Application of Longitudinal Data the Multilevel Models Approach on Diabetes Mellitus

Abstract

Objective: Diabetes mellitus is a metabolic disorder that develops over time and affects the cardiovascular system, eyes, kidneys, nerves, and blood sugar levels. The aim of this investigation was to determine the prevalence of diabetic mellitus patients, identify the associating risk factors using a multilevel longitudinal model, and understand the multilevel model changes for the level-1 and level-2 models. Material and Methods: We examined such types of scenarios using multilevel longitudinal models such as the simple random intercept multilevel model, the random coefficient model, and the null model. Results: There were 248 individuals with diabetes mellitus enrolled in the study for follow-up measurements over 4 time points, among these 248 individuals, 211 had complete data for all four time points. Based on the intraclass correlation coefficient, much of the variability (88.35%) in diabetes mellitus patients was accounted for by the follow-up time in this study, whereas 11.65% of the variability could not be accounted for by the follow-up time. Moreover, the data analysis suggested that sex had a significant effect on diabetes mellitus patients with the progression of time. Conclusion: Based on the results of our study, sex, baseline fasting and educational status had a significant effect on diabetes mellitus patients over time. The educational status of diabetes mellitus patients was found to have a significant effect throughout the follow-up time; this shows that when treating diabetes mellitus patients, the physician should beware of the nature of the disease and how to management diabetes requires a high level of awareness and motivation on part of the patients regarding self-care.