physicians, patients, and intellectual triage

Please, please read this fascinating essay by Maria Bustillos about her daughter’s diagnosis of MS — and how doctors can become blind to some highly promising forms of treatment. The problem? The belief, drilled into doctors and scientists at every stage of their education, that double-blind randomized tests are not just the gold standard for scientific evidence but the only evidence worth consulting. One of the consequences of that belief: that diet-based treatments never get serious considerations, because they can’t be tested blindly. People always know what they’re eating.

See this passage, which refers to Carmen’s doctor as “Dr. F.”:

In any case, the question of absolute “proof” is of no interest to me. We are in no position to wait for absolute anything. We need help now. And incontrovertibly, there is evidence — not proof, but real evidence, published in a score of leading academic journals — that animal fat makes MS patients worse. It is very clearly something to avoid. In my view, which is the view of a highly motivated layperson whose livelihood is, coincidentally, based in doing careful research, there is not the remotest question that impaired lipid metabolism plays a significant role in the progression of MS. Nobody understands exactly how it works, just yet, but if I were a neurologist myself, I would certainly be telling my patients, listen, you! — just in case, now. Please stick to a vegan plus fish diet, given that the cost-benefit ratio is so incredibly lopsided in your favor. There’s no risk to you. The potential benefit is that you stay well.

But Dr. F, who is a scientist, and moreover one charged with looking after people with MS, is advising not only against dieting, but is literally telling someone (Carmen!) who has MS, yes, if you like butter, you should “enjoy” it, even though there is real live evidence that it might permanently harm you, but not proof, you know.

In this way, Dr. F. illustrates exactly what has gone wrong with so much of American medicine, and indeed with American society in general. I know that sounds ridiculous, like hyperbole, but I mean it quite literally. Dr. F. made no attempt to learn about or explain how, if saturated fat is not harmful, Swank, and now Jelinek, could have arrived at their conclusions, though she cannot prove that saturated fat isn’t harmful to someone with MS. The deficiency in Dr. F.’s reasoning is not scientific: it’s more like a rhetorical deficiency, of trading a degraded notion of “proof” for meaning, with potentially catastrophic results. Dr. F. may be a good scientist, but she is a terrible logician.

I might say, rather than “terrible logician,” Dr. F. is someone who is a poor reasoner — who has made herself a poor reasoner by dividing the world into things that are proven and all other things, and then assuming that there’s no way to distinguish among all those “other things.”

You can see how this happens: the field of medicine is moving so quickly, with new papers coming out every day (and being retracted every other day), that Dr. F. is just doing intellectual triage. The firehose of information becomes manageable if you just stick to things that are proven. But as Bustillos says, people like Carmen don’t have that luxury.

What an odd situation. We have never had such powerful medicine; and yet it has never been more necessary for sick people to learn to manage their own treatment.

how not to write a book review, techno-utopian edition

Maria Bustillos’s review of Nick Carr’s new book The Glass Cage is really, really badly done. Let me illustrate with just one example (it’s a corker):

In the case of aviation, the answer is crystal clear, yet Carr somehow manages to draw the opposite conclusion from the one supported by facts. In a panicky chapter describing fatal plane crashes, Carr suggests that pilots have come to rely so much on computers that they are forgetting how to fly. However, he also notes the “sharp and steady decline in accidents and deaths over the decades. In the U.S. and other Western countries, fatal airline crashes have become exceedingly rare.” So yay, right? Somehow, no: Carr claims that “this sunny story carries a dark footnote,” because pilots with rusty flying skills who take over from autopilot “often make mistakes.” But if airline passengers are far safer now than they were 30 years ago — and it’s certain they are — what on Earth can be “dark” about that?

Note that Bustillos is trying so frantically to refute Carr that she can’t even see what he’s actually saying. (Which might not surprise anyone who notes that in the review’s first sentence she refers to Carr as a “scaredy-cat” — yeah, she actually says that — and in its third refers to his “paranoia.”) She wants us to believe that Carr’s point is that automating the piloting of aircraft is just bad: “the opposite conclusion from the one supported by facts.” But if Carr himself is the one who notes that “fatal airline crashes have become exceedingly rare,” and if Carr himself calls the decline in air fatalities a “sunny story,” then he just might not be saying that the automating of flight is simply a wrong decision. Bustillos quotes the relevant passages, but can’t see the plain meaning that’s right in front of her face.

Carr cites several examples of planes that in recent years have crashed when pilots unaccustomed to taking direct control of planes were faced with the failure of their automated systems. Does Bustillos think these events just didn’t happen? If they did happen, then we have an answer to her incredulous question, “If airline passengers are far safer now than they were 30 years ago … what on Earth can be “dark” about that?” That answer is: If you’re one of the thousands of people whose loved ones have died because pilots couldn’t deal with having to fly planes themselves, then what you’ve had to go through is pretty damned dark.

Again, Bustillos quotes Carr accurately: The automation of piloting is a sunny story with a dark footnote. If Carr says anywhere in his book that we would be better off if we ditched our automated systems and went back to manual flying, I haven’t seen it. I’d like for Bustillos to show it to me. But I don’t think she can.

The point Carr is making in that chapter of The Glass Cage is that flight automation shows us that even wonderful technologies that make us safer and healthier come with a cost of some kind — a “dark footnote” at least. Even photographers who rejoice in the fabulous powers of digital photography knows that there were things Cartier-Bresson could do with his Leica and film and darkroom that they struggle to replicate. Very, very few of those photographers will go back to the earlier tools; but thinking about the differences, counting those costs, is a vital intellectual exercise that helps to keep us users of our tools instead of their thoughtless servants. If we don’t take care to think in this way, we’ll have no way of knowing whether the adoption of a new technology gives us a sunny story with no more than a footnote’s worth of darkness — or something far worse.

All Carr is saying, really, is: count the costs. This is counsel Bustillos actively repudiates: “Computers are tools, no different from hammers, blowtorches or bulldozers; history clearly suggests that we will get better at making and using them. With the gifts of intelligence, foresight and sensible leadership, we’ve managed to develop safer factories, more productive agricultural systems and more fuel-efficient cars.” Now I just need her to explain to me how those “gifts of intelligence, foresight and sensible leadership” have also yielded massively armored local police departments and the vast apparatus of a national surveillance state, among other developments.

I suppose “history clearly suggests” that those are either not problems at all or problems that will magically vanish — because if not, then Carr might be correct when he writes, near the end of his book, that “The belief in technology as a benevolent, self-healing, autonomous force is seductive.”

But that’s just what a paranoid scaredy-cat would say, isn’t it?

UPDATE: Evan Selinger has some very useful thoughts — I didn’t see them until after I wrote this post.