Understanding heart rate alarm adjustment in the intensive care units through an analytical approach.

BACKGROUND: Heart rate (HR) alarms are prevalent in ICU, and these parameters are configurable. Not much is known about nursing behavior associated with tailoring HR alarm parameters to individual patients to reduce clinical alarm fatigue. OBJECTIVES: To understand the relationship between heart rate (HR) alarms and adjustments to reduce unnecessary heart rate alarms. METHODS: Retrospective, quantitative analysis of an adjudicated database using analytical approaches to understand behaviors surrounding parameter HR alarm adjustments. Patients were sampled from five adult ICUs (77 beds) over one month at a quaternary care university medical center. A total of 337 of 461 ICU patients had HR alarms with 53.7% male, mean age 60.3 years, and 39% non-Caucasian. Default HR alarm parameters were 50 and 130 beats per minute (bpm). The occurrence of each alarm, vital signs, and physiologic waveforms was stored in a relational database (SQL server). RESULTS: There were 23,624 HR alarms for analysis, with 65.4% exceeding the upper heart rate limit. Only 51% of patients with HR alarms had parameters adjusted, with a median upper limit change of +5 bpm and -1 bpm lower limit. The median time to first HR parameter adjustment was 17.9 hours, without reduction in alarms occurrence (p = 0.57). CONCLUSIONS: HR alarms are prevalent in ICU, and half of HR alarm settings remain at default. There is a long delay between HR alarms and parameters changes, with insufficient changes to decrease HR alarms. Increasing frequency of HR alarms shortens the time to first adjustment. Best practice guidelines for HR alarm limits are needed to reduce alarm fatigue and improve monitoring precision.

Is the Sequence of SuperAlarm Triggers More Predictive Than Sequence of the Currently Utilized Patient Monitor Alarms?

OBJECTIVE: Our previous studies have shown that "code blue" events can be predicted by SuperAlarm patterns that are multivariate combinations of monitor alarms and laboratory test results cooccurring frequently preceding the events but rarely among control patients. Deploying these patterns to the monitor data streams can generate SuperAlarm sequences. The objective of this study is to test the hypothesis that SuperAlarm sequences may contain more predictive sequential patterns than monitor alarms sequences. METHODS: Monitor alarms and laboratory test results are extracted from a total of 254 adult coded and 2213 control patients. The training dataset is composed of subsequences that are sampled from complete sequences and then further represented as fixed-dimensional vectors by the term frequency inverse document frequency method. The information gain technique and weighted support vector machine are adopted to select the most relevant features and train a classifier to differentiate sequences between coded patients and control patients. Performances are assessed based on an independent dataset using three metrics: sensitivity of lead time (Sen L @T), alarm frequency reduction rate (AFRR), and work-up to detection ratio (WDR). RESULTS: The performance of 12-h-long sequences of SuperAlarm can yield a Sen L@2 of 93.33%, an AFRR of 87.28%, and a WDR of 3.01. At an AFRR = 87.28%, Sen L@2 for raw alarm sequences and discretized alarm sequences are 73.33% and 70.19%, respectively. At a WDR = 3.01, Sen L@2 are 49.88% and 43.33%. CONCLUSION AND SIGNIFICANCE: The results demonstrate that SuperAlarm sequences indeed outperform monitor alarm sequences and suggest that one can focus on sequential patterns from SuperAlarm sequences to develop more precise patient monitoring solutions.