Authors (including presenting author) :
Chu HM(1), Law KM(1), Leung YS(1), Ng TM(1)
Affiliation :
(1)Department of Pharmacy, Queen Elizabeth Hospital
Introduction :
In-patient Medication Order Entry (IPMOE), an in-patient computerised physician order entry system developed in Hospital Authority since 2010s, is a closed-looped system for prescribing, medication order verification, drug dispensing, and drug administration.
IPMOE is a relatively recent clinical development. Locally, there is no published before and after studies conducted on the impact of IPMOE in terms of medication errors, adverse drug events nor pharmacist interventions (PIs). Therefore, this study is designed with the objective to investigate the impact of IPMOE on PIs in a selected setting.
Objectives :
To investigate how the implementation of IPMOE affects PIs in medication reconciliation (MR) and to explore changes in the frequency and types of medication errors.
Methodology :
A retrospective observational study was conducted in an acute hospital in Hong Kong, with a 47-bed male medical ward selected as the study ward. The study was conducted to study the period from 1 year before to 1 year after the implementation of IPMOE in the study ward. Patients hospitalised in selected ward with admission MR and/or discharge MR done by pharmacist were included in the study, whereas patients who had not been prescribed with any medications during hospitalisation were excluded. Rates and types of PIs, severity of drug-related problems (DRPs), acceptance of PIs, and the likelihood of detecting the DRPs with pharmacist on ward only were classified, enumerated and compared using chi-square test.
Result & Outcome :
Rate of PIs were found to significantly increase (number of study population with PI(s)/ total number of study population): from 17.8% (n=3071) to 20.3% (n=2980), p=0.014. Similarly, the rate of total number of DRPs/total number of medications reconciled increased from: 2.0% (n=50693) to 2.3% (n=51771), p=0.002. Significant decreases were found in three types of PIs - necessary information missing for complete prescription: from 1.6% (n=879) to 0 (n=900), p< 0.0001, unintended dosage/strength selection: from 12.1% (n=879) to 5.0% (n=900), p< 0.0001 and unintended frequency: from 10.6% (n=879) to 4.5% (n=900), p< 0.0001. The severities of DRPs were comparable in both periods. There were over 90% acceptances of PIs in both periods. The majority of DRPs was determined to be likely be detected only with pharmacist on ward.
