Thread by @interfluidity

summarizing a bit of brilliance from @BCynamon, it's a big problem that the FDA evaluates things like tests only in terms of individual safety and effectiveness, rather than benefit to public health, which amount to very different criteria. 1/

a cheap, fast, unreliable test is much worse than a high-quaity PCR test, from the perspective of a hypothetical individual who might choose either (and can get the PCR results expedited). 2/

but getting everybody frequent cheap, fast, unreliable tests would let us measure, in real time and with great precision, the state of the disease, and conditioning our behavior on a "bad" test result is not so costly if we'll have another test tomorrow. 3/

This point was also made by @ATabarrok here marginalrevolution.com/marginalrevolu…, and i recall @mattyglesias has lamented the insensitivity of medical approvals to dimensions of cost. 4/

In general, it seems like US medical regulation implicitly adopts a hypothetical, if I am an individual who can choose the best existing treatment under the best existing conditions, would this new thing constitute an improvement. 5/

At an individual, ethical level, there's something understandable in that. It seems icky to encourage a market for substandard but cheap things rather than find a way to deliver the best care for all. 6/

But when "expensive" refers to real resources (rather than artificial scarcities like patents whose constraints should be weakened) and the pervasiveness of an intervention has positive public health externalities, cheap becomes not second rate but a prerequisite to effective. 7/

Thanks again to @BCynamon, with whom I speak too infrequently, who argued this case with such verve that I find myself compelled to repeat it. (If I've said anything dumb, of course that's just me.) /fin