Purpose: LARCs are the most effective forms of reversible contraception currently on the market, yet uptake is still quite low among adolescents. As weight concerns are frequently considered by adolescents choosing a contraceptive method, we sought to compare BMI changes within the first two years of use in girls using nonhormonal (copper) and levonorgestrel (LNG) intrauterine devices (IUDs) as well as etonogestrel (ENG) contraceptive implants and to examine the influence of baseline BMI prior to LARC insertion on BMI changes after insertion. Methods: We queried electronic records in our large, urban, health care system to identify girls aged 14-19 who had LARC inserted from 2010-2016. We excluded those with conditions potentially affecting weight, e.g. childbirth, bariatric surgery, systemic illness, etc. We reviewed medical charts and included 216 girls who had complete BMI data; 35% using copper IUD; 32% LNG IUD, 33% ENG implant. We compared baseline BMI (up to 12 months before LARC insertion) to first recorded BMI taken 6 to 24 months after insertion. We used ANOVA with post-hoc SNK tests to compare BMI changes based on type of LARC used. We used Chi-square to compare girls with baseline BMI<85%ile vs. 85-94%ile vs. ≥95%ile by LARC type and examined differences in BMI change by LARC type and baseline BMI category using two-way ANOVA. Results: Of the 216 girls, mean age at LARC insertion was 17.2± 0.2 years; 58% Hispanic; 28% black; 25% obese (BMI≥95%ile); 24% overweight (BMI85-94%ile). Over 6 to 24 months follow up, mean BMI increase for all 216 girls was +0.771 kg/m2, with no difference among the three LARC types. When examining the 146 girls with a follow up BMI recorded within 6 to 12 months after LARC insertion, we found a smaller change in mean BMI for users of copper IUDs (+0.019 kg/m2) than for users of LNG IUDs (+0.607 kg/m2) or ENG implants (+0.840 kg/m2), p <0.01. In the two-way ANOVA of BMI change by baseline BMI category and LARC type, we found a difference by BMI category (p<0.01), with mean BMI increasing more in obese (+1.39 kg/m2) than overweight (+0.74 kg/m2) or normal weight (+0.49 kg/m2) girls. We also found an interaction of LARC type X BMI category (p<0.01) specifically in obese girls, with mean BMI increasing more in hormonal than non-hormonal users LNG IUD (+1.95 kg/m2); ENG implant (+2.07 kg/m2); copper IUD (+0.58 kg/m2) Conclusions: We found a significantly larger BMI increase over the first year of use for adolescents using hormonal LARCs as compared with non-hormonal cooper IUDs. In addition, girls who were obese at initiation of LARC had a significantly larger BMI increase if they used a hormonal LARC compared to those using a non-hormonal copper IUD.