I would just caution people to understand that saying that something is "controlled for" in an observational study (which is something I do all the time, I'm no

I would just caution people to understand that saying that something is "controlled for" in an observational study (which is something I do all the time, I'm not saying it's bad) is not the same as saying that we've eliminated confounders, leaving a causal effect.