An ear-ly intervention to reduce delirium

Where: ICUs around the world.

The issue: Reducing sleep disruption for ICU patients and incident delirium.


ICU patients commonly face sleep disruption, which is associated with delirium and cognitive impairment. Providing earplugs to these patients may be a low-cost way to mollify hospital noise and abate arousal during sleep, potentially cutting patients' risk of developing delirium and incurring its costly consequences.

To further explore earplugs as an intervention, Edward Litton, MBChB, MSc, an intensive care specialist at St John of God Subiaco Hospital in Perth, Western Australia, led a systematic review and meta-analysis of 9 studies published between 2009 and 2015.

How it works

Earplugs likely reduce delirium risk by abating noise and allowing patients to get to sleep quicker and stay asleep longer, Dr. Litton said, noting that the functional MRI scans of people who are severely sleep-deprived show similar patterns to scans of those who are delirious. “If you're not being woken constantly and sleep-deprived, then you're less likely to be confused.... Most importantly, the degree of noise abatement doesn't have to be large to stop those spikes that cause the arousals, and it's the number of arousals from sleep that's probably the major issue.”

The studies in Dr. Litton's analysis used a variety of earplugs, though not all of them are well described. “There is, in fact, a lot of variety in earplugs, and you can get all sorts of different sizings,” he said. “But the more complex you make the intervention, the less likely it's going to be implemented at the [point of care] by nurses and clinicians.”


The analysis included 5 randomized controlled trials with between 12 and 373 participants and 4 nonrandomized, interventional studies with between 100 and 338 participants. In 5 studies reporting incident delirium, earplugs were associated with a 41% decreased risk of delirium. In 4 studies that reported mortality, earplug placement was associated with a 23% reduced risk of hospital mortality, although this was not a statistically significant decrease. Patients were generally adherent to the earplugs—they were better tolerated than eyeshades—and no studies reported any safety concerns.

To Dr. Litton, it makes intuitive sense that if patients sleep better, they're probably likely to heal better. “There are studies dating back to the 1960s where they report earplug use, so it's not a new concept, but it doesn't seem to have gained a lot of traction till recently, and that does surprise me,” he said. “If the point estimate for the odds ratio of reduction in delirium was genuinely anywhere near what we found in our systematic review, you'd have to think that that is 1 of the most cost-effective interventions that could ever be undertaken in the hospital setting.”

Although the analysis found only a nonsignificant reduction in mortality, the results may demonstrate a causal chain that ultimately results in fewer deaths, said Dr. Litton. “There's no doubt the effect size must be small, but it's plausible that if earplugs do generally reduce the incidence and/or severity of delirium in at-risk patients, then that might also be associated with a survival advantage,” as delirium is associated with increased mortality.


A limitation of the analysis is that although Dr. Litton and his team were able to pool studies, the risk of bias was high, and studies generally did not measure the degree of noise abatement on a scientific level.

An implementation challenge is that proper earplug placement requires some clinician education. “There's a bit of an art to putting earplugs in. It sounds silly, [and] it's not hard, but you do need a bit of training about how to place them properly,” said Dr. Litton.

Next steps

Dr. Litton's group recently completed a large point-prevalence study involving about 50 ICUs across Australia and New Zealand, observing noise levels and whether patients used earplugs. Preliminary results show that “basically no one gets earplugs, despite the fact that they're an intervention that is harmless and may be beneficial,” he said. “The question is, why are they not implemented if there is some evidence of benefit? And either it's because that evidence isn't widely known or it's not thought to be of sufficient quality to be implemented.”

Dr. Litton decided to tackle the latter issue by attempting to produce higher-quality evidence through a separate pilot study, which is almost complete, that tested an earplug intervention and addressed gaps in the literature in terms of actually measuring noise abatement and patient comfort. “There are foam, self-expanding, single-use, disposable, one-size-fits-all earplugs, and this is data that we have undertaken but have yet to publish, but it seems like you can get a one-size-fits-all [style] that produces noise abatement and is comfortable for patients,” he said.

Words of wisdom

“I could see a time in the future where earplugs are available in dispensers in the same way that hand hygiene dispensers with a hand wash are available at every bed space,” Dr. Litton said.