Joint modeling of Longitudinal Change of Tumor Cell and Time to Death of Breast Cancer Patients: A Case Study at Gondar University Hospital

Eyerus, Asmare

dc.contributor.author	Eyerus, Asmare
dc.date.accessioned	2017-10-23T10:31:40Z
dc.date.available	2017-10-23T10:31:40Z
dc.date.issued	2017-10-23
dc.identifier.uri	http://hdl.handle.net/123456789/8100
dc.description.abstract	Background: Breast cancer is the major public health problem throughout the world and it is the most common cancer in women and in both developed and developing regions. This study aimed at determining factors that affect longitudinal change of tumor cell and time to death of breast cancer patients. Methods: Hospital based retrospective studies were conducted among breast cancer patients. Both the longitudinal and survival data was extracted from the patient’s chart from the data record from January 2013 to February 2017. The joint linear mixed model and the Cox regression model were used to determine factors that affect the longitudinal change of tumor cell and mortality of breast cancer patients. We analyzed both the separate and joint model of longitudinal and survival model to determine factors that affect the longitudinal change of tumor cell and survival time of breast cancer patients. Results: We compared the separate and joint model by considering their estimates and corresponding significant values and then we found that joint model having the most significant and precise estimates. From the survival sub-model result, patients with clinical stage II as compared to clinical stage I (HR=11.827; CI= 3.018- 46.349), patients with stage III as compared to clinical stage I (HR=18.619; CI=4.872-71.129), patients who live in rural (HR=2.750,CI=1.271-5.929), patients whose educational status are secondary (HR=8.516,CI=3.611-20.008) have significantly increased risk of mortality. From the longitudinal sub-model, clinical stage II and III, alcohol use and rural residence increases the progression change of tumor size and also the interaction effect of visit time by age, education by visit time and stage by visit time contributes to reduce the progression change of the tumor cell. The results also show that the patient’s survival time is associated with patient-specific tumor cell fluctuations such that a patient with higher tumor trend is less likely to survive Conclusion: Having advanced clinical status, secondary and above educational status and rural residence were found to be the predicting factors for mortality of breast cancer patients. Baseline tumor cell, clinical stage, alcohol, the interaction effect of visit by age, visit by education and visit by stage of tumor cell were found to be the predicting factors for longitudinal change of tumour cell of breast cancer patients.	en_US
dc.language.iso	en_US	en_US
dc.subject	Statistics	en_US
dc.title	Joint modeling of Longitudinal Change of Tumor Cell and Time to Death of Breast Cancer Patients: A Case Study at Gondar University Hospital	en_US
dc.type	Thesis	en_US