OBJECTIVES: There has always been a controversy of over-diagnosis in breast cancer screening. Other than costs, screening technologies and related positive results are also important factors which influence anxiety, stress and health related quality of life (HRQoL). The study was aimed to explore the impact of breast cancer screening on HRQoL in China.

METHODS: As a part of the CanSPUC program supported by the central government of China, recruited general population (GP group) aged 40-69 years were evaluated using a structured questionnaire and an algorithm system; if any female evaluated as ‘high risk’ for breast cancer (HR-BC group), then ultrasound combined with mammogram screening were provided. Built on the above platform, a cross-sectional survey was conducted in 12 provinces from September 2013 to December 2014. HRQoL were evaluated using the three-level EuroQol-5-Dimension instrument and utility scores were generated through the Chinese value set.

RESULTS: For female GP group (n=6 595, mean age=54.2) and HR-BC group (n=2 378, mean age=51.6), the mean visual analogue scale (VAS) () were 79.2 (78.8-79.5) and 77.7 (77.1-78.2), respectively; pain/discomfort/anxiety/depression were the mostly reported of any problem (51.4% and 33.1%, P<0.001), the corresponding mean utility scores () were 0.952 (0.950-0.955) and 0.937 (0.933-0.941), respectively (P<0.001). Interview timing for HR-BC group were distinguished, utility scores at pre-screening, in-screening (women knowing either screening result) and post-screening (knowing both results) were 0.928 (0.921-0.935), 0.958 (0.948-0.969) and 0.938 (0.933-0.943), respectively (P<0.001); the corresponding proportions of pain/discomfort/anxiety/depression at three time points were 60.1%, 35.1%, and 49.9%, respectively (P<0.001).

CONCLUSIONS: The impact of breast cancer screening on utility seemed lower than the minimally important difference of EQ-5D (0.074), but statistical differences were observed. The current findings suggested that the short-term impact of screening intervention on HRQoL was not clear yet, and further cohort studies and other instrument measurements would be needed.