While bedside cardiac monitors and other physiological monitoring devices (e.g., continuous pulse oximetry) are designed to alert clinicians to abrupt and acute vital sign changes, we are now learning that these devices contribute to alarm fatigue.1–9 Alarm fatigue occurs when clinicians are desensitized by numerous alarms, most of which are false or clinically irrelevant. Alarm fatigue may lead to inadvertently ignoring alarms because the alarm tones are assimilated into the workflow; silenced alarms without checking the patient; lowering alarm volume; or, in extreme cases, permanently disabling the alarm. These reactions occur because the constant noise and messaging is bothersome to clinicians, their patients, and the patient's family.

Alarm fatigue in the hospital setting is now well recognized as a serious detriment to patient safety. The Association for the Advancement of Medical Instrumentation and the Food and Drug Administration have warned of deaths due to alarm silencing on patient monitor devices.10 A number of other federal agencies and national organizations have issued alerts describing alarm fatigue as a major patient safety concern. For example, the ECRI Institute named alarm fatigue as the number-one health technology hazard in their 2014 report.11 The Joint Commission (TJC) issued an alarm safety alert in 2013 and established alarm safety as a National Patient Safety Goal (NPSG) by issuing NPSG.06.01.01 in 2014.12 TJC established Jan. 1, 2016 as the date when hospitals must establish an alarms management strategy to maintain their accreditation.12 The University of California San Francisco (UCSF) Medical Center created a clinical alarms management committee in May 2014 to address this important clinical issue as well as meet NPSG.06.01.01. In this article, we will describe in detail the process undertaken at the UCSF Medical Center.

Setting

The University of California, San Francisco (UCSF) Medical Center is an academic medical center that provides adult, neonatal, and pediatric care management for critical, acute, and intermediate cases over a wide range of specialties. Three UCSF campuses participated in the clinical alarms management effort: 1) Parnassus: a 590-bed hospital focused primarily on adult services. 2) Mission Bay: includes 183 beds for pediatric specialties; 70 adult beds for patients with orthopedic, urologic, gynecologic, head/neck, gastrointestinal, and colorectal cancers; and a 36-bed birthing center. 3) Mount Zion: provides outpatient surgical services.

Committee Makeup

Prior to the creation of the Clinical Alarms Management (CALM) Committee in May 2014, responsibility for alarm management was decentralized, and individual departments, units, or specialty committees made their own alarm management decisions. While alarm management questions were often directed to individuals with expertise in using equipment with alarms (e.g., clinical nurse specialists, respiratory therapists, physicians, educators), issues were solved at the unit or department level only. The hospital system lacked standardization.

One of the first steps in UCSF's alarm management process was assigning a lead to coordinate the activities of the CALM Committee. The committee selected the patient safety manager, a masters-prepared nurse responsible for patient safety for all three campuses, to link together this diverse committee, which included clinical leaders, administrative leaders, and clinical staff. This was a critical piece to creating and sustaining change within the organization. The patient safety manager maintains a broad clinical perspective within organization, which helped him identify and communicate with key clinicians and drive the initiatives of the CALM Committee.

Alarm management decisions affect many areas of the hospital, including the emergency department, acute and critical care, and procedural areas. Issues may differ depending on the setting, purpose, and age of the patient. Therefore, identifying key stakeholders in each of these areas was a critical early step. Representatives participated in the CALM Committee from several disciplines and departments, including nursing, medicine, clinical engineering, information technology, risk management, respiratory therapy, and materials management. The committee also included two faculty members from the UCSF School of Nursing. They brought research and clinical expertise in bedside cardiac monitoring as well as biomedical engineering expertise in collecting and analyzing physiologic data.

Timeline

In this article, we describe the work of the CALM Committee during a 24-month period beginning in May 2014. The committee met monthly using a web-based conference calling system, which ensured all participants could join the meetings and easily share documents and PowerPoint presentations. Figure 1 illustrates the major stages of the CALM Committee's activities.

Figure 1.

The major stages and initiatives used by the Clinical Alarms Management (CALM) Committee.

Figure 1.

The major stages and initiatives used by the Clinical Alarms Management (CALM) Committee.

Creating the Team and Developing a Strategic Plan

The overarching strategic plan developed by the CALM Committee was to examine, understand, and improve clinical alarm management at the medical center (adult and pediatric) and to meet TJC's NPSG.06.01.01 (Figure 2). The committee developed a strategic work plan to meet these goals. The initial step included collecting data to understand how alarms were being managed within the entire hospital system (adult and pediatric) and across settings (i.e., intensive care, intermediate care, medical/surgical, emergency department, operating room, radiology). The committee's broad membership enhanced the efficiency in collecting these data. The committee also reviewed current policies related to alarm management from all of the clinical areas in order to understand both consistencies and inconsistencies in practices. During this stage, the committee conducted a gap analysis by reviewing incident reports and sentinel events. The goal was to understand patient safety issues where alarms or alarm management were a central issue. The final step in this stage included a focused work plan to drive the strategic plan, as well as the development of action plans and initiatives.

Figure 2.

Strategic plan developed by the Clinical Alarms Management (CALM) Committee. *Joint Commission requirement.

Figure 2.

Strategic plan developed by the Clinical Alarms Management (CALM) Committee. *Joint Commission requirement.

Alarm Inventory

In stage two of the CALM Committee's work, the group conducted a systemwide inventory of machines with alarms (including an asset count) and conducted an alarm risk assessment. Individual committee members scored each piece of equipment, followed by committee consensus. The risk variables they examined included: potential for harm, clinical oversight required, current clinical oversight, use frequency per patient during hospitalization, and urgency. Each variable received a score on a scale of one (lowest) to three (highest). The scores were summed for each piece of equipment or device to produce an overall risk assessment score. The committee used risk assessment scores, asset counts, and clinical judgment to prioritize alarms associated with bedside monitoring (e.g., electrocardiogram and pulse oximetry) as the committee's initial focus, followed by infusion pumps and ventilators. The alarm inventory is represented in Table 1.

The committee's broad membership enhanced the efficiency in collecting these data.

Table 1.

Systemwide inventory of machines/devices with alarms and asset count. The committee conducted an alarm risk assessment. For each piece of equipment, risk variables were scored by individual committee members, followed by committee consensus. The risk variables collected included potential for harm, required clinical oversight, current clinical oversight, use frequency per patient stay, and urgency. Each variable was scored on a scale from 1 (lowest) to 3 (highest). The scores for each piece of equipment or device were then summed. The risk assessment scores, asset counts, and clinical judgment were used to prioritize alarms associated with bedside monitoring (ECG and pulse oximetry) as the committee's initial focus, followed by infusion pumps and ventilators.

Systemwide inventory of machines/devices with alarms and asset count. The committee conducted an alarm risk assessment. For each piece of equipment, risk variables were scored by individual committee members, followed by committee consensus. The risk variables collected included potential for harm, required clinical oversight, current clinical oversight, use frequency per patient stay, and urgency. Each variable was scored on a scale from 1 (lowest) to 3 (highest). The scores for each piece of equipment or device were then summed. The risk assessment scores, asset counts, and clinical judgment were used to prioritize alarms associated with bedside monitoring (ECG and pulse oximetry) as the committee's initial focus, followed by infusion pumps and ventilators.
Systemwide inventory of machines/devices with alarms and asset count. The committee conducted an alarm risk assessment. For each piece of equipment, risk variables were scored by individual committee members, followed by committee consensus. The risk variables collected included potential for harm, required clinical oversight, current clinical oversight, use frequency per patient stay, and urgency. Each variable was scored on a scale from 1 (lowest) to 3 (highest). The scores for each piece of equipment or device were then summed. The risk assessment scores, asset counts, and clinical judgment were used to prioritize alarms associated with bedside monitoring (ECG and pulse oximetry) as the committee's initial focus, followed by infusion pumps and ventilators.

Metrics and Defaults

The CALM Committee collected hospital-level data, as well as determined how to obtain alarm metrics (i.e., number, type, level) and prepare data reports. The Parnassus campus hospital captured alarm data using a sophisticated research-based infrastructure created by School of Nursing researchers.4  Figure 3 demonstrates the research infrastructure and provides a sample of the data available for analysis.

Figure 3.

Illustrated in A is the research infrastructure used to capture alarm data from the bedside electrocardiographic monitor. Figure B shows an example of alarm data available. Displayed is the total number of alarms by day for several parameters, and the level (i.e., crisis, warning, advisory, message/unknown).

Figure 3.

Illustrated in A is the research infrastructure used to capture alarm data from the bedside electrocardiographic monitor. Figure B shows an example of alarm data available. Displayed is the total number of alarms by day for several parameters, and the level (i.e., crisis, warning, advisory, message/unknown).

Our newest hospital, the Mission Bay campus, shared a similar system as Parnassus for capturing alarm data. However, Mission Bay used a new bedside monitoring system, which included an integrated system where GE Monitor alarms (CareScape B×50 Monitor) pass through BedMaster software to a Connexall middleware system, then to a Voalte phone. The committee's goal was to understand how alarms moved within this sophisticated system, and determine the appropriate alarms to send to the nurse's Voalte phones via this complex system. Figure 4 illustrates data available for analysis within this system.

Figure 4.

An alarm report generated from the bedside electrocardiographic monitor in a pediatric unit. Shown are alarm totals (counts) for several alarm parameters during a 1-month period.

Figure 4.

An alarm report generated from the bedside electrocardiographic monitor in a pediatric unit. Shown are alarm totals (counts) for several alarm parameters during a 1-month period.

The last step completed in this stage was obtaining alarm defaults (i.e., on/off, parameters) and alarm levels (i.e., crisis, warning, advisory, message) for devices within the hospital. Because we used different physiologic monitoring systems (i.e., GE, Philips, Masimo), we also determined definitions used for dysrhythmia alarms to gain a better understanding of possible variations. The goal of this step was to determine if we could standardize alarm defaults and alarm levels with the goal of minimizing unnecessary alarms while ensuring patient safety. Figure 5 illustrates an example of the data collection tool we used to collect and compare these data. The committee also used research data collected by the School of Nursing faculty to better understand common nonactionable alarms to determine if adjustments could be recommended. From examining all of these data sources, the committee was able to make policy recommendations regarding default settings, and alarm levels (i.e., crisis, warning, advisory, message).

Figure 5.

In A is the data collection tool used to compare alarm defaults (i.e., on/off, parameters), and definitions by manufacturer (GE or Philips). In B are illustrated alarm levels (i.e., crisis, warning, advisory, message).

Figure 5.

In A is the data collection tool used to compare alarm defaults (i.e., on/off, parameters), and definitions by manufacturer (GE or Philips). In B are illustrated alarm levels (i.e., crisis, warning, advisory, message).

Dysrhythmia and SpO2 settings (i.e., parameter and level [low, medium, high]) were changed in the adult intensive care units based on research data.4 A repeat analysis of dysrhythmia alarms over the course of one month did not reveal new opportunities for making adjustments in dysrhythmia alarms for adult patients, and there were no untoward patient outcomes. Monitoring data for both dysrhythmia and pulse oximetry in pediatric patients revealed a number of opportunities for reducing alarm fatigue, particularly from alarms sent to the nurse's Voalte phone. Figure 6 illustrates the reduction in the number of alarm sent to the Voalte phone following this intervention.

Figure 6.

The number of alarm sent to a nurse's Voalte phone are reduced following an intervention consisting of eliminating nonactionable alarms and adding time delays. The intervention allowed the primary nurse more time to accept an alarm on the phone and additional time to cancel the alarm at the bedside prior to the alarm escalating to the backup nurse's Voalte phone.

Figure 6.

The number of alarm sent to a nurse's Voalte phone are reduced following an intervention consisting of eliminating nonactionable alarms and adding time delays. The intervention allowed the primary nurse more time to accept an alarm on the phone and additional time to cancel the alarm at the bedside prior to the alarm escalating to the backup nurse's Voalte phone.

Understanding Current Evidence and Policy Updates

The next step in CALM Committee's work was to understand evidence-based approaches to ensure our hospitals met current standards and determine areas for improvement. A literature search identified research and best practices for alarm management and gathered both databased (i.e., research articles) and non-databased information (i.e., information from manufacturers, other hospitals). Individual committee members contacted experts at other hospitals to learn their practices and attended webinars on alarm management. Committee members explored a broad range of alarm management topics, including: interventions to minimize false and nonactionable alarms; responsibilities of individual clinician (nurses, respiratory therapists, clinical engineering, providers, and monitor watchers); alarm settings; response to alarms (i.e., by clinician type [nurses versus monitor watcher], changing alarm settings); and education and training.

One example of a topic examined by the committee was skin electrode practices (i.e., type, storage, frequency of changing, packaging) for cardiac monitoring. The literature cites skin electrode management as a possible source of false alarms.13–15 The CALM Committee reviewed the literature for recommendations and compared our current practices to those cited in research studies to determine best practices. We also collected data on all of the units that utilize skin electrodes for cardiac monitoring to determine electrode type, packaging (bulk versus single packet), cost, and total number of electrodes used. Following the literature review and using data from a research study conducted within our facility we standardized electrode management and revised our alarm management policy. The committee decided to use packaging with five skin electrodes in units with a low volume of cardiac monitoring and bulk packaging in units with high use of cardiac monitoring. This ensured electrode freshness with the goal of minimizing poor signal quality.

Education and Communication

The next stage of the CALM Committee's process focused on educating and communicating new policies and procedures to all of the clinicians involved with alarm management. Because alarm management affects broad clinical specialties within the UCSF hospital system (e.g., nurses, respiratory therapists, pharmacists, clinical engineers, physicians, nurse practitioners, and monitor watchers) we conducted a gap analysis regarding educational approaches used to deliver education to nurses and other clinical staff. Table 2 lists the topics covered in the clinical alarms management policy that required education and training for clinical staff in areas affected by clinical alarms. Personnel responsible for managing alarm systems received more focused education on: how to set alarms; when alarm signals can be disabled; when alarm parameters can be changed; who in the organization has the authority to set, change, or disable alarm parameters; procedures for monitoring and responding to alarms; and procedures for checking alarm accuracy. Alarm management education is provided during initial orientation and with annual competency reviews.

Table 2.

The eight broad topics covered in the clinical alarms management policy that required education and training of all clinical staff in areas with clinical alarms.

The eight broad topics covered in the clinical alarms management policy that required education and training of all clinical staff in areas with clinical alarms.
The eight broad topics covered in the clinical alarms management policy that required education and training of all clinical staff in areas with clinical alarms.

In addition, we issued a Patient Safety Bulletin to illustrate case examples where alarm fatigue compromised patient safety. Each issue of the Patient Safety Bulletin described an actual clinical event that triggered an incident report, which led to a root-cause analysis and subsequent policy and/or process change to prevent and/or mitigate a similar future event. The publication's objective was to further organizational learning about adverse events and to encourage staff to identify and report situations that could result in an untoward patient outcome. This would address potential problems that require resolution with broad, interdisciplinary input.

Centralized Alarm Management

For the final stage of this 24-month project, the CALM Committee developed a structure and process to evaluate and improve alarm management throughout the health system. This included multiple clinical departments, radiology, and the operating room. The committee's role was to identify alarm issues, problem solve alarm management, and reach out to key individuals related to alarm management. The CALM Committee was established as the centralized governing and oversight committee for ensuring alarm safety. Topics reviewed by the committee include: appropriate defaults, alarm volume, standardization of alarms across units, (i.e., multiple intensive care units with varied clinical focus), unit type (ICU versus telemetry versus operating room, etc.), and patient type (adult, pediatric).

Summary and Future Directions

The CALM Committee continues to meet and organize activities around alarm management. The initial work of the committee was to gain an understanding of clinical alarms within our hospital system, and identify key stakeholders to participate in the development of a strategic plan around clinical alarms management. Once formed, the committee obtained alarm metrics, including an inventory of where alarms were located within the organization, the number and type of alarms, and default settings. The committee's next steps involved identifying research and evidence-based research related to clinical alarms management in order to benchmark our hospital's current policy and practice and adjust accordingly. The committee then developed and rolled out educational initiatives and developed communication strategies to reach the clinical staff regarding policy changes and competency requirements. This 24-month process positioned us to move forward with strategies and interventions to address alarm fatigue, including research16,17 and quality assurance projects aimed at reducing the high number of false and nonactionable alarms while ensuring patient safety.

References

1.
Bonafide
CP
,
Zander
M
,
Graham
,
CS
,
et al
.
Video Methods for Evaluating Physiologic Monitor Alarms and Alarm Responses
.
Biomed Instrum Technol
.
2014
;
48
(
3
):
220
30
.
2.
Cvach
M.
Monitor Alarm Fatigue: An Integrative Review
.
Biomed Instrum Technol
.
2012
;
46
(
4
):
268
77
.
3.
Cvach
M
,
Rothwell
KJ
,
Cullen
AM
,
et al
.
Effect of Altering Alarm Settings: A Randomized Controlled Study
.
Biomed Instrum Technol
.
2015
;
49
(
3
)
214
22
.
4.
Drew
BJ
,
Harris
P
,
Zègre-Hemsey
JK
,
et al
.
Insights into the Problem of Alarm Fatigue with Physiologic Monitor Devices: A Comprehensive Observational Study of Consecutive Intensive Care Unit Patients
.
PloS One
.
2014
;
9
(
10
):
e110274
.
5.
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
18
.
6.
Gross
B
,
Dahl
D
,
Nielsen
L.
Physiologic Monitoring Alarm Load on Medical/Surgical Floors of a Community Hospital
.
Biomed Instrum Technol
.
2011
;
3
(
suppl
):
29
36
.
7.
Paine
CW
,
Goel
VV
,
Ely
E
,
et al
.
Systematic Review of Physiologic Monitor Alarm Characteristics and Pragmatic Interventions to Reduce Alarm Frequency
.
J Hosp Med
.
2016
;
11
(
2
):
136
44
.
8.
Sendelbach
S
,
Funk
M.
Alarm Fatigue: A Patient Safety Concern
.
AACN Adv Crit Care
.
2013
;
24
(
4
):
378
86
.
9.
Siebig
S
,
Kuhls
S
,
Imhoff
M
,
et al
.
Intensive Care Unit Alarms—How Many Do We Need?
Crit Care Med
.
2010
;
38
(
2
):
451
6
.
10.
Association for the Advancement of Medical Instrumentation
.
A Siren Call to Action: Priority Issues from the Medical Device Alarms Summit
. .
11.
ECRI Institute
.
Top 10 Health Technology Hazards for 2014
. .
12.
The Joint Commission
.
The Joint Commission Announces 2014 National Patient Safety Goal
. .
13.
Clochesy
JM
,
Cifani
L
,
Howe
K.
Electrode Site Preparation Techniques: A Follow-Up Study
.
Heart Lung
.
1991
;
20
(
1
):
27
30
.
14.
Cvach
MM
,
Biggs
M
,
Rothwell
KJ
,
Charles-Hudson
C.
Daily Electrode Change and Effect on Cardiac Monitor Alarms: An Evidence-Based Practice Approach
.
J Nurs Care Qual
.
2013
;
28
(
3
):
265
71
.
15.
Sendelbach
S
,
Wahl
S
,
Anthony
A
,
Shotts
P.
Stop the Noise: A Quality Improvement Project to Decrease Electrocardiographic Nuisance Alarms
.
Crit Care Nurse
.
2015
;
35
(
4
):
15
22
.
16.
Bai
Y
,
Do
D
,
Ding
Q
,
et al
.
Is the Sequence of SuperAlarm Triggers More Predictive than Sequence of the Currently Utilized Patient Monitor Alarms?
IEEE Trans Biomed Eng
.
2016
Jun
30
.
[Epub ahead of print]
.
17.
Pelter
MM
,
Fidler
R
,
Hu
X.
Research: Association of Low-Amplitude QRSs with False-Positive Asystole Alarms
.
Biomed Instrum Technol
.
2016
;
50
(
5
):
329
35
.

About the Authors

Michele M. Pelter, RN, PhD, is an assistant professor at the University of California, San Francisco (UCSF) School of Nursing and director of the UCSF ECG Monitoring Research Lab. Email: michele.pelter@ucsf.edu

James Stotts, RN, MS, is a patient safety manager at the UCSF Medical Center. Email: james.stotts@ucsf.edu

Kevin Spolini, MSN, RN, is a nurse administrative supervisor at the UCSF Medical Center. Email: kevin.spolini@ucsf.edu

Julie Nguyen, PharmD, is a patient safety clinical specialist at the UCSF Medical Center. Email: julie.nguyen@ucsf.edu

Elizabeth Sin, RN, MS, is a patient care manager at the UCSF Medical Center. Email: elizabeth.sin@ucsf.edu

Xiao Hu, PhD, is an associate professor at the UCSF School of Nursing. Email: xiao.hu@ucsf.edu