Estimating admission lifetime and survival for gynaecological cancers at the University College Hospital, Ibadan using Cox regression model.

Abstract

Objective: To estimate the admission lifetime of gynaecological cancer patients in the University College Hospital (UCH), Ibadan and its implication to management and overall outcome using the Cox regression model. Methods: Descriptive and cox regression model in survival analysis were used to analyze data from 823 patients with gynecological malignancies who were treated at the UCH, Ibadan between January 1 1995 and December 31, 2014. The outcome variable for this study was the admission life-time (in days). The variables collected were limited to the age of patients, types of cancer and patients’ status. The study employed some model criteria such as p-value, log rank test, Gehan-Wilcoxon test, Concordance index, R-square, likelihood ratio test, Wald test and score test to check for the efficiency of the results. Results: Of the 823 cases reviewed, 611(74.2%) were right-censored. Cervical cancer had the highest number of patients admitted with 53.5% and was commonest among patients aged 60 years and above (30.1%) while mortality was highest among patients with ovarian cancer. Ovarian cancer and age above 65 years were the only two factors that significantly affected patient’s survival experiences during their admission at the hospital. Conclusion: Patients younger than 60 years and with other gynaecologic cancers, except ovarian, had better chances of survival over a period of 6 months as at the time of admission into the UCH, Ibadan.