Risk factors for fibroadenoma were investigated in a cohort of 265,402 female textile workers in Shanghai, China, who were interviewed at enrollment into a randomized trial of breast self-examination between October 1989 and October 1991 and followed until July 2000. Based on 1,507 women who developed fibroadenoma, relative risks and trends in risk were estimated using Cox regression. The risk of fibroadenoma was highest in women under 35 years of age, and it decreased with age after 35 and dropped markedly at the time of menopause. The risk decreased with increasing number of livebirths and with duration of oral contraceptive use. The risk increased with number of prior benign breast lesions and with decreasing age at the first benign lesion. The risk of a diagnosis of fibroadenoma was elevated in women who received instruction in breast self-examination. The development and persistence of fibroadenomas are dependent on the presence of ovarian hormones, and full-term pregnancies and exposure to exogenous estrogen-progesterone combinations before menopause may reduce risk by enhancing differentiation or reducing estrogen-induced proliferation in the mammary epithelium. The practice of breast self-examination detects some fibroadenomas that would otherwise not be detected.