A Study of the Efficacy of Screening for Breast Cancer Based on the Health Insurance Plan Data

Abstract

In this thes is we would like to assess the benefit of screening in breast cancer detection. For this purpose, the Health Insurance Plan data will be analyzed. If screening is beneficial, we expect to see higher early stage detection in the study group compare to the control group which hopefully will lead to a better survival probability. Our analysis shows that there is a significantly higher combined stage I detection proportion in the study group compared to the control group (50.00% in the control group as opposed to 67.77% in the study group). Furthermore, the study group has a significantly lower proportion of breast cancer deaths than the control group. The Kaplan-Meier estimate shows that the study group has a better survival function than the control group. In addition to these, we shows that there exists no significant lead time. Lastly, we construct a Markov chain model for future analysis.