Similar to the limited findings reported in the adult HIV prevention literature on nonchange in preventive behaviors after intervention [13
], nearly half of adolescents who participated in a demonstrated efficacious STI/HIV risk-reduction program did not increase their condom use after intervention. Those reporting no increase in recent condom use at the six-month followup assessment were more likely to have higher levels of sensation seeking. In addition, adolescents whose condom use did not improve were more likely to have a current boyfriend and a physical abuse history. Additionally, those who did not increase their recent condom use after intervention reported higher levels of fear of condom negotiation across time and no significant increase in partner communication self-efficacy from baseline levels after intervention participation.
Our findings are consistent with research examining the relationship between sensation seeking and sexual risk taking [35
]. However, our findings extend this literature by demonstrating that individuals with higher levels of sexual sensation seeking are at increased odds of not increasing their condom use behavior after participating in a demonstrated efficacious STI/HIV prevention intervention. Research has shown that people high on sensation seeking tend to evaluate risky activity as less risky than those with low or moderate levels of sensation seeking [36
]. Moreover, research on the association between sensation seeking and message manipulation suggests that it may be beneficial to alter intervention strategies to affect the sexual risk taking of high-sensation seekers because they respond to stimuli differently than low-sensation seekers [38
]. Specifically, high-sensation seekers tend to require stronger, novel, and highly arousing messages to hold their attention, whereas low-sensation seekers tend to respond better to more familiar and less intense stimulation. Additionally, high-sensation seekers are more likely to engage in risky behavior to seek out or enhance pleasure, whereas low-sensation seekers might participate in risky behavior for different reasons including the desire to be liked or fit in with one's peers [39
]. Thus, future STI/HIV prevention efforts for adolescents may consider screening for sensation seeking tendencies and testing alternate styles of presenting intervention content to best reach youth with higher levels of sensation seeking.
Adolescents who were in a current relationship with a boyfriend were also less likely to increase their condom use after participating in the intervention. Young women who are in established relationships where condoms have not been consistently used may have greater difficulty negotiating future condom use with their partners [40
]. Additionally, it may be that young women with higher-sensation seeking levels select sexual partners that pose greater risks for adverse sexual health outcomes (e.g., partners with STI history) [36
]. These relationships may also reflect a power imbalance between partners, such that young women have less ability to negotiate condom use [17
]. Future studies should further examine the extent to which specific partner characteristics impact adolescents' responsiveness to sexual health programming. STI/HIV prevention interventions may benefit from providing specific strategies to negotiate condom use in established partnerships where condom use is not normative. Additionally, young women who do not improve recent condom use after participating in an STI/HIV prevention intervention may also need further individualized contact to address specific barriers that exist within the context of their current relationship.
According to the theory of gender and power [17
], the experience of abuse disempowers women to negotiate safer sexual practices in their current sexual relationships because they may fear possible ramifications by male partners [41
]. The findings of this study provide some support for this position by demonstrating that young women reporting a history of physical abuse are at marginally greater odds of not increasing their recent condom use after intervention. Furthermore, young women who did not increase their condom use after participating in HORIZONS reported higher overall levels of fear of negotiating condom use in general, as well as no significant increase in partner communication self-efficacy scores after intervention. Thus, similar to successful interventions designed to reduce traumatic stress and sexual risk among people living with HIV who have histories of abuse [44
], STI/HIV prevention interventions for young women may consider providing more in-depth discussion and instruction on specific strategies to manage and overcome fear or anxiety related to past abuse, as well as fear/anxiety about being assertive in current sexual situations. Doing so may improve the efficacy of STI/HIV prevention programs for those who have a history of abuse.
Many STI/HIV prevention programs designed for young women, including HORIZONS [5
], include a small component on unhealthy relationships (i.e., identifying abusive relationships). However, they may not thoroughly address the multitude of additional factors stemming from prior abuse (i.e., fear and anxiety) that may be hindering young women from living both emotionally and physically healthy lives. Thus, interventions should increase awareness of the co-occurrence of abuse, fear, and sexual risk behaviors, focus on developing cognitive and behavioral skills needed to accurately appraise risk, identify triggers associated with negative affect and sexual risk-taking, and develop strategies to avoid situations that trigger engaging in sexual risk taking.
Specific to the psychosocial mediators targeted in HORIZONS, overall we found that regardless of whether participants did or did not increase their condom use behavior after intervention, they did increase their STI knowledge and condom use skills after participating in the HORIZONS workshops. These findings suggest that adolescents benefited from the specialized activities targeted to increase knowledge and condom use skills in the workshop. However, increases in partner communication self-efficacy scores over time were primarily observed for those that increased their condom use over time. This suggests that, again, partner-level factors may hinder young women from feeling self-efficacious to communicate about sexually related topics, including condom use, with their partners. We did observe significant increases in sexual refusal self-efficacy after intervention but only for the nonchange group. This finding is counterintuitive and requires further exploration, but it should be noted that the nonchange group's followup levels of sexual refusal self-efficacy were comparable to the baseline levels of the change group.