TITLE:
Survival Rate Analysis on Breast Cancer Cases at Univesity College Hospital, Ibadan, Nigeria
AUTHORS:
Olatayo Olusegun Alabi, Aminat Yetunde Saula, Ezra Gayawan, Victor Samuel Alabi, Hamidu Abimbola Bello, Rasaq Yinka Akinbo, Taiwo Abideen Lasisi
KEYWORDS:
Survival Rate, Breast Cancer, Kaplan-Meier Estimator, Log-Rank Test and Multivariate Cox Regression
JOURNAL NAME:
Open Journal of Statistics,
Vol.12 No.2,
April
21,
2022
ABSTRACT: Breast cancer is one of the leading diseases that affect women’s lives.
It affects their lives in so many ways by denying them the required standard of
health needed to carry out all of their daily activities for some days, weeks,
months or years before eventually causing death. This research estimates the
survival rate of breast cancer patients and investigates the effects of stage of
tumor, gender, age, ethnic group, occupation, marital status and type of cancer
upon the survival of patients. Data used for the study were extracted from the
case file of patients in the Radiation Oncology Department, University College
Hospital, Ibadan using a well-structured pro forma in which 74 observations
were censored and 30 events occurred. The Kaplan-Meier estimator was used to
estimate the overall survival probability of breast cancer patients following
their recruitment into the study and determine the mean and median survival
times of breast cancer patients following their time of recruitment into the
study. Since there are different groups with respect to the stages of tumor at
the time of diagnosis, the log-rank test was used to compare the survival curve
of the stages of tumor with considering p-values below 0.05 as statistically
significant. Multivariate Cox regression was used to investigate the effects of
some variables on the survival of patients. The overall cumulative survival
probability obtained is 0.175 (17.5%). The overall estimated mean time until
death is 28.751 weeks while the median time between admission and death is 23
weeks. As the p-value (0.000032) of the log-rank test for comparing stages of
tumor is less than 0.05, it is concluded that there is significant evidence of
a difference in survival times for the stages of tumor. The survival function
plot for the stages of tumor shows that patients with stage III tumor are less
likely to survive. From the estimated mean time until death for the stages of
tumor, it was deduced that stage I tumor patients have an increased chance of
survival. Types of cancer, gender, marital status, ethnic group, occupation and
patient’s age at entry into the study are not important predictors of chances
of survival.