Outpatient Pediatric Hypoxemia Prevalence and Pulse Oximetry Implementation in Malawi

Introduction: Pneumonia remains a leading cause of under-five-year-old death globally, with hypoxemia a major mortality risk factor. The World Health Organization Integrated Management of Childhood Illness (IMCI) guidelines for childhood pneumonia management in low-and-middle-income countries like Malawi include routine outpatient pulse oximetry screening for hypoxemia; however, implementation is limited. To address evidence gaps on the burden of outpatient hypoxemia and pulse oximetry implementation we sought to determine the hypoxemia prevalence among children presenting to an outpatient clinic in Malawi. We also evaluated the measurement burden and hypoxemia yield of varying pulse oximeter implementation strategies within the IMCI framework.


Methods: We conducted a cross-sectional study of children aged 1–59 months in the outpatient department clinic of Salima District Hospital, Malawi. Study staff trained in pulse oximetry enrolled every fifth patient over five days per month over 12 months from 2024-2025, and collected participant demographic and clinical data, and measured room air capillary oxyhemoglobin saturation (SpO₂) (Masimo RadG). Fast breathing was a respiratory rate >60 breaths per minute (bpm) in 1-month-olds, ≥50 bpm in 2–11-month-olds, and >40 bpm in 12–59-month-olds. IMCI pneumonia was cough or difficulty breathing with fast breathing.


Results: An average of 125 children daily (range 56–247) attended the outpatient department clinic. Of 1,134 enrolled children, 12 (1.1%) had a SpO₂ < 94%, including 3 (0.3%) with a SpO₂ < 90%. IMCI pneumonia was present in 218 (19.2%) children and achieved a 67% (8/12) sensitivity and 81% specificity for identifying a SpO₂ < 94% (100% sensitivity for a SpO₂ < 90%). Cough and/or difficulty breathing was reported in 431 (38.0%) children; screening this group detected all SpO₂ < 94% cases (100% sensitivity) but required nearly double the oximeter assessments than in IMCI pneumonia cases.


Conclusions: Hypoxemia was lower than anticipated despite a high burden of respiratory illness. In this outpatient Malawian setting health worker pulse oximeter measurement of children identified with cough and/or difficulty breathing had the highest sensitivity for identifying hypoxemia but could exceed existing healthcare worker testing capacity.