Evidence Based Medicine: Fraud, Double Standards and Ignorance
Ok, as a caveat up front, I am NOT saying there is no credible Evidence Based Medicine (EBM). I am not Anti-Science. I know a few nut-swingers will say this. If you fall into this trap you clearly have poor reading comprehension and your mama dresses you funny.
What I AM, is skeptical of the skeptics, the folks who live and die by what is (or is not) in PubMed. Mainstream medicine got itself into a bit of a pickle a few years ago when holding “complementary and alternative medicine” to some pretty high standards of research validation. I forget the exact number, but 70-80% of what is practiced in your standard hospital or medical office has NO randomized controlled trial (RCT) establishing efficacy. I cannot tell you how many people I’ve run across who will dismiss things like Ancestral Health or gluten free eating due to a paucity of RCT’s , the “gold standard” in medicine… (although even these frequently have huge methodological problems like not keeping people in a metabolic ward for the duration of the study…). These same people think nothing of using methodology which lacks RCT backing (most of standard medical care), yet shoot down new therapy using the same criteria. What they are relying on is, GASP! Observation! That dirty, dirty word to the EBM crowd, yet the very thing which forms a remarkable chunk of the foundation of their practice. I’d call this a double standard but usually that term implies some degree of awareness on the part of the perpetrator. In this situation, there is NO awareness, just an assumption that what makes up current medical practice has been “rigorously studied” when in fact most of medicine has simply been observed to “work” and that was the end of that.
Some of this lack of RCT’s is for ethical reasons. I remember a study looking at survivability of gunshot wounds to the chest in folks who were given aggressive IV fluids. The thinking was that people would do better with a large amount of fluid administration, as an individual with decreasing blood volume easily develops cardiac output problems (and a host of other issues). Interestingly however, somewhat less aggressive therapy seemed to produce better survival rates, possibly due to the ability of the body to better form clots. We don’t know precisely why this is as it’s never been studied and is not likely to be studied for reasons I’ll get to in a moment. The point being, this is NOT a RCT. The study is OBSERVATIONAL. For the EBM crowd this should be jettisoned and largely ignored if they are to be consistent in their treatment of study material. It’d be tough to get a bioethics committee to sign off on a RCT that would really put these questions to rest (we’d need to shoot similar people, in similar ways, apply varying therapy, kill people at various stages of the treatment to track different parameters to verify mechanisms…I guess you could do animal models on this but even that starts getting pretty gnarly). Should this be dismissed as flippantly as Lindeberg’s Paleo vs. Mediterranean diet in Humans trial frequently is amongst the EBM crowd? How about just holding the same standards to all of this material instead of using one’s biases ?
A few days ago I had some back and forth with a nice enough guy on Twitter who looked at this recent Slate piece (More on that later) and declared “This is why I do not advocate Paleo, there is not science to back it up. Show me the science”
Normally I’d simply shoot the guy the Lindeberg paper, a Frasetto paper etc and theye MIGHT be surprised (as these people have never bothered to even look in pubmed for this material) but they will consistently default to the following positions:
-This is a small study, we cannot draw conclusions.
-It is largely observational
And on, and on.
So, I thought about things for a moment and the conversation went something like this:
Me: There are virtually no “good” studies in nutrition. “Good” being defined as a blinded, crossover, metabolic ward trail. If it does not fit this standard it is guess-work and largely worthless.”
Him: “Wait a second! You can’t just throw out all of the observational material that has been generated, there is a lot of value in that!”
Me: “Hey, that is a fantastic point, but if you really buy that then you cannot selectively choose to IGNORE OR DISMISS an observational study just because it does not fit your preconceived notions. Assuming similar study design, ALL these studies have equal merit, right.”
Him: ”Ugh…yea, that makes sense.”
I then I shot him the Lindeberg piece and he commented “Wow, I’ve never read that.” (shocker) Then he followed up with “But, until there are more studies I just can’t follow this stuff.” Am I missing something, but is it not odd that the folks who claim there is “no science” behind any of this stuff, never seem to be up on what has actually been published? If I ask for a critique of the Fresetto work looking at kidney function I just get a blank stare. I mean…if you can come out and strongly proclaim a position on a topic, you might think you have actually READ something on the topic. I know the vegan literature better than the vegans do. The conversation wrapped up with:
Him: “Maybe I should read your book…”
Me: “That’s crazy talk! Reading a book on the topic one is discussing! Send me your mailing address and I’ll send you a book.”
This guy comes from an academic/research background. I’ll give him credit for acknowledging the literature which he has not read, that he has not even read a book on this topic…but I literally had to trap the guy to get this favorable response.
The impetus for this piece has two vectors. One is a recent critical piece on Paleo which was published in Slate the other was yet another in a long list of editorials looking at fraud in science, in this case at one of the premier research institutions, Johns-Hopkins University. Stuff like this started cropping up about 10 years ago, and it seems like it is either happening more often, or we are simply learning about it more frequently. It’s unclear how much fraud occurs in medical research, but it would appear to be “A lot.”
The fraud seems to occur on both sides of the process, the bench researcher “cooking the data”, as well as the review editors who apparently are not doing the job Peer Review is supposed to do (which is provide quality control and maintain standards). That system of peer review is several hundred years old. One can make an agreement that it served a vital role in helping science to crack out of the mysticism of the past, but it is clear this system provides a safe-haven for not only bad research, but also the perpetuation of broken ideas which need to be tossed. Open Source peer review appears to deliver not only integrity, but also is much faster and more accurate. This is a Market based approach to medicine, not dissimilar to the open source approach to many software platforms. This is an important point: Open Source projects tend to work well as the people involved are INTERESTED in the topic at hand. As in: I like to spend time contemplating this topic. They may or may not have financial ties to the topic, but they are not a bored hourly employee nor the director of a lab trying to maintain funding via the publish or perish system. Passion for a topic seems to produce both integrity AND better results. Again, shocker.
This fraud topic creates a rather large bone I have to pick with the EBM “Skeptic” crowd. It’s already annoying when you have folks like the guy above who say there is “no science to back up Ancestral Health”, yet they are completely ignorant and non-conversant on the literature that DOES exist. Add to this a bias which makes shoddy studies supporting the status quo “ok”, yet studies like Lindeberg’s are held in contempt. That’s all annoying, but then when we add the fact we can’t really trust the information coming out of academia. It makes me call “Caca-de-Toro” on much of the EBM scene. It appears we are to trust a system which produces people who dismiss a topic despite complete ignorance of said topic, these people impose double standards and somehow categorize certain material as “acceptable”, dismissing other material (despite equal rigor) and these people “hang their hats” on research that is riddled with fraud.
Again, for the people with reading comprehension issues, I’m not saying science and the scientific method are not valuable, what I AM saying is much of the data being cited is questionable. Double standards are the soup-du-jour, and “Skeptics” are compelled to comment on topics with which they are clearly ignorant. I don’t know about you, but I find all of that “somewhat problematic.”
So, what do we do?
Medicine has done a fantastic job in dealing with infectious disease (antibiotics, vaccinations, public health), and with trauma. Unless your Crown Chakra is so misaligned as to cause brain damage, the strides that have been made in these areas are nothing short of miraculous. Unfortunately however, the “one disease, one cure” efficacy of antibiotics has blinded medicine into thinking a “magic bullet” awaits every disease and condition. Looking at our success with degenerative disease like diabetes, autoimmunity, cancer, neuro-degeneration, and cardiovascular disease paints a bleak picture. We dump massive amounts of resources into these conditions, but with little in the way of return. These “Diseases of Affluence” are complex and largely reflect a gen-environment mismatch that does not lend itself to simple pharmacological solutions. If you disagree with this, please show me how we are treating these issues in fast, cost effective ways. What is missing in medicine is an over-arching theory to both drive hypothesis generation, and allow for better data interpretation. Fortunately, just such a theory exists, and it’s called Evolution via Natural Selection. This is a piece from the Protein Debate we sponsored via the Performance Menu between Prof. T. Colin Campbell (Author of the China Study) and Prof Loren Cordain, discussing the role of protein and human health. This is part of the introduction written by Prof. Cordain:
Although humanity has been interested in diet and health for thousands of years, the organized, scientific study of nutrition has a relatively recent past. For instance, the world’s first scientific journal devoted entirely to diet and nutrition, The Journal of Nutrition only began publication in 1928. Other well known nutrition journals have a more recent history still: The British Journal of Nutrition (1947), The American Journal of Clinical Nutrition (1954), and The European Journal of Clinical Nutrition (1988). The first vitamin was “discovered” in 1912 and the last vitamin (B12) was identified in 1948 (1). The scientific notion that omega 3 fatty acids have beneficial health effects dates back only to the late 1970’s (2), and the characterization of the glycemic index of foods only began in 1981 (3).
Nutritional science is not only a newly established discipline, but it is also a highly fractionated, contentious field with constantly changing viewpoints on both major and minor issues that impact public health. For example, in 1996 a task force of experts from the American Society for Clinical Nutrition (ASCN) and the American Institute of Nutrition (AIN) came out with an official position paper on trans fatty acids stating,
“We cannot conclude that the intake of trans fatty acids is a risk factor for coronary heart disease” (4).
Fast forward 6 short years to 2002 and the National Academy of Sciences, Institute of Medicine’s report on trans fatty acids (5) stating,
“Because there is a positive linear trend between trans fatty acid intake and total and LDL (“bad”) cholesterol concentration, and therefore increased risk of cardiovascular heart disease, the Food and Nutrition Board recommends that trans fatty acid consumption be as low as possible while consuming a nutritionally adequate diet”.
These kinds of complete turnabouts and divergence of opinion regarding diet and health are commonplace in the scientific, governmental and medical communities. The official U.S. governmental recommendations for healthy eating are outlined in the “My Pyramid” program (6) which recently replaced the “Food Pyramid”, both of which have been loudly condemned for nutritional shortcomings by scientists from the Harvard School of Public Health (7). Dietary advice by the American Heart Association (AHA) to reduce the risk of coronary heart disease (CHD) is to limit total fat intake to 30% of total energy, to limit saturated fat to <10% of energy and cholesterol to <300 mg/day while eating at least 2 servings of fish per week (8). Although similar recommendations are proffered in the USDA “My Pyramid”, weekly fish consumption is not recommended because the authors of these guidelines feel there is only “limited” information regarding the role of omega 3 fatty acids in preventing cardiovascular disease (6). Surprisingly, the personnel makeup of both scientific advisory boards is almost identical. At least 30 million Americans have followed Dr. Atkins advice to eat more fat and meat to lose weight (9). In utter contrast, Dean Ornish tells us fat and meat cause cancer, heart disease and obesity, and that we would all would be a lot healthier if we were strict vegetarians (10). Who’s right and who’s wrong? How in the world can anyone make any sense out of this apparent disarray of conflicting facts, opinions and ideas?
In mature and well-developed scientific disciplines there are universal paradigms that guide scientists to fruitful end points as they design their experiments and hypotheses. For instance, in cosmology (the study of the universe) the guiding paradigm is the “Big Bang” concept showing that the universe began with an enormous explosion and has been expanding ever since. In geology, the “Continental Drift” model established that all of the current continents at one time formed a continuous landmass that eventually drifted apart to form the present-day continents. These central concepts are not theories for each discipline, but rather are indisputable facts that serve as orientation points for all other inquiry within each discipline. Scientists do not know everything about the nature of the universe, but it is absolutely unquestionable that it has been and is expanding. This central knowledge then serves as a guiding template that allows scientists to make much more accurate and informed hypotheses about factors yet to be discovered.
The study of human nutrition remains an immature science because it lacks a universally acknowledged unifying paradigm (11). Without an overarching and guiding template, it is not surprising that there is such seeming chaos, disagreement and confusion in the discipline. The renowned Russian geneticist Theodosius Dobzhansky (1900-1975) said, “Nothing in biology makes sense except in the light of evolution” (12). Indeed, nothing in nutrition seems to make sense because most nutritionists have little or no formal training in evolutionary theory, much less human evolution. Nutritionists face the same problem as anyone who is not using an evolutionary model to evaluate biology: fragmented information and no coherent way to interpret the data.
All human nutritional requirements like those of all living organisms are ultimately genetically determined. Most nutritionists are aware of this basic concept; what they have little appreciation for is the process (natural selection) which uniquely shaped our species’ nutritional requirements. By carefully examining the ancient environment under which our genome arose, it is possible to gain insight into our present day nutritional requirements and the range of foods and diets to which we are genetically adapted via natural selection (13-16). This insight can then be employed as a template to organize and make sense out of experimental and epidemiological studies of human biology and nutrition (11).
The Paleo Diet/Ancestral Health concepts were born of observation, an observation that pre-industrial societies show a remarkable absence of the degenerative diseases which plague the developed world. From this observation, various theories have sprung forth (immunologically reactive plant proteins causing Leptin resistance, changes in activity levels altering gene activity, maladaptive sleep patterns, gut dysbiosis and insulin sensitivity, etc.). We are now in the beginning stages of investigating some of these big picture hypotheses, and using the findings to refine our understanding of human health. It is humorous when I hear people dismiss “observational findings”, as these are the seed-crystal of ALL of science. We observe phenomena 1st, endeavor to construct predictive models 2nd.
Modern physics was born a little over a century ago when phenomena were observed that had no model of prediction or explanation. Shallow thinkers with names like Bohr, Einstein, De Broglei, Planck, and Schrodinger hashed out what would become the modern field (no pun intended) of quantum mechanics. It is worthwhile to do a little reading about this time, and the interpersonal conflicts that arose in the physics research community. It was a blood-bath. Personal attack’s, attempts at subterfuge. Substitute “quantum mechanics” for “paleo diet” or “evolutionary medicine” and we can largely transfer these stories of a century ago to our modern equivalent of the emerging science of Darwinian Medicine. The Slate piece I linked to above is a nice illustration, as the author completely dismisses the whole Ancestral Health concept based on the early assumption that our genome has been largely static since the transition from the hunter-gatherer life way to agriculture, and now modern post-industrial living. This original idea was based on this wacky stuff called “The best data we had at the time.” Recent genetic research has shown we’ve seen a remarkable increase in evolutionary pressures as evidenced by a dizzying array of polymorphisms after the advent of agriculture. So, strictly speaking, we are not “hunter gatherers in Pin-Striped suits” but as Stephan Guyenet points out in his series looking at the genetics of the ice mummy, Otzi, the distinction is largely moot from a clinical standpoint. The genetic changes are at best in a middle ground attempting to mitigate the challenges of an agricultural life way, and observationally we still see huge benefit from the Ancestral Health model. Something worth mentioning here is that when we see rapid genetic changes, it is due to a significant selection pressure. A selection pressure is some activity, be it dietary, environmental, or what have you, that causes a large number of the population to die. Or we might have a genetic change that confers a remarkable survival advantage, and this rapidly makes it’s way through a population, or some combination of both. But the implication here is that agriculture was not a benign force to our ancestors, but rather a powerful challenge necessitating rapid evolution to deal with the problem.
So, yes, some early assumptions in the Paleo/Ancestral Health scene ended up not passing the scrutiny of subsequent findings. That’s what science is about, and I’m not sure what field we could look at that the folks who founded it got everything right the first time. Einstein refused to believe we live in a random universe (God does not play dice), and he spent the last decades of his life trying to make the data fit his religious views. Interestingly however, no one calls to invalidate General Relativity due to this gaffe, so again, how about the same standards we apply to other areas of investigation could be applied to the nascent field of Darwinian Medicine?
In closing I’d like to issue a statement and a challenge to the Evidence Based Medicine crowd: If you are conducting research in the biological sciences (medicine is a sub-discipline of biology), and your research does not consider the implications of Evolution via Natural Selection, you are not yet practicing science. You are earnestly data collecting, but you have no overarching framework to know if your questions or interpretations have any merit. If you want to move the field of medicine towards an epistemology with the predictive power of Quantum Mechanics, you must conduct every bit of research, hypothesis generation, and data interpretation with the background music of “what are the evolutionary implications?” This transition will eventually happen; it’s just a question of if it will take a decade or a century.
Happy belated Pi day, and Einstein’s B-day.