You are left being abused as a science denier, quixotic, delusional etc. All the ad hominems which are the ready tools of those unable to make an evidence-based reasoned argument.
It is unpleasant and every now and then you falter. You wonder whether others are seeing something you are missing even though no one can articulate what that something is.
Anthropogenic global warming is an example. Our understanding of the multiple complex systems which constitute climate are shaky. Our comprehension of the complex systems' interactions is vestigial. Our data sets are patchy at best. Our longitudinal data is overly dependent on loose proxies. Our models are assumption-dependent and faulty. Our incentive structures are skewed. We have seen patterns of evidence tampering.
No one is denying that climate changes. No one is denying that there are periods of global warming. No one is denying that some portion of the current post-ice age warming trend might be influenced by pollution (CO2, grit/particulates, methane, or other). What empirical rationalists question is how much might that contribution be (including zero) and more critically, whether any of the proposed policy solutions have any chance of success and whether they are worthwhile given the alternatives.
It is a lonely position to take when every status seeking academic, every knuckle dragging politico, and every chirpy pontificator ascribes the long two week heat wave or recent storm or a heavy rain to "climate change." But empiricism and rationalism dictate a caution unshared by the glib.
You have self-doubts. But then, occasionally, you get affirmation from your position on other debates.
Cholesterol causes heart disease and death by heart attack has been, like AGW, a longstanding claim in academia and public policy despite the evidence being weak and inconsistent. Like AGW, biology and population health are complex systems. There are multiple complex systems, loosely coupled, with undiscovered feedback loops, tipping points, etc. As much as we have studied them, our knowledge is similarly vestigial, data sets patchy, longitudinal data infrequent and dependent on loose proxies, incentive structures misaligned, etc. And similarly, we are prone to levels of conviction far outstripping empirical evidence and reason.
Lowering cholesterol levels in general and "bad" cholesterol levels in particular has been long-standing US public health policy for decades. Based on these public health policies, statins are a $20 billion industry and the related health services are probably another $80 billion. So, for a $100 billion dollar a year business sector spawned by public policy, what is it good for? Like war, Absolutely Nothing!
Double click to enlarge.
Well, more or less, nothing. You can't spend that kind of money without someone benefiting. Just not the intended recipients.
Looking at the studies in the late seventies and early eighties, I could not see the necessary rigor to support the recommendations which were emerging. I am no deeply trained biologist but am sufficiently trained in statistics to see issues and in my academic field of study, economics (the study of incentives) I could clearly see some dangerously misaligned incentive structures.
From the early eighties on, every visit to the doctors office has entailed at least a passing reference to "keep an eye on cholesterol" and "perhaps you might want to reduce these foods from your diet" even though all my cholesterol levels were normal. By the nineties and 2000s it was not uncommon to hear some chatter along prophylactic lines "Perhaps you might want to consider some statins as a precautionary approach."
In the mid- or late eighties, the first long term observational study came in on the effectiveness of cholesterol control (or at least the first that I recall.) No difference in mortality between those controlling their cholesterol levels and those not. Statin users died at the same rate as everyone else. True, their death from cardiac arrest went down, but their death from other causes went up an equal amount. Statins changed your cause of death but not your rate of death.
From that point forward, while waiting further evidence, I began pushing back on the bad cholesterol orthodoxy. Then in the early 2000s a second large population longitudinal study came in replicating the findings of the first. And still you would hear social chatter, media chatter, and even medical chatter about the need to lower cholesterol levels.
As recently as a couple of years ago I had a debate in a doctor's office with a nurse practitioner who was pushing the advisability of a regimen of statins as a precautionary approach since I am now in the heart-attack prone zone of older males (and remembering that I do not have elevated cholesterol.)
So basically for thirty some years, my empirical and rational approach has been at odds, based on the evidence, with the whole of academia, the medical establishment and public policy. Even voicing doubt about the advisability of these policies has been tantamount to proclaiming oneself as a science-denier, a crank and a conspiracy theorist. Someone who is likely to believe that fluoridating water is a communist plot. Such is the fate of the empiricist and the rationalist in times of religious conviction in public policy.
The tide seems to have begun to turn in the past decade. More and more people are willing to say what the evidence indicates. Cholesterol is not the enemy. Possibly a factor under some circumstances for particular individuals but not a universal explanation. From New Research Confirms We Got Cholesterol All Wrong by Baylen Linnekin.
A comprehensive new study on cholesterol, based on results from more than a million patients, could help upend decades of government advice about diet, nutrition, health, prevention, and medication. Just don't hold your breath.But look at those media outlets. A major study but being reported in the Daily Mail and The Express? High class data reported through low prestige channels.
The study, published in the Expert Review of Clinical Pharmacology, centers on statins, a class of drugs used to lower levels of LDL-C, the so-called "bad" cholesterol, in the human body. According to the study, statins are pointless for most people.
"No evidence exists to prove that having high levels of bad cholesterol causes heart disease, leading physicians have claimed" in the study, reports the Daily Mail. The Express likewise says the new study finds "no evidence that high levels of 'bad' cholesterol cause heart disease."
The study also reports that "heart attack patients were shown to have lower than normal cholesterol levels of LDL-C" and that older people with higher levels of bad cholesterol tend to live longer than those with lower levels.
This is probably news to many in government. But it's not news to everyone.
"In fact researchers have known for decades from nutrition studies that LDL-C is not strongly correlated with cardiac risk," says Nina Teicholz, an investigative journalist and author of The New York Times bestseller The Big Fat Surprise (along with a great recent Wall St. Journal op-ed highlighting ongoing flaws in federal dietary advice). In an email to me this week, she pointed out that "physicians continue focusing on LDL-C in part because they have drugs to lower it. Doctors are driven by incentives to prescribe pills for nutrition-related diseases rather than better nutrition—a far healthier and more natural approach."
Biology, nutrition and population-level heath are complex systems which we understand to some small degree but our knowledge is fragmentary, our data sets puny (though large compared to some other fields), our incentive structures are wrong, etc.
Is coffee good for you? Wine? Fats? Carbs? There is no easy and accurate answer. It depends on the person, their circumstances and conditions, and their objectives. We know enough to occasionally make some careful and limited recommendations. But we do not know near enough to make the high-conviction proclamations to which we are accustomed and to the resulting public policies which financially benefit small populations of stakeholders while causing harm to large populations of putative beneficiaries.
We are not completely ignorant and not all academics and government and business people are selfish and evil. But we should, given the history of strong recommendations based on vestigial understanding which are, decades later, rescinded, be cautious about high-conviction proclamations about high complexity systems.
And we should stay strong in cautiously skeptical empiricism and rationalism, no matter how long it takes for the establishment to come around.
Posts
No comments:
Post a Comment