A Predictive Model for Intestinal Protozoal Infections Based on Clinical Characteristics and Social Determinants of Health of Cancer Patients Undergoing Chemotherapy

Intestinal protozoan infection is a major public health concern in Indonesia, particularly among immunocompromised individuals like chemotherapy-treated cancer patients. This study aimed to develop a predictive model for infection risk based on social determinants and clinical characteristics. A cross-sectional study was conducted at Dr. H. Abdul Moeloek General Hospital, Lampung Province, Indonesia, in 2022, involving questionnaires and stool sample analysis. Microscopic examination was performed using sedimentation, iodine staining, and modified Ziehl-Neelsen tests. Statistical analysis showed significant correlations between infection and marital status, as well as erythrocyte levels (p < 0.05). Logistic regression identified key predictors, including age, address, temperature, cancer duration, erythrocytes, hemoglobin, hematocrit, and monocytes. The model explained 47.4% of infection variability (Nagelkerke R2 = 0.474) and demonstrated strong predictive performance (AUC = 85.0%, Hosmer-Lemeshow p = 0.570). Infection probability ranged from 0.35% (low-risk) to 99.77% (high-risk). The dominant risk factors were address, cancer duration, erythrocytes, and hemoglobin, while age, temperature, hematocrit, and monocytes acted as confounders. By targeting these risk factors, healthcare providers can develop effective interventions to reduce the burden of intestinal protozoan infection in this vulnerable population.