Using timbre to improve performance of larger auditory alarm sets

Abstract

Identifiability and perceived urgency were compared for two sets of alarms in a healthcare inpatient setting. One contained currently used alarms where possible, with new sounds added as needed. The other was designed together, was more heterogenous, used timbre to encode intended similarities and explicitly encoded intended urgency across the set. Twenty nurses reported the identity and perceived urgency of the sounds in each set. Participants correctly identified the sound (0.89 vs. 0.77) and alarm category (0.93 vs. 0.82) more often in the new set than in the baseline set. In addition, multiple sounds in the new set were more identifiable. The new sounds also had a larger range of perceived urgency and better urgency match. The results indicate that timbre is well-suited to encode alarm groupings in larger alarm sets and that this, along with increased heterogeneity and explicit urgency mapping, improves alarm set performance.

Practitioner summary: Clinical alarms are frequently misidentified. We found that making alarms more acoustically rich, using timbre to convey alarm groups, and explicitly encoding intended urgency improved identifiability and urgency match. These findings can be used to improve alarm performance across all safety-critical industries.

Keywords:

• Alarms and warnings
• human-machine systems
• patient safety
• health care ergonomics
• sound and noise

Acknowledgements

We would like to thank the nurses at The Ohio State University Wexner Medical Centre (OSUWMC) for their participation in this study. We also would like to thank the members of the OSUWMC Alarm Taskforce, and in particular Traci Mignery, Todd Yamokoski, Lyn Roush, Sharon Steingass, Brenda Vermillion, Jennifer MacDermott, Mike Burke and Nancy Carson, for sharing their clinical perspective on the design of the alarms and their assistance in recruiting participants. Finally, we would like to thank Steve Simula of the Lextant Corporation for his assistance in designing the alarm sounds.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This research was supported by the Institute for the Design of Environments Aligned for Patient Safety (IDEA4PS) at Ohio State University which is sponsored by the Agency for Healthcare Research and Quality [P30HS024379]. The authors’ views do not necessarily represent the views of AHRQ.