ECLIPSE (Exploring the Current Landscape of Intravenous Infusion Practices & Errors) was a research project funded by NIHR (2014-2017) studying medication practices with infusion devices, to document the variety of existing practices and deliver recommendations for best practice in different situations.
There are various studies of the intravenous administration of drugs in hospitals; these suggest that errors often occur in the preparation and administration of these drugs. Depending on the methods used and what is counted as an “error”, published error rates vary from 18% of doses to 173% of doses (error rates of more than 100% are sometimes quoted where studies count more than one error per dose). Many of these errors are very minor and unlikely to affect the patient. However, others can lead to patient harm.
Within the USA, “smart” infusion pumps are used in most hospitals with the aim of preventing such errors. These pumps need more information from hospital staff (such as details of the medication and its concentration, the patient’s weight, etc.); they then alert the user to unusual infusion rates based on those values. Some US research suggests that smart pumps can prevent a significant proportion of errors; other research suggests they have little effect. Smart pump use is much less common in England – only a handful of hospitals use them. There were no prior UK studies of their likely effects on medication errors, and practices in relation to prescribing and administering intravenous medication are different to those in the USA. There is therefore little information available to guide UK hospital staff in deciding how best to prevent errors involving the administration of intravenous medication, and whether or not the costs of smart pumps would be justified by any impact on error reduction.
Our study addresses these gaps in knowledge. We explored how often errors involving administration of intravenous medication occur in UK hospitals, the reasons for these errors, and how they may be linked to, or prevented by, the infusion pumps used. We have compared our results with a similar study of the types and frequency of intravenous medication administration errors in the US, and explored reasons for the differences identified.
First, we conducted a study across 16 English hospitals to find out how often errors occur in the administration of intravenous medication. In each hospital, we studied 3-5 wards (representing critical care, general medicine, general surgery, paediatrics and oncology day care). A small team of staff in each hospital collected details of the intravenous medication being administered on a particular day and compared this to the medication prescribed, to identify any errors. These included errors involving administration of the wrong drug, wrong concentration, wrong infusion rate, wrong method of administration, wrong time of administration, and so on. We also assessed the importance of any errors identified using standard methods. We analysed the data to explore whether there might be differences in error rates or types across different wards or hospitals, or for different types of infusion pump.
Second, we focused in more detail on a smaller number of selected wards and hospitals, to explore the reasons why errors in intravenous medication do or do not occur. Our researchers observed staff administering intravenous medication and setting up infusion pumps in the study wards, and interviewed staff at all levels and with a variety of roles (managers, those responsible for training and device maintenance, as well as nurses and other staff who use infusion pumps and administer intravenous medication).
Finally, we have compared our findings with those of a parallel US study. We had hoped to develop recommendations for best practice for intravenous drug medication administration in English hospitals. However, our key finding is that intravenous medication administration is a complex system, and that a large number of factors affect safety and success: it’s not as simple as just choosing a different pump .
We are reporting findings through research publications and presentations. See elsewhere in this site or contact the team.