Antimicrobial Usage Among Acutely Ill Hospitalized Children Aged 2‒23 Months in Sub-Saharan Africa and South Asia

Understanding patterns of antimicrobial use is critical to supporting antibiotic stewardship and limiting antimicrobial resistance (AMR). We aimed to describe antimicrobial prescribing in acutely ill hospitalized children aged 2-23 months across a range of rural and urban hospital settings in Sub-Saharan Africa and South Asia.

The Childhood Acute Illness & Nutrition (CHAIN) cohort collected data daily throughout hospitalization from children with acute illness aged 2-23 months admitted to 9 hospitals from November 2016 to January 2019. We determined proportions of children receiving antimicrobials, inpatient-days receiving antimicrobials, antimicrobial classes, World Health Organization (WHO) Access, Watch, and Reserve (AWaRe) classifications, and examined factors associated with Watch antimicrobial use.

Of 3101 admissions, 1422 (46%) received antimicrobials before hospitalization. A total of 2816 (91%) children received antimicrobials during 19 398/21 807 (93%) inpatient child-days. Two thousand four hundred seventy-seven (76%), 1092 (35%), and 12 (0.3%) children received Access, Watch, and Reserve antimicrobials, mostly within 48 hours of admission. Three hundred forty-one (11%) admissions received an antimicrobial without any indication. Prior admission, chronic illness, diagnoses of sepsis or meningitis, hypoglycemia, and duration of admission were associated with receiving Watch antimicrobials, while WHO danger signs, severe malnutrition, HIV, and receipt of prior antimicrobials were not, despite their known association with mortality and AMR.

Antimicrobial use was similar across sites, with some overuse and notably limited escalation and de-escalation, likely due to guideline adherence. Guidelines need updating for the absence of relevant antimicrobial sensitivities, to include risk-based antimicrobial prescribing considering mortality risk and prior exposure to antimicrobials and the hospital environment. Hence, clinical trials of risk-differentiated care are needed.