From the perspective of endometrial safety, there has been great controversy about what special management, if any, tamoxifen-treated patients should undergo. Periodic blind endometrial sampling or transvaginal ultrasound has been advocated by some. Because of the problems associated with either of these techniques alone, we recommended an approach that used transvaginal ultrasound and then proceeded to sonohysterography when the endometrial echo on transvaginal ultrasound was not reliably thin and distinct. The American College of Obstetricians and Gynecologists (ACOG), in its committee opinion, stated that patients receiving tamoxifen therapy should only have an annual pelvic exam with pap smear if they remain asymptomatic. Newer data suggest, however, that there are high- and low-risk groups that can be identified by pretreatment screening. Before tamoxifen therapy, 17% of patients have polyps. These patients have 17 times the incidence of atypical hyperplasia than those whose uterus was negative before tamoxifen therapy. Such findings call into question the validity of the only study of raloxifene where uterine safety was the primary endpoint. In that study, any woman with baseline endometrial findings other than pristinely negative (i.e., low risk) was excluded. However, other raloxifene studies without pretreatment screening show relative risk (RR) = 0.8 (95% CI = 0.2, 2.7) for endometrial carcinoma. This compares with the women over 50 years of age in the Breast Cancer Prevention Trial (National Surgical Adjuvant Breast and Bowel Project P-1) with tamoxifen when the RR = 4.01 (95% CI= 1.70, 10.90). The existence of potentially high- and low-risk groups should be taken into account in any future clinical trials looking at the endometrial safety of selective estrogen receptor modulators (SERMs).