243 - Factors Associated with Admission versus Discharge Among Febrile Pediatric Patients with Low Illness Severity in Tanzania

Publication Number: 1230.243

Erin Keizur, UCSF Benioff Children's Hospital Oakland, San Francisco, CA, United States; Erasto M. Sunzula, Muhimbili university of health and allied sciences (MUHAS), Dar es Salaam, Dar es Salaam, Tanzania; Cecilia L. Msemwa, Muhimbili university of health and allied sciences, Dar es salaam, Dar es Salaam, Tanzania; Scott P. Oltman, University of California, San Francisco, School of Medicine, San Diego, CA, United States; Anneka Hooft, University of California, San Francisco, School of Medicine, San Francisco, CA, United States; Teresa B.. Kortz, University of California, San Francisco, School of Medicine, San Francisco, CA, United States

Background: Severe febrile illness remains a leading cause of pediatric morbidity and mortality globally, with the highest burden in low-resource settings. Determining which child requires hospital admission is complex and influenced by many factors. Existing clinical guidelines help support assessment and triage of these children by frontline providers, but they are often limited by reliance on presenting signs and symptoms and lack comprehensive recommendations on dispositions.

Objective: To determine the associations between symptoms, vital signs, and clinical characteristics with disposition in pediatric patients with signs of sepsis and similar presenting illness severity in a tertiary care hospital in Tanzania.

Design/Methods: We conducted a prospective cohort study at Muhimbili National Hospital, Tanzania. Children presenting to the Pediatric Emergency Department (PED) between July 26, 2022- September 20, 2023 with vital signs meeting sepsis criteria were included. We analyzed a subset of these patients with lower severity of illness as measured by a Lambaréné Organ Dysfunction Score (LODS) 0-1 (Table 1). Patients were stratified by disposition (admitted vs discharged) with demographic, historical, and clinical variables compared between the two groups using chi-square tests and multivariable logistic regression. This study was approved by the institutional ethics review board.

Results: After controlling for potential confounders, referral from an outside health facility (OR 8.6, 95% CI 3.54-21.69) and malnutrition (OR 2.78, 95% CI 1.34- 5.77) were the strongest independent predictors of hospital admission (Table 2). Children with sepsis, meningitis, or no identified infection were almost exclusively admitted, while gastrointestinal or respiratory infections were more commonly discharged (Table 3).

Conclusion(s): Nearly all referred children were admitted, suggesting that referral from lower-level health facilities heavily influences disposition at tertiary care centers. Strengthening context-specific triage guidelines, emergency capacity, and referral criteria at lower-level health facilities may help identify patients safe for discharge. Additionally, allocation of resources for community-based nutritional support programs may reduce the need for hospitalizations driven by malnutrition. Further qualitative and observational studies should explore clinical decision-making of providers and evaluate how admission practices differ across levels of the health care system in resource-limited emergency settings.

Table 1
Lambaréné Organ Dysfunction Score (LODS) is a mortality prediction score developed and validated for use in malaria endemic settings and was used as an objective measure of severity of illness in our study.

Table 2
Demographic, clinical, and laboratory characteristics of pediatric patients with LODS 0-1 presenting with febrile illness to the Pediatric Emergency Department (PED) in Tanzania, stratified by hospital disposition (admitted vs discharged).

Table 3
Distribution of primary infectious diagnosis by hospital disposition among pediatric patients with LODS 0-1 presenting with febrile illness to the Pediatric Emergency Department (PED), Tanzania.