P3.064 Patterns of Sexual Partner Accrual Among Adolescent Women

Abstract

Background Younger age and number of sexual partners are consistently identified STI risk factors, but the relative contributions of chronological age and total interval since first coital partner to accrual of partners has not been established in prospective studies. Methods Participants (N = 341; 14–17 years of age at enrollment; 90% African American) were English-speaking adolescent women from lower- and middle-income families residing in areas with high rates of pregnancy and STI. We collected quarterly sexual behaviour data for an average of 3.56 years (range 0.23–8.92 years). Median number of lifetime partners increased from 2 at enrollment to 6 at the last visit. Generalized Additive Mixed Models were used to estimate the cumulative number of partners as a function of time covariates. The cohort effect (enrollment age), the longitudinal effect (follow-up period) and their interaction were considered as predictors. Annual rate of the partner acquisition was estimated by the first derivative of the cumulative number of partners with respect to the “follow-up period”. Results Partner accrual was 2 or more partners per annum within the first 2 years of follow-up, but slowed down to about 1 partner per annum during the years 2 to 5 since enrollment. Rates of partner accrual were similar for all age groups at enrollment and declined at similar rates. Correlation between “enrollment age” and “age at first sex” was 0.26, possibly an indicator of STI and contraceptive care-seeking associated with sexual activity. Conclusions Cross-sectional studies of STI risk among adolescents may confound age-related cohort effects with variation in interval since first sexual intercourse. Our analysis shows that higher rates of partner accrual decrease within two years regardless of the age of first coitus. This suggests that STI prevention based on delay of coitus among adolescent women should be supplemented with efforts to reduce rates of partner accrual.