Factors influencing choices of empirical antibiotic treatment for bacterial infections in a scenario-based survey in Vietnam

JAC Antimicrob Resist. 2020 Dec;2(4):dlaa087. doi: 10.1093/jacamr/dlaa087. Epub 2020 Nov 10.

Abstract

Background: Antimicrobial stewardship (AMS) programmes have been implemented around the world to guide rational use of antibiotics but implementation is challenging, particularly in low- and middle-income countries, including Vietnam. Understanding factors influencing doctors' prescribing choices for empirical treatment can help design AMS interventions in these settings.

Objectives: To understand doctors' choices of antibiotics for empirical treatment of common bacterial infections and the factors influencing decision-making.

Methods: We conducted a cross-sectional survey among medical professionals applying for a postgraduate programme at Hanoi Medical University, Vietnam. We used a published survey developed for internal medicine doctors in Canada. The survey was self-administered and included four clinical scenarios: (i) severe undifferentiated sepsis; (ii) mild undifferentiated sepsis; (iii) severe genitourinary infection; and (iv) mild genitourinary infection.

Results: A total of 1011/1280 (79%), 683/1188 (57.5%), 718/1157 (62.1%) and 542/1062 (51.0%) of the participants selected combination therapy for empirical treatment in scenarios 1, 2, 3 and 4, respectively. Undifferentiated sepsis (OR 1.82, 95% CI 1.46-2.27 and 2.18, 1.51-3.16 compared with genitourinary) and severe infection (1.33, 1.24-1.43 and 1.38, 1.21-1.58 compared with mild) increased the likelihood of choosing a combination therapy and a carbapenem regimen, respectively. Participants with higher acceptable minimum threshold for treatment coverage and young age were also more likely to prescribe carbapenems.

Conclusions: Decision-making in antibiotic prescribing among doctors in Vietnam is influenced by both disease-related characteristics and individual factors, including acceptable minimum treatment coverage. These findings are useful for tailoring AMS implementation in Vietnam and other, similar settings.