This was the first study that assessed the association between nocturnal noise, noise sensitivity, and NRS by using standardized scales and adjusting for relevant confounders. The average NRS level of Chinese adults in Hong Kong was 64.77 ± 12.75. The results revealed that noise sensitivity, as well as lifestyle, sociodemographic attributes, physical and psychosocial health were associated with NRS.
The average nocturnal noise level in the participants’ bedrooms was 51.32 ± 5.61 dBA, which was beyond the suggested level of 30–40dBA at night [27]. NRS was independent of nocturnal noise level while TIB and TST were shown to be negatively associated with noise in this study. This was consistent with a previous study that revealed the relationships of noise, noise annoyance, and objective and subjective sleep parameters [28]. People who live in noisy areas may be more easily influenced by noise, and tend to have less TIB [29]. Furthermore, noise has been shown to fragment sleep, reduce sleep continuity and TST, and increase awakenings and shifts between stages of sleep [30]. These factors may result in less “asleep” time of the individuals.
Consistent with our hypothesis, there was a direct association between noise sensitivity and NRS during the night-time. We found that every unit increase of WNSS (0–100) resulted in a 0.08 unit decrease of NRSS (0–100), but was not significantly associated with changes in physiological sleep parameters after adjusting for nocturnal noise, sociodemographics, lifestyle factors, physical and psychosocial health. It is unlikely that people who are more noise sensitive would have substantially more emotional responses to noise when they are asleep, as they should then also have more changes in their physiological sleep levels under the general stress model. Alternatively, the potential influence of noise sensitivity on NRS may be attributed to the higher vulnerability of noise sensitive people to day-time noise exposure whose stress-related responses have not recovered by the time they go to sleep. Further studies that incorporate day-time noise measurements will be necessary to confirm this hypothesis. In addition to the global scale of the NRSS, higher noise sensitivity was also associated with the less refreshment from sleep and more affective symptoms scales. The refreshment from sleep scale shares similar interpretation as the global scale. For affective symptoms, they have been shown to be influenced by introversion and extroversion [31], which are two personality traits that have also been shown to be associated with noise sensitivity [12].
Individuals who had higher family income had less NRS in this study. Previous study indicated that sleep quality declined with decreasing income [32]. Individuals with higher family income may have less living stress which may help decrease NRS level.
In this study, people who exercised more were less vulnerable to NRS. Previous studies showed that a combined unhealthy lifestyle, which included low or no exercise, would lead to NRS [7], whereas higher intensity exercise was associated with shorter sleep latency and fewer awakenings [33]. Exercise could shorten the N1 stage of non-rapid-eye-movement sleep, which is the stage of very light sleep, while increasing rapid-eye-movement (REM) sleep, sleep continuity, and sleep efficiency [34]. Therefore, exercise could be beneficial for relieving NRS.
Married or cohabited people suffered less NRS when compared with people who were single. A previous study also showed that married adults had better sleep characteristics measured by ActiGraph, such as total sleep time and wake after sleep onset [35]. This may due to the supportive relationship that helps improve sleep quality [36]. Indeed, our study also indicated that people with higher social support had less NRS. Adequate social support is a stress reliever, and it can strengthen individual resilience to stress [37].
This study showed that PHQ was associated with NRS. NRS was suggested to be associated with chronic medical disorders [3]. For example, NRS is common in patients with fibromyalgia syndrome [38]. Individuals who suffer from somatic symptoms like pain may have difficulty falling asleep and more awakenings after sleep onset, which can then increase the risk of poor sleep and NRS.
This study also demonstrated that stress, anxiety, and depression were associated with increased NRS as reported by previous study. Stressful events and psychosocial stressors might induce increased sleep latency and awakenings, and reduced slow wave sleep (SWS) and sleep efficiency [39]. Anxious individuals were reported to have longer sleep latency, a smaller percentage of SWS, more early microarousals, and a lower REM sleep density compared with those with lower levels of anxiety [40]. Lastly, depressive individuals were characterized by a decrease of SWS and sleep efficiency in addition to an increase of the percentage and density of REM [41]. Conclusively, people who are anxious, depressive, and experiencing stressful events may be more vulnerable to NRS.
Lastly, this study indicated that none of the obtained physiological sleep parameters in this study were associated with NRS. Scholars proposed a possible association between alpha activity and NRS among people with chronic fatigue syndrome. However, the fact that people without such symptoms also had NRS caused doubt on this link [1]. As there remains a lack of study on the relationship between objective physiological sleep parameters and NRS, a more precise and functional device which can reveal more sleep parameters is worthy of research to test the relationship between specific physiological sleep parameters and NRS.
Daytime noise, which may also influence NRS and physiological sleep parameters, was not investigated in this study. This would require the assessment of personal noise exposure by asking the participants to carry a noise dosimeter all day. However, this can be highly demanding to the participants and innovation in assessing personal noise exposure would be helpful in this regard. In addition, despite controlling a number of covariates, factors like light were not investigated in this study; further study could be conducted to fill this gap. Moreover, the mechanism underlying why noise sensitive people were more vulnerable to NRS is worth studying since people are considered to be less reactive to the outside environment when sleeping.