Commentary: Indian-American physician says drug costs force tough choices on doctors, patients


How much should your medicine cost?

A physician colleague recently told me the story of a patient who asked if she could put off taking her medication until the end of the month, after she received her paycheck. The price had risen and she didn’t have the money to pay for it along with her other bills. Unsure how to respond, he gave her $20 to cover part of the cost.

Then, he didn’t hear from her for months.

She missed appointments, she wouldn’t answer his calls. When she finally resurfaced, he asked why she hadn’t come back to the clinic. She was embarrassed, she said, because she couldn’t pay him back – until now. She handed him a crisp $20 bill.

As doctors, every day we’re faced with decisions about which of a slew of medications we should use. Would my patient do better with this statin, or that one? Should I use the old antibiotic or a new one that costs twice as much? Often, these decisions have as much financial significance as medical significance. But there’s little guidance on how patients and doctors should think through them.

Consider ticagrelor. It’s a blood thinner used to prevent heart attacks after a patient has a stent placed. Compared with an older blood thinner called Plavix, ticagrelor lowers the risk of recurrent heart attacks by about 2 percent. It costs 25 times as much.

Contrast this with Sovaldi. It’s a new drug with minimal side effects that cures most patients with hepatitis C – a disease previously treated with a cocktail of minimally effective, highly toxic medications. When introduced, Sovaldi cost nearly $90,000 for a 12-week course of treatment.

Should it matter that for one drug, a huge price tag comes with large health gains, and for the other, it barely moves the needle?

Most would agree that the health and economic implications here are substantial, but Medicare and private health insurers often don’t factor them into decisions about whether to cover a new drug – or how much to pay for it. Medicare, for example, is required to cover any cancer drug deemed “safe and effective” by the Food and Drug Administration (FDA) without considering whether its price is commensurate with the value patients receive.

This means a new cancer drug that extends life by a few weeks – and costs tens of thousands of dollars more than the generic – must be covered. Current law prohibits Medicare from directly negotiating prices with drug manufacturers, even though more than 90 percent of Americans think it should. The FDA’s recent shift to expedite drug approval by requiring fewer, smaller clinical trials and using short-term markers for success instead of hard clinical outcomes means that more drugs of marginal value and huge cost may come to market. It also means that patients and taxpayers will be on the hook to pay for them.

Some experts have called for a new pricing scheme in which a drug’s price is more closely tied to the value it provides: Does it work – defined by longer or better life – and if so, how much better does it work compared to existing drugs? “Value-based” pricing sounds simple, but can quickly get complicated. How exactly should we pay for “value”?

“There’s a general sense that drugs cost a lot, but not a lot of specificity on what to do about it,” said Peter Bach, director of the Center for Health Policy and Outcomes at Memorial Sloan Kettering Cancer Center. “Value-based pricing is one way to anchor a drug’s price to its benefit but it’s important to be clear about what we mean by it.”

We could use “outcomes-based pricing” by offering refunds if a drug doesn’t work for particular patients. But this doesn’t help the vast majority of patients who are still stuck paying high prices. For example, Amgen’s proposal to provide refunds to the approximately 7 percent of patients who have a heart attack or stroke on its cholesterol-lowering drug, Repatha, would lower the net price of the drug from $14,000 to $13,620 per year. (Amgen recently announced it would cut the price of Repatha to $5,850, amid growing criticism of escalating drug prices and out-of-pocket costs.)

We could use “indication-based pricing” – paying more for a drug when it’s used to treat a medical condition for which the benefits are large, and less when it’s used for a condition for which it offers little. We could use “mortgage pricing” – yes, like buying a home or an iPhone – for drugs when their benefits accrue slowly over time. Or we could waive co-payments for drugs with big benefits and demand higher co-payments for drugs that are marginally effective.

All of these proposals have been tried in some form and all have limitations – principally because they tinker with how to pay for a drug instead of grappling with the more fundamental question of how much a drug should cost to begin with. Even the Trump administration’s recent proposal for Medicare to pay prices similar to other high-income countries for certain medications would do little to fix the underlying problem of drug prices disconnected with drug value.

One starting point for valuing new drugs is a calculator developed by Bach called DrugAbacus. It’s a user-friendly calculator that allows policymakers (or patients or doctors) to enter how much they value various characteristics of a drug: How novel is it? How bad are the side effects? Does it treat a rare illness?

If a drug works through a mechanism never seen before, it gets points for novelty. But if the side effects are severe – if you cannot take it without vomiting or itching or falling asleep – it gets docked points. By deciding how much we should value each of the calculator’s domains, we can estimate the “value-based” price for a given drug.

“Opponents have said that valuing a drug in quantitative terms is too hard, that it’s like trying to quantify love,” Bach said. “But the math is easy. The harder question is what aspects of a drug we should value most. How much you pay for particular attributes – for example, how well-tolerated it is or how prevalent the treated condition is – will affect the type of innovation you get.”

The problem of high drug prices – or more precisely, prices misaligned with the value patients receive – is increasingly important at a time when patients are being asked to pay more out-of-pocket and policymakers are struggling with how best to lower drug prices. A better understanding of how to value new drugs – what factors should be considered and how they should be weighed – could empower patients, doctors, and policymakers to decide which medications are worth the cost and which cost more than they’re worth.

– – –

Dr. Dhruv Khullar of New York Presbyterian Hospital. (Photo:

Khullar is a physician at NewYork-Presbyterian Hospital, a researcher at the Weill Cornell Medical College’s Department of Healthcare Policy and Research, and director of policy dissemination at the Physicians Foundation Center for the Study of Physician Practice and Leadership.



Please enter your comment!
Please enter your name here