Serial C-Reactive Protein Point-of-Care testing to optimize antibiotic treatment in hospitalized children with signs of infection in Zanzibar: A feasibility study

Bacterial infections are among the leading causes of morbidity and mortality in children, especially in low- and middle- income countries. As a result, antibiotics are frequently prescribed to children with signs of infection, even when cause may not be bacterial. This contributes to antimicrobial resistance (AMR), a global health concern. Integrating serial CRP point-of-care testing with culture and susceptibility testing may improve decision-making by facilitating earlier differentiation between bacterial and non-bacterial infections, promoting more appropriate antibiotic use.The study examined the feasibility of serial CRP POCT to guide discontinuation of antibiotic treatment in children in selected hospitals in Zanzibar, and to determine barriers to methods and procedures for the upcoming randomised controlled trial (RCT) (ISRCTN25937092).This prospective, individually randomized feasibility study was conducted between February 5 and March 3 2024 in two hospitals in Zanzibar. Research assistants and healthcare workers (HCWs) were trained on the use and interpretation of CRP POCT. Eligible neonates and children were randomized to CRP-guided or standard care antibiotic management. Feasibility was assessed using Bowen's framework; clinical outcomes were interpreted as exploratory.Eighty-two children participated, with a 96.0% follow-up rate. The CRP POCT intervention was rated "very important" by 89.7% of 58 HCWs. Adherence to CRP-based antibiotic guidance was high, though early discharges due to caregiver pressure were noted. The intervention integrated successfully into existing workflows and HCWs adapted the flowcharts in practice. Exploratory analysis showed CRP guidance reduced antibiotic treatment days in neonates with signs of infection (5.1 vs 6.6 days), and children aged 6 months to 12 years with febrile illness or diarrhoea (4.8 vs 6.7 days) compared to standard care, but the study was not powered for statistical inferences. This study suggests that the RCT of serial CRP POCT in guiding antibiotic treatment decision is feasible, and operationally implementable in this setting.