• Philippe Lacherez, Eunice Limin Seah, and Penelope Sanderson.
• School of Psychology, The University of Queensland, St. Lucia, Queensland, Australia.
• Hum Factors. 2007 Aug 1;49(4):637-45.

ObjectiveWe explore how accurately and quickly nurses can identify melodic medical equipment alarms when no mnemonics are used, when alarms may overlap, and when concurrent tasks are performed.BackgroundThe international standard IEC 60601-1-8 (International Electrotechnical Commission, 2005) has proposed simple melodies to distinguish seven alarm sources. Previous studies with nonmedical participants reveal poor learning of melodic alarms and persistent confusions between some of them. The effects of domain expertise, concurrent tasks, and alarm overlaps are unknown.MethodFourteen intensive care and general medical unit nurses learned the melodic alarms without mnemonics in two sessions on separate days. In the second half of Day 2 the nurses identified single alarms or pairs of alarms played in sequential, partially overlapping, or nearly completely overlapping configurations. For half the experimental blocks nurses performed a concurrent mental arithmetic task.ResultsNurses' learning was poor and was no better than the learning of nonnurses in a previous study. Nurses showed the previously noted confusions between alarms. Overlapping alarms were exceptionally difficult to identify. The concurrent task affected response time but not accuracy.ConclusionBecause of a failure of auditory stream segregation, the melodic alarms cannot be discriminated when they overlap. Directives to sequence the sounding of alarms in medical electrical equipment must be strictly adhered to, or the alarms must redesigned to support better auditory streaming.ApplicationActual or potential uses of this research include the implementation of IEC 60601-1-8 alarms in medical electrical equipment.

12 keywords...

hide…