Acute respiratory illness among a prospective cohort of pediatric patients using emergency medical services in India: Demographic and prehospital clinical predictors of mortality

Abstract

BACKGROUND: In India, acute respiratory illnesses, including pneumonia, are the leading cause of early childhood death. Emergency medical services are a critical component of India's public health infrastructure; however, literature on the prehospital care of pediatric patients in low- and middle-income countries is minimal. The aim of this study is to describe the demographic and clinical characteristics associated with 30-day mortality among a cohort of pediatric patients transported via ambulance in India with an acute respiratory complaint. METHODS: Pediatric patients less than 18 years of age using ambulance services in one of seven states in India, with a chief complaint of "shortness of breath", or a "fever" with associated "difficulty breathing" or "cough", were enrolled prospectively. Patients were excluded if evidence of choking, trauma or fire-related injury, patient was absent on ambulance arrival, or refused transport. Primary exposures included demographic, environmental, and clinical indicators, including hypoxemia and respiratory distress. The primary outcome was 7 and 30-day mortality. Multivariable logistic regression, stratified by transport type, was constructed to estimate associations between demographic and clinical predictors of mortality. RESULTS: A total of 1443 patients were enrolled during the study period: 981 (68.5%) were transported from the field, and 452 (31.5%) were interfacility transports. Thirty-day response was 83.4% (N = 1222). The median age of all patients was 2 years (IQR: 0.17-10); 93.9% (N = 1347) of patients lived on family incomes below the poverty level; and 54.1% (N = 706) were male. Cumulative mortality at 2, 7, and 30-days was 5.2%, 7.1%, and 7.7%, respectively; with 94 deaths by 30 days. Thirty-day mortality was greatest among those 0-28 days (N = 38,17%); under-5 mortality was 9.8%. In multivariable modeling prehospital oxygen saturation <95% (OR: 3.18 CI: 1.77-5.71) and respiratory distress (OR: 3.72 CI: 2.17-6.36) were the strongest predictors of mortality at 30 days. CONCLUSIONS: This is the first study to detail prehospital predictors of death among pediatric patients with shortness of breath in LMICs. The risk of death is particularly high among neonates and those with documented mild hypoxemia, or respiratory distress. Early recognition of critically ill children, targeted prehospital interventions, and diversion to higher level of care may help to mitigate the mortality burden in this population.