Predictive model for depression risk in CKD patients using NHANES data (2005-2018)

Individuals with chronic kidney disease (CKD) have a higher likelihood of suffering from depression. The objective of this research was to create a nomogram model that can forecast the likelihood of depression in patients with CKD. Data from 3858 individuals with CKD in National Health and Nutrition Examination Survey from 2005 to 2018 was examined. The datasets were divided into training and validation sets, with 80% of the data allocated to training and 20% to validation. Least absolute shrinkage and selection operator regression was applied to select relevant predictive variables associated with depression. Subsequently, multivariable logistic regression analysis was performed to construct the predictive nomogram model. Model evaluation involved the C-index, Hosmer-Lemeshow test, calibration curve, and decision curve analysis. The nomogram model included 8 factors: age, gender, body mass index, smoking status, sleep duration, poverty-to-income ratio, cardiovascular disease, and diabetes mellitus. The model exhibited robust discriminative ability, with area under curve values of 0.735 for the training set and 0.781 for the validation set. The model demonstrated clinical relevance and accuracy in both datasets. In conclusion, our nomogram aids in assessing depression risk in CKD patients, providing valuable support for healthcare providers in optimizing treatment strategies.