private practitioners? Who are “government researchers”? Can you point to medical research that “government” has done that has (a) informed guidelines and (b) did not involve (even by government funding) non-government (e.g., university) researchers?

              As for the ACA, you say it is creating “evidence-based guidelines not guidelines.” Are you suggesting non-evidence based guidelines are to be preferred?

              I’m not trying to attack or debate you here. I quite honestly cannot understand. Can you help me out?

            • @Al- I can think of govt guidelines that were successful. Take the central line initiative. I was in the middle of implementing that in our network. I fought with nurses and docs who did not want to follow the protocol. They decided it couldnt possibly work, despite the numerous very good papers. After tons of discussion we got it implemented ahead of the required date and other area hospitals. It did not become universally accepted until it was govt mandated. it has been very effective.


            • Austin, I appreciate that you want explanations and you want to firm down my position to make it easier for you to understand me. Remember that goes both ways. I don’t understand you either and in part that is due to you not firming down your position on care vs compel which made the former exercise very difficult to understand

               I’ll try and do my best under the circumstances and give you a simple example rather than some of the earlier guidelines that involved DRG’s and left some patients dead.

              The government created guidelines as to the frequency of cholesterol testing. Why should government determine medical necessity status as far as when to draw a cholesterol level?

              My point is government mixes up evidenced-based guidelines with politics and economics. This a new set of words for old stuff. Guidelines are evidenced based. Is this new term going to make them more evidence based or does it have to do with a few more things that so happen to be unrelated to science?  I have no problem with any guidelines that are voluntary and sometimes I don’t have problems with specific guidelines that are mandated based upon true science and the need for safety. Then again I don’t know why some are promoting mandated scientific guidelines that aren’t proven near 100% instead of the states.

            • Steve, to me this mandate is like wearing gloves or a mask in the operating room. This is not mandating how to treat congestive heart failure. Note how it began in the institutions and grew from there finally being mandated. But it wasn’t mandated nationally through an act of Congress was it?

              Just out of curiosity what does the mandate say and can you give me a site so I can read it on line? I am all for doing things the right way, but I want science divorced from economics. I am not advocating paying for anything one wants. I just want to know what the best science is exclusive of cost. The federal government is really more interested in cost and politics which actually means how the one voting gets reelected.

            • @Al- Very few things are mandated by the federal govt when it comes to practice issues. JCAHO has mandates if you want to maintain accreditation. You need to decide for yourself if it is really a private organization or not. What CMS usually does is a small carrot followed by a small stick. If you have a central line infection, you may not be paid for its treatment unless you can verify that you followed the protocol. Good overview at link.

              Evidence that this has been effective. Note that areas of medicine where this is not mandated or not affected by CMS rules still have much higher rates of infection.


              The Leapfrog group has a more detailed history and goes over problems implementing at the local level. Remember, this data was out there by 2009. In my network, we made it mandatory for all lines in all areas of the hospital, not just the ICU. Read the list. It is not that difficult to comply with. Wash your hands, Cover the patient. Those were the only two new parts that most people were not doing. Not until it became mandatory was it universally adopted. Also, note the role of the usUS Dept of Health.


            • Steve, Medicare is one pretty big mandate. It tells doctors how to bill, defines what is medically necessary (the actual thinking process of an Internist), etc. It virtually compels most primary care physicians to join yet if that same physician and his Medicare patient agree not to bill Medicare (in writing) for his labors and have a private arrangement that is considered illegal. 

    • I certainly agree that we need more science (and research about what works well) not less. I think John would agree as well. That is not the issue here. The issue is how is that research applied. Is it done to inform caregivers and patients about better ways of doing things? Or is it imposed from the outside by payers who are trying to reduce their costs? Most of what we’ve seen in recent years is the latter.

      There is nothing wrong with controlling costs, but it has to be done with the interests of individual patients at the forefront. A committee in Hartford Connecticut or in Washington has no knowledge of or concern with individual patients. I think that is the concern John was trying to address.

      Granted there is another issue about how best to transmit new research findings to practicing clinicians. I think I remember a study that said it takes 15 years for new techniques to be adopted. That is too long, and we need to find ways to speed up that process. It suggests that current methods of continuing education are not adequate. This is a topic that needs exploration.

      • What’s the actual evidence that the individual doctor knows better? I’m sure they believe they do. I want to believe they do. But where’s the evidence? Guidelines exist in part because doctors do things that go against evidence.If you compare systems that have strong guidelines to those that don’t, which comes out on top on average and on the extremes (the patients that are truly not average)? It seems to me that those systems that are held out as highly rated believe highly in EBM.

        And if it takes 15 years for new techniques to be adopted, the only real way to speed that up (and it may not be a good idea) is for coercion of some type. Probably of the type you don’t want.

        • MV, physicians use guidelines all the time. Guidelines can be found in text books, periodicals and even specialty books that deal only in guidelines.

        • MV,

          Once again, I am not saying there should be no research, and I am not saying there should be no guidelines. I am concerned about how they are used. If guidelines are used to help clinicians figure out how best to treat a patient, bravo. But if they are developed by third-party payers and imposed on unwilling providers, they are worse than useless. They create an atmosphere of distrust and hostility that permeates the system.

          Why do you think “the only real way to speed that up … is for coercion of some type?” Is that your remedy in other areas of life? Maybe we should bring back the hickory stick in grammar school?

    • I see this chapter as one of the best examples of Austin’s “giant ball of unnecessary fights” argument. Guidelines, protocols and checklists are not bad things. You don’t need centralised NICE-style government intervention to develop guidelines. Under the HSA financing model professional colleges would continue to develop clinical practice guidelines.

      Why trash guidelines and evidence-based medicine? They are not the problem.