Abstract:
Background: Breast cancer is a type of malignant tumor that starts in the breast
cells and commonly occurs in women than men. Breast cancer mostly occurs in the left
breast rather than the right breast and studies show that there is a significant associ ation between time to left and right breast cancer laterally. As far as our knowledge
is concerned, researchers didn’t see the dependency between left and right last stage
breast cancer patients.
Objectives: The main objective of this study was to compare different marginal mod els with copula families for time to last stage of left and right side breast cancer patients
and explore whether there was a statistical dependency between time to left and right
last stage breast cancer. Moreover, to identify the risk factors of last stage breast
cancer of patients.
Methods: A retrospective study design was used to collect relevant data on time to
events of last stage breast cancer patients. A total of 516 patients were included in
this study. In this thesis, mainly nine different copula-based bivariate survival models
were employed. Two-step estimation method was used to estimate the marginal dis tributions and copula parameter. The dependence between the time to right and left
last stage breast cancer of the patient was quantified using the copula parameter, and
the effect of covariates were modeled using the parametric marginal survival model.
Results: Of 516 patients, 59 (11.4%), 55 (10.7%), 146 (28.3%) patients only right,
only left breasts, and both breasts were at the last stage, respectively. The Weibull
marginal distribution with the Clayton copula model fit the last stage breast cancer
dataset. The dependence between time to last stage of right and left breast cancer of
female patients is found to be 0.189.
Conclusion: Being Hypertensive, older patient, positive status of anemia, who resides
in rural and chemotherapy treatment of breast cancer disease were the risk factors that
leads to last stage breast cancer. In conclusion, there is a statistical dependency be tween the time to right and left last stage breast cancer of the patients.
Keywords: Bivariate events, abnormal cell, Weibull-Clayton, Kendall’s
dependenc