Predictive Validity and Moderation Effects of the Strength of Implementation Intentions Scale for Condom Use

Introduction: The use of any alcohol or drug can affect judgment (lowering inhibitions) leading to high-risk sexual behaviors and increasing the likelihood of contracting HIV and other STDs (Trenz et al., 2013). Non-injection drug users are a frequently neglected population and are at substantial risk for HIV infection because of risky sexual practices (Crawford & Vlahov, 2010; Mitchell & Latimer, 2009). A promising and practical route for the prevention of HIV risk among drug users is through the use of implementation intentions, or situation-linked action plans, yet very few studies measure the strength of formed implementation intentions. Research has identified some moderating effects among implementation intentions and behavioral intentions, alcohol use, and amphetamine use (Hagger & Luszczynska, 2014; Rise et al., 2003; Sheeran et al., 2005). This study examined whether the Strength of Implementation Intentions Scale (SIIS) for condom use predicted condom use one year later, and if that relationship was moderated by behavioral intentions to use condoms, alcohol use, and amphetamine use.

Methods: Participants were 590 individuals 18 years or older in court-mandated outpatient drug diversion programs in Southern California. Participants took the baseline survey at the drug diversion facility and took the 1 year follow-up at a drug diversion facility or a location nearby. General linear model regressions were conducted and controlled for condom use at baseline.

Results: Results confirmed that the SIIS for condom use predicted condom use one year later (p < 0.05) and accounted for 9.66% of the variance (entire model accounted for 32.41%), even after controlling for condom use at baseline. However, the relationship between the SIIS and condom use one year later was not moderated by behavioral intentions to use condoms, alcohol use before or during sex, or amphetamine use before or during sex (p > 0.05).

Conclusions: The present study is the first known study to identify the predictive validity of any type of implementation intention assessment over an extended period of time. The significant predictions for condom use provide support for the future use of the SIIS for condom use as a fidelity check or mediator for interventions utilizing implementation intentions to increase condom use. These findings, coupled with the ability of the SIIS to detect intervention effects (Nydegger et al., 2013), imply that it can be used as an outcome measure in situations where a follow-up behavior is not feasible, such as drug diversion programs. Future research should examine various diverse populations to ensure the predictive validity, reliability, generalizability of SIIS, and tailor the SIIS to other health behaviors.