ABSTRACT

Background: Stroke is a major public health and clinical problem that represents the third leading cause of disease worldwide among adult non-communicable diseases Stroke in the developing world is becoming a leading cause of death from infectious diseases that affects mainly adults and adult disability. Ischemic stroke is the most common form of strokes approximately about 80%–85% of all stroke cases.

Methods: A retrospective cohort study was conducted at the Felege Hiwot Referral Hospital among 232 ischemic stroke patients who started treatment between September 2014 and August 2016. Information on relevant variables was collected from adult ischemic stroke patients paper based medical cards and registries. The Kaplan-Meier survival curves and log-rank test used to compare the survival time of different category of predictors, and Cox’s regression model employed to identify the effect of covariates on the survival time of ischemic stroke patients.

Result: A total of 232 adult ischemic stroke, patients were included in the study. Out of the total 232 individuals, 69 (29.74%) died and the remaining 163 (70.26%) were censored at the end of follow-up period.  Based on Cox proportional hazards regression model, age (AHR =1.015, CI: 1.001 – 1.0304), diabetes (AHR=1.353, CI: 1.1938– 1.643), hypertension (AHR =1.607, 95% CI: 1.003–1.0053), structural heart disease (AHR =2.231, 95% CI: 1.158 – 4.593), and history of previous ischemic stroke (AHR = 1.84, 95% CI: 1.088 – 3.113) were significantly associated with the mortality of adult ischemic stroke patients.

Conclusion: Based on the Cox Proportional Hazards regression model the following prognostic factors, age, diabetes, hypertension, structural heart disease, fibrillation, and history of previous ischemic stroke were the risk factors for survival time of ischemic stroke patients

Key words: Cox Proportional Hazard Models, Ischemic Stroke, and Survival Function