This page contains a Flash digital edition of a book.

independently set for each infusion.) Despite common regimens, then, the occurrence of a wide variety of rates and VTBIs cannot be easily explained.

Discussion There are several factors that might cause interruption to an infusion and these require the attention of the nurse or other user. A well-designed infusion pump should help the user facilitate infusion-related tasks. The results shown in this paper can therefore be used to inform and help improve the redesign of a pump and to enhance its safety, in particular to prioritise areas of high use or other concerns.

The analyses of event logs could help understand how devices are used in practice, but unfortunately there is no standard regarding what information should be stored in device logs or the format in which it should be stored. In the same way, not all hospitals have a process of downloading and analysing log fi les regularly as part of their normal practice. They are often only downloaded and analysed when an incident occurs, and the 200-event limit may then be found to be problematic.

As the event times are logged to the nearest minute and without any keystroke data, it is impossible to recreate the exact sequence of actions performed by the user that result in an event. Instead, a higher- level picture of the result of a user action is obtained from the analysis. The lack of low-level interaction logging information can be a problem in incident investigations. As suggested in Thimbleby (2010):

‘Device logs should allow investigators to distinguish between different possible causes of error.’

Furthermore, the infusion pumps analysed did not have the capability of recording the name of the medication being infused or the details of the patient or person setting up the pump. We were unable to identify the most common infusion type. If this information had been available, it would have been possible to include pre-programmed settings in the infusion pump, such as a database of medications from which the user could select pre-defi ned infusion settings. Some smart infusion pumps have DERS installed and have automatic safety limits for medication selected using their drug libraries (Steingass, 2004; Trbovich et al, 2009; Quinn, 2011). Assuming these pumps are used as intended, their log data would be much richer. Analysing logs in the context of such smart infusion systems would open up opportunities to obtain better contextual information on the medications involved in therapies.

The present study provides the estimated use of an infusion pump per department in a particular hospital. These data could be used to create a maintenance program for them. In reality, infusion times vary and the results show that the infusion durations vary across departments. This variance in pump use should be taken into account when planning a maintenance programme.

The research could be expanded to look into other hospitals and produce general guidance or recommendations for maintenance programmes for these devices within the healthcare setting.

Devices need to be able to deliver infusion therapies uninterrupted over a number of hours at a time. When an infusion needs to be delivered using only the energy from the battery, then the study implies that pumps should be able to run on battery for at least 5.5 hours in order to be feasible for use in all departments.

The study also shows that there are combinations of VTBI and infusion rate that are most commonly used, while there are others that are not used at all. This observation could help designers and software developers create interfaces where the most common combinations are more easily set, reducing the number of button presses and risk of error as well as the time needed to set the infusion.

It is also worth noting that there are only a few cases of infusions that use the combination of VTBI and rate in which values are higher than 500 ml or 500 ml per hour. Such information can be used to create a safety feature in the infusion pump software. Even though the use of values such as 600 ml and rate of 600 ml per hour are within the range, the combination of these values to set an infusion are not common, so users might usefully be prompted to double-check the validity of such settings.

Conclusions and future work This study reports quantitative analysis of infusions that occurred in a hospital over a period of 3 years. It provides a quantitative reference point for infusion-related activities. Typical average values for information, such as duration of infusion, number of times infusions are restarted and number of infusions per department, are presented. Nearly half of all infusions were interrupted at least once.

To improve the quality of event logs and the accuracy of what can be inferred from the analysis of logs, a more precise timestamp for activities registered in the

BJN July 2013 CareFusion Supplement 19

Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28