Modeling Time-to-recovery of adult diabetic patients: A Case Study of Jimma University Specialized Hospital

Abstract

Diabetes is a group of diseases marked by high or low level of glucose resulting from defects in insulin production, insulin action or both. There are two main types of diabetes, namely type I and Type II diabetes. The objective of this thesis is to model time-to-first recovery of adult diabetic patients using Cox PH and shared frailty models. A retrospective data was obtained from JUSH diabetic patient clinic. All diabetic patients  18 years of age and who are under treatments in between September 2010 and August 2013 are included in the study. Time of fasting blood sugar level to reach the first normal range, 70-130 mg/dl, of blood since time of treatment or intervention were the response variable. Due to the impact of residential places and unmeasured shared similarities in a cluster, district (Woreda) is used as a random effect (frailty) term in the survival models. In this thesis, Cox PH and shared gamma frailty models were used. The AIC was used to compare the performance of the different models. First, inseparable diabetic mellitus (DM) was analyzed to identify whether diabetic types significantly influencing recovery time of DM. Second, separate types of DM are analyzed to identify factors influencing recovery time of these types of DM. The median recovery time of type-I and type-II diabetic patients were between 2 and 4 months respectively. The minimum and maximum recovery time of type-I diabetic are 1 and 6 months, respectively, whereas for type-II diabetic mini-max recovery time is found to be 1 and 31 months, respectively. Types of diabetic, bodyweight at baseline, fasting blood sugar at baseline, sex and age of patients are significantly associated with time to first recovery of diabetic patients. These variables are important factors that should be considered during the selection phase a treatment (combination of treatments) for diabetes. Moreover, Cox PH with gamma frailty model have resulted in a minimum AIC as compared to Cox PH model without frailty term in the model. This might be due to the shared environmental and residential factors. Hence, Cox PH model with gamma frailty provide a suitable choice for modeling time to first recovery of DM as compared to Cox PH without frailty term in the model.