Sound intensity and noise evaluation in a critical care unit

Nancy Lawson, Kim Thompson, Gabrielle Saunders, Jaya Saiz, Jeannette Richardson, Deborah Brown, Naomi Ince, Marc Caldwell, Diana Pope

Research output: Contribution to journalArticlepeer-review

77 Scopus citations


Background: Sound levels in intensive care units can be high. Unfortunately, high levels of sound tend to result in poor sleep quality, which leads to slower healing, poorer immune response, and decreased cognitive function. Objectives To measure sound levels to which patients in intensive care units are typically exposed. Methods Peak sound pressure levels of alarms on medical devices set at different output levels were measured. Additionally, ambient sound pressure levels for durations of 10 to 24 hours were measured on 12 occasions in patients' rooms in the intensive care unit. Results: Peak levels of equipment alarms measured inside a patient's room were high, and increased as the setting of the alarm level increased. The levels of these alarms when measured in an adjacent room did not increase with alarm output level. Mean sound levels inside the patient's room were generally less than 45 dB(A), but peak levels were often greater than 85 dB(C). Closing the door of the adjacent room did not decrease these peak levels. Peak and mean levels did not differ systematically during 24 hours of measurement. Conclusions: High-intensity equipment alarms disturb patients' sleep but are critical in a medical emergency. However, nurses should not assume that raising the alarm output level will ensure that the alarm is audible from an adjacent room. Ambient noise measurements indicate high peak levels during both day and night.

Original languageEnglish (US)
Pages (from-to)e88-e98
JournalAmerican Journal of Critical Care
Issue number6
StatePublished - Nov 2010
Externally publishedYes

ASJC Scopus subject areas

  • Critical Care


Dive into the research topics of 'Sound intensity and noise evaluation in a critical care unit'. Together they form a unique fingerprint.

Cite this