The Trump Administration has been devolving into chaos with surprising rapidity in the last week, with it’s travel ban suspended, it’s labor nominee going down in flames, Michael Flynn resigning in a scandal that may yet snowball to Watergate proportions, and Trump giving a press conference while apparently in a fugue state.
But it’s equally important to note all the ways the Trump Administration is screwing things up that don’t make their way into the news cycle. And one story which hasn’t gotten too much attention, but which potentially has significant and long term ramifications, is the Trump Administration’s desire to shake up the FDA by slashing the agencies regulations by 75%-80%, presumably with the end goal of ending speeding up the drug approval process.
More than that, though, he wants to fundamentally change the nature of the FDA. Currently, the FDA’s drug approval process essentially has two functions: first ensuring drugs are not safe, and second ensuring that drugs actually work. By contrast, Trump, the Republicans, and various silicon valley moguls that Trump has reportedly been considering to head the FDA, want to eliminate that second function by junking the FDA’s efficacy requirements and instead having the FDA focus strictly on safety while having markets, or some type of rating system from doctors and patients, sort out whether a drug works or not. Putting a veneer of Silicon Valley libertarianism on the idea, one of Trump’s prospective (though later withdrawn) nominees, Balaji Srinivasan, had said that he wanted to turn the FDA into “the Yelp for drugs”, and repeatedly said he would like to make the agency more like tech companies like Uber and Airbnb.
Now, I spent several years recently in China, studying the numerous problems afflicting it’s drug approval process under the CFDA, so suffice it to say that I take a personal interest in the ins and outs of effectively the entry of drugs into the market. And yes, all else equal, speedy approval time is a good metric of a successful drug approval process. Conversely, backlogs and long approval times betray ineffective government, stifles domestic innovation, and can potentially kill thousands by denying them access to life saving drugs.
That said, though, Trump’s proposed shake up of the agency is not only wrongheaded, it’s downright alarming.
First off, they’re trying to “fix” something that isn’t broken
It’s important to note that they’re addressing a non-existent problem. By international standards The United States already has a drug approval process which is already a model of efficiency. Drugs can get more than a month faster in the US than in other industrialized countries, with a full review by the FDA taking on average 322 days compared with 366 days in Europe. Compare and contrast that with the genuinely defective drug approval process in China, where it usually takes between 3-5 years for a drug to gain approval. And generally speaking, backlogs and long approval times have less to do with bureaucratic hurdles and more to do with staffing shortages.
Second, user rating systems are a horribly inappropriate template for drug review
Now, you may be tempted to write the idea of turning the FDA into “Yelp for drugs” as vacuous Silicon Valley technophilia, but it’s actually much worse.
It’s important to note that tech companies are a really bad template for producing quality standards for anything. More often than not, their ratings systems are highly defective. The ones that do work relatively well, like Yelp, deal mainly with consumer facing industries where customers are working off straight forward subjective personal tastes, like restaurants. In that context, a sort of crowd sourced rating system kind of makes sense. But it’s horribly unsuitable for the pharmaceutical industry, which is more or less the exact opposite of that.
The main difference is that there’s nothing subjective about whether or not a drug works. Doctor are able to empirically prove that a drug works through clinical trials, and generally patients are willing to volunteer that information by not dying. All you need, then, is an extensive knowledge of medicine and enough test cases to gain sufficient empirical evidence.
On the other hand, individual patients don’t actually know whether or not a given drug is effective or not because, for the most part, they don’t have the in-depth medical knowledge required to understand the mechanisms of a drug or monitor their own health to that degree. Likewise, individual doctors aren’t in a position to say either. Their only insights into the matter are based on limited numbers of patients who they’re treating in an uncontrolled environment. They don’t have the perspective necessary to make that call. You need long term research based on hundreds or thousands of patients to do that, and you need sometime like the FDA to organize such long term studies.
Third, the proposed system would create all kinds of perverse incentives
The pharmaceutical market place doesn’t operate in a typical market, where you have one firm developing a product and then selling it directly to customers. On the contrary, it’s a multilayered process. On the supply side, you typically start with basic research conducted by one or more public institutions, which then gets taken up by a biotech startup and developed into an product, which is then bought by a larger pharmaceutical company which has the resources to see it through the final stages of testing and then distribute it broadly enough to make a profit. On the consumer side, you have patients who rely heavily on the opinions of doctors when deciding which drugs to take, and insurance companies who reimburse them for the majority of the costs. In this system of six or more players, nobody is in a position to see the complete consequences of their actions, and there are a lot of opportunities for dysfunction.
On the front end, you have biotech startups who are expected to invest hundreds of millions, and potentially even billions, into research and development knowing there’s something like a 97% chance any given project will fail, and even in the best case scenario they won’t be seeing a profit for years. As Harvard Business School professor Gary Pisano pointed out, this is only tenable because, first, biotech startups can rely heavily on public research and resources provided by larger pharmaceuticals, and second because they can raise large amounts of money in IPOs and sales of equity. Now, individual investors don’t actually expect to hold their stake in a biotech startup for the several years it will take for the firm to become profit, nor do they necessarily expect it to succeed. But that’s okay, from their perspective, because as long as they can resell their shares in the biotech startup, they can still turn a profit. Future potential buyers largely work on the same logic, and people can keep buying and reselling that stake in the biotech company until they either bring their product to market (or, more likely, get bought by a large pharmaceutical) or goes bust.
Now, this may sound familiar to anyone who’s studied the subprime mortgage crisis, because it’s essentially the greater fool theory. This isn’t to malign the biotech industry, it’s just a quark of the system that happily tends to work out. However, it does make the system prone to all kinds of fraud it you’re not careful. Founders of biotech companies are under heavy pressure to sell every product they develop as some phenomenal breakthrough, and they have a strong incentive to keep doing so even after they realize they’re ineffective. On the other hand, potential investors aren’t particularly inclined to analyze their investments critically, in fact they have a vested interest in keeping the delusion alive. That’s how the Theranos was able to carry on for years despite having a product they knew didn’t work, at a cost of billions of dollars to all involved. It would have carried on a lot longer had the FDA not been there to point out that the emperor had no clothes.
Meanwhile, on the consumer side, doctors still largely get paid basis of services rendered, and thus have an incentive to prescribe drugs. Even if they’re not acting out of self-interest, they’re not liable for the cost of drugs, literal or physical, so they’ll tend to be very liberal with prescriptions. This is a major defect in the healthcare system which is driving up costs, contributing to the opioid epidemic, and leading to antibiotic strains of bacteria. Beyond this, though, the drug market such that doctors and hospitals do need to maintain a relationship with pharmaceutical companies and wholesalers to maintain access to treatments Hence, they may be loathe to give products bad reviews, lest they be blacklisted in the future. This is to say, there are a lot of reasons individual doctors might not be particularly critical when it comes to drug efficacy.
On the other hand, insurers would have a vested interest in labeling drugs ineffective, but their interest isn’t so much improving treatment as it is finding an excuse to limit coverage. And, if there’s no commonly accepted standard for drug efficacy, there’s no reason insurers couldn’t do this. This is why even major pharmaceutical companies are against the idea of removing efficacy requirements. Whatever inconveniences drug companies face due to the approval process, having a universally accepted for efficacy means they don’t need to argue with insurance companies all the time as to whether or not they’ll reimburse patients who use their products.
Again, this is why you need an impartial actor, like the FDA, to lend a critical eye. Removing efficacy standards would likely just lead to the worst of both worlds, with dubious products flooding the market and insurers limiting coverage.
Fourth, however much you cut down regulations, you can’t reduce approval times if the FDA is a mess
Of course, it’s not just the deliberate ways that Trump and the Republicans are liable to screw up the FDA that are relevant. Whatever your end age may be, you can speed up approval times is you’ve turned the FDA into a mess, and as Vox pointed out last week, that’s exactly what the Trump administration is doing. As I mentioned before, the main reason for backlogs and holdups in the approval process more often than not comes down to staffing shortages. And the Trump administration, with it’s Federal hiring freeze, is already threatening to preempt new hiring that was supposed to take place under the 21st century cures act.
Likewise, the second biggest contributor to a dysfunctional approval process is administrative confusion owing to unclear policies. Again, the Trump Administration seems set to sow all sorts of confusion with broad, impractical policy directives, like his call to eliminate 75-80% of FDA regulations and his mandate to eliminate 2 regulations for every new one created. And of course, the administration’s antagonism towards the civil service, and it’s haphazard attempts to impinge upon their autonomy, is sure to have a demoralizing effect on the professionals who work at the FDA.
Finally, taking the hatchet to the FDA may have broader implications for the world at large
The impact of stripping down the FDA may even go beyond screwing up the American drug market. FDA approval doesn’t just mean access to US markets, it’s also one of the few ways drugs can gain expedited approval in countries with less effective drug regimes. What does it mean for companies who have been able to gain access to Chinese markets largely by virtue of gaining FDA approval when suddenly they find that approval has become meaningless?
More than that, the FDA is also taking an increasingly active role in quality control of drugs coming from abroad. Since 2010, the number of FDA inspections conducted in China and India, the source of the vast majority of active ingredients for pharmaceuticals, has roughly tripled. And for good reason, there’s a counterfeit drug epidemic. Will these functions be cut as a frivolous expenditure under Trump.
Some may imagine that Trump and the Republicans are intentionally trying to break the FDA out of free market dogma. The FDA may not be perfect, and maybe it can be improved upon. But it’s prevented quackery effectively for more than 100 years while not only allowing, but enabling the US to become a leader in the field of pharmaceutical R&D. It really is a key stone in the drug market, and in an era of counterfeit drugs and bacteria which are resistant to anything we can throw at them, it’s more important than ever. Even pharmaceuticals companies agree that removing the reshaping the system in the way Trump is proposing would almost certainly be a disaster.
FDA efficacy requirements may not be the sexiest issue, but it is a vitally important one, and we should be ready to hold the line.