Binary Logistic Regression Analysis in Determining Risk Factors of Acute Respiratory Infection (IRA) on Under-Five Children

Abstract

Acute Respiratory Infection (ARI) is a prevalent ailment and stands as one of the primary causes of mortality among children under the age of five. As reported by the Ministry of Health of the Republic of Indonesia, ARI predominantly affects children aged one to four years, with an incidence rate of 13.7%. Binary logistic regression is utilized to predict the likelihood of ARI occurrences in children under five, employing a linear combination of log-odds pertaining to suspected contributory factors. This study aims to evaluate the risk factors associated with ARI and develop an optimal logistic regression model by analyzing data from 166 participants who visited the Tarus regional health center between July 21 and September 1, 2023. Factors considered in this research as predictors for ARI (dependent variable ) include Immunization Status (X1), exposure to cigarette smoke at residence (X2), exposure to wood smoke at residence (X3), exclusive breastfeeding (X4), and nutritional status (X5). The final analysis revealed that incomplete immunization, exposure to cigarette smoke, and exposure to wood smoke at residence significantly heighten the risk of ARI. The most fitting logistic model obtained was expressed as logit (π_ijk)=-1.1051+1.2297X1+0.8709X2+1.31085X3.