Development of a programme theory for the impact of barcode medication administration on patient safety in UK hospital settings

Abstract Introduction Medication errors most commonly occur during medication administration, with an estimated 37 million medication administration errors (MAEs) annually in English hospitals, of which 7.5% cause moderate/severe patient harm1. Barcode medication administration (BCMA) systems are increasingly implemented in hospital settings, with the aim of decreasing MAEs. However, little is known about the underlying mechanisms that support their use. Aim To develop a programme theory for how BCMA is expected to work in practice that can be tested in future research. Methods This study forms part of a mixed methods realist evaluation2 at three hospital trusts, two in London and one in the Southwest. Phase 1 involved a narrative review to develop a realist programme theory. One researcher extracted contexts (C), mechanisms (M) and outcomes (O) from relevant literature, then worked with the research team and advisory group to build CMO configurations (CMOCs) and preliminary programme theory. In Phase 2, a researcher conducted qualitative interviews with key-informants to refine the preliminary programme theory. The interviews were used to create criteria to choose and refine relevant CMOCs and create a refined programme theory. We worked with a PPI partner who reviewed study documents, interview guide and programme theory. Results The review included 31 studies, of which four were from the UK. We interviewed six key informants (three pharmacists, two nurses and a human-factors specialist). We identified nine proximal and six distal outcomes related to possible MAEs and medication administration timeliness. Findings informed the development of a programme theory (Fig. 1) comprising 13 CMOCs for how BCMA is expected to work. Conclusion Our study was conducted in three hospital trusts using two different electronic health record systems. A limitation is the risk of social desirability bias among staff interviewees. The programme theory shows how contexts (e.g. nurses’ perception of the impact of BCMA on patient safety, use of BCMA for specific drug rounds or medications) can trigger mechanisms and outcomes related to use of BCMA and an increase or decrease in MAEs. The programme theory will be tested using observation of medication administration, analysis of BCMA alert data, and interviews with nurses and patients.