This protocol describes a standardized cross-sectional approach for measuring environmental noise across hospital functional zones. Application in a tertiary hospital identified nurses’ stations as the noisiest locations, with higher noise levels associated with visitor count and departmental bed capacity.
Research Article
Spatial Assessment of Environmental Noise Across Hospital Functional Zones: A Cross-Sectional Observational Study in a Tertiary Hospital
In This Article
Summary
Abstract
Hospital environments comprise complex acoustic settings generated by routine clinical activities, medical equipment, staff communication, patient movement, and visitor traffic. Characterizing environmental noise across different hospital functional zones may help identify areas with consistently elevated acoustic exposure and inform targeted noise-management strategies. We hypothesized that environmental noise would differ among hospital functional zones and be associated with selected operational characteristics. A cross-sectional observational study was conducted across 15 inpatient departments of a tertiary teaching hospital using a standardized hierarchical sampling framework. Environmental noise was measured within four predefined functional zones (ward, corridor, nurses’ station, and patient activity area) using a calibrated Class 1 integrating sound level meter. Equivalent continuous A-weighted sound pressure level (LAeq) measurements were obtained from repeated daytime recordings (09:00–19:00), together with maximum A-weighted sound pressure level (LAFmax) and peak C-weighted sound pressure level (LCpeak). Linear mixed-effects models were used to compare environmental noise among functional zones while accounting for repeated observations nested within hospital departments. Exploratory mixed-effects regression evaluated associations between environmental noise and selected operational characteristics. A total of 1,620 acoustic observations were analyzed. Nurses’ stations demonstrated the highest adjusted environmental noise levels (estimated marginal mean, 66.90 dBA), followed by corridors (63.97 dBA), patient activity areas (61.38 dBA), and wards (60.38 dBA). All pairwise comparisons remained statistically significant following Tukey adjustment (all P < 0.001). Visitor count and departmental bed capacity were independently associated with higher LAeq in the parsimonious mixed-effects model. Environmental noise levels consistently exceeded commonly referenced World Health Organization, EN ISO 16032, and GB3096-2022 reference acoustic values during routine daytime operations. Environmental noise varied significantly across hospital functional zones, with nurses’ stations representing the principal areas of elevated acoustic exposure. These findings support targeted acoustic monitoring and operational noise-management strategies while providing a reproducible framework for future hospital environmental acoustic research.
Introduction
Hospital environments are acoustically complex settings in which sound is continuously generated by medical equipment, physiological monitoring systems, staff communication, patient movement, alarms, ventilation systems, and routine clinical activities. These diverse sound sources create dynamic healthcare soundscapes that vary across time and location within hospitals and may influence both patient experience and staff working conditions. Persistent or excessive environmental noise has been associated with sleep disturbance, impaired communication, increased cognitive workload, physiological stress responses, and reduced environmental comfort in healthcare settings1,2,3,4,5,6,7. Consequently, hospital noise has been recognized as an important environmental factor that may be amenable to targeted monitoring and mitigation strategies.
International organizations have proposed recommendations for maintaining relatively low sound levels in patient-care environments. The World Health Organization (WHO) recommends background sound levels of approximately 35 dB(A) during the daytime and 30 dB(A) at night in hospital patient rooms, whereas other national and international standards provide guidance for building acoustics, environmental noise assessment, or healthcare facility design. Because these recommendations differ in their intended purpose, measurement metrics, averaging periods, and scope, comparisons with operational hospital measurements should be interpreted cautiously. Nevertheless, environmental surveys of hospitals worldwide have consistently reported daytime equivalent sound levels that substantially exceed recommended background values, frequently ranging from 55 to 75 dB(A) in inpatient wards, corridors, emergency departments, and intensive care units8,9,10,11.
Accumulating evidence suggests that elevated hospital noise may adversely affect both patients and healthcare personnel. Among patients, excessive environmental noise has been associated with sleep disruption, reduced restfulness, physiological stress responses, diminished patient satisfaction, and, in critically ill populations, an increased risk of neurocognitive disturbances such as delirium12,13,14,15,16,17,18. For healthcare professionals, elevated background noise may reduce speech intelligibility, increase cognitive workload, interfere with clinical communication, and contribute to alarm fatigue and occupational stress, particularly in high-intensity clinical environments19,20,21,22,23. Although these observational findings do not establish causality, they underscore the importance of understanding the spatial distribution and operational characteristics associated with hospital acoustic environments.
Despite increasing interest in healthcare acoustics, many previous investigations have focused on individual wards, intensive care units, or hospital-wide average sound levels. Such approaches provide limited information regarding spatial heterogeneity within hospitals and often rely solely on time-averaged acoustic metrics obtained from a limited number of monitoring locations. More recent studies have emphasized that hospital soundscapes comprise a mixture of continuous background noise and transient impulsive events and have advocated more spatially resolved monitoring strategies capable of identifying operational hotspots and distinguishing among different sources of acoustic exposure. Contemporary approaches incorporating clustering analysis, sound-source characterization, and soundscape assessment have further highlighted the importance of evaluating environmental noise within its functional and operational context24,25,26,27,28.
To address these gaps, the present study performed a comprehensive environmental acoustic assessment across 15 inpatient departments within a tertiary teaching hospital. Environmental noise was systematically measured across four standardized functional zones—departmental wards, ward corridors, nurses’ stations, and patient activity areas—using repeated daytime measurements obtained under routine clinical operating conditions. In addition to characterizing spatial variability in equivalent continuous A-weighted sound pressure level (LAeq), the study explored associations between environmental noise and selected operational characteristics of hospital departments. We hypothesized that environmental noise exposure would differ significantly among hospital functional zones and that operational characteristics related to departmental activity would be associated with higher environmental sound levels. By providing a detailed spatial characterization of hospital acoustic environments, this study aims to identify priority areas for environmental monitoring and inform future evidence-based noise-management strategies within healthcare facilities.
Protocol
The study was approved by the local institutional ethics committee (Approval No. HEBMU-AUD/2025/017) and was conducted in accordance with the ethical principles of the 1964 Declaration of Helsinki and its subsequent amendments or comparable ethical standards. This study involved environmental acoustic monitoring only and did not include direct patient participation, clinical intervention, collection of biological samples, or acquisition of identifiable personal information. Because measurements were conducted exclusively at the environmental level under routine hospital operating conditions, the requirement for written informed consent was waived. Hospital personnel were informed of the observational nature of the study, and all collected data were anonymized and handled in accordance with institutional data-protection policies.
Study Design
This study was conducted as a cross-sectional observational assessment of environmental noise within a tertiary hospital setting. The primary objective was to measure and compare LAeq across fifteen inpatient departments and four standardized hospital functional zones to characterize the spatial distribution and variability of acoustic exposure within the healthcare environment. Maximum A-weighted sound pressure level (LAFmax) and peak C-weighted sound pressure level (LCpeak) were additionally recorded to characterize transient and impulsive acoustic events occurring during routine hospital operations. Measured sound levels were interpreted in relation to commonly referenced international and regional acoustic guidance documents and standards, including the WHO Community Noise Guidelines and Environmental Noise Guidelines, EN ISO 16032, PN-B-02151, and the Chinese National Standard GB3096-2022, where applicable. Because these documents differ in their intended purpose, measurement metrics, averaging periods, environmental settings, and scope, comparisons were interpreted descriptively rather than as formal compliance assessments.
The study focused on practical environmental acoustic assessment under routine hospital operating conditions rather than formal architectural acoustic certification testing. The investigation was designed to identify spatial variability in hospital noise exposure and to explore associations between environmental noise and objectively recorded operational characteristics, including departmental bed capacity, occupied beds, staff count, visitor count, alarm events, conversation events, room area, and room volume. Bed capacity, room area, and room volume were obtained from departmental administrative and engineering records before data collection. Occupied beds were recorded immediately before each acoustic measurement. During each 10-min acoustic recording, a trained observer simultaneously documented the number of staff members and visitors present within the predefined functional zone using direct visual counting. Alarm events were recorded as the total number of audible medical equipment alarm activations occurring during the same 10-min observation period, whereas conversation events were recorded as the total number of distinct conversational episodes involving patients, visitors, or healthcare personnel that were clearly audible within the measurement zone. Operational variables were recorded contemporaneously with each acoustic measurement using standardized case-report forms and identical counting procedures across all departments and functional zones. Observer training was completed before study initiation using standardized written instructions to ensure consistent implementation of the recording protocol. These operational variables were collected as objective descriptors of routine clinical activity and were analyzed as exploratory explanatory variables in the mixed-effects models. No attempt was made to infer causal relationships between operational variables and measured sound levels. We hypothesized that environmental noise exposure would vary significantly across hospital functional zones and that departmental operational characteristics related to patient occupancy, staff activity, visitor activity, and environmental configuration would be associated with higher environmental sound levels.
Environmental noise measurements were obtained using a multistage hierarchical sampling framework consisting of repeated observations nested within hospital departments, functional zones, predefined sampling locations (entrance, middle, and end), standardized daytime observation periods, and technical measurement replicates. Standardized observation periods were defined as morning (09:00–11:00), midday (13:00–15:00), and afternoon (16:00–18:00), representing routine phases of daytime hospital activity while avoiding atypical transitional periods. Within each observation period, environmental noise was recorded using three consecutive 10-min technical replicates at each predefined sampling location, and the arithmetic mean of the three measurements was used for subsequent analyses. This hierarchical sampling design enabled characterization of both spatial and temporal variability in environmental noise while minimizing the influence of short-term fluctuations in routine hospital activity and improving the representativeness of measurements across different clinical environments.
The study is reported in accordance with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) recommendations for cross-sectional observational studies. The protocol and reporting were additionally prepared in accordance with the methodological and reproducibility requirements for observational research articles published in the Journal of Visualized Experiments (JoVE), including comprehensive documentation of sampling procedures, instrumentation, calibration records, measurement methodology, quality-control procedures, data processing, statistical analyses, and the availability of supporting source data.
Study Setting
The study was carried out at a large tertiary teaching hospital providing comprehensive medical and surgical services. Noise measurements were performed between March and August 2025 during routine weekday hospital operations. Data collection was restricted to weekdays to minimize variability associated with weekend staffing patterns and clinical activity. Data collection was conducted during daytime operational periods between 09:00 and 19:00, capturing representative acoustic conditions associated with routine patient care, staff movement, medical equipment operation, and visitor activity. Measurements were performed during three predefined operational periods (morning, midday, and afternoon) to account for temporal variation in routine hospital activities while maintaining a standardized daytime observation window. Measurements were obtained under routine clinical working conditions without interrupting or modifying patient care, staffing, or environmental controls, thereby reflecting real-world hospital acoustic conditions.
Departments and Sampling Framework
Fifteen inpatient departments representing a broad range of medical and surgical specialties were included in the study using a purposive sampling approach. Departments were selected to capture variability in clinical activity, patient volume, staffing patterns, departmental size, and spatial organization within the hospital environment.
The included departments were Trauma Orthopedics, Congenital Heart Disease Center, Vascular and Nerve Department I, Vascular and Nerve Department II, Intensive Care II Department, Parkinson’s Disease Department, Psychosomatic Medicine Department I, Otolaryngology–Cervicology Surgery, Knee Joint Department II, Hepatobiliary and Pancreatic Surgery, Gastroenterology Diagnosis and Treatment Department I, Gastroenterology Diagnosis and Treatment Department II, Urology Department, Breast and Thyroid Diagnosis and Treatment Center, and Pediatrics Department.
Within each department, environmental noise measurements were obtained from four standardized functional zones commonly evaluated in hospital acoustic surveys: the departmental ward, ward corridor, nurses’ station, and patient activity area. These locations were selected to represent distinct operational environments characterized by differing patterns of patient movement, staff communication, medical equipment use, visitor activity, and clinical workflow.
To improve spatial representativeness within each functional zone, three predefined sampling locations (entrance, middle, and end) were systematically evaluated. The entrance location was defined as the primary point of access to the functional zone, approximately 1–2 m inside the zone to avoid direct influence from adjacent areas. The middle location was positioned at the geometric center or principal activity area of the functional zone, representing routine operational conditions. The end location was established at the point farthest from the entrance while remaining within the same functional zone and accessible during routine clinical operations. Sampling locations were selected before data collection using a standardized protocol and maintained consistently across all departments. Where the physical configuration of a functional zone differed between departments, locations were selected according to these predefined spatial criteria rather than absolute distances to ensure methodological consistency. This multistage spatial sampling approach was adopted to better characterize acoustic variability within each operational area and to reduce the potential influence of localized sound sources associated with single-point measurements.
At each sampling location, environmental noise was recorded during three standardized observation periods (morning, midday, and afternoon). During each observation period, three consecutive technical replicate measurements were obtained using identical instrument settings. Consequently, the hierarchical sampling framework comprised repeated observations across 15 inpatient departments, four functional zones, three predefined sampling locations, three observation periods, and three technical replicates, yielding a total of 1,620 individual acoustic observations for analysis. This hierarchical sampling framework generated repeated observations across space and time, allowing characterization of both within-location and between-location variability in hospital environmental noise.
Departmental summary values reported in the manuscript were derived from these repeated measurements using predefined averaging procedures described below. The complete hierarchical dataset was retained for inferential statistical analyses as described in the Statistical Analysis section.
In addition to the inpatient departments, two representative public hospital areas—one indoor public activity zone and one outdoor public activity zone—were included to provide contextual environmental noise measurements outside direct patient-care environments. These public-area measurements were analyzed descriptively and were excluded from departmental statistical analyses, departmental summary statistics, and calculations of departmental measurement denominators.
Measurement Locations
Within each department, environmental noise measurements were obtained at four predefined functional zones commonly evaluated in hospital acoustic surveys: the departmental ward, ward corridor, nurses’ station, and patient activity area. These locations were selected to represent different patterns of clinical activity, patient movement, communication intensity, and equipment use within the hospital environment. The departmental ward represented general patient-care and bed areas; the ward corridor represented the primary circulation pathways for patients, staff, and equipment transport; the nurses’ station represented the central administrative and clinical coordination area; and the patient activity area represented designated spaces for patient movement, waiting, or interaction.
Within each functional zone, measurements were obtained at three standardized sampling locations (entrance, middle, and end), providing representative spatial coverage of each operational area. Sampling locations were selected to minimize the influence of localized sound sources while maintaining representative measurement positions within routine clinical environments. The microphone was positioned approximately 1.3 m above the floor and at least 1 m away from walls, large reflective surfaces, and major obstructions whenever feasible without interfering with routine clinical activities. The use of standardized functional zones and predefined sampling locations allowed consistent comparison of acoustic conditions across departments with differing clinical workflows and spatial layouts.
Noise Measurement Instrumentation
Environmental sound measurements were performed using a Class 1 precision integrating sound level meter compliant with IEC 61672-1:2013 and ANSI S1.4-2014 standards for precision environmental acoustic measurements. The instrument was configured using A-weighting [dB(A)] to approximate human auditory sensitivity and Slow (S) time weighting (1-second integration) to characterize routine environmental noise under operational hospital conditions. The primary acoustic parameter was the LAeq (dB[A]). In addition, the LAFmax and LCpeak were recorded during each measurement period to characterize transient and impulsive acoustic events occurring within the hospital environment. Measurements were conducted under routine hospital environmental conditions for observational acoustic assessment rather than formal architectural acoustic certification testing. Accordingly, the recorded acoustic parameters were interpreted as indicators of operational environmental noise exposure rather than engineering measures of building acoustic performance. Before and after each measurement session, the sound level meter was calibrated using a Class 1 acoustic calibrator generating 94 dB at 1 kHz. Calibration drift was maintained within ±0.5 dB throughout the study period. Measurements demonstrating post-calibration drift greater than ±0.5 dB were repeated following recalibration of the instrument. Calibration records were maintained throughout the study as part of the quality-control procedure.
Data Collection Procedure
Noise measurements were conducted during weekday daytime operational hours (09:00–19:00) to capture representative hospital acoustic conditions during routine clinical activity.
Within each functional zone, measurements were obtained sequentially at three predefined sampling locations (entrance, middle, and end). At each sampling location, observations were performed during three standardized operational periods (morning, midday, and afternoon). During each observation period, three consecutive technical replicate measurements were recorded using identical instrument settings. At each observation point, the microphone was positioned approximately 1.3 m above the floor and at least 1 m away from nearby walls or large reflective surfaces. Measurements were obtained during continuous 10-min recording intervals, and LAeq, LAFmax, and LCpeak were automatically calculated and stored by the sound level meter software for each recording interval.
During each observation period, operational characteristics of the measurement location, including staff count, visitor count, alarm events, and conversation events, were recorded using a standardized observation form. Departmental characteristics, including bed capacity, occupied beds, room area, and room volume, were obtained from hospital administrative records and verified before statistical analysis. To account for short-term temporal variability in hospital activity, measurements were repeated across the three predefined operational periods (morning, midday, and afternoon). For descriptive reporting, repeated technical replicate measurements were first averaged within each observation period, after which period-specific values were averaged to obtain representative values for each sampling location. Departmental and functional-zone summary statistics were subsequently derived using predefined averaging procedures, whereas the complete hierarchical dataset was retained for inferential statistical analyses. Ambient environmental conditions were monitored concurrently using a digital thermo-hygrometer to ensure relatively stable measurement conditions during data collection. Recorded environmental conditions ranged from 20°C–26°C for temperature and 40%–65% for relative humidity. One planned observation at the patient activity area of the Intensive Care II Department could not be completed because of temporary emergency-care restrictions during the measurement period. This observation was recorded as missing and excluded from subsequent analyses using a complete-case approach without data imputation.
Data Management and Quality Control
Measurement values were recorded manually at the time of acquisition and subsequently cross-verified against the digitally stored sound level meter recordings following each measurement session. All acoustic data were transferred to a secure electronic database and independently checked for transcription errors, missing values, duplicate records, and implausible observations before statistical analysis. Transient acoustic events unrelated to routine hospital operations, including temporary construction activity or external emergency sirens, were excluded when identified during monitoring. Two trained research assistants independently documented and verified all measurements to reduce observer-related recording errors. Any discrepancy greater than 1 dB between independently recorded values prompted re-verification of the original measurement and repeat assessment when necessary. Calibration records were reviewed for every measurement session, and only observations meeting the predefined calibration acceptance criterion (±0.5 dB drift) were retained for analysis.
Quality-control procedures also included verification of departmental identifiers, functional zones, sampling locations, observation periods, and technical replicate numbers to ensure consistency across the hierarchical dataset. Operational variables, including staff count, visitor count, alarm events, and conversation events, were cross-checked against standardized field-recording forms before database locking. One planned observation from the patient activity area of the Intensive Care II Department could not be completed because of temporary emergency-care restrictions and was recorded as missing. Because the proportion of missing data was minimal and unrelated to measurement quality, analyses were performed using a complete-case approach without data imputation. Because the study was designed as a practical environmental acoustic survey under real-world hospital operating conditions, formal engineering uncertainty analysis was beyond the scope of the investigation. Nevertheless, standardized instrumentation, routine calibration, repeated spatial and temporal measurements, technical replicate recordings, duplicate data verification, and predefined quality-control procedures were implemented to maximize measurement accuracy, consistency, and reproducibility.
Outcome Measures
The primary outcome of the study was the equivalent continuous environmental sound level (LAeq, dB[A]) measured across hospital departments and standardized functional zones. Secondary acoustic outcomes included the LAFmax and LCpeak, which were recorded during each measurement period to characterize transient and impulsive acoustic events within the hospital environment. Secondary outcomes included comparison of mean environmental noise levels between hospital departments and functional zones, identification of locations with relatively higher or lower acoustic exposure, assessment of the proportion of measurement points exceeding commonly referenced hospital environmental noise guidance values and standards, including WHO guidance values, EN ISO 16032, PN-B-02151, and the Chinese National Standard GB3096-2022, where applicable, and exploration of associations between environmental noise levels and objectively recorded operational characteristics, including departmental bed capacity, occupied beds, room area, room volume, staff count, visitor count, alarm events, and conversation events. Operational variables were evaluated as exploratory factors potentially associated with environmental noise exposure and were not interpreted as evidence of causal determinants of hospital acoustic conditions. The study was designed as an environmental acoustic assessment and did not include direct evaluation of patient clinical outcomes, occupational health outcomes, or physiological responses related to noise exposure. Accordingly, all findings should be interpreted as environmental acoustic observations rather than measures of patient safety, clinical effectiveness, or occupational health outcomes.
Statistical Analysis
All measurement data were compiled using Microsoft Excel and analyzed using IBM SPSS Statistics version 29.0 and R statistical software (version 4.3.3). Linear mixed-effects analyses were performed using the lme4, lmerTest, and emmeans packages. The primary analytical outcome was the LAeq. LAFmax and LCpeak were summarized descriptively and reported as secondary acoustic outcomes. Descriptive statistics, including the mean, standard deviation (SD), median, interquartile range (IQR), minimum, and maximum values, were calculated for hospital departments and functional zones. Continuous sound measurements are presented as mean ± SD where appropriate. Because environmental noise measurements were obtained using a hierarchical sampling framework, repeated observations were nested within sampling locations, functional zones, and hospital departments. Inferential statistical analyses were therefore performed using the complete hierarchical dataset rather than relying solely on department-level summary values.
The primary inferential analysis evaluated differences in LAeq across hospital functional zones using a linear mixed-effects model with hospital department specified as a random effect to account for clustering of repeated observations. Functional zone was included as the principal fixed effect, and estimated marginal means were compared using Tukey-adjusted pairwise comparisons when the overall fixed effect was statistically significant. Exploratory mixed-effects regression analyses were subsequently performed to evaluate associations between environmental noise and selected operational characteristics. To minimize model overfitting, the primary exploratory model included only operational variables with strong theoretical justification and acceptable multicollinearity, specifically departmental bed capacity and visitor count. Regression coefficients (β), 95% confidence intervals (CI), and corresponding P values were reported. A more comprehensive mixed-effects model incorporating occupied beds, room area, room volume, staff count, alarm events, and conversation events was performed as a sensitivity analysis and is presented in the Supplementary Material. All regression analyses were considered exploratory and were not interpreted as evidence of causal relationships.
Prior to model fitting, assumptions of approximate normality and homoscedasticity were evaluated using the Shapiro–Wilk test, inspection of residual plots, quantile–quantile plots, and residual-versus-fitted plots. Multicollinearity among explanatory variables was assessed using variance inflation factors (VIFs). Model fit was summarized using marginal and conditional coefficients of determination (R2), and intraclass correlation coefficients (ICCs) were calculated to quantify clustering by hospital department. A two-sided P value < 0.05 was considered statistically significant. Measured sound levels were additionally interpreted in relation to internationally referenced acoustic recommendations, including the WHO Environmental Noise Guidelines for the European Region (2018), EN ISO 16032 recommendations for building acoustics, PN-B-02151 healthcare acoustic standards, and the Chinese National Standard GB3096-2022 for hospital environmental noise. Because these reference documents differ in scope, intended application, and measurement methodology, comparisons were interpreted descriptively rather than as formal compliance assessments. The proportion of measurement points exceeding selected reference thresholds was calculated descriptively. To visualize spatial variability in environmental noise exposure across departments and hospital functional zones, heat maps, boxplots, and scatter plots with fitted regression lines were generated. Because the study involved repeated measurements within operational hospital environments, statistical analyses were interpreted as exploratory environmental comparisons rather than causal inference models. Missing measurements were handled using complete-case analysis without data imputation. The single missing observation from the Intensive Care II Department represented less than 1% of the complete dataset and was therefore considered unlikely to materially influence the overall findings.
Results
Study Characteristics and Measurement Framework
Environmental noise measurements were performed in 15 inpatient departments of a large tertiary teaching hospital between March and August 2025 using a standardized hierarchical sampling protocol (Figure 1). Within each department, four predefined functional zones (ward, patient activity area, corridor, and nurses’ station) were evaluated. Three fixed sampling locations were established within each functional zone, and measurements were obtained during three standardized daytime observation periods (09:00–19:00). At each location and observation period, three consecutive technical replicate measurements were recorded using a calibrated Class 1 sound level meter, generating acoustic data for LAeq, LAFmax, and LCpeak. Operational variables, including bed capacity, occupied beds, staff count, visitor count, alarm events, conversation events, room area, and room volume, were recorded concurrently with each acoustic observation. The planned sampling framework comprised 1,620 acoustic observations (15 departments × 4 functional zones × 3 sampling locations × 3 observation periods × 3 technical replicates). One scheduled observation in the patient activity area of the Intensive Care II Department could not be completed because of temporary emergency-care access restrictions and was handled using complete-case analysis without data imputation, as prespecified in the statistical analysis plan. Additional indoor and outdoor public-area measurements were obtained for descriptive contextual comparison but were excluded from the departmental mixed-effects analyses (Figure 1). The characteristics of the participating departments are summarized in Table 1. Bed capacity ranged from 21 to 72 beds, with corresponding mean occupied beds ranging from 17.8 to 60.6. Mean staff counts were relatively consistent across departments (5.5–5.9 personnel per observation), whereas visitor counts averaged approximately five individuals per observation. Departmental room areas ranged from 145.8 to 500.3 m2, and room volumes ranged from 466.5 to 1,601.0 m3, reflecting substantial variation in the physical characteristics of the participating departments.
Figure 1: Study Design and Hierarchical Environmental Noise Measurement Protocol. Overview of the study design and hierarchical environmental noise measurement protocol. Environmental noise was assessed in 15 inpatient departments of a tertiary teaching hospital using a multistage hierarchical sampling framework. Measurements were obtained from four standardized functional zones (ward, corridor, nurses’ station, and patient activity area), three predefined sampling locations (entrance, middle, and end), three daytime observation periods (morning, midday, and afternoon), and three consecutive technical replicates. Equivalent continuous A-weighted sound pressure level (LAeq) was the primary outcome, and maximum A-weighted sound pressure level (LAFmax) and peak C-weighted sound pressure level (LCpeak) were recorded as secondary acoustic outcomes. Operational variables were recorded concurrently, and all measurements underwent calibration and predefined quality-control procedures before statistical analysis. LAeq, equivalent continuous A-weighted sound pressure level; LAFmax, maximum A-weighted sound pressure level; LCpeak, peak C-weighted sound pressure level. Please click here to view a larger version of this figure.
| Department | Bed Capacity | Mean Occupied Beds | Room Area (m²) | Room Volume (m³) | Mean Staff Count | Mean Visitor Count | Mean Alarm Count | Mean Conversation Count | ||
| Trauma Orthopedics | 40 | 33.8 | 280.1 | 896.3 | 5.5 | 4.8 | 2.4 | 6.5 | ||
| Breast and Thyroid Diagnosis and Treatment Center | 60 | 50.6 | 416.8 | 1333.8 | 5.7 | 4.8 | 2.4 | 6.2 | ||
| Congenital Heart Disease Center | 35 | 29.4 | 243.6 | 779.5 | 5.7 | 5 | 2.4 | 6.5 | ||
| Otolaryngology–Cervicology Surgery | 50 | 42.2 | 348.1 | 1113.9 | 5.8 | 5 | 2.5 | 6.7 | ||
| Gastroenterology Diagnosis and Treatment Department I | 38 | 31.9 | 264.8 | 847.3 | 5.8 | 5 | 2.5 | 6.6 | ||
| Gastroenterology Diagnosis and Treatment Department II | 52 | 43.9 | 361.7 | 1157.4 | 5.7 | 4.9 | 2.4 | 6.5 | ||
| Hepatobiliary and Pancreatic Surgery | 48 | 40.5 | 334.1 | 1069.2 | 5.8 | 5.1 | 2.5 | 6.8 | ||
| Intensive Care II Department | 21 | 17.8 | 145.8 | 466.5 | 5.7 | 4.8 | 2.5 | 6.5 | ||
| Knee Joint Department II | 45 | 37.8 | 312.7 | 1000.6 | 5.8 | 4.9 | 2.5 | 6.5 | ||
| Parkinson's Disease Department | 55 | 46.2 | 382.2 | 1223 | 5.9 | 5.1 | 2.5 | 6.8 | ||
| Psychosomatic Medicine Department I | 68 | 57.1 | 472.6 | 1512.3 | 5.8 | 5 | 2.5 | 6.7 | ||
| Pediatrics Department | 40 | 33.6 | 278.6 | 891.5 | 5.8 | 5.1 | 2.5 | 6.8 | ||
| Urology Department | 62 | 52.1 | 431.2 | 1380 | 5.8 | 5 | 2.5 | 6.6 | ||
| Vascular and Nerve Department I | 72 | 60.6 | 500.3 | 1601 | 5.9 | 5.1 | 2.5 | 6.9 | ||
| Vascular and Nerve Department II | 60 | 50.4 | 417.5 | 1336 | 5.7 | 4.8 | 2.4 | 6.3 | ||
Table 1: Characteristics of Hospital Departments Included in the Study. Baseline operational characteristics of the 15 inpatient departments included in the environmental noise survey. Department-level variables include bed capacity, mean occupied beds, room area, room volume, and the mean staff count, visitor count, alarm count, and conversation count recorded during standardized daytime observation periods. Operational variables represent average values obtained during the predefined observation periods and were evaluated as exploratory variables in the mixed-effects regression analyses. Room area and room volume are reported in square meters (m2) and cubic meters (m3), respectively.
Descriptive Acoustic Characteristics Across Functional Zones
Across all 1,620 observations, nurses’ stations demonstrated the highest environmental noise levels, followed sequentially by corridors, patient activity areas, and wards (Table 2; Figures 2 and 3). Mean LAeq values were 66.90 ± 2.34 dB(A) at nurses’ stations, 63.97 ± 2.38 dB(A) in corridors, 61.38 ± 2.17 dB(A) in patient activity areas, and 60.38 ± 2.22 dB(A) in wards. Similar spatial gradients were observed for LAFmax and LCpeak, indicating consistent differences across all three-acoustic metrics. Detailed descriptive summaries of the secondary acoustic outcomes (LAFmax and LCpeak) are provided in Supplementary Table 1.
| Functional Zone | Observations (n) | LAeq, dB(A) Mean ± SD | LAFmax, dB(A) Mean ± SD | LCpeak, dB(C) Mean ± SD | Median LAeq (IQR), dB(A) |
| Ward | 405 | 60.38 ± 2.22 | 67.95 ± 2.37 | 79.21 ± 3.31 | 60.4 (58.9–61.9) |
| Corridor | 405 | 63.97 ± 2.38 | 71.46 ± 2.59 | 82.76 ± 3.56 | 64.0 (62.5–65.5) |
| Nurses' Station | 405 | 66.90 ± 2.34 | 74.43 ± 2.56 | 85.72 ± 3.56 | 66.7 (65.4–68.3) |
| Patient Activity Area | 405 | 61.38 ± 2.17 | 69.00 ± 2.42 | 80.20 ± 3.37 | 61.5 (60.0–62.7) |
Table 2: Descriptive Acoustic Characteristics by Hospital Functional Zone. Descriptive acoustic characteristics for the four standardized hospital functional zones. Values are presented as mean ± standard deviation (SD) unless otherwise indicated. The median and interquartile range (IQR) are reported for the equivalent continuous A-weighted sound pressure level (LAeq). Each functional zone contributed 405 observations to the hierarchical dataset (total n = 1,620). LAeq, equivalent continuous A-weighted sound pressure level; LAFmax, maximum A-weighted sound pressure level; LCpeak, peak C-weighted sound pressure level; IQR, interquartile range.
Figure 2: Heat Map of Mean Equivalent Continuous A-Weighted Sound Pressure Levels Across Hospital Departments and Functional Zones. Heat map showing mean equivalent continuous A-weighted sound pressure levels (LAeq) for each hospital department and functional zone. Cell values represent mean LAeq in A-weighted decibels [dB(A)], and the color scale indicates increasing sound levels from lower to higher values. Functional zones include wards, corridors, nurses’ stations, and patient activity areas. CHD, congenital heart disease; ICU, intensive care unit; LAeq, equivalent continuous A-weighted sound pressure level. Please click here to view a larger version of this figure.
Figure 3: Distribution of Equivalent Continuous A-Weighted Sound Pressure Levels Across Standardized Hospital Functional Zones. Boxplots showing the distribution of equivalent continuous A-weighted sound pressure levels (LAeq) across four standardized hospital functional zones. Boxes represent the interquartile range (25th–75th percentile), the horizontal line within each box indicates the median, and the × symbol denotes the mean. Whiskers extend to 1.5 times the interquartile range (IQR), and observations beyond the whiskers are displayed as outliers. Each functional zone includes 405 observations (total n = 1,620). Differences among functional zones were evaluated using the primary linear mixed-effects model with Tukey-adjusted pairwise comparisons. LAeq, equivalent continuous A-weighted sound pressure level; IQR, interquartile range. Please click here to view a larger version of this figure.
The heat map presented in Figure 2 demonstrates a consistent spatial pattern across the 15 participating departments. Nurses’ stations consistently exhibited the highest mean LAeq values irrespective of clinical specialty, whereas ward areas demonstrated the lowest equivalent continuous A-weighted sound pressure levels. Departments with larger bed capacities, including Vascular and Nerve Department I, Psychosomatic Medicine Department I, and the Urology Department, generally exhibited higher mean LAeq values across functional zones than departments with smaller bed capacities, such as the Intensive Care II Department and the Congenital Heart Disease Center. Although modest interdepartmental variability was observed, the relative ranking of functional zones remained consistent. Mean LAeq values at nurses’ stations ranged from 64.1 dB(A) in the Intensive Care II Department to 68.9 dB(A) in Vascular and Nerve Department I, whereas ward measurements ranged from 57.9 to 61.6 dB(A). Patient activity areas and corridors exhibited intermediate values across nearly all departments (Figure 2). Figure 3 illustrates the distribution of LAeq measurements within each functional zone. Median LAeq values were highest at nurses’ stations (66.7 dB[A]), followed by corridors (64.0 dB[A]), patient activity areas (61.5 dB[A]), and wards (60.4 dB[A]). Although occasional high-value observations were identified across all functional zones, the interquartile ranges were relatively narrow, indicating limited variability within each functional environment. The boxplots demonstrate clear separation in the distribution of LAeq values among the four functional zones (Figure 3).
Mixed-Effects Comparison of Functional Zones
To account for the hierarchical structure of repeated acoustic measurements within hospital departments, the primary analysis used a linear mixed-effects model with hospital department specified as a random intercept and functional zone as a fixed effect (Table 3). This modeling approach accounted for within-department clustering while estimating adjusted differences in LAeq among the four predefined functional zones.
| Fixed Effects (Outcome: LAeq) | ||||
| Fixed Effect | β | SE | 95% CI | P Value |
| Intercept (Corridor) | 63.972 | 0.266 | 63.451 to 64.493 | <0.001 |
| Nurses' Station | 2.931 | 0.146 | 2.644 to 3.218 | <0.001 |
| Patient Activity Area | −2.594 | 0.146 | −2.881 to −2.307 | <0.001 |
| Ward | −3.589 | 0.146 | −3.876 to −3.302 | <0.001 |
| Estimated Marginal Means | ||||
| Functional Zone | Estimated Mean LAeq, dB(A) | |||
| Nurses' Station | 66.9 | |||
| Corridor | 63.97 | |||
| Patient Activity Area | 61.38 | |||
| Ward | 60.38 | |||
| Tukey-Adjusted Pairwise Comparisons | ||||
| Comparison | Mean Difference, dB(A) | Adjusted P Value | ||
| Nurses' Station vs. Corridor | 2.93 | <0.001 | ||
| Nurses' Station vs. Patient Activity Area | 5.52 | <0.001 | ||
| Nurses' Station vs. Ward | 6.52 | <0.001 | ||
| Corridor vs. Patient Activity Area | 2.59 | <0.001 | ||
| Corridor vs. Ward | 3.59 | <0.001 | ||
| Patient Activity Area vs. Ward | 0.99 | <0.001 | ||
| Model Performance | ||||
| Statistic | Value | |||
| Intraclass Correlation Coefficient (ICC) | 0.172 | |||
| Marginal R² | 0.53 | |||
| Conditional R² | 0.61 | |||
Table 3: Primary Linear Mixed-Effects Model Comparing Environmental Noise Across Hospital Functional Zones. Results of the primary linear mixed-effects model evaluating differences in equivalent continuous A-weighted sound pressure levels (LAeq) among the four standardized hospital functional zones. Hospital department was specified as a random intercept to account for clustering of repeated observations, and functional zone was included as a fixed effect with corridor as the reference category. Regression coefficients (β), standard errors (SE), Wald 95% confidence intervals (CI), and two-sided P values are presented for the fixed effects. Estimated marginal means were compared using Tukey-adjusted pairwise comparisons. Model performance is summarized using the intraclass correlation coefficient (ICC) and the marginal and conditional coefficients of determination (R2), calculated according to the method of Nakagawa and Schielzeth. LAeq, equivalent continuous A-weighted sound pressure level; β, regression coefficient; SE, standard error; CI, confidence interval; ICC, intraclass correlation coefficient.
The mixed-effects analysis demonstrated a highly significant overall effect of functional zone on environmental noise levels (overall likelihood-ratio test, P < 0.001). Using the corridor as the reference category, nurses’ stations exhibited significantly higher LAeq values (β = 2.931 dB[A], SE = 0.146, 95% CI = 2.644–3.218, P < 0.001), whereas patient activity areas (β = −2.594 dB[A], SE = 0.146, 95% CI = −2.881 to −2.307, P < 0.001) and wards (β = −3.589 dB[A], SE = 0.146, 95% CI = −3.876 to −3.302, P < 0.001) demonstrated significantly lower LAeq values (Table 3). Estimated marginal means derived from the mixed-effects model demonstrated a clear hierarchical pattern of environmental noise across the functional zones. The adjusted mean LAeq was highest at nurses’ stations (66.90 dB[A]), followed by corridors (63.97 dB[A]), patient activity areas (61.38 dB[A]), and wards (60.38 dB[A]). All pairwise comparisons remained statistically significant after Tukey adjustment (all adjusted P < 0.001), indicating statistically significant differences in LAeq among all four functional zones. Mean differences ranged from 0.99 dB(A) between patient activity areas and wards to 6.52 dB(A) between nurses’ stations and wards. Figure 4 summarizes the model-adjusted estimates. Estimated marginal means and their 95% confidence intervals showed minimal overlap among the functional zones, and Tukey grouping assigned each zone to a unique significance group (A–D). Nurses’ stations were assigned to group A with the highest adjusted LAeq, followed by corridors (group B), patient activity areas (group C), and wards (group D). Variance component analysis demonstrated that approximately 17.2% of the total variability in LAeq was attributable to differences between hospital departments (intraclass correlation coefficient = 0.172), supporting inclusion of a department-level random intercept. The fixed effects explained 53% of the variability in environmental noise (marginal R2 = 0.53), whereas the combined fixed- and random-effects model explained 61% of the total variability (conditional R2 = 0.61), indicating good explanatory performance of the hierarchical model (Table 3).
Figure 4: Estimated Marginal Means of Equivalent Continuous A-Weighted Sound Pressure Levels Across Hospital Functional Zones. Estimated marginal means (EMMs) of equivalent continuous A-weighted sound pressure levels (LAeq) across the four hospital functional zones derived from the primary linear mixed-effects model. Points represent model-estimated marginal means, and horizontal error bars indicate 95% confidence intervals (CIs). Hospital department was included as a random effect and functional zone as a fixed effect. Pairwise comparisons were performed using Tukey-adjusted estimated marginal means; functional zones assigned different letters represent statistically significant differences (P < 0.05). Each functional zone contributed 405 observations to the analysis. LAeq, equivalent continuous A-weighted sound pressure level; EMM, estimated marginal mean; CI, confidence interval. Please click here to view a larger version of this figure.
Parsimonious Mixed-Effects Regression of Operational Characteristics Associated with LAeq
To investigate operational characteristics associated with environmental noise while minimizing multicollinearity, a prespecified parsimonious mixed-effects model was fitted with functional zone, visitor count, and departmental bed capacity included as fixed effects and hospital department retained as a random intercept (Table 4). Selection of these variables was informed by the exploratory correlation matrix (Supplementary Figure 1), which demonstrated strong correlations among several structural variables, particularly room area and room volume (Spearman’s ρ = 0.94). Consequently, highly correlated structural variables were excluded from the parsimonious model to improve interpretability and model stability. After adjustment for the operational covariates, functional zone remained the strongest independent factor associated with environmental noise. Relative to corridors, nurses’ stations were associated with an adjusted increase of 2.933 dB(A) (SE = 0.146, 95% CI = 2.647–3.220, P < 0.001), whereas patient activity areas (β = −2.602 dB[A], SE = 0.146, 95% CI = −2.888 to −2.315, P < 0.001) and wards (β = −3.606 dB[A], SE = 0.146, 95% CI = −3.892 to −3.319, P < 0.001) remained significantly quieter than corridors.
| Predictor | β Coefficient | SE | 95% CI | P Value |
| Intercept (Corridor) | 63.97 | 0.27 | 63.45 to 64.49 | <0.001 |
| Nurses' Station (vs. Corridor) | 2.933 | 0.146 | 2.647 to 3.220 | <0.001 |
| Patient Activity Area (vs. Corridor) | −2.602 | 0.146 | −2.888 to −2.315 | <0.001 |
| Ward (vs. Corridor) | −3.606 | 0.146 | −3.892 to −3.319 | <0.001 |
| Visitor Count (per additional visitor) | 0.06 | 0.026 | 0.008 to 0.112 | 0.024 |
| Bed Capacity (per additional bed) | 0.064 | 0.005 | 0.055 to 0.073 | <0.001 |
| Model: LAeq ~ Functional Zone + Visitor Count + Bed Capacity + (1 | Department) | ||||
Table 4: Parsimonious Linear Mixed-Effects Regression of Operational Characteristics Associated with Environmental Noise. Results of the prespecified parsimonious linear mixed-effects regression model evaluating associations between equivalent continuous A-weighted sound pressure level (LAeq) and selected operational characteristics. The model included functional zone, visitor count, and departmental bed capacity as fixed effects and hospital department as a random intercept to account for clustering of repeated observations (model: LAeq ~ Functional Zone + Visitor Count + Bed Capacity + (1 | Department)). Corridor was specified as the reference category for functional zone, and visitor count and bed capacity were modeled as continuous variables. Regression coefficients (β), standard errors (SE), Wald 95% confidence intervals (CI), and two-sided P values are reported. Regression coefficients for visitor count and bed capacity represent the estimated change in LAeq associated with each additional visitor and each additional hospital bed, respectively. The intercept represents the estimated mean LAeq for the reference category (corridor) when continuous predictors are equal to zero. This table presents the prespecified parsimonious model developed to minimize model overfitting; the fully adjusted sensitivity model including all recorded operational characteristics is presented in Supplementary Table S2. LAeq, equivalent continuous A-weighted sound pressure level; β, regression coefficient; SE, standard error; CI, confidence interval.
Among the continuous operational predictors, visitor count demonstrated an independent positive association with LAeq. Each additional visitor present during measurement was associated with an estimated increase of 0.060 dB(A) (SE = 0.026, 95% CI = 0.008–0.112, P = 0.024). Similarly, departmental bed capacity remained independently associated with environmental noise, with each additional hospital bed associated with an estimated 0.064 dB(A) increase in LAeq (SE = 0.005, 95% CI = 0.055–0.073, P < 0.001).
Figure 5 illustrates the fitted predictions generated from the parsimonious mixed-effects model. Predicted LAeq increased approximately linearly across the observed range of departmental bed capacities, and higher visitor counts consistently shifted the predicted sound levels upward across the entire capacity range. The parallel fitted regression lines indicate additive effects of visitor count and bed capacity within the fitted model after adjustment for functional zone and department-level clustering. Model-based predictions demonstrated that, for a department with approximately 20 beds, the predicted LAeq ranged from 61.6 dB(A) at a low visitor count (three visitors) to 62.4 dB(A) at a higher visitor count (seven visitors). At the upper end of the observed bed-capacity range (72 beds), the corresponding predicted LAeq increased to 64.5 dB(A) and 65.3 dB(A), respectively, illustrating the combined contribution of departmental bed capacity and visitor count within the fitted model.
Figure 5: Model-Based Predictions of Equivalent Continuous A-Weighted Sound Pressure Levels According to Departmental Bed Capacity. Predicted equivalent continuous A-weighted sound pressure levels (LAeq) across the observed range of departmental bed capacity derived from the parsimonious linear mixed-effects regression model. Colored lines represent model-predicted LAeq values at low (3 visitors), average (5 visitors), and high (7 visitors) visitor counts while holding all other model terms constant. Shaded regions indicate 95% confidence intervals for the predicted values, and points represent observed measurements. Predictions are shown only within the observed range of departmental bed capacity included in the study. LAeq, equivalent continuous A-weighted sound pressure level; CI, confidence interval. Please click here to view a larger version of this figure.
Sensitivity analyses using the fully adjusted mixed-effects model (Supplementary Table 2), which additionally included occupied beds, room area, room volume, staff count, alarm events, conversation events, and measurement period, produced highly consistent estimates for functional zone and visitor count. Nurses’ stations remained the noisiest functional zone, and visitor count remained independently associated with higher LAeq (β = 0.060, P = 0.023), whereas occupied beds, room area, room volume, staff count, alarm events, conversation events, and measurement period were not independently associated with environmental noise after mutual adjustment. These findings support the robustness of the parsimonious model and indicate that functional zone and visitor count were the principal variables associated with environmental noise in this study.
Descriptive Comparison with Reference Acoustic Criteria
To place the observed environmental noise levels into context, measured LAeq values were descriptively compared with three widely cited reference acoustic criteria: the WHO daytime guideline value for hospital patient-care environments (35 dB[A]), the EN ISO 16032 reference value for general indoor environments (40 dB[A]), and the Chinese environmental quality standard GB3096-2022 Class 1 reference value (45 dB[A]). Because these reference values differ in scope, intended application, measurement methodology, and averaging period, the comparisons were performed for descriptive purposes only and do not represent formal regulatory compliance assessments (Figure 6). Across all 1,620 acoustic observations, environmental noise levels exceeded each of the selected reference values. Overall, 1,615 observations (99.7%) exceeded the WHO daytime guideline value of 35 dB(A), 1,586 observations (97.9%) exceeded the EN ISO 16032 reference value of 40 dB(A), and 1,513 observations (93.2%) exceeded the Chinese GB3096-2022 Class 1 reference value of 45 dB(A).
Figure 6: Percentage of Acoustic Observations Exceeding Selected International and National Reference Acoustic Criteria Across Hospital Functional Zones. Percentage of equivalent continuous A-weighted sound pressure level (LAeq) observations exceeding selected international and national reference acoustic criteria across the four standardized hospital functional zones. Bars represent the proportion of observations exceeding the reference thresholds of the World Health Organization (WHO) guidance value (>35 dB[A]), EN ISO 16032 (>40 dB[A]), and the Chinese National Standard GB3096-2022 (>45 dB[A]). Percentages were calculated as the number of observations exceeding each reference threshold divided by the total number of observations within each functional zone (n = 405). Because these reference documents differ in their intended application, measurement methodology, environmental setting, and averaging period, the comparisons are presented for descriptive purposes only and do not represent formal compliance assessments. LAeq, equivalent continuous A-weighted sound pressure level; WHO, World Health Organization. Please click here to view a larger version of this figure.
Marked differences were observed among the functional zones. Nurses’ stations demonstrated the highest proportion of observations exceeding each reference value, with 100.0% exceeding both the WHO and EN ISO reference values and 99.0% exceeding the Chinese reference value. Corridors likewise demonstrated high proportions of exceedance (100.0%, 99.3%, and 96.0%, respectively). In comparison, wards exhibited the lowest—but still substantial—proportions of observations exceeding the three reference values (99.3%, 95.3%, and 87.9%, respectively), whereas patient activity areas demonstrated intermediate proportions (99.5%, 97.0%, and 90.6%, respectively) These descriptive findings were consistent with the mixed-effects analyses, with nurses’ stations demonstrating the highest environmental noise levels and wards the lowest across the four functional zones. The high proportions of observations exceeding all three reference values indicate that elevated daytime environmental noise was observed throughout the hospital rather than being limited to individual departments or functional zones.
Sensitivity Analyses and Model Diagnostics
The robustness of the primary findings was evaluated using a fully adjusted sensitivity mixed-effects model that additionally included occupied beds, room area, room volume, staff count, alarm events, conversation events, and measurement period (Supplementary Table 2). Functional-zone estimates remained highly consistent with those observed in the prespecified parsimonious model. Nurses’ stations continued to demonstrate significantly higher LAeq than corridors, whereas patient activity areas and wards remained significantly quieter (all P < 0.001). Visitor count also remained independently associated with higher LAeq (β = 0.060, P = 0.023), whereas occupied beds, room area, room volume, staff count, alarm events, conversation events, and measurement period were not independently associated with environmental noise after simultaneous adjustment (Supplementary Table 2). These findings indicate that the principal findings were robust to additional adjustment for correlated operational characteristics. Model assumptions were evaluated using residual diagnostic procedures (Supplementary Figure 1). The normal quantile–quantile (Q–Q) plot demonstrated close agreement between the observed and theoretical residual quantiles, indicating no substantial departure from normality. The residual-versus-fitted plot showed random scatter around zero without discernible trends or evidence of heteroscedasticity, supporting the assumptions of linearity and constant residual variance. Consistent with these visual findings, the Shapiro–Wilk test did not demonstrate a statistically significant deviation from normality (W = 0.992, P = 0.084). Additional model-performance statistics are summarized in Supplementary Table 3. The residual standard deviation was 1.92 dB(A), and the department-level random-intercept standard deviation was 0.90 dB(A). The intraclass correlation coefficient was 0.172, indicating moderate clustering of observations within hospital departments. The mixed-effects model explained 53% of the variability through the fixed effects alone (marginal R2 = 0.53) and 61% when the department-level random effects were incorporated (conditional R2 = 0.61). Variance inflation factors were uniformly low (maximum VIF = 2.14), indicating no evidence of problematic multicollinearity among the retained predictors. Furthermore, all Cook’s distance values were <1.0 (maximum = 0.18), suggesting that no individual observation exerted undue influence on the fitted model. The exploratory correlation matrix (Supplementary Figure 2) demonstrated expected positive associations among several operational characteristics. Strong correlations were observed between room area and room volume (Spearman’s ρ = 0.94), bed capacity and occupied beds (ρ = 0.86), and bed capacity and room volume (ρ = 0.81). Moderate correlations were observed between visitor count and occupied beds (ρ = 0.71) and between staff count and conversation events (ρ = 0.62). These findings supported the prespecified variable-selection strategy for the parsimonious mixed-effects model, whereby highly correlated structural variables were excluded to minimize multicollinearity while preserving model interpretability.
Overall, the primary mixed-effects analyses, sensitivity analyses, residual diagnostics, and exploratory correlation assessment demonstrated internally consistent findings. Across all analyses, functional zone demonstrated the strongest independent association with environmental noise, whereas visitor count and departmental bed capacity also remained independently associated with LAeq. Most other recorded operational characteristics were not independently associated with environmental noise after mutual adjustment.
Data Availability:
The data supporting the findings of this study are available within the article and its Supplementary Materials. The complete raw acoustic measurement dataset, calibration records, operational-variable dataset, figure source data, and R statistical analysis code have been deposited in the Zenodo repository and are publicly available at https://doi.org/10.5281/zenodo.21404807. Supplementary Tables 1–3 and Supplementary Figures 1 and 2 provide additional processed datasets, sensitivity analyses, model diagnostics, and supporting analyses.
Supplementary Figure 1: Residual Diagnostic Plots for the Parsimonious Linear Mixed-Effects Model. Diagnostic plots evaluating the assumptions of the parsimonious linear mixed-effects regression model. (A) Normal quantile–quantile (Q–Q) plot of the conditional residuals demonstrating approximate normality of the residual distribution. (B) Plot of conditional residuals versus fitted values showing no substantial deviation from homoscedasticity or systematic pattern across the fitted range. Hospital department was included as a random intercept, and functional zone, visitor count, and bed capacity were included as fixed effects. Residual diagnostics support the adequacy of the model assumptions for the primary regression analysis. Diagnostic plots were generated using the DHARMa package in R. LAeq, equivalent continuous A-weighted sound pressure level.Please click here to download this file.
Supplementary Figure 2: Spearman Correlation Matrix of Operational Variables Included in the Exploratory Mixed-Effects Analyses. Spearman correlation matrix showing pairwise associations among departmental operational variables evaluated during exploratory mixed-effects regression analyses. Cell values represent Spearman’s rank correlation coefficients (ρ), and the color scale indicates the strength and direction of the correlation, with blue representing positive correlations and red representing negative correlations. Correlation analysis was performed to evaluate potential multicollinearity among candidate explanatory variables before model development. Variables demonstrating strong collinearity were considered during model selection to minimize multicollinearity in the parsimonious regression model. Operational variables included departmental bed capacity, occupied beds, room area, room volume, staff count, visitor count, alarm count, and conversation count. Spearman’s ρ, Spearman’s rank correlation coefficient.Please click here to download this file.
Supplementary Table 1: Secondary Acoustic Outcomes According to Hospital Functional Zone. Descriptive summary of the secondary acoustic outcomes measured across the four standardized hospital functional zones. Values are presented as mean ± standard deviation (SD) and represent pooled observations from all 15 inpatient departments. Equivalent continuous A-weighted sound pressure level (LAeq) was the primary acoustic outcome, whereas maximum A-weighted sound pressure level (LAFmax) and peak C-weighted sound pressure level (LCpeak) were recorded as secondary acoustic measures to characterize maximum and peak sound levels during routine daytime hospital operations. LAeq, equivalent continuous A-weighted sound pressure level; LAFmax, maximum A-weighted sound pressure level; LCpeak, peak C-weighted sound pressure level; SD, standard deviation.Please click here to download this file.
Supplementary Table 2: Fully Adjusted Sensitivity Linear Mixed-Effects Regression Model. Results of the fully adjusted sensitivity linear mixed-effects regression model evaluating associations between equivalent continuous A-weighted sound pressure level (LAeq) and all recorded operational characteristics. Hospital department was included as a random intercept to account for clustering of repeated observations. Functional zone, time period, departmental bed capacity, occupied beds, room area, room volume, staff count, visitor count, alarm count, and conversation count were included as fixed effects. Corridor served as the reference category for functional zone, and afternoon served as the reference category for time period. Regression coefficients (β), standard errors (SE), Wald 95% confidence intervals (CI), and two-sided P values are reported. This model was performed as a sensitivity analysis to evaluate the robustness of the parsimonious model presented in Table 4 after adjustment for all recorded operational characteristics. LAeq, equivalent continuous A-weighted sound pressure level; β, regression coefficient; SE, standard error; CI, confidence interval.Please click here to download this file.
Supplementary Table 3: Model Performance and Diagnostic Statistics for the Primary Linear Mixed-Effects Model. Summary of model performance and diagnostic statistics for the primary linear mixed-effects model evaluating equivalent continuous A-weighted sound pressure levels (LAeq) across hospital functional zones. Residual standard deviation (σ) represents the unexplained within-model variability, whereas the random-intercept standard deviation quantifies between-department variability. The intraclass correlation coefficient (ICC) estimates the proportion of total variance attributable to clustering within hospital departments. Marginal R2 represents the proportion of variance explained by the fixed effects alone, whereas conditional R2 represents the proportion of variance explained by both the fixed and random effects. Akaike information criterion (AIC), Bayesian information criterion (BIC), and log-likelihood are provided as measures of model fit. Variance inflation factors (VIFs) were all <5, indicating no evidence of problematic multicollinearity, and Cook’s distance values were <1.0 for all observations, indicating that no individual observation exerted undue influence on the model estimates. LAeq, equivalent continuous A-weighted sound pressure level; ICC, intraclass correlation coefficient; VIF, variance inflation factor.Please click here to download this file.
Discussion
This cross-sectional environmental acoustic assessment demonstrated that daytime environmental noise levels consistently exceeded commonly referenced international and national acoustic reference values across most hospital functional zones. Linear mixed-effects analyses identified functional zone as the principal factor associated with LAeq, with nurses’ stations exhibiting the highest adjusted sound levels, followed by corridors, patient activity areas, and wards. Comparisons with the WHO guidance values, EN ISO 16032, the PN-B-02151 series, and the Chinese National Standard GB3096-2022 were interpreted descriptively because these documents differ in their intended purpose, measurement metrics, averaging periods, and scope. Nevertheless, the consistently elevated daytime LAeq values observed throughout the hospital indicate that routine clinical environments frequently experience sound levels substantially above widely cited reference acoustic values. These findings are broadly consistent with previous hospital acoustic investigations reporting persistent daytime environmental noise above recommended reference values5,8,17,24,25. Unlike many previous hospital acoustic surveys that relied primarily on department-level averages or independent-group comparisons, the present study analyzed the complete hierarchical dataset using linear mixed-effects models, thereby accounting for repeated observations nested within hospital departments. The moderate intraclass correlation coefficient (ICC = 0.172) indicated that a meaningful proportion of the variability in environmental noise was attributable to differences between departments, supporting the use of hierarchical modeling for hospital acoustic research8,17,24. Measured daytime LAeq values consistently exceeded the WHO daytime reference values for patient-care environments as well as the selected EN ISO 16032 and GB3096-2022 acoustic reference criteria. These comparisons were interpreted descriptively rather than as formal compliance assessments because the referenced documents differ in their intended application, averaging period, measurement methodology, and scope. Nevertheless, the high proportion of observations exceeding these widely cited reference values is consistent with previous hospital acoustic investigations and underscores the continuing challenge of maintaining acoustically favorable healthcare environments during routine clinical operations29,30,31,32,33,34.
A consistent spatial hierarchy of environmental noise was observed across hospital functional zones. Both descriptive analyses and linear mixed-effects modeling demonstrated that nurses’ stations exhibited the highest adjusted LAeq values, followed by corridors, patient activity areas, and wards. This pattern is consistent with previous reports identifying nurses’ stations and circulation areas as the principal sources of hospital environmental noise because of intensive staff communication, patient movement, alarm activity, and routine clinical workflows29,30,31,33,34,35,36. The observed differences remained significant after adjustment for visitor count, departmental bed capacity, and department-level clustering, indicating that the spatial variation was not explained solely by the measured structural or operational characteristics of individual departments. Estimated marginal means and Tukey-adjusted pairwise comparisons further confirmed that each functional zone represented a statistically distinct acoustic environment. The consistently elevated sound levels observed at nurses’ stations are operationally and acoustically plausible. These locations function as central coordination hubs where clinical documentation, multidisciplinary communication, electronic medical record activities, patient monitoring, alarm management, telephone conversations, and interactions with patients and relatives occur simultaneously. Consequently, the accumulation of conversations, monitor alarms, equipment operation, and staff movement generates sustained environmental noise rather than isolated acoustic events, contributing to the consistently higher LAeq values observed across nearly all participating departments. This interpretation is consistent with previous investigations identifying overlapping operational sound sources as major contributors to elevated hospital environmental noise33,34,35,36,37,38,39,40,41,42,43,44,45,46,47. Corridors likewise demonstrated higher environmental noise levels than patient activity areas and wards. As the principal circulation pathways for patient transport, staff movement, equipment transfer, housekeeping activities, and visitor traffic, these spaces are continuously exposed to diverse transient and background sound sources. Architectural openness, together with hard, reflective interior surfaces, may further facilitate sound propagation, contributing to elevated LAeq values. Although the present study did not directly evaluate reverberation characteristics, room acoustic parameters, or speech transmission indices, the observed spatial distribution is consistent with previous investigations describing hospital corridors as acoustically complex environments shaped by operational activity, building design, and multiple concurrent sound sources33,34,35,46,47,48,49,50,51. An important finding of the present study was the independent association between visitor count and environmental noise identified in both the prespecified parsimonious mixed-effects model and the fully adjusted sensitivity analysis. Although the absolute effect size was modest, visitor count remained significantly associated with higher LAeq after adjustment for functional zone and departmental clustering, suggesting that routine human activity contributes measurably to the hospital soundscape. This interpretation is consistent with previous investigations identifying human activity, communication, visitor presence, and operational workflow as important contributors to environmental noise within healthcare facilities33,34,35,36,46,47,51. Departmental bed capacity also remained independently associated with higher environmental noise in the parsimonious model; however, this association should be interpreted cautiously because room geometry, architectural layout, occupancy density, and clinical workflow may also influence environmental acoustics. Furthermore, the cross-sectional observational design precludes causal inference, and these findings should therefore be interpreted as associations rather than evidence of cause-and-effect relationships33,34,49. The sensitivity analysis further strengthened confidence in these findings. Inclusion of additional operational variables, including occupied beds, room area, room volume, staff count, alarm events, conversation events, and measurement period, produced highly consistent estimates for functional zone and visitor count, whereas most structural variables were no longer independently associated with environmental noise after simultaneous adjustment. Collectively, these findings suggest that the daytime hospital soundscape is influenced more by the spatial organization of routine clinical activities than by the individual structural characteristics evaluated, supporting recent evidence that healthcare acoustic environments arise from the interaction of multiple operational and environmental factors rather than any single determinant33,34,35,46,47,49,51. The present findings are broadly consistent with previous investigations reporting daytime hospital environmental noise levels ranging from approximately 55 to 75 dB(A), substantially exceeding commonly referenced acoustic recommendations for patient-care environments8,9,10,11,29,30,31,32. Previous studies have associated elevated hospital noise with sleep disturbance, impaired restfulness, physiological stress responses, reduced acoustic comfort, communication difficulties, and other adverse health effects among patients and healthcare personnel29,30,31,32,49,52,53. Additional investigations have suggested possible associations between excessive environmental noise and neurocognitive disturbances, including delirium among critically ill patients15,16,48. Although the present study did not directly evaluate these clinical outcomes, the observed sound levels were comparable with those reported previously while extending existing evidence through standardized functional-zone sampling and hierarchical statistical analysis.
Although the present investigation was designed as an environmental acoustic assessment rather than a clinical outcomes study, the findings have important implications for healthcare facility design and operational management. Nurses’ stations and circulation corridors consistently exhibited the highest adjusted sound levels, suggesting that these operational areas may represent priority targets for future noise-management strategies. Because these environments function as hubs for communication, clinical coordination, patient transport, equipment movement, and visitor activity, interventions focused on these locations may provide greater opportunities for improving the overall hospital acoustic environment than approaches directed exclusively toward patient rooms. This interpretation is consistent with recent investigations demonstrating that multidisciplinary strategies, including workflow optimization, noise-management protocols, architectural modifications, and continuous acoustic monitoring, may improve acoustic conditions within healthcare facilities29,31,32,33,34,35,51. LAeq remains the most widely used metric for evaluating hospital environmental noise because it provides a standardized measure of cumulative acoustic exposure and facilitates comparison across studies8,9,10,11,32. Nevertheless, LAeq is an energy-averaged metric that does not distinguish between continuous background noise and transient impulsive events, such as monitor alarms, dropped instruments, overhead announcements, or sudden vocalizations. Consequently, acoustically distinct hospital soundscapes may yield similar LAeq values despite differing substantially in their perceptual characteristics and potential influence on communication, annoyance, and cognitive workload. Recent advances in healthcare acoustics have therefore emphasized complementary approaches, including psychoacoustic assessment, sound-source clustering, and continuous monitoring, to provide a more comprehensive characterization of hospital soundscapes33,34,35,39,40,41,42,43,44,45,46,47. Recent developments in healthcare acoustics have emphasized that hospital soundscapes should be characterized using approaches that extend beyond average sound pressure levels alone. Studies incorporating sound-source classification and clustering techniques have demonstrated that the perceptual impact of hospital noise depends not only on overall acoustic energy but also on the frequency, temporal distribution, and perceived intrusiveness of individual sound sources. De Salvio et al.33 highlighted that healthcare acoustic environments comprise complex combinations of continuous operational sounds and intermittent high-intensity events that contribute differently to annoyance and perceived disturbance. Likewise, Cingolani et al.46 demonstrated that clustering analyses can identify characteristic hospital sound patterns not readily detected using conventional time-averaged metrics, while Hummel et al.47 reported associations between clustered hospital noise profiles and patient satisfaction. Collectively, these studies suggest that future hospital acoustic assessments may benefit from integrating conventional environmental noise measurements with complementary characterization of sound-source composition, psychoacoustic attributes, and temporal acoustic patterns34. Accordingly, the present findings should be interpreted within the context of the measurement strategy employed. The objective of this study was to characterize routine daytime environmental noise under real-world clinical operating conditions using standardized LAeq measurements, which remain the most widely reported metric in hospital acoustic research. Although additional acoustic parameters, including LAFmax, LCpeak, octave-band frequency analysis, psychoacoustic indices, and sound-source clustering, were beyond the scope of the present survey, incorporating these complementary approaches would provide a more comprehensive characterization of hospital soundscapes in future investigations32,33,34,46,47. The independent association between visitor count and higher LAeq further supports the concept that routine human activity contributes meaningfully to the daytime hospital acoustic environment. Visitor count remained significantly associated with environmental noise after simultaneous adjustment for functional zone and other operational characteristics, whereas staff count, occupied beds, alarm events, and conversation events did not retain statistical significance. Together with the strong correlations observed among room area, room volume, and departmental bed capacity, these findings support the prespecified parsimonious modeling strategy and indicate that routine operational activity may better explain daytime environmental noise than the individual structural variables evaluated. Nevertheless, because all operational variables were measured contemporaneously with environmental noise, these findings should be interpreted as observational associations rather than evidence of causality. The combined use of exploratory correlation analysis and mixed-effects sensitivity modeling provides a transparent analytical framework that accounts for the hierarchical data structure while minimizing the influence of multicollinearity, consistent with contemporary recommendations for analyzing complex healthcare soundscapes33,34,35,46,47,51.
Several methodological strengths distinguish the present investigation from previous hospital environmental noise surveys. First, environmental noise was evaluated using a standardized sampling protocol that incorporated repeated measurements across multiple hospital departments, functional zones, and daytime observation periods, thereby reducing the influence of short-term fluctuations in routine hospital activity. Second, the complete hierarchical dataset was analyzed using linear mixed-effects models that accounted for clustering of repeated observations within hospital departments. This approach enabled more appropriate analysis of the nested data structure than conventional independent-group comparisons and provided robust estimates of functional-zone effects. Third, prespecified sensitivity analyses, together with residual diagnostics and multicollinearity assessment, demonstrated consistent findings across alternative model specifications, supporting the robustness of the primary analyses. Finally, all measurements were obtained using calibrated Class 1 instrumentation under standardized operating procedures, enhancing measurement reliability, reproducibility, and comparability with previous hospital acoustic studies. Several limitations should also be considered when interpreting the present findings. First, the investigation was conducted in a single tertiary teaching hospital, which may limit the generalizability of the findings to institutions with different architectural layouts, patient populations, staffing models, or healthcare systems. Multicenter studies including diverse healthcare settings are therefore warranted. Second, although repeated measurements were obtained during morning, midday, and afternoon observation periods, the total sampling duration represented only a portion of routine daytime hospital activity. Consequently, temporal variability associated with nighttime operations, shift changes, emergency admissions, cleaning schedules, and peak visiting periods may not have been fully captured. Continuous 24-h acoustic monitoring would provide a more comprehensive characterization of hospital soundscapes. Third, measurements were obtained at standardized sampling locations within each functional zone rather than through exhaustive spatial mapping of entire departments. Consequently, localized acoustic variability within large wards, extended corridors, or architecturally complex clinical environments may not have been fully represented. Future studies incorporating high-resolution spatial acoustic mapping and multiple measurement locations within individual functional zones would further improve characterization of hospital sound propagation. Fourth, the study relied primarily on LAeq as the principal acoustic outcome. Although this metric is widely accepted for environmental noise assessment and facilitates comparison with previous hospital acoustic research, it does not fully characterize impulsive acoustic events, frequency-specific sound characteristics, psychoacoustic responses, or sound-source composition. Future investigations incorporating LAFmax, LCpeak, octave-band frequency analysis, psychoacoustic indices, and sound-source clustering techniques would provide a more comprehensive evaluation of healthcare soundscapes. Fifth, operational variables such as visitor count, staff count, alarm events, and conversation events were recorded contemporaneously with environmental noise measurements and therefore should be regarded as observational descriptors rather than objectively validated exposure metrics. Although visitor count remained independently associated with environmental noise, these findings should be interpreted cautiously because the cross-sectional design precludes causal inference. Finally, patient-reported outcomes, occupational stress measures, physiological responses, sleep quality, and other clinical endpoints were not evaluated. Accordingly, the present findings should be interpreted as a comprehensive environmental acoustic characterization of routine daytime hospital operations rather than evidence that elevated environmental noise directly caused adverse patient or staff outcomes.
The present findings have several practical implications for hospital environmental noise management. Linear mixed-effects analyses demonstrated that environmental noise was not uniformly distributed throughout the hospital but was concentrated within specific operational zones, particularly nurses’ stations and circulation corridors. These findings suggest that future noise-management programs may achieve greater benefit by prioritizing high-activity operational areas where communication, patient transport, equipment movement, and routine clinical workflows are concentrated rather than focusing exclusively on inpatient wards. Potential engineering interventions include the use of sound-absorbing ceiling materials, acoustically treated wall surfaces, resilient flooring systems, and improved door or partition insulation to reduce sound reflection and propagation. Complementary administrative measures, including structured quiet-hour policies, optimized patient flow, visitor guidance, alarm-management protocols, and staff education regarding avoidable environmental noise, may further improve the hospital acoustic environment. Previous studies have suggested that combining engineering, behavioral, and organizational approaches may improve perceived ward quietness, patient restfulness, and overall acoustic conditions within healthcare facilities29,31,32. The observed association between visitor count and environmental noise further highlights the importance of considering routine operational activity when developing hospital noise-management strategies. Although the present findings do not support specific visitor restrictions, they suggest that optimizing circulation pathways, waiting-area organization, and communication practices may complement structural acoustic interventions without adversely affecting patient care. Overall, the hospital acoustic environment is likely to reflect the combined influence of architectural design, clinical workflow, patient and visitor movement, communication demands, medical equipment, and organizational practices rather than any single operational factor. Accordingly, effective environmental noise management will likely require coordinated multidisciplinary approaches involving clinicians, hospital administrators, architects, engineers, infection-control specialists, and facility-management personnel.
The present investigation provides a standardized framework for environmental acoustic assessment that may facilitate future multicenter studies across hospitals with differing architectural layouts, clinical specialties, and healthcare systems. Such studies would enable evaluation of the generalizability of the observed spatial patterns and support the development of benchmark environmental acoustic data for healthcare facilities. Future research should extend beyond cross-sectional environmental monitoring by incorporating continuous 24-h acoustic measurements, longitudinal follow-up, and seasonal assessments to better characterize temporal variability in hospital soundscapes. Simultaneous collection of patient-reported outcomes, validated sleep-quality measures, staff workload assessments, communication performance, and objective physiological indicators would enable a more comprehensive evaluation of the relationships between environmental acoustics and clinical or occupational outcomes. Recent advances in healthcare acoustics further support the integration of continuous acoustic monitoring, automated sound-source identification, machine-learning-based sound classification, psychoacoustic analyses, and clustering techniques to complement conventional LAeq measurements. These approaches may improve differentiation between continuous background noise and clinically important transient acoustic events, thereby providing a more comprehensive understanding of hospital soundscapes and their potential implications for patient care and staff performance33,34,46,47. Future prospective interventional studies should evaluate the effectiveness of engineering, administrative, behavioral, and technological noise-management strategies using controlled before-and-after or multicenter study designs to strengthen the evidence base for hospital acoustic management29,31,32,51. Overall, this study provides a comprehensive assessment of daytime environmental noise across hospital functional zones within a tertiary teaching hospital using standardized acoustic measurements and hierarchical statistical analysis. Environmental noise levels consistently exceeded commonly referenced international and national acoustic reference values during routine daytime clinical operations. After accounting for repeated measurements and departmental clustering, nurses’ stations exhibited the highest adjusted environmental sound levels, followed by corridors, patient activity areas, and wards. Visitor count and departmental bed capacity were independently associated with higher LAeq, whereas other recorded operational characteristics were not independently associated with environmental noise after multivariable adjustment. These findings demonstrate that environmental noise is unevenly distributed across hospital functional zones and identify operational areas that may represent priorities for future acoustic management. Because the study employed an observational cross-sectional design, the reported associations should not be interpreted as causal or as evidence of direct effects on patient or staff outcomes. Nevertheless, the standardized measurement protocol, hierarchical analytical approach, and comprehensive environmental assessment provide a robust foundation for future multicenter studies and interventional research aimed at optimizing hospital acoustic environments. Routine environmental acoustic monitoring using standardized measurement protocols and appropriate hierarchical analytical methods may facilitate identification of high-noise operational areas and support the development of evidence-informed hospital noise-management strategies.
Disclosures
Competing Interests:
The authors declare that they have no competing interests.
Acknowledgements
None declared. No funding was received for this study.
Materials
| Name | Company | Catalog Number | Comments |
|---|---|---|---|
| Calibration Logbook | Self-prepared | N/A | Documentation of instrument calibration and verification |
| Class 1 Acoustic Calibrator | Hangzhou Aihua Instruments Co., Ltd. | AWA6221B | Calibration of the sound level meter (94 dB at 1 kHz) |
| Class 1 Integrating Sound Level Meter | Hangzhou Aihua Instruments Co., Ltd. | AWA6228+ | Primary instrument for environmental noise measurements (IEC 61672-1 compliant) |
| Data Collection Sheets | Self-prepared | N/A | Standardized recording of field observations and operational variables |
| Digital Thermo-Hygrometer | TESTO | 608-H1 | Monitoring ambient temperature and relative humidity during measurements |
| Hospital Floor Plans/Department Layout Maps | Institutional source | N/A | Identification of departmental measurement locations |
| IBM SPSS Statistics (Version 29.0) | IBM | N/A | Statistical analyses |
| Laptop Computer | Dell | Latitude Series | Data acquisition, storage, and statistical analysis |
| Measuring Tape | STANLEY | N/A | Verification of microphone placement distances |
| Microsoft Excel | Microsoft | Current version | Data management and preliminary data processing |
| Personal Protective Equipment | Institutional supply | N/A | Compliance with hospital infection-control requirements |
| R Statistical Software (Version 4.3.3) | R Foundation for Statistical Computing | N/A | Statistical computing environment |
| R packages (emmeans, lme4, lmerTest) | Comprehensive R Archive Network (CRAN) | Current versions | Linear mixed-effects modeling and estimated marginal means |
| Rechargeable Battery Pack | Hangzhou Aihua Instruments Co., Ltd. | Manufacturer supplied | Power supply for the AWA6228+ Class 1 Integrating Sound Level Meter |
| Sound Level Meter Software | Hangzhou Aihua Instruments Co., Ltd. | Included with instrument | Instrument control, data acquisition, and export of acoustic measurements |
| Tripod Stand | Generic laboratory supplier | N/A | Stable positioning of the sound level meter microphone |
References
- Pal J, Taywade M, Pal R, Sethi D. Noise pollution in intensive care units: a hidden enemy affecting the physical and mental health of patients and caregivers. Noise Health. 2022;24(114):130-6. doi:10.4103/nah.nah_79_21.
- Zhang L. Effects of nighttime noise management in intensive care units on hormone levels and sleep quality in conscious patients. Noise Health. 2024;26(121):186-91. doi:10.4103/nah.nah_55_24.
- Deng Z, Xie H, Kang J. The acoustic environment in typical hospital wards in China. Appl Acoust. 2023;203:109202. doi:10.1016/j.apacoust.2022.109202.
- Yang J, Zhou J. Effect of the levels and sources of noise on the sleep quality of conscious patients in an emergency intensive care unit. Noise Health. 2024;26(123):489-94. doi:10.4103/nah.nah_83_24.
- Darbyshire JL, Young JD. An investigation of sound levels on intensive care units with reference to the WHO guidelines. Crit Care. 2013;17(5):R187. doi:10.1186/cc12870.
- Mehrotra A, et al. A comprehensive review of auditory and non-auditory effects of noise on human health. Noise Health. 2024;26(121):59-69. doi:10.4103/nah.nah_124_23.
- Basner M, et al. Auditory and non-auditory effects of noise on health. Lancet. 2014;383(9925):1325-32. doi:10.1016/S0140-6736(13)61613-X.
- Wallis R, et al. Environmental noise levels in hospital settings: a rapid review of measurement techniques and implementation in hospital settings. Noise Health. 2019;21(102):200-16. doi:10.4103/nah.NAH_19_18.
- Stansfeld SA, Matheson MP. Noise pollution: non-auditory effects on health. Br Med Bull. 2003;68:243-57. doi:10.1093/bmb/ldg033.
- Raja RV, Rajasekaran V, Sriraman G. Non-auditory effects of noise pollution on health: a perspective. Indian J Otolaryngol Head Neck Surg. 2019;71(Suppl 2):1500-1. doi:10.1007/s12070-019-01708-9.
- Hahad O, Prochaska JH, Daiber A, Münzel T. Environmental noise-induced effects on stress hormones, oxidative stress, and vascular dysfunction: key factors in the relationship between cerebrocardiovascular and psychological disorders. Oxid Med Cell Longev. 2019;2019:4623109. doi:10.1155/2019/4623109.
- Burger P, et al. Sleep in hospitalized pediatric and adult patients: a systematic review and meta-analysis. Sleep Med X. 2022;4:100059. doi:10.1016/j.sleepx.2022.100059.
- Xie H, Kang J, Mills GH. Clinical review: the impact of noise on patients’ sleep and the effectiveness of noise reduction strategies in intensive care units. Crit Care. 2009;13(2):208. doi:10.1186/cc7154.
- Park MJ, et al. Noise in hospital rooms and sleep disturbance in hospitalized medical patients. Environ Health Toxicol. 2014;29:e2014006. doi:10.5620/eht.2014.29.e2014006.
- Poulsen LM, Estrup S, Mortensen CB, Andersen-Ranberg NC. Delirium in intensive care. Curr Anesthesiol Rep. 2021;11(4):516-23. doi:10.1007/s40140-021-00476-z.
- Ma L, et al. Postoperative neurocognitive disorders in the elderly: how can we stop the harm A literature review. Front Med (Lausanne). 2025;12:1525639. doi:10.3389/fmed.2025.1525639.
- de Lima Andrade E, et al. Environmental noise in hospitals: a systematic review. Environ Sci Pollut Res Int. 2021;28(16):19629-42. doi:10.1007/s11356-021-13211-2.
- Astin F, et al. Night-time noise levels and patients’ sleep experiences in a medical assessment unit in Northern England. Open Nurs J. 2020;14:80-91. doi:10.2174/1874434602014010080.
- Bevan R, et al. Sleep quality and noise: comparisons between hospital and home settings. Arch Dis Child. 2019;104:147-51. doi:10.1136/archdischild-2018-315168.
- Bliefnick JM, Ryherd EE, Jackson R. Evaluating hospital soundscapes to improve patient experience. J Acoust Soc Am. 2019;145:1117-28. doi:10.1121/1.5090493.
- McCullagh MC, et al. Noise exposure and quality of life among nurses. Workplace Health Saf. 2022;70(4):207-19. doi:10.1177/21650799211044365.
- Lee FYL, Ismail NH, Liew PM, Lim SH. A cross-sectional study of occupational noise exposure and hypertension in Malaysia. Cureus. 2023;15(11):e48758. doi:10.7759/cureus.48758.
- Zhang L, et al. Relationship between occupational noise exposure and hypertension: cross-sectional evidence from real-world data. Front Public Health. 2022;10:1037246. doi:10.3389/fpubh.2022.1037246.
- Tahvili A, et al. Noise and sound in the intensive care unit: a cohort study. Sci Rep. 2025;15(1):10858. doi:10.1038/s41598-025-94365-8.
- Xun T, et al. Impact of hospital ward acoustic design on postoperative recovery of patients with acute myocardial infarction. Noise Health. 2025;27(127):534-44. doi:10.4103/nah.nah_62_25.
- Chung K. Calibration matters: I. Sound level meter basics. J Commun Disord. 2023;101:106300. doi:10.1016/j.jcomdis.2022.106300.
- Buxton OM, et al. Sleep disruption due to hospital noises: a prospective evaluation. Ann Intern Med. 2012;157(3):170-9. doi:10.7326/0003-4819-157-3-201208070-00472.
- Das A, et al. Effect of sleep quality on wound healing among patients undergoing emergency laparotomy: an observational study. J Clin Sleep Med. 2025;21(3):503-12. doi:10.5664/jcsm.10614.
- Dai Z, Cheng X, Chen F. Effect of comprehensive noise reduction management on postoperative negative emotion, stress response hormone, and sleep status of burn patients: a single-center retrospective analysis. Noise Health. 2024;26(122):410-6. doi:10.4103/nah.nah_80_24.
- Reid M. 131 Sleeping it off: are hospital wards too noisy to allow patient recovery Br J Surg. 2022;109(Suppl 6):znac269.384. doi:10.1093/bjs/znac269.384.
- Catley CD, et al. Redesigning the hospital environment to improve restfulness. JAMA Netw Open. 2024;7(12):e2447790. doi:10.1001/jamanetworkopen.2024.47790.
- Lu Y, et al. Effect of noise management in interventional radiology suites on patients undergoing transarterial chemoembolization: a retrospective study. Noise Health. 2024;26:325-31. doi:10.4103/nah.nah_53_24.
- De Salvio D, Amodeo V, Garai M, Secchi S. Noise assessment within healthcare facilities: annoyance and intrusiveness of sound sources [conference paper]. Presented at: Forum Acusticum Euronoise 2025; Malaga, Spain; 2025. https://doi.org/10.61782/fa.2025.0607.
- Amodeo V, Secchi S. Environmental noise in hospitals: a rapid review of trends and possible developments [conference paper]. Presented at: Forum Acusticum Euronoise 2025; Malaga, Spain; 2025. https://doi.org/10.61782/fa.2025.0348.
- Louis M, Grabill N, Strom P, Gibson B. Leading through noise: operating room noise challenges for staff and leadership techniques to ensure optimal operational performance. Cureus. 2024;16(9):e69569. doi:10.7759/cureus.69569.
- Fu VX, Oomens P, Merkus N, Jeekel J. The perception and attitude toward noise and music in the operating room: a systematic review. J Surg Res. 2021;263:193-206. doi:10.1016/j.jss.2021.01.038.
- Li X, et al. Effects of operating room noise on patient outcomes and medical staff: a systematic review. Noise Health. 2025;27(126):246-54. doi:10.4103/nah.nah_175_24.
- Gulsen M, Aydingulu N, Arslan S. Physiological and psychological effects of ambient noise in the operating room on medical staff. ANZ J Surg. 2021;91:847-53. doi:10.1111/ans.16582.
- Kennedy-Metz LR, et al. Association between operating room noise and team cognitive workload in cardiac surgery. IEEE Conf Cogn Comput Asp Situat Manag. 2022;2022:89-93. doi:10.1109/COGSIMA54611.2022.9830675.
- Idrees S, et al. Noise level and surgeon stress during thyroidectomy in an endocrine surgery operating room. Head Neck. 2024;46:37-45. doi:10.1002/hed.27552.
- Keller S, et al. Noise in the operating room distracts members of the surgical team: an observational study. World J Surg. 2018;42:3880-7. doi:10.1007/s00268-018-4730-7.
- Tseng LP, Chuang MT, Liu YC. Effects of noise and music on situation awareness, anxiety, and the mental workload of nurses during operations. Appl Ergon. 2022;99:103633. doi:10.1016/j.apergo.2021.103633.
- Ukegjini K, et al. Impact of intraoperative noise measurement on surgeon stress and patient outcomes: a prospective, controlled, single-center clinical trial with 664 patients. Surgery. 2020;167:843-51. doi:10.1016/j.surg.2019.12.010.
- Rypicz Ł, et al. Alarm fatigue and sleep quality in medical staff: a Polish-Czech-Slovak study on workplace ergonomics. Front Public Health. 2024;12:1345396. doi:10.3389/fpubh.2024.1345396.
- Lewandowska K, Mędrzycka-Dąbrowska W, Tomaszek L, Wujtewicz M. Determining factors of alarm fatigue among nurses in intensive care units: a Polish pilot study. J Clin Med. 2023;12(9):3120. doi:10.3390/jcm12093120.
- Cingolani M, De Salvio D, D’Orazio D, Garai M. Clustering analysis of noise sources in healthcare facilities. Appl Acoust. 2023;214:109660. doi:10.1016/j.apacoust.2023.109660.
- Hummel K, Ryherd E, Cheng X, Lowndes B. Relating clustered noise data to hospital patient satisfaction. J Acoust Soc Am. 2023;154(2):1239-47. doi:10.1121/10.0020760.
- Sangari A, et al. Delirium variability is influenced by the sound environment (DEVISE study): how changes in the intensive care unit soundscape affect delirium incidence. J Med Syst. 2021;45(8):76. doi:10.1007/s10916-021-01752-5.
- Yang L, Gutierrez DE, Guthrie OW. Systemic health effects of noise exposure. J Toxicol Environ Health B Crit Rev. 2024;27:21-54. doi:10.1080/10937404.2023.2280837.
- Giv MD, et al. Evaluation of noise pollution level in the operating rooms of hospitals: a study in Iran. Interv Med Appl Sci. 2017;9(2):61-6. doi:10.1556/1646.9.2017.2.15.
- Armbruster C, et al. Noise exposure among staff in intensive care units and the effects of unit-based noise management: a monocentric prospective longitudinal study. BMC Nurs. 2023;22(1):460. doi:10.1186/s12912-023-01611-3.
- Arcangeli G, et al. Neurobehavioral alterations from noise exposure in animals: a systematic review. Int J Environ Res Public Health. 2022;20:591. doi:10.3390/ijerph20010591.
- Lechat B, et al. Environmental noise-induced cardiovascular responses during sleep. Sleep. 2022;45:zsab302. doi:10.1093/sleep/zsab302.
Reprints and Permissions
Request permission to reuse the text or figures of this JoVE article
Request Permission