To evaluate contraception use and change among young women with early breast cancer. Secondary analysis of a cluster randomized trial. Multi-institutional. Patients with newly diagnosed breast cancer age ≤45 years enrolled from 54 US oncology practices. Sites were randomly assigned to the Young Women's Intervention, an educational intervention for young women with newly diagnosed breast cancer and their oncologists addressing issues specific to this population, including contraception, or a contact-time control physical activity intervention. Participants completed surveys in follow-up, including a 3-month survey regarding contraceptive practices before and after diagnosis. Outcomes of interest included young women's contraceptive use and methods before breast cancer diagnosis and 3 months after study enrollment. Logistic regression models assessed factors associated with use of less than highly effective contraceptive methods categorized according to World Health Organization effectiveness tiers and changes in contraceptive methods. Of 312 women included, 258 (83%) reported contraceptive use before breast cancer diagnosis, and 275 (88%) reported contraceptive use after diagnosis. Use of highly effective methods (e.g., vasectomy, non-hormonal intrauterine devices) increased from 39% before diagnosis to 52% after diagnosis. Use of moderately effective methods (e.g., hormonal methods) decreased from 22% before diagnosis to 3% after diagnosis. Use of less effective methods (e.g., condoms, withdrawal) increased from 22% before diagnosis to 34% after diagnosis. On multivariable analysis, factors associated with using less than highly effective contraception after diagnosis included desire for additional children (odds ratio [OR], 6.33; 95% confidence interval [CI], 3.76-10.66) and discussing contraception with a provider (OR, 1.96; 95% CI, 1.12-3.40). After breast cancer diagnosis, 207 patients (66%) reported no change in contraceptive methods. On multivariable analysis, factors associated with contraceptive method change after diagnosis included age <35 years (OR, 2.96; 95% CI, 1.57-5.58) and provider discussion (OR, 3.59; 95% CI, 1.91-6.78). There was no association in either analysis with study arm. Although most patients used contraception after breast cancer diagnosis, nearly half reported using less than highly effective contraceptive methods with higher failure rates, highlighting the need for early and improved contraceptive counseling for young women with breast cancer. NCT01647607.