Population-based breast cancer screening – the Taiwan experience
The current population-based breast cancer screening in Taiwan has come a long way since the start of phase I in 1995 upon the establishment of the National Health Insurance.
In phase I, which ran from 1995 to 1998, 4,865 women aged 35 and above whose relatives had breast cancer were provided clinical breast examination (CBE), ultrasound and mammography at hospitals. The data projected relative risk reduction in mortality of about 33 percent for this group which is significant, said Associate Professor Amy Ming-Fang Yen, of Taipei Medical University, Taiwan.
Then, the 3-year phase II screening was started in 1999 to test the feasibility of mass screening in the community involving almost 900,000 women aged 35 and above. All received CBE while about 33,000 women also underwent ultrasound scan. The breast cancer detection rate was 1.05 permille or 947 cases.
Phase III was conducted from 2002 to 2004 and comprised two stages—questionnaire in the first stage and mammography in the second stage. The risk score in the questionnaire was derived from CBE data in phase II. [J Med Screen 2006;13(Suppl 1):S23–S27] The questionnaire was circulated to more than 200,000 women aged 50 to 69 to identify those at high risk for breast cancer and refer them for mammography. Eventually, almost 70,000 women were referred for mammography and 149 cases of breast cancer were detected at a detection rate of 2.16 permille.
Mammography screening was extended nationwide in phase IV from 2004 to 2009 where more than 700,000 women aged 50 to 69 were screened every 2 years. From 2010 onwards, the screening age was lowered to 45 and about 1.5 million women were screened until 2012. The detection rates for the periods of 2004-2009 and 2010-2012 were 4.43 permille and 5.41 permille, respectively.
Universal biennial mammography screening, when compared with CBE, was found to reduce the mortality by 41 percent and death due to stage II+ breast cancer by 30 percent. Overdiagnosis was about 13 percent higher with mammography screening compared with CBE. [JAMAOncol 2016;2(7):915–921]
Yen said the mammography screening in Taiwan met WHO’s health system criteria including service delivery with in-reach and out-reach service; teamwork between clinical experts and public health professionals; health information system which covers high-risk to universal screening; technologies supported by evidence-based medicine; financial support from the National Health Insurance and importantly, leadership and governance support from the Health Promotion Administration.
However, she noted there are some issues, which have to be resolved. Firstly, incidence was found to be highest among younger women but the efficacy of mammography in this group is not completely proven. Secondly, only 25 percent of women attend biennial mammography screening regularly. Thirdly, universal screening may not be cost effective.
Advanced biomarkers and new imaging techniques have led to personalized medicine where the population is stratified into different risk groups and undergo screening specific to their risk. [Br J Cancer 2013;108(11):2241–2249,J Epidemiol 2017;27(3):98–106] This type of personalized approach has been shown to reduce the incremental cost-effectiveness ratio (ICER), said Yen.