Exploring determinants of antimicrobial prescribing behaviour using the Theoretical Domains Framework.

BACKGROUND: Few theoretically-based, qualitative studies have explored determinants of antimicrobial prescribing behaviour in hospitals. Understanding these can promote successful development and implementation of behaviour change interventions (BCIs). OBJECTIVE: (s): To use the Theoretical Domains Framework (TDF) to explore determinants of clinicians' antimicrobial prescribing behaviour, identifying barriers (i.e., impediments) and facilitators to appropriate antimicrobial practice. METHODS: Semi-structured interviews with purposively-sampled doctors and pharmacists with a wide range of specialties and expertise in Hamad Medical Corporation hospitals in Qatar. Interviews based on previous quantitative research and the TDF were audio-recorded, transcribed and independently analysed by two researchers using the TDF as an initial coding framework. RESULTS: Data saturation was achieved after interviewing eight doctors and eight pharmacists. Inter-related determinants of antimicrobial prescribing behaviour linked to ten TDF domains were identified as barriers and facilitators that may contribute to inappropriate or appropriate antimicrobial prescribing. The main barriers identified were around hospital guidelines and electronic system deficiencies (environmental context and resources); knowledge gaps relating to guidelines and appropriate prescribing (knowledge); restricted roles/responsibilities of microbiologists and pharmacists (professional role and identity); challenging antimicrobial prescribing decisions (memory, attention and decision processes); and professional hierarchies and poor multidisciplinary teamworking (social influences). Key facilitators included guidelines compliance (goals and intentions), and participants' beliefs about the consequences of appropriate or inappropriate prescribing. Further education and training, and some changes to guidelines including their accessibility were also considered essential. CONCLUSIONS: Antimicrobial prescribing behaviour in hospitals is a complex process influenced by a broad range of determinants including specific barriers and facilitators. The in-depth understanding of this complexity provided by this work may support the development of an effective BCI to promote appropriate antimicrobial stewardship.

The use of theory in the development and evaluation of behaviour change interventions to improve antimicrobial prescribing: a systematic review.

OBJECTIVES: This systematic review (SR) reviews the evidence on use of theory in developing and evaluating behaviour change interventions (BCIs) to improve clinicians' antimicrobial prescribing (AP). METHODS: The SR protocol was registered with PROSPERO. Eleven databases were searched from inception to October 2018 for peer-reviewed, English-language, primary literature in any healthcare setting and for any medical condition. This included research on changing behavioural intentions (e.g. in simulated scenarios) and research measuring actual AP. All study designs/methodologies were included. Excluded were: grey literature and/or those which did not state a theory. Two reviewers independently extracted and quality assessed the data. The Theory Coding Scheme (TCS) evaluated the extent of the use of theory. RESULTS: Searches found 4227 potentially relevant papers after removal of duplicates. Screening of titles/abstracts led to dual assessment of 38 full-text papers. Ten (five quantitative, three qualitative and two mixed-methods) met the inclusion criteria. Studies were conducted in the UK (n = 8), Canada (n = 1) and Sweden (n = 1), most in primary care settings (n = 9), targeting respiratory tract infections (n = 8), and medical doctors (n = 10). The most common theories used were Theory of Planned Behaviour (n = 7), Social Cognitive Theory (n = 5) and Operant Learning Theory (n = 5). The use of theory to inform the design and choice of intervention varied, with no optimal use as recommended in the TCS. CONCLUSIONS: This SR is the first to investigate theoretically based BCIs around AP. Few studies were identified; most were suboptimal in theory use. There is a need to consider how theory is used and reported and the systematic use of the TCS could help.