Fighting antibiotic resistance in Portuguese hospitals: Understanding antibiotic prescription behaviours to better design antibiotic stewardship programmes

Objectives: Since physicians play an important role in antibiotic usage, it is vital to understand their antibiotic-prescribing behaviour and knowledge on antimicrobial resistance in order to develop and implement effective antibiotic stewardship interventions. The aim of this study was to evaluate Portuguese physicians’ knowledge and to understand prescription behaviours, difficulties and barriers in their antibiotic prescription process in order to promote better and well-adapted antibiotic stewardship policies. Methods: This study was conducted in 2016 using a self-administered questionnaire to physicians in two tertiary public hospitals from two different regions in Portugal. Results: Participating physicians [response rate 47.6% (30/63)] identified antibiotic resistance as a global problem; however, one-third did not recognise antibiotic resistance as a major problem on their own hospital. Factors that most influenced antibiotic prescription were ‘microbiology laboratory results’, ‘patient clinical situation’ and patient ‘co-morbidities’. On the other hand, ‘colleagues’ opinion’ and ‘costs control’ were considered as less determining factors. Regarding difficulties and bottlenecks in the antibiotic prescription process, participant physicians reported ‘lack of (or delayed) microbiological results’ and ‘no access to antibiotic susceptibility patterns’ as major barriers. ‘Education and training’ was considered the most effective intervention to improve antibiotic prescription. Conclusion: These results suggest that the design and implementation of antibiotic stewardship interventions should provide better data management and sharing tools between physicians and the microbiology laboratory, especially through the creation of antimicrobial prescribing guidelines according to hospital epidemiology, and easy access to hospital antibiotic susceptibility patterns and epidemiological data.