AIDSImpact.com:: - Santa Fe, USA 2011

Abstract #128 - Reducing concurrency for HIV-prevention: modeling behavioral risk and coital dilution
Authors: Presenting Author: Prof. Larry Sawers - American Unversity
Additional Authors: Prof. Eileen Stillwaggon,
Aim: Background: The concurrency hypothesis asserts that high prevalence of long-term, overlapping sexual partnerships explains extraordinarily high HIV levels in sub-Saharan Africa. For the hypothesis to be valid, mathematical modeling must show that concurrency spreads HIV more effectively than other sexual behaviors. Many models account for the network effect of concurrency that increases HIV incidence, but do not account for the coital dilution effect (non-primary partnerships have lower coital frequency than primary partnerships), that, other things equal, reduces HIV incidence. Elsewhere we demonstrate the weak empirical basis of the hypothesis (JIAS September 2010).
Method / Issue: Methods: We adapt the model of Eaton et al. (AIDS and Behavior September 2010) to accommodate coital dilution by assigning lower coital frequencies to non-primary partnerships based on the empirical work of Morris et al. (PLoS ONE December 2010) and others. As do Eaton and colleagues, we iterate the model daily for 250 years at 12 levels of concurrency (from zero to two-thirds of partnerships concurrent).
Results / Comments: Results: At every level of concurrency, the modified Eaton et al. model leads to epidemic extinction. The model is seeded with 2 percent HIV prevalence. Prevalence falls to zero in 25 to 100 years depending on the level of concurrency. We modeled different decision rules about how to designate a partnership as primary. We performed sensitivity analysis that shows that even modestly lower coital frequencies in non-primary partnerships lead to epidemic extinction.
Discussion: Conclusion: Concurrency cannot play a role in HIV epidemics in sub-Saharan Africa because it cannot spread HIV more rapidly than other forms of sexual behavior. Our modeling, which includes both the network and coital dilution effects of concurrency on HIV incidence, shows that HIV epidemics quickly become extinct even at high levels of concurrency. Thus, concurrency reduction is not an appropriate focus of HIV-prevention strategies.
Go Back