Apart from number of digits (obvious big difference here if it was 40.41 vs 40.42?) I think that the numbers hide the fact that it's more like "I'd bet 5 bucks

Apart from number of digits (obvious big difference here if it was 40.41 vs 40.42?) I think that the numbers hide the fact that it's more like "I'd bet 5 bucks it's somehow positive" than a definite result you can pin your hat on. On the other hand, I think this study should be the poster child of "please don't use p-values" movement. Everything significant, but floats around zero.