kidney stone

Aim: Kidney stone disease, which can affect people of all ages and whose incidence increases day by day, is becoming a public health problem due to treatment costs. This study aims to determine how factors related to kidney stones affect the diagnosis of the disease when taken together, rather than determining their relationship with the disease one by one.Materials and methods: An open-access dataset containing kidney stone status and associated factors was used in the study. Mann Whitney U test and independent sample t-test were used in data analysis. Logistic regression was performed with the backward variable selection method to determine the factors associated with kidney stones. ROC analysis was used to determine the power of the variables that were significant as a result of logistic regression analysis, individually and together, in discriminating kidney stones.Results: According to the results of logistic regression analysis, gravity, cond, and urea calc variables were found to be associated with kidney stones. With ROC analysis, it can be said that urea, calc, and gravity variables with AUC values above 0.60 can distinguish kidney stones. When the combinations of these variables are examined, the AUC values of the binary combinations are between 0.734 and 0.759, while the AUC value obtained for the triple combination is 0.831.Conclusion: According to the results obtained from the article, it can be said that while the factors associated with the disease and used in the diagnosis have little effect on the diagnosis of the disease alone based on the AUC values obtained from the ROC analysis, it can be said that considering them together increases the accuracy in diagnosis. Therefore, considering the factors thought to be associated with the disease together may be more appropriate in diagnosis and may give more accurate results.