Effect of barcode technology on medication preparation safety: a quasi-experimental study

Background: Medication errors are a common problem in hospitals and a major driver of adverse patient outcomes. Barcode verification technology is a promising strategy to help ensure safe medication preparation practices. Objectives: The objectives of this study were (i) to assess the effects of a barcode-assisted medication preparation and administration system regarding the rate of medication preparation errors and (ii) to compare the time spent on medication preparation tasks by medical staff. Methods: A quasi-experimental study with a pre-post design was conducted, from August 2017 to July 2018, in two mixed medical/surgical units of a tertiary teaching hospital. The primary aim was to assess the effects of a barcode-based electronically assisted medication preparation and administration system linked to the hospital's electronic medication administration record regarding the rate of medication preparation errors and time-based staff performance. Data were collected using direct observation. Adjusted and unadjusted logistic models were used for error frequencies and linear regression models for time performance. Results: 5932 instances of medication selection and dosing during 79 medication preparation procedures were observed. The overall medication preparation error incidence decreased from 9.9% at baseline to 4.5% at post-intervention, corresponding to a relative risk reduction of 54.5% (P < 0.001). However, the adjusted effect by registered nurses (RNs) and nurses' work experience of total medication preparation errors showed only borderline significance (odds ratio [OR] 0.64, P = 0.051). For adjusted error-specific analyses, significant error reductions were found in wrong medication errors (OR 0.38, P < 0.010) and wrong dosage errors (OR 0.12, P = 0.004). Wrong patient, wrong form and ambiguous dispenser errors did not occur at post-intervention. Errors of omission (OR 1.53, P = 0.17), additional doses (OR 0.63, P = 0.64) and wrong dispenser boxes (OR 0.51, P = 0.11) did not change significantly. The time necessary to prepare medications for a 24-h period also decreased significantly-from 30.2 min to 17.2 min (beta = -6.5, P = 0.047), while mean preparation time per individual medication dose fell from 24.3 s to 15.1 s (beta = -5.0, P = 0.002). Conclusion: Use of the new barcode technology significantly reduced the rate of some medication preparation errors in our sample. Moreover, the time necessary for medication preparation, both per 24-h period and per single-medication dose, was significantly reduced.