Alcohol intoxication and condom use self-efficacy effects on women's condom use intentions

Abstract

Although research has consistently demonstrated that condom use self-efficacy significantly predicts condom use, there has been little investigation of whether acute alcohol intoxication moderates this relationship. Because alcohol intoxication is often associated with increased sexual risk taking, further examination of such moderating effects is warranted. Using a community sample of young heterosexual women (n=436) with a history of heavy episodic drinking, this alcohol administration experiment examined the effects of intoxication and condom use self-efficacy on women's condom negotiation and future condom use intentions. After a questionnaire session, alcohol condition (control, .10% target peak BAL) was experimentally manipulated between subjects. Participants then read and responded to a hypothetical risky sexual decision-making scenario. SEM analyses revealed that alcohol intoxication directly decreased women's intentions to use condoms in the future. Women with greater condom use self-efficacy had stronger intentions to engage in condom negotiation; however, this effect was moderated by intoxication. Specifically, the association between condom use self-efficacy and condom negotiation intentions was stronger for intoxicated women than for sober women. These novel findings regarding the synergistic effects of alcohol intoxication and condom use self-efficacy support continued prevention efforts aimed at strengthening women's condom use self-efficacy, which may reduce even those sexual risk decisions made during states of intoxication.

Keywords: Alcohol; Condom negotiation; Condom use; Condom use self-efficacy; Sexual risk.

Figure 1

Dotted line represents hypothesized path that was not significant in the final model. Solid lines represent hypothesized paths significant at p < .05 in the final model. Dashed line represents path added to final model based on modification indices. Standardized path values shown.

Figure 2

Values of CUSE represent estimates at mean − 1 SD and mean + 1 SD. Control and alcohol slopes differ significantly (t = 2.01, p < .05).