OBJECTIVES: Accurately measuring health state utilities is crucial for health economic evaluations. Few empirical studies have compared the performance of different methods for measuring utilities among breast cancer patients. The aim of this study was to investigate the extent of agreement between direct and indirect methods for eliciting health state utilities among breast cancer patients. METHODS: Six hundred and twenty-one breast cancer patients were invited for a face-to-face interview in Mainland China. Three methods were considered in the survey, including the time trade-off (TTO), the EQ-5D-5L and the SF-6D. The intraclass correlation coefficient (ICC) and Bland-Altman plots were used to assess agreements among three different measurements. RESULTS: A valid sample of 608 patients (97.9%) completed the questionnaires. The mean age was 48.0±9.6 years, and almost one-third were illiterate or primary school graduates. The mean duration of breast cancer was 38.2±40.9months. The mean health state utilities elicited from the EQ-5D-5L, SF-6Dand TTO were 0.83,0.65 and 0.80 respectively. The overall ICC was 0.46 and Bland-Altman plots showed wide limits of agreement among the three measurements. CONCLUSIONS: The comparison of utilities derived from three instruments indicates a lack of agreement within breast cancer patients.