Background and importance There is little quantitative evidence in the current literature on the incidence of incorrect medication administration of intravenous medications by clinicians. Analysis of infusion pump medication library alert-logs for user initiated corrections of erroneous medication selections adds quantitative data to the investigation of the incidence of lookalike–soundalike intravenous medication administration errors.

Aim and objectives To establish baseline data from infusion pump medication library alert-logs on near miss and user correction lookalike–soundalike medication selection errors during intravenous medication administration.

Material and methods A 12 month facility wide retrospective review of infusion pump medication library alert-logs was conducted to obtain metrics on the set-up phase of intravenous medication administration. Cancelled infusions and user initiated resolutions of incorrect medication selections were analysed. Decision times of clinicians were calculated from the time–date stamps of the pumps’ alert-logs.

Results Incorrect medication selection represented 3.45% (10 017/290 807) of all medication library alerts and 22.40% (10 017/44 721) of all cancelled infusions. Average selection error recognition to cancellation and correction time was 27.00 s (SD 22.25). Medications with longer names were more prevalent among the initial selection errors. Users were more likely to make selection errors with the first part of the medication names (6991/10 017, 69.79%) while the middle part of the medication names were the next most likely to be misidentified (2144/10 017, 21.40%), with name ending confusion being responsible for 8.80% of errors (882/10 017).

Conclusion and relevance The study provides a quantitative appraisal of an area that has been resistant to measurement, and identified a high number of near miss lookalike–soundalike errors. This phenomenon, while largely centred on initial misreading of the beginning of the medication name, also ran through the middle and end portions of medication nomenclature. The value of an infusion pump showing the entire medication name complete with TALLman lettering on the user interface, so that the selection fully matches that of the medication’s pharmacy label, is supported by these findings. FMEA type infusion pump strategies that use multiple distinct user confirmation steps (eg, medication selection followed by therapy selection, and user confirmation of clinical advisories) may reduce the risk of incorrect medication selection, as medications which require the user to pass through such selection and confirmation steps were far less prevalent in the near miss lookalike–soundalike error group.

Conflict of interest No conflict of interest