Politicians and patients are grappling with what to do about high prices for prescription drugs.
To identify a solution, it helps to understand the cause of the problem. A new paper in MIT Press Journals by Professor David Ridley, an economist at Duke University’s Fuqua School of Business, shows that one critical problem is the way Medicare reimburses doctors for drugs. Working with Chung-Ying Lee of National Taiwan University, Ridley studied drug launch prices and found evidence that the system implemented in 2005 has caused a dramatic increase in drug launch prices.
Ridley discusses the pair’s findings, and how Congress could address the problem, in this Fuqua Q&A.
How does the system drive up drug prices?
When health care providers administer drugs to seniors, the providers are reimbursed by Medicare. Since 2005, Medicare has reimbursed health care providers for the drugs they administer based on the average past price of the drug. Because they’ll be paid back either way, most doctors don’t care enough whether an injectable drug they use costs $1 or $1,000. Doctors might be offended by high prices, and they do have to pay more upfront because they need to buy and store the drug prior to use, but they’ll get that money back – and the reimbursements are not based on today’s price, but the price from months earlier. So they get more money today if the price in the past was higher.
Knowing how reimbursement works, manufacturers set high launch prices.
The problem extends beyond Medicare, because private insurers mimic Medicare. It’s easier for providers to understand their reimbursement if it’s similar to Medicare’s, and following Medicare’s lead gives insurers political cover.
It’s well known that Medicare can’t negotiate drug prices. This is not such a problem in the retail drug space, because private insurers negotiate on behalf of Medicare. But it’s a huge problem for the drugs that doctors and other providers administer in an outpatient setting.
Public opinion is the only constraint on these prices, but it works only in extreme cases. Companies that have been most ignorant of that constraint are the non-traditional drug companies. Martin Shkreli, a former hedge fund manager, increased the price of Daraprim by 56 times and became a pariah. But we can’t depend on public opinion to constrain drug prices.
What’s wrong with higher prices?
The current system is bad for patients who often pay 20 percent of the price. It’s bad for taxpayers, who foot the bill for Medicare-funded reimbursements.
It’s also bad for providers, even though they get reimbursed. For some providers – especially small physician groups – it’s hard to pay such high costs up front, even though they’re going to get paid later, because of the amount of money they have to tie up in it. A change to a system with lower prices would help patients without hurting providers.
High prices are good for drug companies and encourage investment in drug development. However, the prices of drugs bear little relation to their value, so drug development incentives are skewed. Artificially high prices in one area lead to lopsided investment.
How did this happen?
The current system replaced another flawed system. Medicare leaders have suspected for many years that we got the fix wrong. Some have proposed experimenting with different mechanisms. But small experiments won’t work. To move the market you need to implement changes at scale. That’s why we wrote this paper, to highlight flaws in the current system using theory and evidence.
What are the alternatives, and their tradeoffs?
The White House is proposing that the price of a drug should be based on low prices in other countries. This is called reference pricing. The likely outcome of U.S. reference pricing is that drug companies will raise prices in other countries. Perhaps actual prices will be flat and list prices will rise in other countries, so the harm might only be to make prices less transparent. But reference pricing would provide little help for Americans and could harm others.
Another proposal to reform reimbursement is to adopt the drug formularies that insurers use in the retail market. Yet another proposal is to not pay separately for drugs, but pay a fixed amount for a bundle of care regardless of which and how many drugs are used.
My preferred mechanism for reimbursement of provider-administered drugs is to tie the price of a drug to its value.
The British pay for drugs based on value, measuring value by the quality of life a drug provides or how long it can extend life. My only complaint is that the British value a year of life at only about $50,000. In the U.S., I think we should assign a higher value to a year of life.
Value-based payment would drive down the prices of many provider-administered drugs, especially those with lower value. For patients, this would mean lower copayments, lower premiums and lower taxes.
Doctors and hospitals will be worried about being reimbursed less. They will have to trust that drug prices will come down. Naturally there will be nervousness until it happens, but drug prices should fall dramatically within a year, especially for lower-value drugs.
We don’t have to adopt the British system. But we have to move away from reimbursing doctors and other health care providers based on their costs. Reimbursement based on costs leads to higher costs.
This issue is not unique to drugs. There are broader questions about how much we should reimburse health care providers in general. Value-based reimbursement is a huge part of our focus at Duke’s Margolis Center for Health Policy. We should move faster toward paying for health care based on its value.