The problem I have with this study is that they don't define what is classified as abused and they don't show the sample size. Even the article mention, I quote : "it's worth nothing that the sample size are a bit small". But they don't mention the sample size and that matters a lot.

Abuse could be : have your husband scream at you in the last 12 month. And then the wife think : I screamed at him because he overcooked the chiken and told him he's an idiot and he screamed back telling me I just have to cook it myself. And then they check yes in the question and they count her instance as police abuse.

Or Abuse could be : As your husband hit you in the past 12 month. In which case it would be quite relevent.

But if they don't tell us how they have defined abuse, there's just no way to know so the interpretation of that study is meaningless.

As for the sample size, it define the margin of error which is also quite relevent to know, yet again they don't mention it.