Public Opinion Supports Covid Vaccine Challenge Trials

A recent study finds broad support for the idea in many countries, including the US.

(Youneedgraphics/Dreamstime)

Back in May, I outlined the moral case for using "challenge trials" to speed up the process of testing and deploy possible coronavirus vaccines. In a challenge trial, volunteers would be given the vaccine and then deliberately infected with the virus, to see if the vaccine gives them immunity.

As economist Alex Tabarrok explains, challenge trials could potentially make vaccines available months earlier than it would be otherwise. And if we can make vaccination happen even a few weeks sooner than would be the case for conventional trials, we can save many lives, and avoid large economic and social costs, as well.

A recent article by Berkeley political scientist David Broockman and several coauthors finds widespread public support for challenge trials in numerous countries, including the US. Here is the abstract (HT: Tyler Cowen]:

A vaccine for COVID-19 is urgently needed. Several vaccine trial designs may significantly accelerate vaccine testing and approval, but also increase risks to human subjects. Concerns about whether the public would see such designs as ethically acceptable represent an important roadblock to their implementation, and the World Health Organization has called for consulting the public regarding them. Here we present results from a pre-registered cross-national survey (n= 5; 920) of individuals in Australia, Canada, Hong Kong, New Zealand, South Africa, Singapore, the United Kingdom, and the United States. The survey asked respondents whether they would prefer scientists to conduct traditional trials or one of two accelerated designs: a challenge trial or a trial integrating a Phase II safety and immunogenicity trial into a larger Phase III efficacy trial. We find broad majorities prefer for scientists to conduct challenge trials (75%, 95% CI: 73-76%) and integrated trials (63%, 95% CI: 61-65%) over standard trials. Even as respondents acknowledged the risks, they perceived both accelerated trials as similarly ethical to standard trial designs, and large majorities characterized them as "probably" or "definitely ethical" (72%, 95% CI:70-73% for challenge trials; 77%, 95% CI 75-78% for integrated trials). This high support is consistent across every geography and demographic subgroup we examined, including people of diverging political orientations and vulnerable populations such as the elderly, essential workers, and racial and ethnic minorities. These findings bolster the case for these accelerated designs and can help assuage concerns that they would undermine public trust in vaccines.

I do not claim that such widespread public support proves that challenge trials are morally justified. As the author of Democracy and Political Ignorance, I recognize that public opinion on policy issues is often influenced by ignorance, partisan and ideological bias, and other dubious considerations. Thus, my moral justification for challenge trials doesn't depend on how popular they are.

That said, the popularity of challenge trials is potentially significant because it might alleviate politicians' and business leaders' fears that holding such trials would lead to a public backlash. If Broockman and his coauthors are correct, the vast majority of people are likely to support challenge trials, despite the possible risks.

One limitation of the study is that the authors did not ask respondents whether they approve of paying challenge trial volunteers, which I argue is also justified, but some may object to on "commodification" and "exploitation" grounds, similar to those raised against legalization of organ markets. But even if payment to participants is limited to compensation for time and expenses and free health care for any complications they might suffer, it is likely we can still attract a large number of volunteers. The impressive 1 Day Sooner website has already signed up over 38,000 challenge trail volunteers from around the world (that figure may grow even higher by the time you read this).

Covid vaccines are, of course, already being tested in the US and elsewhere. But the progress is impeded by the inherent slowness of the process, and in some cases by a shortage of volunteers from specific demographic categories. Challenge trials could help address both problems.

The faster the better. Time is money. And in this case it's literally a matter of life and death, as well.

Comments

[Victor]

Phase 3 clinical trials are necessary to prove efficacy and safety in the population groups that will be vaccinated. There are some obvious problems with replacing phase 3 with challenge trials:

1. you would need to challenge a set of participants representative of the population, including 75 y.o. people with diabetes and such. No researcher would take such a risk.
2. what kind of virus challenge would be representative of the real life exposures? How many viral particles? Through the nose, through the eyes? Nobody knows. I believe, in primate challenge experiments, people just use super high viral doses, which is the only way to know that the exposure was sufficient, if you don’t have a plenty of subjects. Obviously, there will be no placebo arm. Maybe you didn’t challenge your participants correctly, but you will not know.
4. Collecting safety data takes time anyway. You need to wait those few months after the vaccination.

So, it’s just not worth it.

[bernard11]

I don’t have a particular moral objection, but I do share the concerns Victor expresses. If you don’t have a good sample, the results are suspect, and you are taking enormous risk.

In other words, my concerns are with the accuracy of the results.

[Á àß äẞç ãþÇđ âÞ¢Đæ ǎB€Ðëf ảhf]

1 and 2 are exactly the same for all trials. Challenge trials are no different.

4 also applies, and the sooner you start, the sooner you get good data for 4.

3 is a mystery to us all.

[bernard11]

1 and 2 are exactly the same for all trials. Challenge trials are no different.

No, because in a challenge trial you deliberately infect all the subjects. In an ordinary trial you don’t do that. Volunteers get the vaccine, but are not intentionally infected.

[Á àß äẞç ãþÇđ âÞ¢Đæ ǎB€Ðëf ảhf]

You can certainly choose which volunteers to infect. Nothing prevents it.

Typical statist. Doesn’t understand “voluntary”, thinks it means chaos or mandatory, nothing in between, nothing actually voluntary, nothing actually controlled.

[bernard11]

You can certainly choose which volunteers to infect. Nothing prevents it.

No. You can’t. red the description. You deliberately infect every participant. If you choose not to infect people over 65, say, your test is useless.

Typical statist. Doesn’t understand “voluntary”, thinks it means chaos or mandatory, nothing in between, nothing actually voluntary, nothing actually controlled.

Oh, fuck off. You have no idea what I’m talking about. There is nothing “statist” about suggesting that the results of this kind of trial may be worthless.

[BadLib]

You infect every participant, not every applicant.

Not all applicants are accepted into the study. As well, obviously there may not be enough volunteers, let alone approved participants, with a specific attribute that is thought to be relevant.

If too few people with some relevant attribute, such as being over 65 or having diabetes or suffering from heart disease or being of some ethic group, are accepted into the challenge trial, perhaps the vaccine isn’t approved on an accelerated basis for people with that attribute.

Meanwhile, the traditional trial would presumably be continuing with many tens of thousands of participants and short term side effect data would be available from that study. The challenge trial is focusing on efficacy.

[Victor]

Pardon, 2 and 3 got merged, and 4 remained 4. I don’t see how to edit

[bernard11]

Just looked at Tabarrok’s article. He is talking about a trial with a total of 100 subjects, of whom 50 get the vaccine. That strikes me as terribly low. Side effects that affect only small percentages of vaccine recipients have a good chance of not showing up at all.

[Don Nico]

Victor,
the increased risk for people with diabetes is not proven on the basis of mechanism nor is it seen statistically in a sample of 5.5 billion people

[BadLib]

People are not required the vaccine that has only been approved early due to challenge trials with whatever caveats that may entail. Just require informed consent before receiving such a vaccine if the concerns are valid.

If enough challenge trial participants over 75 or with diabetes can be found, the vaccine simply could be contraindicated for such people who would have to wait until a vaccine had been approved via a “traditional” trial (which, presumably, included participants over 75 and with diabetes) later.

As far as (4), how long do you wait? We don’t know – esp. for the more novel vaccines. Every month of delay will cost tens of thousands of premature deaths (fortunately, mostly among old people so the reduction in Quality Adjusted Life Years is not as bad as it could be).

[Kenneth Almquist]

With regard to point 1, the proposal by Alex Tabarrok does not call for testing on a representative sample of the population. With regard to point 2, the proposal does use a placebo control. All of the volunteers are exposed to the virus, even those who were given a placebo rather than a vaccine.

The idea is worth considering only if there are a number of vaccine candidates ready for phase 3 trials but you don’t have enough volunteers to conduct phase 3 trials on all of them, at least not right away. In that case, the result of a challenge trial covering all of the vaccines could be used to help decide which vaccines to put into phase 3 trials first.

The challenge test is not an alternative to a phase three trial, both because it doesn’t use a representative population and because it doesn’t have enough volunteers to provide much information about safety.

[Krayt]

“It’s just not worth it.”

By what definition? When people are dying hand over fist, you have to compare those deaths vs. the deaths from a drug rushed out too soon (which, being monitored, will be stopped shortly anyway.)

The issue isn’t testing. It’s protracted testing. How many continued dying as drugs take up to 10 years for approval, vs. a relative handful who die because of early release of drugs that ultimately wouldn’t make the cut?

Best not to ask that question, lest the FDA turn out to be the largest mass murderer the past century.

[Á àß äẞç ãþÇđ âÞ¢Đæ ǎB€Ðëf ảhf]

The only moral problem is whether they are voluntary and informed. Everything else is nobody els’s business, especially the “government’s”, which is shorthand for “thugs who enforce my opinion”.

[DaivdBehar]

No data on the vaccines. We know that 99% of the infected will have no symptoms or the mildest, briefest symptoms. Why gamble on the unknown? There are probably 100 million infected people in the US, if the rare total population testing studies are an indication. The vaccines should be given to COVID health workers, and to vulnerable elderly people. Leave everyone else alone.

[Don Nico]

“We know that 99% of the infected will have no symptoms or the mildest, briefest symptoms. ”
Where in the world did you get such faulty information? It is plain wrong

[DaivdBehar]

From the denominator left out of biased news reports. Projecting the results of studies of entire populations, the number of infected people is likely 100 million in the US, today.

I suggested testing a valid sample of the population each week. The CDC has yet to do that. That proposal is based on knowledge of 5th grade math, fractions.

[AtR]

Most of the vaccines under consideration are being tested for their ability to reduce mild symptoms, and not for their ability to prevent infection.

See https://www.forbes.com/sites/williamhaseltine/2020/09/23/covid-19-vaccine-protocols-reveal-that-trials-are-designed-to-succeed/#52faeb785247.

[Stephen Lathrop]

Seems like a rigorous challenge trial would need to include highly vulnerable subjects, and vary the exposure until it got high enough reliably to deliver fatalities. Without evidence that your challenge is sufficient to kill people, what have you really learned?

Any volunteers?

[ReaderY]

The first rule of medical ethics is to do no harm. In medical situations, people are desperate. Permitting patient harm permits exploitation.

Suppose a rich had an ordinarily rare disease. Should society permit deliberately Infecting enough people to provide a reliable trial of a drug? And when that fails, find new people and try again on another? From the point of view of strict classical economics, the answer is yes. Money is utility, and the life of one billionaire is worth exactly the lives of a billion people each with a dollar to their name. If you kill off half of them to save the life of the billionaire, from the perspective of classical utility society has come out ahead. And how will people with no money eat except by agreeing to what the billionaire wants? There is a reason why social morality regards human life as an inherent value distinct from its economic utility, and places limits on the kinds of agreements people can make.

[Krayt]

Meanwhile, back in reality, the correct measurement is net lives saved. Everything else is murder. For ethics, are the test subjects volunteering with an understanding of the risks?

You’ve introduced money as an equivalency for lives, and in a very class warfare way, well done!

This heart drug took an extra 3 years to prove it was safe, so sacrificed a million lives worldwide, which otherwise would have been far, far less, even if the drug proved ultimately dangerous, and was pulled (the study does not stop.)

But oh my god, the hay that politicians and lawyers can make with that! And let’s build up memes that this is the ultimate wrong, those dirty billionaires!

Thanks for nothing, politicians and lawyers.