| Article Access Statistics|
| Viewed||2429 |
| Printed||100 |
| Emailed||0 |
| PDF Downloaded||23 |
| Comments ||[Add] |
|Year : 2022
: 24 | Issue : 113 | Page
|Relationships between long-term residential exposure to total environmental noise and stroke incidence
Larisa I Yankoty1, Philippe Gamache2, Céline Plante3, Sophie Goudreau3, Claudia Blais4, Stéphane Perron5, Michel Fournier3, Martina S Ragettli6, Marianne Hatzopoulou7, Ying Liu1, Audrey Smargiassi8
1 School of Public Health, Centre of Public Health Research, University of Montreal and CIUSSS du Centre-Sud-de-l’Île-de-Montréal, Montreal, Canada
2 Quebec National Institute of Public Health, Quebec, Canada
3 Montreal Regional Department of Public Health, Montreal, Canada
4 Quebec National Institute of Public Health, Quebec, Canada; Faculty of Pharmacy, Laval University, Quebec, Canada
5 School of Public Health, Centre of Public Health Research, University of Montreal and CIUSSS du Centre-Sud-de-l’Île-de-Montréal, Montreal, Canada; Quebec National Institute of Public Health, Quebec, Canada
6 Swiss Tropical and Public Health Institute, Basel, Switzerland; University of Basel, Basel, Switzerland
7 Department of Civil Engineering, University of Toronto, Toronto, Canada
8 School of Public Health, Centre of Public Health Research, University of Montreal and CIUSSS du Centre-Sud-de-l’Île-de-Montréal, Montreal; Quebec National Institute of Public Health, Quebec, Canada
Click here for correspondence address
|Date of Submission||02-Jun-2021|
|Date of Decision||08-Oct-2021|
|Date of Acceptance||20-Oct-2021|
|Date of Web Publication||25-Jul-2022|
Background: Noise has been related to several cardiovascular diseases (CVDs) such as coronary heart disease and to their risk factors such as hypertension, but associations with stroke remain under-researched, even if CVD likely share similar pathophysiologic mechanisms. Aim: The objective of the study was to examine the association between long-term residential exposure to total environmental noise and stroke incidence in Montreal, Canada. Materials and Methods: We created an open cohort of adults aged ≥45years, free of stroke before entering the cohort for the years 2000 to 2014 with health administrative data. Residential total environmental noise levels were estimated with land use regression (LUR) models. Incident stroke was based on hospital admissions. Cox hazard models with age as the time axis and time-varying exposures were used to estimate associations, which were adjusted for material deprivation, year, nitrogen dioxide, stratified for sex, and indirectly adjusted for smoking. Results: There were 9,072,492 person-years of follow-up with 47% men; 26,741 developed stroke (21,402 ischemic; 4947 hemorrhagic; 392 had both). LUR total noise level acoustic equivalent for 24 hours (LAeq24h) ranged 44 to 79 dBA. The adjusted hazard ratio (HR) for stroke (all types), for a 10-dBA increase in LAeq24h, was 1.06 [95% confidence interval (CI): 1.03–1.09]. The LAeq24h was associated with ischemic (HR per 10 dBA: 1.08; 95% CI: 1.04–1.12) but not hemorrhagic stroke (HR per 10 dBA: 0.97; 95% CI: 0.90–1.04). Conclusion: The results suggest that total environmental noise is associated with incident stroke, which is consistent with studies on transportation noise and other CVD.
Keywords: Cardiovascular disease, cerebrovascular disease, environmental noise, traffic-related air pollution
|How to cite this article:|
Yankoty LI, Gamache P, Plante C, Goudreau S, Blais C, Perron S, Fournier M, Ragettli MS, Hatzopoulou M, Liu Y, Smargiassi A. Relationships between long-term residential exposure to total environmental noise and stroke incidence. Noise Health 2022;24:33-9
|How to cite this URL:|
Yankoty LI, Gamache P, Plante C, Goudreau S, Blais C, Perron S, Fournier M, Ragettli MS, Hatzopoulou M, Liu Y, Smargiassi A. Relationships between long-term residential exposure to total environmental noise and stroke incidence. Noise Health [serial online] 2022 [cited 2022 Dec 1];24:33-9. Available from: https://www.noiseandhealth.org/text.asp?2022/24/113/33/351965
| Background|| |
Cardiovascular effects of environmental noise (i.e., noise from outdoor sources) are a growing concern among the general public and policy-makers. The pathophysiologic mechanism believed to underlie the effects of noise on the cardiovascular system is based on the general stress theory. Exposure to environmental noise levels and noise-induced sleep disturbances would trigger a cascade of reactions in the brain and the endocrinoimmunologic system, leading to modifications of physiologic parameters such as blood pressure, lipid and glucose levels, etc.- Such physiologic disturbances in the long would initiate or worsen cardiovascular diseases (CVDs) such as coronary heart diseases (CHDs) including myocardial infarction (MI), cerebrovascular diseases such as stroke and their metabolic risk factors such as hypertension (HT).- Annoyance induced by noise exposure has also been associated with atrial fibrillation which could lead to CVD-like stroke. Although the pathophysiologic mechanism related to different cardiovascular effects of noise is likely the same, associations between noise and stroke have been poorly studied compared to other problems such as HT and MI., Yet, stroke is the second leading cause of death and disability in the world and this trend is expected to increase in the coming years.
The few studies that have examined the relationship between environmental noise and the incidence of stroke and stroke mortality mainly assessed road and aircraft noise and they showed inconsistent results. Sørensen et al. found positive associations between road noise and all stroke types, after adjusting for road-traffic air pollution such as nitrogen dioxide (NO2). Sørensen et al. reported that contrary to ischemic stroke, hemorrhagic stroke was not associated with road noise. It is worth noting that these two studies by Sørensen et al., were based on overlapping samples. Yet other studies reported no association between road noise and stroke incidence.,,
With regard to aircraft noise, a recent meta-analysis (seven studies) found a moderate borderline statistically significant association between aircraft noise and the onset of all stroke types. A meta-analysis on stroke mortality with three studies found a positive but not statistically significant pooled relative risk with aircraft noise. Pyko et al., however, reported a reverse association between aircraft noise and stroke incidence (all types combined) in a recent cohort study (not included in the meta-analysis by Weihofen et al.).
Associations between environmental noise and stroke onset were usually based on transportation noise (road traffic and aircraft mostly) estimated with propagation models.,, To our knowledge, no study has estimated associations between total environmental noise (all outdoor noise sources including road traffic, aircraft, neighborhood, etc.) and stroke onset. Studies recommended however to consider exposure to the combination of different noise sources in relation to cardiovascular problems because people are often exposed to several noise sources in their lives rather than specific and individual noise sources., Differences in estimates of association between transportation noise and stroke may be influenced by differences in total noise exposure levels. It would therefore be important to use exposure models that consider all environmental noise sources to examine the relationship with stroke. Land use regression (LUR) models are appropriate for this purpose, as they are based on real measurements of total environmental noise. Differences in associations between environmental noise and stroke incidence may also be influenced by population studied and adjustments made. Very few studies have been conducted in North America and whether results from European studies are generalizable is unknown because of the differences in urbanization, climatic conditions, types of construction, etc. In addition, not all studies have controlled their association for traffic-related air pollutants. Finally, it remains to clarify whether environmental noise is more related to ischemic or to hemorrhagic strokes.
Therefore, the objective of the present study was to examine the association between long-term residential exposure to total environmental noise and the incidence of stroke and its subtypes (ischemic and hemorrhagic) in the adult population of Montreal, Canada, adjusting for traffic-related air pollution.
| Methods|| |
We use the Quebec Integrated Chronic Disease Surveillance System (QICDSS) to create a retrospective open cohort of adults living on the Island of Montreal between April 1, 2000 and March 31, 2014. The QICDSS links five health administrative databases sharing a common health insurance number such as the health insurance registry which contains information on birth date and sex of all people born and living in Quebec and residential six-character Canadian postal codes. Among other databases included, hospital discharges is the most important for the present study. QICDSS start date is 1996.
To be included in the cohort, individuals were to be aged 45 years and over; people in this age group are more likely to develop CVD and its metabolic risk factors. Subjects also had to reside in Montreal for at least 4 years at the same residential address and to have no history of hospitalization for stroke. We used previous data of at least 4 years prior to joining the cohort to exclude prevalent cases of stroke (for individuals who entered in 2000, for example, 1996–2000 data were used). This 4-year period was chosen to ensure a long follow-up and is also in line with methods of previous studies., Participants were followed from entry into the cohort until their stroke event, moving outside of Montreal, death, or the end of the study.
Incident stroke diagnosis
We identified incident stroke cases based on the first hospital admission with the appropriate diagnostic codes from the International Classification of Diseases, ninth (ICD-9) and tenth revision (ICD-10-CA, Canadian version), as suggested by the Public Health Agency of Canada. They included for ischemic stroke, ICD-9: 362.3, 433.x1, 434.x1, 436; and for ICD-10-CA: H34.1, I63 (I63.6 excluded), I64. For hemorrhagic stroke they included, for ICD-9: 430, 431 and for ICD-10-CA: I60, I61. Primary and secondary causes of hospital admission were included in our analyses because analyses performed with the primary diagnosis only showed similar results. We studied both hemorrhagic and ischemic strokes combined and separately.
Total environmental noise exposure
We assessed the long-term exposure to total environmental noise from all outdoor sources using yearly estimates from LUR models. We used the term “total environmental noise” to refer to noise from all outside sources (multiple sources) to distinguish from noise from transportation sources only. Noise exposure was estimated at the residential six-digit postal codes of the subjects throughout their follow-up (2000–2014). When a person moved, a different yearly noise exposure value corresponding to the new residential postal code was assigned. Thus, a person followed for the 15 years (i.e., 2000–2014) and residing at the same address for the entire follow-up would have the same yearly noise estimate applied for all days of each year. If a person moved within a year, two different noise values were assigned, one for the days at the old address and one for the days at the new address in that year.
Six-digit postal codes correspond to a block side in dense areas and to larger areas outside the urban core of Montreal. The median area of six-digit postal code polygons for Montreal is 4478 m2. Population numbers are not available for postal codes but in the urban core, they can contain as few as 50 people.
Annual average total environmental noise levels (from all sources including transportation, neighborhood, construction, etc.; in decibels A, dBA) were based on the LUR models developed by Ragettli et al. The models are based on 204 full-week measurements in the summer of 2010 and spring of 2014. Continuous 2-minute noise measurements were used to model the A-weight equivalent noise levels over 24 hours (LAeq24h), day-evening-night levels (Lden), and night levels (Lnight). Lden represents noise levels over 24 hours with a weight of +5 dBA on the evening levels and +10 dBA on the night levels. The Lnight is for noise levels from 23:00 to 7:00. The Lden, Lnight, and LAeq24h were modeled with various land use variables related to roads, railways, aircrafts, and other variables of the built environment. LUR models explained 68%, 59%, and 69% of the spatial variability of environmental noise levels LAeq24h, Lden, and Lnight, respectively, in Montreal. We used the average noise estimates of 2010 and 2014 and assumed that there was no change in noise levels estimated between 2010 and 2014 and those of previous years (i.e., until 2000). The estimated LAeq24h, Lden, and Lnight levels beyond the range of measurements of Ragettli et al (about 0.4%) were set to the lowest and highest measurement values used to develop the models (48.4–74.8 dBA for LAeq24h; 52.2–79.4 dBA for Lden and 44.5–71.9 dBA for Lnight).
We obtained information on age and sex from the health insurance registry. We used the neighborhood socioeconomic status (SES) represented by the Montreal region material deprivation index, as individual information was not available in our database. The material deprivation index was developed based on Canadian census information on education, income and employment status; it is calculated for each Montreal dissemination area (the smallest census units reporting on SES; they include 400 to 800 individuals and regroup several postal codes). We assigned the quintiles of material deprivation index (Q1 = the least deprived, Q5 = the most deprived) to each participant via the residential postal code at the entry of the cohort and throughout the follow-up. We used values from the 2001, 2006, and 2011 census for the years 2000 to 2003, 2004 to 2008, and 2009 to 2014, respectively. We were unable to consider individual confounding factors related to lifestyle habits such as smoking, physical activity, diet, etc., due to the lack of information in our data file. However, we performed an indirect adjustment for smoking (i.e., sensitivity analysis) as described below.
Air pollution is also a possible confounding factor in the relationship between environmental noise and CVD. We estimated the long-term exposure to traffic-related air pollution (average annual levels of NO2) of each subject using the LUR model developed by Crouse et al. for Montreal. As data were only available for the years 2000 to 2006, we calculated the average NO2 levels for those years at each postal code and assigned this average to each year of follow-up, that is, from 2000 to 2014.
We used the Pearson coefficient r to evaluate the correlation between the exposure variables. Cox proportional hazard models with time-varying exposure levels were used to estimate the associations between total environmental noise (LAeq24h, Lden, and Lnight from LUR models) and the incidence of stroke (ischemic and hemorrhagic strokes combined). Associations were assessed for individuals with complete information on covariates. We used time-varying exposure levels to consider changes that occurred when people moved during the follow-up as mentioned above. We used age (in days) as the underlying time scale. All models were stratified by sex to consider the difference in the onset of stroke. Restricted cubic spline variables with three knots (quantiles 0.10, 0.50, and 0.90) were used to evaluate the linearity of the relationships with continuous exposure variables. The statistical significance of the nonlinear terms was evaluated with the likelihood ratio test (P < 0.05). We used product terms (i.e., interaction) between the exposure variables and age to assess the proportional hazard assumption (P < 0.05). We also assessed the effect of the interaction between noise estimates and NO2 with product terms. With each exposure variable, we first estimated the crude association. We then adjusted for calendar year (to consider potential changes in the risk over time); subsequent models additionally included either material deprivation or NO2. The full model was stratified by sex and adjusted for calendar year, material deprivation, and NO2. The full models were assessed separately for ischemic and hemorrhagic stroke. The hazard ratios (HRs) for stroke were expressed per 10 dBA increase in noise levels with 95% confidence interval (CI) for the continuous noise levels. Categorical analyses were also performed using the following noise levels categories: LAeq24h and Lnight: <55 dBA; 55 to 59.9 dBA; 60 to 64.9 dBA; ≥65 dBA; Lden: <60 dBA; 60 to 64.9 dBA; 65 to 69.9 dBA; ≥70 dBA.
Because individual information on smoking was not available in our database, we performed a sensitivity analysis (indirect adjustment) to assess its potential influence on the relationship between environmental noise and the incidence of stroke by using the method of Steenland and Greenland adapted by Villeneuve et al. [Figure S2].
| Results|| |
A total of 1,152,077 individuals who met the inclusion criteria were followed from 2000 to 2014. Of these people, we excluded 86,823 (7.5%) for missing data on exposure variables and material deprivation quintiles. The remaining 1,065,254 individuals with complete information provided 9,072,492 person-years of observation which were used for all the analyses carried out. Of this number, 119,788 (11.2%) were censored for death and 148,391 (13.9%) for moving out of Montreal. A total of 464,185 moves (from 312,291; 29.3% individuals) occurred within Montreal during the study period. During the follow-up, 26,741 (2.5%) persons were diagnosed with stroke (21,402 with ischemic stroke alone, 4947 with hemorrhagic stroke alone, and 392 with both). [Table 1] summarizes the characteristics of the study population.
|Table 1 Descriptive statistics of the cohort subjects aged 45 years and older on the Montreal Island (Canada) for the years 2000 to 2014|
Click here to view
Noise levels (from LUR models) were normally distributed [Table 2]. The average total noise levels (LAeq24h, Lden, and Lnight) were >55 dBA. Males and females were both exposed to similar noise levels [Table S1]. Proportions of older participants in high noise categories were slightly higher than younger people and those more materially deprived were more often exposed to higher noise levels.
|Table 2 Distribution of total noise exposure levels (time-varying exposures) of the Montreal population (Canada), for the years 2000 to 2014|
Click here to view
Mean NO2 levels were 13.3 parts per billion (ppb) (standard deviation = 3.3). A weak correlation was found between NO2 and LUR noise levels (Pearson r varying from 0.17 to 0.19) [Table S2].
We observed positive associations between total environmental noise levels (LAeq24h, Lden, and Lnight) and stroke incidence (all types combined) [Table 3]. Based on the use of restricted cubic splines, we found no departure from linearity for the associations with total noise levels from LUR models [Figure S1]. The effects of the interaction between noise estimates and NO2 were not statistically significant (P > 0.05). For the crude models, we found HRs of 1.16 (95% CI: 1.13–1.19), 1.14 (95% CI: 1.11–1.17), and 1.20 (95% CI: 1.16–1.23) per 10 dBA increase, respectively, in LAeq24h, Lden, and Lnight [Table 3]. After stratifying by sex and adjusting for the year, we did not observe changes in the HRs. Additional adjustment for the material deprivation index or air pollution (NO2) slightly influenced the HRs but they remained statistically significant. After stratifying by sex and fully adjusting for the year, material deprivation index and NO2, the HRs were, respectively, 1.06 (95% CI: 1.03–1.09), 1.06 (95% CI: 1.02–1.09), and 1.08 (95% CI: 1.04–1.11) for 10 dBA increase in LAeq24h, Lden, and Lnight. In the full model, the association with NO2 was positive (HR: 1.19 per 5 ppb; 95% CI: 1.17–1.22) (data not shown). Associations between noise levels and stroke incidence were unchanged with further adjustment for smoking (indirect adjustment) [Table S3].
|Table 3 Associations* between total environmental noise levels and stroke incidence in people aged 45 years and older living in Montreal, Canada (2000–2014), per 10 dBA increase in noise levels|
Click here to view
We observed that total environmental noise was positively associated with ischemic stroke unlike hemorrhagic stroke. The HRs for ischemic stroke stratified by sex and adjusted for the year, the material deprivation index, NO2, and indirectly adjusted for smoking were 1.08 (95% CI: 1.04–1.12), 1.07 (95% CI: 1.03–1.10), and 1.08 (95% CI: 1.04–1.12) per 10 dBA increase in LAeq24h, Lden, and Lnight, respectively [Table S4]. The adjusted HR for hemorrhagic stroke was 0.97 (95% CI: 0.90–1.04) per 10 dBA in LAeq24h. Associations with the other noise indicators (Lden and Lnight) and hemorrhagic stroke were similar [Table S4].
Analyses with noise levels in categories (categories of LAeq24h, Lden, and Lnight) suggested dose–response relationships with stroke onset (all types combined) and the ischemic type [Table S5]. The HRs stratified by sex and adjusted for the year, NO2, and material deprivation ranged from 1.02 (95% CI: 0.99–1.06) for the lowest noise levels (55–59.9 vs. <55 dBA) to 1.07 (95% CI: 1.01–1.14) for the highest levels (≥65 vs. <55 dBA) in LAeq24h for all strokes. The results were similar for ischemic stroke [Table S5]. However, no clear dose–response relationship was observed with hemorrhagic stroke [Table S5].
| Discussion|| |
We found positive associations between total environmental noise levels and stroke incidence (all types combined) and more specifically with ischemic stroke in a Canadian city. Our estimates of association between total environmental noise levels and stroke incidence (all types combined) are comparable to results reported by Sørensen et al. who found a positive association between road noise and stroke (incidence rate ratio, IRR: 1.14 per 10 dBA in road Lden; 95% CI: 1.03–1.25). The meta-analysis by Cai et al. also reported a similar effect size but the association was not statistically significant (HR: 1.05 per 10 dBA in Lden; 95% CI: 0.97–1.14).
However, our estimates differ from those of other studies which reported weaker or no association between road noise and stroke incidence. For example, in a case–control study from Germany, Seidler et al. found an odds ratio (OR) of 1.02 (95% CI: 1.00–1.03) per 10 dBA in LAeq24h (road noise) for all strokes combined. Pyko et al., Cai et al., and Carey et al. found no association between road noise and all strokes combined. The adjusted HRs were respectively: 1.00 (95% CI: 0.92–1.09) per 10 dBA in Lden (Sweden cohort); 0.98 (95% CI: 0.94–1.02) per 3.9 dBA in Lden (joint analysis of three European cohorts: HUNT, EPIC-Oxford, and UK Biobank); and 0.93 (95% CI: 0.83–1.05) comparing people exposed to >65 versus <55 dBA in Lnight (London cohort). This may be due to the fact that some of these studies directly adjusted for numerous individual covariates not considered in our study such as smoking status, alcohol consumption, occupation, education, and physical activity. Differences in methods used to assess noise exposure could also explain differences between studies. Indeed, most studies that have examined the relationship between environmental noise and CVD such as stroke did not estimate associations with total (all sources) environmental noise like we did, but rather with specific noise sources (mostly from roads), using propagation models.,,,,, Few studies have examined the combined effect of the three main transportation noise sources on stroke (all types) and they also found contrasting results., Pyko et al. reported a nonstatistically significant positive association (i.e., HR per 10 dBA in Lden: 1.42; 95% CI: 0.87–2.32). Heritier et al. found no association with night-time exposure to all three noise sources combined and stroke (all types) mortality (associations presented graphically in the study).
We also found a positive association between total environmental noise and ischemic stroke but not with hemorrhagic stroke, which is in line with the few other studies that have performed analyses by stroke type. For example, Sørensen et al. found in a Danish cohort (women and men), an adjusted (NO2 included) IRR of 1.15 (95% CI: 1.04–1.26) per 10 dBA in Lden (road traffic noise) at diagnosis of ischemic stroke; they did not find an association with intracerebral hemorrhagic stroke (IRR not adjusted for air pollution: 0.99 per 10 dBA in Lden; 95% CI: 0.81–1.20). In a Switzerland cohort study on cardiovascular mortality, Heritier et al. found that ischemic stroke deaths were positively associated with road and aircraft noise. Their adjusted HRs per 10 dBA in Lden were 1.05 (95% CI: 1.00–1.10) and 1.07 (95% CI: 1.02–1.13) for road and aircraft noise, respectively. Hemorrhagic stroke was not associated with these two noise sources (HR for 10 dBA road Lden: 1.00, 95% CI: 0.97–1.04; HR for 10 dBA aircraft in Lden: 0.99, 95% CI: 0.95–1.03). In a case–control study from Germany, Seidler et al. found an OR of 1.02 (95 CI%: 1.01–1.04) and 0.98 (95% CI: 0.94–1.02) per 10 dBA in LAeq24h (road), respectively, for ischemic stroke and hemorrhagic stroke. Ischemic and hemorrhagic stroke are two types of stroke that differ pathophysiologically. The first is caused by an arterial blockage (clot, atheroma plaque), whereas the second is caused by an arterial rupture (often from a pre-existing aneurysm and high blood pressure). Though it is possible that the pathophysiologic mechanism underlying associations between noise and stroke involves arterial blockages and not arterial rupture, it is more likely that too few hemorrhagic strokes were found to detect an association.
We also observed in our study that the traffic-related air pollutant NO2 had a minimal confounding effect on the relationship between total environmental noise and stroke incidence [Table 3]. This is in line with the results of previous studies suggesting an independent role of noise and air pollution on the onset of CVD.,
The large study population and the long follow-up period is one of the strengths of this study; subjects were followed on average for 8.5 years. Our study is the first to use a LUR model to estimate total environmental noise exposure in relation to stroke. The total noise better reflects people’s exposure in real life as it has strong correlation with noise measurements. Our study is also in line with recent studies that suggest using the combination of different noise sources to which individuals are exposed when assessing associations with CVD.,
It is however important to highlight some of the limitations of our research. First, we did not have the precise address of the participants nor information on housing characteristics (room orientation, number of openings, soundproofing, etc.). Our associations could therefore suffer from misclassification errors. However, the bias induced is nondifferential (i.e., affects all participants equally) and would thus likely underestimate the real risk. Our noise models did not consider individuals’ exposure at locations other than the residence, such as the workplace, school, and during transit; this likely contributes to misclassification of exposure.
Our LUR estimates could also induce another misclassification error which can decrease the associations toward the null. Indeed, the estimation of noise levels with the LUR was based on data from 2010 to 2014 and we assumed that the noise levels were constant for the other years (2000-2009 and 2011-2013). The noise levels to which people are exposed to could have changed over time but we did not have data to assess this variation. However, little changes took place in the residential infrastructure and noise sources during this follow-up period.
We used LUR models to estimate A-weighted noise levels (LAeq24h, Lden, Lnight). Although most studies published to date have used similar indicators, such indicators ignore differences in spectral and temporal characteristics of different noise sources. Indeed A-weighted LAeq24h and Lden indicators may be more suitable for road traffic noise because this source is relatively continuous and not characterized by impacts. However, noise from sources like trains and airplanes may be more intermittent and this may not be captured by conventional indicators. Furthermore, A-weighted indicators may reduce the influence of some noise sources. Thus, our LUR may capture only partially the influence of noise on stroke. Furthermore, part of the associations observed may also be due to characteristics of the urban environment that correlates with noise and urbanicity and it is thus difficult to separate their effects.
Although selection bias is unlikely because we used linked administrative health data (with few missing data, especially for hospital data that we used here) and in Canada there is universal health coverage, we most likely have residual confounding in our analysis. Indeed, we did not have individual information on SES and lifestyle variables; we only performed an indirect adjustment for smoking, of our associations between total environmental noise (LAeq24h, Lden, Lnight) and stroke.
| Conclusion|| |
The results of our study showed that total environmental noise is positively associated with stroke in particular with ischemic type. Notwithstanding some limitations, our results are consistent with associations between noise and other CVD like MIs that likely share similar pathophysiologic mechanisms. Our results also support the independence of the effects of traffic-related air pollution and total environmental noise on the incidence of CVD such as stroke.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Basner M, McGuire S. WHO Environmental Noise Guidelines for the European region: a systematic review on environmental noise and effects on sleep. Int J Environ Res Public Health 2018;15:519.
Münzel T, Schmidt FP, Steven S, Herzog J, Daiber A, Sørensen M. Environmental noise and the cardiovascular system. J Am Coll Cardiol 2018;71:688–97.
Münzel T, Steven S, Hahad O, Daiber A. Noise and cardiovascular risk: nighttime aircraft noise acutely triggers cardiovascular death. Eur Heart J 2021;42:844–6.
Recio A, Linares C, Banegas JR, Díaz J. Road traffic noise effects on cardiovascular, respiratory, and metabolic health: an integrative model of biological mechanisms. Environ Res 2016;146:359–70.
Hahad O, Beutel M, Gori T et al.
Annoyance to different noise sources is associated with atrial fibrillation in the Gutenberg Health Study. Int J Cardiol 2018;264:79–84.
Pyko A, Andersson N, Eriksson C et al.
Long-term transportation noise exposure and incidence of ischaemic heart disease and stroke: a cohort study. Occup Environ Med 2019;76:201–7.
van Kempen E, Casas M, Pershagen G, Foraster M. WHO Environmental Noise Guidelines for the European region: a systematic review on environmental noise and cardiovascular and metabolic effects: a summary. Int J Environ Res Public Health 2018;22:1–59.
Sørensen M, Hvidberg M, Andersen ZJ et al.
Road traffic noise and stroke: a prospective cohort study. Eur Heart J 2011;32:737–44.
Sørensen M, Lühdorf P, Ketzel M et al.
Combined effects of road traffic noise and ambient air pollution in relation to risk for stroke? Environ Res 2014;133:49–55.
Cai Y, Hodgson S, Blangiardo M et al.
Road traffic noise, air pollution and incident cardiovascular disease: a joint analysis of the HUNT, EPIC-Oxford and UK Biobank cohorts. Environ Int 2018;114:191–201.
Carey IM, Anderson HR, Atkinson RW et al.
Traffic pollution and the incidence of cardiorespiratory outcomes in an adult cohort in London. Occup Env Med 2016;73:849–56.
Weihofen VM, Hegewald J, Euler U, Schlattmann P, Zeeb H, Seidler A. Aircraft noise and the risk of stroke. Dtsch Arzteblatt Int 2019;116:237–44.
Cai Y, Ramakrishnan R, Rahimi K. Long-term exposure to traffic noise and mortality: a systematic review and meta-analysis of epidemiological evidence between 2000 and 2020. Environ Pollut 2021;269:116222.
Halonen JI. Transportation noise and cardiovascular health: role of multiple noise sources. Occup Environ Med 2019;76:199–200.
Aguilera I, Foraster M, Basagaña X et al.
Application of land use regression modelling to assess the spatial distribution of road traffic noise in three European cities. J Expo Sci Environ Epidemiol 2015;25:97–105.
Blais C, Jean S, Sirois C et al.
Quebec Integrated Chronic Disease Surveillance System (QICDSS), an innovative approach. Chronic Dis Inj Can 2014;34:226–35.
Lee DS, Chiu M, Manuel DG et al.
Trends in risk factors for cardiovascular disease in Canada: temporal, socio-demographic and geographic factors. CMAJ 2009;181:E55–66.
Ragettli MS, Goudreau S, Plante C et al.
Statistical modeling of the spatial variability of environmental noise levels in Montreal, Canada, using noise measurements and land use characteristics. J Expo Sci Environ Epidemiol 2016;26:597–605.
Pampalon R, Hamel D, Gamache P, Raymond G. A deprivation index for health planning in Canada. Chronic Dis Can 2009;29:178–91.
Tetreault L-F, Perron S, Smargiassi A. Cardiovascular health, traffic-related air pollution and noise: are associations mutually confounded? A systematic review. Int J Public Health 2013;58:649–66.
Crouse DL, Ross NA, Goldberg MS. Double burden of deprivation and high concentrations of ambient air pollution at the neighbourhood scale in Montreal, Canada. Soc Sci Med 2009;69:971–81.
Thiebaut ACM, Bénichou J. Choice of time-scale in Cox’s model analysis of epidemiologic cohort data: a simulation study. Stat Med 2004;23:3803–20.
Harrell FE. Regression Modeling Strategies: With Applications to Linear Models, Logistic Regression, and Survival Analysis [Internet]. New York, NY: Springer. Springer Series in Statistics; 2001. Available at http://link.springer.com/10.1007/978-1-4757-3462-1
Accessed June 3, 2016.
Steenland K, Greenland S. Monte Carlo sensitivity analysis and Bayesian analysis of smoking as an unmeasured confounder in a study of silica and lung cancer. Am J Epidemiol 2004;160:384–92.
Villeneuve PJ, Goldberg MS, Burnett RT, van Donkelaar A, Chen H, Martin RV. Associations between cigarette smoking, obesity, sociodemographic characteristics and remote-sensing-derived estimates of ambient PM2.5: results from a Canadian population-based survey. Occup Environ Med 2011;68:920–7.
Seidler AL, Hegewald J, Schubert M et al.
The effect of aircraft, road, and railway traffic noise on stroke - results of a case-control study based on secondary data. Noise Health 2018;20:152–61.
] [Full text]
Heritier H, Vienneau D, Foraster M et al.
Diurnal variability of transportation noise exposure and cardiovascular mortality: a nationwide cohort study from Switzerland. Int J Hyg Environ Health 2018;221:556–63.
Heritier H, Vienneau D, Foraster M et al.
Transportation noise exposure and cardiovascular mortality: a nationwide cohort study from Switzerland. Eur J Epidemiol 2017;32:307–15.
Donnan GA, Fisher M, Macleod M, Davis SM. Stroke. Lancet 2008;371:1612–23.
Heritier H, Vienneau D, Foraster M et al.
A systematic analysis of mutual effects of transportation noise and air pollution exposure on myocardial infarction mortality: a nationwide cohort study in Switzerland. Eur Heart J 2019;40:598–603.
Smargiassi A, Plante C, Morency P et al.
Environmental and health impacts of transportation and land use scenarios in 2061. Environ Res 2020;187:109622.
School of Public Health, Centre of Public Health Research, University of Montreal and CIUSSS du Centre-Sud-de-l’Île-de-Montréal, Montreal, QC H2L 2W5
Source of Support: None, Conflict of Interest: None
[Table 1], [Table 2], [Table 3]