186. Early Life Origins of Risky Sexual Behavior in Adolescence

Abstract

Purpose Risky sexual behavior contributes to unplanned teenage pregnancy, sexually transmitted infection (STI) and adverse social, emotional and physical health outcomes in adolescence into adulthood. We have little understanding of early determinants of risky sexual behavior. This study aimed to determine the prospective relationship between problem behavior through childhood, as measured by the Child Behavior Checklist (CBCL), and risky sexual behavior in adolescence. Methods A prospective cohort study (the Western Australian Pregnancy Cohort [Raine] Study) has collected extensive socio-demographic, individual, family and environmental data from over 2800 parents, the children they gave birth to, and individuals in the children’s lives from the antenatal period, birth and every 1-3 years following. Approximately 1200 of these children (now adolescents) reported on sexual experience at the 16/17-year follow-up and were dichotomized into high and low sexual risk groups according to sexual intercourse debut and other sexual behavior. For each childhood follow-up, continuous CBCL z-scores were summarized using medians, inter-quartile ranges and ranges. Mann-Whitney U tests were used to compare outcomes between sexual risk groups. Further, CBCL raw scores at each follow-up were dichotomized with reference to published cut-off values (T scores ≥70), indicating clinically recognized behavioral problems. The Cox proportional hazards model was used to compare the relative risk of sexual debut over time between participants who were identified with behavioral problems at each childhood follow-up and those without. Results Initial analyses show significantly higher z-scores and T scores for Total Behavior Problems and the Externalizing Behavior subscale in participants who reported sexual intercourse debut compared to those who did not (p<0.05). In addition, participants who were identified with clinically recognized delinquent and aggressive behavioral problems during the 5, 8, 10 and 13-year follow-ups were significantly (p<0.005) more likely to have experienced sexual debut with hazard ratios ranging from 1.69 to 4.06 on univariate analyses. These associations remained significant in multivariate analyses when adjustments for multiple potential confounders at each age were considered. Conclusions The development of problem behavior in early childhood is an important contributor to risky sexual behavior in adolescence, but the nature and significance of this relationship is, as yet, unknown. These unique data indicating early predictors of risky sexual behavior in adolescence may help determine how and at what ages interventions may be effective. Sources of Support: NHMRC