Antibiotic resistance quietly reshaping hospital safety worldwide rarely receives the granular, pre-and-post measurement that would reveal whether targeted interventions actually move the needle. This study from Vietnamese district hospitals offers exactly that kind of accountability data — and the setting matters, because Southeast Asia carries among the highest burdens of multidrug-resistant organisms globally, in both clinical and community populations.
Four Vietnamese district hospitals — two urban facilities in Hanoi, two rural ones in Ca Mau — served as the study sites. Cross-sectional prevalence surveys were conducted immediately before and at the conclusion of a six-month antimicrobial stewardship (AMS) bundle launched in March 2022. The intervention combined institutional stewardship committees, standardized prescribing guidelines, clinician education, and regular prescribing audits with direct feedback. Bacterial carriage was assessed through stool or rectal swabs for extended-spectrum β-lactamase (ESBL)-producing Enterobacterales and carbapenemase-producing Enterobacterales (CPE), alongside nasal swabs for methicillin-resistant Staphylococcus aureus (MRSA). Phenotypic susceptibility testing and molecular characterization were both employed, strengthening the microbiological precision of the findings.
The study's design — repeated cross-sectional rather than longitudinal cohort — limits causal inference; changes in carriage prevalence between survey waves could reflect patient turnover, seasonal variation, or community-level shifts rather than intervention effects alone. Still, the bundled AMS approach tested here mirrors what the WHO and global stewardship networks recommend for resource-limited settings, making this one of the more ecologically valid tests of such programs in the region. For health-conscious adults, the takeaway is systemic: antibiotic overuse in any healthcare system increases the reservoir of resistance that eventually affects everyone. Programs demonstrating measurable impact, even modest reductions in ESBL carriage rates, contribute incrementally to a field that urgently needs scalable, low-resource models. This work is best read as confirmatory in design ambition and potentially instructive in implementation detail.