This study tested a modified version of the Theory of Planned Behavior (TPB) for predicting both safer sex intentions and actual engagement in safer sex in a sample of HIV-negative heterosexual methamphetamine users. According to this model, the three constructs of attitudes toward condoms, social norms regarding condom use, and perceived control beliefs for safer sex behaviors should significantly predict one’s intention to engage in safer sex (Ajzen, 1991
). Our results provide clear support for this model. Specifically, we found that more positive attitudes toward condoms, greater expectations from their peers to engage in safer sex behaviors, and greater control over negotiating safer sex and using condoms all significantly predicted intention to use condoms during sex. Further, our results expand this model by identifying 3 additional factors that predict intention to engage in safer sex in this population. Specifically, greater desire to stop unsafe sex was associated with greater intention to practice safer sex, whereas higher levels of methamphetamine use and greater intentions to engage in sex were associated with reduced intention to practice safer sex behaviors.
Also consistent with the theory of planned behavior, safer sex intentions significantly predicted behavioral outcomes (i.e., future safer sex behavior). That is, individuals who reported more intention to engage in safer sex actually used condoms a greater percentage of the time they had sex. We acknowledge that these data are correlational in nature, and therefore urge caution in interpreting our results as causal. Nonetheless, these results may have implications for the development of interventions for increasing safer sex practices in methamphetamine-using individuals. First, it would be interesting to test whether or not reducing methamphetamine use increases intentions to practice safer sex, which in turn decreases risky sexual behaviors. That is, interventions designed to reduce risky sexual behavior may wish to incorporate cognitive-behavioral and contingency management components for reducing methamphetamine use. Indeed, others have demonstrated the efficacy of a cognitive-behavioral intervention with contingency management for reducing methamphetamine use and sexual risk behaviors in a population of methamphetamine-dependent gay and bisexual men, although the authors did not examine whether change in methamphetamine use mediated change in sexual risk behavior (Shoptaw et al., 2005). Future studies may want to examine this effect.
We found that desire to stop risky sex behaviors was associated with intentions to use condoms during sex. We believe this finding is interesting because it suggests that motivational interviewing (MI) (Miller & Rollnick, 1991
) may be useful for reducing risky sex behaviors in this population. In particular, MI seeks to produce “change talk”, or a desire to change, which in turn strengthens one’s commitment to actual change. We believe use of MI to elicit “change talk” for risky sexual behaviors in methamphetamine-using individuals may be an important component of risk-reduction interventions. MI may be used prior to an existing intervention for reducing sexual risk behaviors or can be integrated into these interventions as a means of augmenting their effects. Again, our study was correlational, and these suggestions should be tested scientifically to determine their merit.
Weaker intentions for having sex were one of the best predictors of intention to engage in safer sex behaviors. Specifically, individuals who did not intend to have sex were more likely to report intentions to engage in safer sex behavior. While this intuitively makes sense, other factors may also have played a role in this relationship. Specifically, individuals who had lower intentions to have sex also tended to report stronger social norms for safer sex (r = -.15), reduced use of methamphetamine (r = .15), and increased desire to stop unsafe sex (r = -.12). Therefore, it seems likely that a combination of these factors played a role in explaining this relationship. Researchers may wish to further examine factors that explain a relationship between sex intentions and safer sex intentions.
There are limitations to this study, including the volunteer sample and high attrition rate. Indeed, of participants completing baseline measures, 46% did not complete their follow-up measures. This prevented us from testing the full model in all participants who completed baseline measures. While it would be ideal to know how our expanded model performs in predicting safer sex intentions and sexual risk behavior among a full sample of participants, we believe the accuracy of our model for predicting safer sex intentions among the sub-sample of 228 participants (i.e., 48.2% of variance explained) suggests this model may perform particularly well. Nonetheless, as discussed in our previous manuscript (Mausbach et al., 2007
), methamphetamine-using individuals are difficult to track over time, and 20% of participants in this sample were incarcerated over the course of the study. Those participants who did not complete follow-up assessments reported less desire to stop unsafe sex, suggesting this sample was “treatment seeking” and more likely to change sexual behavior over the 6-month study period. Whether these results apply to non-treatment seeking methamphetamine users is unclear, and future research should replicate these results in this sample.
In sum, we find strong support for a model of sexual risk behavior in a sample of HIV-negative heterosexual methamphetamine users, as indicated by our model explaining nearly 50% of the variance in safer sex intentions. In addition to increasing efficacy for, altering normative beliefs and improving positive attitudes toward safe sex behaviors, our model suggests additional targets for interventions designed to increase safer sex practices in this population. These targets include increasing one’s desire to change sexual risk behavior, reducing amount of methamphetamine used, and reducing depressive symptoms. We anticipate interventions incorporating techniques targeting these components will demonstrate efficacy for increasing safe sex practices in this high risk population.