Assessing the environmental burden of disease due to road traffic noise in Hesse, Germany Matthias Lochmann 1 HLNUG (Hessian Agency for Nature Conservation, Environment and Geology) Rheingaustraße 186 65203 Wiesbaden Germany Janice Hegewald 2 Melanie Schubert 3 Andreas Seidler 4 Institute and Policlinic of Occupational and Social Medicine, Faculty of Medicine, Technische Universität Dresden, 01307 Dresden, Germany

ABSTRACT As guidance for informed decision-making, we estimated the environmental burden of disease at- tributable to road-traffic noise in Hesse. Using detailed road-traffic-noise exposure data provided by HLNUG, we calculated the DALYs due to road-traffic noise > 40 dB(A) L 24h (unweighted average 24 h noise level) and other noise metrics for endpoints with known dose-response functions and evidence in the literature (NORAH-study on disease risks and WHO reviews). For Hesse, we found a total of 26,501 DALYs attributable to road-traffic noise or 435 DALY per 100,000 persons for the reference year, 2015. The end points "Annoyance" and "Sleep disturbance" contribute more than 70 % of the burden. Further, we estimated that a hypothetic uniform road-traffic-noise reduction of 3 dB would prevent 23% of this burden of disease. We are planning to suggest an alternative approach to extract an annoyance function from raw data used in the WHO-review. Our findings imply that the burden attributable to street-traffic-noise is of the same order of magnitude as, for example, the more fully researched environmental risk factor particulate matter. HLNUG is evaluating expanding the BoD- approach including uncertainty assessment to other environmental risk factors and its use for inform- ing decision makers.

1 INTRODUCTION In our contribution to INTER-NOISE 2022, we summarize the outcomes of a project which is re- ported in [1] with more epidemiological details. Here, we focus on the technical/ acoustical aspects. Road-traffic noise is the largest source of noise pollution in Europe by far [2]. According to the World Health Organization (WHO), about 40% of the European population is exposed to road traffic- weighted day–evening night-noise levels (LDEN) of ≥55 dB(A) [3]. Exposure to road-traffic noise

1 Matthias.Lochmann@hlnug.hessen.de

2 janice.hegewald@tu-dresden.de

3 melanie.schubert@tu-dresden.de

4 andreas.seidler@tu-dresden.de

worm 2022

has short- and long-term adverse effects on physical and mental health and well-being. Cardiovascu- lar and metabolic effects [4], sleep disorders [5] as well as mental disorders [6], and severe annoyance [7] are associated with prolonged exposure to road-traffic noise. One method of quantifying health loss is to calculate disability-adjusted life years (DALYs). DALYs are a measure that sums the pro- jected years of life lost due to mortality and years of healthy life lost due to non-fatal morbidity. The WHO and the World Bank introduced this concept in their Global Burden of Disease (GBD) study to quantify and compare health loss in different world regions [8, 9]. DALYs aid decision-making and help to derive recommendations for intervention measures.

2 DATA SOURCES

worm 2022

Figure 1: Road noise mapping according to END (left) and PLUS (right) at the example of a hessian community.

2.1 Exposure data The HLNUG is in charge to perform the noise mapping according to the European Noise Directive (END). The required noise measures are the time-weighted day-evening-night level 𝐿 𝐷𝐸𝑁 and the unweighted night level 𝐿 𝑛𝑖𝑔ℎ𝑡 . However, the HLNUG also performed a more detailed noise mapping [10]

1. based on a more complete dataset of road traffic data provided by Hessen Mobil, covering not

all, but more roads than required by END 2. calculating down to 𝐿 𝐷𝐸𝑁 > 40 dB(A) and 𝐿 𝑛𝑖𝑔ℎ𝑡 > 40 dB(A)

auterbaéh LDEN [dB(A)] >40. G-«. _es _ Mo mss. a>: 002505 1 6 ReRB ” a 45 ‘ddometer

These two differences are obvious in the comparison of both mapping versions in Figure 1. The third main difference between the END- and PLUS-exposure data as used here consists in the way of his- togramming: the assignment of exposure levels to inhabitants of buildings. The END requires dis- tributing the inhabitants of a house to a set of façade points, and assigning the respective exposure levels as indicated in Figure 2.

Figure 2: Illustration of the distribution of façade points on a building; taken from [11, p. 6]. At every façade point, the exposure levels have to be calculated. For our mapping “END” in 2016, all of them were assigned to inhabitants; for the upcoming noise mapping round 2022, only the loudest half of the façade points is used (this method is not considered here). For our mapping “PLUS” only the façade point with the highest level was considered for all inhabitants.

worm 2022

However, virtually all dose-response studies considered here, imply the assignment of the loudest façade level to all inhabitants of the building instead. This leads to an important upshift of the expo- sure distribution compared to the END-required assignment. We decided to use the loudest façade assignment for our PLUS-data and rather to be compatible to the dose response-literature, than to the official END-requirements. In [10] however, also the assignment to the respective façade level was used. For burden estimations, we used histogrammed exposure data, as shown in Figure 3. In order to be able to use the 𝐿 𝐷𝐸𝑁 -exposure data with the dose-response functions derived for unweighted contin- uous sound pressure levels over 24 h ( 𝐿 24ℎ ), we transformed the exposure data using the conversion, 𝐿 𝐷𝐸𝑁 − 3.3 𝑑𝐵 = 𝐿 24ℎ [14]. 2.2 Dose-Response-functions We selected the health outcomes for our DALY calculations based on systematic reviews conducted for the “WHO Environmental Noise Guidelines for the European Region” project [4, 5, 7, 12–18] and on more recent evidence for depression [6, 19]. We determined sufficient evidence for a relationship between road-traffic noise exposure and the following outcomes: annoyance, sleep disorders, ischemic heart disease, and stroke. We did not con- sider hypertension because van Kempen, et al. [4] concluded there is only “very low” evidence for a relationship between incident hypertension and road-traffic noise. On the other hand, we decided to tentatively assign a function for depression. We did so, although the effect from the published studied only approaches significance, without reaching it so far. However, the total contribution of depression being rather small, we considered its use as informative until further evidence might be collected. All

selected dose-response functions are plotted in Figure 4. The units for effect size are either “percent- age of high annoyance” (%HA, for endpoint annoyance), “percentage of high sleep disturbance” (%HSD, for endpoint sleep), and “percentage of excess risk” (%ExcessRisk all other considered end- points). The annoyance functions taken from [7] have negative slopes at low levels, which is obvi- ously not reflecting reality. Therefore, we are going to suggest an alternative way of evaluating the raw data of [7]. The value

%𝐸𝑥𝑐𝑒𝑠𝑠𝑅𝑖𝑠𝑘= ሺ𝑅𝑅−1ሻ∗100 (1) is used for plotting the function, whereas for further calculations, we used the relative risk (“RR”)- notation, which is also found in most studies.

worm 2022

Figure 3: Exposure data histograms for the different road noise mappings: END and PLUS and addi- tionally a reduction scenario reduced by 3 dB are considered. The distributions are shown as functions of the noise measures 𝐿 24ℎ (used in the Norah-study [20]), 𝐿 𝐷𝐸𝑁 (used in the WHO-cited studies [5, 16]) and 𝐿 𝑛𝑖𝑔ℎ𝑡 (used by both) in order to be compatible with the dose-response-studies used here.

3 METHODS FOR BURDEN ESTIMATION 3.1 Clinical endpoints For each of the cardiovascular diseases and depressive disorders (all clinical endpoints), according to [21] we calculated a value for the population attributable fraction

𝑃𝐴𝐹= σ 𝑝 𝑙 ሺ 𝑅𝑅 𝑙 −1 ሻ 𝑙 1 + σ 𝑝 𝑙 ሺ𝑅𝑅 𝑙 −1ሻ 𝑙

(2)

for all noise levels with index l and the relative risk 𝑅𝑅 𝑙 at the respective level above 40 dB(A) 𝐿 24ℎ using 40 dB(A) 𝐿 24ℎ as the counterfactual value (Norah-functions) or the respective level above 53 dB(A) 𝐿 𝐷𝐸𝑁 using 53 dB(A) 𝐿 𝐷𝐸𝑁 as the counterfactual value (WHO-functions). Here, p l is the pro- portion of the population in the l -th exposure band. For the next step, we used the all-risk, endpoint-specific (with index i ) burden 𝐷𝐴𝐿𝑌 𝑖 for Germany in 2015, taken from the WHO report [22] and adapted to the specific needs of this calculation. For details and handling of the age specifications of different populations see [1]. The total attributable burden stemming from clinical endpoint then amounts to

L24h LDEN Lnight 3 a 3S s 8 S2 data source s 8 — END = — Plus 3 = ] PLUS -3 48 1 g 8 2 ae w oJ—! J 40 50 60 70 80 40 50 60 70 80 40 50 60 70 80 Sound pressure level on (loudest/every) facade [dB(A)]

𝑛

(3)

𝐷𝐴𝐿𝑌 𝑟𝑜𝑎𝑑 = ෍𝑃𝐴𝐹 𝑖 ⋅𝐷𝐴𝐿𝑌 𝑖

𝑖=1

with i as index for the n different endpoints.

worm 2022

Figure 4: Our selected dose-response-functions: Percentage for high annoyance, high sleep disturb- ance, or excess disease risk as functions of the respective noise metric as given in literature. Some study publications provided the parametric uncertainties for the slope. For these functions, the shaded areas indicate the derived confidence band.

3.2 Highly Annoyed and Highly Sleep-Disturbed The DALYs for annoyance and sleep disturbance due to road-traffic noise were calculated using the number of road noise-annoyed residents

𝑛

(4)

𝑡𝑜𝑡𝑎𝑙 𝐻𝐴 𝑜𝑟 𝐻𝑆𝐷= ෍𝑝 𝑖 ⋅𝑟 𝑖

𝑖=1

where r i is the number of Hessian residents exposed to a certain level of noise ( i ), and p i is the %HA or %HSD expected for the i -level of noise. For annoyance we used the noise metrics 𝐿 𝐷𝐸𝑁 starting with r i at 40 dB(A) and using 𝐿 𝐷𝐸𝑁 < 40 dB(A) as counterfactual. We calculated the number of se- verely sleep-disturbed persons analogously with nightly noise levels ( 𝐿 𝑛𝑖𝑔ℎ𝑡 22–6 h) of 40.0 dB(A) and above using 𝐿 𝑛𝑖𝑔ℎ𝑡 < 40 dB(A) as the counterfactual. Persons under the thresholds of 𝐿 𝐷𝐸𝑁 < 40 or 𝐿 𝑛𝑖𝑔ℎ𝑡 < 40 dB(A) did not contribute to the proportion of people who were highly annoyed. As these two endpoints are non-fatal, the DALY contribution consists of the product of this preva- lence and the disabilty weights for annoyance and sleep disturbance, respectively. We used the DW of 0.02 for being highly annoyed and the DW of 0.07 for sleep disturbances [3, 23]. No confidence intervals were provided with the %HA and %HSD formulas, so we were unable to calculate confi- dence intervals (parameter uncertainty) for these estimates.

eg L24h LOEN Light i $ 2 60 & ez ‘endpoint 5 Noah vO 40 Nera opresson 2 WHO annoyance full YWH0 sroyanes Sled = vo sctmic < E vino seer 3” E ino store 8 * 6 3 40 50 60 70 8040 50 60 70 8040 50 60 70 80 of ‘Sound pressure level on loudest facade [dB(A)]

4 RESULTS The burden contributions attributable to road noise under different approaches are illustrated in Figure 5. Contributions are given for each of the four categories annoyance, depression, heart disease and

sleep disturbance. For annoyance and heart disease, we evaluated two or three possible approaches/ pathways respectively.

Figure 5: Attributable burden contributions for the four endpoints, for three different exposure sce- narios and different dose-response functions. Confidence intervals are derived only from the para- metric uncertainty of the corresponding dose-risk-function if indicated, set to 0 otherwise.

worm 2022

As last step, we calculated the total number of DALY for the attributable burden in Hessen. For the main scenario, we chose the exposure data of the PLUS-mapping, for annoyance the “WHO annoy- ance full”-contribution, for heart disease the “Norah Cardiovascular disease”-contribution. For com- parison, we varied the exposure data to either a hypothetical, global reduction by 3 dB, or the END- data as described above. Table 1 shows the total estimates of these three scenarios. Table 1: Estimates for the burden attributable to road traffic noise in Hesse in the reference year, in three different sensitivity analysis scenarios.

exposure scenario total burden

specific burden [DALY/100.000 pers]

[DALY]

main (PLUS) 26,501 434,9

reduction (PLUS-3dB) 19,884 326,3

END 6,595 108,2

Our estimates imply, that the effect of an overall 3-dB(A) reduction in road-traffic noise would pre- vent nearly a fourth of the road-noise attributable burden. Using only END-mapping data would un- derestimate the burden by a factor of four. The burden attributable to street-traffic-noise is of the same order of magnitude as, for example, the more fully researched environmental risk factor partic- ulate matter [24].

5 CONCLUSIONS Although a set of questions and uncertainties remain, we are confident, that estimating the burden of traffic noise, together with further uncertainty assessment [25] and its communication, might be an

8 8 8 3 8 8 0. attr. burden [DALY/year/100,000 pers] END BB Noran cvo Noah dopresin Wo annoyance tt ; WHO annoyance slates Ho chor Ho seep BB wi stroke 5 coe oo 30? oo PLUS PLUS -3 dB endpoint ore veo 30 00% ooo No ser?

adequate tool for informing decision makers. HLNUG is evaluating expanding the BoD-approach including uncertainty assessment to other environmental risk factors.

References [1] J. Hegewald, M. Schubert, M. Lochmann, and A. Seidler, “The Burden of Disease Due to Road

Traffic Noise in Hesse, Germany,” International Journal of Environmental Research and Pub- lic Health , vol. 18, no. 17, p. 9337, 2021, doi: 10.3390/ijerph18179337. [2] Managing exposure to noise in Europe . Luxembourg: Publications Office of the European Un-

ion, 2017. [Online]. Available: https://op.europa.eu/en/publication-detail/-/publication/ 01060a58-92ad-11e7-b92d-01aa75ed71a1 [3] F. Theakston, “Burden of disease from environmental noise; Quantification of healthy life

years lost in Europe,” WHO, 2011. [Online]. Available: http://www.euro.who.int/__data/as- sets/pdf_file/0008/136466/e94888.pdf [4] E. van Kempen, M. Casas, G. Pershagen, and M. Foraster, “WHO Environmental Noise Guide-

lines for the European Region: A Systematic Review on Environmental Noise and Cardiovas- cular and Metabolic Effects: A Summary,” International Journal of Environmental Research and Public Health , vol. 15, no. 2, p. 379, 2018, doi: 10.3390/ijerph15020379. [5] M. Basner and S. McGuire, “WHO environmental noise guidelines for the european region: A

systematic review on environmental noise and effects on sleep,” International Journal of Envi- ronmental Research and Public Health , vol. 15, no. 3, p. 519, 2018, doi: 10.3390/ijerph15030519. [6] J. Hegewald et al., “Traffic Noise and Mental Health: A Systematic Review and Meta-Analy-

sis,” International Journal of Environmental Research and Public Health , vol. 17, no. 17, 2020, doi: 10.3390/ijerph17176175. [7] R. Guski, D. Schreckenberg, and R. Schuemer, “WHO Environmental Noise Guidelines for the

European Region: A Systematic Review on Environmental Noise and Annoyance,” Interna- tional Journal of Environmental Research and Public Health , vol. 14, no. 12, p. 1539, 2017. [8] C. J. L. Murray, A. D. Lopez, and World Health Organization, The global burden of disease: A

comprehensive assessment of mortality and disability from diseases, injuries, and risk factors in 1990 and projected to 2020 ; summary . Cambridge: World Health Organization; Harvard School of Public Health, 1996. [9] B. Cao, “WHO methods and data sources for global burden of disease estimates 2000-2011,”

WHO, Geneva Global Health Estimates Technical Paper WHO/ DDI/DNA/GHE/2020.3, 2020. [Online]. Available: http://www.who.int/healthinfo/statistics/GlobalDALYmethods_2000_ 2011.pdf?ua=1 [10] N. van der Pütten and L. Herok von Garnier, “Möglichkeiten und Herausforderungen einer er-

weiterten Umgebungslärmkartierung,” Lärmbekämpfung , vol. 13, no. 4, p. 120, 2018. [11] vanderpuetten, Berechnungsmethode zur Ermittlung der Belastenzahlen durch Umgebungs-

lärm: BEB , 2021. [12] I. van Kamp et al., “Study on methodology to perform an environmental noise and health as-

sessment-a guidance document for local authorities in Europe,” 2018, doi: 10.21945/RIVM- 2018-0121. [13] C. Clark and K. Paunovic, “WHO Environmental Noise Guidelines for the European Region:

A Systematic Review on Environmental Noise and Cognition,” International Journal of Envi- ronmental Research and Public Health , vol. 15, no. 2, p. 285, 2018, doi: 10.3390/ijerph15020285.

worm 2022

[14] C. Clark and K. Paunovic, “WHO Environmental Noise Guidelines for the European Region:

A Systematic Review on Environmental Noise and Quality of Life, Wellbeing and Mental Health,” International Journal of Environmental Research and Public Health , vol. 15, no. 11, p. 2400, 2018, doi: 10.3390/ijerph15112400. [15] D. Jarosińska et al., “Development of the WHO Environmental Noise Guidelines for the Euro-

pean Region: An Introduction,” International Journal of Environmental Research and Public Health , vol. 15, no. 4, p. 813, 2018, doi: 10.3390/ijerph15040813. [16] A. L. Brown and I. van Kamp, WHO environmental noise guidelines for the European region:

A systematic review of transport noise interventions and their impacts on health : Multidiscipli- nary Digital Publishing Institute, International Journal of Environmental Research and Public Health , vol. 14. [Online]. Available: http://www.mdpi.com/1660-4601/14/8/873 [17] M. J. Nieuwenhuijsen, G. Ristovska, and P. Dadvand, “WHO Environmental Noise Guidelines

for the European Region: A Systematic Review on Environmental Noise and Adverse Birth Outcomes,” International Journal of Environmental Research and Public Health , vol. 14, no. 10, p. 1252, 2017, doi: 10.3390/ijerph14101252. [18] WHO, Environmental noise guidelines for the European Region . Copenhagen, Denmark:

World Health Organization, Regional Office for Europe, 2018. [19] A. M. Dzhambov and P. Lercher, “Road Traffic Noise Exposure and Depression/Anxiety: An

Updated Systematic Review and Meta-Analysis,” International Journal of Environmental Re- search and Public Health , vol. 16, no. 21, p. 4134, 2019, doi: 10.3390/ijerph16214134. [20] A. Seidler et al., “Aircraft, road and railway traffic noise as risk factors for heart failure and

hypertensive heart disease—A case-control study based on secondary data,” International Journal of Hygiene and Environmental Health , vol. 219, no. 8, pp. 749–758, 2016, doi: 10.1016/j.ijheh.2016.09.012. [21] J. A. Hanley, “A heuristic approach to the formulas for population attributable fraction,” Jour-

nal of epidemiology and community health , vol. 55, no. 7, pp. 508–514, 2001, doi: 10.1136/jech.55.7.508. [22] WHO, “World Health Organization. Global Health estimates 2019: Disease Burden by Cause,

Age, Sex, by Country and by Region, 2000–2019,” Geneva, Switzerland, 2020. [23] M. Tobollik, M. Hintzsche, J. Wothge, T. Myck, and D. Plass, “Burden of Disease Due to

Traffic Noise in Germany,” International Journal of Environmental Research and Public Health , vol. 16, no. 13, p. 2304, 2019, doi: 10.3390/ijerph16132304. [24] C. J. Murray, “Global burden of 87 risk factors in 204 countries and territories, 1990–2019: a

systematic analysis for the Global Burden of Disease Study 2019,” The Lancet , vol. 396, no. 10258, pp. 1223–1249, 2020, doi: 10.1016/S0140-6736(20)30752-2. [25] A. B. Knol, A. C. Petersen, J. P. van der Sluijs, and E. Lebret, “Dealing with uncertainties in

environmental burden of disease assessment,” Environmental health : a global access science source , vol. 8, p. 21, 2009, doi: 10.1186/1476-069X-8-21.

worm 2022