A Pooled Analysis And Comparison Of Breast Cancer Screening Programs And Screening Schemes In Shandong

Objective Data of four population-based breast cancer screening programs supported by the central budget of our country in Shandong Province was collected and a pooled analysis was done. Our studies were divided into two parts, In part 1, we aim to analyze the results of breast cancer screening programs in 2008-2011,and to evaluate the accuracy of clinical breast examination(CBE), breast ultrasonography(BUS) and mammography(MAM) for breast cancer in urban and rural areas of Shandong province. The purpose of part 2 was to compare the screening cost of two schemes and the treatment cost in patients with different clinical stages. The results would help to explore a compatible screening strategy for breast cancer in Shandong Province. And the investigations would be recommended to the government to popularize the programs of early detection and treatment of breast cancer.MethodsA pooled analysis of four population-based breast cancer screening programs in Shandong Province was done in the first part. The detection rate and proportion of early stage cancer was computed to assess the screening quality. Also, we valuate the accuracy of CBE, BUS and MAM independently and concurrently for breast cancer by computing the value of sensitivity, specificity, positive predictive value (PPV) and other indexes.In second part, the screening cost of two schemes used in urban and rural areas were summarized and compared. And treatment cost of a random sample of medical records of 360 breast cancer patients who had been treated in Shandong Tumor Hospital at the same time with different clinical stage were also explored. The results of screening project were described and discussed and the further correlation analysis was used to identify the features in costs of treating breast cancer with different clinical stage.Data cleaning and statistical analysis were performed using spss 18.0. The characteristics of the populations were described and the rates were compared by using x2 tests and Fisher’s exact probabilities. The screening expenses were described and Kruskal-Wallis test was conducted to compare average cost of direct hospitalization expenses for first treatment, average first outpatient expenses in different clinical stages. All analysis were using P< 0.05 as having significant difference.ResultsPart 1:A total of 64864 women and 67 breast cancer cases were identified in 4 programs with detection rate of 103.3/10 0000, and the proportion of early stage cancer was 41.8%. Those two rates in urban program were much more higher than in rural program. Based on the results of screening from the pooled analysis, the values of sensitivity of CBE, BUS and MAM alone were 52.2%,77.6% and 77.8%, respectively. And the values of specificity were 80.2%,75.2% and 75.5%; and those of PPV were 67.9%,70.7% and 63.6% separately. Sensitivity of combined tests in parallel with the values between 83.6% and 98.1%, is significantly higher than those in serial whose values were 37.0%-55.6%; and specificity and PPV of combined tests in serial is higher than those in parallel.Part 2:For each positive patients that were diagnosed by the program, the average cost was 145000 yuan in urban program used 3 tests indepently and concurrently, and the cost in urban and rural program with the scheme that CBE and BUS firstly and MAM when indicated were 131000 and 174000 yuan separately. The average screening cost of breast cancer screening program in Shandong were still high.Conclusions1. The screening programs in our province was of high quality relatively with high detection rates and proportions of early stage cancer. And the quality in urban area was better than that in rural areas, which pointed out that on the one hand, the incidence rate was higher in urban than in rural areas; and on the other hand, the improvement of screening quality and ability of detection and treatment of breast cancer in rural areas is in urgent need.2. The value of sensitivity of CBE alone was lowest in all three methods, and there were no significant difference in BUS and MAM. The sensitivity of combined tests in parallel were higher than those in serial, and so was with the value of false positive rate.3. The scheme of CBE, BUS and MAM used together is suitable in more richer urban areas with enough resources, and the scheme that CBE and BUS at the first stage, followed by MAM when indicated, can be promoted in average urban areas. But the cost is still too high to develop the breast cancer screening in some low incidence regions with limited health resources such as rural areas in our province. There remains a need to improve public awareness about breast cancer prevention and treatment and to continue improving the ability of earlier-stage detection in rural areas. Lower costs were spent in early staging patients compared with the advanced patients, which furtherly suggested the importance of early detection and treatment of breast cancer.4. To improve the technical level and screening quality is still the main problem we are facing. Also we are lacking of long-term and follow-up data of breast cancer screening in our country, and the research about health economics of breast cancer screening program is in great need.