• What’s a year of life worth?

    New paper out in Health Affairs, “In A Survey, Marked Inconsistency In How Oncologists Judged Value Of High-Cost Cancer Drugs In Relation To Gains In Survival“:

    Amid calls for physicians to become better stewards of the nation’s health care resources, it is important to gain insight into how physicians think about the cost-effectiveness of new treatments. Expensive new cancer treatments that can extend life raise questions about whether physicians are prepared to make “value for money” trade-offs when treating patients. We asked oncologists in the United States and Canada how much benefit, in additional months of life expectancy, a new drug would need to provide to justify its cost and warrant its use in an individual patient. The majority of oncologists agreed that a new cancer treatment that might add a year to a patient’s life would be worthwhile if the cost was less than $100,000. But when given a hypothetical case of an individual patient to review, the oncologists also endorsed a hypothetical drug whose cost might be as high as $250,000 per life-year gained. The results show that oncologists are not consistent in deciding how many months an expensive new therapy should extend a person’s life before the cost of therapy is justified. Moreover, the benefit that oncologists demand from new treatments in terms of length of survival does not necessarily increase according to the price of the treatment. The findings suggest that policy makers should find ways to improve how physicians are educated on the use of cost-effectiveness information and to influence physician decision making through clinical guidelines that incorporate cost-effectiveness information.

    I don’t want to spend a lot of time on the methods, because this pretty much tells you all you need to know. When asked, most oncologists thought that a drug should cost less than $100,000 per year of life gained to be cost-effective. When confronted with a patient (even a hypothetical one), however, they endorsed giving drugs that cost up to $250,000 per additional year of life.

    The authors conclude that we should do a better job of helping doctors understand how cost-effectiveness information should be used. I agree with that, but want to add to it. It’s necessary, but not sufficient.

    I think society needs to have this discussion. Almost no one has $250,000, or even $100,000, saved up if they need it to extend their life for one year. So when we say we think that’s the reasonable number, we’re asking others to pay for it, either through government programs or private insurance. Is $100,000 a reasonable amount to pay for an additional year of life? Is it too low? Is $250,000 enough?

    There has to be a maximum. Surely $1 trillion is too much for a year of life; that would bankrupt the country. So there is a limit, and this isn’t a theoretical exercise. It’s something we really need to decide at some point. But it’s a conversation America seems determined never to have.


    • This is a great point. Even within the cost-effectiveness literature, a QALY in the United States is typically considered to be worth $50,000, or a little more than the GDP per capita of the United States. Put another way, we consider one year of life at full health for one person to be worth slightly more than that person can, on average, be expected to produce in value.

    • This is an interesting result, but only from an academic point of view. The underlying assumption (stated, but seemingly without much awareness of its implications) was that the oncologist should be the decision maker. This is why the oncologist needs to be educated about cost-effectiveness. That’s already starting from an odd premise: that the preferences of the physician should take precedence over the preferences of the patient or of the society. The value of a life year differs across individuals. There’s no reason to think that a physician is uniquely qualified to determine the value of a life-year, either for society or for a particular patient. Obviously public programs need to establish guidelines, but that is for the program to determine.

      • This is exactly right. The physician is the wrong person to make this choice. The doctor’s job is to do the “best” for the patient regardless of anything else. The doctor can’t say to the patient: it’s my decision that a year of your life isn’t worth it to the government. The government has to say that.

        Of course, even though making the government the bad guy is exactly the right thing, Ms. Palin is going to be screaming “death panels” again. Sigh.

    • I think these numbers exist when used to calculate insurance benefits and to plan other policies. Why does this continue to be a problem when it comes to health care? It is surely a conversation we need.


    • The oncologist doesn’t pay for the drug. The oncologist earns money by prolonging treatment. The oncologist will be in favor or any treatment that prolongs treatment regardless of cost. The oncologist should not be the person making this decision.

    • Here’s another way to interpret the results of this study: when we ask people an abstract question, they tend to give one answer, but when that confronted with a particular case, they respond differently. In the words of Joseph Stalin, “One death is a tragedy; a million deaths is a statistic.” So the conclusion is that asking people artificial questions about dollars per QALY is a poor way to determine their real preferences.

    • I agree with comments suggesting that oncologists should not be the ones making the decision. Their role, along with other experts, is to tell us how much we can extend life for x dollars. I strongly agree with the suggestion that there needs to be a national discussion of how much public money we should be willing to spend to extend someone’s life by X amount. We should make that discussion a step towards a transparent health insurance system where we say,that we will spend x public dollars to add a year of life ( and how we estimate the impact of different procedures on life expectancy) in Medicaid and Medicare and require private health insurance policies to be equally transparent

    • Given that both health care in general and drugs in particular cost significantly more in the US than in most other countries it would be interesting to see the results of a similar study in other countries.

    • Put a 20% co-pay on oncology treatment and that will help the decision-making.

      That may sound cold to some, but a bankrupt country sounds even colder to me.

      • Rather than a co-pay based on 20% of the treatment cost, how about a co-pay based on 20% of the patient’s or family’s net worth? See the ramifications?

    • Even if you remove the financial incentive completely and pay docs a flat salary, they are still going to tend towards aggressive chemotherapy far too often to fit into the “guidelines”

      The economist response to this is to say “you idiots, its a simple fix. Just tell the patient that an extra year of life is not worth 200k.” The assumption is that the widgets will magically fall into place and the quota/goal will be met.

      Of course, economists have never had to sit down with a patient and tell them that they are dying. As soon as the doc says what the economist wants him to say, the patient walks out the door and goes down the street to another doc.

      Expecting doctors to sit down with patients and say “your extra year of life is not worth the extra treatment cost” is absurd. It wont happen. Even if you make all doctors employees of the federal government and pay them a flat salary it STILL wont happen. Thats what the economists dont understand.

      There’s no simple one step fix to this problem. You have to attack it from multiple angles:

      1) Remove the financial incentive to give extra treatments that are fruitless. Refuse reimbursement for all chemo treatments done in the last X months of life. Better yet, pay a flat fee to docs for every diagnosis, not for every treatment. If a patient is diagnosed w/ AML, the doc gets a flat $X fee that he can use as he sees fit to treat the entire life of the patient for his cancer.

      2) Incentivize doctors to discuss end of life treatment plans with patients. Research shows that the more often that patinets discuss this with doctors, the more often they choose the less invasive interventions.

      #1 and #2 by themselves are nowhere near enough to meet the “goals” that economists will set down for end of life care. #3 is key

      3) Dont force the doc to be the “bad guy” by refusing treatment and dont let the patient make the stupid choice to continue fruitless treatment. Set up a guideline panel that is iron-clad, no exceptions. If you are 85 years old with severe diabetes and congestive heart failure, and you get AML, you get no chemotherapy period., except for palliative/hospice care. No appeals, end of discussion. Make the government the bad guy. Have the panel make the right decision that nobody else could.

      • I agree with making the government the bad guy, Start with a highly visible discussion of what the government should pay for an added year of life (e.g. televised congressional hearings) . Then translate that standard into legislation. As much as possible, make congress and the President the focus, not the panel whose job it will be to determine what kinds of care extend life long enough to meet the standard. Require private insurance companies to clearly, explicitly, and visibly describe their standards. (They would be free to accept the government standard or develop their own). Require doctors to obtain a statement from patients explicitly acknowledging the standard set by their insurance. Pay for end of life counseling.

    • Ron and docjones’ plans would lead to a lot of dead people. Not the patients, but the executives, administrators, and panel members who would wind up getting shot by the decedents’ family members.

    • I heard a Dartmouth economist whose name I can’t recall offer an interesting solution-offer the patient money!
      Would you like treatment with Provenge which will extend your life with prostate cancer by about 4 months at a cost to the system of $100 000, or would you like $25 000 cash added to your estate?