Although the majority of ecological and experimental studies have suggested a potential role of phytoestrogens in breast cancer prevention, findings from epidemiological studies have been inconsistent. Part of the inconsistencies may be attributable to the difficulty in measuring intake levels of phytoestrogens. Overnight urine samples from 250 incident breast cancer cases and their individually matched controls were analyzed for urinary excretion rates of isoflavonoids, mammalian lignans, and citrus flavonoids. The study subjects were a subset of the participants in the Shanghai Breast Cancer Study, a large population-based case-control study conducted in Shanghai from 1996-1998. To minimize potential influence of treatment on the exposure of interest, urine samples from breast cancer cases were collected before cancer therapy. Urinary excretion of total isoflavonoids and mammalian lignans was substantially lower in breast cancer cases than in controls. The median excretion rate of total isoflavonoids was 13.97 nmol/mg creatinine in cases and 23.09 in controls (P = 0.01), and the median excretion rate of total lignans was 1.77 in cases and 4.16 in controls (P < 0.01). The risk of breast cancer was reduced with increasing excretion of total isoflavonoids (P for trend, 0.04) and total lignans (P for trend, <0.01), with adjusted odds ratios of 0.62 (95% confidence interval, 0.39-0.99) and 0.40 (95% confidence interval, 0.24-0.64) observed for the highest versus the lowest tertile of total isoflavonoid and lignan excretion, respectively. The adjusted odds ratio was 0.28 (95% confidence interval, 0.15-0.50) for women who had a high excretion rate of both total lignans and isoflavonoids compared with those with a low excretion of both groups of phytoestrogens. No association was observed with citrus flavonoids. The results from this study suggest that high intake of certain phytoestrogens may reduce the risk of breast cancer.