In the community of men who have sex with men (MSM) the prevalence of the HIV-1 infection is still high. Promiscuity and condom fatigue are making unprotected anal intercourse (UAI) more common and sexually transmitted infections (STIs) presumably harder to track. Yet, MSM communities are peculiar in the sense that men can adopt fixed (insertive or receptive) or versatile (both practices) roles. Some old theoretical work predicted that the transmission of HIV-1 would be enhanced in MSM populations engaged more in role versatility than in role segregation, in which fixed roles are predominantly adopted. These predictions were based on the assumption that the probability of acquisition from unprotected insertive anal (UIA) sex was neglectable, which is an inappropriate assumption. This chapter shows that the increase of the HIV-1 prevalence among MSM due to role versatility holds under a stronger assumption of bidirectional virus transmission.
TopMethods: A Stochastic Model With Different Role Classes
The following is assumed:
- •
The MSM population practices UAI.
- •
There is random mixing among individuals and sexual-role classes.
- •
A corollary of the previous assumption is that there is not role assortativity, which is when versatile MSM preferentially choose other versatile MSM as partners, or when insertive MSM preferentially choose receptive over versatile MSM as partners, or when receptive MSM preferentially choose insertive over versatile MSM as partners. Versatile individuals have the same preference of being insertive or receptive when their partners are also versatile.
By defining 
as the prevalence of the HIV-1 infection at the MSM population (proportion of the population that is infected), and B, V and T as the proportion of males within the MSM population that have a receptive (B), versatile (V) and insertive (I) role (so that 
), the prevalence is given by:
(1) where
,
and
are the probabilities that an individual within the population is infected given that his sexual role is receptive, versatile or insertive, respectively. These conditional probabilities are defined as follows:
(2)(3)(4) where
is the long-term probability of acquisition from URA sex with infected insertive partners and
is the long-term probability of acquisition from UIA sex with receptive infected partners. In other words, the probability that an individual within the population is infected given his sexual role is defined by the expected prevalence of the infection at each of his potential-partners’ sexual-role classes (
,
and
) weighted by the long-term probability of acquisition from UAI with insertive
or receptive
infected partners.