We are often told that we cannot leave the pharmaceutical market unregulated, as uninformed consumers will make poor choices. I don’t doubt that there would be many examples of poor choices in a free market, but that’s not the same as claiming that the choices would be poor, on average.
This article caught my eye:
Nearly 200 covid-19 vaccines are in development and some three dozen are at various stages of human testing. But in what appears to be the first “citizen science” vaccine initiative, Estep and at least 20 other researchers, technologists, or science enthusiasts, many connected to Harvard University and MIT, have volunteered as lab rats for a do-it-yourself inoculation against the coronavirus. They say it’s their only chance to become immune without waiting a year or more for a vaccine to be formally approved.
Among those who’ve taken the DIY vaccine is George Church, the celebrity geneticist at Harvard University, who took two doses a week apart earlier this month. The doses were dropped in his mailbox and he mixed the ingredients himself.
Church believes the vaccine designed by Estep, his former graduate student at Harvard and one of his proteges, is extremely safe. “I think we are at much bigger risk from covid considering how many ways you can get it, and how highly variable the consequences are,” says Church, who says he has not stepped outside of his house in five months.
Before going any further, let me emphasize that I have zero knowledge as to whether this is a good idea, and I don’t plan to discuss this vaccine. The effectiveness of this vaccine has no bearing on this post.
Instead, I’m going to claim that the people taking this vaccine are relatively bright and welli nformed on the subject. These are not the sort of people whom the FDA was set up to protect from making foolish decisions about their own healthcare. The “uninformed consumer” rationale for regulation does not apply to them, even if, ex post, their decision turns out to have been a mistake.
There are costs and benefits to taking unapproved vaccines. Given that fact, there’s a question as to whether the FDA should allow but discourage vaccines still in development (as they do with certain approved drugs like hydroxychloroquine) or ban them outright, as they do with drugs or vaccines not formally approved for any use.
Reading this article on the Harvard and MIT scientists self-vaccinating makes me wonder if the FDA’s official policy is actually the policy that maximizes expected utility.
If I were told tomorrow that I had lung cancer, my next question would not be, “What do you recommend I do?” Instead my question would be, “What do the majority of cancer doctors do when they themselves are faced with a similar diagnosis.” Based on what I’ve read, the answer to these two questions would be quite different.
READER COMMENTS
Alan Goldhammer
Aug 3 2020 at 4:11pm
Scott – there are two outstanding questions about a COVID-19 vaccine. Is it safe and is it effective. the answers to both of those questions require clinical trials of varying sizes. One of the lost opportunities has been the inability to draw a strong correlation between the immune response of the vaccinated individual and broad immunity to the virus in “the wild.” More could be done in this area that would actually lead to faster vaccine development in the future. Safety is more complicated as large trials are required to find rare but potentially dangerous safety signals. If there is a rare side effect occurring at the rate of 1 in 10,000 people, one needs a minimum of 30,000 in the trial to reliably find this out. Since most vaccines are given to healthy individuals, trials are over-powered for safety. COVID-19 presents a special case because of the lack of good treatment and the high mortality in certain populations. It may be that it is in society’s best interest to provide vaccines earlier than would be customary because of this.
The COVID-19 vaccine landscape is further complicated because of the number of totally new mRNA and DNA vaccines that have not been used before. We don’t know whether they will work or how long lasting immunity will be. Will they present different side effects from more traditional vaccines that have been widely deployed. What happens if the first vaccine turns out to be the wrong choice? It’s all complicated and while I have my own opinions about which of the current candidates is best and that I would take, I am glad that I am not running the FDA right now.
Having read the RADVAC group’s effort on the citizen’s vaccine, I applaud them for their DIY approach. I do worry about sterility issues with such homemade approaches, The nasal delivery method does get around some of the issues that would be really bad were it an injectable. What isn’t clear is whether the short peptides that are part of their vaccine preparation are sufficiently antigenic as is the whole Spike protein that is the basis of the major vaccine projects. The website says about 20 people have taken this vaccine. It would be informative if they had antibody titer data to present.
robc
Aug 3 2020 at 6:40pm
I dont dont remember you commenting on it, what did you think of Robin Hanson’s variolation experiments suggestion back in March?
Even today, that seems a better way to get immune ( for the young and healthy) than an experimental vaccine.
Alan Goldhammer
Aug 3 2020 at 8:03pm
I commented on his blog at the time. If this was to be a serious project the military should have taken advantage of the T.R. Roosevelt outbreak to use the carrier as the variolation test site. It can easily be supplied and people on and off loaded. The military also has a vested interest in keeping personnel. That they did not entertain this possibility means a) they did not consider it or b) considered and dismissed it. My own feeling based on the epidemiology of the disease is that it too risky. There is no way to predict what group of people will react badly to SARS-CoV-2 and progress to severe COVID-19. We also know that the big problem is overloading the hospital system. Massive variolation would likely exacerbate this since we seem to be in the middle of this experiment right now.
We may get very lucky and the herd immunity percentage is on the low side in terms of number of infections. If we take the number of infected as 10X that of the reported number we stand at maybe 40 million plus Americans who have been infected. The low number for herd immunity might be 25% which means we are 1/2 way home. Unfortunately, this likely means another 120-150 thousand fatalities. Even at herd immunity there would still likely be sporadic but smaller outbreaks.
I was optimistic that we could put a testing regime in place with good public health measures. Both attempts have been failures.
Dylan
Aug 3 2020 at 4:35pm
I’m fine with people having the right to do stupid and possibly dangerous things, so I don’t really have a problem with this on that level. My question is, where does the knowledge come from? Let’s assume this vaccine actually works, it almost certainly doesn’t, because most experimental drugs don’t work, but for the sake of argument assume it does. How would we ever actually figure that out? There is nothing anywhere close to a clinical trial here. Estep even admits he doesn’t know how many of the 70 people he’s sent the drug to have taken it. George Church hasn’t left the house in 5 months, now that he’s taken the drug, is he suddenly going to start going to big gatherings? There’s nothing like a control arm, people are self-selected, it is a small group, etc… The best you can figure out is if it has any immediate and showstopping side effects that hit a large proportion of the people taking the drug.
That’s where I think the utilitarian argument comes in for drug regulation. The fact is that we have very little idea of what will work in drug discovery before testing it in people. We don’t even have a good idea after testing in people, unless the trial is well designed. See all the issues with HCQ benefits that are seen anecdotally, but disappear in RCTs. Regulation ensures that we have the right incentive structure in place to actually do the long, hard, expensive work of figuring out what works.
If you want an example of what drugs look like without this, take a look at the unregulated stem cell market. There are ambiguities in FDA regulation that have let stem cell clinics proliferate without any evidence to back up their claims, and they make some pretty big claims. The story I link to includes links to employers requiring employees to make use of these clinics because they are sometimes cheaper than FDA approved options (not that most of these clinics are by any means cheap). These places stay in business by anecdotal evidence, observational trials, and the desperation of patients that don’t have better options. The thing is, there’s next to zero good evidence that these work. I used to follwo a couple of companies that were doing real work in the space, but early promising results didn’t pan out.
Scott Sumner
Aug 4 2020 at 1:52pm
Two mistakes:
“Let’s assume this vaccine actually works, it almost certainly doesn’t, because most experimental drugs don’t work”
This confuses drugs that treat diseases with vaccines.
As for testing, no one is suggesting that we stop doing testing. People are free to use HCQ even as it currently undergoes testing.
Dylan
Aug 4 2020 at 2:29pm
I don’t think there is as big of a distinction here. Sure, we’ve got a longer history with vaccines than we do with other classes of drugs, so that might de-risk development a bit, but that certainly doesn’t translate to these fancy new RNA vaccines, which we’ve not had success with so far.
As I said above, I’ve got no problem with one offs like the example. I think people are stupid for doing it, not only because it’s dangerous, but because if you’re going to do something dangerous like this, would be nice to at least help advance human knowledge a bit.
Here’s where the utilitarian argument comes in though. Drug trials are already pretty hard to recruit for. One of the big selling points is getting access to experimental drugs that you otherwise couldn’t get. You only have a 50% chance of actually getting the drug, but that’s better than no chance. However, if you can instead buy an experimental drug, I think many people would prefer to do that, partially because they don’t really understand how unlikely it is to work. That’s one block in a delicate incentive structure that allows for clinical trials to work the way they do, and when you start removing blocks the whole thing is likely to tumble.
Mark Z
Aug 3 2020 at 5:41pm
When Ralph Steinman (who won the Nobel Prize in Medicine a few years ago) was diagnosed with pancreatic cancer he treated himself fairly extensively with immunotherapies he himself largely designed. Incidentally, many researchers at various institutions aided him in his self-treatment endeavors (here’s an article about Steinman’s self-treatment: https://www.reuters.com/article/us-nobel-medicine-experiment/insight-nobel-winners-last-big-experiment-himself-idUSTRE7956CN20111006). If the thinking in that case was that Steinman was an expert on the topic and thus knew what he was doing, one could argue that even someone who doesn’t know what he’s doing should be able to hire someone like Steinman who does know what he’s doing to do it for him. (According to the article, somehow – I have no idea how – what Steiman did wasn’t in violation of FDA rules)
Garrett
Aug 3 2020 at 7:30pm
This reminds me of an idea Matt Levine had to replace the accredited investor requirement:
I think this would work for non-FDA approved drugs as well.
Scott Sumner
Aug 4 2020 at 1:34pm
Love it!!
Matthias Görgens
Aug 5 2020 at 5:00am
Scott, I warmly suggest you subscribe to Matt Levine’s column Money Stuff, if you haven’t already done so. It’s full of gems like these.
Phil H
Aug 4 2020 at 2:12am
I still think this is part of the massive straw man argument that dogs this site and its response to the FDA and medical regulations in general. The FDA doesn’t say you can’t sell unlicensed vaccines. It just says you can’t can’t claim that they are effectively vaccines when you sell them.
There’s probably another law that says you’re not allowed to sell nasty viruses, and it’s the combination of those laws that creates the prohibition. But the essence of the FDA regulation is that you’re not allowed to make an unsupported claim in your marketing. It’s no more or less than a highly specified and beefed-up version of the rule that you’re not allowed to mislabel goods for sale – a rule that has been hallowed in common law for many centuries.
So all of this argumentation is sadly misdirected, I think.
Scott Sumner
Aug 4 2020 at 1:38pm
So you are saying the article is wrong?
“The US Food and Drug Administration (FDA) requires authorization to test novel drugs in the form of an investigational new drug approval. But the Radvac group did not ask the agency’s permission, nor did it get any ethics board to sign off on the plan.
Estep believes Radvac is not subject to oversight because the group’s members mix up and administer the vaccine themselves, and no money changes hands. “If you are just making it and taking it yourself, the FDA can’t stop you,” says Estep. The FDA did not immediately respond to questions about the legality of the vaccine.”
Phil H
Aug 5 2020 at 8:29am
Well… I’m not sure how you think that passage you quote does anything but reinforce my point. At least as Estep sees it:
“If you are just making it and taking it yourself, the FDA can’t stop you”
Which is what I said. The FDA can’t stop you doing anything. What the FDA can do is deny you a license to market your drug with the claim that your drug is effective for indication X. And the FDA will deny you the license if you have failed to follow their administrative requirements.
It’s possible that this “trial” will not be admissible evidence for an application to the FDA to market their vaccine. But there’s no evidence here that the FDA is “banning” anything.
Phil H
Aug 5 2020 at 8:32am
Sorry, just to respond slightly more fully on that first sentence: It’s incomplete.
“The US Food and Drug Administration (FDA) requires authorization to test novel drugs in the form of an investigational new drug approval [if you wish to subsequently get FDA approval to market your drug]” is what it would say if it were complete. And of course, everyone does wish to get that approval. Still, this is not the same thing as “banning them outright”.
Hambone
Aug 4 2020 at 7:33pm
Sure. Technically, if you chain someone up in your basement for a week with no access to water, they just die of dehydration…
In the medical device industry we see FDA non-approval as equivalent to a ban. Sometimes, you can sell something off-label (*if* it’s already approved for some other indication) but even then there are a ton of rules that the FDA has around that, and you can get fined into the ground for breaking any one of them, or just for selling too much of a product off-label and not running a trial. But for a new drug or device, there is no overlap in the Venn diagram of “was not approved” and “is currently being sold”. The FDA is a necessary piece in that arrangement, even if there are other actors that help enforce it. In its absence, physicians, hospital purchasers and insurers would come up with some way of determining safety and effectiveness, because people would still care. But since the FDA is there, whether something is “safe” and “effective” comes down to how 4 all-too-human FDA reviewers (usually from the bottom 25% of their grad school classes, by the way) subjectively interpret the data from your trial. That’s nuts.
JonB
Aug 4 2020 at 3:06pm
It may be useful to remember that there are multiple local minima in complicated decision spaces…..The argument that the FDA is near the global maximum in regulatory wisdom is painful beyond belief….speaking from personal experience in dealing with regulatory staff who are less competent than many local PCPs in terms on intuitive understanding of risk and benefit.
I suspect the solution is regulatory competition. Produce a system where exit is more important than voice. Consumers and their insurances companies choose a regulatory system which suits their personal needs.
Currently, it is illegal to run a drug study without FDA approval. This is a mistake.
Capt. J Parker
Aug 5 2020 at 1:15pm
JonB said:
I thinks this needs elaboration. I’m sure the FDA clinical trials regulatory scheme has lots of things in need of improvement. But, IMHO the requirements for receiving Informed Consent from the patient, which FDA enforces, need to be, if anything, tightened and more rigorously enforced.
Capt. J Parker
Aug 5 2020 at 12:34pm
FDA requires drugs and medical devices to be safe and effective. These are two quite different criteria with efficacy usually being the the criteria that takes much more time and much more money to prove. Many (most I think, but I am not sure of this) other government drug regulation schemes focus mainly on the drugs being safe and the consideration of efficacy is left up to the medical community and the public.
In the COVID 19 era it is worth debating if the safe and effective criteria is utility maximizing of if safe alone should be FDAs criteria. Vaccines for COVID19 have already passed the safety threshold, or at least passed it enough for FDA to allow the vaccines into large scale efficacy trials. If the “safe but yet to be proved effective” vaccines were available to the public at large by prescription my bet is that would be more utility maximizing than the current approach.
If FDA really is “allowing but discouraging” Hydroxychloroquine use to treat COVID19 then FDA is doing something Congress never gave them authority to do. Off label use of Hydroxychloroquine to treat COVID19 is not within FDA’s authority to regulate. FDA can prevent manufacturers and distributors of the drug from promoting it for use in treating COVID19 but, if your physician wants to prescribe it to you for COVID19 FDA has no legal right to do one thing on another about it.
JonB
Aug 5 2020 at 6:35pm
“Informed Consent” is a often a wizard word like “inflammation” or “justice” used by the elite to signify adherence to guild professional standards, e.g. advertising.
These forms and processes are typically generated by committee, typically producing overly lengthy and jargon-filled paperwork. Although, in AGGREGATE, these forms are mean to be useful, in the real world the complexity of the physician-patient interaction is poorly captured by these forms. Asymmetrical information is everywhere, in both directions.
Who decides which directions need to be tightened and more rigorously enforced? How does one judge whether these changes are having the desired effect? Quis custodiet ispsos custodes? Who watches the FDA?
We need as much innovation in informed consent as we do in drug design. Markets fail, more markets,.,,,,,,
Capt. J Parker
Aug 6 2020 at 10:18pm
I must respectfully disagree with you JonB. Informed Consent for participation in a clinical trial means you have been informed that the treatment you are being offered lies outside the current standard of care and has not been demonstrated to be safe and effective.
The complexities of the physician patient relationship do not absolve the physician of informing the patient when the treatment being proposed is experimental.
The question of who decides what needs to be tightened is a good one. My partial answer is: the consumers of medical care should have some input into the decision.
In my own personal history, I was having a routine screening procedure done. Just before receiving anesthesia the physician informed me that in an effort to improve training of physicians in his field he would like to have his assistant “follow the case.” Having some experience with clinical trials and Informed Consent I recognized that I was being asked for verbal consent, probably for participation in a Non-Significant Risk study. I began asking questions and got a lot of vague answers. After getting assurances that no experimental drugs, devices or procedures were involved I said OK, fine, let’s get on with it, your assistant can “follow the case.” I’m still unclear what I consented to. Probably some relaxation of HIPPA rules so my data could be tracked but, who knows.
Looking back I think that whole incident was total BS. I was asked to consent to something – who knows what – while lying on the procedure table. I remember having the fleeting thought at the time to ask the physician if his IRB knew what he was up to but thought best not to tick off the guy about to perform a procedure on me, keeping in mind “the complexities of the patient physician relationship.”
Comments are closed.