Whew, I'm glad you don't think it's causal! I wasn't necessarily saying you thought that -- but by "linking", I meant saying that type had anything to do with chronic pain OR saying that chronic pain has anything to do with type.
To clarify: it's not that you're saying that they're more likely to have chronic pain -- it's that they're more likely to report that they do? If we're going to rationalize the idea of linking functions to that, I'd say that Se would make more sense. And straying from MBTI -- the Sp instinct would make more sense as linking to that.
But I wouldn't be questioning the correlation if I knew that the source of your statistic was trustworthy -- and you didn't answer my question before. So..?