Arrhythmia Alarms Research Articles

Learning From Alarms: A Robust Learning Approach for Accurate Photoplethysmography-Based Atrial Fibrillation Detection Using Eight Million Samples Labeled With Imprecise Arrhythmia Alarms.

Atrial fibrillation (AF) is a common cardiac arrhythmia with serious health consequences if not detected and treated early. Detecting AF using wearable devices with photoplethysmography (PPG) sensors and deep neural networks has demonstrated some success using proprietary algorithms in commercial solutions. However, to improve continuous AF detection in ambulatory settings towards a population-wide screening use case, we face several challenges, one of which is the lack of large-scale labeled training data. To address this challenge, we propose to leverage AF alarms from bedside patient monitors to label concurrent PPG signals, resulting in the largest PPG-AF dataset so far (8.5 M 30-second records from 24,100 patients) and demonstrating a practical approach to build large labeled PPG datasets. Furthermore, we recognize that the AF labels thus obtained contain errors because of false AF alarms generated from imperfect built-in algorithms from bedside monitors. Dealing with label noise with unknown distribution characteristics in this case requires advanced algorithms. We, therefore, introduce and open-source a novel loss design, the cluster membership consistency (CMC) loss, to mitigate label errors. By comparing CMC with state-of-the-art methods selected from a noisy label competition, we demonstrate its superiority in handling label noise in PPG data, resilience to poor-quality signals, and computational efficiency.

Fewer emergency department alarms is associated with reduced use of medications for acute agitation

Background and objectivesPatient monitoring systems provide critical information but often produce loud, frequent alarms that worsen patient agitation and stress. This may increase the use of physical and chemical restraints with implications for patient morbidity and autonomy. This study analyzes how augmenting alarm thresholds affects the proportion of alarm-free time and the frequency of medications administered to treat acute agitation. MethodsOur emergency department's patient monitoring system was modified on June 28, 2022 to increase the tachycardia alarm threshold from 130 to 150 and to remove alarm sounds for several arrhythmias, including bigeminy and premature ventricular beats. A pre-post study was performed lasting 55 days before and 55 days after this intervention. The primary outcome was change in number of daily patient alarms. The secondary outcomes were alarm-free time per day and median number of antipsychotic and benzodiazepine medications administered per day. The safety outcome was the median number of patients transferred daily to the resuscitation area. We used quantile regression to compare outcomes between the pre- and post-intervention period and linear regression to correlate alarm-free time with the number of sedating medications administered. ResultsBetween the pre- and post-intervention period, the median number of alarms per day decreased from 1332 to 845 (−37%). This was primarily driven by reduced low-priority arrhythmia alarms from 262 to 21 (−92%), while the median daily census was unchanged (33 vs 32). Median hours per day free from alarms increased from 1.0 to 2.4 (difference 1.4, 95% CI 0.8–2.1). The median number of sedating medications administered per day decreased from 14 to 10 (difference − 4, 95% CI −1 to −7) while the number of escalations in level of care to our resuscitation care area did not change significantly. Multivariable linear regression showed a 60-min increase of alarm-free time per day was associated with 0.8 (95% CI 0.1–1.4) fewer administrations of sedating medication while an additional patient on the behavioral health census was associated with 0.5 (95% CI 0.0–1.1) more administrations of sedating medication. ConclusionA reasonable change in alarm parameter settings may increase the time patients and healthcare workers spend in the emergency department without alarm noise, which in this study was associated with fewer doses of sedating medications administered.

Performance comparison of 6 in-hospital patient monitoring systems in the detection and alarm of ventricular cardiac arrhythmias

BackgroundPatient monitoring devices are critical in alerting potential cardiac arrhythmias during hospitalization; however, there are concerns of alarm fatigue due to high false alarm rates. ObjectiveTo evaluate sensitivity and false alarm rate of hospital-based continuous electrocardiographic (ECG) monitoring technologies. MethodsSix commonly-used multiparameter bedside monitoring systems available in the United States were evaluated: B125M (GE HealthCare), ePM10 and iPM12 (Mindray), Efficia and IntelliVue (Philips), and Life Scope (Nihon Kohden). Sensitivity was tested using ECG recordings containing 57 true ventricular tachycardia (VT) events. False positive rate testing used 205 patient-hours of ECG recordings containing no cardiac arrhythmias. Signals from ECG recordings were fed to devices simultaneously; high-severity arrhythmia alarms were tracked. Sensitivity to true VT events and false positive rates were determined. Differences were assessed using Fisher’s exact tests (sensitivity) and Z-tests (false positive rates). ResultsB125M raised 56 total alarms for 57 annotated VT events and had the highest sensitivity (98%, P<0.05), followed by iPM12 (84%), Life Scope (81%), Efficia (79%), ePM10 (77%), and IntelliVue (75%). B125M raised 20 false alarms, significantly lower (P<0.0001) than iPM12 (284), Life Scope (292), IntelliVue (304), ePM10 (324), and Efficia (493). The most common false alarm was VT, followed by non-sustained VT. ConclusionWe found significant performance differences among multiparameter bedside ECG monitoring systems using previously-collected recordings. B125M had the highest sensitivity in detecting true VT events and lowest false alarm rate. These results can assist in minimizing alarm fatigue and optimizing patient safety by careful selection of in-hospital continuous monitoring technology.

Usability Study on Patient Monitoring Systems: An Evaluation of a User Interface Based on User Experience and Preference.

BACKGROUND The poor ergonomic design of patient monitoring systems can cause user errors and patient harm. This paper presents the results of a comparative usability study based on user experience and the results of a user preference survey. MATERIAL AND METHODS We conducted a usability study of 3 patient monitoring systems: Mediana M50, Philips IntelliVue MP70, and Philips IntelliVue MX700. Thirty-nine Coronary Care Unit nurses and 19 Pulmonology and Allergy Care Unit nurses participated in this usability study. User experience was assessed with the Post-Study System Usability Questionnaire and the National Aeronautics and Space Administration Task Load Index. A user preference survey was conducted to evaluate the subjective medical device design preferences for the M50 system's user interface. RESULTS Nurses from the Coronary Care Unit recognized a higher system usability for the MP70 than the M50 (P=0.001) system, and a lower workload for the MP70 compared with the M50 (P=0.005) system. There was no significant (P>0.05) difference in perceived system usability and workload between the M50 and MX700 systems for the nurses from the Pulmonology and Allergy Care Unit. Nurses preferred to activate the arrhythmia alarms except for the ST alarms and missed the beat alarm. They also preferred having a wave freeze function, standby mode, and early warning scoring function, which provides a signal for a patient's deterioration in health. CONCLUSIONS The study provides valuable data on a user interface evaluation based on user experience and preference. The outcome of this study will be helpful for designing next-generation patient monitors with improved patient safety.

A contrastive learning approach for\xa0ICU false arrhythmia\xa0alarm reduction

The high rate of false arrhythmia alarms in Intensive Care Units (ICUs) can lead to disruption of care, negatively impacting patients’ health through noise disturbances, and slow staff response time due to alarm fatigue. Prior false-alarm reduction approaches are often rule-based and require hand-crafted features from physiological waveforms as inputs to machine learning classifiers. Despite considerable prior efforts to address the problem, false alarms are a continuing problem in the ICUs. In this work, we present a deep learning framework to automatically learn feature representations of physiological waveforms using convolutional neural networks (CNNs) to discriminate between true vs. false arrhythmia alarms. We use Contrastive Learning to simultaneously minimize a binary cross entropy classification loss and a proposed similarity loss from pair-wise comparisons of waveform segments over time as a discriminative constraint. Furthermore, we augment our deep models with learned embeddings from a rule-based method to leverage prior domain knowledge for each alarm type. We evaluate our method using the dataset from the 2015 PhysioNet Computing in Cardiology Challenge. Ablation analysis demonstrates that Contrastive Learning significantly improves the performance of a combined deep learning and rule-based-embedding approach. Our results indicate that the final proposed deep learning framework achieves superior performance in comparison to the winning entries of the Challenge.

Experience with a Wearable Cardioverter-defibrillator in 436 Patients

The aim of the present study was to review the safety and efficiency of wearable cardioverter-defibrillators (WCDs) under current guideline-directed medical therapy (GDMT). We retrospectively analyzed 436 consecutive WCD patients seen in the years 2014–2020. Detected automatic arrhythmia alarm (AA) episodes were validated and classified as correct or incorrect. The positive predictive value (PPV) was calculated. GDMT was optimized in our outpatient clinic to maximal tolerated doses. During a total wear time (WT) of 23,527 days, 3,135 AAs were transmitted from 206 of 436 (47.2%) patients. Visual analysis revealed correct diagnoses of non-sustained ventricular tachycardia (VT) in 38 AAs from 6 patients (total PPV, 1.21%; PPV in VT patients, 41%); the remaining AAs were artifacts. No appropriate or inappropriate shocks and fatalities were recorded. LVEF significantly improved (P < .001) during the WT from 25% (range, 20%–30%) to 40% (range, 34%–46%). Defibrillators were implanted in 109 patients (27%). The PPV for VT of the WCD was very low. There were fewer instances of true VT than previously reported, and no shocks (appropriate or inappropriate) were delivered. The majority of patients greatly improved with GDMT, and device implantation rates were lower than previously reported. Improvements in arrhythmia detection algorithms are warranted. Based on our results, WCDs are rarely needed for lifesaving shocks under optimal GDMT.

Real-Time Arrhythmia Detection Using Hybrid Convolutional Neural Networks.

BackgroundAccurate detection of arrhythmic events in the intensive care units (ICU) is of paramount significance in providing timely care. However, traditional ICU monitors generate a high rate of false alarms causing alarm fatigue. In this work, we develop an algorithm to improve life threatening arrhythmia detection in the ICUs using a deep learning approach.Methods and ResultsThis study involves a total of 953 independent life‐threatening arrhythmia alarms generated from the ICU bedside monitors of 410 patients. Specifically, we used the ECG (4 channels), arterial blood pressure, and photoplethysmograph signals to accurately detect the onset and offset of various arrhythmias, without prior knowledge of the alarm type. We used a hybrid convolutional neural network based classifier that fuses traditional handcrafted features with features automatically learned using convolutional neural networks. Further, the proposed architecture remains flexible to be adapted to various arrhythmic conditions as well as multiple physiological signals. Our hybrid‐ convolutional neural network approach achieved superior performance compared with methods which only used convolutional neural network. We evaluated our algorithm using 5‐fold cross‐validation for 5 times and obtained an accuracy of 87.5%±0.5%, and a score of 81%±0.9%. Independent evaluation of our algorithm on the publicly available PhysioNet 2015 Challenge database resulted in overall classification accuracy and score of 93.9% and 84.3%, respectively, indicating its efficacy and generalizability.ConclusionsOur method accurately detects multiple arrhythmic conditions. Suitable translation of our algorithm may significantly improve the quality of care in ICUs by reducing the burden of false alarms.

Computer Assisted Patient Monitoring: Associated Patient, Clinical and ECG Characteristics and Strategy to Minimize False Alarms

This chapter is a review of studies that have examined false arrhythmia alarms during in-hospital electrocardiographic (ECG) monitoring in the intensive care unit. In addition, we describe an annotation effort being conducted at the UCSF School of Nursing, Center for Physiologic Research designed to improve algorithms for lethal arrhythmias (i.e., asystole, ventricular fibrillation, and ventricular tachycardia). Background: Alarm fatigue is a serious patient safety hazard among hospitalized patients. Data from the past five years, showed that alarm fatigue was responsible for over 650 deaths, which is likely lower than the actual number due to under-reporting. Arrhythmia alarms are a common source of false alarms and 90% are false. While clinical scientists have implemented a number of interventions to reduce these types of alarms (e.g., customized alarm settings; daily skin electrode changes; disposable vs. non-disposable lead wires; and education), only minor improvements have been made. This is likely as these interventions do not address the primary problem of false arrhythmia alarms, namely deficient and outdated arrhythmia algorithms. In this chapter we will describe a number of ECG features associated with false arrhythmia alarms. In addition, we briefly discuss an annotation effort our group has undertaken to improve lethal arrhythmia algorithms.

Effect of bundle set interventions on physiologic alarms and alarm fatigue in an intensive care unit: A quality improvement project

ObjectiveTo determine if the implementation of an evidence-based bundle designed to reduce the number of physiologic monitor alarms reduces alarm fatigue in intensive care nurses. DesignThis quality improvement project retrospectively reviewed alarm data rates, types, and frequency to identify the top three problematic physiologic alarms in an intensive care unit. An alarm management bundle was implemented to reduce the number of alarms. The Nurses’ Alarm Fatigue Questionnaire was used to measure nurses’ alarms fatigue pre- and post-implementation of the bundle. SettingA combined medical surgical intensive care unit at an accredited hospital in the United States. ResultsThe top three problematic alarms identified during the pre-implementation phase were arrhythmia, invasive blood pressure, and respiration alarms. All three identified problematic physiologic alarms had a reduction in frequency with arrhythmia alarms demonstrating the largest decrease in frequency (46.82%). When measuring alarm fatigue, the overall total scores increased from pre- (M = 30.59, SD = 5.56) to post-implementation (M = 32.60, SD = 4.84) indicating no significant difference between the two periods. ConclusionAfter implementing an alarm management bundle, all three identified problematic physiologic alarms decreased in frequency. Despite the reduction in these alarms, there was not a reduction in nurses’ alarm fatigue.

ECG alarms during left ventricular assist device (LVAD) therapy in the ICU

BackgroundIn hospitalized patients with left ventricular assist device (LVAD), electrical interference and low amplitude QRS complexes are common, which could impact the accuracy of electrocardiographic (ECG) arrhythmia detection and create technical alarms. This could contribute to provider alarm fatigue and threaten patient safety. ObjectivesWe examined three LVAD patients in the cardiac intensive care unit (ICU) to determine: 1) the frequency and accuracy of audible arrhythmia alarms; 2) occurrence rates of technical alarms; and 3) alarm burden (# alarms/hour of monitoring) MethodsSecondary analysis. ResultsDuring 593 h, there were 549 audible arrhythmia alarms and 98% were false. There were 25,232 technical alarms and 93% were for artifact, which was configured as an inaudible text alert. ConclusionFalse-arrhythmia and technical alarms are frequent in LVAD patients. Future studies are needed to identify both clinical and algorithm-based strategies to improve arrhythmia detection and reduce technical alarms in LVAD patients.

Research of Methods to Reduce Alarm Fatigue of Monitoring System

In order to solve alarm fatigue, the algorithm optimization strategies were researched to reduce false and worthless alarms. A four-lead arrhythmia analysis algorithm, a multiparameter fusion analysis algorithm, an intelligent threshold reminder, a refractory period delay technique were proposed and tested with collected 28 679 alarms in multi-center study. The sampling survey indicate that the 80.8% of arrhythmia false alarms were reduced by the four-lead analysis, the 55.9% of arrhythmia and pulse false alarms were reduced by the multi-parameter fusion analysis, the 28.0% and 29.8% of clinical worthless alarms were reduced by the intelligent threshold and refractory period delay techniques respectively. Finally, the total quantity of alarms decreased to 12 724. To increase the dimensionality of parametric analysis and control the alarm limits and delay time are conducive to reduce alarm fatigue in intensive care units.

Double Trouble: Patients With Both True and False Arrhythmia Alarms

Patients with both true and false arrhythmia alarms pose a challenge because true alarms might be buried among a large number of false alarms, leading to missed true events. To determine (1) the frequency of patients with both true and false arrhythmia alarms; (2) patient, clinical, and electrocardiographic characteristics associated with both true and false alarms; and (3) the frequency and types of true and false arrhythmia alarms. This was a secondary analysis using data from an alarm study conducted at a tertiary academic medical center. Of 461 intensive care unit patients, 211 (46%) had no arrhythmia alarms, 12 (3%) had only true alarms, 167 (36%) had only false alarms, and 71 (15%) had both true and false alarms. Ventricular pacemaker, altered mental status, mechanical ventilation, and cardiac intensive care unit admission were present more often in patients with both true and false alarms than among other patients (P < .001). Intensive care unit stays were longer in patients with only false alarms (mean [SD], 106 [162] hours) and those with both true and false alarms (mean [SD], 208 [333] hours) than in other patients. Accelerated ventricular rhythm was the most common alarm type (37%). An awareness of factors associated with arrhythmia alarms might aid in developing solutions to decrease alarm fatigue. To improve detection of true alarms, further research is needed to build and test electrocardiographic algorithms that adjust for clinical and electrocardiographic characteristics associated with false alarms.

Team-Based Intervention to Reduce the Impact of Nonactionable Alarms in an Adult Intensive Care Unit.

Nonactionable alarms comprise over 70% of alarms and contribute a threat to patient safety. Few studies have reported approaches to translate and sustain these interventions in clinical settings. This study tested whether an interprofessional team-based approach can translate and implement effective alarm reduction interventions in the adult intensive care unit. The study was a prospective, cohort, pre- and postdesign with repeated measures at baseline (preintervention) and post-phase I and II intervention periods. The settings for the most prevalent nonactionable arrhythmia and bedside parameter alarms were adjusted during phases I and II, respectively. The number of total alarms was reduced by 40% over a 14-day period after both intervention phases were implemented. The most prevalent nonactionable parameter alarms decreased by 47% and arrhythmia alarms decreased by 46%. It is feasible to translate and sustain system-level alarm management interventions addressing alarm fatigue using an interprofessional team-based approach.

Blind Source Separation for the Aggregation of Machine Learning Algorithms: An Arrhythmia Classification Case

In this work, we present an application of the blind source separation (BSS) algorithm to reduce false arrhythmia alarms and to improve the classification accuracy of artificial neural networks (ANNs). The research was focused on a new approach for model aggregation to deal with arrhythmia types that are difficult to predict. The data for analysis consisted of five-minute-long physiological signals (ECG, BP, and PLETH) registered for patients with cardiac arrhythmias. For each patient, the arrhythmia alarm occurred at the end of the signal. The data present a classification problem of whether the alarm is a true one—requiring attention or is false—should not have been generated. It was confirmed that BSS ANNs are able to detect four arrhythmias—asystole, ventricular tachycardia, ventricular fibrillation, and tachycardia—with higher classification accuracy than the benchmarking models, including the ANN, random forest, and recursive partitioning and regression trees. The overall challenge scores were between 63.2 and 90.7.

Single-modal and multi-modal false arrhythmia alarm reduction using attention-based convolutional and recurrent neural networks.

This study proposes a deep learning model that effectively suppresses the false alarms in the intensive care units (ICUs) without ignoring the true alarms using single- and multi- modal biosignals. Most of the current work in the literature are either rule-based methods, requiring prior knowledge of arrhythmia analysis to build rules, or classical machine learning approaches, depending on hand-engineered features. In this work, we apply convolutional neural networks to automatically extract time-invariant features, an attention mechanism to put more emphasis on the important regions of the segmented input signal(s) that are more likely to contribute to an alarm, and long short-term memory units to capture the temporal information presented in the signal segments. We trained our method efficiently using a two-step training algorithm (i.e., pre-training and fine-tuning the proposed network) on the dataset provided by the PhysioNet computing in cardiology challenge 2015. The evaluation results demonstrate that the proposed method obtains better results compared to other existing algorithms for the false alarm reduction task in ICUs. The proposed method achieves a sensitivity of 93.88% and a specificity of 92.05% for the alarm classification, considering three different signals. In addition, our experiments for 5 separate alarm types leads significant results, where we just consider a single-lead ECG (e.g., a sensitivity of 90.71%, a specificity of 88.30%, an AUC of 89.51 for alarm type of Ventricular Tachycardia arrhythmia).

Routine Continuous Electrocardiographic Monitoring Following Percutaneous Coronary Interventions.

The clinical utility of routine electrocardiographic monitoring following percutaneous coronary interventions (PCI) is not well studied. We prospectively evaluated the incidence, cost, and the clinical implications of actionable arrhythmia alarms on telemetry monitoring following PCI. One thousand three hundred fifty-eight PCI procedures (989 [72.8%] for acute coronary syndrome and 369 [27.2%] for stable angina) on patients admitted to nonintensive care unit were identified and divided into 2 groups; group 1, patients with actionable alarms (AA) and group 2, patients with non-AA. AA included (1) ≥3 s electrical pause or asystole; (2) high-grade Mobitz type II atrioventricular block or complete heart block; (3) ventricular fibrillation; (4) ventricular tachycardia (>15 beats); (5) atrial fibrillation with rapid ventricular response; (6) supraventricular tachycardia (>15 beats). Primary outcomes were 30-day all-cause mortality. Cost-savings analysis was performed. Incidence of AA was 2.2% (37/1672). Time from end of procedure to AA was 5.5 (0.5, 24.5) hours. Patients with AA were older, presented with acute congestive heart failure or non-ST-segment-elevation myocardial infarction, and had multivessel or left main disease. The 30-day all-cause mortality was significantly higher in patients with AA (6.5% versus 0.3% in non-AA [P<0.001]). Applying the standardized costing approach and tailored monitoring per the American Heart Association guidelines lead to potential cost savings of $622 480.95 for the entire population. AA following PCI were infrequent but were associated with increase in 30-day mortality. Following American Heart Association guidelines for monitoring after PCI can lead to substantial cost saving.

Reduction of false alarms in the intensive care unit using an optimized machine learning based approach

This work attempts to reduce the number of false alarms generated by bedside monitors in the intensive care unit (ICU), as a majority of current alarms are false. In this study, we applied methods that can be categorized into three stages: signal processing, feature extraction, and optimized machine learning. At the stage of signal processing, we ensured that the heartbeats were properly annotated. During feature extraction, besides extracting features that are relevant to the arrhythmic alarms, we also extracted a set of signal quality indices (SQIs), which we used to distinguish noise/artifact from normal physiological signals. When applying a machine learning algorithm (Random Forest), we performed feature selection in order to reduce the complexity of the models and improve the efficiency of the algorithm. The dataset used is from Reducing False Arrhythmia Alarms in the ICU: the PhysioNet/Computing in Cardiology Challenge 2015. Using the performance metric “score” from the Challenge, we achieved a score of 83.08 in the real-time category on the hidden test set, which is the highest in all published work.

Reducing False Arrhythmia Alarms Using Different Methods of Probability and Class Assignment in Random Forest Learning Methods.

The literature indicates that 90% of clinical alarms in intensive care units might be false. This high percentage negatively impacts both patients and clinical staff. In patients, false alarms significantly increase stress levels, which is especially dangerous for cardiac patients. In clinical staff, alarm overload might lead to desensitization and could result in true alarms being ignored. In this work, we applied the random forest method to reduce false arrhythmia alarms and specifically explored different methods of probability and class assignment, as these affect the classification accuracy of the ensemble classifiers. Due to the complex nature of the problem, i.e., five types of arrhythmia and several methods to determine probability and the alarm class, a synthetic measure based on the ranks was proposed. The novelty of this contribution is the design of a synthetic measure that helps to leverage classification results in an ensemble model that indicates a decision path leading to the best result in terms of the area under the curve (AUC) measure or the global accuracy (score). The results of the research are promising. The best performance in terms of the AUC was 100% accuracy for extreme tachycardia, whereas the poorest results were for ventricular tachycardia at 87%. Similarly, in terms of the accuracy, the best results were observed for extreme tachycardia (91%), whereas ventricular tachycardia alarms were the most difficult to detect, with an accuracy of only 51%.

ECG Monitoring during End of Life Care: Implications on Alarm Fatigue

Excessive numbers of clinical alarms in the intensive care unit (ICU) contribute to alarm fatigue. Efforts to eliminate unnecessary alarms, including during end of life (EOL) care, are pivotal. This study describes electrocardiographic (ECG) arrhythmia alarm usage following the decision for comfort care. We conducted a review of electronic health records (EHR) in patients who died and had comfort care orders that were in place during our study. The occurrences of ECG arrhythmia alarms among these patients were examined. We found 151 arrhythmia alarms that were generated in 11 patients after comfort care was initiated: 72% were audible, 21% were manually muted, and 7% had an unknown audio label. Level of alarm: 33% crisis, 58% warning, 1% message, and 8% were labeled unknown. Our report shows that ECG monitoring was commonly maintained during the EOL care. Since the goal of care during this phase is for both patient and family comfort, it is important for the clinicians to weigh the benefits versus harms of the continuous ECG monitoring.

Epidemiology of patient monitoring alarms in the neonatal intensive care unit

ObjectiveTo characterize the rate of monitoring alarms by alarm priority, signal type and developmental age in a Level-IIIB Neonatal Intensive Care Unit (NICU) population.Study DesignRetrospective analysis of 2 294 687 alarm messages from Philips monitors in a convenience sample of 917 NICU patients, covering 12 001 patient-days. We stratified alarm rates by alarm priority, signal type, postmenstrual age (PMA), and birth weight (BW) and reviewed and adjudicated over 21 000 critical alarms.ResultsOf all alarms, 3.6% were critical alarms, 55.0% were advisory alarms, and 41.4% were device alerts. Over 60% of alarms related to oxygenation monitoring. The average alarm rate (±SEM) was 177.1±4.9 [median: 135.9; IQR: 89.2–213.3] alarms/patient-day; the medians varied significantly with PMA and BW (p<0.001) in U-shaped patterns, with higher rates at lower and higher PMA and BW. Based on waveform reviews, over 99% of critical arrhythmia alarms were deemed technically false.ConclusionsThe alarm burden in this NICU population is very significant; the average alarm rate significantly underrepresents alarm rates at low and high PMA and BW. Virtually all critical arrhythmia alarms were artifactual.