That's cool. It still seems sus AF for the reasons that I added to my original comment.

For example, they're discounting research based on a "publication bias" but who determines the bias? What about their own bias?

Their other tools are equally questionable: random-effects, machine learning...

Perhaps most importantly they can't undermine the results for low income groups which seem to be the most important for this type of thing:

First, the meta-analytic association between economic inequality and mental health was negative only in low-income samples. This finding was replicated using Gallup data: In low-income contexts, a one-point increase in the Gini coefficient predicted a mental-health decline equivalent to moving from the 39 th to the 50th percentile of the within-country income distribution (for details on the benchmarking procedure, see SI, p. 37). This suggests that inequality may be particularly harmful to low-income populations, possibly by undermining community cohesion70 , fostering adverse social comparison71 , or fueling perceptions of unfairness 31 . Thus, even if inequality does not noticeably affect overall population mental health, it may still exacerbate disparities between income groups 55.

And, if their study doesn't apply to low income groups, does it actually apply to like semi-low-income? They're literally pulling every trick possible to generate evidence against the (obvious) theory work, but they still can't find anything to undermine the result for poor people. For some reason they think this doesn't matter, but honestly it makes me feel like their research doesn't matter. Especially when the title of their paper omits what I consider the most important part. I guess if the title were "No meta-analytical effect of economic inequality on well-being or mental health except for poor people", then nobody would care/fund. Sensational titles get sensational funding.