I wrote this after reading the Cowen/Pahlka interview. Cowen says: Maybe the agencies should be just shut down and rebuilt by the next administration? Pahlka: That would be too much chaos. Cowen: So maybe do it in gradual way? Pahlka: We’ve tried, it does not work. And my reaction was: “Oh my, I know these discussions from the 90′s.”
It was a cool natural experiment: A bunch of countries tried reforming at the same time, used different approaches, got different outcomes. So maybe there’s something to learn there.
Or to take the FDA example: You want to do a shake-up. How exactly would you go about that? Organizations are already in equilibrium. If you shake them, they just return to the previous state. Is the only way to shut them down and rebuild them from scratch? Or is there a less destructive, gradual approach? Again, we may get some insights from previous reform attempts, even if the problems don’t match in 1:1 way.
Congress should repeal Kefauver-Harris so that “treatment X has not been proven efficacious by a wildly expensive high church effort at proof” stops being a valid reason for the FDA to ban something (the way it is (and has been since 1962) in the status quo).
The FDA’s ban on “treatments of unproven efficacy” has never been coherent, never helped people, and slowed medical innovation way way way down.
The FDA’s ban on “treatments of unproven safety” should be much much cheaper, and slowly scale up based on the size of the N who have tried a treatment, and it should only involve an attempt to measure the lack of safety, so that informed consent can be brought to bare in specific cases, by a trusted clinician, about how much danger is “worth it” in a specific case.
Chemotherapy, for example, is often brutal, and the side effects can be fatal… its just that this is a clinical cost worth clinically paying in some cases, for some cancers, with some prognoses. It isn’t “safe” it is just “probably safer than letting the cancer run to completion, and hopefully better than alternative treatments that the doctors even know about, contingent on the specific personalized diagnosis, by one or more doctors, regarding a specific tragic situation”.
For some diseases, nothing can save you from “doctor has low skill”. Medicine is intrinsically dangerous.
The FDA is a fig leaf on this… and it is a very very heavy fig leaf, that mostly only mitigates anxiety in exchange for OCD rituals, even while the form and cost of the rituals harm science and medicine and technology and “the production of more consumer surplus for patients (by getting them highly effective treatments at competitive prices based on an economic race to the marginal cost of providing the treatment)”.
The rent on health is too damn high. The FDA needs more judges, and tort lawyers of high integrity, and economists, and data scientists running statistical clearinghouses… not more molecular biologists.
This would be a huge change, and it would also probably work.
I wrote this after reading the Cowen/Pahlka interview. Cowen says: Maybe the agencies should be just shut down and rebuilt by the next administration? Pahlka: That would be too much chaos. Cowen: So maybe do it in gradual way? Pahlka: We’ve tried, it does not work. And my reaction was: “Oh my, I know these discussions from the 90′s.”
It was a cool natural experiment: A bunch of countries tried reforming at the same time, used different approaches, got different outcomes. So maybe there’s something to learn there.
Or to take the FDA example: You want to do a shake-up. How exactly would you go about that? Organizations are already in equilibrium. If you shake them, they just return to the previous state. Is the only way to shut them down and rebuild them from scratch? Or is there a less destructive, gradual approach? Again, we may get some insights from previous reform attempts, even if the problems don’t match in 1:1 way.
Congress should repeal Kefauver-Harris so that “treatment X has not been proven efficacious by a wildly expensive high church effort at proof” stops being a valid reason for the FDA to ban something (the way it is (and has been since 1962) in the status quo).
The FDA’s ban on “treatments of unproven efficacy” has never been coherent, never helped people, and slowed medical innovation way way way down.
The FDA’s ban on “treatments of unproven safety” should be much much cheaper, and slowly scale up based on the size of the N who have tried a treatment, and it should only involve an attempt to measure the lack of safety, so that informed consent can be brought to bare in specific cases, by a trusted clinician, about how much danger is “worth it” in a specific case.
Chemotherapy, for example, is often brutal, and the side effects can be fatal… its just that this is a clinical cost worth clinically paying in some cases, for some cancers, with some prognoses. It isn’t “safe” it is just “probably safer than letting the cancer run to completion, and hopefully better than alternative treatments that the doctors even know about, contingent on the specific personalized diagnosis, by one or more doctors, regarding a specific tragic situation”.
For some diseases, nothing can save you from “doctor has low skill”. Medicine is intrinsically dangerous.
The FDA is a fig leaf on this… and it is a very very heavy fig leaf, that mostly only mitigates anxiety in exchange for OCD rituals, even while the form and cost of the rituals harm science and medicine and technology and “the production of more consumer surplus for patients (by getting them highly effective treatments at competitive prices based on an economic race to the marginal cost of providing the treatment)”.
The rent on health is too damn high. The FDA needs more judges, and tort lawyers of high integrity, and economists, and data scientists running statistical clearinghouses… not more molecular biologists.
This would be a huge change, and it would also probably work.