Individual and Neighborhood Factors Associated with Sexual Risk Typologies in Two Urban African American Longitudinal Cohorts

Introduction: While previous work suggests that adolescent sexual risk behaviors are complex and multidimensional, few previous studies have identified sexual risk profiles among urban youth despite their disproportionately poor sexual outcomes. Moreover, our understanding of the risk factors for these profiles is limited. In order to prevent the negative outcomes associated with risky sexual behaviors among men and women who have grown up in an urban environment and reduce health disparities, it is critical to identify factors throughout the life course that predict specific sexual risk profiles.

Methods: Utilizing two urban cohorts followed longitudinally from ages 6 to 19-20, we first identified sexual risk typologies. We applied latent class analyses separately for males (n=777) and females (n=926) on seven age 19-20 behavioral indicators – ever had sex in one’s lifetime, early sex (before age 14), currently sexually active, multi-partner sex, sex without a condom, sex while intoxicated, and exchanging sex for goods or money. Next we validated these classes by examining their association with sexual and behavioral outcomes (e.g., pregnancy, sexually transmitted infections, substance use). Finally, we examined factors across the life course that differentiated these typologies, including individual and neighborhood factors.

Results: For women fit statistics indicated a three-class solution: a no risk class (28%), a moderate risk class (57%) characterized by sex without a condom with a single partner and a high risk class (28%) characterized by sex without a condom, sex while intoxicated, and multi-partner sex. For men fit statistics indicated a four-class solution. An estimated 23% of the men were in a no-risk class; 14% were in a class with no current risk, but initiated sexual activity early, before the age of 14. The multiple risk class (11%) was characterized by early sex, unprotected sex, sex while high or intoxicated, and multiple sexual partners in the past month. Like women, the largest class for males (52%) was characterized by condomless sex with a single partner. Both census data and individual reports of neighborhood differentiated classes for men and women, including neighborhood composition and community disadvantage index. For individual factors, high residential mobility and aggressive behavior were the most salient predictors for both men and women.

Discussion: The results of this study offer evidence to improve interventions targeting sexual risk behavior for urban youth by identifying gender-specific risk factors at various levels associated with specific high-risk behavioral profiles. This knowledge will allow for a more tailored and efficient approach to prevention to ultimately reduce health disparities in our urban centers.