3 ms·

When Will the FDA Approve Paxlovid?

23 points by feross 5y ago

newsbinator 5y ago
>Shouldn’t I be equally skeptical of Paxlovid now that it’s having its own super-promising early trials? > No. For one thing, this isn’t amateur hour anymore. The ivermectin trials were random people who bungled their experiments or just plain made them up. They had sample sizes of (going through the first few on my notes) 25, 116, and 66 people. The Paxlovid trial was run by the best scientists Pfizer’s money can buy, and had a sample size of 1,219 (it would have been 3,000 if they hadn’t stopped it early).
- tasogare 5y ago
  > The Paxlovid trial was run by the best scientists Pfizer’s money can buy Well, buying scientists is exactly the problem at hand with Pfizer et al. And before one jump to call me names, it's a very similar situation to tabacco industry buying experiments on cigarettes, or oil industry and the climate change. What science need is more independent trials.
  - credit_guy 5y ago
    > What science need is more independent trials. Your wish is hereby granted: https://www.nih.gov/research-training/medical-research-initiatives/activ/covid-19-therapeutics-prioritized-testing-clinical-trials https://www.nih.gov/research-training/medical-research-initi...
- ncmncm 5y ago
  For benefit of those who did not read his long post about ivermectin: the best estimate from available data is that any benefit seen comes from treating co-morbid worm infestations. (There is lots of test-tube evidence suggesting ivermectin was worth trying; but actual clinical effectiveness is all that really matters, in the end, and that failed.)
  - merpnderp 5y ago
    Wasn’t it a guestimation of 50/50 that relied on doctors not realizing treating people for covid who also had worms with steroids would cause the worms to go into hyper growth? Great article but that last bit about assuming doctors in areas where worm infections are common don’t know they can’t treat the infected with steroids, seems a big stretch.
shusaku 5y ago
> Or perhaps there’s enough evidence, but it takes a long time to process it? But then how come the prediction markets are already 90% sure what decision they’ll make? Is this kind of a Grossman-Stiglitz Paradox? I guess the FDA should just consult the prediction market to figure out what decision they will make!
stevespang 5y ago
The FDA has always been about "paralysis by analysis" and vice versa . . .
LargoLasskhyfv 5y ago
When they have a less obvious name that doesn't read like Pavloxid.
ncmncm 5y ago
People are talking about the delay as if it might have something to do with side effects. But how does delaying approval change anything? They are not learning anything more about side effects while they sit with no trials running. AFaICS, the only difference is that the side effects, if any, are delayed exactly as long as the treatments, with all the deaths that would have been prevented accumulating, day by day. At some point the harm from non-treatment exceeds the potential/expected harm of side effects from treatment in people who did not die even though untreated. (Side effects that would have been expected in people who die, instead, probably should not count.) Is that when they approve it? What was achieved? Also: since they halted the trial, did they pull the treatment from people who were getting it? And still not give any to the control patients? How is that more ethical than continuing the trial?

23 points by feross 5y ago

>Shouldn’t I be equally skeptical of Paxlovid now that it’s having its own super-promising early trials? > No. For one thing, this isn’t amateur hour anymore. The ivermectin trials were random people who bungled their experiments or just plain made them up. They had sample sizes of (going through the first few on my notes) 25, 116, and 66 people. The Paxlovid trial was run by the best scientists Pfizer’s money can buy, and had a sample size of 1,219 (it would have been 3,000 if they hadn’t stopped it early).

tasogare 5y ago
> The Paxlovid trial was run by the best scientists Pfizer’s money can buy Well, buying scientists is exactly the problem at hand with Pfizer et al. And before one jump to call me names, it's a very similar situation to tabacco industry buying experiments on cigarettes, or oil industry and the climate change. What science need is more independent trials.
- credit_guy 5y ago
  > What science need is more independent trials. Your wish is hereby granted: https://www.nih.gov/research-training/medical-research-initiatives/activ/covid-19-therapeutics-prioritized-testing-clinical-trials https://www.nih.gov/research-training/medical-research-initi...
ncmncm 5y ago
For benefit of those who did not read his long post about ivermectin: the best estimate from available data is that any benefit seen comes from treating co-morbid worm infestations. (There is lots of test-tube evidence suggesting ivermectin was worth trying; but actual clinical effectiveness is all that really matters, in the end, and that failed.)
- merpnderp 5y ago
  Wasn’t it a guestimation of 50/50 that relied on doctors not realizing treating people for covid who also had worms with steroids would cause the worms to go into hyper growth? Great article but that last bit about assuming doctors in areas where worm infections are common don’t know they can’t treat the infected with steroids, seems a big stretch.

shusaku 5y ago

> Or perhaps there’s enough evidence, but it takes a long time to process it? But then how come the prediction markets are already 90% sure what decision they’ll make? Is this kind of a Grossman-Stiglitz Paradox? I guess the FDA should just consult the prediction market to figure out what decision they will make!

stevespang 5y ago

The FDA has always been about "paralysis by analysis" and vice versa . . .

LargoLasskhyfv 5y ago

When they have a less obvious name that doesn't read like Pavloxid.

ncmncm 5y ago

People are talking about the delay as if it might have something to do with side effects. But how does delaying approval change anything? They are not learning anything more about side effects while they sit with no trials running. AFaICS, the only difference is that the side effects, if any, are delayed exactly as long as the treatments, with all the deaths that would have been prevented accumulating, day by day. At some point the harm from non-treatment exceeds the potential/expected harm of side effects from treatment in people who did not die even though untreated. (Side effects that would have been expected in people who die, instead, probably should not count.) Is that when they approve it? What was achieved? Also: since they halted the trial, did they pull the treatment from people who were getting it? And still not give any to the control patients? How is that more ethical than continuing the trial?