By Rachel Sachs
In recent days, it seems like the din of voices arguing that the FDA should approve pharmaceuticals more speedily and on less evidence has grown louder. It is a central theme of the 21st Century Cures Act, which the House may vote on today and which I seemingly will never finish blogging about (most recent post here, with links to previous ones). It is the premise that underlies other legislation recently introduced into Congress. And it was the topic of a Wall Street Journal opinion piece just last week. In the view of these critics, sure, the FDA has some role to play in ensuring safety and some basic level of effectiveness. But the current standard for demonstrating effectiveness is, in their view, much too strict. Instead, we ought to approve drugs more quickly and allow insurance companies and physicians to decide which products have enough supporting evidence to merit reimbursement.
Here’s the problem: that is not the way we’ve set up the system. FDA approval is often linked to insurance coverage. Medicaid must cover essentially all FDA-approved drugs, and Medicare similarly has limited ability to decline to cover FDA-approved drugs. Even private insurers are generally required to cover at least some prescription drugs, although in some cases this may be on a more limited basis. Take Exondys, a drug that recently won accelerated approval from the FDA for the treatment of a small number of patients with Duchenne Muscular Dystrophy (I’ve blogged about Exondys here). Because Exondys was approved based on a surrogate endpoint and not actual evidence of clinical improvement (Exondys’ label actually says that “[a] clinical benefit of Exondys 51 has not been established”), it would seem to be a poster child for these arguments about the FDA. Allow insurers to cover it or not as they choose, since we don’t yet know if it works. Yet many insurers are legally required to pay the $300,000 a year on average the company is charging for the drug.
I’ll put it another way. If we lower the FDA’s approval standards and do nothing to coverage requirements, we will all almost certainly end up paying more money for drugs that don’t work. The pronoun “we” here is important: because an enormous amount of these expenditures will come through Medicare and Medicaid, which are funded by all of us as taxpayers, it costs all of us financially when ineffective or unsafe drugs are approved by the FDA. Many people who argue for a decrease in FDA standards also believe that we spend too much through Medicare and Medicaid, yet they don’t seem to put these two pieces of the argument together.
Of course, there’s a way out of this dilemma. We could decouple FDA approval from insurer coverage. We could give Medicare, Medicaid, and private insurers greater ability to walk away from the table when there isn’t enough evidence for a drug, or when the manufacturer isn’t offering a significant enough discount. However, this means that we’d be restricting access to certain drugs for patients. And patients don’t like that. This is one of the key problems with giving Medicare the authority to negotiate drug prices, a very popular position which has even been embraced by the Republican President-Elect. Without the ability to walk away from the table – and to restrict patient access – that authority is largely meaningless. But it is difficult to build support for a provision which would restrict access in this way.
We actually already have a system for regulating insurance coverage of health technologies that’s decoupled from the approval system. It’s the system we apply to medical devices. After gaining FDA approval, which is typically far less onerous than the process most pharmaceuticals must undergo, device manufacturers must separately apply to insurers (including CMS) for coverage of their products. On its face, this resembles the system many of the above FDA critics would like to see. But medical device companies don’t like the bifurcated system. It creates additional uncertainty for them, as they don’t know whether insurers will accept as sufficient the information they used to traverse the FDA approval process, and performing additional studies can be costly. AdvaMed, the primary industry organization for medical device companies, recently released a report making these arguments and calling for some type of combined review program. (NB There actually is a parallel review program between FDA and CMS, which was recently permanently reauthorized after a pilot period. But the report only mentions it obliquely. If you know why, email me!)
But as long as we’ve coupled FDA approval and insurance reimbursement for pharmaceuticals, Congress cannot alter half of the system and hope the other side magically resolves itself. Congress itself has created this interconnected system, and it is critical that our elected representatives understand that system. Otherwise, well-intentioned efforts to alter one piece may end up causing more harm elsewhere.