Scarce supervision, strong pressures: Systemic drivers of antibiotic prescribing by intern physicians in Bangladesh

Background: Antimicrobial resistance (AMR) is a global health crisis driven by the inappropriate use of antibiotics. Prescribing habits are often shaped during internship, and studies suggest interns usually follow attending physicians’ guidance when choosing antibiotics. However, evidence from low- and middle-income countries (LMICs), including Bangladesh, remains scarce. This study explored antibiotic prescribing practices and systemic drivers among medical interns.

Methods: We conducted a cross-sectional survey between July and August 2025 in 13 public tertiary- care hospitals in Bangladesh. A structured, pre-tested questionnaire was administered in person and online to consenting interns. The tool assessed timing of antibiotic initiation, empirical use, clinical triggers, routes of administration, and systemic influences. Data were analysed using descriptive statistics.

Results: Among 393 respondents, 330 (84%) reported prescribing antibiotics independently. A total of 71.2% initiated empirical therapy after clinical diagnosis but before laboratory confirmation, while only 9.7% routinely waited for culture results, and 42.2% did not wait at all. Commonly cited reasons for empirical initiation were fear of treatment delays (54%), severe or worsening symptoms (48.1%), guidance from senior staff (28%), and hospital traditions (26.7%).

Clinical symptoms most frequently prompting empiric use were fever (12.5%), wound infections (14.8%), and productive cough (8.4%). Beta-lactam antibiotics were prescribed by 96.2% of intern physicians, while 27% reported prescribing fluoroquinolones irrespective of symptoms. Regarding administration routes, 80.4% reported prescribing intravenous therapy, 64.9% oral therapy, and 45.3% both.

Systemic constraints strongly shaped prescribing choices. Pharmacy stock influenced antibiotic selection in 51% of cases. Time constraints due to heavy patient load were reported by 52.4% of interns, and hierarchical influence from senior physicians by 49.1%. Institutional support was limited: 257 interns (65.4%) reported no clear hospital antibiotic guidelines, and only 19.1% found existing guidelines easily accessible. Notably, 44.3% of interns indicated that accessible protocols would improve their practice.

Conclusion: Intern physicians in Bangladeshi tertiary hospitals frequently prescribe antibiotics autonomously, often empirically and without microbiological confirmation. Their prescribing decisions are shaped less by evidence than by systemic and institutional pressures, including lack of guidelines, pharmacy shortages, time constraints, and hierarchical expectations. The predominant use of broad-spectrum beta-lactams further underscores risks for antimicrobial resistance escalation. Enhanced supervision, accessible guidelines, improved diagnostic capacity, and pharmacy support are urgently needed to promote rational antibiotic use and strengthen stewardship.