The following is a guest post by Yevgeniy Feyman, a senior research assistant in the department of health policy and management at the Harvard T.H. Chan School of Public Health. His research focuses on a range of topics including hospital quality and Medicare Advantage. Previously, he was deputy director of health policy at the Manhattan Institute for Policy Research.
The first post in this series made an argument that drug pricing should be tied to value and offered an idea for how we might think about paying for curative therapies. In this post, I’ll focus on pricing ideas for innovative drugs that aren’t curative, as well as generic drugs.
Innovative, but non-curative drugs
Not all drugs are cures. Many drugs treat late-stage diseases, illnesses with few other treatments available, or are just simple maintenance therapies that do everything from keeping cholesterol to blood sugar in check. Some of these might be “innovative” in the full sense of the word, using the latest scientific developments to offer incredible benefits – Gleevec, for instance, was a revolutionary treatment for leukemia patients with a particular genetic biomarker. Other drugs, like the much hyped PCSK9 inhibitors geared towards statin-intolerant patients might fall into a grey area – they’re highly innovative, but the clinical benefits have yet to be seen.
As with curative therapies the goal should be to pay for value. But here, value might be much more relative. In the case of Sovaldi, it would make little sense to tie payment for the drug to the average cost of existing treatments for instance. That would grossly undervalue the therapy and would ignore its benefits.
A drug that extends a cancer patient’s life by a few weeks, however, is a different story. The value of this kind of late-stage drug might be better assessed in relation to existing options. A system that guaranteed insurers a price referenced to existing therapies, for instance (as some European countries do), would still incentivize innovation – it would just underscore that there is a limit to how much we’re willing to pay for any given innovation.
But there are other approaches too. For instance, if we agree on the value of a quality-adjusted life year for different disease states — a big “if” — we could use that dollar value to benchmark drug prices. This would help make drug pricing more consistent, incentivizing drugs that add more value. Some might raise a reasonable objection here – assuming a uniform value of a life-year could mean that treatments for patients with late-stage (or hard-to-treat) diseases fall below this value threshold. That’s why we might want this this threshold to vary across diseases. Simply put, we may not want to apply the same value threshold to late-stage cancer treatments, for instance, as we would to diabetes treatments.
Yet another approach might tie drug payments to specific, expected outcomes. A drug that promises a reduction in cholesterol that achieves only half that reduction might receive half the payment, for instance. Similarly, a drug approved using non-clinical endpoints (surrogate endpoints, like tumor size for instance), might justify a lower price than one approved based on clinical endpoints due to the uncertainty about actual patient outcomes. Some payers are already entering into outcome-based payment models – Cigna is doing so with a new anti-cholesterol drug, and a heart failure drug, for instance. The point is simple – there must be some limit to what we as a society are willing to pay for a given outcome.
Expensive generic drugs
Generic drugs are a different animal altogether from innovative drugs. After a patent for a drug expires, other companies are able to make their own replicas of that drug. The pricing here is different too – there’s no concern about incentivizing innovation in generics. There’s little R&D involved in getting a generic drug to market, and there’s nothing novel about the drug (by definition). This means that, we should be less concerned about the incentives for innovation –regulatory action won’t be as potentially harmful as with innovative drugs. But, since the exclusivity period is over, it means that competition also can be an effective tool in keeping prices down in the generic market.
Generally, after generics hit the market drug prices fall significantly, with more competitors pushing the price down further. But recently, there’s been some public concern about generic drug prices in instances in which competition is insufficient. Turing Pharmaceuticals’ decision to raise the price of Daraprim by 5,000 percent is probably the most widely known. (The more recent EpiPen debacle underscores the importance of competition, but is more complicated than the Daraprim issue.)
Despite these high-profile cases, the generic market seems to be functioning mostly as intended. An ASPE report looking at Medicaid FFS spending on generic drugs found that nearly 65 percent of drugs from 2013 to 2014 experienced a price decrease. A little over 14 percent of generic drugs experienced price increases of over 20 percent. Why the generic market malfunctions on occasion can vary, but likely has to do with a lack of competition.
Drug shortages, for instance – which can happen for a variety of reasons – can allow manufacturers to demand higher prices. Regulatory backlogs matter, too. Up until recently, the FDA had a significant filing backlog for generic drugs. This may no longer be an issue — the filing backlog has cleared. But “filing” is only the first step in getting a generic drug to market. A manufacturer also needs FDA approval, which might still take a while – generics that launched in 2015 took close to four years for approval. The FDA’s recent commitment to a 15-month review cycle for generics is intended to speed that up; whether they succeed is yet to be seen.
The basic point is that action to stem generic price increases might focus on fixing problems in the market rather than explicitly regulating prices. The Daraprim case, for instance, might have had a role for antitrust litigation. More generally, it might be wise to expedite generic approvals for companies with strong track records for drug safety, in this case approving the manufacturer rather than the specific drug. Overall, though, it isn’t clear that the generic drug market is in any real danger of massive price increases with a few notable exceptions.
Thinking about how to reform drug pricing first means accepting two realities: 1) Innovative drugs will be expensive, and that’s OK, to a point. 2) The prices we pay for innovative drugs cannot be unlimited, and must still be tied to value. The ideas above would help move us closer to a system that explicitly pays more for valuable drugs – though they’re far from exhaustive.
Over at Forbes, for instance, Matthew Herper recommends more transparency on drug prices, faster approval of similar drugs that have seen large price hikes, among other ideas. Peter Bach, of Memorial Sloan Kettering, has called for indication-specific pricing of cancer drugs. Elsewhere, Paul Howard (my former colleague at the Manhattan Institute) has argued for allowing Medicaid programs engage in bond offerings to help pay for curative therapies, similar to the “health care loans” idea.
A unifying theme among these ideas is that they don’t treat all high priced drugs equally. They recognize that high prices arise in different subsectors for different reasons. In doing so, they move the conversation beyond the blanket “price controls” vs. “no price controls” talking points.