The following originally appeared on The Upshot (copyright 2015, The New York Times Company).
When anyone proposes reducing prescription drug prices — as Hillary Rodham Clinton and Bernie Sanders recently have — the most commonly heard criticism is that it would squelch innovation. But not all pharmaceutical innovation is valuable. Though some drugs are breakthroughs, some offer only marginal benefits at exorbitant cost.
There is a way to keep prices low while encouraging drug companies to innovate: Look to Europe and elsewhere, where drug prices are a fraction of those in the United States. Germany, Spain, Italy and a half dozen other countries have pushed drug prices lower with a system called reference pricing. It has led to drug price decreases and significant savings in the Canadian province of British Columbia as well as in Germany, Italy,Norway, Spain and Sweden. A study published in the American Journal of Managed Care found that price reductions ranged from 7 percent to 24 percent.
Here’s how it works: Drugs are grouped into classes in which all drugs have identical or similar therapeutic effects. For example, all brands of ibuprofen would be in the same class because they contain the same active agent. The class could include other nonsteroidal anti-inflammatory agents like aspirin and naproxen because they are therapeutically similar. The insurer pays only one amount, called the reference price, for any drug in a class. A drug company can set the price of its drug higher, and if a consumer wants that one, he or she pays the difference.
Setting the reference price low enough puts considerable pressure on drug manufacturers to reduce prices for drugs for which there are good substitutes. If they don’t, consumers will switch to lower-cost products. In British Columbia and in Italy, the reference price is set at the lowest-price drug in the class; Germany uses an average price across drugs; Spain also uses an average, but only of the lowest-priced products that account for at least 20 percent of the class’s market.
In pushing prices down, reference pricing doesn’t suppress innovation; it encourages a different form of it. The market still rewards the invention of a cutting-edge drug with novel therapeutic effects. Such a drug might be placed in a new class and therefore could be priced high. But, within classes, the market also rewards innovations that lead to lower-priced drugs, because consumers switch to them to avoid out-of-pocket costs. In these ways, reference pricing promotes cost-effectiveness.
Consider, for example, the price of new anti-cholesterol drugs known as PCSK9 inhibitors: about $14,000 a year. A recent report from the Institute for Clinical and Economic Review (ICER) received considerable attentionwhen it argued that the drugs were priced too high for the value they offered patients. Reducing the prices to close to $2,000 would make them both cost effective and would help keep American health spending below a widely accepted growth target, according to ICER’s analysis.
Reference pricing could help drive down the prices of the PCSK9 inhibitors if they were put in the same therapeutic class as other, cheaper generic cholesterol drugs, like statins. If this happened, PCSK9 manufacturers — Amgen and Regeneron Pharmaceuticals — would face powerful incentives to reduce their prices.
However, some people might reasonably argue that PCSK9 inhibitors are superior to statins and therefore should not be grouped with them. Because ICER’s price is based on cost-effectiveness, it incorporates such performance differences by recommending a higher price for more effective drugs, though in the case of PCSK9s, a lower price than the manufacturers may want.
The promise of reference pricing goes beyond prescription drugs. In apaper presented at Brookings this month, the Harvard economist Amitabh Chandra, the University of Michigan law professor Nicholas Bagley and I proposed extending the approach to a wider range of medical technologies.
We suggested that Medicare should pay more for a new therapy for a given condition only if that new therapy is better than existing therapies. (In 2010, David Leonhardt wrote about a similar idea.) In no case, we proposed, should Medicare pay more for a therapy than a generally accepted cost-effectiveness standard. If patients wanted cost-ineffective therapies, they could pay the difference out of pocket, a departure from current Medicare policy.
As it stands, other countries are far ahead of the United States in pricing drugs to promote cost-effective pharmaceutical innovation. But interest is growing here in new approaches. Peter B. Bach, a physician at Memorial Sloan Kettering Cancer Center, recently proposed a variation on reference pricing that considers how the cost-effectiveness of a cancer drug varies by what disease it is used to treat. He noted that the drug Tarceva costs the same whether it is used to treat patients with a kind of lung cancer or patients with pancreatic cancer. But the results are wildly different. On average, Tarceva extends a lung cancer patient’s life by just over three months; it extends a typical pancreatic cancer patient’s life by a mere week and a half.
Dr. Bach’s insight is that we should be paying for what we care about — life gained — not the drug itself. He therefore proposed that the price of Tarceva be sharply reduced for pancreatic cancer patients to bring the cost per duration of life gained in line with that of lung cancer patients.
Critics of Dr. Bach’s idea, ours and the approach of ICER claim they would restrain innovation that could benefit patients. However, they are devised specifically to reward smarter innovation, which is precisely what we need.
