Improving the Accuracy of the Discharge Medication Information using an Electronic Discharge Referral System
21 October 2005
North Sydney/Central Coast Area Health Service
This project was a finalist in the Baxter 2005 NSW Health Awards, Safety of Healthcare category.
Entries from the Baxter 2005 NSW Health Awards - full list
Author
M Duguid, Pharmacy Department Royal North Shore Hospital, NSCCAHS
For more information on this project please contact N.Golan
Abstract
The project aimed to improve the accuracy of medication information provided to community care providers at time of discharge and reduce potential for medication errors using a web-based Electronic Discharge Referral System (eDRS). Discharge prescriptions were screened by pharmacists and amendments in discrepancies incorporated into the eDRS business processes.
The project was evaluated for consistency of information, percentage and potential severity of errors detected on discharge summaries in the eDRS compared with the "Old" paper system for patients discharged between April 2003 and April 2005. It was found that the proportion of prescriptions discrepancies between original discharge prescription and computer version was 4% (eDRS) versus 20.45% ("Old"). The average severity of these errors was coded as Class 2 (eDRS) versus Class 3.3 ("Old"), where Class 1 has no clinical impact and Class 5 has major severity. The most frequent type of error detected was Class 2 (eDRS) versus class 4 ("Old").
Aim
To improve the accuracy of medication information provided to community care providers at time of discharge and reduce potential for medication errors using a web-based Electronic Discharge Referral System (eDRS).
Background
It has been shown that pharmacists screening discharge prescriptions reduce medication errors on discharge and improve the quality of discharge information supplied to community care providers (1,2). A previous study at Royal North Shore Hospital (RNSH) demonstrated nearly 12% of patients having at least one error on their discharge prescription (1). At RNSH, pharmacists screen all discharge prescriptions presented to the Pharmacy Department against the current medication charts. Any discrepancies are followed up with the prescriber who is responsible for amending the carbon copy of the discharge summary with any changes. Failure to do this can potentially lead to medication errors after discharge and an inaccurate medication history.
The piloting of a web-based eDRS in Northern Sydney Health (NSH) hospitals provided the opportunity to redesign processes to improve the accuracy of the medication information in the discharge summary.
Methodology
A Pharmacy eDRS working party team was established comprising pharmacists and the eDRS Project Team. Current processes were reviewed in conjunction with key hospital staff to identify issues surrounding legal, manual and automated processes to be addressed for the implementation of better practice in supporting continuity of care for medication. It was quickly determined that discharge medication processes used across NSH were not supported well by the software product.
Several enhancements were made including printing of a prescription (to allow for checking pre-discharge), development of a user friendly template and an audit facility. An additional enhancements of a lookup interface to the Pharmacy formulary is currently in development. The working party reported directly to the eDRS Project Implementation Committee.
A template for the medication section was designed in consultation with the senior and junior medical staff. Approval was gained from Pharmaceutical Services at NSW Health to use the template as a non-handwritten prescription. A Deed of Agreement was approved that allowed pharmacist to amend the discharge medication template after consultation with the prescriber. The system also enabled pharmacists to document any interventions made to the prescription in the electronic record of the summary and provided the facility for pharmacists to communicate medication related issues to GPs and other community care providers.
Planning and Implementation
eDRS was introduced into the acute aged care ward at RNSH in March 2004 as the first of the pilot sites in NSH. To date eDRS is live in 10 wards across the area. The project team provided training on site for the users via a mobile laptop fleet approximately 2 weeks before each ward go-live and at the change of doctors at each term. A specific training package was developed for pharmacy staff. Pharmacists also assisted with training prescribers on the wards.
Throughout the pilot phase there were regular meetings to review the system and a number of improvements were made to the medication section.
The major change in practice for pharmacists was the ability to amend the discharge summary on line following verbal discussion with the prescriber of discrepancies identified in discharge prescriptions/summaries. Pharmacists were also required to sign off on the summary. This ensured that any summaries containing information on medication were checked prior to a patient’s discharge.
Outcomes and Evaluation
Data was collected retrospectively from the medication records of aged care patients who were discharged in April 2005 (eDRS arm) and April 2003 ("Old" system arm). The discharge prescription (pharmacy record) was compared to the discharge summaries sent to the patient’s General Practitioner (GP). Discrepancies were categorised into administrative (ie prescribers name, signature, date missing) and clinical ie medication errors.
The medication errors were classified for potential severity, class 1 (No clinical impact) to class 5 (Major severity).
Outcomes
Fifty and 44 summaries were reviewed in the eDRS and "Old" system arm respectively.
Clinical/ Medication errors:
- Consistency.
- Discrepancies between original discharge prescription and computer version was 4% (eDRS) versus 20.45% ("Old").
- Severity The average severity of these errors was Class 2 (eDRS) versus Class 3.3 ("Old").
- The most frequent type of error was Class 2 (eDRS) versus class 4 ("Old").
Administrative Errors:
- There were 0% administrative errors in the eDRS compared to 9.1% of errors in the ‘old’ system
Conclusion:
The results demonstrated that pharmacists checking discharge prescriptions and amending discrepancies in discharge summaries using a web-based eDRS improved the accuracy of medication information provided to community care providers and reduced potential for medication errors.
The results also concur with the results of a survey of GPs and Nursing Homes undertaken earlier in the trial where 78% of General Practitioners surveyed reported a major advance in communication between primary and secondary care, with 97% reporting that the medication section contained clear and concise data on patient’s medication.
Future Scope
A major enhancement to the system is the integration of the eDRS with the pharmacy information system software to allow prescribers to select drugs from the pharmacy database/formulary at the time of prescribing. A part of this enhancement will also allow a specific report to be sent automatically to a patient’s nominated community pharmacy. This enhancement has been supported by NSW Health.
The eDRS will be rolled out to all hospital discharges across NSCCH therefore incorporating the important quality step of pharmacists screening all discharges and signing off of discharge summaries prior to their transfer to the community care providers. The benefits of this improved screening of discharge medication orders cannot be understated from the point of view of patient safety, appropriateness of care, and economic savings.
There is also the potential for including recommendations for Home Medicines Review in those patients identified at risk of adverse medication events post discharge.
References
- Duguid M, Gibson M, O’Doherty R. Review of discharge prescriptions by pharmacists integral to continuity of care. Journal of Pharmacy Practice and Research 2002;32:94-95
- Whitty JA, Green B, Cottrell WN. A Study to determine the importance of ward pharmacists reviewing discharge prescriptions. Aust J Hosp Pharm 2001;31: 300-2.
