The association between aircraft noise levels and deprivation Xiangpu Gong 1 Centre for Environmental Health and Sustainability, University of Leicester, Leicester, UK National Institute for Health Protection Research Unit in Environmental Exposures and Health, University of Leicester, Leicester, UK Nicole Itzkowitz 2 MRC Centre for Environment and Health, Department of Epidemiology and Biostatistics, School of Public Health, Imperial College London, London, UK Glory O Atilola 3 MRC Centre for Environment and Health, Department of Epidemiology and Biostatistics, School of Public Health, Imperial College London, London, UK Kathryn Adams 4 Centre for Environmental Health and Sustainability, University of Leicester, Leicester, UK Calvin Jephcote 5 Centre for Environmental Health and Sustainability, University of Leicester, Leicester, UK Marta Blangiardo 6 MRC Centre for Environment and Health, Department of Epidemiology and Biostatistics, School of Public Health, Imperial College London, London, UK John Gulliver 7 Centre for Environmental Health and Sustainability, University of Leicester, Leicester, UK Anna Hansell 8 Centre for Environmental Health and Sustainability, University of Leicester, Leicester, UK

ABSTRACT There is limited evidence on deprivation distribution of noise exposure. Noise data (Lden, Laeq24, Lnight, Levening, and Lday) were available for London Heathrow airport for 2014-18. These were

1 xg82@leicester.ac.uk 2 n.itzkowitz@imperial.ac.uk 3 g.atilola@imperial.ac.uk 4 kaa39@leicester.ac.uk 5 cj191@leicester.ac.uk 6 m.blangiardo@imperial.ac.uk 7 jg435@leicester.ac.uk 8 ah618@leicester.ac.uk

Jai. inter noise 21-24 AUGUST SCOTTISH EVENT CAMPUS ? O ? . GLASGOW

linked with different measures of deprivation: the Carstairs deprivation index (UK Census-derived), fuel poverty rate and the avoidable death rate. We used a random effects model, accounting for year and % ethnic minority to quantify the association. Our findings consistently indicate that areas with the least deprivation are the quietest. We found that quintiles 2 – 5 of all deprivation variables exhibit a positive association, indicating that the least deprived areas are consistently the quietest. There is some evidence of a dose-response relationship with aircraft noise in terms of avoidable death rate. Carstairs index quintiles are significantly associated with all four metrics. Fuel poverty has a significant but relatively weak relationship with aircraft noise, compared to Carstairs and avoidable death rate. There is less conclusive evidence of gradients for Carstairs index and fuel poverty rate. Results will be discussed with community groups near Heathrow prior to Internoise 2022. As air transport increases post-pandemic, information on noise exposures as well as views from community groups can inform future airport policies 1. INTRODUCTION

Aircraft noise has been a major source of environmental pollution. In 2011, nearly 3.2% of the European Union's population was exposed to aircraft noise levels greater than Lden 55 dB[1], a figure that could continue to rise over the next 30 years if noise management strategies do not improve or improve slightly, as the number of flights is expected to increase by nearly 42% between 2017 and 2040 [1]. There is a growing body of studies that have examined the adverse association between aircraft noise and human psychological and physiological health [2]. Aircraft noise exposure is associated with an increased risk of developing adverse physiological problems in people [3, 4]. There is also evidence that actual aircraft noise exposure and noise annoyance may be negatively related to psychological outcomes [5, 6]. Deprivation is a significant confounding factor in studies of the health effects of aircraft noise pollution. Yet existing evidence on the relationship between aircraft noise and deprivation is limited and inconclusive. One study found a negative correlation between socioeconomic status and LAeq24h in Montreal, Canada, whereas a study from Chicago published in 2021 documents an unclear relationship between income and daytime noise exposure [7]. A recently published study found that in the vicinity of 90 airports in the US found no evidence that lowest income census block groups were more likely to be noisy [8]. Another study, which examined evidence at the individual level for people living near London Heathrow airport, concluded that people with the highest household income had a greater likelihood of living within a 50 dB contour of aircraft noise [9]. A review of the evidence examining social inequalities in noise exposure from all sources found a mixed relationship between deprivation and noise exposure [10]. Moreover, deprivation is a multidimensional concept, encompassing numerous facets of an individual's life over the course of their lives [11]. Most studies on the deprivation distribution of aircraft noise exposure have focused exclusively on material deprivation. Evidence on the relationship between aircraft noise and other dimensions of deprivation, such as health inequality, is needed to understand the pathway by which deprivation may confound aircraft pollution health studies. Therefore, we attempted to investigate the relationship between aircraft noise and deprivation in this study. 2. METHODOLOGY

We used modelled aircraft noise data for London Heathrow airport, providing daily equivalent continuous aircraft noise levels (Laeq) at postcode level for eight different time periods (04:30-06:00, 06:00-07:00; 07:00-15:00; 15:00-19:00; 19:00-22:00; 22:00-23:00; 23:00-24:00; 24:00-04:30) throughout the day for the period of 2014-2018. This includes 164,012 postcodes that encompass a rectangular area surrounding Heathrow airport, as shown in Figure 1.

We used these eight noise levels to calculate yearly mean aircraft noise levels, which resulted in four metrics: Laeq24, Lnight, Levening, and Lday. We measured deprivation using three variables: Carstairs index of multiple deprivation (Census Output Areas level (COA), 2011 only), fuel poverty rate (Lower Layer Super Output Areas level (LSOA), 2014-2018), and avoidable death rate per 100,000 (Local Authority District level (LAD), 2014-2018).

 Carstairs index is a commonly used area-level measure of material deprivation in health

studies [12]. It was calculated using four variables from the 2011 census, including male unemployment, low social class, non-car ownership, and overcrowding. This variable has the highest spatial resolution among the three deprivation indicators chosen for this study, due to its geography being Census Output Area (COA, the highest spatial resolution of English Census geography of average population of 310 individuals). This indicator is time invariant as only 2011 values were available. Data was obtained via UK Data service (link: https://www.data-archive.ac.uk/ ) .  Annual fuel poverty rate is used to measure the percentage of households that were unable to

maintain standard thermal comfort and safety [13]. Fuel poverty has been increasingly recognised as a distinct form of social and health inequality [14]. It has been hypothesised that cold may be associated with excess winter deaths [15]. A cold home due to fuel poverty has been linked to respiratory problems, arthritis, and rheumatism in people of all ages, as well as mental health problems in adolescents[16]. This indicator is annual, covering the period 2014-2018. The geographic level is Lower Layer Super Output area (LSOA) level (Census geography category with average population of 1500 individuals) and covers the period 2014-2018. We extracted fuel poverty from GOV.uk (link: https://www.gov.uk/government/collections/fuel-poverty-statistics ) .  We used yearly avoidable death rate per 100,000 to measure health inequality. Mortality is an

outcome that can be clinically quantified, and avoidable mortality is amenable to policy intervention [17]. Avoidable death rate could therefore be used to capture the geographical disparity in health [17]. The definition of avoidable death rate can be found from https://consultations.ons.gov.uk/health-and-life-events/avoidable-mortality- definition/results/consultationresponse- reviewoftheavoidablemortalitydefinition_finalwithchangesimplemented.doc . The data is at Local Authority District (LAD) level (mean population of approximately 179,361.6 per local authority district), covering each year 2014-2018. We downloaded the data from Office for National Statistics (link: https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/causesofdeath/ bulletins/avoidablemortalityinenglandandwales/previousReleases ) .

Figure 1 Geography of the targeted postcodes. We adjusted for the percentage of non-white population per Local Authority District since ethnic concentration may be related to both deprivation and aircraft noise levels. This data was obtained from NOMIS, via Office for National Statistics (link: https://www.nomisweb.co.uk/ ) . We specified a random effects model to estimate the association between noise levels and quintiles of deprivation at postcode level. We clustered the variance at COAs, LSOAs, or LADs, depending on the geographic levels of deprivation variables. The equations are specified as: 𝑛𝑜𝑖𝑠𝑒 ௜௝௧ = 𝐶𝑎𝑟𝑠𝑡𝑎𝑖𝑟𝑠 ௝ + 𝑦𝑒𝑎𝑟 ௧ + 𝑒𝑡ℎ𝑛𝑖𝑐 ௞௧ + 𝑢 ௝ + 𝑒 ௜௝௧ (1) where 𝑖 represents individual postcode, 𝑗 represents individual Output Areas, 𝑘 represents individual Local Authority district and 𝑡 represents year. 𝑛𝑜𝑖𝑠𝑒 ௜௝௧ , 𝐶𝑎𝑟𝑠𝑡𝑎𝑖𝑟𝑠 ௝ and 𝑒𝑡ℎ𝑛𝑖𝑐 ௞௧ represents noise levels (continuous), quintiles of Carstairs index and quantiles of % ethnic minority population. 𝑛𝑜𝑖𝑠𝑒 ௜௝௧ = 𝐴𝑣𝑜𝑖𝑑𝑎𝑏𝑙𝑒 ௝௧ + 𝑦𝑒𝑎𝑟 ௧ +𝑒𝑡ℎ𝑛𝑖𝑐 ௞௧ + 𝑢 ௝௧ + 𝑒 ௜௝௧ (2) where 𝑖 represents individual postcode, 𝑗 represents individual LSOA, 𝑘 represents individual Local Authority district and 𝑡 represents year. 𝑛𝑜𝑖𝑠𝑒 ௜௝௧ , 𝐴𝑣𝑜𝑖𝑑𝑎𝑏𝑙𝑒 ௝௧ and 𝑒𝑡ℎ𝑛𝑖𝑐 ௞௧ represents noise levels (continuous), quintiles of avoidable death rate and quantiles of % ethnic minority population. 𝑛𝑜𝑖𝑠𝑒 ௜௝௧ = 𝐹𝑢𝑒𝑙𝑝𝑜𝑣𝑒𝑟𝑡𝑦 ௝௧ + 𝑦𝑒𝑎𝑟 ௧ + 𝑒𝑡ℎ𝑛𝑖𝑐 ௝௧ + 𝑢 ௝௧ + 𝑒 ௜௝௧ (3) where 𝑖 represents individual postcode, 𝑗 represents individual Local Authority district, and 𝑡 represents year. 𝑛𝑜𝑖𝑠𝑒 ௜௝௧ , 𝐹𝑢𝑒𝑙𝑝𝑜𝑣𝑒𝑟𝑡𝑦 ௝௧ and 𝑒𝑡ℎ𝑛𝑖𝑐 ௝௧ represents noise levels (continuous), quintiles of fuel poverty rate and quantiles of % ethnic minority population. Each postcode uniquely belongs to an Output Area, Lower Layer Super Output Area, and Local Authority district, which enables us to link data. Each regression examined a single pair of noise exposure and deprivation.

aha Heathrow Aiport Local Authorities in Greater London Postcodes © Local Authorities outside Greater London ~_~ ae

3. RESULTS

Table 1 shows the descriptive summary of the variables used in the analysis. There were 820,060 observations in the dataset. Lday and Lnight noise levels were unavailable for 41,099 observations, while 58,491 observations did not have any Laeq24, Leve, and Lden noise levels due to no flight activity above the postcodes at the time. The arithmetic means noise levels for LAeq24, Lday, Leve, and Lnight were 43.09, 44.49, 43.74, and 37.58 dB, respectively. The mean Carstairs index, fuel poverty, avoidable death rate per 100,000 persons, and % non-white were 0.936, 10.18, 133.3 and 36.26 on average.

(1) (2) (3) (4) (5) VARIABLES N mean sd min max

Laeq24 761,569 43.09 5.549 27.00 73.76 Lday 778,961 44.49 5.582 28.87 75.28 Leve 761,569 43.74 5.416 26.00 73.95 Lnight 778,961 37.58 6.317 20.30 70.25 Avoidable death rate per 100,000 persons 820,060 133.3 25.10 78 209.9 Carstairs Index 820,060 0.936 3.104 -4.876 28.31 Fuel poverty rate 820,060 10.18 3.635 1.800 29.60 % non-white population 808,990 36.26 14.51 4.4 68.9

Table 1 Descriptive summary of the variables

Table 2 illustrates the pairwise correlations between variables involved in analysis. There were very high correlations between the noise parameters LAeq24, Lday, Leve and Lnight. There is a fairly weak relationship between three deprivation variables and % non-white people (all below 0.5).

Variables (1) ( 2) (3) (4) (5) (6) (7) (8) (1) Laeq24 1.00 (2) Lday 1.00 1.00 (3) Leve 0.97 0.96 1.00 (4) Lnight 0.92 0.90 0.86 1.00 (5) Carstairs index 0.01 0.02 -0.03 0.06 1.00 (6) fuel poverty -0.08 -0.08 -0.09 -0.06 0.36 1.00 (7) Avoidable death rate 0.23 0.23 0.18 0.34 0.42 0.08 1.00 (8) % non-white populatio n -0 . 10 - 0.09 -0.12 -0.09 0.49 0.36 0.41 1.00 Table 2 Pairwise correlations between noise metrics, deprivation measures and control variable Tables 3 and 4 demonstrate the main results from regressions. Every column displays a relationship between one deprivation variable and one noise metric while controlling for quantiles of % non- white. The dependent variables were Laeq24 (24-hour average) and Lday (7:00 and 19:00) in Table 3, and Levening (19:00 – 23:00) and Lnight (23:00 – 04:30) in Table 4. We found that quintiles 2 – 5 of all deprivation variables exhibited a positive association, indicating that the least deprived areas were consistently the quietest. Quintiles of avoidable death rates were linked to Lday, Lnight, and Laeq24, but not to Levening. Notably, there was some evidence of a dose- response relationship with aircraft noise using the avoidable death rate index. For example, in Table 4, quintile 2 to 5 (Q2 to Q5) of avoidable death rate where Q5 has the highest rate were exposed to noise levels that were 0.31, 0.91, 1.07, and 1.23 dB louder at night than Q1. Table 3 shows that during daytime and 24-hour periods, Q4 and Q5 of avoidable death rate are significantly noisier than Q1, and Q2. This implies a trend in which areas with a higher avoidable death rate tend to be nosier. Our evidence shows that the Carstairs index quintiles were significantly associated with all three deprivation metrics. The strongest relationship was with night aircraft noise levels (coefficients: Q2

1.98, Q3 2.39, Q4 1.85, and Q5 2.31). Fuel poverty had a significant but relatively weak relationship with aircraft noise, compared to Carstairs and avoidable death rate. There is less conclusive evidence of gradients for the associations of noise with Carstairs index and fuel poverty rate. When Carstairs index was used, the noisiest two quintiles during day were Q2 (coefficient: 1.26) and Q5 (coefficient: 1.05), while during night these were Q3 (coefficient: 2.39) and Q5 (2.31). However, during evening, they were Q2 (coefficient: 1.19) and Q3 (coefficient: 0.76). The third and fourth quarters of the fuel poverty rate consistently experienced the loudest noise during day, evening, and night. (1) (2) (3) (5) (6) (7) Dependent variables Laeq24 Lday

Carstairs Index Q1 – least deprived (base)

Carstairs Index Q2 1.18*** 1.26*** (0.22) (0.26) Carstairs Index Q3 1.00*** 0.98*** (0.29) (0.33) Carstairs Index Q4 0.53** 0.46** (0.21) (0.23) Carstairs Index Q5 0.98*** 1.05*** (0.22) (0.24)

Avoidable death rate Q1 – least deprived (base)

Avoidable death rate Q2 0.17 0.15 (0.11) (0.10) Avoidable death rate Q3 0.49*** 0.51*** (0.18) (0.14) Avoidable death rate Q4 0.54** 0.59*** (0.22) (0.19) Avoidable death rate Q5 0.52** 0.55** (0.26) (0.22) Fuel poverty rate Q1 – least deprived (base)

Fuel poverty rate Q2 0.06*** 0.04** (0.02) (0.02) Fuel poverty rate Q3 0.08*** 0.06*** (0.02) (0.02) Fuel poverty rate Q4 0.09*** 0.06* (0.03) (0.04) Fuel poverty rate Q5 0.07** 0.06* (0.03) (0.03) Constant 43.35*** 43.80*** 44.04*** 44.66*** 45.09*** 45.37*** (0.13) (0.68) (0.13) (0.15) (0.71) (0.14)

Observations 751,948 751,948 751,948 768,798 768,798 768,798 Number of postcodes 154,173 154,173 154,173 162,004 162,004 162,004 % non-white quintiles YES YES YES YES YES YES

Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 Table 3 The association between quintiles of deprivation measures and 24-hour and daytime aircraft

noise

(1) (2) (3) (5) (6) (7) Dependent variables Leve Lnight

Carstairs Index Q1 – least deprived (base)

Carstairs Index Q2 1.19*** 1.98*** (0.22) (0.28) Carstairs Index Q3 0.76** 2.39*** (0.32) (0.40) Carstairs Index Q4 0.19 1.85*** (0.23) (0.23) Carstairs Index Q5 0.62** 2.31*** (0.24) (0.23) Avoidable death rate Q1 – least deprived (base)

Avoidable death rate Q2 0.20 0.31** (0.23) (0.15) Avoidable death rate Q3 0.54 0.91* (0.38) (0.48) Avoidable death rate Q4 0.56 1.07** (0.42) (0.51) Avoidable death rate Q5 0.55 1.23** (0.53) (0.55) Fuel poverty rate Q1 – least deprived (base)

Fuel poverty rate Q2 0.10*** 0.16*** (0.03) (0.03) Fuel poverty rate Q3 0.08** 0.23*** (0.03) (0.03) Fuel poverty rate Q4 0.09** 0.20*** (0.04) (0.04) Fuel poverty rate Q5 0.04 0.13*** (0.04) (0.04)

Constant 44.41*** 44.63*** 44.86*** 36.45*** 37.65*** 38.12*** (0.13) (0.65) (0.18) (0.15) (0.86) (0.15)

Observations 751,948 751,948 751,948 768,798 768,798 768,798 Number of postcodes 154,173 154,173 154,173 162,004 162,004 162,004 % non - white quintiles YES YES YES YES YES YES Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1

Table 4 The association between quintiles of deprivation measures and evening- and night-time

aircraft noise

4. DISCUSSION

We examined the association between noise levels (Laeq24, Lday, Leve, and Lnight) and three measures of deprivation in our study (Carstairs index, avoidable death rate and fuel poverty) while controlling for ethnic minority concentration per Local Authority district. Our findings consistently indicated that areas with the least deprivation were the quietest, independent of deprivation variables used. There was some evidence that areas experiencing the greatest material deprivation experienced the loudest aircraft noise (i.e. Q5 of the Carstairs index was the noisiest during daytime). However, for other deprivation metrics this was Q3 or Q4. Taken together, these findings suggested that aircraft noise near Heathrow has disproportionately impacted relatively less affluent areas (i.e. Q3, Q4 and Q5). The existing evidence for the link between deprivation and exposure to aircraft noise is limited, and ambiguous [18]. It is notable that studies used different definitions of deprivation and also used differing noise metrics. A recently published study examining 90 airports in the US, found that census block groups with a higher proportion of low educated residents were less likely to be exposed to daytime noise levels greater than day-night average sound level (DNL) 65 dB than census block groups with a high proportion of high educated residents [8]. It also noted that there was no evidence that lowest income census block groups were more likely to be noisy. Another study, looking at Heathrow airport matched annual average aircraft noise contours (2001) and London Travel Demand Survey, concluded that individuals with the highest household income had a greater likelihood of living within an annual average 50 dB contour in 2010 [9]. This is not inconsistent with our findings, which looked at noise at different times of day and looked at numerical levels rather than a binary measure of above/below 50dB. Heterogeneity in the research design (ecological versus individual-level), as well as the models used to estimate noise may affect the results – as indicated in our own analyses and in differences with published studies. Moreover, living in close proximity to good transportation links, employment patterns, and housing market dynamics may help offset unwillingness to live in areas with higher noise levels [8, 19]. One important motivation for this study is that deprivation, as a significant confounder in noise and health studies, is multifaceted [11]. Many studies' use of material deprivation may be insufficient to capture the multidimensional nature of deprivation. For instance, Siddiqi [20] argued that declining absolute economic conditions alone may not fully explain the increase in mortality among middle- class whites in the United States [21, 22], middle-aged (45-54) population in England [23], and young men (15-44) in Scotland [24]. Other factors, such as cumulative disadvantage [25] and diminishing opportunities for good health [26], may also contribute to mortality and morbidity. We included three deprivation measures in our study, which all showed a positive association with aircraft noise levels near Heathrow airport. Our results reveal, in general, fuel poverty has a positive but relatively weak relationship with aircraft noise, as compared to the Carstairs index and avoidable death rate. Although aircraft noise had a stronger relationship with the Carstairs index, the results present only a slight difference between relatively affluent (i.e. Q2) and deprived areas (i.e. Q4 and Q5). A discernible difference between wealthy (Q1 and Q2) and impoverished areas (Q4 and Q5) was observed only when the avoidable death rate was used. If this finding is replicated elsewhere, it may imply that deprivation could confound the health outcomes of aircraft noise exposure near London's Heathrow airport via different pathways (i.e. poor economic condition, excess winter deaths and health inequality). More studies are needed to confirm this. Our study has several strengths and limitations. One strength is its large dataset, which includes annual averaged noise levels for 164,012 postcodes near Heathrow airport. Additionally, we included a variety of deprivation domains, including not only poverty, but also health inequalities. Among the limitations are that we used area-level not individual-level noise and deprivation estimates therefore the ecological fallacy may apply; also that Carstairs index was only available for one year (the year of the Census). Heathrow airport is situated close to highly populated areas, some of which are very wealthy, so may not be representative of other airports.

5. CONCLUSION

We found a positive association between deprivation, as measured by the Carstairs index, avoidable death rate, and fuel poverty, and aircraft noise levels in 164,012 postcodes surrounding London Heathrow airport. However, the relationship did not have a clear gradient except for avoidable death rate. Our study is one of very few to investigate the relationship between aircraft noise and inequality. Our next steps are to discuss findings with community groups near the airport. As air transport increases post-pandemic, information on noise exposures as well as views from community groups can inform future airport policies 6. REFERENCES

1. European Environmental Agency, European Aviation Environmental Report 2019 , E.E. Agency. 2019. 2. Clark, C., Aircraft noise effects on health. Centre for Psychiatry, 2015. 3. Van Kempen, E., et al., WHO environmental noise guidelines for the European region: a systematic review on environmental noise and cardiovascular and metabolic effects: a summary. International journal of environmental research and public health, 2018. 15 (2): p. 379. 4. Vienneau, D., et al., Association between transportation noise and cardio-metabolic diseases: an update of the WHO meta-analysis. 2019. 5. Floud, S., et al., Medication use in relation to noise from aircraft and road traffic in six European countries: results of the HYENA study. Occupational and environmental medicine, 2011. 68 (7): p. 518-524. 6. Gong, X., et al., Association between Noise Annoyance and Mental Health Outcomes: A Systematic Review and Meta-Analysis. International Journal of Environmental Research and Public Health, 2022. 19 (5): p. 2696. 7. Huang, Y.-K., et al., Community daytime noise pollution and socioeconomic differences in Chicago, IL. Plos one, 2021. 16 (8): p. e0254762. 8. Simon, M.C., et al., Sociodemographic Patterns of Exposure to Civil Aircraft Noise in the United States. Environmental health perspectives, 2022. 130 (2): p. 027009. 9. Tonne, C., et al., Socioeconomic and ethnic inequalities in exposure to air and noise pollution in London. Environment international, 2018. 115 : p. 170-179. 10. Dreger, S., et al., Social inequalities in environmental noise exposure: A review of evidence in the WHO European Region. International journal of environmental research and public health, 2019. 16 (6): p. 1011. 11. Hajat, A., et al., Confounding by socioeconomic status in epidemiological studies of air pollution and health: challenges and opportunities. Environmental health perspectives, 2021. 129 (6): p. 065001. 12. Allik, M., et al., Developing a new small-area measure of deprivation using 2001 and 2011 census data from Scotland. Health & place, 2016. 39 : p. 122-130. 13. Liddell, C. and C. Morris, Fuel poverty and human health: a review of recent evidence. Energy policy, 2010. 38 (6): p. 2987-2997. 14. Simcock, N., G. Walker, and R. Day, Fuel poverty in the UK: Beyond heating. People, Place and Policy, 2016. 10 (1): p. 25-41. 15. Mercer, J.B., Cold—an underrated risk factor for health. Environmental Research, 2003. 92 (1): p. 8- 13. 16. Dear, K.B. and A.J. McMichael, The health impacts of cold homes and fuel poverty . 2011, British Medical Journal Publishing Group. 17. Tang, K.K., D. Petrie, and D.S.P. Rao, Measuring health inequality with realization of potential life years (RePLY). Health Economics, 2009. 18 (S1): p. S55-S75. 18. Agency, E.E., Unequal exposure and unequal impacts: social vulnerability to air pollution, noise and extreme temperatures in Europe. EEA Report NO 22/2018, 2018. 19. Brainard, J.S., et al., Exposure to environmental urban noise pollution in Birmingham, UK. Urban Studies, 2004. 41 (13): p. 2581-2600. 20. Siddiqi, A., et al., Growing sense of social status threat and concomitant deaths of despair among whites. SSM-population health, 2019. 9 : p. 100449.

21. Dasgupta, N., L. Beletsky, and D. Ciccarone, Opioid crisis: no easy fix to its social and economic determinants. American journal of public health, 2018. 108 (2): p. 182-186. 22. Case, A. and A. Deaton, Mortality and morbidity in the 21st century. Brookings papers on economic activity, 2017. 2017 : p. 397. 23. Joyce, R. and X. Xu, Inequalities in the twenty-first century: introducing the IFS Deaton Review. May 2019. 2019. 24. Allik, M., et al., Deaths of despair: cause-specific mortality and socioeconomic inequalities in cause- specific mortality among young men in Scotland. International journal for equity in health, 2020. 19 (1): p. 1-10. 25. Rehder, K., J. Lusk, and J.I. Chen, Deaths of despair: conceptual and clinical implications. Cognitive and behavioral practice, 2021. 28 (1): p. 40-52. 26. DeVerteuil, G., Deaths of despair and the social geographies of health denial. Geography Compass, 2022. 16 (2): p. e12607.