ObjectiveGiven high unmet need for contraception in Ethiopia, this study aimed to determine prevalence and factors associated with early discontinuation of long-acting reversible contraceptives (LARC).MethodsThis institution-based cross-sectional study was conducted with 389 participants using exit interviews with clients presenting for removal of LARC at the family planning clinic of four government hospitals in Addis Ababa.SPSS version 26 was used for analysis. Descriptive statistics, bivariate, and multivariate logistic regression were computed.ResultAmong the 389 clients, 236 (60.7%) discontinued early. In multivariate regression, lack of pre-insertion counseling on side effects (AOR = 3.5, p = 0. 000; 95% C.I = 1.8—6.8), presence of side effects (AOR = 1.9, p = 0. 017; 95% C.I = 1. 1- 3.4), history of abortion (AOR = 3.5, p = 0. 001; 95% C.I = 1. 6- 7.4); and no prior contraception use (AOR = 2.9, p = 0. 000; 95% C.I = 1. 6- 5.3) were positively associated with early discontinuation. Whereas insertion outside of Saint Paul’s Hospital Millennium Medical College (AOR = 0.4, p = 0. 000; 95% C.I = 0. 2- 0.6), and influence on choice of contraceptives by others (AOR = 0.2, p = 0. 000; 95% C.I = 0. 2- 0.4) were negatively associated with early discontinuation.ConclusionEarly discontinuation of LARC was high among study participants. Counseling about possible side effects and giving women the opportunity to decide their own choice of contraception might help in reducing early discontinuation.