To assess contraceptive switching and discontinuation among participants enrolled in a contraceptive access project over 3 years and to identify variables associated with contraceptive change. The HER Salt Lake study enrolled individuals between 2015 and 2017 from four clinics in Salt Lake County into a prospective, longitudinal cohort. All participants were able to switch or discontinue at no cost (between March 2016 and March 2020). We collected eight follow-up surveys over 3 years after enrollment. Each survey wave included questions about method use in the previous 4 weeks. We categorized participants in three ways, allowing for time-varying outcomes by wave: 1) those who reported using the same method as previous wave (continuers), 2) those who reported using a different method from previous wave (switchers), and 3) those who reported using no contraceptive method at that wave (discontinuers). We report the frequency of outcomes and conducted multinomial regression models assessing predictors of switching and discontinuation. Among 4,289 participants included in this analysis, 2,179 (50.8%) reported at least one instance of switching or discontinuation, and 2,110 (49.1%) reported continuing with their baseline method at the end of the study. Those reporting method change (switching or discontinuing) reported an average of 1.93 change events over the study follow-up period (range 1-8). Among those reporting any method change, 522 participants (23.9%) reporting at least one instance of both switching and discontinuation. Among those reporting any instance of discontinuation (n=966), 498 (51.6%) never reported uptake of a subsequent method. Among those who did report a subsequent method (n=468), 210 (44.8%) reported restarting a previously used method, and 258 (55.1%) reported starting a new method. Although we identified overlap among variables associated with switching and discontinuation, other predictors were discordant between switching and discontinuation. New contraceptive users commonly switch and discontinue methods. User behavior is associated with certain demographic characteristics and pregnancy planning. ClinicalTrials.gov , NCT02734199.