Invited Perspective: Diabetes and Road Traffic Noise at the Most and Least Exposed Façade.

A wide variety of environmental exposures have been linked to diabetes (Dendup et al. 2018). Recently, a meta-analysis of five cohort studies reported a positive association between road traffic noise and type 2 diabetes (Vienneau et al. 2019a). In a new report in Environmental Health Perspectives, Thacher et al. (2021) evaluated potential associations of long-term exposure to road, railway, and aircraft noise with type 2 diabetes in a nationwide cohort from Denmark. A major strength of this study is the inclusion of transportation noise at the most and least exposed residential façade. Health effects of transportation noise are believed to be mediated partly through sleep disturbance (Héritier et al. 2018; Münzel et al. 2017). As modeled day- and nighttime transportation noise are generally very strongly correlated with one another, it is impossible to disentangle effects of these exposures. However, many dwellings have quiet(er) façades, which could be where the bedroom would be located. Noise at the least exposed façade may better describe exposure during sleep, and studies including transportation noise at the most and least exposed façades could help to indicate whether daytime noise and noise during sleep are biologically relevant exposures for disease development.

Thacher et al. (2021) observed a moderate correlation (Spearman ) between road traffic noise during the day, evening, and night () at the most and least exposed façade. This could be explained by several factors. Road traffic noise is mainly produced by the combustion engine and contact of wheels with the road and is transmitted by pressure waves. As noise waves spread out over the area, their energy decreases relatively quickly, and noise levels drop (Davies et al. 2009). Moreover, dense and complex road networks and building configurations can affect noise levels because noise waves can be reflected and refracted by buildings and noise barriers.

In the study by Thacher et al. (2021), associations of road traffic noise at the most and least exposed façades with diabetes were quite similar in the minimally adjusted model. However, after adjustment for socioeconomic status (SES), green space, and ambient fine particulate matter , associations of 10-y mean road traffic noise at the most exposed façade attenuated from [95% confidence interval (CI): 1.07, 1.08] to 1.03 (95% CI: 1.03, 1.04) per 10-dB increase. Associations of 10-y mean road traffic noise at the least exposed façade, on the other hand, did not change after adjustment for these covariates [ (95% CI: 1.07, 1.09) per 10-dB increase], indicating that correlations with these covariates differ between both exposures. Correlations with nitrogen dioxide () were slightly stronger with road traffic noise at the least exposed façade, but associations persisted after adjustment for .

No information about lifestyle factors was available in this study, similar to other administrative cohorts. One may think that the lack of adjustments for lifestyle factors may have led to biased effect estimates because factors such as diet and physical activity are important predictors of type 2 diabetes. However, lifestyle factors can only be considered confounders if they are related to the outcome and the exposure. In the complete absence of a relationship between lifestyle factors and road traffic noise, additional adjustment for lifestyle factors would not affect associations. Given that associations of road traffic noise at the least exposed façade were not affected by adjustment for individual- and area-level SES and that SES is, to a certain extent, related to lifestyle factors, the impact of the lack of adjustment for lifestyle factors might be limited.

When road traffic noise at the most and least exposed façades were included simultaneously in the model, Thacher et al. (2021) observed that the HR of noise at the most exposed façade attenuated, whereas the HR of noise at the least exposed façade barely changed. If the assumption that bedrooms are generally located at the least exposed façade holds, this finding suggests that any hazardous effect of road traffic noise on diabetes would occur mainly through effects during sleep.

Two studies evaluated associations of traffic noise with health outcomes and had information about the location of the bedroom (Babisch et al. 2014; Foraster et al. 2014). Foraster et al. (2014) reported stronger associations of outdoor traffic noise at the postal address with hypertension, compared to associations of outdoor traffic noise at the bedroom façade or indoor traffic noise in the bedroom. For systolic blood pressure, associations of traffic noise exposure at the postal address disappeared after adjustment for , whereas associations of traffic noise exposure at the bedroom façade and indoors in the bedroom remained (Foraster et al. 2014). Babisch et al. (2014) found a positive association of road traffic noise exposure of the living room—but not with road traffic noise exposure of the bedroom—with hypertension. Both studies reported that of the participants had the bedroom at the rear of their house (Babisch et al. 2014; Foraster et al. 2014). The location of bedrooms is likely dependent on room structure and building types; therefore, the percentage of bedrooms at the least exposed façade will vary by study area and could affect results.

A few other studies have evaluated associations of transportation noise at the most and least exposed façades with health outcomes without information about the location of the bedroom. A study in Denmark observed no clear association of road traffic noise at the least exposed façade with redemption of prescriptions for sleep medication, whereas road traffic noise at the most exposed façade was positively associated with redemption of prescriptions for sleep medication (Roswall et al. 2020). In the Swiss National Cohort, associations of road traffic noise with myocardial infarction mortality were twice as strong at the most exposed façade as at the least exposed façade (Vienneau et al. 2019b). Two studies in Denmark reported stronger associations of road traffic noise at the most exposed façade with incident stroke and mortality due to all causes, cardiovascular disease, and stroke compared with associations of road traffic noise at the least exposed façade (Sørensen et al. 2021; Thacher et al. 2020). Only for ischemic heart disease mortality stronger associations were observed with road traffic noise at the least exposed façade (Thacher et al. 2020). These differences in associations may indicate that the importance of the timing of road traffic noise exposure differs between health outcomes.

In conclusion, the well-conducted study by Thacher et al. (2021) makes an important contribution to the existing literature about transportation noise and diabetes. Future studies that include transportation noise at the most and least exposed façades could help to indicate whether daytime noise and noise during sleep are biological relevant exposures for disease development.