The study you're citing doesn't actually have a methodology that one can interrogate about how it reached this figure, and there's some fairly common patterns like opening a marriage in response to infidelity, that would create a misleading correlation between open marriages and unstable ones.

That's vastly different from entering into a relationship with the intent to be non-monogamous (and perhaps already having multiple partners), but is probably much more common, because, for whatever reason there's less of a taboo on sleeping around while married than on having multiple committed romantic relationships.