Hospital Noise, Especially Alarms, Most Disruptive to Sleep

Hospital Noise, Especially Alarms, Most Disruptive to Sleep

Pauline Anderson

July 16, 2012 — A new study has quantified how disruptive typical hospital sounds are to the sleep of patients, and shows, perhaps not surprisingly, that electronic sounds that are intentionally alarming, such as phones and IV alarms, are more arousing than the sound of human voices, even at the same level.
Researchers report that sleep disruption is more likely during non-REM (rapid eye movement) stage 2 (the most abundant stage) than non-REM stage 3 (slow wave, the deepest stage). Arousal to sounds during REM sleep caused a greater and more sustained elevation in heart rate (HR), they also found.
These results point to the need for improved in-hospital acoustics to facilitate optimal care — something that will be increasingly in demand as the baby boomer generation ages and requires more institutional health services, the researchers suggest.
As it stands, surveys show that noise and bad sleep quality consistently top the list of complaints from hospitalized patients, said study author, Orfeu M. Buxton, PhD, assistant professor in the Division of Sleep Medicine, Harvard Medical School, and associate neuroscientist in the Department of Medicine, Brigham and Women's Hospital, Boston, Massachusetts.
"There's nothing quite as terrible as being sick or awaiting surgery or recovering and having to listen to someone else's TV or health issues being dealt with overnight," Dr. Buxton told Medscape Medical News.
Findings of this study were published online June 12 in the Annals of Internal Medicine.
Noise Stimuli
Disrupted sleep is associated with hypertension, cardiovascular and coronary heart disease, impaired immune function, elevated stress hormone responses, attention and memory deficits, and depressed mood.
For the study, researchers enrolled 12 healthy adults — 8 women, mean age 27 years, mean body mass index (BMI) 21.8 kg/m². Study participants slept at home on a regular schedule for at least 4 days before the study.
For the 3-day investigation, participants stayed at the Massachusetts General Hospital Sleep Laboratory. During the first night, researchers verified the absence of sleep disorders and established baseline sleep recordings. On the next 2 nights, participants were exposed to various noises and noise levels while researchers documented electroencephalogram (EEG) arousals and electrocardiogram HR accelerations.
The protocol involved 14 noise stimuli, which included "good" conversation (voices discussing a positive patient outcome), "bad" conversation (the same voices discussing a negative patient outcome), door closing, telephone ringing, toilet flushing, ice machine dispensing, IV alarm going off, laundry cart rolling, jet roaring overhead, and traffic moving.
The sounds came from 4 overhead speakers placed about 7 feet from the bed. Once the patient was in a steady sleep stage, a 10-second sound was played — starting at the level of about a whisper and increasing by 5 decibels — until sleep was disrupted by an arousal, the sleep stage changed, or the maximum exposure level — about 70 decibels, or the equivalent of yelling — was reached.
"If the subject didn't have an arousal or changed stage or wake up, then we would wait 30 seconds, or 1 more epoch, and present another noise," said Dr. Buxton.
Louder Sounds
As expected, the study showed that louder sounds were more apt to cause sleep disruption. It also found that the most arousing noises were such things as a phone ringing and an IV alarm going off. Within the lowest tested ranges, these sounds produced sleep disruptions more than 50% of the time.
"We characterize those alarm noises as either intentionally alerting or intentionally obnoxious," said Dr. Buxton. In the real hospital environment, he said, such alarms often signal nothing more than the patient rolling over and accidentally setting off the bell.
"The amazing thing to me is that those sounds are so obnoxious that even when you control for the peak level of the sound, the peak decibel level in that 10 seconds, the intentionally obnoxious noises are still alerting even at the level of a whisper."
Conversations and voice paging were equally highly alerting. Sounds of ice machines and laundry carts were arousing at relatively low levels.
Sounds that come from outside the hospital — for example, jets taking off and street noises — were the least arousing among the stimuli, they found.
The researchers found unique arousal probability profiles for each sleep stage — REM, non-REM2 (or N2), and non-REM3 (or N3).
"What was really exciting for us was to see [that] at the level of speech, say 50 or 55 decibels across those 14 sounds, even controlling for the peak level, there's a huge dispersion in the arousal probability," said Dr. Buxton.
"If you look at deep sleep (non-REM3), phones and IV alarms will wake you up more than 90% of the time, voices will wake you 80% of the time, but the sounds of traffic are only disrupting sleep 10 or 15% of the time or less," Dr. Buxton added. "So there’s a 90% difference in the probability of arousal controlling for the peak of the sound. This suggests to us that the brain during non-REM is processing something about the content of the noise."
But when participants moved to REM sleep, that discrimination disappears, said Dr. Buxton. "There's much less dispersion in REM, as if you're no longer paying as much attention to the content" of the noises.
Cycle in Sleep
People cycle between REM and non-REM sleep about every 90 minutes, with REM sleep returning at the end of the sleep period, added Dr. Buxton. "So in the morning, we're relatively less able to discriminate amongst sounds and have a relatively low probability of arousal. A phone in the beginning of the night is more likely to wake you up than the same phone ringing in the morning."
Arousal effects of noise on sleep included HR elevations — even when disruptions were brief. The stage in which the arousal occurred substantially predicted the magnitude of the HR increase (P < .0001).
"We demonstrate that evoked arousals elicit HR acceleration from all stages of sleep, but a greater magnitude (10 beats/min) and faster onset of HR accelerations from REM, with lesser magnitude and less rapid accelerations in stages N2 and N3,” the authors write.
Baseline (pre-arousal) HR did not predict the magnitude of the response (P = .94). The study night was not significant (P = .83), reflecting lack of habituation of the electrocardiogram HR response, said the authors.
Heart rate effects might be particularly relevant to the critical care settings of a hospital. ICUs, said Dr. Buxton, "are the most heavily alarmed, panicked places, where everyone is hooked up to so many different things that can alarm at any one point." He noted that in real life, hospital patients are likely exposed to more than one noise at a time — and even more frequently than was the case in this study.
The researchers recommend greater focus on a protective sleep environment. Possible changes include designing more acoustically sound facilities, improving alerting technology (eg, central monitoring stations), providing less disruptive night care routines, and educating healthcare personnel on the negative effect of noise on patient health.
Such measures do work. Dr. Buxton cited one study showing that demand for sedative medications was reduced by 49% (P = .0041) and by 62% in patients over 64 years of age (P = .005) through such steps as putting ice machines behind closed doors, playing a lullaby in the evening, dimming the lights, installing a sound meter in the hallways and alerting staff when their voices were raised, and doing medication checks at the same time.
Informative Study
Invited to comment, Michael Decker, PhD, associate professor at Georgia State University, Atlanta, Georgia, and spokesperson for the American Academy of Sleep Medicine, said he found the study design intriguing and the results informative with regard to the physical impact of sleep disruption.
"If we are chronically interrupted throughout the night by noise and our heart rate is accelerated and our blood pressure shoots up, that is not necessarily a good thing for the healing process," he said.
"Intuitively, we've known that loud noise is bad for patients trying to sleep, but I think this is one of the first studies to define just how bad it can be."
The impact of in-hospital sleep deprivation may be more severe in older patients, said Dr. Decker. He noted that the amount of slow wave sleep (non-REM3) decreases with age, and that older patients are much more vulnerable to being aroused by environmental noises.
Dr. Decker pointed out that systems that are designed to help keep patients safe at night — IV monitors and heart rate alarms — are the same systems that wake them up from sleep and can impair the sleep process. "We have to think about how to design alarms and how to alert medical staff to problems" while being minimally disruptive, he said.
Ann Intern Med. Published online June 12, 2012. Abstract

Medscape Medical News © 2012 WebMD, LLC
Send comments and news tips to news@medscape.net.

-->