Constructing Latent Classes of Sexual Abstinence Motives Endorsed by South African High School Learners

Phylicia T Bediako, Linda Lee Caldwell, Stephanie T. Lanza, Lisa Wegner, Joachim Jacobs, Jacqueline A. Miller

Introduction: Black and Coloured (i.e., mixed race, or multiracial) South Africans represent a substantial proportion of individuals with HIV/AIDS in their country and worldwide. The higher burden of adverse outcomes in these two groups could be related to earlier sexual initiation and more prolonged exposure to sexual risk throughout the sexual life course. This study investigated how early adolescents maintained one type of risk-reducing behavior, sexual abstinence, despite living in a high-risk environment.

Methods: We analyzed data from 9,117 eighth grade learners who reported that they had never had sex. Latent class analysis (LCA) was used to construct distinct “profiles” based on eleven binary indicators representing learners’ reasons for sexual abstinence. Multiple group and covariate analyses were used to identify predictors of membership in motive profile classes.

Results: The highest-ranking motives for abstinence included future protection (82.4%), fear of STIs (82%), and fear of HIV (81.7%). The lowest-ranking motives included friends’ disapproval of sex (54.3%) and not wanting to have sex (57.1%). Based on fit statistics and interpretability, a 7-class solution was selected as the best fitting model. In descending size, the classes were All Motives, Low Social and Moral Reasons, No Personal Readiness Concerns, Protecting Future Only, Physical Reasons Only, No Motives, and No Physical Concerns. The results of the covariate and multiple-group analyses revealed that class membership was significantly associated with socio-demographic characteristics, parental presence, educational aspiration, loss of a close friend or relative, substance use, relationship status, and sex-related experiences such as HIV testing. Membership in the All Motives class appeared to be associated with decreased risk-taking and fewer risk factors compared to membership in the other classes. The Protecting Future Only class was most closely associated with risk factors for sexual behavior, even more so than the No Motives class.

Conclusion: We found variation in the nature of the abstinence motive profile classes as well as in the types of characteristics and behaviors that predicted membership in each of the classes. Understanding the connections between motives, experiences, characteristics, and behaviors offers the opportunity to identify potentially vulnerable groups and also address the portion of sexual risk that is related to engaging in other risky behaviors. Programs can attempt to weaken the link between other risk behavior and sexual risk before learners start engaging in sexual risk behaviors and support the maintenance of safe sexual risk behaviours.

This abstract was submitted to the 2017 Society for Prevention Research Annual Meeting