|
|
||||||||
Corresponding author: Maria Cvach, MSN, RN, CCRN, The Johns Hopkins Hospital, 600 N. Wolfe Street, Baltimore, MD 21287 (e-mail: mcvach1{at}jhmi.edu).
| Abstract |
|---|
|
|
|---|
Purpose Excessive numbers of monitor alarms and fear that nurses have become desensitized to these alarms was the impetus for a unit-based quality improvement project.
Methods Small tests of change to improve alarm management were conducted on a medical progressive care unit. The types and frequency of monitor alarms in the unit were assessed. Nurses were trained to individualize patients alarm parameter limits and levels. Monitor software was modified to promote audibility of critical alarms.
Results Critical monitor alarms were reduced 43% from baseline data. The reduction of alarms could be attributed to adjustment of monitor alarm defaults, careful assessment and customization of monitor alarm parameter limits and levels, and implementation of an interdisciplinary monitor policy.
Discussion Although alarms are important and sometimes life-saving, they can compromise patients safety if ignored. This unit-based quality improvement initiative was beneficial as a starting point for revamping alarm management throughout the institution.
Notice to CE enrollees:A closed-book, multiple-choice examination following this article tests your understanding of the following objectives:
|
When alarm frequency is high, nurses are at risk for becoming desensitized to the alarms that are intended to protect their patients. Cardiac monitor algorithms are intentionally set for high sensitivity at the expense of specificity. As a result, numerous false alarms occur.1 A 2006 American College of Clinical Engineering survey of more than 1300 health care professionals showed that a large percentage of respondents believed that what are commonly called "nuisance" alarms occur frequently (81%), disrupt patient care (77%), and can reduce trust in alarms, causing clinicians to disable them (78%).1 In other studies,2–4 researchers have reported a high percentage (86%–99.4%) of false-positive alarms produced by physiological monitors, stating that alarms result in a change in the management of the patient less than 1% of the time.
The probability of responding to an alarm is lower if the false-alarm rate is high, and alarms in use today do not convey the intended sense of urgency.3,5–7 Nurses in intensive care units stated that the primary problem with alarms is that they are continuously going off and that the largest contributor to the number of false alarms in intensive care units is the pulse oximetry alarm.8–10 "Smart alarms," which analyze multiple parameter changes in a patients condition, may be a solution, but not all monitors currently in use have such features.
Excessive numbers of cardiac monitor alarms and fear that nurses have become desensitized to these alarms were identified as a safety concern by nurses in a 950-bed, northeastern academic medical center. An interdisciplinary alarm management task force was created and charged with (1) evaluating excessive equipment alarms that obscure and desensitize clinicians, (2) standardizing the hospitals approach to alarm management, (3) assessing the reliability of secondary or adjunct alarm notification devices, (4) determining the educational needs of clinicians regarding alarm management, and (5) assessing new technology and systems that may improve alarm management. The purpose of this article is to describe a unit-based quality improvement initiative that enabled the task force to quantify the frequency of cardiac monitor alarms on a single unit and to perform small tests of change to improve management of monitor alarms.
| When alarm frequency is high, nurses can become desensitized and develop "alarm fatigue."
|
| Description of the Test Unit |
|---|
|
|
|---|
A 12-lead electrocardiography hard-wire monitoring system was in use on the unit, with a bedside display and the capability of switching to telemetry mode as needed. This type of monitoring system allows multilead continuous analysis. Although multilead analysis is performed, typically leads II and V1 are viewed. The only time this is changed is when there is difficulty viewing either lead because of its shape or voltage, in which case an alternative lead is selected.
During telemetry, the patient is placed on a 5-lead analysis monitoring system and can be monitored only within the antenna range, which spans the length of the MPCU hallway. A central monitor is available at the nursing station. When patients are transported off the unit, a transport monitor is used and staff trained in physiological monitoring interpretation accompany the monitored patient. Nurses on this unit monitor many types of physiological parameters, including invasive and noninvasive blood pressure, central venous pressure, and oxygen saturation, as well as dysrhythmias. ST-segment analysis is available but is activated only for cardiac conditions such as chest pain of suspected cardiac origin or ischemia. The unit does not have an assigned house staff physician; consequently, a physician is not physically present on the unit at all times. MPCU nurses therefore must be skilled in customizing and responding to physiological monitor alarms.
| The largest contributor to the number of false alarms was the pulse oximetry alarm.
|
The MPCU has a high degree of patient activity, particularly on the day and evening shifts. Patients move about frequently in their rooms as they engage in physical and occupational therapy, visit with friends and family, and leave the unit for tests and procedures. Physiological monitor alarms add to the chaos created by other types of extraneous noise.
As the test unit, the MPCU participated in quality monitoring activities aimed at quantifying the numbers of crises, warning, and system warning monitor alarms on the unit. These activities were followed by small tests of change by using a systematic approach to reducing the number of alarms and to test the effects of published "best practices" related to management of physiological monitor alarms.
| Methods |
|---|
|
|
|---|
The initial process began with collection of baseline alarm data. Two types of alarms occur with the MPCU monitoring system: patient status alarms and system status alarms. Patient status alarms are divided into 4 types: crisis, warning, advisory, and message. System status alarms are triggered by mechanical or electrical problems (Table 1
). Clinical engineers used a monitoring device that counted all cardiac alarms and system status alarms sounding at the central monitoring station for an 18-day period in January 2006. After analyzing the data and the existing monitoring system, the task force decided to focus on managing the most serious monitor alarms and system status alarms; that is, the crisis alarms, warning alarms, and system warnings. Advisory alarms such as those for low oxygen saturation (pulse oximetry), premature ventricular contractions, and ST-segment changes, which accounted for more than 90% of cardiac monitor alarms, were not included. A future project is planned to address advisory alarms in the cardiology care unit. The hospitals monitor vendor attended meetings of the task force and assisted the team by providing requested monitor information, literature, and clinical consultants during the initiative.
|
In April 2006, MPCU nursing staff completed a pretest questionnaire (Appendix 1
, www.ajcconline.org) developed by the CUSP team to assess monitor knowledge and unit noise level.
The same questionnaire was administered after the tests of change.
|
The second test of change was to revise the default settings for the units monitor alarms, including parameter limits and levels, so that alarms that occurred were actionable and clinically significant. Cardiac monitors have default parameter limits and levels set by the manufacturer. These are modified by the hospitals clinical engineering department per unit request before installation. Default settings take effect each time a patient is discharged and a new patient is admitted to the room with that monitor. The hospital did not have an institutional standard for default settings when this quality improvement initiative was started. The focus was on determining (1) the most frequent alarms, (2) duplicate alarms, and (3) perceived "nuisance" alarms.
Before making any changes to alarm parameter limits and levels, the Alarm Management Task Force verified the definitions of each setting to ensure that the ramification of any changes made would be anticipated. Initial default changes were made in June 2006 and additional changes were made in November 2006. Data collected between changes showed a reduction in alarms. Table 2
shows the default changes made as of November 2006.
|
| The task force intended for all alarms to be actionable.
|
Another default change was to increase the premature ventricular contraction limit. This limit was increased from 6 to 10 per minute in order to decrease the number of alarms caused by benign premature atrial, junctional, and ventricular beats. Because an increase in premature beats also could be indicative of electrolyte disturbances, the task force agreed on 10 as the default limit. This limit ensured that an increase in ectopy would trigger the nurse to evaluate for electrolyte disturbances.
The task force also noted duplicate alarms when reviewing the data. The alarms for high and low heart rate were the same as the bradycardia and tachycardia alarms. Although how the monitor calculates bradycardia varies somewhat from how it calculates heart rate low, and how the monitor calculates tachycardia differs from how it calculates heart rate high, the task force thought that the monitor alarm did not need to sound twice; therefore, the alarms for heart rate high and low were moved to message level and the alarms for bradycardia and tachycardia were increased to warning level.
The final test of change, a software addition to the units physiological monitoring system in November 2006, allowed a remote auto view of alarms at each patients bedside. With this type of notification, a crisis alarm sounds at the central monitor station and at all bedside monitors. This additional software displays the patients vital signs and rhythm in a split-screen view, allowing all staff to see these alarms and act swiftly. The sound of the alarm is unique and can be customized to be a onetime, double-beep alarm or a continuous, double-beep alarm. When this software addition was introduced, the units quality improvement/safety representative reinforced the new default changes and encouraged nurses to customize and individualize monitor alarms on the basis of each patients needs. Nurses were trained to respond to this alarm as they would respond to any crisis alarm. Additionally, MPCU has a secondary notification system that uses a one-way pager carried by the charge nurse for alarm notification.
The final MPCU alarm data collection occurred between December 2006 and January 2007. These data were analyzed and compared with previous results. In May 2007, the postintervention survey was completed and compared with the preintervention survey that had been completed 1 year prior.
| Results |
|---|
|
|
|---|
|
|
Because an institutional standard for alarm response was lacking, the preintervention survey also requested information on how staff responded when an alarm was heard. Responses by staff to this question were to check the patient, verify the alarm, troubleshoot the leads/electrodes, and notify the house officer.
The survey tool was used to check nurses perceived compliance with alarm management. The results of the survey may not be representative of the average nurse employed on the unit because completion of the survey was not mandatory.
| Discussion |
|---|
|
|
|---|
The approach in this quality improvement initiative was to perform small tests of change, which included ensuring proper training of staff about monitoring systems, regular assessment and individualization of alarm parameters, ensuring audibility and accountability of alarms, revision of monitor alarm defaults, and software modification to provide a remote auto view of alarms at the patients bedside. As a result of the positive effects of retraining, a skills checklist for alarm management competency (Appendix 2
, www.ajcconline.org) is now used on the unit to educate all new nurses about the monitor system and to ensure that they are knowledgeable about monitors and alarm management.
This checklist includes information on electrode preparation and placement, monitor skills, and troubleshooting that supplements the practical points of the hospitalwide interdisciplinary policy on monitors.
|
| This initiative resulted in a 43% reduction in critical physiologic monitor alarms.
|
This unit-based initiative was beneficial as a starting point for revamping alarm management throughout the institution. Overall, the CUSP team found that modifying alarm parameter defaults and educating nurses about the importance of individualizing monitor alarm parameters helped decrease excessive monitor alarms. The CUSP team also learned that even with proper tailoring of alarms, crisis alarms such as asystole, ventricular tachycardia, and ventricular fibrillation can occur falsely. Some of the main reasons for this undesired effect are poor skin preparation or electrode interface, movement of the patient, and lack of adherence of electrodes. When false crisis alarms occur, these "nuisance" alarms can result in nurse desensitization, which may lead to delayed action by the nurse the next time the monitor alarms.2,4,9
Other valuable lessons learned include the following: (1) unit staff should analyze their alarm parameters and alarm levels to determine if they are appropriately set and avoid duplicative alarms; (2) alarm parameters should be set to actionable levels to decrease the number of false or "nuisance" alarms occurring and increase the likelihood of the alarm being an actionable alarm so it will not be ignored; (3) nurses must be trained to individualize alarm parameters and levels so alarms that occur are meaningful and actionable; and (4) institutions would do well to establish an institutionwide standard for management of physiological monitor alarms. These lessons were brought back to the Alarm Management Task Force for discussion and consideration by other monitored units.
| Alarms that are viewed as false-positive or nuisance alarms may delay a nurses response.
|
The results of these small tests of change on a single unit provided the basis for applied learning to other adult and pediatric units in the institution. The test unit has continued its quest for quality by partnering with the Clinical Engineering Department to pilot a new middleware system that enables physiological monitor alarms to be communicated through a variety of systems such as pagers, cell phones, and LED marquee signs. This system will broaden the ability to receive alarm notification through multiple channels versus just one-way pagers.
Although this quality improvement initiative led to standardization of monitor education and implementation of a hospitalwide monitor protocol, additional benefits included increasing nurses knowledge of monitors and decreasing the number of nuisance alarms. Complete buy-in from coworkers or staff was essential to achieving a true culture change in alarm management. Manipulation of monitor defaults and staff training are not sufficient to sustain change unless the unit is held accountable for maintaining a zero-tolerance for nuisance alarms and troubleshooting these alarms as soon as they occur.
| This unit-based initiative was a starting point for revamping alarm management in the institution.
|
| ACKNOWLEDGMENTS |
|---|
This article is followed by an AJCC Patient Care Page on page 38.
None reported.
Now that youve read the article, create or contribute to an online discussion on this topic. Visit www.ajcconline.org and click "Respond to This Article" in either the full-text or PDF view of the article.
For more about monitor alarms, visit the American Journal of Critical Care Web site, www.ajcconline.org, and read the article by Korniewicz et al, "A National Online Survey on the Effectiveness of Clinical Alarms" (January 2008).
| REFERENCES |
|---|
|
|
|---|
Related articles in AJCC:
This article has been cited by other articles:
![]() |
L. Bell Monitor Alarm Fatigue Am. J. Crit. Care., January 1, 2010; 19(1): 38 - 38. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |