Accelerated failure time modelling in the analysis of type 2 diabetic patient data

Aims: Type 2 Diabetes is a metabolic disorder and one of the most common non-communicable diseases. The aim of this study is to identify the significant prognostic factors affecting the survival of Type 2 diabetic patients using various Accelerated Failure Time (AFT) Models. Method: Multiple Accelerated Failure Time (AFT) models including Log-logistic, Lognormal, Weibull, and Exponential were used to analyze a dataset of Type 2 Diabetic patients. The appropriate model is selected on the basis of the minimum Akaike Information Criterion (AIC) and Bayesian information criterion (BIC) values. Out of 1,256 patients, 756 patients were enrolled in this study, with a follow-up period spanning 13 years (from 2005 to 2018). Results: The Weibull AFT model is found to be the best-fitted AFT model on the basis of AIC and BIC values, and further modeling and analysis have been done accordingly. Based on modeling, the time ratio (TR) for a family history of diabetes, BMI \(\ge\) $\ge$ 30, alcohol consumption, smoking addiction, Hypertension and vegetarian diet are found to be significant prognostic factors for this study. Conclusions: Significant predictors were identified using the Weibull AFT model. Prognostic factors such as family history of diabetes, alcohol consumption and smoking addiction, and vegetarianism are found to significantly affect the survival of Type 2 diabetic patients (P-value < 0.05).