Introduction: Medication errors are common, with studies reporting at least one error perpatient encounter. At hospital discharge, medication errors vary from 15%-38%. However,studies assessing the effect of an internally developed electronic (E)-prescription systemat discharge from an emergency department (ED) are comparatively minimal. Additionally,commercially available electronic solutions are cost-prohibitive in many resource-limitedsettings. We assessed the impact of introducing an internally developed, low-cost E-prescriptionsystem, with a list of commonly prescribed medications, on prescription error rates at dischargefrom the ED, compared to handwritten prescriptions.
Methods: We conducted a pre- and post-intervention study comparing error rates in a randomlyselected sample of discharge prescriptions (handwritten versus electronic) five months pre andfour months post the introduction of the E-prescription. The internally developed, E-prescriptionsystem included a list of 166 commonly prescribed medications with the generic name, strength,dose, frequency and duration. We included a total of 2,883 prescriptions in this study: 1,475 inthe pre-intervention phase were handwritten (HW) and 1,408 in the post-intervention phase wereelectronic. We calculated rates of 14 different errors and compared them between the pre- andpost-intervention period.
Results: Overall, E-prescriptions included fewer prescription errors as compared to HWprescriptions.Specifically, E-prescriptions reduced missing dose (11.3% to 4.3%, p <0.0001),missing frequency (3.5% to 2.2%, p=0.04), missing strength errors (32.4% to 10.2%, p <0.0001)and legibility (0.7% to 0.2%, p=0.005). E-prescriptions, however, were associated with a significantincrease in duplication errors, specifically with home medication (1.7% to 3%, p=0.02).
Conclusion: A basic, internally developed E-prescription system, featuring commonly usedmedications, effectively reduced medication errors in a low-resource setting where the costs ofsophisticated commercial electronic solutions are prohibitive. [West J Emerg Med. 2017;18(5)943-950.]