When comparing the agreement strength of infusion pump alarm perceptions of AACN nurses versus the survey results of the 3 general device surveys, there seems to be a meaningful response variability (Table 3). While most AACN nurses agreed that pump nuisance alarms disrupt patient care, HTF 2011 survey respondents believed this is less of an issue (91% vs 71%). Similarly, nuisance alarms occurring frequently appears to be less of an issue for HTF 2011 (77%) versus AACN survey respondents (87%). While CCN 2017 (100%) and TCICU 2015 (98%) survey results indicated that nuisance alarms reduced trust in alarms, fewer AACN nurses (83%) thought this would cause staff to inappropriately silence alarms. The survey responses pertaining to staff sensitivity and response time for alarms indicated meaningful differences, with HTF 2011 survey responses 29% higher (66%) and TCICU 2015 survey responses 38% lower (34%) than the AACN survey result of 47%. The most profound variation in survey results was related to instances when alarms cannot be heard and are missed. AACN nurse responses were 97% to 137% higher (69%) for infusion pumps specifically, as opposed to more general alarm perceptions reported in the CCN 2017 (35%), TCICU 2015 (32%), and HTF 2011 (29%) surveys.
When AACN nurses were asked to rank 5 infusion pump alarm-related issues from 1 (most important) to 5 (least important), the mean rankings ranged from 2.26 to 2.86 (Table 4). Frequent false alarms leading to reduced attention or response was the most important issue (2.26), followed by difficulty in identifying source of alarm (2.69). Again, there was variation in the rankings of reported alarm-related issues when comparing the AACN survey with the CCN 2017, TCICU 2015, and HTF 2011 survey results. AACN and HTF 2011 nurses both ranked frequent false alarms leading to reduced attention or response the number 1 issue, while the CCN 2017 and TCICU 2015 surveys ranked it fourth. While inadequate staff to respond to alarms as they occur was the number 1 issue for CCN 2017 nurses, it ranked fourth for AACN nurses, and fifth for both TCICU 2015 and HTF 2011 respondents. Similar response variability was reported regarding difficulty understanding the priority of an alarm, with CCN 2017 and TCICU 2015 nurses ranking it the second most important issue. HTF 2011 and AACN survey respondents ranked it the third and fifth most important issue, respectively.
This study was a preliminary attempt to measure nurses' perceptions related to infusion pump alarms and compare them with the findings of 3 other studies that measured clinical alarms in general. The intent was to assess whether CCNs perceive infusion pump alarms differently than clinical alarms.
Based on the limited demographic data reported in the 3 comparison studies, it is unclear how similar or different the 4 study populations are. Cho et al6 found no statistically significant (P ≤ .05) difference in clinical alarm fatigue in ICU nurses when considering demographic characteristics. While demographic factors typically would be expected to have an impact on a study's results, these findings indicate that demographic factors may not play a large part in alarm perceptions.
Infusion Pump Alarm Perception Questions
The 10 pump alarm perception questions showed respectable reliability (Cronbach's α= 0.75).23 There was overwhelming agreement that infusion pump alarms disrupt patient care (91%) and occur frequently (87%). Considering that most infusion pump alarms interrupt infusion therapy and require direct action by the clinician, it is not surprising that nurses perceive pump alarms to disrupt patient care. This frequency perception is supported by a recent infusion pump alarm study that reported an average of 159 pump alarms every 24 hours on a 16-bed critical care unit.3 Possible solutions to help augment the management of pump alarms might be the use of nursing unit central monitoring systems; intelligent alarm systems that automatically change settings and suppress alarms; alarm delays; escalation; and distributed alarm systems, which forward critical infusion alarms to mobile devices.9,24 It is encouraging that 85% of the nurses agreed that using unit monitors to visually display pump alarms could potentially be useful in improving alarm recognition and management. To date, however, there is no outcome research on the use of central monitoring specifically for infusion pump alarms, and forwarding of clinical alarms to cell phones has had mixed results.24,25 Vockley25 suggested that pushing pump alarms to cell phones exacerbated alarm fatigue, if alarms were not prioritized and selectively forwarded based on patient- and infusion-specific parameters.
Despite infusion pump companies' efforts to improve alarm recognition by having single-channel pumps, distinctive alarm tones, differentiating light signals, and real-time alarm status dashboards, 59% of the nurses agreed that when multiple pumps are being used on a patient, it can be confusing to determine which is alarming. In an emergent event when every second counts, confusion over which pump is alarming can be a matter of life and death for the critically ill patient. This may be an area that could benefit from hospital and infusion pump company partnerships to brainstorm, test, and implement creative solutions to improve pump alarm identification, escalation, and management.
Nurses' Perceptions of Infusion Pump Alarms and Clinical Alarms
The results of the current study showed differences between nurses' perceptions of infusion pump alarms and their perceptions of clinical alarms, in general, as described by Korniewicz et al,16 Funk et al,19 Sowan et al,20 and Petersen and Costanzo.21 The statement that elicited the most different results was “there have been frequent instances where alarms could not be heard and were missed.” Thirty-five percent of the CCN 2017 nurses, 32% of the TCICU 2015 nurses, and 29% of the HTF 2011 nurses agreed with this statement when applied to clinical alarms; 69% of survey responses of AACN nurses agreed with the statement when applied specifically to infusion pump alarms. Looking at alarm volume produced by infusion pumps alone, Kurnat-Thoma and Shah3 reported that the decibel sound level readings for 1 to 3 pumps at a room doorway ranged from 58 to 75 dB, while the overall noise in patient care areas was a mean of 56 dB with a peak of 76 dB in the progressive care unit. Thus, pump alarms may not be audible over general unit noise. However, Tegnestedt et al4 found that monitor, ventilator, dialysis machine, and infusion pump alarms were all similar in volume, measuring between 82 and 85 dB. If these devices have no significant difference in actual alarm volumes in the clinical setting, then perhaps the nurses' perceptions of pump alarms not being heard and often missed are related to perceived criticality of these alarms. Specifically, physiological monitors and ventilators might be perceived as higher priority or more urgent than pump alarms (eg, a fatal arrhythmia or obstructed airway alarm vs an infusion complete alarm), and, therefore, nurses may be less critically “in tuned” to pump alarms and more likely to respond to them more slowly. Ultimately, the hospital environment has more device noise today than it did when the WHO recommendations2 were published. It will be necessary to reevaluate and determine optimal and realistic device alarm volumes and tones and the ability to differentiate different devices and levels of acuity. The AAMI Foundation is helping to address this through alarm design research based on auditory perception and cognition.26 The researchers are identifying alarm acoustics that are easier to learn and localize, which will influence future national standards on medical device alarms.
With regard to the ranking of issues that affect response to infusion pump alarms, nurses clearly identified frequent false alarms leading to reduced attention or response as the most important issue. In infusion pumps, nuisance alarms can take the form of frequent air-in-line alarms as a result of microbubbles; occlusion alarms as a result of kinked administration sets or positional intravenous catheters; hold, expired, or inactivity alarms; and battery alarms. Evidence suggests that increased training and education can help reduce the number of nuisance alarms.3,7,21,27 In a study involving a pediatric ICU, Manrique-Rodríguez et al27 found that providing user training and support tools, as well as continuously monitoring results, decreased the number of unnecessary pump alarms and that alarms that sounded were taken more seriously. Kurnat-Thoma and Shah3 characterized pump alarms and associated nurse perceptions across 6 care units, concluding that there was a need to improve staff education specific to clustering tasks around medical device alarms. Nurses have identified that lack of training contributes to the problem of alarm management, suggesting that increased training using real clinical scenarios with a focus on which alarms are nonactionable and how to set patient specific alarms may be useful.7,21
This study has several limitations that should be considered. The sample was one of convenience and, for the most part, represented a self-selected group of CCNs. Approximately 80% of the survey respondents were attending a presentation on pump alarms. Of these, approximately 60% completed the survey before the presentation; 40% completed it following the presentation. Because the presentation included content on types and frequency of alarms, completing the survey after the presentation could have skewed respondents' perceptions and the survey results. While a randomized sample may produce similar results, the authors cannot assume that the present study results would apply to any other sample. Psychometric evidence to support the infusion pump-specific adapted HTF 2011 survey is limited, and ideally, the survey should undergo additional validity and reliability testing to improve confidence in the accuracy of the survey findings.
The use of the terms nuisance and false across all the surveys is another limitation, because the terms have not been defined and validated, and nurses taking the surveys may have different interpretations of them. For example, the Merriam-Webster dictionary28 defines nuisance as “annoying or unpleasant,” a false alarm as “an alarm sound that occurs when no valid triggering event has taken place,” and the terms nuisance, false, false positive, clinically insignificant, and nonactionable are often used interchangeably without clear definition.7,9,10 The first step to measure and understand perceptions around clinical alarms is to define and use consistent terminology. The AAMI Foundation is creating a standardized taxonomy regarding alarms, and some researchers are avoiding the use of nuisance, false, and true alarms because of their various interpretations.7,8
Furthermore, the classification of alarms may not apply to all equipment-related clinical alarms. For example, 1 classification is actionable, requiring intervention, as opposed to nonactionable, having no clinical relevance or requiring no clinical intervention.9 Because most infusion pump alarms cause an interruption in the therapy and/or require clinical intervention, this type of classification may be less applicable to infusion pumps. A more meaningful way to classify pump alarms might be based on those that do not interrupt the infusion and those that do (eg, delivery interruption alarms). It also has been suggested that alarms be delineated based on level of priority (eg, high = urgent, medium = quick response, low = attention needed).10 But how is an urgent versus a quick response operationalized? With infusion pumps, it's likely that the specific infusion and the patient's condition will determine the priority of addressing a delivery interruption alarm. For example, a bag-empty alarm for a stable patient on maintenance fluids would be a low priority, while a bag-empty alarm for a critically ill patient on epinephrine would require an urgent response. It will be important to define and classify critical pump alarms so that meaningful interventions and/or technologies can be implemented and evaluated. Possible technology solutions include (1) providing nurses with the ability to modify priority alarms, alarm delays, and escalations based on the patient and the infusion; and (2) providing real-time infusion alarm dashboards to prioritize alarm responses and avoid preventable alarms, such as bag-empty alarms.
Results of this study offer only a glimpse into nurse perceptions related to infusion pump alarms and by no means offer a complete picture. To date, no research has quantified alarm fatigue in nurses or connected alarm perceptions to actual measurement of nurse fatigue. Furthermore, there are only limited data on the frequency, duration, and other characteristics of infusion pump alarms,3,29–31 and early research suggests that pump alarms may represent only a small portion of the clinical alarm issue, with ICU pumps being in an alarm state as little as 0.3% of total infusion time.31 A benchmarking measurement standard for pump alarms should be established, and future work needs to collect data on infusion pump alarms across different pump types, models, and manufacturers. It is also hoped that further research, using robust sampling techniques and instrumentation, will be able to add more detail to the infusion pump alarm story and offer insights into modifiable factors that potentially will be able to reduce the infusion pump alarm burden on nurses and patients without compromising patient safety.
Infusion pump alarms are problematic for CCNs, with the bulk of the problem stemming from a perceived high frequency of nuisance alarms that are believed to disrupt patient care. Compared with nurse perceptions of clinical alarms, perceptions of infusion pump alarms appear to be different. Based on these results, it may not be appropriate to apply data and recommendations about clinical alarms, in general, to infusion pump alarms. Future studies employing validated and reliable data collection tools and rigorous sampling techniques are needed to better understand nurse perceptions of infusion pump alarms and the impact of those perceptions on patient care. Once a better understanding of verifiable and repeatable infusion pump alarm issues is obtained, appropriate corrective solutions related to relevant and associated noise can be designed.
Health care industry collaboration is encouraged to establish pump alarm taxonomy, measurement standards, and benchmark data and to conduct necessary research to help identify creative solutions to improve pump alarm management. Potential solutions may include identifying non–decibel noise-producing alternatives to alert nurses to a patient's alarming pump; employing real-time dashboards, alarm prioritization, and patient-specific alarm customization technologies; and educating nurses on specific practices and pump configurations to help reduce unnecessary alarms. Recognizing the sampling and instrumentation shortfalls of the current study, it is recommended that further studies be conducted using more generalizable sampling techniques and psychometric evidence to support instrumentation.
1. Ryan KM, Gagnon M, Hanna T, Mellow B. Noise pollution: do we need a solution? An analysis of noise in a cardiac care unit. Prehosp Disaster Med. 2016;31(4):432–435.
2. Berglund B, Lindval T, Schwela DH; World Health Organization. Occupational and Environmental Health Team. Guidelines for Community Noise. Geneva, Switzerland: World Health Organization; 1999. http://www.who.int/iris/handle/10665/66217
. Accessed June 6, 2017.
3. Kurnat-Thoma E, Shah K. A community hospital's evaluation of alarm management safety factors. J Nurs Adm. 2016;46(12):675–682.
4. Tegnestedt C, Günther A, Reichard A, et al Levels and sources of sound in the intensive care unit—an observational study of three room types. Acta Anaesthesiol Scand. 2013;57(8):1041–1050.
6. Cho OM, Kim H, Lee YW, Cho I. Clinical alarms
in intensive care units: perceived obstacles of alarm management and alarm fatigue
in nurses. Healthc Inform Res. 2016;22(1):46–53.
8. Bonafide CP, Lin R, Zander M, et al Association between exposure to nonactionable physiologic monitor alarms and response time in a children's hospital. J Hosp Med. 2015;10(6):345–351.
9. Cosper P, Zellinger M, Enebo A, Jacques S, Razzano L, Flack MN. Improving clinical alarm management: guidance and strategies. Biomed Instrum Technol. 2017;51(2):109–115.
10. Cvach M. Monitor alarm fatigue
: an integrative review. Biomed Instrum Technol. 2012;46(4):268–277.
16. Korniewicz DM, Clark T, David Y. A national online survey on the effectiveness of clinical alarms
. Am J Crit Care. 2008;17(1):36–41.
17. American College of Clinical Engineering (ACCE) Healthcare Technology Foundation. Impact of Clinical Alarms
on Patient Safety. Plymouth Meeting, PA: ACCE Healthcare Technology Foundation; 2006. http://www.thehtf.org/documents/White%20Paper.pdf
. Accessed March 9, 2018.
18. American College of Clinical Engineering (ACCE) Healthcare Technology Foundation. 2011 National Clinical Alarms
Survey: Perceptions, Issues, Improvements, and Priorities of Healthcare Professionals. Plymouth Meeting, PA: ACCE Healthcare Technology Foundation; 2011. http://www.thehtf.org/documents/2011_HTFAlarmsSurveyOverallResults.pdf
. Accessed March 9, 2018.
19. Funk M, Clark JT, Bauld TJ, Ott JC, Coss P. Attitudes and practices related to clinical alarms
. Am J Crit Care. 2014;23(3):e9–e18.
20. Sowan AK, Tarriela AF, Gomez TM, Reed CC, Rapp KM. Nurses' perceptions and practices toward clinical alarms
in a transplant cardiac intensive care unit: exploring key issues leading to alarm fatigue
. JMIR Hum Factors. 2015;2(1):e3. doi: 10.2196/humanfactors.4196.
21. Petersen EM, Costanzo CL. Assessment of clinical alarms
influencing nurses' perceptions of alarm fatigue
. Dimens Crit Care Nurs. 2017;36(1):36–44.
22. Sowan AK, Gomez TM, Tarriela AF, Reed CC, Paper BM. Changes in default alarm settings and standard in-service are insufficient to improve alarm fatigue
in an intensive care unit: a pilot project. JMIR Hum Factors. 2016;3(1):e1.
23. DeVellis RF. Scale Development: Theory and Applications. 3rd ed. Los Angeles, CA: Sage; 2012.
25. Vockley M. Passion for safety underpins healthcare system's infusion pump upgrade. Biomed Instrum Technol. 2017;51(1):46–50.
27. Manrique-Rodríguez S, Sánchez-Galindo AC, López-Herce J, et al Risks in the implementation and use of smart pumps in a pediatric intensive care unit: application of the failure mode and effects analysis. Int J Technol Assess Health Care. 2014;30(2):210–217.
29. Gorges M, Markewitz BA, Westenskow DR. Improving alarm performance in the medical intensive care unit using delays and clinical context. Anesth Analg. 2009;108(5):1546–1552.
APPENDIX Infusion Pump Alarm Survey
Keywords:© Copyright 2018 by Infusion Nurses Society
alarm fatigue; alarm survey; clinical alarms; critical care nurses; false alarms; infusion pump alarms; nonactionable alarms; nuisance alarms; nurse alarm perceptions