Using binary logistic regression to identify the most important factors affecting the incidence of diabetes during the year 2025

Abstract

The study aimed to identify the most important factors affecting the incidence of diabetes during the year 2025. Binary logistic regression was used to explain the relationship between the dependent variable and the independent variables, and the maximum likelihood estimation method was used to estimate the parameters of the logistic model through data representing a group of diabetic patients aged from one month to 19 years obtained from the records of the Diabetes Hospital in Al-Khums City for the year 2025 AD. The study concluded that the logistic discrimination model is appropriate in studying the factors affecting diabetes and it was found that the estimated parameters of the logistic model are significant. It was also found that the variables (weight and age) are influential in the dependent variable, which represents the condition of the person infected or not infected, and that the rest of the variables have no effect, and that the overall percentage of correct classification for the accuracy of prediction of the dependent variable based on the estimated model is (71.3).