Hospital Soundscapes

Abstract

Every patient admitted to a hospital carries an invisible vulnerability. Pain, fatigue, and fear strip away the ordinary defences through which healthy people filter unwanted stimulation. In that condition, sound becomes something more than noise. It becomes part of the experience of being ill, and part of the experience of healing.
This thesis examines the acoustic environments of four clinical departments — Emergency, Intensive Care Unit, Oncology, and Haematology — across four Dutch hospitals. Its central argument is that hospital sound cannot be adequately understood through decibels alone. The study proposes and demonstrates an integrated approach that combines psychoacoustic metrics derived from the Zwicker model with ISO 12913-based perception surveys, to characterise not only how loud a department is, but how it is perceived.
Data were collected through 124 calibrated acoustic measurements and 86 in-situ perception surveys, administered simultaneously to staff and patients, enabling direct pairing of objective and subjective datasets. The findings show that hospital departments differ not only in volume but in acoustic profile. The Emergency department recorded the highest values across every dimension measured and was perceived as chaotic and acoustically inappropriate. Haematology achieved the quietest and most acoustically favourable profile, an outcome attributable to private room enclosure and soft-close hardware rather than reduced staffing or patient activity. The Intensive Care Unit and Oncology occupied intermediate positions with distinctly different acoustic characters: the ICU registered lower mean annoyance overall yet was perceived as persistently harsh, owing to the tonal character of monitoring alarms reflected in the highest tonality values of any department; Oncology carried a higher mean annoyance distributed more evenly across equipment and environmental sources, rendering it perceptually more tolerable in quality despite the comparable sound level.
The strongest statistical relationship identified is between equivalent sound level and perceived appropriateness, with a Spearman correlation of -0.94, p < 0.01. This finding reinforces the core argument: in clinical environments, how appropriate a soundscape feels is more strongly predicted by its acoustic character than by its level.
The ISO 12913 framework demonstrated diagnostic utility, separating departments on the soundscape circumplex and integrating objective with perceptual data. Method A proved feasible in active clinical settings. Three adaptations are identified as necessary for hospital research specifically: a proxy or observational pathway for patients unable to complete the perception questionnaire; longitudinal sampling in place of single-session surveys, since hospital acoustic exposure is sustained rather than momentary; and explicit treatment of clinical role as a perceptual moderator rather than untreated background context.
The thesis translates these findings into department-specific intervention priorities and a business case linking acoustic quality to patient experience, staff retention, alarm safety, and clinical recovery. The evidence points toward a model of soundscape-informed design in which psychoacoustic profiling, alongside conventional noise measurement, becomes a standard component of hospital acoustic assessment.
Hospitals will never be silent. Nor should they be.