Fatal prescription charts

A junior doctor starts a new short‐term contract in an unfamiliar hospital and attends the half‐day training on the tailor‐made electronic prescription chart. During a long and busy shift a prescription for morphine is required. Listed on the database under morphine, the first of many options, is a 200 mg modified release tablet. Under the illusion that the top of a list is a safe place to start, a single dose is selected and given by a junior nurse on a short staffed ward who assumes that the patient has built up to this dose. The patient dies of the overdose. Apart from the preventable death of the patient, the prescription is fatal for the prescriber's career and manslaughter charges are considered. Clicks on drop down menus, more prone to mistakes than written or typed instructions, can select the wrong dose or the wrong product within a fraction of a second. This can end a life and have life‐time consequences for others. System alerts can flag potential errors, but these are so common that alert or alarm fatigue jeopardises patient safety 1.

The aviation industry knows well that there are multiple safety considerations when a pilot flies in a new cockpit. Standardized training, manuals and practice with simulators make air travel safe. General flying training is analogous to teaching clinical pharmacology. But multiple designs of cockpits or prescription charts can lead to unnecessary errors, however good the background knowledge is. Though the NHS Chief Pharmacist has not supported UK standardization, research from Australia showed benefits 2. The variability of the available systems with a lack of minimum standards is a problem. The US Food and Drug Administration (FDA) (Office of Surveillance and Epidemiology, Center for Drug Evaluation and Research) estimates some 7000 US deaths per year are caused by medication errors, with 11% of these directly related to confusion with names, dosages or abbreviations 3. The FDA identified the problems with Computerized Prescriber Order Entry Medications to make appropriate recommendations that are directly relevant to the UK 3.

Quality control and safety standardization are not unknown in medicine, though less common. Errors in laboratory testing are reduced by training, monitoring and investigation. Cases where an inborn error of metabolism was missed, or the wrong information from genetic testing given, stimulated procedural reviews hastened by the information revealed when courts award damages. The need to standardize the safety of computer programs is illustrated by radiotherapy; the medical physics community lacks quality guidelines for software and mistakes in radiation dose can occur 4.

Standards in UK laboratories are monitored with quality control samples supervised by an international independent charitable body, United Kingdom National External Quality Assessment Service (UK NEQAS). They aim to ensure tests are comparable, safe and clinically useful to the patient wherever or whenever they are performed. UK NEQAS reports to the UK Accreditation Service (UKAS), which in turn reports to NHS England under the supervision of the Department of Health (DoH).

Since 1994 ERNDIM (the European Research Network for evaluation and improvement of screening, Diagnosis and treatment of Inherited disorders of Metabolism) has developed a consensus between European Biochemical Genetics Centres on reliable and standardized procedures. ERNDIM has expanded diagnostic proficiency testing to cover most of Europe for the diagnosis, treatment and monitoring of inherited metabolic diseases. It established a European directory of laboratories and promotes education. Since 2005 ERNDIM has been a partner in the Eurogentest project, necessary when some 20 000 laboratories provided genetic tests cheaply without external quality control. Now there are only some 180 European qualified laboratories for this service. When a mistake is identified, a critical incidence report is submitted to ERNDIM. If the laboratory does not give a satisfactory response within a deadline, it loses its certificate of competence and is not allowed to operate.

Large computer projects often disappoint. The eye watering $34 billion spent on US electronic health records, together with substantial other investment over eight years, generated a substandard system 5. In response, a new US law requires health IT products to have a certified Application Programming Interface (API). It is hoped that this will generate a uniform healthcare API to produce economies of scale from the current plethora of systems. Despite common national goals, the opportunity to develop a single system is lost by the creativity of Balkanized, disparate and expensive programs. Improving care and reducing medication errors depends on both the system adopted and how it is implemented locally 6. Standardization is the key in that if a substandard system is adopted nationally, then the effort to correct this can be focused on just one system, rather than the multiple systems that can exist in even a single hospital.

So how far have we got with prescribing safety? Attempts to standardize the UK paper prescription chart failed. Although one can argue about the best format this should take, there is no disadvantage to having one chart for the whole UK; it could always be modified once adopted. Uniformity foundered in the debates between the Royal Colleges, Royal Pharmaceutical Society of Great Britain and the regional views from England, Scotland and Wales. The main lesson was that no single person or organization was responsible for harmonization, a fault common to bad quality assurance 7. The NHS ducked the issue. Since then the expansion in the numbers and variety of electronic prescription charts is remarkable, similar to the explosion in cheap unsupervised genetic testing before regulation. The waste of resources is phenomenal; each NHS Trust reinvents the wheel with expensive contractors selling new software to handle common national problems. Without central control, commercial interests own copyright of the multiple computerized charts. The poor prescriber who changes hospitals spends hours learning each system. But the biggest loser is the patient. A chart is like a gun; the pen or keyboard the live ammunition. Such things should not be in the hands of inadequately trained operators without adequate safety supervision.

In order to save lives, time and money, the NHS needs to get a grip and behave like the national system that it is. At present UK hospitals can have a different electronic prescribing system in their Emergency Department, Intensive Care and general wards, each of which cannot communicate either with each other or the hospital pharmacy dispensing system. ​​If within‐hospital systems cannot communicate, imagine the problem of transferring data to new systems. The original Doomsday Book remains legible after 1000 years, though the first electronic version was unreadable after a decade. The European Falsified Medicines Directive 2011/62/EU comes into force in February 2019 and will require manufacturers to contribute to a computer‐based verification of the authenticity of a medicine when dispensed. Such a barcode‐based formulary might contribute to prescription standardization when classifying medicinal products.

If local NHS computer programs are so different that a single prescribing system is not possible, then each variant of the electronic chart needs to be certified with a clear line of responsibility to the DoH, backed by legislation. As a minimum, the layout of the chart, the drug names and the alert systems should be standardized. If the DoH remains complacent, then more damages will be paid as liability cases based on software faults increase 8.

National standardization is the ideal. Both Australia and New Zealand have national prescription charts, Australian Health Ministers required a common inpatient chart for all public hospitals by 2006 to reduce the risk of prescribing and administering error 9. New Zealand health authorities recognized standardization as a safety issue in many industries and that prescribers often move between hospitals 10. Their standard paper chart led to standardization of electronic prescribing 11. The inability of the NHS to follow suit is expensive, a drain on staff time and potentially fatal for patients. It is simple to review the experience of the national systems of Australia and New Zealand and the work done by the FDA. This should be implemented nationally in the UK.

Competing Interests

There are no competing interests to declare.

I thank the many colleagues who advised on the content of this article.

Warren, J. B. (2018) Fatal prescription charts. Br J Clin Pharmacol, 84: 417–418. doi: 10.1111/bcp.13473.