Why local antibiotic resistance data matters - Informing empiric prescribing through local data collation, app design and engagement in Zambia.

BACKGROUND: Control of antimicrobial resistance (AMR) relies on local knowledge and local intervention implementation. Effective antibiotic stewardship requires locally-suitable prescribing guidelines. We aimed to use a novel digital tool (the ZARIApp) and a participatory approach to help develop locally-relevant empiric antibiotic prescribing guidelines for two hospitals in Lusaka, Zambia. METHODS: We produced an AMR report using samples collected locally and routinely from adults within the prior two years (April 2020 - April 2022). We developed the ZARIApp, which provides prescribing recommendations based on local resistance data and antibiotic prescribing practices. We used qualitative evaluation of focus group discussions among healthcare professionals to assess the feasibility and acceptability of using the ZARIApp and identify the barriers to and enablers of this stewardship approach. RESULTS: Resistance prevalence was high for many key pathogens: for example, 73% of 41 Escherichia coli isolates were resistant to ceftriaxone. We identified that high resistance rates were likely due to low levels of requesting and processing of microbiology samples from patients leading to insufficient and unrepresentative microbiology data. This emerged as the major barrier to generating locally-relevant guidelines. Through active stakeholder engagement, we modified the ZARIApp to better support users to generate empirical antibiotic guidelines within this context of unrepresentative microbiology data. Qualitative evaluation of focus group discussions suggested that the resulting ZARIApp was useful and easy to use. New antibiotic guidelines for key syndromes are now in place in the two study hospitals, but these have substantial residual uncertainty. CONCLUSIONS: Tools such as the free online ZARIApp can empower local settings to better understand and optimise how sampling and prescribing can help to improve patient care and reduce future AMR. However, the usability of the ZARIApp is severely limited by unrepresentative microbiology data; improved routine microbiology surveillance is vitally needed.

Antimicrobial usage at a large teaching hospital in Lusaka, Zambia.

Antimicrobial resistance is a growing global health concern. Antimicrobial stewardship (AMS) curbs resistance rates by encouraging rational antimicrobial use. However, data on antimicrobial stewardship in developing countries is scarce. The objective of this study was to characterize antimicrobial use at the University Teaching Hospital (UTH) in Lusaka, Zambia as a guiding step in the development of an AMS program. This was a cross-sectional, observational study evaluating antimicrobial appropriateness and consumption in non-critically ill adult medicine patients admitted to UTH. Appropriateness was defined as a composite measure based upon daily chart review. Sixty percent (88/146) of all adult patients admitted to the general wards had at least one antimicrobial ordered and were included in this study. The most commonly treated infectious diseases were tuberculosis, pneumonia, and septicemia. Treatment of drug sensitive tuberculosis is standardized in a four-drug combination pill of rifampicin, isoniazid, pyrazinamide and ethambutol, therefore appropriateness of therapy was not further evaluated. The most common antimicrobials ordered were cefotaxime (n = 45), ceftriaxone (n = 28), and metronidazole (n = 14). Overall, 67% of antimicrobial orders were inappropriately prescribed to some extent, largely driven by incorrect dose or frequency in patients with renal dysfunction. Antimicrobial prescribing among hospitalized patients at UTH is common and there is room for optimization of a majority of antimicrobial orders. Availability of certain antimicrobials must be taken into consideration during AMS program development.