Papers of the Week: 25/2012

Clearly much water (23 weeks worth, apparently) has passed under the bridge since I last posted on new papers crossing my rss/email desk.  So here are the latest batch:

  1. I am more than a little fascinated by the interplay of race, SES, gender and any other stratifier you can mention in determining infection risks.  And in the field of STIs, I appear not to be the only one.  One angle on this is to look at multiple low-power identities and see how they interact.  For example, here‘s a paper that focuses on the intersection of race/ethnicity and sexual orientation.  Risks definitely rise with multiple minority statuses, but the pattern is non-simple and varies by sex.  It’s never as simple as you’d think.
  2. More on concurrent partnerships in Africa.  A recent study found no association between (self-reported) concurrent partnerships and HIV incidence in rural South Africa and while it is generally accepted that concurrency can theoretically drive an HIV epidemic, empirical evidence remains scant that it does so.  One reason for this may be that while concurrency is risky, its prevalence is low.  This idea is supported by a paper out of Malawi from last year which finds that concurrency is long-lasting when it occurs, but that it occurs infrequently (only in 9% of the sample).  Lots more evidence is needed on this, but these are the right questions to be asking.
  3. This one is only “new to me”. Dynamic models of sexual relationships (and other contact networks, I think that sexual networks are simpler than most) need to be a big new field in ID Epi.  If that’s going to happen we (public health people) are going to need to read lots of network analysis stuff (aside: here‘s an intro from Nick Christakis and collaborator Kirsten P Smith (meta-aside: Smith’s disseration from Penn looks really interesting – international comparisons of STI rates, but I can’t see it published yet).  If you want more theoretical details, James Moody – one of the key people behind the Add Health network work mentioned in the above article – wrote a paper in Social Forces a decade ago, which outlines things nicely.  Beach reading, if I ever made it to the beach.  And enjoyed reading once there.

Next time I’ll try and make these papers a little more up-to-date, but this’ll have to do for now.

Concurrency vs. Rapid Partner Change

So here’s an interesting paper. The authors looked at self-reported Sexually Transmitted Infections (STIs) and self-reported sexual history over the past year in a cross-sectional study in St. Petersburg, Russia. This design might raise a red flag or three regarding the reliability of the data, but unless you can tell me how misreporting is likely to be correlated with (what I think is) the key finding, that doesn’t undermine their work all that much.

What they report is that short relationship gaps (1-90 days) put people at higher risk than overlapping (concurrent) relationships. The authors suggest:

The present study’s findings may seem counterintuitive, but we might explain them with a scenario in which people with short partnership gaps have more frequent sexual intercourse with their STI-infected partner than do individuals with overlapping relationships, provided that both groups have similar rates of condom use, which was the case in our study

Specifically, they argue that concurrent partnerships require you to be having sex with multiple people at once, and thus you can’t give each partner as much attention as you can in a serially monogamous relationship.

I thought that this paper was interesting because it helped me to remember that while concurrency is dangerous for STIs, closely packed serial relationships can be just as bad, if the gap between relationships is less than the incubation window period plus the period of high infectiousness around breakout. This is especially true if, as seems reasonable (although the interwebs/PubMed is not helping me with longitudinal evidence of sexual behaviour within relationships – sounds like a paper to me) people have more frequent sex at the beginning of relationships.

An Aside: Interestingly, another paper from the same study noted that your partners’ concurrency status is associated with your disease status, but given that, yours doesn’t actually matter. As ever, who you’re involved with is more important than how many you’re involved with…

Paper: Zhan W, Krasnoselskikh TV, Golovanov et al. Gap between Consecutive Sexual Partnerships and Sexually Transmitted Infections Among STI Clinic Patients in St Petersburg, Russia. AIDS and Behavior, 2011; epub ahead of print. DOI: 10.1007/s10461-011-9932-z. Link (gated).