[srinivgp]

The paper you linked to says, in its summary section, "For the general public, evidence shows that physical distancing of more than 1 m is highly effective and that face masks are associated with protection, even in non-health-care settings, with either disposable surgical masks or reusable 12–16-layer cotton ones, although much of this evidence was on mask use within households and among contacts of cases."

But I assume you mean that the meta analysis didn't claim masks were useless, those three underlying papers did? I'm not sure what those three were, though, maybe I missed it. I found all the papers used but there were like a dozen in the non-health-care setting.

[timr]

That is their interpretation of the meta-analysis, yes ("summary section" == "discussion section"). That is not a statement of fact, but of the authors' opinions about the facts reported. The data stands on its own -- though in this case, since it's a meta-analysis with fancypants statistical re-weighting of the data, the line between "data" and "analysis" is decidedly blurred.

> But I assume you mean that the meta analysis didn't claim masks were useless, those three underlying papers did? I'm not sure what those three were, though, maybe I missed it. I found all the papers used but there were like a dozen in the non-health-care setting.

The data they have presented, aggregating ~all of the prior research literature on masks, shows a weak effect, at best. Only by mixing data from different settings (mostly hospitals) and different types of masks (from cloth to respirators) and re-weighting the conclusions do they arrive at this claim of even weak effectiveness.

Do masks work? Based on the literature, the only honest answer is "maybe they have a small effect, but the evidence isn't very good."

Compare this to what you see repeated in the media and by "experts". The truth is...most of these people haven't read the papers, and are just repeating what they've heard.