Smart pump alerts: All that glitters is not gold

https://doi.org/10.1016/j.ijmedinf.2011.10.010Get rights and content

Abstract

Introduction

The implementation of smart pump technology can reduce the incidence of errors in the administration of intravenous drugs. This approach involves developing drug libraries for specific units and setting hard and soft limits for each drug. If a programming error occurs and these limits are exceeded, an alarm sounds and the infusion can be blocked. A detailed analysis of these alarms is essential in order not to bias the results in favor of a positive impact of this technology.

Purpose

To evaluate the results of the first analysis of the use of smart infusion pumps and to assess the significance and practical implications of the alarms sounded.

Methods

The study was performed by a multidisciplinary team that consisted of a clinical pharmacist, a pediatrician from the pediatric intensive care unit (PICU), and the chief nurse of the unit. A library of 108 drugs was developed over a 7-month period and introduced into 40 syringe pumps and 12 volumetric pumps (Alaris® with Plus software) before being applied in 6 of the 11 beds in the PICU. After four month's use, data were analyzed using the Guardrails® CQI v4.1 Event Reporter program.

Results

Following the first four months of implementation, compliance with the drug library was 87%. By analyzing the alerts triggered, we were able to detect problems such as the need to increase user training, readjust limits that did not correspond to clinical practice, correct errors in the editing of the drug library and including a training profile.

Conclusion

It is difficult to obtain accurate data on the true impact of this technology in the early stages of its implementation. This preliminary analysis allowed us to identify improvement measures to distinguish, in future evaluations, the alarms triggered by a real programming error from those caused by incorrect use.

Highlights

► Smart pump technology has shown to be effective in avoiding administration errors. ► The development of a practical drug library is key to the success of this technology. ► Information gathered in the first months of implementation can help to improve the quality of the drug library and to detect early usersî training needs. ► An exhaustive analysis of the alarms triggered allows to refine the technology and to distinguish alarms associated with real errors from the irrelevant ones.

Introduction

Drug therapy is associated with high morbidity and mortality, largely because of errors that can occur at different stages of the medication loop [1], [2]. Of these medication errors, 38% occur during the administration phase [3], [4], [5] and we know that only 2% of administration errors are intercepted [4]. The potential impact of medication errors depends on the route of administration, the type of drug, the dosage and the characteristics of the patient. For example, high-risk drugs or those with a narrow therapeutic margin administered intravenously to critically ill pediatric patients are highly likely to cause damage in the event of an error [6], [7], [8], [9], [10], [11], [12].

Intravenous therapy requiring careful control of flow is delivered using infusion pumps. Although these systems enhance the operational safety of high-risk medicines or those with a short half-life, they are not free from programming errors that sometimes lead to administration of much higher doses than prescribed [13], [14]. The use of smart infusion pumps, which have built-in safety software with a drug library, reduces programming errors that could lead to administration errors with varying impacts on patients [15].

Drug libraries are specifically designed for each particular unit and take into account the characteristics of their patients and the drugs that are usually prescribed according the pathologies managed. They are based on lists of drugs for which concentrations, maximum and minimum doses and infusion times are individually defined. These characteristics determine the administration rates that the hard and soft limits are based on.

If a user exceeds the defined dose for a soft limit by mistake, an alarm sounds to alert him/her that the dose or rate of administration may not be right for a particular patient, but it allows the infusion to continue. By contrast, accidentally exceeding a hard limit will sound an alarm that will lead the user to either cancel the infusion or reprogram delivery [16].

Infusions can be started either in ml/h to deliver a volume over a period of time or in dosage units for the same purpose, inside or outside the drug library. The main difference between these programs, besides starting infusions within the safety software or not, is that when using dosage units, health professionals know the amount of drug in mg that is being administered while when entering rates in ml/h, users must bear in mind the intravenous mixture concentration so that the total mg administered can be calculated (i.e. mcg/kg/min versus ml/h).

Smart pumps can avoid errors in the administration process. However, the implementation of this technology can last for at least one year. During the early stages of this process unnecessary alarms are common [17]. Progressive use of this tool makes the number of unnecessary alarms drop, thus increasing the clinical significance of the alarms generated and enabling us to interpret alarms as potential administration errors that have been avoided.

There are limited published data about the main factors causing unnecessary alarms related to smart pump technology that may bias the results in favor of the effectiveness of these systems in reducing medication errors.

A systematic analysis of the alarms sounded during the use of infusion pumps in routine practice conditions provides essential data that will enable us to refine this technology and create a useful and effective tool for intercepting errors in the management of intravenous medication.

Section snippets

Objectives

We aimed to evaluate the results of the first analysis of the use of smart infusion pumps in a pediatric intensive care unit (PICU) and to assess the clinical significance and practical implications of the alarms reported.

Material and methods

This was a descriptive study conducted in a PICU with 11 beds and approximately 500 admissions annually (35% with heart disease, including patients in the postoperative period of cardiac surgery, 30% in the postoperative period of other surgery, and the remaining 35% with medical conditions). This PICU belongs to a 310 bed women and children hospital, with 127 beds for maternity and 183 beds for children. It is a national reference centre in terms of pediatric heart transplant. The

Results

During the first four months, 50,203 infusions were initiated, and 43,731 of these were started through our drug library, giving an overall compliance with the use of drug library of 87.1%. A total of 4296 infusions (8.6%) were started in ml/h and the remaining 2176 (4.3%) in dosage units, both outside the drug library. An individual analysis of each profile showed good compliance for both cardiac therapy (21,105 infusions (87.7%)) and analgesia-sedation (20,074 infusions (89.1%)). However, in

Discussion

Our results revealed that, overall compliance with a smart infusion system based on an integrated drug library was good, which is the first goal to accomplish when implementing this technology. Of the total number of infusions initiated with infusion pumps, 87.1% were carried out through the drug library. We can consider this percentage to be high if we take into account the substantial change that the introduction of this technology has had on the daily work of staff of the unit, not only in

Conclusions

The analysis of data generated during the use of smart infusion pumps at an early stage of their implementation, allowed us to identify warnings caused by incorrect use of the pump and by setting limits that were too strict. By adjusting these limits and improving user training, we would be able to distinguish between alarms associated with a real programming error and those caused by incorrect use.

In all events, this technology has been proven to detect and intercept administration errors that

Authorsî contributions

Silvia Manrique-Rodríguez, Amelia Sánchez-Galindo, Cecilia M. Fernández-Llamazares, Jesús López-Herce, Lara Echarri-Martínez and Vicente Escudero-Vilaplana have contributed to analyze and interpret data stored in pumps during the first month of use.

Silvia Manrique-Rodríguez and Amelia Sánchez-Galindo have written the article.

Cecilia M. Fernández-Llamazares, Ángel Carrillo-Álvarez and María Sanjuro-Sáez have contributed to the revision of this article.

Conflict of interests

Authors declare that there are no conflict of interests.

Acknowledgements

The authors wish to thank the pediatric intensive care staff for their collaboration in this project, Samuel García for his help in getting and analyzing the information stored in the pumps and Dr. Bazire and Dr. Boyle for their contributions in translating this article.

Summary table

What is already known of this topic

  • Administration errors are difficult to intercept and can be costly, especially when both intravenous medications and pediatric critical patients are involved.

  • Implementation of

References (31)

  • D.W. Bates et al.

    Variability in intravenous medication practices: Implications for medication safety

    J. Comm. J. Qual. Patient Saf.

    (2005)
  • S. Manrique-Rodríguez et al.

    Bombas de infusión inteligentes: luces y sombras de una tecnología en desarrollo

    Med. Clin. (Barc)

    (2011)
  • M. Imhoff et al.

    Smart alarms from medical devices in the OR and ICU

    Best. Pract. Res. Clin. Anaesthesiol.

    (2009)
  • J.M. Aranaz et al.

    Estudio nacional sobre los efectos adversos ligados a la hospitalización (ENEAS)

    (2006)
  • D.W. Bates

    Preventing medication errors: A summary

    Am J Health Syst Pharm.

    (2007)
  • M.E. Stefl

    To err is human: Building a safer health system in 1999

    Front Health Serv. Manage.

    (2001)
  • G.A. Pepper

    Errors in drug administration by nurses

    Am. J. Health Syst. Pharm.

    (1995)
  • L.L. Leape et al.

    Systems analysis of adverse drug events. ADE prevention study group

    JAMA

    (1995)
  • I. Hatcher et al.

    An intravenous medication safety system: Preventing high-risk medication errors at the point of care

    J. Nurs. Admin.

    (2004)
  • L.C. Hadaway

    Managing i.v. therapy: “high-alert” drugs keep nurse managers ever watchful

    Nurs. Manage.

    (2000)
  • National Quality Forum

    Serious Reportable Events in Patient Safety: A National Quality Forum Consensus Report

    (2002)
  • R. Kaushal et al.

    Medication errors and adverse drug events in pediatric inpatients

    JAMA

    (2001)
  • J.E. Sullivan et al.

    Medication errors in paediatrics–the octopus evading defeat

    J. Surg. Oncol.

    (2004)
  • I.C. Wong et al.

    Incidence and nature of dosing errors in paediatric medications: a systematic review

    Drug Saf.

    (2004)
  • Institute for Safe Medication Practices [Internet] [cited October 17, 2011]. Available from:...
  • Cited by (22)

    • Usability of smart infusion pumps: A heuristic evaluation

      2022, Applied Ergonomics
      Citation Excerpt :

      Problems in a smart pump's hardware and software design affect how effectively and safely nurses can use them for infusion procedures. Prior research validates our findings about problems in smart pump design (Giuliano, 2018; Pinkney et al., 2010; Manrique-Rodríguez et al., 2012). Some design problems we identified can be addressed through re-design.

    • Implementing smart pump technology in a pediatric intensive care unit: A cost-effective approach

      2014, International Journal of Medical Informatics
      Citation Excerpt :

      Factors that limit the success of this technology are the ability to program infusions without drug library integration if adherence to the new tool is poor, and the limited technology integration found in most healthcare institutions [25–27]. To guarantee a good compliance with the drug library is the first goal to accomplish when implementing smart pump technology in order to draw conclusions about its efficacy in intercepting errors [15]. There was a high percentage of use of safety software in our study (92%), which allowed for the interception of 92 infusion-related programming errors of different types that could potentially have reached the patient out of 574,312 infusions started using the drug library.

    • The Role of "Smart" Infusion Pumps in Patient Safety

      2012, Pediatric Clinics of North America
      Citation Excerpt :

      Whether or not the technology is viewed as attractive depends on the actual cost of all implementation and ongoing maintenance. Even then, a published analysis of alerts during the first 4 months of smart pump implementation in a pediatric intensive care unit identified numerous problems that necessitated retraining of nursing staff, and changes to previously selected limits because of clinical alert irrelevance and errors in library programming.17 Readers should not underestimate the cost associated with changes to a smart pump library.

    View all citing articles on Scopus
    View full text