Signal Extraction Problems in Public Health

Among the many criticisms Nocera and McLean of the government’s response to Covid-19, outlined in their recent book The Big Fail, is that public health officials weren’t exactly on the level with the American public. Health officials presented their claims with far more certainty than the evidence justified. Going back to why the “pandemic playbook” that had so many years of effort put into its creation turned out to be all but useless, Nocera and McLean note:

One significant problem for the public health officials trying to combat COVID-19 was that it was a different kind of virus. All of the government’s planning had been done in anticipation of an influenza pandemic. There was much less known, especially in the early months, about how the coronavirus spread and how deadly it was. And Fauci was never willing to acknowledge that uncertainty.

I don’t think it’s quite true to say Anthony Fauci was never willing to acknowledge uncertainty. He often did – but usually only after the fact, when explaining why he had changed his position on something. That’s fine in and of itself – we want public officials to change their minds after learning new information. But what Fauci was absolutely terrible at doing was acknowledging any degree of uncertainty at any given present moment, and especially allowing that a high degree of uncertainty about present conditions meant there was wide scope for reasonable disagreement about current policy. Whenever there was pushback to a pronouncement of his, the prospect that such pushback might be justifiably grounded in uncertainty disappeared, and those who disagreed with his take were described by him as motivated by “an antiscience bias that people are – for reasons that are inconceivable and not understandable – they just don’t believe science and don’t believe authority.”

Take the issue of masking. In the early days of the coronavirus pandemic, many public health officials confidently made pronouncements to the public that there was no benefit to wearing masks to preventing transmission of the disease. It later emerged that they did think masks were beneficial for preventing the spread of the disease, but they were worried there wouldn’t be enough masks for medical personnel if they openly acknowledged that. So, to prevent people from rushing out to stock up from the limited supply of masks and ensure more were available for hospital staff, they deliberately downplayed the value of masks, at least at first.

Then came a change of guidance from above:

In April 2020, the CDC, in reversing its original position that no one needed a mask, said that Americans should immediately start wearing masks, including cloth masks. Overnight, tens of millions of Americans began wearing them, and hundreds of entrepreneurs began churning them out, many pivoting from their normal business, which had been shut down. But as more was learned about the coronavirus, some experts began to realizer that cloth masks didn’t do much good. “We have known for many months that COVID-19 is airborne and therefore a simple cloth mask is not going to cut it,” said Lena Wen, a public health professor at George Washington University. Finally – finally – in early 2022, nearly two years into the pandemic, the CDC acknowledged that “loosely woven cloth produces offered the least protection” from the virus. That kind of grudging change didn’t inspire confidence.

Additionally, at various points Dr. Fauci seemed to acknowledge that his public-facing statements weren’t meant to reflect what he actually believed to be true, but instead were tuned to be what he thought would most effectively get people to do what he wanted, even if what he said wasn’t true or justified by the evidence. As Fauci said about his approach to vaccination rates:

When polls said only about half of all Americans would take a vaccine, I was saying herd immunity would take 70 to 75 percent … Then, when newer surveys said 60 percent or more would take it, I thought, “I can nudge this up a bit,” so I went to 80, 85.

(More than anything, this reminds me of the Bart Simpson’s famous line “I only lied because it was the easiest way to get what I wanted!”)

Perhaps in these and many other such cases, public health officials felt their best move was to put forth a sort of Noble Lie approach – “We don’t really know for sure if masks and lockdowns will be beneficial on net. They might be, and we think it’s worth trying, but we’re not sure. But if we acknowledge that uncertainty publicly, that might undermine people’s willingness to accept either of them, so we should act as if things are settled and that anyone who disagrees is simply refusing to ‘follow the science.'”

But this creates something akin to what economists call a signal extraction problem. As soon as public health officials start making their pronouncements based on more than what they know to be true, or based on things they actually know are untrue but aimed to get people to move in the “right” direction, they have contaminated the signal such messages are supposed to send. From that point on, any time public health advice is given, people will wonder “Okay, are they saying this because they actually believe it’s true, based on good reasons? Or are they overplaying their certainty because they think that’s the best way to get people to do what they want? Or are they saying something they don’t actually believe, because they are worried an honest answer won’t get the response they want?”

It’s common for people to point out that a serious problem in America is that people are losing trust in institutions. But it’s worth asking if this loss of trust is because Americans are failing to show institutions the proper trust they have earned, or because the players running those institutions have been failing to act in a trustworthy way?

And trust in institutions can make a huge difference. As Nocera and McLean describe in their book:

As a point of comparison, [Dr. Jay] Bhattachary liked to use the example of Sweden. Sweden was controversial because it eschewed lockdowns and kept its society running. But when vaccines became available, “Sweden got 97 percent of adults to take the vaccine without any mandates,” Bhattachary said. “Why? Because people trusted the government. And the reason they trusted the government was that officials were honest with what they knew and what they didn’t know. And they didn’t force people to do things that were outside their capacity to manage.”

This decline in trust in institutions in America has been going on for a long time. As George Will put it in his book The Conservative Sensibility:

In 1964, 76 percent of Americans trusted government to do the right thing “just about always or most of the time.” Today, fewer than 20 percent do. The former number is one reason [President Lyndon] Johnson did so much; the latter is one consequence of his doing so.

I am inclined to think the loss of trust in institutions is one of the biggest problems facing American society right now. But I also believe that a good deal of that trust has been lost because the elites in institutions believe and act as though the masses have been failing them – ordinary people aren’t showing the elites the kind of deference and respect the elites believe they deserve.

(5 COMMENTS)