This is a perfect example of why our current media feels so exhausting. There's so many conflicting viewpoints and studies that all seem reasonable.

It's exhausting because the stories exaggerate the findings. Most everything in science is nuanced. Small sample, preliminary finding, caveats, unclear results…

Really, this is more of an issue with modern storytelling which is often motivated to be sensational, or at least to present a relatable story.

Or in this case a meta study. It's hard to say how meaningful the results are because the purposes and methods of the underlying studies, meaning it's not likely they were designed to control for other factors that would impact what this study was looking at.

With a low enough sample size anything can be true.

0.8*1.2 = 0.96, so that's a net 4% decrease in deaths :)