Fikire Enquoselassie

Modeling for the Survival of Patients with Diabetes Mellitus: the Case of Yekatit Xii Hospital, Ethiopia Fikire Enquoselassie

The result from Kaplan-Meier estimation revealed that the survival of a patients is significantly related with alcohol use, tobacco use, BMI, type of diabetic disease diagnosed, diabetic complications, BP, cholesterol level, pre-existing health condition, and family history of DM. The Logrank result also indicated that the survival probability of diabetic patient was not statistically different in experiencing the death event among groups classified by sex, place of residence and region. The stratified Cox proportional hazard model and the Weibull regression model result indicated that overweight, alcohol use, tobacco use, high BP, high cholesterol level, complication of DM, high level of FBS and positive family history of DM significantly contributed to shorter survival time of a patient. In conclusion, the findings of this study showed that initial weight, BMI, alcohol use, tobacco use, BP, type of diabetic disease diagnosed, cholesterol level, complication of DM, FBS and family history of DM were major factors related to survival time of diabetic patient. It is recommended that people should be aware on the burden of these risk factors and well informed about the disease.