Pandemic Ethics and Status Quo Risk (forthcoming in PHE)

My latest paper, 'Pandemic Ethics and Status Quo Risk', has just been accepted for publication in Public Health Ethics. Here's the abstract:

Conservative assumptions in medical ethics risk immense harms during a pandemic. Public health institutions and public discourse alike have repeatedly privileged inaction over aggressive medical interventions to address the pandemic, perversely increasing population-wide risks while claiming to be guided by “caution”. This puzzling disconnect between rhetoric and reality is suggestive of an underlying philosophical confusion. In this paper, I argue that we have been misled by status quo bias—exaggerating the moral significance of the risks inherent in medical interventions, while systematically neglecting the (objectively greater) risks inherent in the status quo prospect of an out-of-control pandemic. By coming to appreciate the possibility and significance of status quo risk, we will be better prepared to respond appropriately when the next pandemic strikes

The central idea is that heuristics of ambiguity-aversion and favouring inaction over (potentially risky) action can be expected to backfire terribly in circumstances -- such as a pandemic -- in which "business as usual" is leading us towards disaster.  Instead, I suggest that our policy and institutional responses to such emergency circumstances need to be rebalanced towards (i) liberalizing access to experimental treatments and vaccines, and (ii) requiring an explicit cost-benefit analysis to justify any sort of vaccine obstructionism (e.g. failure to immediately grant Emergency Use Authorization to any credible candidate vaccine early in the pandemic, and of course any post-authorization suspensions).

Other key points of the paper:

The Indefensibility of Post-Vaccine Lockdowns

Reasonable people may disagree about the justifiability of early-pandemic lockdowns (while awaiting the availability of vaccines), but this is just nuts:

Austrian officials’ decision to impose a lockdown that will last at least 10 days and as many as 20 came after months of struggling attempts to halt the contagion through widespread testing and partial restrictions. Starting Monday, public life in the country is to come to a halt, with people allowed to leave their homes only to go to work or to procure groceries or medicines.

What's the justification for this?  When vaccines are freely available to all, Covid isn't a serious threat except to those who refuse the vaccine, and thereby accept personal responsibility for the consequences. If policymakers are worried about hospital over-crowding, unvaccinated adults suffering complications from Covid should go to the back of the line.  If the unvaccinated are not willing to accept the risk of death due to a lack of hospital beds, they can either (i) get vaccinated, or (ii) stay home or take other precautions while local case rates are high.  But if they insist on risking their health, and get seriously ill as a result, they've no-one to blame but themselves.  It's simply not reasonable to infringe upon everyone's liberties for fear of harms that individuals have it within their own power to mitigate or avoid.

JCVI endorses Status Quo Bias

The UK's Joint Committee on Vaccination and Immunisation recently recommended against vaccinating children under 16 against Covid, despite granting that "the benefits from vaccination are marginally greater than the potential known harms." (Of course, aggregated over a subpopulation of millions, even "marginal" improvements in risk profile can result in several saved lives and scores or hundreds fewer hospitalizations.  And, as Deepti Gurdasani makes clear in this thread,* all the evidence should lead us to expect the "unknown" risks from Covid to outweigh those from the vaccine, so taking uncertainty into account should lead us to regard vaccination as all the more important.)

So what's behind the JCVI's verdict?  They are at least admirably transparent:

In providing its advice, JCVI also recognises that in relation to childhood immunisation programmes, the UK public places a higher relative value on safety compared to benefits.

It's important to be clear on what this really means. Note that this is not invoking any kind of philosophically defensible harm/benefit asymmetry.  (Many people think it's more important to reduce suffering than to promote happiness, but that's not what this is about.)  Vaccines aren't to make you happy. The "benefits" they provide are specifically safety benefits, i.e. against other health risks.  So what the JCVI is really saying is that they place higher value on protecting people from potential harms from vaccines than on protecting people from potential harms from COVID.

That is deeply messed up.

I just hope that greater philosophical clarity here will help people to see how messed up it is (and so change these institutions' values in future).  Every time some dopey bureaucrat claims they're prioritizing "vaccine safety" over "benefits", they need to be met with the response: No, you're prioritizing safety from vaccines over safety from COVID.

That's clearly indefensible.  We just need to make it clear that this is in fact what they are doing.  Don't let them obscure the reality of status quo risks behind a weasel-word like "benefits".  The choice isn't between "safety vs benefits", it's "safety [against lesser vaccine risks] vs safety [against greater covid risks]".

* = Thanks to Dan Fogal for the pointer.

The Ethics of "Off-Label" Vaccinations for Kids

The WSJ reports that some parents hope to get their kids (under 12) vaccinated against Covid, as "the FDA’s approval generally means vaccines are eligible for off-label use, meaning beyond approved populations."  However, "the FDA and Centers for Disease Control and Prevention have emphasized that the safest thing for this group of children is to wait for more data to be analyzed."

I'm curious whether it's really true that waiting is "safest", or whether these advisories ignore status quo risk. I don't have the empirical knowledge to answer what really would be safest here, but questions worth asking include:

Pandemic Paralysis

I'm continually appalled by how easily people move from "the consequences of doing X are uncertain" to "X must be banned!", even when there would seem every reason to expect that X actually has high expected value and ought to be encouraged, if anything.

The latest example is X = J&J booster shots for the immunocompromised (or, indeed, everyone):

The weasel words there ‘isn’t enough data to determine’ indicate a typical failure to think in Bayesian terms and use all the information available and a typical failure to think in terms of patient welfare and expected cost and benefits.

Notice also the illiberal default. Instead of saying ‘we don’t have data on the J&J vaccine and the immunocompromised so we are not at this time recommending or not recommending boosters but leaving this decision in the hands of patients and their physicians’ they say ‘we don’t have data and so we are forbidding patients and their physicians from making a decision using their own judgment.’

It's an ongoing problem in our pandemic response that prudent precautionary measures like this get blocked, without any positive medical justification, because the Powers That Be don't follow decision theory and instead insist that we all must sit passively on the tracks, ignoring the oncoming train, while they order routine safety checks on the service ladder.

Apologies for sounding like a broken record on this, but this problem of pandemic paralysis is (i) really serious, and (ii) insufficiently appreciated.  So it's worth repeating again: mere uncertainty is not sufficient reason to reject a pandemic policy proposal.  Critics need to offer reasons for thinking that the potential downsides outweigh the potential upsides (in expectation).

For an extended version of the argument (with many real-world examples), see my paper: 'Pandemic Ethics and Status Quo Risk'.  It argues at length for the banal -- and yet almost universally rejected or neglected -- claims that (1) it would be a serious moral error to focus on a single source or type of risk (neglecting greater overall risks) in evaluating pandemic policy; and (2) we should be open to the possibility that higher-variance options may best reduce overall risk.

'Risky Research' Redux

I'm looking forward to participating in 1DaySooner's Zoom panel discussion on 'What is the Upper Limit of Risk in Clinical Trials?' next week (May 4th, @6pm ET) -- you can register here if you're interested in attending.

My basic view is that there is no absolute upper limit: given informed consent, the risk just needs to be proportionate, i.e. outweighed by the social value of the information gained from the research.

Indeed, this strikes me as entirely straightforward.  There are two key values that public policy should be guided by: beneficence (promoting the overall good) and autonomy (respecting individuals' choices about their own lives).  Conflicts between the two values can be morally tricky.  But if both of these values point in the same direction, as they do in the case of valuable research involving willing volunteers, then it really should be a no-brainer.  There's just no good reason to engage in anti-beneficent paternalism.  So: let's please stop doing that!

I think that's the simplest case for "risky research".  In my paper with Peter Singer, we additionally proposed a principle of risk parity according to which, "if it is permissible to expose some members of society (e.g. health workers or the economically vulnerable) to a certain level of ex ante risk in order to minimize overall harm from the virus, then it is permissible to expose fully informed volunteers to a comparable level of risk in the context of promising research into the virus."  Again, it just makes no sense to block willing volunteers from taking on some level of risk if such obstruction effectively condemns a far greater number of unwilling people to even greater harms.

What principled value here could outweigh the combined force of autonomy and beneficence?  I look forward to hearing what my fellow panelists have to say...

Follow Decision Theory!

Back in January, I wrote that there's no such thing as "following the science" -- that scientists and medical experts "aren't experts in ethical or rational decision-making. Their expertise merely concerns the descriptive facts, providing the essential inputs to rational decision-making, but not what to do with those inputs."

It's worth additionally emphasizing that this question of how to convert information into rational decisions is not something about which academic experts are entirely at sea. On the contrary, there's a well-developed academic subfield of decision theory which tells us how to balance considerations of risk and reward in a rational manner.  The key concept here is expected value, which involves multiplying (the value of) each possible outcome by its probability.  For example, we know that (all else equal) we should not accept a 50% chance of causing 10 extra deaths for the sake of a 1% chance of averting 100 deaths, for the latter's expected value (one death averted) does not outweigh the former's expected cost (5 extra deaths).

Imagining an Alternative Pandemic Response

I received my first shot of the Moderna vaccine yesterday -- which naturally got me thinking about how this should've been accessible much, much sooner.  I don't think anyone's particularly happy about the way that our pandemic response played out, but there's probably a fair bit of variation in what people think should've been done differently.  What alternative history of COVID-19 do you wistfully yearn after?  Here's mine (imagining that these lessons were taken on board from the start)...

Appeasing Anti-Vaxxers

As the ongoing pandemic obviously causes immense harms, there are correspondingly immense benefits to vaccinating people sooner. Our actual policies have failed at this in a number of ways (from failing to encourage experimental vaccination, to gratuitous delays in approving successful vaccines even after the trial data were received). Now some countries are suspending use of the AZ vaccine due to (poorly-grounded) fears about rare side-effects, seemingly oblivious to the fact that there's a much more serious (and high-probability) "side-effect" to non-vaccination, namely, COVID-19.   This all seems bad enough, on straightforwardly utilitarian grounds.  But I now want to argue that it's even worse than that: even if these delays did some good, by reassuring the vaccine-fearful, they would still be wrong.

Against Anti-Beneficent Paternalism

In a previous post, I argued that "undue inducement" worries are typically deeply misguided, and that banning good compensation is contrary to the interests of the very people that it's intended to help.  In this post, I want to raise a different objection: that even if allowing and/or incentivizing beneficent actions (such as kidney donation, or challenge trial participation) would "induce" some people to perform beneficent acts that they might later regret (or that they wouldn't have agreed to if thinking more clearly), it may nonetheless be the case that banning this would be morally even worse.

There's No Such Thing as "Following the Science"

Ezra Klein quotes a Harvard epidemiologist's criticism of the FDA for blocking rapid at-home Covid tests: "They are inadvertently killing people by not following the science."

I agree with the spirit of the criticism (and was heartened to read that Biden’s surgeon general nominee agrees that the FDA has been "too conservative"), but it's worth clarifying that the FDA's failure here is fundamentally ethical, not scientific.

The Risk of Excessive Conservatism

In 'Lessons from the Pandemic', I summarized what I took to be some of the biggest mistakes of the pandemic response, and tried to give a sense of the scale of the potential damage done, along with some concrete suggestions for how we might have done vastly better.  Some readers (e.g. here) seemed of the opinion that only those with "authority" should express such opinions, which I obviously disagree with.  But to better help such readers, it might be helpful to bracket any particular empirical details or examples and focus instead on the most general overarching claim of my post: that excessive conservatism risks immense harm in a pandemic.

One doesn't need a medical degree to see that this more modest (yet still important) claim is true.  For it does not require us to establish that some unconventional pandemic policy truly would be much better; it suffices to note that an unconventional pandemic policy easily could be much better -- i.e., there's a non-trivial probability of this -- and since excessive conservatism would dismiss such unconventional proposals out of hand, such conservatism poses a significant risk of immense harm.  Since it is worth guarding against significant risks of immense harm, it is worth guarding against excessive conservatism in a pandemic. 

To turn this into a more pointed critique of the medical/policy establishment (and elite public opinion), we can simply observe that there is no evidence that said establishment (or elite opinion) is suitably aware of this risk, or that they have taken suitable steps to guard against excessive conservatism.  Quite the opposite, I think: the publicly-available evidence (including establishment pronouncements, presented justifications, and policy decisions) all give off the strong appearance of extreme conservatism.  And of course we have background knowledge that institutions tend to be conservative, and that conventional medical ethics in particular is extremely conservative -- as evidenced by opposition to utilitarian policies like kidney markets.  So we should all go into a pandemic with a moderately high prior for the view that mainstream institutions and opinion are likely to be excessively conservative.  What I'm trying to do is to draw attention to how grave a risk this is, especially in a pandemic.

Is it conceivable that their conservatism is all for the best in the end?  Sure, it's possible.  Is that any kind of problem for my critique?  No, of course not.  We should press for clear cost-benefit analysis from pandemic policy-makers even if it turns out that such improved epistemic procedures would -- remarkably -- yield the same substantive policies at the end of the day.  For there is at least a non-trivial chance that we could do much, much better.

Lessons from the Pandemic

It's generally recognized that our (American) response to the Covid-19 pandemic was disastrous. But I think far fewer appreciate the full scale of the disaster, or the most significant causal levers by which the worst effects could have been avoided.  (Yes, Trump was bad.  But his public health disinformation and politicization of masking -- while obviously bad -- may prove relatively trivial compared to the mammoth failings of our public health institutions and medical establishment.)  Much of the pandemic's harm could have been mitigated had our institutions been properly guided by the most basic norms of cost-benefit analysis.  Consider:

Epistemic Calibration Bias and Blame-Aversion

People typically treat having an importantly false belief as much more problematic than failing to have an importantly true belief.  They're more concerned about being over-confident than being under-confident in their credences.  But why?  Is such an epistemic asymmetry warranted?

I'm dubious.  The ideal is to be epistemically well-calibrated: to have just the degree of confidence in an important proposition that is warranted by your evidence, such that in the long run exactly X% of your "X% confident" beliefs turn out to be true -- no more and no less.  Moreover, it seems to me that we should be equally concerned about miscalibration in either direction.  If we are underconfident (or withhold judgment entirely) when our evidence strongly supports some important truth, that's just as bad, epistemically speaking, as being correspondingly overconfident.

Scale and Symmetry in Covid Debates

One curious feature of some public debate about Covid policy is when people object to a disliked policy proposal by appealing to a consideration that counts at least as much against the alternative.  Here I'll just highlight a couple of especially striking examples of this: scale and unknown risks.

(1) Scale:  Back when people were debating whether society's response might end up being worse than the disease, it wasn't unusual to see health boosters emphasize the sheer scale of the health costs that would be incurred along the path to herd immunity through natural infection.  "Even a fatality rate of just 0.01% for younger adults would translate into thousands of deaths across that population."  That kind of thing.

Which invites the obvious response: Yes, the scale of a pandemic makes the policy stakes really high!  For example, if you lower everyone's quality of life by an average of 1/3 for a year, that translates into more than 100 million life-years of equivalent value lost in the US alone (cf. estimated health gains of a few million life-years from covid prevention measures).

New Paper on Pandemic Ethics

'The Ethics of Deliberate Exposure to SARS-CoV-2 to Induce Immunity', forthcoming in the Journal of Applied Philosophy, by Robert Streiffer, David Killoren, and myself.

Abstract:

We explore the ethics of deliberately exposing consenting adults to SARS-CoV-2 to induce immunity to the virus (“DEI” for short). We explain what a responsible DEI program might look like. We explore a consequentialist argument for DEI according to which DEI is a viable harm-reduction strategy. Then we consider a non-consequentialist argument for DEI that draws on the moral significance of consent. Additionally, we consider arguments for the view that DEI is unethical on the grounds that, given that large-scale DEI would be highly likely to result in some severe illnesses and deaths, DEI amounts to a form of killing. Our thesis is that incorporating a DEI program alongside the status-quo “calibrate-the-curve” responses could have significant advantages at the early stages of pandemics. These potential advantages mean that, at a minimum, research into DEI would have been justified early in the COVID-19 pandemic, and that DEI programs should be explored as potential additions to our overall approach to emerging pandemics in the future.

Quick summary of the harm-reduction potential of DEI:

Combining Experimental Vaccines + Variolation

Current estimates suggest that the US will end up with around 500k Covid deaths (itself but a tiny fraction of the total harm done), with eventual herd immunity reached half via vaccination and half via natural infection.  I expect we could've done much better via variolation (low-dose viral inoculation), for reasons I've previously discussed.  But I've recently learned that an even better option was available -- though it's not one that I've seen anyone suggest before.

NY Mag reports that "We Had the COVID-19 Vaccine the Whole Time" (well, since February).  Many now wonder: Would it have been prudent to distribute the experimental vaccine, at least to high-exposure or high-risk individuals, even before it was proven safe and effective?  Quite possibly.  After all, the risks from an untested vaccine of this sort are much lower than the risks from COVID-19 for many individuals.  The main potential downside would be the "moral hazard" of vaccinated individuals behaving more recklessly and spreading COVID if the vaccine turned out to be either (i) ineffective, or (ii) effective only at suppressing symptoms but not viral transmission.

But this risk could in turn be avoided by a combination policy of vaccination followed by variolation (with quarantine).  If the vaccine fails, variolation acts as a "fail-safe" to provide natural immunity (while minimizing unwitting community transmission, in stark contrast to all our uncontrolled infections).  And if the vaccine worked (as turns out to actually be the case) then the risks from variolation are further minimized.  Given even a moderate chance of vaccine effectiveness, the expected value of variolation is significantly bolstered (due to reduced risk).

Against Bad Government [Pandemic Edition]

Libertarians like to rail against "Big Government", but that often comes off as needlessly ideological and indiscriminate.  Governments are capable of doing good things, after all, and it's worth supporting them when they do.  (Granted, there are more sweeping arguments for why, say, a market-based economy is a better idea than communist central-planning, no matter how "good" the communist leaders might be.  But attacks against "big government" tend to have a wider target than just this.)

So I prefer to stake a claim against bad government.  When governments do bad things, it's the badness, not the mere fact of intervention, that I'm opposed to.  Framing it this way may help to make for a more receptive audience, too: few liberals are impressed by sweeping complaints about "big government", but nobody could deny that particular policies might be bad.  And from there we might even be able to identify systematic reasons why certain governments, at certain times, could be expected to generally do more harm than good, and hence why we might reasonably prefer that they had less power over us (at least if we're not able to remedy the deeper problem).

The Optimal Use of Suboptimal Vaccines

There's an interesting piece in the NYT warning against hasty FDA approval of a (potentially) suboptimal covid vaccine.  I'm especially interested in the third reason they offer:

Third, the F.D.A. must consider the impact of an emergency authorization on existing vaccine studies. There are almost 40 continuing vaccine clinical trials, and several have been identified as promising. Those trials are blind, which means that the participants do not know whether they are getting the vaccine or a placebo. Would some participants drop out and opt for the newly authorized vaccine, undermining those studies? Would the trials be able to recruit additional participants who would risk getting a placebo once there was an authorized vaccine?

It would seem pretty messed up if we had to deprive millions of people of a potentially life-saving vaccine (say, one that might cut covid fatalities by half) because that was the only way to ensure that we could find unvaccinated volunteers to test (potentially) better vaccines (to find one that would ultimately bring about the swiftest end to the pandemic).  Ideally, we should want most people to make use of the suboptimal vaccine in the meantime, whilst incentivizing/compensating research participants to forego the current vaccine and instead stick with their trials, to help society find an even better vaccine.

If only our society had invented some means of exchange that could be used to provide the necessary incentive/compensation!  Failing that, I guess we must simply watch as more people die...

Against Prudish Research Ethics

We're all familiar with prudishness as it applies to sexual ethics: the prude thinks certain sex acts are immoral, even between happily consenting adults.  They also hold that sex work is inherently degrading, and that others should not be allowed to offer monetary compensation in exchange for one's sexual labour.  The prude is not willing to tolerate others engaging in consensual and mutually beneficial exchanges in this arena if they don't stem from what the prude regards as the "right" motivations and take place within "approved" institutional arrangements (e.g. marriage).  It's a deeply illiberal perspective that has thankfully fallen out of favour in recent decades.  We may, of course, have reasonable concerns about the exploitation of sex workers in practice.  But it's increasingly recognized that the best response to such practical concerns is to improve the options available to those in desperate circumstances, not to deprive them of (what they evidently regard as) their current best option.  So I think it's fair to say that liberals have won out over sexual prudes in our current cultural milieu.

Sadly, the reverse appears true within the arena of research ethics.  Research prudes think that certain kinds of medical research (e.g. involving voluntary infection) are unethical, even when all involved are happily consenting adults.  They disapprove of offering monetary compensation to research participants, to make participation worth one's while when it otherwise would not be.  They are not willing to tolerate others' engaging in consensual and mutually beneficial research arrangements if participation doesn't stem from what the research prude regards as the "right" (i.e. non-financial) motivations.  It's a deeply illiberal view that unfortunately still predominates, with cultural bastions like the New York Times routinely dismissing controversial ("queer") research possibilities as "unethical", without argument.  We may, of course, have reasonable concerns about the exploitation of research participants in practice.  But it's depressing how hastily people assume that the best response to such concerns is to paternalistically deprive others of an option that they might well have reasonably preferred over their available alternatives.

Perhaps the most important difference between the two arenas is that the research prude's illiberalism is vastly more harmful.  Medical research has immense positive externalities.  So preventing it has immense negative externalities.  You're not just harming the would-be research participants (not to mention undermining their autonomy), you're also harming all those who end up suffering from medical conditions that could have been cured or prevented had the research gone ahead.  Missed opportunities are rarely salient, and so do not provoke the outrage that they truly deserve.  But on any reasonable estimate, the death toll of research prudishness is surely monstrous.