Conflating contemporary events and calling them causes and effects is a risky business, a sure way to make statistics lie.

The correlation, I submit, is not in the sexuality of the couplings but in the frequency of anal sex. At least, this seems much more probable to me, as it obviates the need to discover some mysterious aspect of homosexuality that spontaneously generates the disease when it occurs. That's just magic-think--there is no evidence I have ever seen that there is some cellular or molecular kickback mechanism that makes AIDS happens if intimate same-sex contact occurs. Your numbers also do not differentiate between Male Sex with Males (high probability of using the anal approach) versus Female Sex with Females (lower probability of same due to easier plumbing).

Nor does your assertion take into account frequency between multiple-partner series and same-partner series, which is an important differentiator.

I think it far more probable that the correlated vectors would turn out to be promiscuity and anal penetration, rather than homosexuality itself.

A