Despite decades of use, the long-term safety of breast implants in women remains a concern. While the incidence of breast cancer among women has increased dramatically in the past decade, the implant-related risk of carcinoma of the breast only recently has received widespread attention. An additional concern is that the presence of the implant may delay tumor detection. This study allows examination of breast cancer risk and detection issues among patients with long-term exposure. We conducted a record linkage cohort study of cosmetic breast implant patients. We abstracted the records of the private practices of 35 broad-certified plastic surgeons in Los Angeles County, California. We included 3182 white women who received cosmetic breast implants between 1953 and 1980. Spanish-surnamed women, nonresidents of Los Angeles County, and patients with prior subcutaneous mastectomy or breast cancer were excluded. Cancer outcomes through 1991 have been ascertained through record linkage with the Los Angeles County Cancer Surveillance Program. With a median follow-up of 14.4 years, 31 breast cancer cases were observed, compared with 49.2 expected, based on Los Angeles County population-based incidence rates (standardized incidence ratio = 63.0 percent; 95 percent confidence limits: 42.8 and 89.5 percent). The distribution of stage of disease at diagnosis among women with implants did not differ from that of all similar breast cancer patients in Los Angeles County. In Los Angeles County, augmentation mammaplasty patients experience a significantly lower than expected risk of breast cancer and no delay in breast cancer detection after an average of 14.4 years of exposure. While the linkage methodology allows the possibility of failing to detect diagnosed cancer cases and does not permit collection of some pertinent risk factors, the six other published epidemiologic studies on the topic also report breast cancer risk to be at or below the expected rate.