Drug administration errors are a common problem; they contribute to unintended patient harm and may in part be due to distractions during the drug administration round. In this study, a prospective sample of 38 drug rounds on an acute surgical ward were observed, and their duration, time spent dealing with interruptions and nature of the interruptions were recorded. On average, 11% of each drug round was spent dealing with interruptions. There were one or more interruptions in two-thirds of the rounds studied (average 2.61 interruptions per round), with a mean duration of 1 minute per interruption. The interruptions came from doctors (21%), other nurses (17%), patients (11%), telephone enquiries (8%), relatives (3%) or were initiated by the nurse conducting the round (21%). This pattern contradicts the subjective impressions of nurses in previous questionnaire studies, but it was notable that the longest individual interruptions were from conversations with patients (mean 249 seconds) and phone calls (mean 212 seconds). The data confirm the frequency of interruptions, and their potential as a safety hazard. Objective information from direct observation will prove valuable in designing possible solutions to the problem. These will require local knowledge and frontline staff involvement to be sustainable.

Lean can substantially and simultaneously improve compliance with a bundle of safety related processes. Given the interconnected nature of hospital care, this strategy might not translate into improvements in safety outcomes unless a system-wide approach is adopted to remove barriers to change.

An incident is more likely to be reported if harm results. Surgical complications are not generally perceived to be "reportable incidents," but they are addressed in Mortality and Morbidity meetings (M&M). Integrating M&M and incident reporting data will result in more comprehensive healthcare safety systems.