Are noise reduction interventions effective in adult ward settings? A systematic review and meta analysis.

OBJECTIVE: High levels of environmental noise in hospitals disturbs sleep. We aimed to identify, critically appraise and summarise primary research that reports studies that tested interventions to reduce night-time noise levels in ward-settings.

DESIGN: Systematic review and meta-analysis in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement.

DATA SOURCES: Key health-related and specialist acoustic databases (CINAHL, EMBASE, MEDLINE, The Cochrane Library, PsycINFO, NHS Evidence, Knovel, Journal of the Audio Engineering Society, Journal of the Acoustical Society of America, IEEE Xplore Digital Library, Acta Acustica and Acoustics in Practice) were searched from their inception to April 2017, with no language restrictions.

REVIEW METHODS: Experimental, quasi-experimental and observational study designs assessing the effect of noise reduction interventions on patient outcomes and/or environmental noise levels were included. Two reviewers independently conducted a quality appraisal using a published framework.

RESULTS: In total, 834 records were identified with nine studies meeting inclusion criteria. Quality appraisal showed that the level of evidence was generally weak. A range of noise reduction interventions were identified: one study implemented a single intervention, whilst the remainder were complex, multi-faceted interventions. Findings from individual studies showed mixed results but preliminary evidence suggests that noise reduction interventions can reduce environmental noise levels in ward settings and improve patients' sleep ratings. Quantitative data from 6 studies were pooled. A random effects meta-analysis determined that a synthesised estimate for the standardised mean difference in total hours sleep (no intervention - intervention) was -0.11 h (95% CI -0.46 to 0.25 h; p = 0.556), with moderate statistical heterogeneity. A random effects meta-analysis determined that a synthesised estimate for the standardised mean difference in awakenings per night (no intervention - intervention) was 0.05 (95% CI -0.20 to 0.29; p = 0.715), with negligible statistical heterogeneity. A random effects meta-analysis determined that a synthesised estimate for the odds ratio for disturbed nights (no intervention: intervention) was 0.75 (95% CI 0.55 to 1.01; p = 0.059), with low statistical heterogeneity.

CONCLUSIONS: Individual studies show that noise reduction interventions are feasible in ward settings and suggest they have potential to improve patients' in-hospital sleep experiences. However meta-analyses show insufficient evidence to support the use of such interventions at present. There is a lack of appropriately designed studies to test intervention effectiveness. Robust studies are required to identify the most effective interventions to address this significant and ubiquitous problem.

WHAT IS KNOWN ABOUT THE TOPIC: WHAT THIS PAPER ADDS.