The FDA Debate, and the Infant Fish Oil Saga

Scott’s article Adumbrations Of Aducanumab went viral, which takes aim at the FDA for purportedly delaying potentially life-saving treatments.

Most people seem to be in agreement with Scott that the FDA is too inefficient, too slow, and too stringent. Yet there needs to be some sort of impartial filtering mechanism for separating the wheat from the chaff, but a purely decentralized option, such as customer reviews, would seem to be a bad idea for matters pertaining to possible life or death, especially given how easy it is to manipulate Amazon reviews, for example.

The libertarian argument is that the FDA is too stringent, but all drugs cost taxpayers money. A more lenient FDA means more taxpayer money spent on drugs that may not work. But an actual libertarian solution would be no FDA but also no Medicare and Medicaid. Or drastic tort reform, as Scott himself describes:

I don’t understand the exact legalities of this, but it seems like Medicare and Medicaid have to cover anything the FDA approves. The situation with private insurances is more complicated but still not great. My guess is that if a private insurance covers an Alzheimer’s patient, and a doctor says that aducanumab is “medically necessary” for that patient, and the insurance doesn’t cover it, and the patient’s Alzheimer gets worse, that patient can sue the insurance company for failing to provide standard of care. What makes it standard of care? Because the FDA approved it, and Medicaid and Medicare are giving it to their patients, of course!

I dunno what the ideal balance is. No one wants to waste money, especially taxpayer money, on drugs that don’t work or are even harmful, yet no one wants to see people die or get sicker due to bureaucracy. But given how much healthcare costs have surged over the past few decades, it’s evident Americans are paying and spending too much on healthcare, and I don’t see how relaxing FDA restrictions would help in this regard unless approved drugs were super-effective at treating disease and saving money, but the evidence suggests most drugs are not.

I think Scott overestimates the efficacy of drugs. Being a doctor, he wants to help his patients get better, and drugs are a way of achieving that. But the examples he gives about infant cholestasis and AIDS anti-viral drugs are cherry picking. He also overestimates, IMHO, the efficacy of the Covid vaccine at preventing Covid and saving lives. With Delta and other variants, who were previously vaccinated are getting sick and spreading Covid. Maybe the vaccines have saved a hundreds of thousands of lives, but even so, it would still be cherry picking on Scott’s part.

Scott a few days later posted an update in regard to the infant fish oil story. Although the FDA would not approve Omegaven until 2018, exceptions could be made on a case by case basis, and no infants were actually denied Omegaven, nor did doctors have to smuggle it in. The nice thing about having thousands of readers is you get to outsource all your fact-checking, but ideally the fact-checking would occur before the article is published and not after.

Regarding Covid vaccines being delayed, I think the FDA was overcautious because the stakes were very high , and if the vaccines did not work, then the credibly would be ruined and there would be no way of doing it over, so there was only once chance to get it right. Already with the Delta variant the vaccines are not stopping the spread of the disease.

A common narrative you sometimes see is that the FDA is denying potentially life-saving treatments, and that thousands of lives would be saved if the FDA were faster or didn’t exist. But some of the most costly FDA-approved drugs do not work that well. They do not cure disease or add many years of life, but just prolong survival for a few more months, maybe a year, at a great financial cost, not to mention the side effects of the drugs. The examples Scott gives of FDA approved drugs saving lives or adding decades of life (such as Omegaven), is more of an exception than the norm.

Contrary to Scott’s thesis, the evidence suggests the FDA, due to public pressure, is approving too many drugs:

Overall cancer survival has barely changed over the past decade. The 72 cancer therapies approved from 2002 to 2014 gave patients only 2.1 more months of life than older drugs, according to a study in JAMA Otolaryngology–Head & Neck Surgery.

And those are the successes.

Two-thirds of cancer drugs approved in the past two years have no evidence showing that they extend survival at all, Prasad said.

The result: For every cancer patient who wins the lottery, there are many others who get little to no benefit from the latest drugs.

Inaction is perceived as failure, so approving ineffective drugs is perceived as progress even if the drugs don’t work.

Without fail a desperate patent with terminal cancer who enters a clinical trial, gets sicker and dies. In over a decade of following this stuff, I have not once, ever, on any forum, seen someone with terminal cancer who enters a clinical trial be cured. At best it’s a few extra months or life, which is no better or even worse than the placebo/control group.

That is not to say clinical trials cannot yield good drugs, but it’s just so rare that when it happens it is literally a once-in-a-quarter-century occurrence or even rarer. People with advanced colon cancer for example are treated with the same cocktail of drugs as 15 years ago. The last major breakthrough for treating metastatic colon cancer was Avastin (which is not chemotherapy)…which was in 2003.

The drugs which do work, which by ‘work’ adding a up to a year or so of survival for someone with terminal cancer, are extremely expensive. It’s not like cancer patients are writing checks to hospitals and drug companies, but rather it’s paid for by insurance. So more such drugs are presumably funded by higher premiums for healthy people. If everyone is going to be subsidizing the sick, it may as well be for treatments that will do more than add up to a few months of life.

Note: this is for the drugs that make it through the FDA gauntlet, so you can only imagine how bad the drugs that fail are, if a few months of added survival time is considered a ‘success’.

Regarding the fish oil example again, it’s not like the FDA is necessary: people can and do seek treatments that are outside of the purview FDA. The FDA only pertains to domestically manufactured and sold drugs, but patients and doctors can avail themselves of imported drugs on a case by case basis. If I had to reform the FDA, I would reduce restrictions for inexpensive but effective drugs, like the examples Scott lists, but be much less inclined to approve expensive drugs that do not add much marginal benefit.