Development And Health Economic Evaluation Of Breast Cancer Screening Strategy Among Chinese Women

ObjectiveScreening methods, screening start-stop age,and screening interval were important factors to constitute the breast cancer screening strategy. Implementing different breast cancer screening strategies may have different screening benefits. Therefore, in the reasonable and effective use of the country’s limited resources for health conditions and in view of the characteristics of Chinese women breast cancer, to evaluate the sensitivity(Se), specificity(Sp) and its factors of clinical breast examination(CBE), breast ultrasound(BUS) and mammography(MAM), explore how to construct a variety of different breast cancer screening strategies and conduct cost-effectiveness analysis. To choose the best suitable breast cancer screening strategy for Chinese women and provide policy suggestions for improving the allocation of health resources in breast cancer prevention and control. Methods1. A total of 33234 women participated in the “Chinese Women Breast Cancer Screening Optimization multicenter Program” between 2008 and 2010. The Se and Sp and its 95% confidence interval(95%CI) of CBE, BUS and MAM by single use, in series or in parallel were calculated by χ2 test. The association between epidemiologicalfactors and sensitivity and specificity were also measured by χ2 test. We used 35 and 40 as the screening start age; 59, 64, and 69 as the screening end age; three methods of single or combination as the screening methods, and once every two years or once every three years as the screening interval to build the Chinese women breast cancer screening strategies.2. The main purpose of screening is early detection. The screening effect directly shows that screening diagnosis of breast cancer pathological staging has certain moved forward relative to the clinical diagnosis of breast cancer. 182 breast cancer cases were diagnosed by “The Central Fiscal Transfer Payment the Local Chinese Women Breast Cancer Screening Program” as the screening group. 435 breast cancer patients with clear pathological information were randomly selected as the non-screening control group from Breast Cancer Retrospective Study in seven areas in 2008. Meanwhile, according to the breast cancer incidence regularity and the progress regularity under the screening intervention, we used the Treeage Pro 2011 software to build natural history of breast cancer disease progression model and breast cancer screening and diagnosis Markov model. The model cycle was 1 year. The queue simulation method was used to estimate screening effects and costs of 100000 Chinese women in the next 50 years with different screening strategies, such as Life years saved and quality adjusted life years saved. According to different screening ages, we calculated the incremental cost-effectiveness ratio(ICER) between 22 kinds of screening strategies and compared the incremental cost-effective ratio between 22 kind of screening strategies and non-screening strategy.3. We evaluated the factors which may affect the breast cancer screening health economics, including multiple factors of the tornado diagram sensitivity analysis, breast cancer incidence rate, all-cause mortality rate, and discounting rate for one-way sensitivity analysis. We also estimated breast cancer preclinical sojourn time for Chinese women based on these screening data. Results1. Compared with the Se(86.0%, 95%CI: 79.2%-92.8%) and Sp 94.3%(95%CI: 94.0%-94.5%) of single mammography, supplement CBE or BUS could improve the Se and Sp of screening methods. The Se and Sp of CBE and MAM in parallel and BUS and MAM in parallel were 92.2%(95%CI: 86.9%-97.4%) and 93.7%(95%CI: 93.4%-94.0%), 95.0%(95%CI: 90.7%-99.3%) and 93.2%(95%CI: 93.0%-93.5%), respectively. According to different start-stop age, screening methods, and screening interval, we build 132 kinds of breast cancer screening strategy.2. There were 4 kinds of strategies that were identified as the optional undominated strategies, namely, triennially BUS and MAM in parallel screening between 35-59 years, biennially BUS and MAM in series screening between 35-59 years, biennially CBE, BUS and MAM in parallel screening between 35-59 years, and biennially CBE, BUS and MAM in parallel screening between 35-69 years. Compared with no screening, the ICER was 36774 RMB(90706 RMB) per quality adjusted life year saved(QALY). The cost-effectiveness of different breast cancer screening strategies in different age groups showed that biennial screening was better than triennially screening, screening in parallel was better than that in series. All the undominated screening strategies included BUS which hinted that BUS had a certain guiding value in the current level of breast cancer incidence and economy in China.3. Tornado diagram sensitivity analysis found that the breast cancer incidence annual percentage of changing had the greatest effect in the screening costs, followed by the screening start age and end age and the breast cancer diagnosis expenses. And the results of one-way sensitivity analysis based on different populations’ breast cancer incidence rate, all-cause mortality rate and discounting rate were changing very small, which showed that the model was stable. According to the data from multicenter breast cancer screening program, it was estimated that Chinese women breast cancer mean preclinical sojourn time was 10.46 years(95%CI: 8.10-13.51). The mean preclinical sojourn time of women under of 50 years was 9.37 years(95%CI: 5.71-15.38) and 10.84 years(95%CI: 8.04-14.61) for women aged 50 years and over. And the study has not found any influential covariate in preclinical sojourn time and progress probability. ConclusionSupplement the CBE or BUS on the basis of MAM screening alone can effectively improve the accuracy of screening methods. Three screening methods in parallel had the highest sensitivity, which can diagnose breast cancer in early stage.Women aged 35-69 years with once every 2 years of CBE, BUS and mammography in parallel screening strategy was the most cost-effectiveness. Compared with no screening, the ICER was 36774 RMB(90706 RMB) per QALY. This study provided scientific basis for Chinese women breast cancer screening. It may be a consideration to implement this breast cancer screening strategy in areas with high breast cancer incidence rate in China. The study also found that young women had shorter sojourn time and faster preclinical progress probability than older women,which also remind us that people in high risk should be screened as soon as possible in sojourn time. The better diagnosis breast cancer at early stage, the more obvious the screening effect will be.