A Study on Significant Predictors for Prediction of Undiagnosed T2DM Using Binary Logistic Regression Model
Type 2 Diabetes Mellitus (T2DM) is a chronic disease that can cause premature deaths worldwide. Malaysia is one of the many countries that facing this serious epidemic. The World Health Organization (WHO) has also estimated that Malaysia would have 2.8 million people having T2DM disease in 2030. This study aims to identify significant predictors for prediction of undiagnosed T2DM patients in one of the highest prevalence states of T2DM. Binary logistic regression model proposed to predict the presence of T2DM among undiagnosed respondents. The selection of significant predictors using univariate, multivariate and backward stepwise selection was implemented in this study. The study concludes that four predictors were found significant for prediction of undiagnosed T2DM patients.