The last 25 years have seen considerable technological advancements in the monitoring of severely ill patients in the intensive care unit (ICU) 1. These include sophisticated medical devices with auditory alarms. Such alarms aim to improve patient safety by alerting medical staff to information that requires a response, such as unsatisfactory or changing physiological parameters, and malfunctioning medical equipment. The question is are these alarms really a good fit for intensive care?
Even though it seems paradoxical, the results of numerous recent studies indicate that the presence of a high number of alarms poses a potential risk to the integrity and safety of patients 2. This is not only due to organic disorders caused by high levels of noise that sometimes exceed 120 dB but also because the sensory overload from an excessive number of alarms can lead professionals to become exhausted and desensitized 3. This ‘alarm fatigue’ can decrease caregivers’ alertness and confidence in the urgency of these alarms, and make it difficult to identify which alarms are clinically significant or relevant. A lack of response to relevant alarms may result in severe consequences for the clinical conditions of patients 4.
Medical staff describe alarms as being noisy, blatant, or a nuisance, causing cognitive stress and disturbing patients and staff alike. In order to attend to alarms, patient care and workflow is interrupted and clinicians’ overall workload is significantly increased (even if no action is required, just recognizing, acknowledging and evaluating an alarm takes time). Nuisance alarms can therefore lead to clinicians taking inappropriate actions such as lowering the alarm volume, extending alarm limits outside a reasonable range, or disabling alarms. However, relevant alarms that are neglected, silenced or ignored by health professionals can impact patient safety. Moreover, there is no standardization of alarm sounds among manufacturers 5. This further exacerbates the problem of alarm fatigue as caregivers need to be able to distinguish the audible alarms and react based on their perceived importance. Thus, there is the irony that alarms whose purpose is meant to protect patients have instead led to increased unit noise, alarm fatigue, and consequently a major patient safety concern.
The excessive number of alarms in ICUs have been shown in numerous studies. For example, a sample survey from the ICU at John Hopkins showed hospital staff are exposed to an average of 350 alarms per bed per day 6. This equates to tens of thousands of alarms per hospital every day. Indeed, as new devices are introduced, the number of alarms to which a healthcare professional may be exposed may be as high as 1000 alarms per shift 7. Alarm fatigue in a study in a Brazilian hospital in 2018 showed that more than 66% of alarms were recorded without a response and less than 26% of alarms were attended within 5 minutes 4.
Since 2014, The Joint Commission (TJC) of the Food and Drug Administration (FDA) has made alarm management a National Patient Safety Goal because there continues to be sentinel events related to alarm management and fatigue 2. For example, there were 566 reports to the FDA of patient deaths related to monitoring device alarms between 2005 and 2008 8. The Emergency Care Research Institute (ECRI), an organization specializing in patient safety and the use of electro-medical equipment, has also continually found alarms to be the number one technological danger in the health field 9. This is due to the high number of adverse events among inpatients of hospitals in the USA, including death, cardiorespiratory arrest, and cardiac arrhythmias. Accordingly, several organizations based in the USA, in addition to TJC and ECRI, such as the Advancement of Medical Instrumentation (AAMI Foundation) and the American Association of Critical-Care Nurses have been focusing their efforts to highlight this issue and improve alarm safety in healthcare.
Research suggests that 80%–99% of alarms are false 10. This is largely due to alarms being designed by default to be high in sensitivity but low in specificity ‘because a manufacturer is much more likely to be held liable for an alarm system that fails to annunciate a clinically significant alarm condition than for any problems caused by annunciating false-alarm conditions’ 11. Clinicians also usually follow the ‘better-safe-than-sorry’ logic in setting alarm systems in clinical environments. This means that although today’s monitors can detect a large number of conditions, these functions typically come with a high rate of false positives. For example, a study of alarms in a children’s hospital found that 87% of ICU alarms and 99% of ward alarms were non-actionable 12. Clinicians therefore need to make a conscious decision as to which events should be detected to decrease the incidence of non-actionable alarms. For example, several studies have shown that arrhythmia detection is frequently enabled without any indication 7.
Nuisance alarms may create a ‘cry wolf’ affect where the sheer number of false positives can cause distrust and result in inaction at a time when it is necessary. False or nuisance alarms occur when alarm parameters are set outside of clinically actionable limits or are set to a generalized population instead of the appropriate patient-specific conditions. For example, adult defaults used for pediatric patients; a configuration for a medical patient used for a surgical patient; or alarm thresholds not adjusted to a patient’s baseline values upon admission. Moreover, over monitoring of patients who no longer need such intense monitoring can trigger false alarms. By setting limits at the ‘action required’ point, the need for repeated adjustments can be limited. For example, the Boston Medical Center reduced its weekly audible cardiac alarm rate by 89% by adjusting their monitor alarms for bradycardia, tachycardia and heart rate limits 1. This therefore highlights the need for ICUs to individualize care and not use a ‘one size fits all’ approach.
The other side to reliance on alarms is that they can create a false sense of security. Harmful inaction can occur as a result of over trusting and heavily relying on alerts. Therefore, if action is only taken when alarms are sounded, false negatives may also impact patient safety.
Thus, we need to create smart alarms and smart displays in which the system does not over alert the clinician but is also sophisticated enough to identify those cases in which a deterioration may occur. Ideally, alarms that look for patterns over multiple parameters are required to indicate ‘real’ problems. Indeed, hospitals have expressed a need for a single assessment indicator to assess pulse rate, exhaled carbon dioxide, respiratory rate, oxygenation etc. 6. In fact, patient data captured from multiple sources can allow for continuous surveillance. In contrast to traditional patient monitoring, such a systematic, goal-directed process can detect physiological changes in patients early, interpret the clinical implications of those changes and alert clinicians so they can intervene rapidly 13. The CLEW platform was designed with this in mind and provides information about patients’ expected deteriorations as well as a designated platform to support the clinical workflow. The deterioration probability is calculated in near real time for each patient using pre-developed models for various deterioration types. As CLEW models were created using sophisticated analytics and machine learning, the system dramatically reduces false and irrelevant alerts. Clinicians can see at a glance on one smart display the full clinical situation of the patient with the patient’s data from all units and highlighting all alerts. These individually set alarms of multiple parameters for each patient can therefore generate genuinely actionable alerts with few false alarms. This is crucial in reducing alarm fatigue, improving alarm management, providing value-based care, and saving lives.
- Cvach M. Clinical Alarms and the Impact on Patient Safety. initiatives-patientsafety.org.
- The Joint Commission. National Patient Safety Goals Effective January 2019. Hospital Accreditation Program. Available from: https://www.jointcommission.org/-/media/tjc/documents/standards/national-patient-safety-goals/historical/npsg_chapter_hap_jan2019.pdf?db=web&hash=D7E1C2DA08C73CE3F9C8120305A3A8AB Accessed November 15, 2020
- Andrade-Méndez B, Arias-Torres DO, Gómez-Tovar LO. Alarm fatigue in the Intensive Care Unit: Relevance and response time. Enferm Intensiva. 2020; 31:147-153. DOI: 1016/j.enfi.2019.11.002
- Oliveira AEC, Machado AB, Santos ED, Almeida EB. Alarm fatigue and the implications for patient safety. Rev Bras Enferm [Internet]. 2018;71(6):3035-40. https://www.scielo.br/pdf/reben/v71n6/0034-7167-reben-71-06-3035.pdf
- Bridi AC, Louro TQ, da Silva RC. Clinical Alarms in intensive care: implications of alarm fatigue for the safety of patients. Rev Lat Am Enfermagem. 2014;22(6):1034-40. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4309240/pdf/0104-1169-rlae-22-06-01034.pdf
- MacDonald I. Hospitals rank alarm fatigue as top patient safety concern. Fierce Healthcare 2014. https://www.fiercehealthcare.com/healthcare/hospitals-rank-alarm-fatigue-as-top-patient-safety-concern
- Ruskin KJ, Hueske-Kraus D. Alarm fatigue: impacts on patient safety. Curr Opin Anaesthesiol. 2015;28(6):685-90. DOI: 1097/ACO.0000000000000260
- Cvach M. Monitor alarm fatigue: an integrative review. Biomed Instrum Technol. 2012;46(4):268-77. DOI: 2345/0899-8205-46.4.268
- Institute ECRI. Top 10 health technology hazards for 2013. Guidance article. 2012;41(11). https://www.ecri.org/Resources/Whitepapers_and_reports/2013_Health_Devices_Top_10_Hazards.pdf
- Bach TA, Berglund L-M, Turk E. Managing alarm systems for quality and safety in the hospital setting. BMJ Open Quality 2018;7:e000202. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6069923/
- AAMI Foundation. Human factors engineering-design of medical devices (ANSI/AAMI HE75) 2009/(R)2013. Arlington, VA: AAMI, 2013. http://my.aami.org/aamiresources/previewfiles/HE75_1311_preview.pdf
- Bonafide CP, Lin R, Zander M, Graham CS, Paine CW, Rock W, Rich A, Roberts KE, Fortino M, Nadkarni VM, Localio AR, Keren R. Association between exposure to nonactionable physiologic monitor alarms and response time in a children’s hospital. J Hosp Med. 2015;10(6):345-51. https://europepmc.org/article/med/25873486
- Capsule Technologies Inc. Continuous Clinical Surveillance: A Solution for Patient Safety. Sept. 2020. Available from: CapsuleTech.com