Systematic Review of meta-analyses for noise Yingxin Chen 1 Centre for Environmental Health and Sustainability, University of Leicester University Road, Leicester, LE1 7RH Claire Blackmore 2 Centre for Environmental Health and Sustainability, University of Leicester University Road, Leicester, LE1 7RH Eminson, Katie 3 Centre for Environmental Health and Sustainability, University of Leicester University Road, Leicester, LE1 7RH Xiangpu Gong 4 Centre for Environmental Health and Sustainability, University of Leicester University Road, Leicester, LE1 7RH Anna Hansell 5 Centre for Environmental Health and Sustainability, University of Leicester University Road, Leicester, LE1 7RH

ABSTRACT

In the context of constructing a burden of disease (BOD) toolkit for noise exposures in England, we sought to identify meta-analyses to provide exposure-response coefficients that would update those available from the WHO Noise Guidelines for the European Region published in 2018. We conducted a systematic review of systematic reviews relating to noise exposure and selected health outcomes published in 2017-20. We used the AMSTAR checklist to score all selected systematic review papers at the same time as data extraction. A new review needed to have at least a moderate score on AM- STAR to be recommended as an alternative/update to the WHO analyses. Twenty-three papers were included in the review, of which 16 studies provide quantitative effects estimates. Eight considered environmental noise associations with metabolic outcomes, seven with mental health outcomes, six with cardiovascular outcomes, and five studies considered wellbeing, sleep, and annoyance. We rec- ommend one new meta-analysis on diabetes and one on hypertension, plus a further new meta-anal- ysis on ischemic heart disease (IHD) to be considered for use in place of the evidence recommended in the 2018 WHO Environmental Noise Guidelines. Our review also suggested there are now enough studies available to conduct a meta-analysis for noise and dementia.

1 yc310@leicester.ac.uk

2 clb91@leicester.ac.uk

3 ke79@leicester.ac.uk

4 xg82@leicester.ac.uk

5 ah618@leicester.ac.uk

worm 2022

1. INTRODUCTION

In the context of developing a new UK Burden of Disease Toolkit for noise for use by the UK Health Security Agency, this review assessed whether the newer evidence should take precedence over the exposure-response relationships for major outcomes proposed in the WHO systematic reviews on environmental noise. 2. METHODS

We conducted a search of PubMed and Medline from 2017 to December 2020. Our search terms named noise sources of interest, the health outcomes of interest, papers that were termed epidemio- logical or health studies, and papers titled included “review”, “meta-analysis”. We limited number of studies using filter “review papers” to identify systematic review studies. A standardized data extrac- tion form was used to collect methodologic variables from each included review. We used the online version AMSTAR checklist to score all selected systematic review papers at the same time as data extraction. 2.1. Paper identification

We conducted a search of PubMed and Medline from 2017 to 15 June 2020, that was updated (on 07 Mar 2021) to December 2020. We also searched conference proceedings for the International Con- gress and Exposition on Noise Control Engineering (Inter-noise, 2017-19), Euronoise (2018), Inter- national Commission on Biological Effects of Noise (ICBEN, 2017), The International Commission for Acoustics (ICA, 2019), the International Congress on Sound and Vibration (ICSV, 2017-19). We identified further papers through colleagues and our own reference libraries. Our search terms named noise sources of interest, the health outcomes of interest, papers that were termed epidemiological or health studies, and papers titled included “review”, “meta-analysis”. We limited number of studies using filter “review papers” to identify systematic review studies. Our eligibility criteria are below (See Table 1). Table 1: Eligibility criteria to screen studies for the review. Components Inclusion criteria Exclusion criteria Study type Systematic review and meta-analysis for epidemiological studies. e.g., ecological studies, cross-sectional studies, prospective and retrospective cohort studies, case-control studies and observational or experimental studies of people exposed to environmental noise.

Review studies for intervention, in-vitro, toxicological, animal, or controlled exposure studies were excluded.

Population Members of the general population as well as specific seg- ments of the population particularly at risk e.g., Children, pregnant women

Animals, occupational popula- tions

Exposure Environmental noise exposure from road, rail, aircraft, transport, industrial, wind turbine, construction.

Studies participants exposed to occupational or laboratory- based noise were excluded Outcome Assessment of serval health outcomes, including effects on sleep, annoyance, cognitive impairment in school children, quality of life, cardiovascular disease (ischemic heart dis- ease, hypertension, stroke), metabolic disease (e.g., Obe- sity, diabetes mellitus).

Health outcomes of unclear clin- ical health relevance, e.g., epige- netics, metabolomics, methyla- tion was excluded.

worm 2022

Two researchers (CB & YC) scanned titles and read abstracts (sometimes with reference to full paper if not clear) to identify papers for the list for full paper reading. Duplicates, reviews for the WHO Environmental Noise guidelines and papers with no quantitative synthesis of results were excluded, as well as those out of scope (due to exposure or outcome issues). 2.2. Data extraction and preparation

A standardized data extraction form (see table 1) was used to collect the following methodologic variables from each included review, conducted by at least two reviewers (YC, CB, AH, KE, XG). This comprised: first author (publication year), type of review, journal, data source, years covered, exposure of interest, outcome, location included, whether it had quantitative synthesis, number of papers and estimates, study results and comments.

2.3. Quality assessment

We used the online version AMSTAR checklist (https://amstar.ca/Amstar_Checklist.php; see appen- dix 5 for detail) to score all selected systematic review papers at the same time as data extraction, with each paper scored by at least two reviewers (YC, CB, AH, KE, XG). Any discrepancies in scor- ing were resolved in discussion. (The WHO reviews were also extracted and scored using AMSTAR as comparison) For a new review to be recommended as an alternative/update to the WHO analyses, it needed to have at least a moderate score on AMSTAR. 3. RESULTS

3.1. Papers Identified

worm 2022

Combining both initial (2017 to 15 June 2020) and update (to December 2020) searches and after evaluation by reading full text, we found 23 papers that were included in the review (see Figure 1).

Figure 1: Literature search flow chart showing number of papers identified.

Figure 2 presents the summary of key features of included review studies. Most studies reviewed were on road traffic noise exposure (n=15), followed by railway noise (n=9), air traffic noise (n=9) and combined environmental noise (n=7). The majority of the studies reviewed concerned metabolic health effects (n=8), mental health (n=7) and CVDs (n=6). Three studies had a high AMSTAR rating, 11 studies were rated as “moderate” using AMSTAR scoring and nine were low or critically low.

worm 2022

15

0 2 4 6 8 10 12 14 16RoadRailAirWind turbineCombined/AmbientDiabetesMental healthCVDHypertensionQoL/Annoyance/SleepCognitionHighModerateLowCritically low

11

9 9

7 8 7 6

6

5

5

3

2 3

3

Health outcomes

AMSTAR scores

Source of noise

Figure 2: Key features of review studies 3.2 Effects of noise on metabolic outcomes (diabetes)

Overall, eight studies systematically reviewed environmental noise associations with diabetes (1-8). Among these, six conducted meta-analyses for metabolic outcomes to derive exposure-re- sponse associations for environmental noise (2, 4-8). Of the six quantitative studies, there were three studies rated “moderate” on AMSTAR scoring (2, 5, 6): den Braver and colleagues found higher exposure to residential noise associated with higher risk of T2DM, but this was not statistically sig- nificant on meta-analysis (OR:1.49; 95% CI: 0.78, 2.82; I 2  = 75.8%) (6). This review considered the built environment, also looked at physical activity, food, etc. However, this review did not separate noise exposure by source and only 4 papers were included in the meta-analysis, two of which had already been included in WHO meta-analysis. Therefore, we did not consider this could be used as an alternative to the WHO review. Sakhvidi found an overall 6% increase (OR:1.06, 95% CI:1.03, 1.09) in risk of diabetes associated with 5dB increase in exposure to noise regardless of source(5), with the strongest association for air traffic (combined OR: 1.17; 95% CI: 1.06, 1.29), followed by road traffic noise (combined OR: 1.07; 95% CI: 1.02, 1.12) per 5 dB increase in exposure(5). Vienneau pooled RRs from 6 studies resulting in a respective 11% (road traffic, 95%CI: 1.08, 1.15, I 2 =0.0% for 5 studies), 20% (aircraft, 0.88, 1.63, I 2 =0.0%, for 3 studies) and 1% (railway, 0.94 - 1.04,I 2 =0.0%, for 2 studies) increased risk of diabetes incidence per 10 dB increment in Lden (2). However, this is a conference proceeding and has not been peer-reviewed, unlike the Sakhvidi paper which has similar results (similar ratios and confidence intervals) from a similar number of studies. The similarities in results from Vienneau and Sakhvidi were considered encouraging and strength- ened a case to update the WHO exposure-response coefficient using the Sakhvidi paper (Table 2).

Table 2: Comparison of WHO 2020 review, Sakhvidi 2018 review and Vienneau 2019 review on exposure-response coefficients for diabetes

Author year van Kempen, 2020

Sakhvidi,2018

Vienneau ， 2019

(WHO) per 10dB

per 5dB

per 10dB

Years covered Jan 2000 - Aug 2015

1 Jan 2014 - 1 Mar 2019

prior to 13 Sep 2017

Number of estimates/ pa-

3 for diabetes inci-

10 estimates (6 studies) 9 estimates (15 pa-

pers included

dence

pers) Road – Diabetes

1.08 (1.02, 1.14)

1.11 (1.08, 1.15) (5 studies I 2 =0.0%)

1.07 (1.02, 1.12) (3 studies I 2 =33.6%) Rail – Diabetes (n estimates, I 2 )

(n estimates, I 2 )

(1 study)

0.97 (0.89, 1.05)

1.01 (0.94, 1.04) (2 studies I 2 =0.0%)

0.98 (0.95, 1.02) (2 studies, I 2 =0%) Aircraft – Diabetes

(1 study)

0.99 (0.47, 2.09)

1.20 (0.88, 1.63) (3 studies I 2 =0.0%)

1.17 (1.06, 1.29) (4 studies, I 2 =0%) 3.3 Effects of noise on cardiovascular outcomes and hypertension

(n estimates, I 2 )

(1 study)

Overall, six studies considered environmental noise associations with cardiovascular outcomes (1-3, 9-11), of which four conducted meta-analyses (2, 9-11). Vienneau’s 2019 meta-analysis estimated pooled RRs for ischemic heart disease (IHD) incidence, giving a respective 2% (road traffic, 95%CI: 1.00, 1.04, I 2 =23.2%), 3% (aircraft, 95%CI: 0.98, 1.09, I 2 =89.4%) and 1% (railway, 95%CI: 0.99, 1.03, I 2 =23.3%) increased risk of IHD incidence per 10 dB increment in Lden(2). The review received a moderate AMSTAR score. As this review includes more studies than WHO review for road and aircraft noise exposures (though one fewer for railway), it could be considered for use instead of WHO for IHD incidence as an outcome (IHD). Table 3: Comparison of WHO 2020 review and Vienneau 2019 review on exposure-response coeffi- cients for IHD incidence

Author, year van Kempen, 2020 (WHO) Vienneau ， 2019 Year cover Jan 2000 - Aug 2015 1 Jan 2014 - 1 Mar 2019 Num. of paper included 22 studies 21 estimates (13 papers) Road – IHD incidence

1.08 (1.01, 1.15) (7 studies I 2 not clear)

1.02 (1.00, 1.04) (13 studies, I 2 =23.2%) Rail – IHD incidence

(n estimates, I 2 )

N/A 1.01 (0.99, 1.03) (3 studies I 2 =23.2%) Aircraft – IHD incidence (n estimates, I 2 )

(n estimates, I 2 )

1.09 (1.04, 1.15) (2 studies I 2 =48.4%)

1.03 (0.98, 1.09) (5 studies, I 2 =89.4%) Three studies systematically reviewed environmental noise effect on hypertension(12-14). Dzham- bov and colleagues conducted a meta-analysis from 14 estimates from 9 studies (7 cohort and 2 case- control). They found "Low" quality evidence, with RR 1.018 (95% CI: 0.98, 1.05, I 2 =46%) per 10dB(A) increase in noise overall(13), and "moderate" evidence from a subgroup analysis of cohort studies (RR: 1.02; 95%CI: 0.99, 1.05, I 2 =31%) per 10dB noise increase in incident of hypertension. This review could potentially use as an update of WHO meta-analysis for hypertension.

worm 2022

3.4 Effects of noise on mental health outcomes

Overall, seven studies systematically reviewed environmental noise effects on mental health out- comes(15-21). Among them, four conducted quantitative analyses(16, 17, 19, 20). There were three systematic reviews identified that provide a quantitative estimate for association with depression and with anxiety(16, 19, 20). Dzhambov’s review found for 10dB increase in road traffic Lden, the odds of depression increased by 4% (95% CI: -3%, 11%), and odds of anxiety increase by 12% (95% CI: -4%, 30%) (16). Hegewald’s 2020 meta-analysis estimated pooled RRs for depression giving a re- spective 3% (road traffic, 95%CI: 0.99, 1.06, I 2 =60.2%), 14% (aircraft, 95%CI: 1.12, 1.15, I 2 =0.0%) and 2% (railway, 95%CI: 0.95, 1.08, I 2 =95.6%) increased risk of depression per 10 dB increment in L den (19). Lan and colleagues pooled RRs for anxiety from 12 studies resulting in a respective 8% (road traffic, 95%CI: 0.99, 1.18, I 2 =51%), 8% (aircraft, 95%CI: 0.99, 1.18, I 2 =91%) and 1% (railway, 95%CI: 0.97, 1.15, I 2 =95.6%) increased risk of anxiety per 10 dB increment in Lden(20). While AMSTAR score were moderate or above, there are a number of caveats to using any of these reviews in a burden of disease calculation as outcome assessment methods were heterogeneous. Studies with largest influence (seen on Forest plot) were studies using register-based outcomes of depression/anx- iety. 3.5 Effects of noise on wellbeing, sleep, and annoyance

Five studies systematically reviewed environmental noise effect on wellbeing, sleep and annoyance (1, 15, 21-23) but none of these performed meta-analysis. Three of the five studies were reviews on health effects of wind turbine noise [21-23]. Clark’s 2020 review did not perform a quantitative review from 24 studies investigating environmen- tal noise exposure effects on mental health, wellbeing, and quality of life, as outcome measures were heterogeneous[15]. Van Kamp’s review described the results of a scoping synthesis of the literature into the effects of environmental noise on sleep from 17 studies and annoyance from 22 studies [1]. This review reported “moderate” to “high” quality evidence for an association between environmen- tal noise and annoyance; and “moderate” evidence for road traffic noise and cortical awakenings and self-reported sleep disturbance. The new reviews identified suggest there are now enough studies available for a new update of the meta-analysis on wellbeing, annoyance and sleep measures for aircraft noise, road traffic, and rail traffic noise. 4. Discussion

4.1. Recommendation for diabetes (DM), hypertension and ischemic heart disease (IHD)

We recommended for the UK Health Security Agency burden of disease (BOD) toolkit for noise exposures in England: one new meta-analysis on diabetes (Sakhvidi, 2018) and one on hypertension (Dzhambov, 2018), plus a further new meta-analysis on IHD (Vienneau, 2019) to be considered for use in place of the evidence recommended in the 2018 WHO Environmental Noise Guidelines sys- tematic review. 4.2. Recommendations for updated exposure-response risk estimates for outcomes consid- ered

• For mental health, we identified three moderate/high quality meta-analysis on depression/anxiety, but we were concerned about combination of heterogeneous outcome measures in these studies so would not recommend for use in the BoD toolkit. We recommended for future work, a new meta-analysis of noise and dementia as a new condition to be added into the BoD toolkit.

worm 2022

• For wellbeing, annoyance, and sleep, new reviews did not provide quantitative summary of esti- mates so these cannot be used in the BOD toolkit as alternative to WHO reviews. • For cognition, neither of the reviews provided quantitative effects estimates because of large amount of heterogeneity between studies. Thus, we do not recommend these reviews to use as alternative to WHO as no quantitative synthesis was provided. 4. CONCLUSIONS

We recommend noise effect estimates from one new meta-analysis on diabetes and one on hyperten- sion, plus a further new meta-analysis on IHD to be considered for use in place of the noise effect estimates provided by the WHO systematic reviews supported the WHO 2018 Environmental Noise Guidelines for the European Region. Our review also suggested there are now enough papers availa- ble to conduct a future new meta-analysis for noise and dementia. 5. ACKNOWLEDGEMENTS

This study acknowledges funding from the National Institute for Health Research (NIHR) Health Protection Research Unit in Environmental Exposures and Health, a partnership between the UK Health Security Agency, the Health and Safety Executive and the University of Leicester. The views expressed are those of the author(s) and not necessarily those of the NIHR, the UK Health Security Agency, the Health and Safety Executive or the Department of Health and Social Care. 6. REFERENCES

1. van Kamp I, Simon S, Notley H, Baliatsas C, van Kempen E. Evidence Relating to Environmental Noise Exposure and Annoyance, Sleep Disturbance, Cardio-Vascular and Metabolic Health Outcomes in the Context of IGCB (N): A Scoping Review of New Evidence. International Journal of Environmental Research & Public Health [Electronic Resource]. 2020;17(9):26. 2. Vienneau D, Eze IC, Probst-Hensch N, Röösli M. Association between transportation noise and cardio-metabolic diseases: an update of the WHO meta-analysis. 2019. 3. Rugel E, Brauer M. Quiet, clean, green, and active: A Navigation Guide systematic review of the impacts of spatially correlated urban exposures on a range of physical health outcomes. Environmental Research. 2020:109388. 4. An R, Wang J, Ashrafi SA, Yang Y, Guan C. Chronic Noise Exposure and Adiposity: A Systematic Review and Meta-analysis. American journal of preventive medicine. 2018;55(3):403-11. 5. Zare Sakhvidi M, Zare Sakhvidi F, Mehrparvar A, Foraster M, Dadvand P. Association between noise exposure and diabetes: A systematic review and meta-analysis. 2018;166:647-57. 6. den Braver NR, Lakerveld J, Rutters F, Schoonmade LJ, Brug J, Beulens JWJ. Built environmental characteristics and diabetes: a systematic review and meta-analysis. BMC Med. 2018;16. 7. Wang H, Sun D, Wang B, Gao D, Zhou Y, Wang N, et al. Association between noise exposure and diabetes: meta-analysis. Environmental science and pollution research international. 2020;27(29):36085-90. 8. Wang Z, Zhao L, Huang Q, Hong A, Yu C, Xiao Q, et al. Traffic-related environmental factors and childhood obesity: A systematic review and meta-analysis. Obes Rev. 2020. 9. Khosravipour M, Khanlari P. The association between road traffic noise and myocardial infarction: A systematic review and meta-analysis. Sci Total Environ. 2020;731:139226.

worm 2022

10. Weihofen VM, Hegewald J, Euler U, Schlattmann P, Zeeb H, Seidler A. Aircraft Noise and the Risk of Stroke: A Systematic Review and Meta-analysis. Deutsches Ärzteblatt International. 2019;116(14):237. 11. 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. 12. Fu W, Wang C, Zou L, Liu Q, Gan Y, Yan S, et al. Association between exposure to noise and risk of hypertension: a meta-analysis of observational epidemiological studies. J Hypertens. 2017;35(12):2358-66. 13. Dzhambov AM, Dimitrova DD. Residential road traffic noise as a risk factor for hypertension in adults: Systematic review and meta-analysis of analytic studies published in the period 2011-2017. Environmental Pollution.240:306-18. 14. Dzhambov AM, Dimitrova DD. Children's blood pressure and its association with road traffic noise exposure - A systematic review with meta-analysis. Environ Res. 2017;152:244-55. 15. Clark C, Crumpler C, Notley AH. Evidence for Environmental Noise Effects on Health for the United Kingdom Policy Context: A Systematic Review of the Effects of Environmental Noise on Mental Health, Wellbeing, Quality of Life, Cancer, Dementia, Birth, Reproductive Outcomes, and Cognition. Int J Environ Res Public Health. 2020;17(2). 16. Dzhambov AM, Lercher P. Road Traffic Noise Exposure and Depression/Anxiety: An Updated Systematic Review and Meta-Analysis. Int J Environ Res Public Health. 2019;16(21). 17. Schubert M, Hegewald J, Freiberg A, Starke KR, Augustin F, Riedel-Heller SG, et al. Behavioral and emotional disorders and transportation noise among children and adolescents: a systematic review and meta-analysis. International journal of environmental research and public health. 2019;16(18):3336. 18. Zare Sakhvidi F, Zare Sakhvidi MJ, Mehrparvar AH, Dzhambov AM. Environmental Noise Exposure and Neurodevelopmental and Mental Health Problems in Children: a Systematic Review. Curr Environ Health Rep. 2018;5(3):365-74. 19. Hegewald J, Schubert M, Freiberg A, Romero Starke K, Augustin F, Riedel-Heller SG, et al. Traffic Noise and Mental Health: A Systematic Review and Meta-Analysis. Int J Environ Res Public Health. 2020;17(17). 20. Lan Y, Roberts H, Kwan MP, Helbich M. Transportation noise exposure and anxiety: A systematic review and meta-analysis. Environ Res. 2020;191:110118. 21. Lotinga MJ, Perkins RA, Berry B, Grimwood CJ, Stansfeld SA, editors. A review of the human exposure-response to amplitude-modulated wind turbine noise: health effects, influences on community annoyance, methods of control and mitigation. Proceedings of the 12th ICBEN Congress on Noise as a Public Health Problem; 2017. 22. Liebich T, Lack L, Hansen K, Zajamšek B, Lovato N, Catcheside P, et al. A systematic review and meta-analysis of wind turbine noise effects on sleep using validated objective and subjective sleep assessments. J Sleep Res. 2020:e13228. 23. Freiberg A, Schefter C, Girbig M, Murta V, Seidler A. Health effects of wind turbines on humans in residential settings: Results of a scoping review. 2019;169:446-63.

worm 2022