I understand where you're coming from too, and I realise that this sounds like splitting hairs, but what science has demonstrated is that in a single large RAT, the median overall survival (the point when half of participants had died) was 7 months later for people who exercised more.

That's a very important step towards demonstrating causality — I'm not dismissing the science at all — but it's not enough to get us to reasonable certainty. Despite all the best efforts in an RAT, I'm sure the researchers themselves would be the first to admit that there's still a significant risk that they're seeing correlation more than causality. Repeating the result in at least a couple more large scale RATs will be needed to be confident of causality.

My point is that there's no need to wait for that additional research, because exercise brings other demonstrated benefits regardless, so it's a win for those, and also a potential win for the future if additional studies bear out the first findings.