Correlation or Causality?

New study from Melbourne psychologists finds that women who are depressed have more sex. The article seems to imply that the relationship is causal - that the depression triggers the need for sex. For instance, see this:

""When people are depressed they feel more insecure about their relationships and concerned that their partner may not care about them or find them valuable," Dr Allen said.

"Having sex helps them feel that closeness and security."

That's a completely credible theory. I can easily see how it would work. Sexual intimacy has the potential to solve a lot of loneliness problems. Heck, even in the Christian religion, you see this expressed in the first pages of Genesis when God says to the solitary Adam, "It is not good for man to be alone" and thus makes him a partner with whom he enjoys, among other things, a sexual relationship. Interestingly, even in the perfect world, it is not enough to simply be in communion with a loving God - man still needs a partner.

But, putting aside the credibility of the theory, it sounds like all the team really has uncovered is a correlation between depressed women and sex. How do they know that the causality runs in the direction of depression causing demand for sex? The article opens up with these two paragraphs:

A survey of Melbourne women presented at an international mental health conference has concluded that females who suffer from mild to moderate depression have a third more sexual activity than those who are not.

They also had more sexually liberated attitudes, a bigger variety of sexual experiences and, if single, were more likely to partake in casual sex, Dr Sabura Allen, a clinical psychologist at Monash University, said.

I wonder if the team considered the possibility that it was the engaging in promiscuity that was causing the depression, and not the other way around. Or perhaps there are feedback effects - that loneliness triggers demand for sex which, under certain conditions, can increase the loneliness and lead to a somewhat compulsive appetite for sexual partners. Point is, finding a correlation between two variables is a necessary but not a sufficient condition for proving causality. Psychology has had a major influence on economics, particularly with Daniel Kahneman and Amos Tversky's work, but if nowhere else, I think empirical psychology could stand to be influenced dramatically by our emphasis on the problems of endogeneity.