> But that's not what the paper studied.

This is a recurring problem with science reporting. Falsifiability and reproducibility demand that good studies focus on very specific, well-defined effects. Then it gets reported as leading to much broader conclusions on the topic as a whole.

Counterintuitively, "missing the forest for the trees" is exactly what you want a study like this to do. You can let a later meta-analysis figure out how to make a forest out of the trees.