4 ms·

I wouldn't say it's ridiculous. This is generally how a scientific study is done. One "object" per study. An article like this is not made to stand alone and sh

I wouldn't say it's ridiculous. This is generally how a scientific study is done. One "object" per study. An article like this is not made to stand alone and should in general not be used for non-science purposes, such as policy-making. First order comparative studies (such as experiments) of course exists, but comparisons are generally done to my knowledge in meta-studies. For laypeople and policy makers we of course need comparisons and validations before we let them guide our choices.

amluto 6y ago
I disagree. There is no way this can be meta-analyzed in a way that can reliably inform policy, and waiting until we discover (or not) which vaccines remain effective in the face of variants due to an outbreak in a vaccination is horrible public health and economic policy. To the contrary, the policy makers have access to insane amounts of money. The US has spent trillions overall due to Covid and billions on vaccine development. Paying the researchers or a different lab $1M to run an important study is a no-brainer. Or do it DARPA/DoD style: announce a list of studies that need doing, let people apply for grants to do them, and keep the list updated! The fact that researchers are bumbling around doing good quality science, getting nice publications, and failing to answer the right questions is an abject failure of science policy. edit: here’s your meta-analysis: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7654888/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7654888/ This sucks. As the authors note, the individual studies were not usefully standardized and mostly can’t be compared. At the very least, the FDA could have said, as part of the initial Covid vaccination trial goals, that a specific set of identical measurements were required, and that all candidates were expected to make those particular measurements. Additional measurements would, of course, be welcome. This would have cost the public a grand total of $0. Maybe $500k for someone to write the thing down and take some feedback to make sure the list made sense, but that seems high.

amluto 6y ago

I disagree. There is no way this can be meta-analyzed in a way that can reliably inform policy, and waiting until we discover (or not) which vaccines remain effective in the face of variants due to an outbreak in a vaccination is horrible public health and economic policy. To the contrary, the policy makers have access to insane amounts of money. The US has spent trillions overall due to Covid and billions on vaccine development. Paying the researchers or a different lab $1M to run an important study is a no-brainer. Or do it DARPA/DoD style: announce a list of studies that need doing, let people apply for grants to do them, and keep the list updated! The fact that researchers are bumbling around doing good quality science, getting nice publications, and failing to answer the right questions is an abject failure of science policy. edit: here’s your meta-analysis: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7654888/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7654888/ This sucks. As the authors note, the individual studies were not usefully standardized and mostly can’t be compared. At the very least, the FDA could have said, as part of the initial Covid vaccination trial goals, that a specific set of identical measurements were required, and that all candidates were expected to make those particular measurements. Additional measurements would, of course, be welcome. This would have cost the public a grand total of $0. Maybe $500k for someone to write the thing down and take some feedback to make sure the list made sense, but that seems high.