The Pharmaceutical Journal
Vol 279 No 7473 p407
13 October 2007

This article
Reprint   Photocopy

PDF 30K, Acrobat Reader

Reducing medication errors 2007

Automate to improve patient safety

Pharmacists need to embrace new technologies in order to reduce medication errors, Laurence Goldberg, independent pharmaceutical consultant and former non-executive director of the National Patient Safety Agency, told participants.

Mr Goldberg highlighted that adverse drug events cost the NHS approximately £2bn per year in hospital stays and over £400m a year in clinical negligence settlements.

“So if anybody tells us we haven’t got any money to embrace this new technology they are living in a dream world. The issue is perhaps we are not putting together a strong enough business case,” he said.

There is evidence to show that automation removes picking errors and therefore reduces dispensing errors, said Mr Goldberg. He explained that, as well as being used in hospitals, robots can also be used across the secondary/primary care interface and within primary care.

“We can put these robots in a community pharmacy linked to a GP practice,” he said. He suggested that take home and outpatient prescriptions issued electronically in hospitals could also be dispensed by robots in community pharmacies.

Mr Goldberg predicted that within the next 10 to 15 years most hospitals will have integrated systems that combine electronic prescribing with automated dispensing and bar-coded or radio frequency ID tag administration. “There is a cost to pay but trusts are now beginning to see that there are savings to be made,” he argued.

Mr Goldberg also talked about automated systems in development, including robots that compound minibags, prefilled syringes and infusions under aseptic conditions.

One participant had experience of using a decentralised automated system in a London hospital. She pointed out that when the robot breaks down it can be difficult to locate items manually. She also said that robots do not eliminate human error, such as the wrong drug being requested in the first place. This, said Mr Goldberg, would be resolved when robots become linked to e-prescribing.

Trigger drugs help identify ADEs

A list of trigger drugs to aid identification and reporting of adverse drug events in UK hospitals has been developed by pharmacists at King’s College Hospital, London.

Pharmacist Georgina Boon explained that, traditionally, adverse drug event (ADE) reporting is a reactive process and there is significant under-reporting. Triggers, such as drugs, may be indicators of possible adverse events, she said.

US research has shown that identifying triggers increases ADE reporting rates compared with traditional methods, such as notes review.

A list of drugs was adapted from a previously published list. Over six weeks, 115 trigger drugs were prescribed for 88 patients on medical and surgical wards. The indication for the drug and reason for the prescription was documented to determine whether an ADE had occurred, explained Ms Boon.

A multidisciplinary panel of three doctors and three pharmacists independently reviewed the data and decided whether each trigger drug prescribed was as a result of an ADE.

Of the 115 trigger drugs, 44 per cent were associated with an ADE in 37 patients. Of these, 12 patients experienced an ADE that could have been prevented, explained Ms Boon.

The specificity of each trigger was calculated as the frequency with which an ADE occurred for that particular drug.

Five trigger drugs showed 50 per cent or more specificity for an ADE:

• Beriplex (100 per cent)
• Naloxone (100 per cent)
• Vitamin K (77 per cent)
• Calcium Resonium (60 per cent)
• Hydroxyzine (50 per cent)

“However, only two of the ADEs that we identified were reported by staff through normal hospital reporting systems,” she added.

Gillian Cavell, deputy director of pharmacy, medication safety, explained that the trigger list was launched within the trust in October 2006.

The number of reports associated with trigger drugs over the six months following the launch more than doubled from nine to 25. However, she pointed out that, if the original work were to be extrapolated, she would have expected to see 150 reports over six months.

“Further promotion of the trigger list needs to be done in order to improve reporting rates with the aim of preventing medication-related adverse events in the future,” said Ms Cavell.

New reporting system

A medication error reporting scheme has been developed by Gerry Armitage, senior research fellow at Bradford Institute for Health Research. The reporting form is simple and focuses on using information to promote learning at the point of reporting. The scheme was piloted in Bradford Teaching Hospitals NHS Foundation Trust.

Mr Armitage looked at 1,000 reports collected over five years. He also conducted interviews with pharmacists, nurses and doctors who had reported errors. From this he developed a new reporting scheme.

Of the 1,000 forms analysed, 25 per cent did not specify any causative factors. Mr Armitage therefore decided to include on the new form a list of contributory factors, which he collated from the report analysis, interviews and literature. Tick boxes are provided to rate from 1 to 10 how important each factor was in contributing to the error.

Mr Armitage emphasised the richness of information gained from a multidisciplinary perception of error and said that the design of any reporting scheme should accommodate all health care professionals.

“Reporting will not eliminate those horrendous cases … but as part of an integrated risk management system I think it can be a help,” he concluded. Mr Armitage is currently putting together a business case to develop an electronic reporting form.