From The problem with rapid Covid testing, Mayank Gupta writes:
The absolute number of false positives would rise dramatically under slightly inaccurate, broad surveillance testing. At least initially, the number of people going to the doctor to ask what to do would also rise. One can imagine if doctors truly flubbed and didn’t know how to advise patients accurately, a lot of individual patients would lose trust in the medical system (testing, doctors, or both). The consequence of this would be more resistance to health public policy measures in the future.
For a reminder of why rapid testing is valuable, see Alex Tabarrok. Note also the evidence from the NBA that people who need useful tests can be more innovative than the medical system.
This seems like the tip of an important iceberg.
Our medical system would probably produce more health if doctors were more willing to admit they’re uncertain and/or confused. And there’s nothing obviously good about misleading patients into overestimating how much the medical system knows.
But that would mean doctors get less prestige. Patients likely reinforce doctors’ desire for certainty, by using the appearance of certainty as a proxy for competence.
I expect these attitudes deter a substantial amount of innovation.
Here are some other examples which I think demonstrate the pattern. Keep in mind that there’s almost always some special-interest group promoting the pattern, but that’s only part of the problem. Their ability to do so is fairly dependent on voters/patients rewarding something that’s only loosely connected to health.
Measuring the Spread of COVID-19
There’s been a lot of attention devoted to counts of how many people have tested positive for COVID-19, and drastically less attention to estimates of how many people have been infected. The number of infections is more valuable evidence of how bad the pandemic is. Yet the number of positive tests is measured much more precisely. So we get the perverse situation where incumbent politicians are less popular if more infections get detected.
Masks
Around March or April, most U.S. authorities switched from sounding confident that facemasks, at least as worn by ordinary people, were useless, to being quite confident that widespread use was essential. As far as I can tell, the evidence both before and after this switch should have been summarized as “our best guess is that they work”.
SlateStarCodex provides a moderately good argument that the medical system prefers a dialect that leaves no room for expressing uncertainty: So masks haven’t been proven to work beyond a reasonable doubt. Just like the legal term for “not proven guilty beyond a reasonable doubt” is “not guilty”, the medical communication term for “not proven effective beyond a reasonable doubt” is “not effective”.
Vaccines
A vaccine has been available since early July. It’s safety is adequate to convince some smart, risk averse, medical researchers to use it on themselves.
It sure looks like a vaccine would need to be a lot worse than most existing vaccines to be riskier than having people go unvaccinated.
Yet there’s almost a consensus among influential mainstream commentators that a vaccine shouldn’t be used outside of trials until it’s been proven “safe”.
People talk as if “safe” were a well-understood goal, yet I never seem to run across a clear description. I’m pretty sure that it’s neither being used to mean less harm than the available alternatives, nor zero detectable harm. One thing that it might mean is the lowest risk of bad publicity for the FDA.
E.g. Politico:
The FDA has countered that it previously signaled it would hold Covid-19 vaccines to a higher bar given the stakes of the pandemic and the need to rebuild public confidence that any emergency authorization will be grounded in science.
Commentators often cite anti-vaxxer risks, but seem rather reluctant to provide scientific evidence for claims about how potential anti-vaxxers would react. My limited attempts at understanding anti-vaxxers suggest that a moderately important part of their motivation is the impression that authorities are lying to them about the risks of vaccines.
These commentators seem to imply that vaccines are commonly as risky as the COVID-19 risks that people are currently taking, even though there’s little evidence to support that. How is trust in authorities helped by sounding more alarmist than is justified by the evidence?
I’ll guess the FDA wants to create the impression that it has scientifically determined a definition of “safe” that reflects what’s good for the country, but which is too hard for laymen to understand. The reluctance of FDA, and pro-FDA commentators, to quantify safety suggests that’s an illusion.
Note that I don’t know to what extent vaccines are being delayed by manufacturing constraints compared to FDA approval. I can imagine that nearly all the vaccine delay is due to the speed of manufacturing, so that the FDA approval delays are only costing a few lives. But I can also imagine that the RADVAC vaccine and some others could have been delivered in moderately large volume in August, saving a substantial fraction of the people who are being infected today.
The FDA has decided that COVID-19 vaccines need to be 50% effective. I’m guessing that the FDA has chosen a threshold that many vaccines can meet, so most likely this won’t be a problem. But no sensible person would be 100% certain of that. What would happen if all the vaccines under development turn out to be 40% effective? They’d still save many lives if widely used, but the FDA would have backed itself into a corner. The FDA must have some strong need for looking certain about what criteria to use if they’re willing to take such a risk.
More Vaccines
Speaking of vaccines, until fairly recently, flu shots were restricted to facilities that were inconvenient for many people. Then regulations were relaxed to allow pharmacies to administer some vaccines. This might have increased vaccination rates. I presume that made us slightly healthier. What prevented that from happening earlier? I’m guessing it was FUD about whether pharmacists were as reliable as doctors. Ok, I guess I’m unsure how well this supports my thesis.
Alzheimer’s
Dale Bredesen seems to have figured out how to reverse some (most?) early stage Alzheimer’s. His approach has some difficulties: it’s complicated and inconvenient.
The medical establishment has largely ignored Bredesen’s work. I’ll suggest that that reaction is partly due to some key uncertainties that his protocol would introduce.
E.g. is vitamin D an important part of the protocol? I’d say there’s a fairly good chance that it’s not relevant, and he’s just including it because it’s got a great risk/reward ratio. Including enough treatments with this kind of risk/reward ratio will end up including a few that turn out to be really important (don’t forget that vitamin D also has a great risk/reward ratio for minimizing COVID-19 harm).
It wouldn’t be surprising if doctors were merely as pessimistic about getting patient compliance as they often have been about preventing diabetes or getting patients to quit smoking. But it looks like there’s something more going on.
Bredesen articulated enough of a protocol that an average doctor could replicate maybe 80 or 90% of Bredesen’s case results. But he ran into IRB trouble due to concerns that doctors wouldn’t have clear instructions for what to do in other cases. So if the patient’s LDL-P is high, the doctor is supposed to identify and fix the cause. But a full protocol for that would be a book in itself, which mainstream medicine mostly gives up on (in part because it’s likely to involve trial and error; but there’s also the statin industry, whose income might depend on people not knowing what causes high LDL-P).
So we’re left with a situation where the top Google result for Alzheimer’s says:
While there is no cure for Alzheimer’s disease or a way to stop or slow its progression, there are drug and non-drug options that may help treat symptoms.
I.e. a confident-sounding conclusion that is probably false.
Sleep Apnea
Snoring and sleep apnea can sometimes be treated with a mouthguard-like appliance.
The odd thing is that it’s relatively easy to get if you want it for snoring (not a disease?), but if you want it for apnea (a disease), then you’ll face an additional hassle to get authoritative evidence as to whether you need it:
I mentioned some of my snoring issues to my dentist and asked about this dental device I’d heard of. He first told me that if I need the device for snoring then all I had to do was ask. However, if I felt I needed the device for apnea then it would require a sleep lab/overnight stay result and a full prescription. This is how the medical industry works sometimes and he admitted that even though the device ALSO fixes sleep apnea, he cannot prescribe it for that specifically without those lab results. So I told him I just need to stop the snoring (wink wink) and I was finally on my way to a potential solution.
Conclusion
This pattern might be the result of a universal human desire. Or it might reflect something more temporary, such as a remnant of High Modernism, when technocrats believed society was on the verge of justified confidence in a wider variety of technological answers than was actually the case.