This site is intended for health professionals only

The trials and tribulations of electronic prescribing

DK Hughes
BSc(Hons) MSc MRPharmS
Principal Pharmacist/Clinical Services Manager

KT Farrar
MPharm MRPharms
Director of Pharmacy

AL Slee
MSc MRPharmS
Principal Pharmacist/Medicines Management & IT
Wirral Hospital NHS Trust, UK
All also Honorary Clinical Lecturers
School of Pharmacy and Chemistry
Liverpool John Moores University UK

Information for Health set targets for UK acute trusts to have achieved EPR (electronic patient record) level 3, which includes electronic prescribing, by 2005.(1) Most family doctors in the UK prescribe by computer; yet currently in the hospital service only a few sites have extensive application. Why is this?

Implementing computerised prescribing in a hospital setting is a considerable undertaking. A number of problems can limit the process, including the complex nature of hospital prescribing and administration and the cultural aspects of such a major change. It is essential that service managers engage key multidisciplinary professionals in the application.
Electronic prescribing is technically complex, and sufficient resources must be made available to provide the infrastructure for training of all grades of staff and to maintain and develop databases.

Implementation issues

Complexity of hospital-based prescribing
Almost every patient who comes into hospital will have a prescription for a medicine or diagnostic agent. Prescribing in hospital is more complex than in primary care as it often involves frequent changes of medication as the patient’s condition improves or deteriorates. It certainly involves prescription of medicines by a wide variety of routes, particularly the intravenous route, and, perhaps most importantly, requires a record of administration to be made.

A patient’s hospital journey may involve periods of treatment in different clinical areas. The safe and accurate prescribing and administration of medicines requires good communication and extensive documentation. Many specialist areas, such as intensive care or theatres, have evolved their own bespoke prescription sheet, designed to facilitate prescribing and administration of medicines in their specific area. As a result they are resistant to move to other forms of prescribing.

Cultural difficulties
Our own experience of computerised prescribing would suggest that the largest obstacle to overcome is one of culture. The decision to purchase our system was made in the late 1980s by a team of clinicians (doctors and nurses) who were looking for a clinically driven system which provided, as a byproduct, the management information required by the changing NHS contracting system. The investment made in time and effort to establish a culture within the hospital that allowed use of the computer system beyond entry and retrieval of laboratory orders was considerable. Without such extensive preparation, prescribing “online” may have remained an aspiration.

Management issues
The implementation requires extensive multidisciplinary cooperation. Within each professional group, you need strong clinical champions who, while having different professional requirements, share a common vision. Underpinning the professional staff goals it is essential that senior hospital management lead the implementation. To do this, they need to comprehend the basic ­clinical and technical needs of their professional staff and must be integrated into the clinical decision-making processes. Managers must be well informed, especially in cases of intra- or interdisciplinary disputes.

The introduction of computer systems into complex healthcare environments requires major changes to take place. There will always be resistance to such change. The major issues facing clinicians and managers in healthcare today are trying to ensure that we obtain maximal benefit from any healthcare investment and that we deliver high-quality clinical care for patients. A key feature of the Wirral process was to measure certain indicators of quality and efficiency before and after implementation. Relatively simple data capture included accurate measurements of the time taken to receive prescription orders into the central dispensary, an analysis of common prescription errors and a qualitative assessment of handwritten prescriptions. This had a number of benefits, including engaging clinical staff in the belief that implementation of computerised prescribing could provide clear benefits.

It also provided a useful baseline for the post­implementation audit, which was an essential tool in convincing the sceptics of the benefits.

There are many other factors that are important for successful implementation and maintenance. These include provision of appropriately trained ­clinical staff who are integrated into the technical support departments to maintain and develop prescribing pathways. It is essential that sufficient resources are made available to maintain accuracy and advances in therapy and drug safety. Such tasks are often neglected.(2) There is a large training requirement, which itself requires the necessary personnel infrastructure. For example, the Wirral Hospital employs 4,500 staff; some 3,800 have some specific access to the Hospital Information System. Medical staff require a minimum of six hours’ training, with at least three hours specifically for prescribing, before authorisation of passwords and sign-on codes. Robust and detailed procedures are required for security access and computer downtime.


Illegible writing
In an analysis of litigation cases involving drug therapy, the Medical Defence Union (UK) cites illegible writing by doctors as a considerable problem.(3) Writing in lower case and the use of abbreviations were often found to be the root cause of error. It is reasonable to expect that electronic prescribing will largely ­eliminate this problem.

In a pre- and postcomputerisation study undertaken locally, a total of 1,217 prescriptions from 123 patients were scrutinised in the first phase and 963 prescriptions from 141 patients in the second phase of the study. Prescriptions were evaluated for specific essential information about therapy, including:

  • Drug name.
  • Dose.
  • Route of administration.
  • Administration time.
  • Start date.
  • Prescriber’s signature.

The data was subjected to statistical analysis using the chi-square test. The computer system proved superior to handwritten prescriptions in all criteria relating to legibility and completeness (see Table 1).


Medication errors/risk reduction
Adverse drug events (ADEs) are the most common type of iatrogenic injury occurring in hospitalised patients.(4,5) The social and economic burden of injuries resulting from ADEs is enormous. Johnson and Bootman estimate the cost of drug-related morbidity and mortality in the USA due to ADEs to be more than $76.6bn annually.(6) Evidence that computer systems can prevent adverse drug events is now overwhelming,(7) and can no be longer ignored. Computerised decision support may be especially important. (8)

Formulary management
Information in a written formulary supported by retrospective feedback from pharmacists is the most common form of formulary management used in the UK. Given the nature of junior doctor rotational employment, the processes are inherently inefficient. It requires repeated education of replacement staff, whose experience in other institutions may dilute formulary measures, with some formulary changes becoming ignored. Computerised, physician order entry can successfully aid formulary management.(9) Carefully designed, locally agreed order entries can facilitate both the implementation and maintenance of formulary changes.

Many sophisticated hospital computerised prescribing systems with educational and interactive decision support are now available and are in the process of being implemented worldwide. These systems have been shown to be one of the most effective methods of improving the safety of medicines usage. Although application of such systems is an essential component of modern healthcare and will expand rapidly, their implementation is complicated by logistical, financial and cultural difficulties.

The authors would like to acknowledge the invaluable help of Samantha Lane, who was previously employed as a preregistration pharmacist/clinical pharmacist at Wirral Hospital and was instrumental in analysing the quality of prescribing pre- and postcomputerisation. We would also like to thank Sarah Moore, statistician at the Clinical Practice Research Unit, Wirral Hospital, for her ­assistance with the statistical analysis of the data.


  1. NHS. An information strategy for the modern NHS 1998–2005. London: NHS Information Authority; 1998.
  2. Bates DW, Leape LL. Pharmacies and prevention of potentially fatal drug interactions-in reply. JAMA 1996;275:1086-7.
  3. BMJ. Illegible writing kills patients. BMJ 1992;305:604.
  4. Brennan TA, Leape LL, Laird NM, et al. Incidence of adverse events and negligence in hospitalised patients. N Engl J Med 1991;324:370-6.
  5. Leape LL, Brennan TA, Laird NM, et al. The nature of adverse events in hospitalised patients. N Engl J Med 1991;324:377-84.
  6. Johnston JA, Bootman JL. Drug-related morbidity and mortality. Arch Intern Med 1995;155:1949-56.
  7. Yan Q, Hunt CA. Preventing adverse drug events (ADEs): the role of computer information systems. Drug Information J 2000;34:1247-60.
  8. Bates DW. Using information technology to reduce rates of medication errors in hospitals. BMJ 2000;320;788-91.
  9. Farrar K, Slee A. Formulary management – effective computer management systems. Pharm J 1999;262: 363-5.

American Informatics Association
Heath Informatics Journal
Healthcare Computing
NHS IM&T Electronic Library
British Journal of Healthcare Computing

Be in the know
Subscribe to Hospital Pharmacy Europe newsletter and magazine