Confounders are included and therefore controlled for, and causality can be assumed since the DV is an event/hazard. This is also the reason you end up with the

Confounders are included and therefore controlled for, and causality can be assumed since the DV is an event/hazard. This is also the reason you end up with the seemingly high average...