Dan Klein harps on a too-rarely-asked question: Counterparts to the FDA function in other countries. The FDA could adopt a standing policy that drugs permitted in Europe, Canada, Japan, Australia, etc. automatically become permitted in the US. Why doesn’t the FDA adopt such a policy? Is it because a drug that is safe and effective in Australia, Canada, or France may not be safe and effective in the United States?
Since approvals are made by national agencies, nationally aggregated incentive determines willingness to fund the (expensive) review process. The U.S. is often the largest market in sales and patients and places high valuations on life and thus has the highest aggregated incentive to fund "correct" decisions. Under a policy that the FDA would accept approval in certain other countries, drug companies would have a strong incentive to find the review agency with the least ability to stage a critical review. It probably won't be the U.S FDA.
So yes, I might know better than Billy's parents if I have more time and money to spend investigating.
If you grant the FDA is capable of making a good judgment on that matter, why argue in favor of the idea that the judgment it *has made* is bad? Shouldn't you accept the decision they've come to, on the basis of trust based on evidence?
You and I have been arguing past each other. I took as a premise that the FDA requires duplication (at great expense) of previous findings in other countries, but that the findings are typically in agreement. You seem to be talking about the case where the findings are in disagreement, which I assumed to be rare on account of stability of the causal mechanism. Hmm.
Here is a scenario for separate drug regulations that actually operates in the real world, but by no mean a large part of why things are organized as they are. Drug effects can vary racially and Japan's regulatory apparatus uses its control of the market to force studies of Japanese samples, especially in the case of "me-too" drugs. (By "scenario," I mean merely what happens; I make no claims about market failure.)
BTW, other COX II inhibitors are still on the market. Vioxx was not the only one. Merck fudged the safety data on it, in their submission to the FDA. That might be part of the reason they decided to withdraw it themselves.
There is a pretty simple answer to this -- America is clearly the leader in bio-medicine. The NIH is the largest biomedical funding research agency in the world. The FDA most likely has much more qualified people making these decisions than its counter parts. The FDA also recently has come under scrutiny for some conflicts of interest. I'm not sure why I would expect drug regulatory industries in smaller countries to be more instituationally sound. In an ideal world, one supreme drug-regulatory agency would make the decision for everyone -- but we live on Earth.
This type of "multiple review" inefficiency is in lots of systems. Peer Review. Job Interviews, etc. It's just not financially important enough to whine about.
Peter, while drug spending has better support in the literature for giving health benefits than overall spending, yes my relative skepticism about drug benefits makes me see less at stake in drug regulation. But something is at stake, and it is a nice clear example of paternalism to help us understand paternalism more generally.
I have friends who work in the Canadian equivalent of the FDA. They actually run their inspections differently from the FDA, on purpose, to cover off things that the FDA might miss. So they think they're smarter than the FDA, eh?
Merck made that decision. One can speculate about what the FDA might have done, but in this case the manufacturer acted first.
You are absolutely right and I am wrong. Wikipedia also tells me that an FDA advisory committe voted 17-15 that Vioxx should be brought back because the benefits outweighed the costs. Updated.
I will avoid the tort reform debate on why the costs to Merck outweighed the benefits to Merck.
Robin, you made a comment here indicating that you think drugs are less valuable than most of us think.I'm wondering whether that view can be reconciled with your interest here in more access to drugs and the view Eliezer mentioned:
According to this source, the FDA's longer approval process prevents 5,000 casualties per year by screening off medications found to be harmful, and causes at least 20,000-120,000 casualties per yearAre the benefits of new drugs low, but still positive enough to be an exception to your general rule that the marginal benefit of medical spending is zero? Or is the benefit of new drugs unrelated to how much is spent on medicine?
"Reference Vioxx above: Phil Goetz thinks it would be an acceptable trade-off for him, but the FDA has decided that no one gets to make that decision." Merck made that decision. One can speculate about what the FDA might have done, but in this case the manufacturer acted first.
Zubon wrote: "The standards I hold for my own use are different from the standards I hold for making something illegal. I do not drink alcohol; I do not like the benefit-cost trade-off there; I do not believe the safety testing on alcohol would be so negative as to drive it off the market. Reference Vioxx above: Phil Goetz thinks it would be an acceptable trade-off for him, but the FDA has decided that no one gets to make that decision.
Again, two separate questions: should you use x? Should everyone be legally barred from using x?"
I agree with you. My argument there was about someone's proposal to test only for safety.
I think that all drugs should be legal, and that the FDA should be more like, say, UL -- it would certify drugs so that people and doctors can have more confidence in them, but no one should be legally prohibited from choosing what to put in his own body.
I have friends who work in the Canadian equivalent of the FDA. They actually run their inspections differently from the FDA, on purpose, to cover off things that the FDA might miss.
Caledonian, is there reason to believe that the FDA has the authority to consider other health agencies' certification procedures? I do not know the relevant legislation, but I would not be surprised to find that Congress has eliminated that option.
That would address the second point, if the FDA is not allowed to exert its judgment on that question. On the first point, it would pass the recursive buck to, "Why does Robin Hanson think his judgment is better than those Congressmen?" But that seems like a dead question, given how likely anyone on the internet is to defer to the legislature's judgment.
Presumably the FDA is competent to assess whether other health agencies' certification procedures are adequate. Granting that presumption for the sake of argument: The FDA does not take those health agencies' certification procedures into account when making its decisions.
If you grant the FDA is capable of making a good judgment on that matter, why argue in favor of the idea that the judgment it *has made* is bad? Shouldn't you accept the decision they've come to, on the basis of trust based on evidence?
Since approvals are made by national agencies, nationally aggregated incentive determines willingness to fund the (expensive) review process. The U.S. is often the largest market in sales and patients and places high valuations on life and thus has the highest aggregated incentive to fund "correct" decisions. Under a policy that the FDA would accept approval in certain other countries, drug companies would have a strong incentive to find the review agency with the least ability to stage a critical review. It probably won't be the U.S FDA.
So yes, I might know better than Billy's parents if I have more time and money to spend investigating.
duplication (at great expense) of previous findings in other countries
A lot of people on this thread seem to believe that, but it's not true: they use the same studies.
If you grant the FDA is capable of making a good judgment on that matter, why argue in favor of the idea that the judgment it *has made* is bad? Shouldn't you accept the decision they've come to, on the basis of trust based on evidence?
You and I have been arguing past each other. I took as a premise that the FDA requires duplication (at great expense) of previous findings in other countries, but that the findings are typically in agreement. You seem to be talking about the case where the findings are in disagreement, which I assumed to be rare on account of stability of the causal mechanism. Hmm.
Here is a scenario for separate drug regulations that actually operates in the real world, but by no mean a large part of why things are organized as they are. Drug effects can vary racially and Japan's regulatory apparatus uses its control of the market to force studies of Japanese samples, especially in the case of "me-too" drugs. (By "scenario," I mean merely what happens; I make no claims about market failure.)
BTW, other COX II inhibitors are still on the market. Vioxx was not the only one. Merck fudged the safety data on it, in their submission to the FDA. That might be part of the reason they decided to withdraw it themselves.
There is a pretty simple answer to this -- America is clearly the leader in bio-medicine. The NIH is the largest biomedical funding research agency in the world. The FDA most likely has much more qualified people making these decisions than its counter parts. The FDA also recently has come under scrutiny for some conflicts of interest. I'm not sure why I would expect drug regulatory industries in smaller countries to be more instituationally sound. In an ideal world, one supreme drug-regulatory agency would make the decision for everyone -- but we live on Earth.
This type of "multiple review" inefficiency is in lots of systems. Peer Review. Job Interviews, etc. It's just not financially important enough to whine about.
Peter, while drug spending has better support in the literature for giving health benefits than overall spending, yes my relative skepticism about drug benefits makes me see less at stake in drug regulation. But something is at stake, and it is a nice clear example of paternalism to help us understand paternalism more generally.
I have friends who work in the Canadian equivalent of the FDA. They actually run their inspections differently from the FDA, on purpose, to cover off things that the FDA might miss. So they think they're smarter than the FDA, eh?
Damn straight. Good for them.
Merck made that decision. One can speculate about what the FDA might have done, but in this case the manufacturer acted first.
You are absolutely right and I am wrong. Wikipedia also tells me that an FDA advisory committe voted 17-15 that Vioxx should be brought back because the benefits outweighed the costs. Updated.
I will avoid the tort reform debate on why the costs to Merck outweighed the benefits to Merck.
Robin, you made a comment here indicating that you think drugs are less valuable than most of us think.I'm wondering whether that view can be reconciled with your interest here in more access to drugs and the view Eliezer mentioned:
According to this source, the FDA's longer approval process prevents 5,000 casualties per year by screening off medications found to be harmful, and causes at least 20,000-120,000 casualties per yearAre the benefits of new drugs low, but still positive enough to be an exception to your general rule that the marginal benefit of medical spending is zero? Or is the benefit of new drugs unrelated to how much is spent on medicine?
"Reference Vioxx above: Phil Goetz thinks it would be an acceptable trade-off for him, but the FDA has decided that no one gets to make that decision." Merck made that decision. One can speculate about what the FDA might have done, but in this case the manufacturer acted first.
Zubon wrote: "The standards I hold for my own use are different from the standards I hold for making something illegal. I do not drink alcohol; I do not like the benefit-cost trade-off there; I do not believe the safety testing on alcohol would be so negative as to drive it off the market. Reference Vioxx above: Phil Goetz thinks it would be an acceptable trade-off for him, but the FDA has decided that no one gets to make that decision.
Again, two separate questions: should you use x? Should everyone be legally barred from using x?"
I agree with you. My argument there was about someone's proposal to test only for safety.
I think that all drugs should be legal, and that the FDA should be more like, say, UL -- it would certify drugs so that people and doctors can have more confidence in them, but no one should be legally prohibited from choosing what to put in his own body.
PhilB
I have friends who work in the Canadian equivalent of the FDA. They actually run their inspections differently from the FDA, on purpose, to cover off things that the FDA might miss.
Caledonian, is there reason to believe that the FDA has the authority to consider other health agencies' certification procedures? I do not know the relevant legislation, but I would not be surprised to find that Congress has eliminated that option.
That would address the second point, if the FDA is not allowed to exert its judgment on that question. On the first point, it would pass the recursive buck to, "Why does Robin Hanson think his judgment is better than those Congressmen?" But that seems like a dead question, given how likely anyone on the internet is to defer to the legislature's judgment.
Presumably the FDA is competent to assess whether other health agencies' certification procedures are adequate. Granting that presumption for the sake of argument: The FDA does not take those health agencies' certification procedures into account when making its decisions.
If you grant the FDA is capable of making a good judgment on that matter, why argue in favor of the idea that the judgment it *has made* is bad? Shouldn't you accept the decision they've come to, on the basis of trust based on evidence?
A lot of bureaucrats were involved in the decisions regarding what the US drug policies should be.
Why does Robin Hanson think his judgement is better than those bureaucrats'?
The argument from majjority consensus cuts both ways.