Background Prescription errors are common in neonatal intensive care units (NICU). Computerised physician order entry (CPOE) is one of the most effective interventions to decrease these errors but its implementation is expensive and time consuming. Completion of CPOE in our NICU is planned for 2015. Meanwhile, alternative options are necessary in order to improve the quality of prescriptions and to decrease medicines errors.

Purpose To assess whether a preformatted medical order sheet and a taught course had an effect on the quality of prescriptions and the frequency of errors during the prescription stage.

Materials and methods A two-phase observational study, pre- (Phase 0) and post-intervention (Phase I), over 4 consecutive months, was conducted in an 11-bed NICU.

Interventions included:

• introduction of a new preformatted medical order sheet (enhancement in completeness of prescriptions);

• a taught course on appropriate prescription and medicines errors.

Errors were identified every morning during the prescription process (medical round), using National Coordinating Council for Medication Error Reporting and Prevention taxonomy. Error rates between pre- and post-intervention phases were compared with a χ² test.

Results 83 patients were included in Phase 0 and 81 in Phase I.

505 prescriptions were analysed in Phase 0 and 523 in Phase I.

The rate of prescription errors decreased from 26.9% (Phase 0) to 15.3% (Phase I) (p < 0.05). Dose errors decreased from 11.7% to 4.6%, unit errors from 1.6 % to 0.4%, frequency errors from 3.4% to 1.3% and ambiguous prescriptions from 8.1% to 3.8%. Complete prescriptions increased from 0.2% to 33.3%.

Conclusions We demonstrated that a cheap and simple method to implement interventions can also improve completeness and intelligibility of prescriptions and decrease medicines errors.

No conflict of interest.