Chronic Kidney Disease (CKD) is a major global health issue that can lead to serious complications and long-term medical care. This study aims to identify key clinical factors associated with CKD status using binary logistic regression analysis. The dataset, obtained from Kaggle, contains 400 patient records with various clinical and demographic attributes. The dependent variable is CKD status (positive or negative), while the independent variables include age, blood pressure, hemoglobin level, urine albumin level, and serum creatinine. Initial analysis involved descriptive statistics and multicollinearity checks, followed by model estimation and evaluation using likelihood ratio and Wald tests. The final model identified four significant predictors: blood pressure, hemoglobin, urine albumin, and serum creatinine. The model achieved a high classification accuracy of 95.50% and an Area Under the ROC Curve (AUC) of 98.78%, indicating excellent predictive performance. These results highlight the importance of these clinical indicators in early CKD detection and support their use in risk assessment models for kidney disease screening
Keywords: Chronic Kidney Disease, Binary Logistic Regression, Likelihood Ratio Test, Wald Test, Classification Accuracy