Remember That Study That Claimed That Masks Don’t Work?

Not so much, it seems.

The publisher of the study, has stepped back from the study and repudiated the statements made by the study's lead:

The Cochrane Review has apologized for an evidence review that led many to conclude, inaccurately, that masks don’t work.

The idea that masks don’t help slow COVID is an “inaccurate and misleading interpretation” of the report they published in January, Karla Soares-Weiser, editor-in-chief of the Cochrane Library, wrote in an update posted to the their website on Friday. The international organization publishes summaries of evidence on various health topics, and are now blaming a poorly-worded summary of one report for the fact that many people came away with the idea that the face coverings don’t help.

Their analysis was based on a dozen studies that compared people wearing medical masks to their barefaced counterparts, couched inside a broader look at physical pandemic measures. Despite a line cautioning that the shakiness of the data “hampers drawing firm conclusions,” the authors concluded that masking in the community made “little to no difference.”

This is not surprising.  While meta-analysis such as this one, were multiple studies are aggregated can be valuable if done well, this was not.  They were grouping together studies of apples and studies of oranges.

Also, note that meta-analysis is not the best method in all cases.

Where you can actually understand the basic science behind a phenomenon, and model it, and construct models of it to conform performance, this is the best studies, and it has been conclusively shown that N95 type masks do reduce transmission, and there is also significant evidence that elsastomeric masks with replaceable N99 filters are even better.

It should also be noted that the lead author for the study, Tom Jefferson, writes for the Brownstone Institute, which is a wack-doodle anti-mask and anti-vaccine outfit that grew directly from the infamous "Great Barrington Declaration".