A new study modeling results from other data sources indicates that well over half of new HIV infections in five major metro areas (Baltimore, LA, Miami, San Francisco, and NYC) among men who have sex with men can be linked to their primary partners (e.g. boyfriends, lovers, husbands, etc.).

Here's the basic 411:

Methods: We modeled HIV transmissions, using data from MSM in five US cities from the National HIV Behavioral Surveillance System, the HIVNET Vaccine Preparedness Study, and other published data. Annual HIV transmissions were estimated by partner type (main or casual) and by sex type (receptive anal intercourse, insertive anal intercourse, or oral sex).

Results: Sixty-eight percent [95% confidence interval (CI) 58-78) of HIV transmissions were from main sex partners because of a higher number of sex acts with main partners, more frequent receptive roles in anal sex with main partners, and lower condom use during anal sex with main partners. By sex type, 69% (95% CI 59-79) of infections were from receptive anal intercourse, 28% (95% CI 19-38) were from insertive anal intercourse, and 2% (95% CI 0-5) were from oral sex. The model-based estimated HIV incidence rate was 2.2% (95% CI 1.7-2.7) per year. Sensitivity analyses demonstrated estimates of transmission from main sex partners as low as 52% (95% CI 41-62) and as high as 74% (95% CI 68-80).

Conclusion: According to our model, most HIV transmissions among MSM in five US cities are from main sex partners. HIV prevention efforts should take into account the risks of HIV transmissions in male partnerships, and couples-based HIV prevention interventions for MSM should be given high priority in the US HIV prevention research portfolio.

We've known for sometime that a great deal of HIV transmission risk can be located not just within anonymous partners (as the sex panic around HIV would indicate), but also within their primary relationships. I talk to many men who engage in short-term, serial monogamy and opt out of condom use very quickly in the relationship. I didn't find it in my quick read of the article, but it's not immediately obviously how "main partner" got defined as they use it -- e.g. how long a relationship would qualify.

Also of note here is the grave disparity between RAI (receptive anal intercourse) risk and IAI (insertive AI) risk -- 69% of infections were the result of RAI, while 28% were the result of IAI. Obviously, reporting is a concern here (stigma may push some to report IAI instead of RAI) -- but it's worth noting the chasm between the two. Although as someone mentioned in the discussion on CHAMP's listserv, this disparity isn't as a gross as we might expect it given the data we generally rely on that says that the risk disparity is more like a factor of 10 difference.

In any case, we should seriously consider implications for this data in the way we approach prevention in the US. Australia started doing couple prevention eons ago (the famous "Talk. Test. Test. Trust." campaign), and the US is quickly catching on -- there are numerous efforts being developed or already underway. But data like this highlights that the epidemic doesn't neatly fit many people's preconceived notions that link STI's with unabashed promiscuity.

If you have journal access, you can find the article here in the journal AIDS.