I know that middle-brow dismissals are frowned upon but I want to write this anyway: I will never believe these types of studies without either a control group or better measuring of confounding variables. Health & happiness and marriage both have giant confounding variables like meeting a great partner and being a good partner. The authors claim that they accounted for this but I don't see how. The study is here[1].
> an unmeasured confounder would need to be associated with both incident marriage and mortality by risk ratios of 2.45 each, above and beyond the measured covariates, to fully explain away the observed association between marriage and mortality.
I may be interpreting that incorrectly and would appreciate corrections but why is it so unreasonable that meeting a great partner and having kids and thus wanting to strive to be healthier happier person would decrease all-cause mortality by a factor of 2.45 over someone who hadn't accomplished those things?
> an unmeasured confounder would need to be associated with both incident marriage and mortality by risk ratios of 2.45 each, above and beyond the measured covariates, to fully explain away the observed association between marriage and mortality.
I may be interpreting that incorrectly and would appreciate corrections but why is it so unreasonable that meeting a great partner and having kids and thus wanting to strive to be healthier happier person would decrease all-cause mortality by a factor of 2.45 over someone who hadn't accomplished those things?
[1] https://www.sciencedirect.com/science/article/pii/S259011332...