There is an increasing need for accurate prediction methods of assessing individual risk for breast cancer for both clinical and research purposes. The purpose of this study is to compare the Gail and Claus model risk estimates of breast cancer among women with a family history of breast cancer. This study presents risk estimates from two models of breast cancer risk in 491 women 18 to 74 years of age with a family history of breast cancer who were recruited to risk counseling clinical trials in Seattle, Washington between 1996 and 1997. These trials included women from the general population and additional samples of Ashkenazi Jewish, African-American, and lesbian women. We estimated and compared lifetime (to age 79) and 5-year risk for developing breast cancer using the National Surgical Adjuvant Breast and Bowel Project adaptation of the Gail model and the Claus model. About one-quarter of participants fell into the Gail "high" risk category (> or =1.7% risk of developing breast cancer in the next 5 years). The average lifetime risk was estimated at 13.2% by the Gail model and 11.2% by the Claus model. Estimates from the two models were moderately and positively correlated (r = 0.55) with the Gail model yielding a higher estimate than the Claus model for most participants. If women with a family history of breast cancer are being counseled regarding decisions on genetic testing, tamoxifen use, or other preventive measures, presenting both Claus and Gail estimates may be the best option.