Separate and Joint analysis of longitudinal measurement of sugar level and time to event of diabetes mellitus patients: An instance of Debre Berehan Referal Hospital

Authors

  • Wudneh Ketema , Samrawit Eshetu

Abstract

Diabetes mellitus is a metabolic issue where by glucose can't successfully get moved out of the blood .It is an ongoing infection of sugar, fat and protein digestion brought about by either outright absence of insulin or insulin opposition. The point of this finding was to conduct statistical analysis on jointly model and separately analysis of longitudinally measured fasting glucose level and survival analysis of diabetic patients. A total number of 384 DM patients include in this study. Among those 34(8.85%) were died. Separate and joint modeling of longitudinal measurement of sugar level and time to Event was fitted. The statistical tools in analyzing separate and joint model are R and SAS software. The outcomes for discrete and joint models were unique and furthermore, the assessments of the affiliation boundaries in the joint investigation were critical, giving proof of relationship between the two sub models. The log level of glucose fast and time to death was positive association, which indicates a unit increase in log level of glucose fast the risk of death is increasing. When evaluating the generally execution of both the different and joint models as far as model parsimony, goodness of fit, and the statistical significance of both the association parameters the joint model performs better. Thus, joint model was preferred for simultaneous analyses of repeated measurement and survival data.

Published

2020-11-01

Issue

Section

Articles