OBJECTIVES: This study aimed to capture response shift and true change in health-related quality of life (HRQOL) scores of breast cancer patients after surgery. METHODS: A data set from a prospective study to identify predictive factors of HRQOL (Taira N, Shimozuma K, et al: Breast Cancer Res Treat, 2011) was analyzed, which included HRQOL scores of physical well-being (PWB) and emotional well-being (EWB) scales of the FACT-G in 191 female breast cancer patients during a two-year postoperative period (at baseline [1 month after surgery]), 6, 12, and 24 months postoperatively). Oort’s structural equation modeling approach was used to investigate three aspects of response shift: (a) a change in the respondent’s internal standards of measurement (i.e., recalibration); (b) a change in the respondent’s importance of values (i.e., reprioritization) and (c) a redefinition of the target construct (i.e., reconceptualization). RESULTS: All three aspects of response shift were observed. Recalibration and reprioritization were occurred in three items of PWB (‘nausea’, ’trouble with family’, ’side effects’). Reconceptualization was observed from PWB to EWB in two items (‘nausea’ and ‘pain’) and from EWB to PWB in two items (‘sadness’ and ‘nervousness’). True change, which was calculated after adjustment of response shift, was observed in PWB (the across occasion difference of common factor mean [alpha] = 0.238, P <0.001) during first 6 months, and in PWB (alpha = 0.605, P <0.001) and EWB (alpha = 0.234, P <0.05) during first 12 months, while observed data analyses indicated statistically significant change in PWB and EWB during first 12 months. CONCLUSIONS: Captured response shifts in this study may be affected by various pre- and post-operative events such as notification of cancer and/or received treatments. These results will help improve reliability of HRQOL measurements in a longitudinal study.