Integrative Obfuscation

The marketing of so-called CAM or integrative medicine continues. These terms are just that – marketing. They are otherwise vacuous, even deceptive, and meant only to conceal the naked fact that most medical interventions that hide under the CAM/integrative umbrella lack plausibility or credible evidence that they actually work.

Medicine that works is simply “medicine” – everything else needs marketing.

Last week in the British Medicine Journal (BMJ) Hugh MacPherson, David Peters, and Catherine Zollman wrote a very telling editorial entitled Closing the evidence gap in integrative medicine, which Edzard Ernst has rightly characterized “a masterpiece in obfuscation.”

The essence of the editorial can be boiled down to this – proponents of integrative medicine are disappointed that scientific research has not validated their failed modalities. Therefore they want to weaken the rules of evidence so that they can get the results they desire.

But first, they begin their marketing with the branding of “integrative medicine”, taken from

“medicine that reaffirms the importance of the relationship between practitioner and patient, focuses on the whole person, is informed by evidence, and makes use of all appropriate therapeutic approaches, healthcare professionals and disciplines (conventional and complementary) to achieve optimal health and healing”

This is not integrative medicine, which is rather the attempt at mixing unproven, disproved, and often highly implausible methods into science-based medicine. But proponents can’t go around saying directly that they want to use treatments that probably don’t work, so they brand a nice sounding term with a collection of virtues that actually have nothing to do with what they are selling.

This is obfuscation because a definition such as this should indicate what distinguishes one category from another – what makes integrative medicine different from medicine. The practitioner-patient relationship has been central to medicine since Hippocrates. The “bio-psycho-social” (whole person) approach to medicine is also nothing new – that’s what I was taught in medical school before the term “integrative” was invented. “Informed by evidence?” – I hope so. In fact, that’s a bit weakly stated. And mainstream medicine already makes use of all “appropriate” therapies – the addition of the world “complementary” is just adding another vague term without any operational definition.

So what is the difference between “integrative” medicine and medicine? None that I can detect by that statement, except the suggestion (well-hidden) of using unproven treatments. The brilliance of the marketing, of course, is that by defining your made-up brand of medicine with these virtues, you suggest that your competition lacks them – without ever having to say it.

Let’s get to the meat of the editorial. They write:

Yet when it comes to deciding whether an intervention, and which type of intervention, might be helpful for a particular patient, a worrying gap exists between the perceived potential for using integrative approaches in areas of poorly met clinical need and the availability of supporting evidence derived from good research.

Throughout the editorial the unstated major premise is that the authors know that the “integrative” modalities they want to promote work. They simply need to figure out a way to support what they already know to be true with something that can be marketed as scientific evidence. Nowhere do they give a hint that they are concerned about using science to figure out IF a treatment works.

They also do not give the slightest consideration to the fact that the evidence gap may be the result of the fact that the treatments simply do not work.

In a response to the editorial, David Colquhoun correctly points out their “special pleading” – “If your treatments cannot pass the test, the test must be wrong.” The test they are complaining about is the randomized controlled trial.

They make the usual excuses for why rigorous scientific studies are not adequate to evaluate integrative methods – they are too complex, the studies are artificial and don’t apply to the real world, and they interfere with choice and the therapeutic relationship. This is flawed reasoning on multiple levels.

First, many “integrative” methods are quite amendable to the double-blind randomized trial. Anything that can be taken as a pill – such as herbs or homeopathy – can be compared to an identical-looking placebo. Procedures are more challenging, but researchers have figured out ways to compare acupuncture to sham and even placebo acupuncture, using non-penetrating needles in an opaque sheath.

But on a deeper level they miss the point of scientific research entirely. They complain that the therapeutic relationship is part of the treatment (as if that is different than regular medicine) and therefore it cannot be separated from the treatment itself in clinical trials. But the entire point of scientific studies is to isolate variables as much as possible so that we can come to conclusions about whether or not each variable has a net positive or negative health effect in certain situations and conditions. This is what makes medicine scientific an allows it to move forward – knowing what works and is safe and what isn’t. The art of medicine is then putting it all together in a positive therapeutic interaction.

James May, chairman of HealthWatch, nailed it when he wrote in response:

Effective medicine is best measured with RCTs. Caring is not. ‘Integrative medicine’ therefore risks both damaging how we measure effective medicines (RCTs), as well as reducing caring to measurables. A better term for this might be ‘disintegrative medicine’.

He is saying, as I just did, that the authors miss the point about the purpose of medical research.

In the place of the randomized controlled trial, the authors recommend so-called “pragmatic” studies, comparison research, and case studies. These all have one thing in common – they are not blinded. They therefore do not separate out the variable of the intervention in question from the therapeutic ritual and non-specific effects. These kinds of studies are useful for guiding the application of treatments that have already been shown to work. But they are useless in determining if a treatment works.


Over the last couple of decades there has been a persistent effort to promote unscientific medical modalities with increasingly sophisticated marketing – and to the horror of many, the marketing is working. Proponents have come up with an steady stream of euphemisms for medicine not supported by science – “alternative”, “complementary”, “integrative”, and “functional”. They have marketed them as “natural” and “holistic” and assured the public that if we just researched these methods they would be proven to work. They marketed their modalities to the public, to regulators, and to academia.

Now, after decades of research, after the NCCAM spent 2.5 billion dollars of taxpayer money – the research overwhelmingly shows that these implausible methods don’t work. (It should not have been surprising that treatments that had a very low prior probability turned out not to work  – the improbable, it turns out, is improbable.) For years we were told that we just needed more research and better designed research. Acupuncture studies need better placebo-controlled acupuncture.

Now “integrative” proponents are switching gears (after the research they were asking for did not turn out as they hoped), increasingly arguing that the problem is not their improbable modalities but the research technology itself. They want to switch to the kind of studies that are not even designed to determine if a treatment works or not, and are almost guaranteed to produce the results they want – results they can use to confuse the public and politicians, and to lubricate their passage into the halls of academia.

To this skeptic, this is an old and familiar strategy. When science fails to validate your belief system, blame science.

Posted in: Science and Medicine

Leave a Comment (17) ↓

17 thoughts on “Integrative Obfuscation

  1. Michelle B says:

    Excellent post.

    ‘Integrative medicine’ therefore risks both damaging how we measure effective medicines (RCTs), as well as reducing caring to measurables. A better term for this might be ‘disintegrative medicine’.

    I am so going to use ‘disintegrative medicine’ from now on to describe SCAM treatments.

  2. DevoutCatalyst says:

    Given the daggers that many alt practitioners throw at medicine’s back, how is that aspect ever going to be integrated? You cannot placate the alternative thinker, you cannot have a rational discussion with inerrancy. If I’m in a car wreck, and an acupuncturist is onboard as first responder, hopefully I’ll retain the capacity to kick him in the ass.

  3. Crescendo says:

    Dr. Novella, this blog post reminded me of an earlier piece you had written entitled “Changing the Rules of Evidence.”

    Required reading.

  4. Crescendo says:

    Damn, I didn’t notice that you have already linked to it!

    I can make an omelette for dinner with all this egg on my face.

  5. Dr Benway says:

    Steve, you write so well. Maybe you can help me articulate something I’m struggling with.

    I find myself wanting to say things in defense of the therapeutic relationship that echo CAM rhetoric. But I don’t want to come across as minimizing the need to be science-based.

    I reject the dichotomy of “science” vs “other ways of knowing.” Either one reasons from evidence or authority.

    Where we set the evidential bar depends on context. If I email the Audubon people to report that I saw osprey fishing on the pond today, they’ll likely accept that as evidence. If I tell you guys I saw a ghost, that won’t be good enough. It won’t count for much without independent corroboration or associated physical evidence.

    The evidence of my own sensations in response to various ice cream flavors is of use to me, though it can’t be corroborated by others. Yet it’s possible to demonstrate inter-subjective agreement among many people concerning these feelings. Humans know chocolate from strawberry.

    Likewise, people know a good friend who listens is necessary, else life is unbearable.

    It pains me to say this, but in the realm of helping kids with developmental problems, the effort to be “evidenced based” is presently the cancer that is killing us.

    Seasoned SPED teachers sit at desks for hours each day writing stuff down. They develop IEP novellas filled with micro-skills and tedious micro-goals. “Joey will complete three addition problems within the time alloted with no more than three verbal cues to stay on-task.” They have to write this stuff down because if it’s not on paper, it didn’t happen. Or if it did happen, it doesn’t count as objective data unless it’s on paper.

    If the SPED teacher is at the computer, who’s with the kids? Less experienced TAs who come and go, who feel no particular attachment to individual students and limited responsibility for their progress.

    Why do children learn the capitol of Vermont? Because it’s of burning interest? No. Kids learn most of the boring stuff we want them to learn because there exists a relationship with a caring adult sharing the learning experience with them. No relationship; no learning.

    It’s easy to spot a teacher with good kid-relationship skills verses one who is out of touch. Putting this difference into words is hard. Operationally defining “patience,” “curiosity”, “encouragement,” “listening,” is possible, but much is lost in translation. Like porn, a good relationship is difficult to define but obvious when you see it.

    The third party payors or managers can’t see relationships in action. So they insist on “measurable goals” on paper. Sadly, the tracking of these little clumps of leaves causes us to miss the forest of what matters to kids.

    Psychological development is a slow, gradual evolution punctuated by radical steps forward. An autistic kid in a residential setting for three years makes no apparent progress. He still can’t tolerate a conversation lasting more than ten minutes. Then over a two-week period he makes a friend for the first time. The two friends spend half a day working on a fort outside. No one could have predicted this would happen when it did.

    Blue-Cross/Blue-Shield want weekly progress meetings regarding a 13 year-old boy in a long-term residential program. He’s post TBI due to MVA four months ago (epidural hematoma, induced coma 2-3 weeks, one seizure, about a dozen fractures) He can’t remember stuff and he minimizes his need for help. IQ about 108 now; was in a gifted program couple years ago. But no gross focal deficits.

    Why isn’t he home? Because his parents have no control over him. ADHD, setting fires, smoking pot, truancy, not coming home at night, sexually active, wants to kill his dad. The usual. I told BC/BS nothing will change in a week. Srsly. The lady said, if he’s not making *measurable gains* each week, the funding dies.

    I predict if he goes home now he’ll be back with his deliquent drug abusing buddies. He’ll take stupid risks and get more seriously hurt. This may happen even with a longer period of residential care. But there’s a chance that a more gradual transition home will allow him time to find friends who don’t do crimes. His parents will have back-up so they can practice setting limits with follow-through. He can experience that it’s not so bad letting them be the boss. A life may be saved.

    My argument isn’t science. I can’t cite studies comparing post-TBI deliquent teens in a several-month residential program verses no treatment or outpatient care. Yet I feel I’m *not* talking like a chiropracter or naturopath. I’m not invoking woo. There ought to be wide inter-subjective agreement among normal humans about this stuff.

    My question to y’alls: how to remain true to the ethos of science without pooping all over variables that are difficult to quantify or measure over short periods of time? How to respect the need for third party corroboration while preventing third parties from sucking the life out of the therapeutic relationship? How do we *prove* what we know is true, that the relationship matters more than micro-goals in the long run to emotionally impaired kids?

  6. Mojo says:

    This redefining of “Integrative Medicine” as particular aspects of normal medicinal practice is reminiscent of an approach I’ve seen ID proponents use a couple of times. For example, I’ve recently seen someone define artificial selection as “intelligent design”, and then argue that because artificial selection works there is evidence for ID. It’s a sort of “Humpty Dumpty” approach to argument.

    Ironically, artificial selection was one of the things that Darwin based his argument for natural selection on.

  7. kausikdatta says:

    Excellent post, Dr. Novella.

    Dr. Benway, I wish I had the right answers to all your highly valid questions and concerns. One question in particular struck a chord with me because of a recent experience involving two close family members.

    How to respect the need for third party corroboration while preventing third parties from sucking the life out of the therapeutic relationship?

    How, indeed?

    To my mind, this is a key lacuna which the peddlers of the pseudo-scientific woo called CAM exploit fully to worm their way in. Think of the quote from that Dr. Novella mentioned:

    medicine that reaffirms the importance of the relationship between practitioner and patient…

    Would I be wrong in asserting that this is why CAM has gained a lot of ground over the years?

    Through their lying, deceitful, hand-waving ways and nonsensical blathering, the CAM practitioners (think of your friendly neighborhood homeopath) are quick to establish a rapport with the patient (if only to prey upon the patient’s vulnerabilities, but the patient doesn’t realize that).

    From the patient’s perspective, here is a person who appears to care for the former’s physical complaints and troubles, and who promises to heal by using a system of what the patient vaguely understands as ‘medicine’. This person also throws in a lot of new-agey stuff (that makes the patient feel good about himself/herself), along with a smattering of fear-mongering about conventional medicine and its ‘side effects’ – and the patient is hooked to CAM for life, truly believing that this CAMster has his/her best interests in mind.

    Now, contrast this experience to that of the patient of a proper (and properly trained) physician within a modern health care system. Unless you are seriously ill and taken to the ER by the paramedics, you walk the ER and you wait for hours before anyone even attends to you. The first person who finally gets to you is not interested (and not even feigning an interest) in your complaints, but just does the routine stuff by rote. Or – in a different situation – you call up your doctor’s office, and get an appointment for a fortnight or a month later. In both places, after an interminable wait at the front office or in an exam room, your designated doctor sits with you for five minutes, rushing through the gross physical examinations, not really listening to your complaints or taking a detailed patient history. Before you know it, you have a scrip and out you go to make the co-pay and leave.

    Now tell me: which one do you think the patient would trust more? Which one would the patient be interested in forging a relationship with?

    Even in Dr. Benway’s anecdote, the BC/BS rep did not care enough (or was not competent enough) to understand the depth of the situation involving this particular patient. This is not an isolated incident. It is happening all over this country.

    But can there be a serious overhaul of the entire US healthcare system that it so badly needs?

  8. Excellent questions. You lay out many of the factors that must be balanced – again, putting it all together is the art of medicine. You also cannot have an exact science of one individual patient.

    I think much of the problem comes from CAM/integrative proponents exploiting just this confusion. We need rigorous science to establish which interventions work, which principles are valid, and which are nonsense.

    But we can use common sense, personal knowledge, the therapeutic relationship, and interpersonal skills to apply the science of medicine to individual care.

    We do need better negotiations or regulations to allow us to track treatments and outcomes to justify third party reimbursements (whether private or government) without being buried in paperwork.

    These are real and difficult problems, but are somewhat tangential to the proper role of science in determining what works.

  9. daedalus2u says:

    It is clear that what matters to BC/BS is the funding. That is why “if he’s not making *measurable gains* each week, the funding dies.”

    The “goal” of the one week time horizon is to minimize the chance that a week of time will go by without *measurable gains*. In other words that funding is stopped at the earliest possible moment. As soon as he hits a plateau, the funding will stop.

    A way to address this with the TBI kid might be to add a new diagnosis, one that is only apparent now after there have been all these “measurable gains”.

  10. Dr Benway says:

    You guys are great. Lemme vent a little more, then I gotta do my job.

    After one-week with skateboard v minivan kid, the “goals” that made sense to me were simply questions I’d like answered. So my to-do list falls on the shoulders of the grown-ups, not the kid, at this point.

    1. Let’s have a chat with dad. What’s up with the kid wanting to kill him? When did things go so terribly wrong? Can the usual parent-child empathic bond be restored?

    2. Let’s engage parents and school contacts with the friendship problem. How do we discourage contact with drug-using peers and encourage contact with kids who do their homework?

    3. Whatever “substance abuse” is biologically, this kid’s got it. Yesterday he took a trip to the store. He was given a dollar. He immediately wanted Red Bull. Then it was cigarettes. Then a lighter. He hasn’t had a smoke in months, but he perseverates on getting a nicotine patch. He’s all about taking something to alter or control his feeling states. Many first degree relatives with same issues. Maybe court involvement (mandatory urines) would help. Let’s talk to his public defender for the fire setting charge, which I believe was dropped.

    Well, per BC/BS, the goal must be something the child does. The reviewer is looking at a box on a computer screen, and that’s what the box says. Fine. Let’s ask the kid not to hit people or run away, even though that’s a “so what?” right now.

    Titanic. Deck Chairs. All friggin’ day long.

  11. kausikdatta says:

    Dr. Benway, you have come up with an excellent course of action with your to-do list! :D I don’t seem to remember if you mentioned your specialty – sorry if I missed it. But perhaps this kid needs extensive psychiatric help, if he is not getting that already?

    1. Let’s have a chat with dad. What’s up with the kid wanting to kill him? When did things go so terribly wrong? Can the usual parent-child empathic bond be restored?

    Absolutely. I feel it is important to figure out how the kid got in the first place the negative emotions (pardon my layman’s terms, I am not a psychiatrist/psychologist) he is venting now. If it can be addressed at source, perhaps the parent-child bond can be repaired to the child’s advantage.

    2. Let’s engage parents and school contacts with the friendship problem. How do we discourage contact with drug-using peers and encourage contact with kids who do their homework?

    Once again, absolutely. But perhaps if the parental contact and empathy are enhanced, the child would not seek out drug-using peers as much. The school should probably pitch in big time using positive reinforcement techniques, in order to steer the kid towards behavior that is not harmful for himself or others.

    [Time to insert a favorite pet-peeve of mine here ;) – schools nowadays are way too short on instilling a sense of discipline and identity into their pupils; I feel that the modern promotion of unbridled individuality in children in the absence of a framework (such as personal discipline) – particularly at an early age when the faculties of reason and decision-making are not fully developed – can only lead to a sense of confusion and disenfranchisement in the said children, a sense that they are ill-equipped to cope with. Hence they turn to consciousness-altering substances for an escape. But of course, I may be talking through the back of my neck here: IANAP]

    3. Whatever “substance abuse” is biologically, this kid’s got it… He’s all about taking something to alter or control his feeling states. Many first degree relatives with same issues…

    The question to be answered is why. Why does a 13 year old kid need altered consciousness? What is he missing, what is he seeking?

  12. daedalus2u says:

    Two words, nitric oxide.

  13. Tsuken says:

    Great post – and comments – about an embarrassingly terrible editorial. In. The. BMJ?? Srsly?!?

    The limitations of RCTs are often held up as representing problems – and by implication a reason to not rely (strongly? at all?) on their results. It seems to me however that the limitations do not necessarily equal problems.

    We are all aware that there are major differences between RCT subjects and real world patients. For a psychiatric example, 40-60% of our patients have substance abuse/dependence problems (in our demographic here anyway), and the rates of other psychiatric comorbidity are high. They are also generally the illest of the ill. Trial subjects however are well enough to have capacity for informed consent, have no substance abuse, and are diagnostically “clean”. The anti-RCT position would be to point out this disconnect and consequently ignore or minimise the RCT results.

    However, what would be the effect of this difference in samples? Unsurprisingly, the treatment that worked in the RCT, doesn’t work in the real world – or at least not so well. However that’s exactly the converse of what the CAM/disintegrative/functional/oogie-boogie proponents are lamenting: that their treatment which “works so well” in the real world, doesn’t stack up in RCTs (and of course RCTs are fatally flawed whereas clinical experience and the beliefs of patients are infallible). But … the RCT demonstrates a treatment to be efficacious. It might not prove especially effective in the real world, and yes, that’s a limitation of RCTs. However, it seems to my concrete (medical-industrial-complex-tool) mind that if a treatment cannot be demonstrated in an RCT to be efficacious, it simply cannot be effective in the real world – where things are harder.

    As I pointed out to my medical student yesterday, just because a patient gets better while taking/using a particular treatment, does not mean the treatment works. As an example, most antidepressant medications have an NNT of around 4. Say I give 4 people an antidepressant, and get lucky and all of them get better. Only 1 actually did so because of the pill. The others responded to perhaps their social situation improving, or support from my caring team of nurses, or the fact that I’m a nice bloke who they trusted and told them they’d be better soon, or that they would have got better anyway.

    Unfortunately we always see the confusions of correlation with causation, and anecdote with data, that the SBM authors do such a good job of pointing out and arguing against. Trouble is, it’s hard to argue/explain that in the real world – even to other doctors sometimes. I’m about to try next Friday ;-)

  14. Wicked Lad says:

    That’s a compelling and thought-provoking post, Steve. I’m a lay person–a lay person among lay people–and your post raised a question I’ve wondered about from time to time: How are surgical techniques tested? I can’t imagine rigorous RCTs are ethical for surgery.

    Assuming we don’t use RCTs, how does that affect the field? Is progress hindered? Do surgeons face more uncertainty over the efficacy of their treatments than, say, internists?

    And does this teach us anything about how to regard SCAM treatments? Or are surgical techniques so much more plausible than most SCAM treatments that different standards of evidence are appropriate?

  15. bruabra says:

    Quick questions for Dr. Tsuken, regarding his “pointing out” to his student:

    “just because a patient gets better while taking/using a particular treatment, does not mean the treatment works. As an example, most antidepressant medications have an NNT of around 4. Say I give 4 people an antidepressant, and get lucky and all of them get better. Only 1 actually did so because of the pill.”

    Are you still giving depressed people antidepressants? Because from your reasoning they’re just as good as acupuncture (and please note that both interfere with neurotransmitters modulation).

    “Bias is like the butt*ole, everyone have their own” Dr. Abramoff

  16. With procedures that cannot ethically be blinded, like surgery, you do have to do different kinds of studies than with drugs where you can do a placebo control. But that does not mean you cannot have scientific rigor.

    You can study outcomes with surgical intervention vs best medical care, but because such studies are not blinded outcomes have to be as objective as possible. You cannot study whether or not the patient feels better, but you can study mortality and survival rates, for example. Or you can use quantitative functional scales.

    For many surgical procedures the question is fairly concrete. Removing an inflamed appendix, for example, is about saving the life of the patient.

    I would not, however, justify a surgical procedure without a plausible mechanism based upon unblinded subjective outcomes.

Comments are closed.