The following originally appeared on The Upshot (copyright 2018, The New York Times Company). It also appeared on page B7 of the print edition on December 11, 2018.
Americans are generally uncomfortable with pharmaceutical prices — which are the highest for brand drugs among wealthy nations — and with drug companies’ profits. But if policies were adopted to reduce drug prices, could there be negative consequences? On balance, would such policies be good or bad?
I asked three health policy experts to consider these questions in the context of four specific drug pricing policies.
Most Americans, including those on Medicare, get drug coverage from private insurers, which negotiate prices with drug manufacturers. (Insurers often outsource that task to pharmacy benefit management companies.)
In a competitive market, insurers that drive the hardest bargains should be able to reduce premiums and cost-sharing. That would then attract more people to enroll, increasing insurers’ revenue.
But “consumers turn out to be fairly bad at shopping and do not respond to price decreases,” said Fiona Scott Morton, an economist at the Yale School of Management. She was an author of a study showing that consumers in Medicare tend to stick with whatever plans they originally selected, even as premiums rise.
Aaron Kesselheim, an associate professor of medicine at Harvard Medical School, said: “A key disadvantage of Medicare’s drug benefit is that it has no real system in place for holding down drug prices. Drug manufacturers with monopoly products can raise prices to whatever extent the market will bear.”
Medicare rules that require insurance plans to cover all drugs in certain classes, including for cancer, help undermine plans’ ability to negotiate prices downward. Medicare “can’t walk away from the table” and refuse to cover a drug, he said.
Nicholas Bagley, a professor of law at the University of Michigan, pointed out that for many drugs — most commonly, generics — competition has lowered prices. “But for drugs that lack clinical substitutes — including some patented drugs and complex biologics — competition doesn’t work,” he said.
This is by design. Patents that the government provides to drug companies necessarily create monopolies. Therefore “we need another solution if we want private plans to get lower prices,” Mr. Bagley said.
Other government programs, including Medicaid and the Department of Veterans Affairs, obtain steeper discounts than Medicare drug plans. In part this is because of government regulations that mandate price reductions. But those programs also negotiate directly with manufacturers, obtaining additional discounts. For example, the V.A. pays about 40 percent less for drugs than Medicare drug plans do.
“The V.A. obtains larger discounts in part because it can — and does — institute a more restrictive formulary than Medicare, meaning some drugs that are more costly but no better than alternatives are not as easily obtained in the V.A. as they are from a Medicare plan,” Dr. Kesselheim said.
Ms. Scott Morton said: “Direct negotiation by Medicare would only be effective if it could say, ‘No, we won’t cover your product’ to a drug manufacturer. Is that really plausible? I’m not sure that’s what Medicare beneficiaries would want.”
Medicare could save money if it could negotiate with manufacturers and exclude high-priced drugs. That would benefit taxpayers, but it would mean reduced access to drugs for the program’s current beneficiaries and slower innovation of drugs in the future.
“If Medicare doesn’t pay attention to a drug’s benefits during those negotiations, the pharmaceutical industry won’t have the right incentives to develop the drugs we need most,” Mr. Bagley said.
One way to better align a drug price with its clinical value is to follow the lead of other countries. Some simply refuse to pay if a drug is considered too expensive relative to the benefits it delivers. In Britain, for example, the National Institute for Health and Care Excellence (NICE) advises the country’s National Health Service about which drugs to cover.
The institute considers a drug’s cost-effectiveness, as well as the type of condition it addresses and whether it’s for a particularly vulnerable population. There’s more leeway for drugs administered toward the end of life, for example.
Drug manufactures have cut prices to get their products covered in Britain. In April 2014, NICE initially did not recommend Eli Lilly’s lung cancer drug Alimta on cost-effectiveness grounds. But when Eli Lilly dropped Alimta’s price, it gained NICE’s approval.
“If we stopped covering drugs that are 1 percent better but 1,000 percent more expensive, drug manufacturers would steer their research investments toward more effective drugs,” Mr. Bagley said. “The obstacle is political.”
Ms. Scott Morton said: “The main reason we do not have a NICE-like system in the U.S. is that drug manufacturers lose when insurers know which drug is most cost-effective. It is in pharma’s interest to protect their profits and lobby vigorously against any government body that would reveal which drugs have the highest value.”
Another consideration, according to Dr. Kesselheim: “We still might be willing to pay a lot for extremely effective new drugs.”
That is, if we’re willing to pay according to value, some very high-value medications will command very high prices. “A system with less wasteful drug spending for many other conditions should be able to better afford high prices in those circumstances,” he said.
There is a more nuanced variation on Britain’s take-it-or-leave-it approach. An insurer or public program could use information about the price-benefit trade-off to establish not whether to cover a drug, but how generously.
The insurer or program could make a drug more accessible by cutting cost-sharing or by clearing away administrative hurdles. Often, for a lower price, insurers are more willing to provide easier access, which increases sales of that drug — a trade-off of volume for price.
This happened with the cholesterol medication Praluent, made by Regeneron and Sanofi and approved by the Food and Drug Administration in 2015. It was originally priced at $14,000 per year. At that price, the pharmacy benefit management firm Express Scripts — which manages drug benefits for 25 million consumers — covered the drug only if a patient went through a complex review process.
In March 2018, the Institute for Clinical and Economic Reviewestimated that a cost-effective price for the drug would be, at most, $8,000 per year for high-risk patients.
In exchange for more favorable coverage by Express Scripts, Regeneron and Sanofi agreed this year to lower the price of Praluent to the level the Institute for Clinical and Economic Review recommended. More patients will have access to the drug, and it will cost insurers less per patient than originally priced.
“This kind of value-based insurance design is the holy grail, at least for new drugs,” Mr. Bagley said. “But we wouldn’t want to pay a value-based price for aspirin. Doing so would cause us to pay a lot more for it than we do today.”
Dr. Kesselheim said: “One reason imposing higher cost-sharing on less effective drugs isn’t as successful as it could be today is that drug manufacturers undermine the approach. They offer coupons to patients that counteract increased cost-sharing levels or provide physician offices with strategies to circumvent prior authorization paperwork.”
“Drug manufacturers offering financial assistance is an illegal kickback in Medicare,” Ms. Scott Morton said. “We would likely lower drug costs by extending that rule to the private sector.”
Ultimately, all these approaches have limitations. “There isn’t one best model for drug pricing,” Mr. Bagley said. “What works for generics won’t work for patented drugs, and we shouldn’t pay for antibiotics the same way that we pay for cancer drugs.”
That’s why it’s hard to boil realistic solutions down to a bumper-sticker slogan. Balancing prices and access to drugs for the patients of today with the innovation that will benefit those of tomorrow will take ingenuity, as well as a lot of political will.