N95 masks work. We have a handful of studies of sufficient power behind them to show that.
There are a lot of studies that have insufficient power to prove that surgical masks "work". There are a few studies of sufficient power to show genuine improvement from surgical masks--generally in the cohorts with greater risks.
As far as I know, there are zero sufficiently powered studies showing that surgical masks don't work or have a negative effect.
"Cannot prove for all cohorts" does NOT magically mean "does not work". The fact that some studies show that the masks have a provable effect shifts the Bayesian prior significantly towards "likely positive effect but difficult to prove without expensive trial".
After all, surgeons started wearing masks long before Covid-19.
Comments
N95 masks work. We have a handful of studies of sufficient power behind them to show that.
There are a lot of studies that have insufficient power to prove that surgical masks "work". There are a few studies of sufficient power to show genuine improvement from surgical masks--generally in the cohorts with greater risks.
As far as I know, there are zero sufficiently powered studies showing that surgical masks don't work or have a negative effect.
"Cannot prove for all cohorts" does NOT magically mean "does not work". The fact that some studies show that the masks have a provable effect shifts the Bayesian prior significantly towards "likely positive effect but difficult to prove without expensive trial".
After all, surgeons started wearing masks long before Covid-19.