Who again is supposed to cut the rate of growth of Medicare spending?

A federal advisory committee said Wednesday that there was adequate evidence that the drug Provenge prolongs the lives of men with advanced prostate cancer, an endorsement that makes it more likely that Medicare will pay for the $93,000 treatment.

The Medicare Coverage Advisory Committee voted that it had “intermediate confidence” in the strength of the data that the drug, developed by Dendreon, improves survival.

The median result in trials is four extra months of life, and the link is here.  If you think that four months of life are worth $93,000 of public money, that is a defensible position.  But how many times can you apply that argument?  I saw this treatment as a good candidate for allocation by market mechanisms, but I suppose that's not how it will be.


The company that makes it was the fast money darling stock of 2006-2009, due to oddities in the FDA approval process.

$93,000 for intermediate-confidence of an extra four months of still-dying-from-cancer time has got to be about $600K of other people's money per Value-Added of Quality-Adjusted-Life-Years($/VAQALY or ICER).

In the UK, the maximum NICE threshold for pharmaceutical funding is something like $60K - ten times less.

Sigh - does anyone doubt this is all headed for total collapse?

"Four months is a long time when you're dying."

If your time was worth more when you were closer to death wouldn't this imply strongly negative real interest rates?

In the context of Medicare the argument tends to become more political. But for a moment, ignore Medicare, and consider a private insurer. How does a private insurer take decisions like these? What is the framework to use when deciding if or not a certain treatment is to be covered or not?

Life-years extended? Life-years extended per dollar? Probability of success?

Is 4 additional years equally worthwhile in a 80 year old man as in a 20 year old boy?

"Kinda sorta maybe this shouldn't be subject to the democratic process." A dictatorial process then?

Um... I'm still baffled why anyone thinks Medicare or Social Security are going to get cut in any way. We're about to have the largest population of retirees in this nation's history and they're going to cut their own benefits? Why are we even discussing this? It's more likely we'll get rid of the Department of Defense.

The fact that the drug costs $93,000 is a pretty good indicator that market mechanisms are not working for this drug. In fact, the company has a government-granted patent monopoly on production of the drug, so they can charge whatever they please. No market there.

If the drug was priced based on its cost of production in a competitive marketplace, it would cost hundred of times less than that.

You need on of our good old fashioned European death panels.

Lots of politicking involved, but mostly does a good job.

James, the marginal cost of the drug is probably close to zero, but the risk-adjusted development costs were massive. If the company can't recover it's R&D costs with a risk premium, it would never develop the drug in the first place. There's a market failure called natural monopoly.

Expensive drug: poor people die sooner.

No drug: everyone dies sooner.

Those costs borne by government are entirely indefensible if they make the system insolvent, even if we value every life equally.

We all die, and no amount of medical care changes that. More care equals higher costs. Costs mean productive resources diverted from alternative uses. Resources must be paid for. Medical care is a private good with miniscule positive externalities. Such a good is most efficiently paid for by the beneficiary. Any insurance scheme should be actuarily sound according to the expected losses.

But let's pretend the actual reduction is from a baseline rate of 8% to Lipitor's 4%. So the NNT for Lipitor compared to placebo is 100/4=25. 25 people have to get Lipitor to prevent one heart attack. 25 people is $45,625/yr. Is $45,625 worth it to prevent one heart attack, or, at that price, should we maybe just stop eating donuts?

This seems the classic example of where the economically sound option may not be the ethically palatable one.

The solution to overpriced or overvalued end of life costs is simple:

Establish an end of life medical cost exclusion, or a variant of it, as follows:

The last 6 months of end of life care (medicare costs) comes out of the estate of the deceased(with an exclusion or cap if there is a surviving spouse).

Watch the kids talking with the parents about "do you really want to prolong your life this way; wouldn't you be happier if..."

People want things for others if they are free, not if they have to pay for them. My daughter, a pathologist, tells me of the following case as told to her by another doctor. Patient is comatose, will require millions to be kept alive, kids have a choice to pull the tube. Kids say no. Hospital says: OK, we have to transfer him to a nursing home; state will pay for the care, but you need to pay $35 for part of the payment for the ambulance. Kids say no, balking at the $35. They then say Pull the tube.


How would people's incentives change if the last 6 months of end of life care were not on the house?

I meant to say: I'd call my track B plan "voluntary death panels."

You are doing it wrong. Let's accept figures of 4 months, and Medicare paying $93,000 for it.

But $93,000 has nothing to do with manufacturer's marginal cost of producing drugs to provide those 4 months over long term (= actual economic cost)! Most of the price is just incentives to cover R&D costs averaged over multiple drugs via patent system.

How much will making this drug cost once patent expires, generics come in, and some experience of mass scale manufacturing kicks in? $9300? $930? $93? These are the kind of numbers we always get in the end.

I'd even say that - if we don't drastically "overpay" for new drugs, we're economically inefficient in the long term.

I find it very ironic that you of all people are missing distinction between marginal costs and price.

Now drug patent system is pretty bad in some ways, it promotes wasteful ad campaigns, lawsuits, R&D of knockoffs that narrowly escape patent claims etc. but that's a totally different story. Even given all that, $93000 is completely wrong number.

Make the pharmaceutical testing procedure cheaper.

Drug companies have to spend a lot of money -- and take a lot of time -- doing clinical trials. Remove that burden from them. Let the medical profession do it for them.

Anybody who isn't satisfied with standard treatment can volunteer for experimental treatments. If they volunteer they don't get to choose which experimental treatment they get, and they might get assigned to a control group and get the standard treatment.

As people drop out of the program, publish their sanitised results on the internet. Continue the study with as many people as want experimental treatment, until you get results showing which version is better or results showing that the more expensive version is not enough better to justify the expense. Publish the results on the internet.

It's silly to have the drug company which stands to lose as much as a billion dollars if the treatment is not better, do the clinical studies which prove it is better. That adds to their expense and encourages shoddy work.

If a large number of scattered physicians try the new approach, they'll get the sort of results they'd get after the treatment became standard, apart from the problem that it's done on volunteers for experimental programs. If there aren't enough volunteers, that's a sign that patients are satisfied with the standard treatment.

If it is done this way it will be cheaper for drug companies to develop new drugs, and there will quite likely be fewer new drugs declared safe and effective, much less improvements on standard practice.

I had no idea the end-of-life months were on the house. I've been through the process several times now, and it was always financially wrenching. We all must have been doing it wrong.

I would spend ALL of my money for 4 months worth of life.

Does this entitle me to spend $93,000 of yours?

Really?? If spending all your money would leave your spouse in poverty, you would spend all your money knowing it merely prolongs your suffering?

My dad had prostate cancer and as he put it, "I just wanted to put your mother in the ground", but he ended treatment with the late 90s cancer drug trial. He spent $10K a week for a couple of weeks with no benefit, applied to the drug company for aid and was turned down, and then told his doctor he wasn't going to follow his advise and spend his last $50K on the drug.

[This was before the Medicare Drug benefit; but he had a drug benefit that didn't cover the drug. And the current Medicare drug program doesn't cover all drugs; it depends on which plan. Of course, Provenge is not a drug.]

And the $50K, plus life insurance benefit, went to the government because his heir at the time was my mother who was impoverished from nursing home care needed because of her dementia. She did not know who he was, but she knew a nice man was no longer coming to see her.

My dad had a lot of experience with end of life after a life of pastoring, and sitting with the dying, and he had talked with hospice, but not soon enough.

None of his doctors spent any real time taking about end of life planning, but only suggested he talk to hospice services. Clearly his doctors were not comfortable in talking about death.

Besides, they weren't paid by Medicare to talk about end of life, nor was anyone they might have referred my dad to going to be paid by Medicare. But when a provision was put in the health reform to pay for such consultation, Palin called it death panels and Republicans objected to it as Medicare dictating euthanasia.

It's not public money. It's our money. They take it and then give some of it back to some of us if we spend it the way they require.

There are things that could be called 'public money.' This isn't one of them.

All you are seeing is another reason why we shouldn't try to cram non-public expenditures into the definition of "public money."

"I find it very ironic that you of all people are missing distinction between marginal costs and price."

I don't know how much anyone is missing. However, the R&D gets paid for. The issue is the distortion that public funding creates in medical decision-making. A majority of medical expenses take place at the end of life. If not for Medicare et.al. would this be more skewed toward prevention or lower cost comfort efforts or even 'go have fun.' All that money spent to develop this drug that must now be recouped to develop the next 4.1 month drug might have been spent elsewhere, like on aging research, or something else that actually is a public good.

Yes, Andrew.

The marginal cost to the patient is zero, so this drug will be overconsumed. In fact, it will be consumed by every patient with prostate cancer save a few weird exceptions. How can the insurance/health care system afford this? Oh, and all other similar drugs.

If a drug obviates the need for a more costly surgery later, it makes sense.

The price doesn't merely represent the MC, but also development costs. We permit the monopoly because those fixed costs must be paid for, otherwise we don't get the drug at all.

This is a very old economic question (Vernon and Grabowski) with a brand new real world example.

I'm glad people brought up the costs to families and quality of life issues. All relevant.

The ethics is not relevant at all. If it were, there would not be a starving or malnourished person on the planet. Feeding people their entire lives is more important than extending their life a few months at the end. Food, though, is not a natural monopoly, which makes the provision more economically interesting.

Patients need incentives to not request this or similar treatments. See Howard's comment above- a reasonable proposal to build such incentives into the system. See, also, Bill's comment immediately above those.

The real question isn't whether or not to use a treatment like this.

The real question is, "Who decides?"

My vote is for the patient and their family and letting them decide ahead of time in their choice of private insurance. This sort of thing is one reason why people don't want a "government" solution.

Thomas Sowell, My proposal, that the estate of the deceased cover the last six months of medical care, let's the family decide.

In fact, they can, if they want, buy insurance to cover the costs of the last six months of life.

Not socially insuring the six month tail risk alters decisions in a way that will promote efficiency, reduce costs, increase choice, and provide an insurable risk that people can plan for, or assume, as they choose.

And, it will save a lot of money.

"The median result in trials is four extra months of life ... If you think that four months of life are worth $93,000 of public money..."

Four months is a median. Before making any judgement about the value of the treatment, one needs to look at the mean. If 10 people had their lives extended by 20 years and 11 people died within 4 months, then 4 months is a median, but the treatment on average would have bought more than ten years worth of life.

Comments for this post are closed