[fshbbdssbbgdd]

It’s good of you make this correction, we need to be accurate when discussing the research at hand.

That said, it’s worth pointing out that “adjusting for” a factor isn’t quite as effective as it sounds: https://dynomight.net/control/

Having built some regression models in my time, throwing a ton of extra variables in doesn’t fill me with confidence that I am “controlling” much of anything.

[roughly]

Following from the above, I'll presume you've got a specific methodological critique based on the process detailed in the paper?

[fshbbdssbbgdd]

The paper in question: https://jamanetwork.com/journals/jamanetworkopen/fullarticle...

It’s a research letter, which is peer-reviewed by JAMA but shorter than a traditional, full-fledged research article.

The full text of the letter’s description of the adjustments:

> with adjustment for known and suspected risk factors for depression, including

[followed by a list of over a dozen factors]

The data sharing supplement says that the code is on GitHub (hmm!), but doesn’t include a link (oh…).

That’s all I have to work with. I can’t make a specific criticism of the adjustments in this study. I’m left with my preconceived notions of how this kind of research is often done, even with peer-review.

When I read the words “we controlled for X in our population study of Y”, I usually take the lazy shortcut of not updating my internal model of the relationship between X and Y at all, because I’ve fooled myself too many times before. If they mention a natural experiment or instrumental variable I get excited.