Recommendation of New Medical Alarms Based on Audibility, Identifiability, and Detectability in a Randomized, Simulation-Based Study

Christopher Bennett, Roman Dudaryk, Nichole Crenshaw, Judy Edworthy, Richard McNeer

Research output: Contribution to journalArticle

Abstract

OBJECTIVES: Accurate and timely identification of existing audible medical alarms is not adequate in clinical settings. New alarms that are easily heard, quickly identifiable, and discernable from one another are indicated. The "auditory icons" (brief sounds that serve as metaphors for the events they represent) have been proposed as a replacement to the current international standard. The objective was to identify the best performing icons based on audibility and performance in a simulated clinical environment. DESIGN: Three sets of icon alarms were designed using empirical methods. Subjects participated in a series of clinical simulation experiments that examined the audibility, identification accuracy, and response time of each of these icon alarms. A statistical model that combined the outcomes was used to rank the alarms in overall efficacy. We constructed the "best" and "worst" performing sets based on this ranking and prospectively validated these sets in a subsequent experiment with a new subject sample. SETTING: Experiments were conducted in simulated ICU settings at the University of Miami. SUBJECTS: Medical trainees were recruited from a convenience sample of nursing students and anesthesia residents at the institution. INTERVENTIONS: In Experiment 1 (formative testing), subjects were exposed to one of the three sets of alarms; identical setting and instruments were used throughout. In Experiment 2 (summative testing), subjects were exposed to one of the two sets of alarms, assembled from the best and worst performing alarms from Experiment 1. MEASUREMENTS AND MAIN RESULTS: For each alarm, we determined the minimum sound level to reach audibility threshold in the presence of background clinical noise, identification accuracy (percentage), and response time (seconds). We enrolled 123 medical trainees and professionals for participation (78 with < 6 yr of training). We identified the best performing icon alarms for each category, which matched or exceeded the other candidate alarms in identification accuracy and response time. CONCLUSIONS: We propose a set of eight auditory icon alarms that were selected through formative testing and validated through summative testing for adoption by relevant regulatory bodies and medical device manufacturers.

Original languageEnglish (US)
Pages (from-to)1050-1057
Number of pages8
JournalCritical care medicine
Volume47
Issue number8
DOIs
StatePublished - Aug 1 2019

Fingerprint

Reaction Time
Metaphor
Nursing Students
Statistical Models
Noise
cyhalothrin
Anesthesia
Equipment and Supplies

ASJC Scopus subject areas

  • Critical Care and Intensive Care Medicine

Cite this

Recommendation of New Medical Alarms Based on Audibility, Identifiability, and Detectability in a Randomized, Simulation-Based Study. / Bennett, Christopher; Dudaryk, Roman; Crenshaw, Nichole; Edworthy, Judy; McNeer, Richard.

In: Critical care medicine, Vol. 47, No. 8, 01.08.2019, p. 1050-1057.

Research output: Contribution to journalArticle

@article{a38dadee8c7f4c4fa070a7047e9b6b0f,
title = "Recommendation of New Medical Alarms Based on Audibility, Identifiability, and Detectability in a Randomized, Simulation-Based Study",
abstract = "OBJECTIVES: Accurate and timely identification of existing audible medical alarms is not adequate in clinical settings. New alarms that are easily heard, quickly identifiable, and discernable from one another are indicated. The {"}auditory icons{"} (brief sounds that serve as metaphors for the events they represent) have been proposed as a replacement to the current international standard. The objective was to identify the best performing icons based on audibility and performance in a simulated clinical environment. DESIGN: Three sets of icon alarms were designed using empirical methods. Subjects participated in a series of clinical simulation experiments that examined the audibility, identification accuracy, and response time of each of these icon alarms. A statistical model that combined the outcomes was used to rank the alarms in overall efficacy. We constructed the {"}best{"} and {"}worst{"} performing sets based on this ranking and prospectively validated these sets in a subsequent experiment with a new subject sample. SETTING: Experiments were conducted in simulated ICU settings at the University of Miami. SUBJECTS: Medical trainees were recruited from a convenience sample of nursing students and anesthesia residents at the institution. INTERVENTIONS: In Experiment 1 (formative testing), subjects were exposed to one of the three sets of alarms; identical setting and instruments were used throughout. In Experiment 2 (summative testing), subjects were exposed to one of the two sets of alarms, assembled from the best and worst performing alarms from Experiment 1. MEASUREMENTS AND MAIN RESULTS: For each alarm, we determined the minimum sound level to reach audibility threshold in the presence of background clinical noise, identification accuracy (percentage), and response time (seconds). We enrolled 123 medical trainees and professionals for participation (78 with < 6 yr of training). We identified the best performing icon alarms for each category, which matched or exceeded the other candidate alarms in identification accuracy and response time. CONCLUSIONS: We propose a set of eight auditory icon alarms that were selected through formative testing and validated through summative testing for adoption by relevant regulatory bodies and medical device manufacturers.",
author = "Christopher Bennett and Roman Dudaryk and Nichole Crenshaw and Judy Edworthy and Richard McNeer",
year = "2019",
month = "8",
day = "1",
doi = "10.1097/CCM.0000000000003802",
language = "English (US)",
volume = "47",
pages = "1050--1057",
journal = "Critical Care Medicine",
issn = "0090-3493",
publisher = "Lippincott Williams and Wilkins",
number = "8",

}

TY - JOUR

T1 - Recommendation of New Medical Alarms Based on Audibility, Identifiability, and Detectability in a Randomized, Simulation-Based Study

AU - Bennett, Christopher

AU - Dudaryk, Roman

AU - Crenshaw, Nichole

AU - Edworthy, Judy

AU - McNeer, Richard

PY - 2019/8/1

Y1 - 2019/8/1

N2 - OBJECTIVES: Accurate and timely identification of existing audible medical alarms is not adequate in clinical settings. New alarms that are easily heard, quickly identifiable, and discernable from one another are indicated. The "auditory icons" (brief sounds that serve as metaphors for the events they represent) have been proposed as a replacement to the current international standard. The objective was to identify the best performing icons based on audibility and performance in a simulated clinical environment. DESIGN: Three sets of icon alarms were designed using empirical methods. Subjects participated in a series of clinical simulation experiments that examined the audibility, identification accuracy, and response time of each of these icon alarms. A statistical model that combined the outcomes was used to rank the alarms in overall efficacy. We constructed the "best" and "worst" performing sets based on this ranking and prospectively validated these sets in a subsequent experiment with a new subject sample. SETTING: Experiments were conducted in simulated ICU settings at the University of Miami. SUBJECTS: Medical trainees were recruited from a convenience sample of nursing students and anesthesia residents at the institution. INTERVENTIONS: In Experiment 1 (formative testing), subjects were exposed to one of the three sets of alarms; identical setting and instruments were used throughout. In Experiment 2 (summative testing), subjects were exposed to one of the two sets of alarms, assembled from the best and worst performing alarms from Experiment 1. MEASUREMENTS AND MAIN RESULTS: For each alarm, we determined the minimum sound level to reach audibility threshold in the presence of background clinical noise, identification accuracy (percentage), and response time (seconds). We enrolled 123 medical trainees and professionals for participation (78 with < 6 yr of training). We identified the best performing icon alarms for each category, which matched or exceeded the other candidate alarms in identification accuracy and response time. CONCLUSIONS: We propose a set of eight auditory icon alarms that were selected through formative testing and validated through summative testing for adoption by relevant regulatory bodies and medical device manufacturers.

AB - OBJECTIVES: Accurate and timely identification of existing audible medical alarms is not adequate in clinical settings. New alarms that are easily heard, quickly identifiable, and discernable from one another are indicated. The "auditory icons" (brief sounds that serve as metaphors for the events they represent) have been proposed as a replacement to the current international standard. The objective was to identify the best performing icons based on audibility and performance in a simulated clinical environment. DESIGN: Three sets of icon alarms were designed using empirical methods. Subjects participated in a series of clinical simulation experiments that examined the audibility, identification accuracy, and response time of each of these icon alarms. A statistical model that combined the outcomes was used to rank the alarms in overall efficacy. We constructed the "best" and "worst" performing sets based on this ranking and prospectively validated these sets in a subsequent experiment with a new subject sample. SETTING: Experiments were conducted in simulated ICU settings at the University of Miami. SUBJECTS: Medical trainees were recruited from a convenience sample of nursing students and anesthesia residents at the institution. INTERVENTIONS: In Experiment 1 (formative testing), subjects were exposed to one of the three sets of alarms; identical setting and instruments were used throughout. In Experiment 2 (summative testing), subjects were exposed to one of the two sets of alarms, assembled from the best and worst performing alarms from Experiment 1. MEASUREMENTS AND MAIN RESULTS: For each alarm, we determined the minimum sound level to reach audibility threshold in the presence of background clinical noise, identification accuracy (percentage), and response time (seconds). We enrolled 123 medical trainees and professionals for participation (78 with < 6 yr of training). We identified the best performing icon alarms for each category, which matched or exceeded the other candidate alarms in identification accuracy and response time. CONCLUSIONS: We propose a set of eight auditory icon alarms that were selected through formative testing and validated through summative testing for adoption by relevant regulatory bodies and medical device manufacturers.

UR - http://www.scopus.com/inward/record.url?scp=85069889428&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85069889428&partnerID=8YFLogxK

U2 - 10.1097/CCM.0000000000003802

DO - 10.1097/CCM.0000000000003802

M3 - Article

VL - 47

SP - 1050

EP - 1057

JO - Critical Care Medicine

JF - Critical Care Medicine

SN - 0090-3493

IS - 8

ER -