Association of Scottish Trust Chief Pharmacists
Electronic prescribing: the way forward?
The Association of Scottish Trust Chief Pharmacists held a conference on electronic prescribing in Stirling on November 1,
aimed at increasing understanding of the issues around electronic prescribing in Scotland. Vince Summers reports
In a keynote address, Mr CHARLIE KNOX (head of computing and information technology strategy, Scottish Executive Health Department) explained how electronic prescribing was viewed in terms of the National Health Service in Scotland strategy for IT. The current focus was on getting the basics of clinical support in place, especially in hospitals. These would begin to allow the development of more advanced systems such as electronic prescribing and digital X-ray systems.
The basics required for electronic prescribing included adequate networks, a developing electronic patient record (EPR), drug databases that could be linked between primary and secondary care, and methods of interchanging data between the two sectors to allow medication histories to move with the patient. The evaluation of the electronic prescribing pilot in Ayr hospital would also provide valuable information for developing and implementing systems in other sites.
Evaluation of Ayr pilot
The evaluation of the pilot at Ayr hospital was presented by a team consisting of Mrs MICHELE CALDWELL (chief pharmacist, Ayrshire and Arran acute NHS trust, and ASTCP chairman), Mrs FRANCES FOWLIE (clinical pharmacist) and Mr JAMES SNELL (project manager for electronic prescribing, Ayr hospital). They outlined the reduction in the risks that an electronic prescribing and administration (EPA) system could make possible and the process of implementing the system for orthopaedics in Ayr hospital. These included issues with staff training, software development, IT support and hardware problems.
The evaluation had indicated that EPA was at least as safe as the current paper-based system. The incidence of most types of medication error was reduced, but errors of omission increased. This seemed to be due to medicines not being prescribed on admission because details of dose and frequency were not available and the system would not allow prescribing without them. EPA in Ayr had been shown to realise some of the expected benefits in improving the quality and safety of prescribing and reducing the risk of the medicines process.
EPA at Ayr had also highlighted the problems of fully testing and validating complex software for a patient critical system. It was becoming increasingly difficult to check that software performed fully as expected in a test environment, and so problems would not be found until they occurred in live situations. Software updates could also be a problem as the improvements on the previous version might be accompanied by errors not previously present. There was a need to develop some formal method of software validation analogous to medicines licensing.
Summing up the main theme of the morning session, Mr BILL SCOTT (chief pharmaceutical officer, Scottish Executive Health Department) highlighted the need for a national drug database/dictionary or a system to link the currently available ones and a method of validating critical software in order to allow the benefits of electronic prescribing to be more widely available.
Decision support
Giving a presentation on the possibilities for decision support in electronic prescribing, Mr CLIVE MOSS-BARCLAY (chief pharmacist, Royal Oldham hospital) said that 6.7 per cent of people admitted to hospital were subject to serious or potentially serious medication error. In addition, information overload in respect of medicines was increasing: for example, the British National Formulary listed 17,000 possible drug interactions. Decision support at the time of dispensing could improve the quality and effectiveness of prescribing, prevent information overload and present unbiased, evidence-based information at the appropriate time.
Decision support aimed to remove from the system any inappropriate choices to prevent unsafe practice, act in anticipation of the needs of the prescriber, provide access to appropriate information in a timely manner and provide solutions to complex problems. Clinical decision support should be designed to: initiate investigations, eg, radiology, pathology; propose actions or deny the user inappropriate actions, or warn the user about the potential effects of a "risky "decision; provide access to relevant data; carry out complex processes (eg, design of total parenteral nutrition regimens, complex dosage calculations) and process multiple data sets (patient, drugs, disease) to develop action plans based on critical care pathways. This would really start to assist in improving the quality of the medicines process.
Mr IAN CLARKE (Wolfson computer laboratory, University of Birmingham, Queen Elizabeth medical centre) described the system developed for prescribing and administration of medicines in the hospital's renal unit. It had been developed with considerable input from clinicians and had a rules-based decision support system designed into it. Laboratory results and other patient information were also part of the system and were integrated into some of the rules of decision support for prescribing. The system worked on hand-held or static computers and seemed to be well received by users. The system had now been taken on by a large software supplier and was being developed in conjunction with them, including developing its use in clinical areas other than the renal unit.
Mr SEAN BRENNAN (Northgate Solutions; formerly Department of Health) described the levels of the electronic patient record and where electronic prescribing was likely to fit into this as an essential requirement. Barriers to implementing electronic prescribing had been identified from workshops held with those who were currently operating systems. It was important to learn from the experiences of those who had already implemented systems in order to make it less problematic for subsequent developments. Barriers included software development, time-scales for procurement and implementation, cultural and organisational issues, level of investment, lack of a long term vision and strategy for EPR and EPA, the complexity of the prescribing process, failure to transfer knowledge and experience, senior management and clinician commitment, and inappropriately worded legislation and official guidance.
Extent of implementation
Mr VINCE SUMMERS (chief pharmacist, Borders General hospital) presented the results of a survey of current implementation of electronic prescribing in United Kingdom hospitals. Of 215 questionnaires sent to trust chief pharmacists throughout the UK in June, 188 (87 per cent) had been returned by September. Despite this high level of interest, the vast majority of hospitals (89.4 per cent) had no electronic prescribing currently, although most of this group (86 per cent) had plans for piloting, introducing and evaluating EP systems within the next five years.
Although 11 per cent of hospitals answered that they had a EP system, most did not have full electronic prescribing systems. They had discharge systems without any of the capabilities that would be expected of EP (such as decision support), or systems designed and used only in specialised areas (eg, renal or chemotherapy), or limited pilot schemes. Only 2 per cent of responders operated what would be regarded as full electronic prescribing systems. Many of the comments received on the survey forms highlighted the presence of the barriers described above in hospitals across the UK.
Outlining legal aspects of electronic prescribing, Mr STEPHEN LUTENER (head of pharmacy law, Royal Pharmaceutical Society) said that the Medicines Act did not currently allow for electronic signatures on prescriptions and this would need a change in legislation. However, the Act allowed hospitals to administer medicines under the direction of a doctor rather than by means of a formal prescription and this was likely to allow electronic prescribing in secondary care.