That there isn't evidence proving this relationship doesn't make it false.
In scientific method everything is false until you can prove the claim. Its been tried and never has been proven.
I suspect that the models most people dive are so conservative that instances of DCS are sufficiently rare to mask the relationship.
So there is no statistically significant evidence then. Same thing. So if it makes any difference at all its too small to be picked up in any studies. Going from that i'd say you cant claim it improves the safety margin.
To make such a claim you need to supply proof.