Introduction
The national strategic programme, published by the Department of Health, is concerned with major developments in the deployment and use of Information Technology (IT) in the NHS. In line with the plan for IT and other Intensive Therapy Units (ITUs) around the country, the ITU at the Royal Sussex County Hospital (RSCH) implemented the electronic Clinical Information System (CIS) in 2002 with the prescribing side of the system being fully operational from August 2004. It was noted that during the initial period following the implementation of full electronic prescribing, certain errors were recurring and potentially putting patient safety at risk so the aim of this audit was to establish a baseline level of error so that solutions could be formulated and implemented to improve practice.

Methodology
Data collection was performed from 30 August 2004 to 30 September 2004; approximately one month after full electronic prescribing was implemented. The errors were identified by the pharmacist during the clinical screening of the electronic prescriptions and recorded by way of a customised data collection form.

Results
The majority of prescribing errors occurred in the first four days of the ITU admission. The number of errors as a result of the electronic prescribing documented during the audit was 67 and the number of general clinical pharmacy interventions over the same time period was 60. This represents an increase in the pharmacist’s workload of 112 percent. The most common drugs associated with errors were salbutamol nebules, ranitidine and paracetamol. The most common errors were duplication of prescriptions (n=31) by doctors and also nurses, possibly accidentally, and incorrect frequency (n=12). As the most common error was duplication, the action taken in 49 percent of cases was to cancel those duplicated prescriptions (n=33). Other courses of action were for the SHO to re-prescribe the drug entry (n=15) and the computer’s default for a specific drug was altered in an attempt to minimise error with that entry. Errors that were made and could not be undone used "sticky notes" or comments on the drug tab to justify the action.

Conclusion
The audit fulfilled the aim of establishing a baseline level of error with electronic prescribing on the CIS. The types of errors have been identified and steps can now be taken to minimise them. The two main areas have been targeted for improvement as result of this audit are training and programming. The issues that require alterations to programming could be reviewed and implemented where possible to facilitate ease of use. Training needs of staff are being addressed and it is planned that the pharmacist in conjunction with the system programmers will design a competency-based workbook in time for the new team of SHOs that are due to rotate to the ITU in February 2005. The workbook will include examples of errors from this audit as a learning tool. There should be a re-audit at a suitable time interval to ascertain if the measures taken as a result of this audit have been effective.