Liaisons dangereuses: HIV risk behavior and prevention in steady gay relationships

Abstract

This dissertation studied sexual risk-taking behavior among gay men in steady relationships. The main targets of this study were to establish: (a) whether steady relationships form a risk environment for HIV-infection; (b) some of the determinants of risky and protective behavior between steady partners; and (c) an effective way to reduce sexual risk-taking in steady relationships. The following provides an overview of the main conclusions of the dissertation:

Steady relationships as a risk environment for HIV-infection

The rate of sexual risk-taking in relationships was established by measuring the rate of unprotected anal intercourse (UAI) between steady partners and correcting it for negotiated safety (i.e. testing for HIV and reaching agreements about safe sex outside the relationship) and negotiated safety compliance. The latter refers to whether men have UAI with casual partners outside their relationship while they have agreed to practice negotiated safety. The rates of risky UAI between steady partners were high (39%) and these rates are significantly higher than the rates of UAI with casual partners (20%).

The share of steady versus casual partners as the source of HIV infection was examined in a pool of seroconversions between the years 1984 and 2000 in the data of the Amsterdam Cohort Studies among Homosexual Men. We found that younger gay men who seroconverted later in the AIDS epidemic were more likely to have contracted HIV from a steady partner than from a casual partner. The steady partner was the source of HIV infection in 67% of the seroconversions between 1994 and 2000 among younger gay men.

Determinants of risky and protective behavior in steady relationships

To examine the determinants of risky and protective sexual behavior in relationships, we consistently applied regression analyses with one multinomial outcome variable examining the contrasts between risky UAI and one of the following protective behaviors: negotiated safety, condom use, and avoidance of anal sex. Results indicated that low relationship satisfaction was associated with more risky UAI and that high commitment to the relationship was associated with more practice of negotiated safety, but not with more condom use. Perceiving UAI as a symbol of trust and believing that the partner desired UAI were associated with less condom use but with a higher likelihood that men practiced negotiated safety. Anticipated sexual excitement from having UAI was associated with risky behavior in steady relationships, but that this perceived excitement was, surprisingly, not associated with the practice of negotiated safety.

The first incident of risky UAI in relationships occurs mostly within the first three months of the relationship, and almost half do not discuss having UAI with their partner before it occurs.

Reducing sexual risk-taking in steady relationships

To reduce sexual risk-taking in steady relationships we developed an online intervention that targeted single gay men and promoted the practice of negotiated safety with future steady partners. The Internet intervention content was based on empirical data and the intervention modules were operationalized according to the Information, Motivation, Behavioral-skills (IMB) model.

We tailored the intervention based on the individual needs of the participants, according to their informational-, motivational- and perceived-skill deficiencies for practicing negotiated safety. The Internet was chosen as the communication tool for the intervention. The Internet enabled us to obtain the data necessary for tailoring and directly present men with the tailored advice – all during the same online contact.

To examine the efficacy of this tailored intervention, we conducted a randomized controlled trail online examining the effect of a tailored versus non-tailored version of the intervention. Results showed better cognitive impact for the tailored intervention than the non-tailored one: men in the tailored condition had higher intentions to practice negotiated safety with future steady partners. Most importantly, only the tailored condition brought about behavioral change at the six months follow-up: men in the tailored condition where more likely to practice negotiated safety with their new steady partner. The intentions to practice negotiated safety at time 1 mediated the effect on behavior at follow-up.