Comorbid of chronic kidney disease (CKD) patients who undergoing dialysis in Indonesia using firth logistic regression

• Main Authors: Amini, Fatonah, Oktora, Siskarossa Ika
• Format: Conference Proceeding
• Language: English
• Subjects: Bias; Comorbidity; Diabetes; Diabetes mellitus; Dialysis; Health care facilities; Health centres; Hemodialysis; Hypertension; Kidney diseases; Personal health; Public health; Regression analysis; Signs and symptoms
• ISSN: 0094-243X; 1551-7616
• DOI: 10.1063/5.0041667

Cases of non-communicable diseases (NCD) in Indonesia tend to increase, one of which is Chronic Kidney Disease (CKD). CKD has become a global health problem because of the difficulty of treatment and expensive medical costs. CKD often refers to a a silent killer disease that develops without providing previous symptoms. One of the alternative treatments is dialysis. According to the 2018 Riskesdas, more than fifty percent of CKD patients who are undergoing dialysis is the productive age people. Ten percent to forty percent of these patients also have comorbid CKD like diabetes mellitus and hypertension. The study aims to analyze the determinant of CKD patients who undergo dialysis in Indonesia. Based on the 2018 Riskesdas, 17 percent of CKD patients undergo dialysis. Because of this condition, logistic regression modeling with an imbalance of the binary response (success and failure categories) could create a bias in the predicted probabilities. The stronger the imbalance of the response variable is, the more severe is the bias in the predicted probabilities; therefore, the firth logistic regression method is used to avoid bias. Based on the results, diabetes mellitus status, hypertension status, age, and the lack of fruits and vegetable consumption significantly affect the dialysis status of CKD patients. Thus, the community is always expected to adopt a healthy lifestyle and increase awareness of personal health by regularly checking their health at the health facility.