Ever since the passage of the Kefauver-Harris Amendment, a.k.a. the “Drug Efficacy Amendment,” or Drug Amendments of 1962, to the Federal Food, Drug, and Cosmetic Act in the wake of the thalidomide debacle, the FDA has been mandated to hold to two standards when licensing drugs. It must require evidence that demonstrates convincingly that the drug is (1) safe for its intended use and (2) effective against the disease or symptoms that it is meant to treat. (Before this amendment, as hard as it is to believe, the FDA was only mandated by law to require evidence that a drug was safe before licensing it.) In brief, the amendment required drug manufacturers to submit evidence showing that the drugs they were applying to the FDA to license were safe and effective, expanded Food and Drug Administration (FDA) oversight over drug manufacturers through more rigorous postmarketing surveillance and oversight of manufacturing processes to prevent contamination and assure that what the label said was in a drug was actually in the drug, and set rules for drug advertisements and labels to minimize deceptive advertising. As part of this postmarketing surveillance, drug manufacturers were required to report adverse reactions to new medications to the FDA. Indeed, the Kefauver-Harris Amendment was a driving force in the the development of randomized controlled clinical trials (RCTs), both scientific and ethical, because increasingly after passage of the amendment, the FDA required high-quality RCTs to support applications to license new drugs, until the RCT became the gold standard science required. (Yes, I’m riffing.)
While the FDA has long been a bogey man for quacks and the “health freedom” movement, arguably in 1962 it be came the bogey man, as the Kefauver-Harris Amendment vastly expanded its purview and power. Since then, the conflict between commerce and the desire to bring drugs to market quickly versus FDA regulations and drug approval processes has increased. In addition, the desire on the parts of advocates or sufferers of serious diseases to access promising new therapies quickly versus the FDA’s more conservative mandate to make sure that such drugs are safe and effective before marketing them have frequently clashed, most notably when the AIDS pandemic hit in the 1980s, leading to narratives such as the one in Dallas Buyers’ Club, a movie featuring an inaccurate and highly deceptive plot in which a scrappy underdog with AIDS—played by the charismatic Matthew McConaughey, no less!—defies the FDA to bring unapproved drugs into the country to cure himself and others of the new and deadly disease. Ever since the 1980s, a common narrative coming from the “health freedom” movement—or, as I like to call it, the movement that wants freedom from any pesky laws and regulations governing medicine—has taken up the narrative that the FDA is “keeping cures from the people” through its hidebound insistence on sufficient pesky RCT evidence showing safety and efficacy as a requirement to approve new drugs. Indeed, narratives like this were behind the whole “right-to-try” movement that purported to let patients with terminal illnesses access experimental drugs, as long as they have passed only the lowest bar of evidence, a phase I clinical trial demonstrating safety. The claims of their supporters notwithstanding, these laws have done nothing for such patients.
While there are real and legitimate criticisms of how the FDA regulates drugs, I periodically feel the need to remind our readers that the real goal of the health-freedom movement has always been to weaken, neutralize, or even eliminate the FDA. Right-to-try has been just one tool to achieve this, under the banner of “freedom” and outrageously deceptive arguments by libertarians claiming that, because of how slow the process is to approve drugs, the FDA kills more people than it saves. Lately, I’ve been seeing publications aligned with President Donald Trump’s and Secretary of Health and Human Services Robert F. Kennedy Jr.‘s “make America healthy again” (MAHA) movement dredging up old libertarian tropes, attacks, and lies about the FDA, most recently in an article in Reason by Zach Weissmueller entitled The COVID Reckoning Doesn’t Go Nearly Far Enough. You might think that this isn’t about libertarian tropes about the FDA, but the splash page on social media gives the game away:
require something greater than replacing the public
health establishment: ending the FDA’s monopoly.”
At least not-a-friend-of-the-blog Jeffrey Tucker over at the “free market” (and very antivax) Brownstone Institute was a lot more honest about what MAHA is thinking when he titled his post What if the FDA were eliminated?, a more accurate reflection of libertarian sentiment. Just to tweak your memory, Jeffery Tucker is a far-right-wing neo-Confederate prone to conspiracy theories about vaccines that rival those of Mike Adams. You might recall that Tucker was a behind-the-scenes driving force behind the Great Barrington Declaration, the October 2020 manifesto calling for a “natural herd immunity” in the name of reopening society so that business could get back to making money.
Yes, MAHA is recycling old libertarian nonsense that I was writing about in 2014 when Nick Gillespie published an article at The Daily Beast entitled Kill the FDA (Before it Kills Again), not to mention in 2006, when Reason‘s Ronald Bailey just wanted to “suggest that the FDA’s increased obsession with safety may be killing more people than it saves.” More recently, RFK Jr., soon after coining the slogan “make America healthy again” (MAHA), in essence declared war on the FDA. Although he didn’t go nearly as far as Weissmueller, Gillespie, Bailey, and other libertarians have in calling for its dismantling he did echo many of the same tropes long popular among the “health freedom” movement.
example when the FDA ever “suppressed” vitamins,
clean foods, sunshine, or exercise.
Let’s take a look at Weissmueller’s article first, as it wraps this old quack “health freedom” message in a post-COVID wrapper to justify it, and it even abuses Bayesian analysis, a favorite topic of this blog going back to its very inception in 2008.
Zach Weissmueller’s revisionist history of the FDA: “Bayesian analysis proves I’m right.” (It doesn’t.)
Zach Weissmueuller’s article is a marvel of recycling and repackaging “health freedom” tropes in a deceptive way. Basically, it’s revisionist history—not the rigorous, necessary historical revisionism that historians continually engage in as new evidence about the past comes to light, but rather very obviously ideologically motivated historical revisionism designed to demonize public health authorities in general and the FDA in particular.
Unsurprisingly, Weissmueller starts out with what is the now-standard narrative coming from MAHA and the health freedom movement, namely that federal public health authorities at the Centers for Disease Control and Prevention (CDC), National Institutes of Health (NIH), and the FDA have “lost the public’s trust.” In these narratives, always conveniently missing is the tsunami of misinformation, disinformation, and ideologically-aligned attacks on public health coming from quacks, cranks, and contrarians, aided and abetted by social media algorithms. In any event, here we go (again):
The public health establishment lost America’s trust during the COVID-19 pandemic with its bureaucratic incompetence, “noble lies,” and authoritarian mandates.
It was about time. Agencies like the National Institutes of Health (NIH) and the Centers for Disease Control and Prevention (CDC) have been failing the American public for decades. But the agency with the worst track record is the Food and Drug Administration (FDA), which, long before COVID, caused Americans to lose hundreds of thousands of life years by slowing down drug development. After COVID, the American public had finally started to catch on.
President Donald Trump put a new team in charge of the federal health apparatus, some of them truth-telling contrarians who risked their careers to expose the sins of the old guard. But it’s also a messy picture because the man in charge of Health and Human Services has controversial theories about the dangers of vaccination with little factual basis.
One notes that Weissmueller doesn’t actually specify the specific “noble lies” that the federal public health establishment supposedly told in the early months and years of the pandemic, probably because the vast majority of them were not “lies” unless you were a COVID contrarian, but he does pivot to focus the distrust of federal health agencies that developed during the pandemic to the FDA in particular. Of course, he can’t resist telling a not-so-noble lie of his own in the process, namely his nonsense claim that the FDA has long caused “Americans to lose hundreds of thousands of life years by slowing down drug development,” an evidence-light (very light), emotionally charged claim that has long been a staple of the libertarian fringe that wants to eliminate the FDA. The “logic,” such as it is, of this argument is that the power of the free market will lead to the unleashing of creativity and innovation that will bring cheap and highly effective treatments and cures to the people; that is, if you completely ignore history going back to the very founding of the FDA in 1906 and back into the 19th century.
Also, while I do give Weissmueller credit for at least acknowledging obliquely in the most milquetoast terms possible that RFK Jr. is a raging antivax conspiracy mongering loon, his admission would be more convincing if he changed the word “controversial” to “dangerous and disproven” and “with little factual basis” to “based on bad science, pseudoscience, and raving conspiracy theories.” Amusingly, while crediting President Trump this time around with empowering cranks whose views he happens to like, euphemistically calling them “truth-telling contrarians who risked their careers to expose the sins of the old guard,” Weissmueller also conveniently forgets that the President during most of the first year of the pandemic was…Donald J. Trump!
Whoever’s in charge of the FDA, however, what Weissmueller wants is this:
Nobody, no matter their integrity or political persuasion, should have the power to decide what you or I put in our bodies. The only way to fix America’s regulatory health care apparatus is to dismantle it. A new guard has replaced the old. The COVID reckoning is here. But it hasn’t gone far enough.
There we go! I’m only surprised that it took him so long to get to the lede!
Over the years, I’ve expressed a little bit of sympathy towards the point of view that people should have the freedom to decide what is put into their bodies. However, I am also a realist, based on long experience and observation of what happens when regulations are too lax or even nonexistent. I also know from that long experience and observation that the “health freedom” movement has always been far less about the freedom of consumers to decide what is put into their bodies than it has been about freedom for unscrupulous quacks to sell whatever snake oil that they can market those consumers and to use whatever distortions, deceptions, misinformation, and lies that they want in order to persuade them to purchase their nostrums and thereby yield hefty profits.
And here’s where the revisionist history gets super-duper revisionist, this time the history of the Kefauver-Harris Amendment:
The story of an overreaching public health state doesn’t begin with Robert F. Kennedy Jr. or Dr. Anthony Fauci. It begins with a law passed in 1962.
The FDA didn’t always throw up as much red tape as it does today. In its early days, its only job was to sign off on safety. Drug makers would submit safety evidence, and if the agency didn’t object within 60 days, the drug could go to market and be sold right over the counter. Back then, it was the drug company that decided whether to require a doctor’s prescription or not based on liability concerns.
This voluntary arrangement between drug manufacturers, patients, and doctors worked fairly well, says Jeffrey Singer, a practicing surgeon and senior fellow in the Department of Health Policy Studies at the Cato Institute.
“If [the drug manufacturer] thought that their drug was a little too risky and some people could have problems and they’ll get sued, they would say to the pharmacist, we only want our product to be sold to people who have a prescription,” he says.
It was a system based on informed consent and voluntary exchange.
How nice! This view makes me think of a Norman Rockwell painting of a kindly doctor with his patient. (It was always a male back then.) Who could possibly object to a system like this? Hardly anyone would object if what Weissmueller describes were an accurate reflection of the actual system in place before 1962. As I like to say, informed consent requires that accurate scientific information addressing the efficacy versus the risks and safety of a treatment be provided to patients in a dispassionate manner that allows their questions to be addressed. Remember, too, that when the Kefauver-Harris Amendment became law, medicine was still in its longstanding paternalistic era, where “doctor knows best,” and one didn’t question one’s doctor. (Think Dr. James Kildare or Dr. Marcus Welby rather than Dr. Nick Riviera.) Moreover, Weissmueller conveniently neglects the part of the Kefauver-Harris Amendment that empowered the FDA to regulate medical advertising for licensed drugs, specifically requiring accurate information about what is in the drug, adverse events found in clinical trials, and an accurate accounting of the expected efficacy compared to side effects and adverse events. Does he really think that such a provision wouldn’t have been part of the amendment if misleading claims about pharmaceutical products weren’t a problem? Seriously, misleading claims from manufacturers about pharmaceuticals are still a problem.
Also, I’ve discussed Dr. Singer before. Let’s just say that Dr. Singer is a general surgeon. Don’t get me wrong. My base training is as a general surgeon too; that is, before I subspecialized. However, Dr. Singer has no detectable significant expertise in these matters beyond standard medical school learning and likes to liken medicine to religion, doctors to priests, and accuse doctors of “scientific gatekeeping.” He’s presented as an expert by Weissmueller, but he is anything but. He is, however, a senior fellow at the Cato Institute and on the Board of Scientific Advisors of the American Council on Science and Health (ACSH). Let’s just say that these groups, like Reason, have a fairly predictable ideology that supports anti-FDA rhetoric, for instance, this article by Dr. Singer on the Cato Institute website entitled Abolish the FDA.
But let’s move on:
The economist Milton Friedman described this phenomenon in a 1999 interview with the Hoover Institution. “The pressure on the FDA is always to be late in approving,” Friedman said, “and there is enormous evidence that they have caused more deaths by their late approvals than they have saved by their early approvals.”
Funny, but I perused that interview with Friedman. He doesn’t actually present any evidence to support the claim that the FDA has caused more deaths than through its late approvals than it has saved through early approvals, other than to claim that there is “enormous evidence” for it. Basically, what Friedman is saying is utter bullshit. Really, there is no other word. Not that that stops Weissmueller from continuing:
Cancer researchers calculated that a five-year delay in approving a lung cancer drug with a modest one-percent cure rate would cost 282,000 life-years, while stricter trial regulations would save only 16. University of Chicago economists found that when the FDA did speed up approvals in the early ’90s, faster access saved between 180,000 and 310,000 life-years, compared to 56,000 life-years lost to drugs eventually withdrawn for safety reasons.
The first study from the U. of C. is a modeling study that doesn’t actually demonstrate that anything like what was modeled has actually happened, making it a hypothetical thought experiment rather than useful data. Indeed, the use of loaded language in the article, such as “regulatory fundamentalism,” is the tell that this article is more advocacy and ideology than strictly science:
When patients have poor outcomes or have been misled, regulators and others may be interrogated and censured by congressional committees and may be vilified by the press. “Regulatory fundamentalists,” who followed the exact letter of the regulations, are able to defend themselves by claiming to be faultless because they fully complied with and enforced all regulations. “Regulatory rationalists,” who instead exercised judgment in aiming to fulfill the intent of the regulations while attempting to facilitate progress, are much less able to defend themselves.51 This leads to enrichment of regulatory bodies with fundamentalists and fosters a fundamentalist reinterpretation of the regulations. Institutions, investigators, companies, and regulatory bodies then feel threatened and try to protect themselves from the uncertainty of both current and future fundamentalist reinterpretations of the regulations by taking steps that magnify the impact of the reinterpretations at the local level. This leads to greatly increased compliance rigidity, and this in turn leads to a massive regulatory traffic jam, with increased costs and delays and with decreased innovation and progress.
Or maybe people attracted to regulatory positions tend to be the sort to take their job seriously and actually enforce the regulations. Again, the loaded language tells you where the authors are coming from. What’s interesting to me are the sorts of “solutions” that they propose. Here are a few:
- Preclinical toxicology: require only LD10 in rodents
- Preclinical pharmacology: require only cytochrome p450 interactions
- Permit PI to do protocol override/appropriate modifications without IRB/FDA approval
- Stop the requirement for rapid reporting of all deviations/violations to IRB and sponsor
- Accept higher level of risk for lethal diseases than for benign and nonlethal diseases
Of course, we already accept higher levels of risk in the clinical trials used to approve drugs for cancer and other deadly diseases than we do for drugs for nonlethal diseases and more benign conditions. That’s why drugs as toxic as chemotherapy can be tested in patients suffering from deadly cancers. As for the rest, these seem… unwise. Toxicity in rodents often maps unreliably to toxicity in humans and there’s more to pharmacology than cytochrome p450 interactions. Don’t get me wrong. There are a number of ways that clinical trial documentation and regulatory burden could be reduced without compromising the safety of patients enrolling in such trials. As one might expect in a study cited by libertarians, what is proposed in this paper goes way too far.
The second article cited is not a scientific study at all, but a 2007 article in Reason by Ronald Bailey. (Funny how Weissmueller seems unable to cite more recent evidence.) Bailey does, however, cite a 2005 study by economists, not medical researchers, published in the National Bureau of Economic Research. Interestingly, this paper doesn’t really directly address what Weissmueller claims. Rather, it looks at what happened after the FDA switched to the Prescription Drug User Fee Act (PDUFA), which was passed in 1992, that empowered the FDA to collect user fees from companies to fund the drug approval process and thereby sped up the drug approval process. (Ironically, it is also the PDUFA that has led to many MAHA types accusing the FDA of regulatory capture because it depends on user fees paid by drug companies for much of its revenue.) In any case, let’s just say that this study is old and has a number of limitations that to me lead me to view it more as very likely confusing correlation with causation. Moreover, Weissmueller failed to cite other data that have shown a correlation between faster drug approvals and increased adverse events, although the data are mixed and other studies show no statistically significant difference before and after PDUFA. Suffice to say that the picture is nowhere near the slam dunk in favor of speeding up drug approvals that Weissmueller argues.
Naturally, Weissmueller lists the litany of usual COVID contrarian complaints about missteps of the FDA during the pandemic, some legitimate, some exaggerated, even including the usual conspiracy theory that the FDA slow-walked the approval of COVID-19 vaccines until after the 2020 election in order to hurt Donald Trump’s chances of winning, even quoting not-a-friend-of-the-blog Dr. Vinay Prasad, who, before he was all about casting doubt on the safety and efficacy of COVID vaccines was all about promoting them because of their potential to stop the hated “lockdowns” and other public health interventions that drove proto-MAHA bonkers during the pandemic. Weissmueller also rather misleadingly quotes friend-of-the-blog and admired science advocate Dr. Paul Offit, who had expressed disagreement with what he considered to be the premature recommendation for a third COVID booster, which was a scientifically arguable point but in Weissmueller’s hands is presented as the most flagrant political interference in FDA decision-making that there could possibly have been.
Of course, medical regulatory agencies and public health officials all over the world were struggling with deciding what to do in the face of a pandemic due to a novel pathogen, and of course mistakes were made. Unfortunately, an honest and science-based acknowledgment of FDA mistakes is not what is happening under the current regime, as we (particularly Dr. Jonathan Howard) have been documenting ever since Trump took office for the second time. Instead, Weissmuller asks:
What if the FDA had run safety tests but had zero involvement in vaccine approval?
Some Americans might have chosen to accept the risk of getting the COVID vaccine early on in the pandemic, before it had been tested. Others might choose to wait years for better data on its effects to surface. We should be allowed to make our own medical decisions based on our risk tolerance, vulnerabilities, and individual health profiles.
Instead, the federal government prevented many of us from trying the vaccine, and then it attempted to coerce many of us into getting it.
This is one of the most brain-dead takes on FDA approval that I’ve ever seen, and I’ve seen a lot of them from libertarians. First of all, to make substantial progress at ending the pandemic, any vaccine would have to be effective, not just safe. Second, Weissmueller seems to forget that vaccine supply was highly constrained when it was first granted an emergency use approval by the FDA in December 2020. With such a constrained supply initially, decisions had to be made regarding who would be eligible for the vaccine first, and it made perfect sense that frontline healthcare workers, staff at long-term care facilities such as nursing homes, and first responders should have been at the front of the line, at least until supplies expanded. I admit that I was very quick to get on the waiting list at my cancer center to get the vaccine and got my first dose in mid-December 2020. Moreover, it was not the FDA that mandated vaccines. The FDA cannot do that. It can only approve or not approve vaccines for certain population groups. Weissmueller is deceptively conflating the FDA with the entire federal government as a means of pulling a truly silly “What if?” out of his nether regions.
The part of Weissmueller’s article that does the most mental gymnastics and contortions, though, is the part where he begins by listing RFK Jr.’s depredations against science as HHS Secretary, including firing the Advisory Committee on Immunization Practices (ACIP) and replacing its members with antivaxxers, but then pivots to approvingly citing Dr. Marty Makary, who is now FDA Commissioner:
He’s also signaled a shift toward Bayesian methods for evaluating drug effectiveness, in which the FDA acknowledges the limits of its own knowledge and acts more like a sports bettor, synthesizing information from many sources to wager on the likelihood that a treatment works for specific patients.
Hint: That is not quite what Bayesian methods do. Indeed, one of our founders, Dr. Kimball Atwood, wrote extensively about Bayesian methods as applied to trials of what was then known as “complementary and alternative medicine” (CAM). The earliest example that I can find is the most detailed as well. It was posted in 2008, a mere month and a half into this blog. First, though, let’s look at what Weissmueller thinks that Bayesian methods are:
Statistician Aaron Brown agrees with Prasad that many studies leading to FDA approval are poor or inconclusive. But Brown says even better clinical trials won’t tell us conclusively how a new drug will perform in the wild. The best approach is Bayesian—let doctors and patients try new drugs, and then we can gradually learn from their effects.
“Frequentists look for decisions,” Brown explains. “Approve the drug, don’t approve the drug. Make everybody get the vaccine. Let nobody take the vaccine. Bayesians want to hand you a list of odds and say, ‘OK, if you take this pill, there’s a 10 percent chance you’ll get cured, a two percent chance it’ll kill you.’ The recent FDA action is primarily directed at Bayesian practice. They wanna bring more Bayesian tools into what is basically a frequentist FDA.”
The Bayesian approach also takes into account the heterogeneity of human physiology. A drug that fails in a broad clinical trial, or that brings outsized risk for one class of patients, might be transformative—or at least worth trying—for another subset of the population.
Brown wonders whether a more Bayesian FDA might have avoided the pitfalls of the one-size-fits-all approach of COVID-era vaccine trials and recommendations.
As you might expect, Aaron Brown is a regular contributor at Reason. He holds an M.B.A. in Finance and Statistics from the University of Chicago and an S.B.in Applied Mathematics from Harvard. As far as I can tell, he has no advanced training or experience in clinical trial design, epidemiology, or other disciplines relevant to making such broad pronouncements about drug regulation. Moreover, as longtime—as in really longtime—readers know, we at SBM—including myself—tend to lean more Bayesian than frequentist, which is why we are able to describe why frequentist trials of homeopathy are not good evidence that homeopathy works. So what’s the problem with making FDA approval more “Bayesian”?
There wouldn’t be a huge problem if it were done correctly, and what Brown describes is anything but. First, he lays down a bunch of straw men, such as “Approve the drug, don’t approve the drug. Make everybody get the vaccine. Let nobody take the vaccine.” That is an utterly nonsensical description of how the FDA approval process works. Moreover, then Brown states, “Bayesians want to hand you a list of odds and say, ‘OK, if you take this pill, there’s a 10 percent chance you’ll get cured, a two percent chance it’ll kill you.'” Seriously, these are the sorts of discussions that, for example, every oncologist has with cancer patients every day when they discuss whether a patient should undergo adjuvant chemotherapy after curative surgery: This is your chance of relapse and death with chemotherapy. This is your chance of relapse and death without chemotherapy. This is the chance that chemotherapy will make a difference. This is the chance that it will make no difference. This is the chance that you will suffer adverse reactions like X and Y. Based on this, I think it is/isn’t a good idea for you to take the chemotherapy. Seriously, I can tell from Brown’s characterization that he is utterly clueless about how science- and evidence-based medicine is actually practiced. By his own definition, we are Bayesians:
Brown wonders whether a more Bayesian FDA might have avoided the pitfalls of the one-size-fits-all approach of COVID-era vaccine trials and recommendations.
“If the FDA were entirely Bayesian,” he says, “They’d say, ‘OK who needs this [COVID vaccine] the most?’ Old people with risk factors who have high exposure to COVID—they need it the most … it can have a lot of bad effects and it’s still a good deal for them. So we give it to a bunch of them and we watch them very carefully … We’re seeing, how do they feel? When did they take the vaccine? When did they get COVID if they got it? What symptoms did they have? And we do trial and error … We’d never make it mandatory. We just slowly move to [encouraging vaccination], and we let people make their own decision based on their own circumstances.”
Again, this is utterly deceptive. Early on, when vaccine supply was constrained, groups were prioritized in just this manner, with the elderly and those with chronic diseases prioritized because they were most at risk of hospitalization and death from COVID. Brown is deceptively conflating postmarketing surveillance and adjustment of vaccine recommendations by ACIP based on the data thus obtained with the initial RCTs of the vaccine that demonstrated safety and efficacy, while making an argument against mandates that his revisionist history doesn’t support. Those RCTs were a starting point, not an endpoint. They were enough to recommend emergency use approval. Then more research was done to fine-tune recommendations, identify adverse events not detected in the initial clinical trials, and calculate more accurate estimates of efficacy and safety, particularly as the virus mutated to evade vaccine-induced (and, always ignored by people like Weissmuller and Brown, post-infection) immunity.
Coming back to Bayesian reasoning, though, as Dr. Atwood described so well, all Bayesian reasoning involves, boiled down to its essence, is that, before a study is done, we estimate, based on the best available existing evidence, the prior probability that the study will show that the therapy has an effect, called P(B). I’ll now quote Dr. Atwood:
We might not know P(B), but it is a constant. Thus on the right side of the equation we can direct our attention to the terms in the numerator, which predict certain things: if the prior probability of a hypothesis is high, it will not require much in the way of confirming data to reassure us of that opinion. If the prior probability of a hypothesis is small, it will require a large amount of credible, confirming data to convince us to take it seriously. If the prior probability is exceedingly small, it will require a massive influx of confirming data to convince us to take it seriously (yes, extraordinary claims really do require extraordinary evidence). The simplest result, albeit one that many find discomfiting, is found if P(A) approaches zero: no amount of “confirming data”—especially of the error-prone sort generated by a clinical trial—should convince us to accept the hypothesis.
P(A) for homeopathy or reiki, for instance, is so small that it fits that last definition.
If you look at prior articles by Brown, you’ll see that he likens the estimation of prior probability that a treatment will work to betting lines on sports betting sites like FanDuel or by prediction markets like Kalshi. While it’s true that there will always be some subjectivity in determining prior probabilities, in Bayesian reasoning we try to come to the best estimate we can, not the seemingly highly subjective guess that Brown implies:
Bayesians accept that a patient’s response to a treatment is not random, just unknown. Bayesians are like bookies. They want to hand patients a list of bets they can make with different payouts—cure, death, unpleasant side effects, no change, etc.—and prices attached. Patients can consult with doctors and other experts to choose good bets based on their own additional information, if any, and their personal preferences. Importantly, the Bayesian statistician considers herself one of the people for whom she gives advice—she bets for herself and her loved ones using the same odds she quotes for the public.
This has little to do with regulation, and, again, where does Brown think that we get these lists of “different payouts” and “prices attached” if not from more frequentist clinical trials?
Of course, to Weissmueller, the real goal has little to do with Bayesian reasoning. Rather, it is this:
The problem isn’t who’s in charge; it’s that anybody is in charge at all. Health care decisions are best left to patients and their doctors.
But the health care system is so complex and interdependent that the entire system needs to be unravelled.
What does that involve? I’m sure you can guess:
Brown says the FDA would do a lot less harm if it filled a purely advisory role, stripped of its power to keep drugs off the market.
“I would love the FDA to decide it was only an informational agency,” he says. “You have to change the goal of the FDA. The goal of FDA is no longer telling people what to do. It’s no longer saying these things are forbidden, these things are required.”
In this scenario, if you want to take drugs that have FDA approval, go right ahead. If you want to try drugs that don’t have FDA approval, have at it. It’s more likely you’d have to pay for them out of pocket, but nobody should have the right to stop you.
Again, this assumes that the patient is a reasonably rational actor (usually true) and that there is no tsunami of misinformation to counter the information provided by the FDA under this scenario, misinformation that even trained scientists and physicians sometimes have difficulty recognizing and rebutting. You and I know how reasonable an assumption that is.
But what about our old friend Jeffrey Tucker?
Quoth Jeffrey Tucker: Abolish the FDA!
Unsurprisingly, Jeffrey Tucker begins his article What If the FDA Were Eliminated? with similar COVID contrarian tropes about the failures of the FDA and public health establishment during COVID, adding his own misinformation about how supposedly “deadly” COVID vaccines were. I won’t bother to quote it directly, as it’s all so tiresome and nothing that I (or any regular readers) haven’t heard a thousand times before. Instead, let’s jump right into what Tucker prescribes:
A worthy thought experiment: how would drug approvals and food safety be managed in the absence of such government agencies? The thesis: the free and competitive marketplace would likely be far more strict and exacting than these government agencies. Private solutions would emerge as the standard bearers of approvals, in a way similar to how the private Underwriters Laboratory (founded 1894) codifies the safety of appliances, the Better Business Bureau (founded 1912) polices fraud in business, and actuaries in many sectors assess and price risk.
Anyone in a free market can sell anything. Doing so profitably over the long term and earning consumer trust is an entirely different matter. Markets have their own way of regulating safety, efficacy, and quality, often in ways that are more strict than government agencies have traditionally permissioned.
I note that Weissmueller alluded to, but did not dwell on, a similar idea in his article:
If Makary makes the drug approval process less stringent, that would be progress, but who knows if the next administration will undo these reforms? He wouldn’t be the first FDA director to try and fail at slashing red tape.
That’s why the better outcome would be for the FDA close up shop and disappear.
As is often the case, Tucker presents a highly biased retelling of the history of the events leading up to Congress passing The Biologics Control Act of 1902, enacted in response to two separate incidents in 1901 involving the deaths of a total of 22 children from vaccines contaminated with tetanus, 13 from a diphtheria antitoxin, and 9 due to contaminated smallpox vaccine. Tucker makes much of the history indicating that the manufacturers wanted this law in order to boost public confidence in their products and also because they were afraid that state governments would take over the manufacture of such products. According to Tucker:
Thus did the creation of the agency formed by the Congressional action (the Hygienic Laboratory, part of the Public Health and Marine Hospital Service, under the Treasury Department, and later to become the National Institutes of Health) served to create a private cartel of drug and vaccine manufacturers, crowding out private solutions and disabling the normal market-based rule of caveat emptor or buyer beware.
This is exactly what the biggest players in the industry intended. It was a brilliant strategy. Pretend to be deeply vexed by a government crackdown while pulling the strings behind the scenes of a new agency that the public trusts more than the industry. This was not only the birth of a new path toward public management of this one industry; it was the origin of the regulatory state itself insofar as it intervenes directly in the consumer marketplace.
One wonders if Tucker would have preferred state governments taking over the manufacture of such products. According to him, after the popularity of Upton Sinclair’s novel The Jungle (a book about the meatpacking industry of the time that everyone should read) the passage of the Pure Food and Drug Act of 1906 supposedly followed the same playbook, although Tucker can’t help but make the evidence-free claim that the law resulted in less safe meat, citing an article by himself that doesn’t actually demonstrate that the inspections mandated by law spread bacteria, even though it makes the claim. The regulatory well thus poisoned, Tucker tries to answer his own question regarding what would keep food and drugs safe if there were no FDA. It will surprise no one that he makes the tired old libertarian argument that the magic of the free market will do it:
Now to the original question: what would regulate these industries if the FDA were not around at all? In electronics and business quality, we find the answer. These two industries face strict control emerging organically from consumer demand alone.
Institutions like Underwriters Laboratory and the Better Business Bureau are titans in these sectors with no assistance from government. User ratings emerging in the digital era have a vast impact on sales success, as any Amazon seller can tell you. And in industries like sports, household construction, and driving skill, private insurers exercise a dominant influence through financial carrots and sticks, as directed by actuaries assessing risks.
The very existence of the FDA has crowded out such elaborate and complex systems in the case of food and drugs, which is precisely why their safety and efficacy is the subject of such huge public controversy.
Of course, the rebuttal to such ahistorical nonsense, a delusion so ridiculous that only die-hard libertarians can entertain it as potentially true, is to ask: Why hadn’t such a system arisen organically, thanks to the magic of the free market, before the government passed The Biologics Control Act of 1902 and the Pure Food and Drug Act of 1906? Why hadn’t, as Tucker suggests would happen in a free market, reputational damage due to selling of unsafe products led to vaccine manufacturers and meat packers going out of business, thus demonstrating the lack of a need for these laws? As for the idea that insurers would take up the burden by deciding which drugs they will and won’t reimburse for, this, too, is ludicrous, because, after all, where do the data come from upon which insurers base their coverage decisions?
I also like his referring to “user ratings emerging in the digital era,” because it allows me to point you to my previous rebuttal of such nonsense that I wrote in 2017 in response to a potential FDA Commissioner candidate having advocated online reporting systems for drug efficacy and safety that he likened to a “Yelp for drugs.” The CliffsNotes version is that, for products and services like restaurants, rides, or home improvement, it’s very much apparent to the average end user whether the service “worked.” Think about it. Either the meal was good, or it wasn’t. Either it was worth the price to the purchaser, or it wasn’t. Either the service was good, or it wasn’t. Either the ride showed up on time and got you where you wanted to go at a reasonable price, or it didn’t. Either the home repair company fixed your home’s problem, or it didn’t. People can judge these things without specific special expertise. For drugs and devices, that’s just not so easily done. Whether a drug is working or not is not necessarily obvious to the user (although it can be sometimes). Also, thanks to something we discuss very frequently here at SBM, whether a drug “works” or not can also be affected by patient expectations and placebo effects, hence the need for randomized, double-blind clinical trials, often placebo-controlled. The idea of online user ratings for drugs as a means of verifying efficacy and safety is ludicrous on the surface.
Even Tucker himself can’t answer his own question, but he has a magical faith that the FDA is Evil:
Today there are many efforts on the part of nonprofits working to educate the public on many topics related to individual and public health. They work at cross purposes with these agencies that have worked to foist onto the markets what the market would have otherwise been inclined to either filter out, deprecate, or reject entirely.
Could we do without the FDA? We would likely be far better off.
How does Tucker know this? He doesn’t. He simply expects that his magical mystical all-powerful invisible hand of the free market will do it.
During the height of the pandemic and the rise of antivax disinformation targeting COVID-19 vaccines, I used to like to say that “everything old is new again.” What I meant at the time was very simple. None of the misinformation and conspiracy theories arising to spread fear, uncertainty, and doubt about COVID vaccines was anything new. All the fear mongering antivax narratives about COVID vaccines could be traced back to old antivax disinformation, even the claim that there was “DNA contamination” in the vaccines. Although the margins might change based on specific circumstances and diseases, the core of the false narratives about vaccines never changes.
The same is true of the narratives behind the “health freedom” movement. To libertarian “health freedom” warriors like Weissmueller and Tucker, the FDA is always evil and aligned with Big Pharma. Things would always be better without the FDA, and without the FDA the invisible hand of the free market would come down from on high to magically guarantee that our food, medicines, and vaccines are safe and effective. In reality, it takes some serious revisionist history to make such a deluded argument, as Weissmueller and Tucker demonstrate. The FDA before Trump’s second administration was far from perfect, and there are perfectly reasonable arguments to be had based on science, medicine, economics, and ethics about the appropriate balance between regulatory speed versus drug safety, but that is not what either of these health freedom warriors are offering. Moreover, what history clearly tells us is that, absent reasonable government regulation, all that “health freedom” really means is the freedom of quacks, charlatans, and fraudsters to make money conning the public without pesky government interference, which is clearly what the result of Weissmueller and Tucker’s preferred policies with respect to food and drug regulation would unleash.