Spatiotemporal Assessment Of The Effect of Indoor Noise On Annoyance, Health And Comfort In Hospital Wards And Intensive Care Unit

Hospital-noise levels can induce physiological responses, and affects sleep quality, which could contribute to cardiovascular-related health problems. Till date, high resolution hospital noise exposure assessment studies have not received much attention in Oman. This study aims at assessing sound pressure levels across hospital wards and intensive care unit (ICU) rooms to determine annoyance, and potential health effects based on perception and risk estimates. An indoor exposure assessment using high precision noise sensors was conducted in a female medical ward (FMW), isolated ward (SLW), emergency ward (EMW), and intensive care Unit (ICU) in a public hospital in Muscat city, Oman. Self– administered questionnaire was randomly distributed among respondents using both online and eld survey approach to ascertain annoyance, health effects, and potential risks associated with exposure. The study found that 24–hr noise levels (L Aeq ) ranged from 55.2–61.7 dB(A) in the hospital wards and ICU rooms, which exceeded WHO’s hospital indoor rooms critical limit of 35 dB(A) by 58% – 76%. A total of 150 participants took part in the survey. Among the respondents, 53% reported moderate annoyance at the hospital wards while 56% felt sensitivity to the noise levels. Noise annoyance was reported by the majority of the patients across the various wards and emergency rooms as causing slight annoyance (50%) and intermittent sleep disturbances (49%). The majority (73%) of the medical staff have complained that the current noise levels affect overall work performance (p = 0.004), while 70% of them have further complained of it as a cause of workplace distraction (p = 0.011). Quantitative risk estimates showed that both the % of highly annoyed (HA) persons (16%), and highly sleep–disturbed (HSD) persons (9%) were very high in FMW, while ICU was found to have the lowest risk. However, the greater number of the respondents (87%) believed there are possibilities of mitigating (p <0.001) the current noise levels.


Introduction
Noise pollution exposures continue to pose a serious public health threat in a built environment especially as businesses, infrastructural developments continue to grow in response to ever increasing global population (Goines and Hagler, 2007;To et al., 2015;Yuan et al., 2019). While, noise levels were recognized as a common environmental hazard to people living closer to airports, road-ways, harbors, wind farms, and industrial areas (Münzel et al., 2018), there has also been a growing concern about its impacts on health care microenvironments (Montes-González et al., 2019). Short-term noise exposure was found as a contributor to physiological problems (vascular and cerebral dysfunctions), mental stress (Daiber et al., 2020), metabolic syndrome (Huang et al., 2020), annoyance, and sleep disturbance (Park et al., 2015). These health effects may exacerbate especially in indoor commercial areas and workplaces as most people spend about 90% of their time in these environments with continuous noise exposures  Sułkowski, 2009). Since noise pollution is considered as everyday environmental problem, it is imperative to adopt intervention measures to reduce these exposures to help minimize the associated adverse health effects and discomforts. Several mitigation approaches have been developed, these include the use of acoustic plasters (Magrini and Lisot, 2015), installation of transducers (Sohrabi et al., 2020), others have also proposed reduction of air ow in naturally ventilated buildings (Oldham et al., 2004) and use of porous sound absorbers (Cobo and Simón, 2019). In addition, since outdoor noise pollution from road tra c, aircraft and industries are also major contributors to indoor noise levels, it is therefore important to tackle these sources as well. Here, increasing urban greenery, use of noise of barriers made with sustainable materials, and sustainable urban tra c planning, and pavement designs have been identi ed as effective mitigation measures (Ferrini et  . Interestingly, an investigation was carried out to evaluate sleep disturbances caused by the noise produced in hospital indoor environments using home settings as control among both adults and children. The study reported low sleep quality in pediatric rooms (77.1%) compared to homes (88.9%) with measured median noise levels of 48.6 and 34.7 dBA, respectively (Bevan et al., 2019). It is known that the degree of noise levels could be in uenced by spatio-temporal, socio-cultural and lifestyle variabilities among populations (Gjestland, 2007;Minoura et al., 2015).
In the case of Oman, there is limited knowledge about noise exposures and their annoyance levels in hospital environments. This study was conducted at a Government Teaching Hospital (GTH), one of the major government hospitals in Oman with more than 3000 staff. The main objective of this present study is to assess the noise levels, reported annoyance, and health effects on populations dwelling in various wards and emergency rooms in GTH in Oman. This was achieved by using a mixed approach of eld measurement, social survey via questionnaires, and quantitative risk approaches. Also, the observed noise levels is compared with both local and international standards, and recommendations were made based on hospital management and urban planning point of view.

Study Area
This study was carried in Muscat, the capital city of Oman. According to National Centre for Statistics and Information (NCSI), the total population of Oman is about 4.5 million people, while Muscat city has a current resident population of about 1.3 million inhabitants (NCSI, 2020). The city has a normal average outdoor temperature (23-34 ºC) and humidity levels of 67% especially during winter and spring seasons except summer where elevated conditions are experienced (Amoatey et al., 2020b). These harsh summer climatic conditions are incorporated into architectural designs of most buildings including hospitals to protect people from exposure to harsh outdoor environmental conditions. However, these enclosed building designs are associated with increasing indoor noise levels (Amoatey et al., 2020c;Kraus and Juhásová Šenitkov, 2017). The GTH is rated the second largest hospital in Oman with yearly in-patients and out-patient population of 31,853 and 408, 163, respectively, served by a total of 3,084 staff including medical doctors and nurses as indicated in Table 1 (PSD, 2018). The aerial view of the GTH and its neighborhood is presented in Fig. 1. The hospital is located at the center of a University campus with no urban residential building or industry located in close proximity except residential apartments for hospital staff and student accommodation.
However, the forecourt of the hospital has networks of several roads, sometimes. creating heavy vehicular tra c during morning rush hours thereby serving as one of the major sources of tra c noise. Especially the roundabout (about 200 sqm) lying just about 15 m closer to the hospital main gate is the main source of tra c noise levels to the University community. Due to the high tra c population at University, the hospital has made a provision for more than three different small car parks for both patients and visitors.
These car parks exit the main road to Muscat City which lies about just 650 m away from the hospital.

Noise Measurements
Measurements were conducted across different wards and emergency rooms in the GTH. for a period of four days (Table 1). using a high-level precision noise meter. These monitoring locations were female medical ward (FMW), isolated ward (SLW), emergency ward (EMW), and intensive care unit (ICU) ward as indicated by points (i.e. position of the noise meter) P1, P2, P3, and P4, respectively. The rationale for the selection of these four main wards is due to the frequent availability of patients and hospital staff, making them more vulnerable to noise pollution. The detailed description of measurement dates, and locations of the four wards are shown in Table 2   sample was estimated by taking into account the maximum margin error (5%), and con dence interval (CI) of 95% according to the following formula: n = sample size, p = assume the proportion of people likely to experienced annoyance and sleep disorders, z = the con dence interval required, and E = the maximum margin error.
A survey was conducted among medical staff, nurses, medical students at GTH on 29th October 2019 through their emails. Since most of the patients and accompanying people who visit the hospitals do not have emails, eld surveys through interviews were adopted by ensuring that the survey included all important noise hotspots of the hospital (wards, outpatient departments, and reception hall).

Questionnaire
A questionnaire survey based on noise level and perception. First, a structured questionnaire was designed in both Arabic and English languages as these are the most commonly used languages in Oman. The rst component of the questionnaire was based on noise exposures where topics such as sources of noise, duration of low/high noise, and mitigation were included. To achieve the overall health impact of noise pollution, the second part of the questionnaire covered a wide range of different topics such as the impact of noise on daily activities (e.g. work), noise annoyance, sensitivity, sleep disturbance, potential health effects approach. Being the main focus of the study and to ensure a clear response from the interviewees, noise annoyance and sleep disturbance were assessed by using 4-point verbal Likert scales (Lei, 1994). Using the 4-point verbal for annoyance question; "To what extent are you annoyed from hospital noise?" The order of responses ranging from high to low /no annoyance were provided as "Extremely", "Moderately", "Slightly" and "Not at all" (Lei, 1994). Statistical analysis through chi-square test was performed to determine the level of signi cance (p < 0.05) of the survey data using 95% con dence limits (CI). The statistical test was carried out by using SPSS software (Version 23).

Risk Estimates
For risk estimates only, the study employed WHO's quantitative exposure-response models shown in Eqs. 1 and 2 to predict the percentage risk of highly annoyed (%HA) and highly sleep-disturbed (%HSD) personswhile the measured noise levels under this study were expressed as L Aeq , %HA, and %HSD models require the input noise data to be in day, evening, night (L den ), and nighttime (L night ), respectively. To

Noise Exposure Assessment
The assessment focused on evaluating levels of noise exposures among four major wards (i.e. FMW, SLW, EMW, and ICU) in SQUH based on a total of 288 high-resolution data points for different sound level indicators as shown in Table 3. It can be observed that there was a signi cant increase in noise levels (L Aeq ) across all the wards during the measurement campaign. Among the wards, FMW recorded the highest noise levels of 61.7 dB(A) followed by SLW (58.8 dB(A)), while the lowest of 55.2 dB(A) was found in ICU. It is evident that the measured noise levels across the wards were very high as L 90 was found ranging from 49.3 to 57.3 dB(A), and L 10 having minimum and maximum noise levels of 55.8 dB(A) to 64.2 dB(A), respectively during the 24-hour measurement duration. In general, the noise levels (L Aeq ) among all the wards were higher than WHO's critical limit of 55 dBA recommended for a typical outdoor urban environment with exception of ICU where narrow exceedance was found (WHO, 2011).
Considering the GTH wards as indoor microenvironments, the WHO's community noise (L Aeq ) exposure guidelines recommend 35 dB(A) as acceptable limits that could minimize moderate annoyance and sleep disturbances to the exposed population/building occupants (WHO, 1999). The same guideline has set more stringent threshold limits of 30 dB(A) speci cally to hospital wards to help improve sleep performance levels among patients (WHO, 1999). Another more relaxed indoor residential noise exposure levels guideline of 45 dB(A) has been developed to serve public health advice in order to reduce speech interference and annoyance levels (US EPA, 1974). Based on the WHO (1999) standard of 35 dB(A), the noise levels in the hospital wards were exceeded by 76%, 68%, 65%, and 58% for FMW, SLW, EMW, and ICU, respectively (Table 3). It can therefore be emphasized that these sound pressure levels are the actual noise levels found within the GTH. This is because measurements were taken within indoor spaces of the wards thereby minimizing the intrusion of external noise sources such as tra c (Amoatey et al., 2020a).

Health Risk
Elucidating the estimates of annoyance and related sleep disturbances outcomes from noise exposures in the GTH is crucial in mitigating long-term health risks among the affected population including patients and staff. Table 5 provides a summary of %HA and %HSD from 24-hour exposure noise levels (L Aeq ) observed among various wards at the GTH. It was revealed that the estimated %HA (16.2%) and %HSD (9.2%) persons were very high in FMW, while ICU was found to have the lowest risk of annoyance and sleep disturbances. These ratiometric risk estimates offer an opportunity to ascertain a larger number of people in SQUH including similar hospitals in Oman as other gulf countries who might be at risk from noise annoyance and sleeping disturbance problems.  7%). The participants were mostly accompanying persons (25%), patients (21%), nurses (19%), and students (19%) with both doctors and visitors being < 10%. From a noise exposure assessment standpoint, 63.3%, 26%, and 10.7% of the population who took part in the survey were found in the wards, OPD, and at the reception hall, respectively. The majority (58.7%) of the participants do not know about their chronic disease history. However, the proportion of those who have complained of chronic diseases (p < 0.001) such as headache, sleeping di culties, anxiety, tinnitus, and heart disease problems during the study period was approximately 21.3%, 9.3%, 6%, 4%, and 0.7%, respectively (Table 6). complained of the noise levels as causing extreme annoyance (Table 6). In the response to the question about their sensitivity (p = 0.14) to noise exposures, a large proportion of the respondents (56%) reported feeling sensitivity while 44% of them did not feel any sensitivity about the current noise levels. The perception about future noise mitigation measures showed that a signi cant (p < 0.001) number of the respondents (87%) believed there are possibilities of reducing the current noise levels (Table 6). For questions about sources of noise at the GTH, respondents stated that visitors walk (footsteps) outsides the wards (49.3%), ringing of phone (42%) and operation of medical equipment (40.6%) produces moderate noise levels. Also, about 25.3% of the respondents reported that visitors walk (footsteps) in the wards, and alarms of medical devices generate high noise levels (Fig. 6).
The study further conducted across survey speci cally among patients, accompanying persons, medical staff at the GTH to better understand the extent to which noise exposures affects them including quality of work in the case of medical staffs ( Table 7). The reason is that these groups spend more than 80% of their time in the hospital indoor environment, making them more vulnerable to noise levels compared to visitors and students (Alhorr et al., 2016). Here, most of the patients (50%) are of the view that the noise levels cause slight annoyance, while 37% of them expressed feeling annoyed (p = 0.03). A very high (87.5%) proportion of the patients stated that the noise levels do not cause any disruption in communication (p < 0.001) between them and the medical staff, while 46.87% of them mentioned that it sometimes causes sleep disturbances (p = 0.03). However, 41% of the patients responded that the noise levels cannot affect their illness recovery while 22% believes that it can somewhat affect the rate of recovery ( Table 7).
The medical staff seems more vulnerable to occupational noise exposures as about 72.5% of them complained that the current noise levels affect their overall working performance levels (p = 0.004). Again, when a question about how noise affects their day-to-day work quality, 70% stated that it causes loss of concentration during work (Table 7). For the accompanying persons, 40.5% of them feel noise exposure sometimes causes sleeping disturbances (p = 0.06), and this is similar to the proportion (40% ) who reported the current noise levels to cause slight annoyance. Also, about 66.5% of the accompanying persons admitted that noise can aggravate patient health while 14.5% of them do not know of such potential health impacts (Table 7).

Discussions
The high noise levels from this study have exceeded both WHO (1999) and US EPA (1974) recommended critical limits for hospital rooms/wards, and residential indoor dwellings, respectively by 58-76%. The study observed signi cant differences in sound pressure levels (L Aeq ) between full-capacity children and female wards, represented by 61.7 dB (A) compared to the same type of ward (58.8 dB (A)) with a fewer number of patients. This clearly indicates that the sources of these noise exposures are attributed to human-induced activities, and thus warrant immediate mitigation actions. It must be acknowledged that the current noise levels could cause annoyance and facilitates the delaying of recovery time of patients especially the aged populations with underlying chronic diseases (Loupa, 2020). The health risk estimates from this study revealed that a quite substantial proportion of the exposed population in the GTH have been exposed to hazardous noise levels. These include patients, accompanied persons, and medical staff who have been affected by high annoyance and sleep disturbance problems. These predicted health outcomes may exacerbate considering the disparities among the exposed population as well as the duration and intensity of the noise exposure across the hospital wards.
The outcome from this study revealed some general consistencies with previous studies from the literature. Hospital noise levels (L Aeq ) in some speci c microenvironments including ICU, emergency departments, waiting rooms, and pediatric clinics were found ranging from 37-

Conclusions
This study presents spatiotemporal noise exposure levels measured across different wards and emergency rooms in the GTH in Muscat, Oman. To ascertain the annoyance levels and health impacts of these exposures, a survey study using online and eld questionnaires were carried out among SQUH patients, medical staff, and the accompanied patients. It was found that the 24-hour noise levels (L Aeq ) in the hospital wards have exceeded WHO (1999) standard of 35 dB(A) by 76%, 68%, 65%, and 58% for FMW, SLW, EMW, and ICU, respectively. The noise levels were highest in EMW for both morning and night time periods while FMW also experienced similar highest noise levels during the visiting hours of the day. Thus, none of the wards including ICU had noise levels found below WHO's standards set for hospital rooms, indoor residential areas, and urban residential environments thereby posing a serious public health threat to the patients. Noise annoyance has been reported by the majority of the patients across the various wards and emergency rooms as causing slight annoyance and intermittent sleep disturbances. The majority (72.5%) of the medical staff have complained that the current noise levels affect overall work performance, while 70% of them have further complained as a major cause of workplace distraction. Quantitative risk estimates using WHO's dose-response model showed that both %HA (16.2%) and %HSD (9.2%) persons were very high in FMW, while ICU was found to have the lowest risk of annoyance and sleep disturbances. Overall, the noise was caused by human-induced activities including visitors walk (footsteps) outside wards (49.3%), ringing of phone (42%), and operation of medical equipment (40.6%). However, a very large number of the respondents (87%) believed there are possibilities of mitigating the current exposure levels to help improve acoustic comfort in the hospital.

Con ict of Interest
Authors declare no potential competing con ict of interest

Ethics Approval and Consent to Participate
Participants who enrolled for the online surevy were provided with an option to "accept" OR "decline" completing the questionnaire .Verbal consent was saught for participants who enrolled for the eld interview.

Availability of Data and Materials
Not applicable   Visiting periods temporal variations of hourly noise (LAeq) levels across GTH wards Night times temporal variations of hourly noise (LAeq) levels across the GTH wards (FMW, SLW, EMW, ICU represents female medical ward, isolated ward, emergency ward and intensive care unit ward, respectively).

Figure 6
Respondents impression towards the sources of noise at the hospital