While a body of empirical literature describes relationships between teens' contraceptive use behaviors and individual-level characteristics, relatively few studies have examined influences of sexual partners on adolescent contraceptive use. This is a critical gap, given that contraceptive use occurs in the context of a relationship between two people. The purpose of this study is to examine sexual partner, sexual relationship and individual-level influences on contraceptive use consistency among sexually active adolescent girls at high risk for pregnancy and sexually transmitted infections. Data for this study are from 13-17 year old participants in Prime Time, a clinic-based intervention study aiming to reduce sexual risk behaviors and improve condom and contraceptive use among teen girls at high risk for pregnancy. The Prime Time intervention included 18 months of case management and a 16-week peer leadership program. 111 non-pregnant Prime Time participants completing study surveys at baseline (T1) and 12 months (T2) comprise the sample for this study (T1 data collected 9/99-12/02, T2 data collected 9/00 –1/04). Contraceptive outcomes, measured at T2, included consistency of condom use, consistency of hormonal method use, and consistency of any birth control method use; outcomes were reported using 6-month contraceptive calendars. Predictor variables assessing individual characteristics at T1 were girls' attitudes about pregnancy and birth control, perceived benefits of birth control, and number of male sex partners in the past 6 months. Predictor variables assessing partner and relationship characteristics at T2 were having the same sex partner at T1 and T2, age difference between partners, communication with partner about sexual risk, and perceived partner attitude about birth control. Multivariate regression models predicting contraceptive consistency outcomes controlled for participant age, race, emergency contraceptive use history, and Prime Time study condition. In multivariate analyses, consistency of hormonal method use at T2 was significantly predicted by consistent hormonal use at T1, participants' positive attitudes about birth control, perceived partner attitude supporting birth control use, and having the same sex partner at T1 and T2. Consistency of using any birth control method at T2 was significantly predicted by participants' desire to avoid pregnancy, perceived partner attitude supporting birth control use, and minimal age differences between partners. In multivariate analyses, none of the individual, partner or relationship variable examined in this study predicted consistency of condom use at T2. This study underscores the importance of considering individual, partner and relationship characteristics in predicting contraceptive consistency. To promote consistent contraceptive use among sexually active adolescent girls, providers must find ways to address partner attitudes and relationship characteristics, as well as the beliefs and practices of individual girls.