March 31, 2009

Background:

On December 2-3, 2008, the US Centers for Disease Control and Prevention (CDC) sponsored a Consultation on Serosorting Practices among Men who Have Sex with Men (MSM). Serosorting is most often defined as a person choosing a sexual partner known to be of the same HIV serostatus, often to engage in unprotected sex, in order to reduce the risk of acquiring or transmitting HIV. The consultation sought to explore the extent of serosorting among MSM and the level of HIV risk associated with serosorting under different conditions. Approximately 25 external consultants attended this consultation, including social and behavioral scientists, public health personnel, clinicians, laboratory scientists, and community members from a variety of organizations. Participants reviewed available data and discussed implications for prevention messages as well as defined gaps requiring future research.

Presenters at the consultation presented on:

• Patterns of serosorting among persons infected with HIV, including a review of the biology of sexual HIV transmission, with an emphasis on the likelihood of infection with an additional strain of HIV (superinfection) and other potential harms such as sexually transmitted infections (STIs).
• Serosorting among persons who are not infected with HIV as well as “negotiated safety,” a strategy studied in Australia for committed partners not infected with HIV.
• Perspectives on serosorting from three MSM community members who discussed their real-world experiences and challenges to help contextualize the scientific data.

Summary

Broad issues discussed during the consultation included the definition of serosorting and related terms; the extent to which serosorting is intentional; the effectiveness of serosorting given that decisions may be made based on assumptions, rather than knowledge, about serostatus; and the complex contextual environment (relational, psychosocial and structural) for MSM in the United States. Consultants discussed the meaning of various terms, such as serosorting, strategic positioning, seroguessing, and seroadaption; when and how those terms are being (or should be) used; and the risks associated with some of these behaviors. The consultants perceived seroadaptation to be a broader and more inclusive term than serosorting and, more consistent with the broader goal of improving the sexual health of gay and bisexual men in the United States. Discussions also focused on possible ways to frame prevention messages, including breadth and content of the messages as well as specific messages for partnerships in which both members are infected with HIV, partnerships in which both partners are not infected with HIV, and partnerships in which the members are serodiscordant or serostatus is unknown. General concern was expressed that using serosorting in some particular contexts would not reduce HIV risk and might actually increase it.

The concern with regard to unprotected sex between men who are infected with HIV focused on the potential for increased STIs and superinfection. While STIs are much more common among MSM who are infected with HIV than among those not infected with HIV, there is not yet a demonstrated difference in STIs between those who practice serosorting and those who do not. Untreated STIs were discussed as factors that could increase risk both to the overall health of men who are HIV infected and for superinfection. The consultants discussed the benefits of STI screening and treatment to reduce transmission from men infected with HIV.

Regarding superinfection, the consultant noted that few cases have been reported in the literature and that the risk appears highest soon after seroconversion and may decline over the course of infection. Some data discussed by the consultants also indicate limited clinical consequences of superinfection. However, the consultants noted the lack of research among African American and Latino men and on superinfection in general. The consultants discussed how superinfection risks should be framed in communications to patients and the larger community. The consultants felt that there is a great deal of public confusion on the risks, and expressed the belief that the medical and prevention communities may have overemphasized the likely risk of superinfections, as well as the severity of its consequences.

Regarding serosorting among men who are not infected with HIV, presenters and consultants highlighted conflicting data and opinions. On the one hand, HIV transmission cannot occur between two seronegative persons, and serosorting is a community-developed strategy that is already operating in some MSM networks and might be enhanced through public health research and messaging. However, ascertainment of serostatus can be difficult, especially because MSM in the United States, particularly African American and Hispanic/Latino MSM, have high rates of HIV prevalence and incidence, and large percentages of unrecognized infection. The consultant expressed support for research into home-based HIV-detection technologies that might help in some cases of ascertainment serostatus (but could be problematic with very recent infection). The consultants again noted the lack of research data about serosorting practices among African American and Latino MSM.

The consultants also discussed Australian data on a negotiated safety strategy. Important elements of negotiated safety are captured by a marketing campaign titled “Talk, Test, Test, Trust” and include at least a 6-month relationship, shared knowledge of HIV-negative serostatus supported by two tests separated in time, clear agreement about sex inside and outside the relationship, and sex with casual partners being “safe.” Data presented showed a protective effect among MSM who were not infected with HIV and who strictly followed its tenets. The limitations of negotiated safety were discussed with casual or one-time partners as well as early in relationships, when changes occurred in long-term relationships, and when negotiated safety agreements were not revisited. Additionally, the transferability of these findings to the United States was discussed in light of the racial/ethnic and HIV epidemic differences between the United States and Australia. Some of the prevention topics that consultants discussed regarding serosorting between men who are not infected with HIV, or perceive themselves not to be infected, included:

• The recommended frequency of HIV tests;
• The technological limits for detecting acute HIV infection;
• An understanding of what prevention options are being replaced by serosorting;
• Whether serosorting practices vary by partner type (main/casual);
• The heightened risk of seroconversion over time among MSM who are not infected with HIV but who practice serosorting;
• The potentially detrimental consequences of higher-risk MSM networks overlapping with lower-risk MSM networks;
• The role of condoms in serosorting.

Conclusion:

The consultation highlighted the need to clearly define seroadaptation practices, and particularly serosorting, among MSM, and to understand the differing risks based on a variety of contextual factors. Several areas for future research were recommended. The information obtained in this consultation will inform potential future CDC activities, the development of prevention messages or guidance documents, and the creation of future funding announcements. As follow-up to this consultation, CDC will develop a fact sheet on serosorting among MSM that will include definitions of various seroadaptation practices, the state of the science with regard to behavioral patterns and their potential HIV risk under different situations and with different types of partners, and recommendations for future research. The fact sheet will be released later this year.