The Role of Traffic-Related Air Pollution on Neurodegenerative Diseases in Older People: An Epidemiological Perspective

Abstract

Traffic-related air pollution is ubiquitous and almost impossible to avoid. It is important to understand the role that traffic-related air pollution may play in neurodegenerative diseases, such as dementia, Alzheimer’s disease, and Parkinson’s disease, particularly among older populations and at-risk groups. There is a growing interest in this area among the environmental epidemiology literature and the body of evidence identifying this role is emerging and strengthening. This review focuses on the principal components of traffic-related air pollutants (particulate matter and nitrogen oxides) and the epidemiological evidence of their contribution to common neurodegenerative diseases. All studies reported are currently observational in nature and there are mixed findings depending on the study design, assessment of traffic-related air pollutant levels, assessment of the neurodegenerative disease outcome, time period of assessment, and the role of confounding environmental factors and at-risk genetic characteristics. All current studies have been conducted in income-rich countries where traffic-related air pollution levels are relatively low. Additional longer-term studies are needed to confirm the levels of risk, consider other contributing environmental factors and to be conducted in settings where air pollution exposures are higher and at-risk populations reside and work. Better understanding of these relationships will help inform the development of preventive measures and reduce chronic cognitive and physical health burdens (cost, quality of life) at personal and societal levels.

Keywords

Air pollution Alzheimer’s disease dementia mild cognitive impairment Parkinson’s disease

INTRODUCTION

Ambient air pollution is a major concern for health worldwide. Exposure to air pollution from either indoor or outdoor sources can increase the risk of morbidity and mortality [1]. The Global Burden of Disease study estimated that exposure to fine particulate matter (particulate matter with an aerodynamic diameter of <2.5 μm) is the fifth leading risk factor for mortality in the world and accounts for 4.2 million deaths and > 103 million disability-adjusted life years in 2015 [2]. Although air pollution is a universal issue, it is likely to cause the greatest harm in individuals with underlying vulnerability who are exposed to high level of air pollution. The severity of air pollution becomes even more evident in extraordinary situations like the current Coronavirus pandemic. Several studies indicate significant relations between air pollution and COVID-19 infections as well as related fatality rates [3, 4]. In recent years the evidence has accumulated that air pollution exposure has not only a harmful impact on the respiratory and cardiovascular system, but also on the brain and cognitive processes through vascular and inflammatory mechanisms [5]. An increasing number of neurotoxicological studies have demonstrated that exposures to airborne particles induce oxidative stress and widespread neuroinflammation and other neurotoxic reactions affecting multiple brain regions in animals [6]. Neuropathological examinations following autopsies of children and young adults living in urban areas with high levels of ambient air pollutants have also revealed increased biomarker of accelerated brain aging [5].

TRAFFIC-RELATED AIR POLLUTION

Ambient air pollution is a mixture of gases and suspended particulate matter (PM) in the air. Many of these pollutants are from anthropogenic sources, directly emitted from different processes such as combustion from traffic and industrial processes. In urban areas, air pollution is mostly generated by traffic, constituting up to 80% of airborne concentrations of PM in the urban environment [7]. Traffic-related air pollutants (TRAP) have substantially increased in recent decades [8]. TRAP can either be from exhaust or non-exhaust emissions. Pollution from exhaust emissions primarily consists of products of fuel combustion, such as diesel or gasoline. Non-exhaust emissions are the result of wear and tear of vehicle parts, such as brakes and tires. The main pollutants of TRAP are nitrogen oxides (NOx) particularly, nitrogen dioxide (NO₂); PM ultrafine (aerodynamic diameter <0.1 μm: UFP), fine (<2.5 μm: PM_2.5), and coarse (<10 μm: PM₁₀) and carbon dioxide (CO₂). These pollutants are considered primary pollutants as they are directly emitted. Additionally, when primary pollutants react with the atmosphere, secondary by-products such as ozone (O₃) and secondary aerosols are generated. The non-exhaust emissions are mainly PM generated by abrasion. The chemical composition of TRAP and the degrees of traffic exhaust is directly related to the quality and type of fuel (diesel or gasoline) used as well as type and age of the vehicle.

The two fractions of PM predominantly implicated in neurodegenerative effects are PM_2.5 and ultrafine PMs. Both PM_2.5 and UFPs are mostly from tailpipe and brake emissions from motor vehicles, aircrafts, and marine vessels [9]. Notably, due to their small size, UFPs are ideal for lung deposition, penetration, and exhibit effects extending beyond the respiratory tract [10, 11]. Distance from roadways are important predictors for both PM_2.5 and UFPs; however, UFPs usually show a stronger gradient from roadways than PM_2.5 [12]. Studies showed that UFP concentrations decrease exponentially with downwind distance from roadsides [13].

NEURODEGENERATIVE DISEASES

Neurodegenerative diseases are a heterogeneous group of disorders that are characterized by the progressive degeneration of the structure and function of the central nervous system or the peripheral nervous system with unknown etiology. These conditions, which are characterized by cognitive and functional decline, include mild cognitive impairment (MCI), dementia, Alzheimer’s disease (AD), and Parkinson’s disease (PD). These diseases affect millions of people worldwide. In the Global Burden of Disease Study in 2016, the estimated number of people living with dementia and AD was 43.8 million, and with PD was approximately 6 million worldwide. Globally, AD and dementia ranked third among the neurological disorders for the highest burden of disease, as assessed by disability-adjusted life years (DALYs) [14].

The onset of neurodegenerative diseases can occur at any stage throughout life but the risk of being affected by a neurodegenerative disease increases rapidly with age, particularly after the third decade of life. It is predicted that this prevalence will continue to increase as the global population grows and the aged demographic increases [14]. There is strengthening evidence of genetic predisposition to some neurodegenerative diseases with the Apolipoprotein E (APOE) 4 allele appearing to be an independent risk factor for AD.

METHODS

The aim of the review was to provide an over-view of the state-of-the–art on environmental epidemiological studies on the association between traffic-related air pollution exposure and neurodegenerative diseases in the elderly. A literature search using PubMed was performed using keywords such as, “Air pollution”, “Traffic”, “Cognition”, “Cognitive decline”, “Neurodegenerative diseases”, and “Dementia”, and by assessing the reference lists of other literature reviews and studies. After the removal of duplicate studies and the exclusion of animal studies, and non-English language studies, we assessed titles, abstracts and keywords. We only included papers which had their full text assessed according to the inclusion and exclusion criteria in duplicate. The review was focused on pollutants directly emitted from traffic-related sources such as NO_x/NO₂, PM₁₀ and PM_2.5. We did not include studies investigating other pollutants such as O₃ and SO₂.

HYPOTHESIZED PATHWAYS BETWEEN TRAP EXPOSURE AND NEURODEGENERATIVE DISEASES

Long-term exposure to TRAP may have adverse impacts on the central nervous system. Animal models exposed to components of TRAP have provided much information to inform current understanding of the potential mechanisms underlying the effects of air pollutants on the central nervous system. Subject to the pollutant, dose, exposure period, gender, age, and health status, a complex and diverse range of effects have been detected. These include the formation of free radicals and subsequent oxidative stress; neuronal damage; RNA and DNA damage; and identification of tau and amyloid-β (Aβ)₄₂ proteins which are early hallmark signs of AD and PD [15 –17]. Little is known about how the air pollutants affect the development of tau and Aβ₄₂; however, animal studies which modelled inhalation exposure to TRAP have detected increased activity of powerful matrix metalloproteinases and degradation of tight junction proteins in the brain’s microvasculature resulting in altered brain-blood barrier permeability and expression of neuroinflammatory markers [18]. The mechanisms and physiological pathways are currently not well understood but advancements in the field are underway. Based on these findings, a range of hypotheses have been proposed that suggest PM, diesel nanoparticles, and gases such as ozone in air pollution may enter the central nervous system via multiple pathways. These include by direct translocation via the olfactory nerve or through the olfactory/nasal mucosa by triggering a breakdown of the epithelial barrier and entering the brain; or by indirect pathways whereby the impact of air pollutants on peripheral parts of the body may exert effects on the brain [19]. An example of this is when air pollutants penetrate and damage the deeper parts of the lungs resulting in the release of cytokines and non-cytokine circulating signals into the blood stream which travel to brain and trigger microglial (the brain’s resident immune cells) activation and stimulate neuroinflammation [19], resulting in elevated reactive oxygen species and pro-inflammatory cytokines in the brain. However, the immune function of microglia, which is important for maintaining brain health, can become reprogrammed under these conditions and become a chronic source of elevated neurotoxic cytokines which are toxic to neurons, resulting in neuronal dysregulation [20]. This dysregulated immune response, via the microglia, may result in the accelerated accumulation of tau and Aβ₄₂, a neurotoxic fragment of the amyloid-β protein precursor, and alpha-synuclein in the brain. These depositions can lead to neuronal dysfunction and subsequently have an important role in neurodegenerative diseases [15 , 22].

There is also a growing body of research that suggests that cardiovascular and respiratory risk factors and conditions may possibly mediate the relationship between exposure to air pollutants and neurodegeneration [23 –25]. Global cortical Aβ deposition has also been linked with various cardiovascular diseases risk factors such as hypertension, diabetes, hypercholesterolemia, and cerebrovascular conditions including stroke [26]. Further, ambient air pollution exposure has been linked to stroke and cerebrovascular diseases [27].

EPIDEMIOLOGICAL EVIDENCE OF ASSOCIATIONS BETWEEN TRAP AND NEURODEGENERATIVE DISEASES

There is growing epidemiological evidence that TRAP exposure might lead to neurodegenerative diseases [28]. Higher exposures to a range of TRAP has been found to be associated with increased risk of neurocognitive diseases in several studies. Table 1 summarizes the key characteristics and results from recent TRAP and neurodegenerative diseases studies. Table 2 summarizes and highlights the diverse pollutants assessed with the type of exposure modelling used, alongside common neurodegenerative diseases. We report the findings from these studies in this review grouped by the neurocognitive outcome and then by study design.

Table 1

Overview of papers on traffic related air pollution and neurodegenerative outcomes: Key characteristics and results

Author, publication year [Ref]; setting	Study type; time period	Sample type, age groups and size	Results
Case-control studies
Kirrane et al. 2015 [31]; USA	Nested case control; 1999–2010	Farmers and spouses; aged 12–92 years; n = 83,343	0.71 μg/m³ increase in PM_2.5 and incidence of PD (OR = 1.34; 0.93 to 1.93).
Wu et al. 2015 [41]; Taiwan	Case-control; 2007–2010	Alzheimer’s Disease (AD) n = 249; Vascular dementia (VaD) n = 125; controls n = 497	Highest vs lowest tertile of PM10 was associated with increased AD risk (OR = 4.17, 2.31 to 7.54) and increased VaD risk (OR = 3.61, 1.67 to 4.33).
			Non-carriers of APOE ɛ4: increased risk of AD (OR = 4.24, 2.11 to 8.54) and VaD (OR = 2.60, 1.12 to 6.02).
			Carriers of APOE ɛ4: increased risk of AD (OR = 3.50, 1.08 to 11.29) but the sample was not large enough to examine associations with VaD.
			Interaction analysis not statistically significant.
Ritz et al. 2016 [30]; Denmark	Case control; 1996 to 2009	Mean age 62 years; n = 1696.	IQR increases of 2.97 μg/m³ increases in NO₂ associated with risk of PD (OR = 1.09; 1.03 to 1.16).
			Subjects living for≥20 years in the capital city had higher risks (OR = 1.21; 1.11 to 1.31) than for those living in provincial towns (OR = 1.10; 0.97 to 1.26)
Chen et al. 2017 [32]; Taiwan	Nested case-control; 1995–2013	National health insurance research dataset; aged > 40 years; with PD n = 1060, without PD n = 4020	Higher levels of PM₁₀ exposure (>65 μg/m3) compared to lower levels of PM₁₀ (≤54 μg/m3) and the incidence of PD (OR = 1.35; 1.12 to 1.62). No significant associations with NOx, NO₂, NO.
Li et al. 2019 [33]; Taiwan	Nested case-control; 2005–2013	National administrative dataset; aged≥65 years; incident cases of vascular dementia n = 831, controls n = 3324	Increase from the lowest to the highest quartile of NO₂ associated with increased risk of vascular dementia (HR = 2.22; 1.35 to 3.65). No significant associations between increased PM₁₀ levels except when they increased from the 25^th to 50^th percentiles (HR = 1.27; 1.01 to 1.61).
Cohort studies
Chang et al. 2014 [44]; Taiwan	Retrospective cohort; 2000–2010	National health insurance research dataset;>50 years; n = 29,547	Increases from the lowest to highest quartile of NO₂ exposure and dementia incidence (HR = 1.54; 1.34 to 1.77).
Palacios et al. 2014 [36]; USA	Prospective cohort; 1990–2008	Mean age at baseline = 58±7 years; n = 115,767 women	Increases from the lowest to highest quartiles and incidence of PD: PM_2.5 (HR = 1.08; 0.81 to 1.45), PM₁₀ (HR = 0.99; 0.84 to 1.16) and PM_2.5–10 (HR = 0.92; 0.71 to 1.19).
Jung et al. 2015 [45]; Taiwan	Retrospective cohort; 2001–2010	National health insurance research dataset; aged≥65 years; n = 95,690	4.34 μg/m³ increases in PM_2.5 associated with an increased risk of incident AD (HR = 2.38; 2.21 to 2.56).
Kioumourtzoglou et al. 2016 [35]; USA	Population-based cohort; 1999 to 2010	Medicare enrollees; aged≥65 years; n = 9.8 million	1 μg/m³ increase in PM_2.5 associated with increased risk of incident hospitalization for dementia (HR = 1.08; 1.05 to 1.11), AD (HR = 1.15; 1.11 to 1.19), and PD (HR = 1.08; 1.04 to 1.12).
Oudin et al. 2016 [28]; Sweden	Cohort; 1993–1995 to 2008–2010	General population sample; aged 55 to 85 years; n = 1806	Higher annual levels of NO₂ (third quartile > 17–26 μg/m³) compared with lower levels (lowest quartile 4.8–9 μg/m³).
			Higher NO₂ levels associated with dementia (HR = 1.49; 1.04 to 2.14); AD (HR = 1.51; 0.96 to 2.37) and vascular dementia (HR = 1.46; 0.83 to 2.61).
Tzivian et al. 2016 [29]; Germany	Cohort (cross-sectional); 2006-2008	Mean age = 64±8 years; n = 2050	IQR increases and MCI incidence: 2.09 μg/m³ increases in PM₁₀ (OR = 1.11; 95% CI 0.99 to 1.23); 1.44 μg/m³ increases in PM_2.5 (OR = 1.16; 1.05 to 1.27); 6.11 μg/m³ increases in NO₂ (OR = 1.10; 0.97 to 1.25); 15.7 μg/m³ increases in NOx (OR = 1.10; 0.96 to 1.26).
			Total traffic load > 5000 cars/day in 100m buffer (OR = 1.00; 0.94 to 1.07).
Cacciottolo et al. 2017 [39]; USA	Cohort; 1999–2010	European women in USA aged 65–79 years; n = 3647	Risk of incident all-cause dementia was higher in areas with high PM_2.5 (3-year average > 12 μg/m³) compared to lower PM_2.5 (HR = 1.92; 1.32 to 2.80).
			Risk changed with APOE 4 status: APOE 3/3 (HR = 1.68; 0.97 to 2.92); APOE 3/4 (HR = 1.91; 1.17 to 3.14); and APOE 4/4 (HR = 3.95; 1.18 to 13.19).
Chen et al. 2017 [34]; Canada	Population-based cohort; 2001–2013	Registry; aged 55–85 years; n = 243,611	IQR increases and incident dementia: 4.8 μg/m³ increase in PM_2.5 (HR = 1.03; 1.02 to 1.04) and 14.2 ppb increase in NO₂ (HR = 1.10; 1.08 to 1.12).
Palacios et al. 2018 [37]; USA	Prospective cohort; 1988–2000	Mean age at baseline = 57±10 years; n = 50,352 men	Increases from the lowest to highest quintile and incidence of PD: PM_2.5 (HR = 0.97; 0.72 to 1.32), PM₁₀ (HR = 0.85; 0.63 to 1.15) and PM_2.5–10 (HR = 0.88; 0.64 to 1.22).
Andersson et al. 2018 [47]; Sweden	Prospective cohort; 1993–1995 to 2008–2010	General population sample; aged 55–85 years; n = 1721	Increases from lowest quartile (<9 μg/m³) to third quartile (17–26 μg/m³) of NOx levels associated with risk of incident dementia (HR = 1.48; 1.03 to 2.03). Interaction models between ambient noise and NOx were not significant.
Carey et al. 2018 [42]; UK	Retrospective cohort; 2005–2013	Primary care data; aged 50–79 years; n = 130,978	Highest quintile of NO₂ (>41.5 μg/m³) compared to those living in the lowest quintile (<31.9 μg/m³) were at higher risk of dementia (HR = 1.40; 1.12 to 1.74) and AD (HR = 1.50; 1.08 to 2.08); but not vascular dementia (HR = 1.01; 0.66 to 1.39).
			IQR increase (0.58 μg/m³) in PM_2.5 associated with borderline increased risk for dementia (HR = 1.08; 0.99 to 1.18).
Cerza et al. 2019 [43]; Italy	Prospective cohort; 2001–2013	Administrative dataset; aged 65–100 years; n = 350,844	Risk for first hospitalization for VaD: 10 μg/m³ increases in PM₁₀ (HR = 1.06; 1.02 to 1.10), 5 μg/m³ increases in PM_2.5 (HR = 1.07; 1.01 to 1.12), 10 μg/m³ increases in NO₂ (HR = 1.05; 1.03 to 1.07), 20 μg/m³ increases in NOx (HR = 1.08; 1.06 to 1.10); living within 50 meters of a major road (HR = 1.17; 1.10 to 1.24), living within 50–100 meters (HR = 1.11; 1.03 to 1.19) and living within 101–200 meters (HR = 1.10; 1.03 to 1.17) compared to living beyond 300 meters.
			Risk for first hospitalization for AD: 10 μg/m³ increases in PM₁₀ and (HR = 0.95; 0.91 to 0.99), 5 μg/m³ increases in PM_2.5 (HR = 0.91; 0.87 to 0.94), 10 μg/m³ increases in NO₂ (HR = 0.91; 0.89 to 0.94) 20 μg/m³ increases in NOx (HR = 0.96; 0.94 to 0.98); and living within 50 meters of a major road (HR = 0.97; 0.90 to 1.04) compared to living beyond 300 meters.
Grande et al. 2019 [46]; Sweden	Prospective cohort; 2001–2004	Population-based; aged > 60 years; n = 3363	Risk for incident dementia per 0.88 μg/m³ increases in PM_2.5 (HR = 1.54; 1.33 to 1.78), and per 8.35 μg/m³ increases in NOx (HR = 1.14; 1.17 to 1.75). Comorbid cardiovascular disease increased the magnitude of risks: heart failure (PM_2.5 HR = 1.93; 1.54 to 2.43; NOx HR =1.43; 1.17 to 1.75), and ischemic heart disease (PM2.5 HR = 1.67; 1.32 to 2.12; NOx HR = 1.36; 1.07 to 1.71).
Lee et al. 2019 [38]; USA	Retrospective cohort; 2000–2013	Medicare enrollees; aged≥65 years; n = 13 million	1 μg/m³ increases in annual average PM_2.5 associated with an increased risk of dementia hospitalization (HR = 1.049; 1.048 to 1.051) and VaD hospitalization (HR = 1.086; 1.082 to 1.090).
Oudin et al. 2019 [40]; Sweden	Cohort; 1993–1995 to 2008–2010	General population sample; aged 55 to 85 years; n = 1567	Increase from the lowest to the third quartile of NOx associated with incidence of VaD or AD (HR = 1.52; 1.08 to 2.12); incidence of AD only (HR = 1.72; 1.12 to 2.65).
Yuchi et al. 2020 [48]; Canada	Population-based cohort; 1994–2003	Administrative health dataset; all adults 45–84 years in Vancouver; 7.4 million person-years observations.	Living near major roads (<50 meters) or a highway (<150 meters) associated with PD (HR = 1.07; 0.96 to 1.18), NAD (HR = 1.14; 1.07 to 1.20), and AD (HR = 1.19; 0.95 to 1.49).
			IQR increase in PM_2.5 associated with risk of PD (HR = 1.09; 1.02 to 1.16), NAD (HR = 1.02; 0.98 to 1.05) and AD (HR = 0.90; 0.76 to 1.07).
			IQR increase in NO₂ associated with risk of PD (HR = 1.12; 1.05 to 1.20), NAD (HR = 1.02; 0.99 to 1.06) and AD (HR = 0.84; 0.70 to 0.99).
			Effect estimates did not alter when environmental noise or greenness were included.

Table 2

Overview of papers on traffic related air pollution and neurodegenerative outcomes: exposures; models; and outcomes

Exposure models used: I, Interpolation approach; E, Emissions approach with dispersion modelling; L, Land Use Regression model; D, Distance to (major) road.

Only one study reported investigations between TRAP exposure and MCI on a cross-sectional level. This longitudinal study in Germany reported borderline associations with interquartile (IQR) increases in PM₁₀ (Odds Ratio [OR] = 1.11; 95% CI 0.99 to 1.23), NO₂ (OR = 1.10; 0.97 to 1.25), and NOx (OR = 1.10; 0.96 to 1.26) and moderate associations with PM_2.5 (OR = 1.16; 1.05 to 1.27) (Table 2). Adjustment for APOE 4 status slightly decreased the strength of association between PM_2.5 and MCI but it remained significant. Total traffic load of > 5000 vehicles per day (adjusted for background NO₂ levels) in a 100-metre buffer around the residential location was not significantly associated with MCI (OR = 1.00; 0.94 to 1.07). The association between PM_2.5 and MCI were stronger among participants with no or moderate alcohol consumption (OR = 1.27; 1.07 to 1.50) than in participants with high alcohol consumption (OR = 0.96; 0.75 to 1.21); in current and former smokers (OR = 1.39; 1.12 to 1.71) than in non-smokers (OR = 1.01; 0.85 to 1.21); and, in participants exposed to higher outdoor noise levels (≥60 dBA) at their residence (OR = 1.30; 1.01 to 1.67) than those exposed to lower noise levels (<60 dBA) (OR = 1.10; 0.93 to 1.29) [29].

The majority of studies examined TRAP exposure and dementia and dementia subtypes: vascular dementia, AD, and PD. This section will first report on the findings from case-control studies. In Denmark an investigation of 1,696 patients with PD reported that 2.97 μg/m³ increases in NO₂ were associated with a higher risk of PD (OR = 1.09; 1.03 to 1.16). For subjects living for≥20 years in the capital city, the risks were higher (OR = 1.21; 1.11 to 1.31) than for those living in provincial towns (OR = 1.10; 0.97 to 1.26), and there were no associations among those who resided in rural areas [30]. In the US, a nested case-control study in an occupational cohort observed a non-significant increased risk between 0.71 μg/m³ increase in PM_2.5 and incidence of PD (OR = 1.34; 0.93 to 1.93) [31]. In Taiwan, a study reported a risk between higher levels of PM₁₀ exposure (>65 μg/m³) compared to lower levels of PM₁₀ (≤54 μg/m³) and the incidence of PD (OR = 1.35; 1.12 to 1.62). There were no significant associations with the other TRAP that were assessed (NOx, NO₂, NO) [32]. In a similar study, it was reported that increased NO₂ from the lowest quartile to the highest quartile was associated with increased risk of vascular dementia (HR = 2.22; 1.35 to 3.65). There were no significant associations between increased PM₁₀ levels except when they increased from the 25th to 50th percentiles (HR = 1.27; 1.01 to 1.61) [33].

The case-control studies demonstrated mixed strengths and directions of associations between NOx, NO₂, and PM₁₀ and these dementia and subtypes outcomes. Building on these findings, this paper will now discuss the results from prospective and retrospective cohort studies to further elucidate what the current evidence in this field indicates. In Canada, a longitudinal study of residents aged 55–85 years reported that the risk of incident dementia was associated with every 4.8 μg/m³ increase in PM_2.5 (Hazard Ratio [HR] = 1.03; 1.02 to 1.04), and with every 14.2 ppb increase in NO₂ (HR = 1.10; 1.08 to 1.12) [34]. In 50 US cities, a longitudinal analysis of all Medicare enrollees aged≥65 years reported that a 1 μg/m³ increase in PM_2.5 was associated with increased risk of incident hospitalization for dementia (HR = 1.08; 1.05 to 1.11), AD (HR = 1.15; 1.11 to 1.19), and PD (HR = 1.08; 1.04 to 1.12). They also found higher risks associated with 5 μg/m³ increases in PM_2.5 and hospitalization for dementia (HR = 1.46; 1.29 to 1.66), AD (HR = 2.00; 1.70 to 2.35), and PD (HR = 1.44; 1.22 to 1.70) [35]. Conversely, a prospective cohort study of 115,767 women (nurses) found no significant associations between increases from the lowest to highest quartiles of PM_2.5 (HR = 1.08; 0.81 to 1.45), PM₁₀ (HR = 0.99; 0.84 to 1.16), and PM_2.5–10 (HR = 0.92; 0.71 to 1.19) and PD [36]. Also, a prospective cohort study of 50,352 men (health professionals) also found no significant associations between increased cumulative exposure from the lowest to highest quintile of PM_2.5 (HR = 0.97; 0.72 to 1.32), PM₁₀ (HR = 0.85; 0.63 to 1.15), and PM_2.5–10 (HR = 0.88; 0.64 to 1.22) and PD [37]. Another longitudinal analysis of 13 million Medicare enrollees reported 1 μg/m³ increases in annual average PM_2.5 were associated with an increased risk of hospitalization for dementia (HR = 1.049; 1.048 to 1.051) and vascular dementia (HR = 1.086; 1.082 to 1.090). The magnitude of risk increased with urbanization with lower risk in rural (HR = 1.036; 1.031 to 1.041) compared with urban areas (HR = 1.052; 1.050 to 1.054) [38]. In a study of older women in the US, participants who resided in locations with high PM_2.5 (3-year average exposure was >12 μg/m³), the risk for all-cause dementia was increased by 92% compared with those who lived in areas with lower levels of PM_2.5 (HR = 1.92; 1.32 to 2.80). Their analysis indicated that this risk appeared to be altered depending on which APOE allele they had, namely those with APOE 3/3 (HR = 1.68; 0.97 to 2.92); APOE 3/4 (HR = 1.91; 1.17 to 3.14); and APOE 4/4 (HR = 3.95; 1.18 to 13.19). Carriers of APOE 4 allele appear to be at higher risk of developing dementia if they lived in areas with higher levels of air pollution although in this study the 95% CI overlapped and the differences in risk between the groups may not be clinically significant [39]. This has been supported by recent findings from Sweden and Taiwan where APOE 4 status interactions models with levels of NOx [40] and PM₁₀ [41] and dementia and AD incidence did not demonstrate that APOE 4 status had modifying effects.

In a London-based primary-care based cohort study, mid-older aged adults living in areas with the highest quintile of NO₂ (>41.5 μg/m³) compared to those living in the lowest quintile (<31.9 μg/m³) were at higher risk of dementia (HR = 1.40; 1.12 to 1.74) and AD (HR = 1.50; 1.08 to 2.08); but not vascular dementia (HR = 1.01; 0.66 to 1.39). With each 7.5 μg/m³ increase in NO₂ (adjusted for other air pollutants and noise) the risks for dementia increased (HR = 1.15; 1.04 to 1.28). Participants living in the highest quintile of PM_2.5 (>0.75 μg/m³) compared to the lowest quintile (<1.04 μg/m³) had an increased risk of dementia (HR = 1.26; 1.04 to 1.54) and AD (HR = 1.46; 95% CI 1.08 to 1.98) but not vascular dementia (HR = 0.99; 95% CI 0.68 to 1.44). With each 0.58 μg/m³ increase in PM_2.5, there was a borderline increased risk for dementia (HR = 1.08; 0.99 to 1.18). There were no significant associations between living closer to a major road and the incidence of any of these conditions. The strength of this analysis was the availability of individual risk factors on the participants which included important comorbid behaviors or conditions identified as risk factors for neurocognitive decline such as smoking, alcohol consumption, body mass index, history of depression, and cardiovascular conditions [42].

An Italian administrative cohort dataset followed up older adults without dementia for 12 years until their first hospitalization for numerous forms of dementia (including forms related to Creutzfeldt-Jakob disease, senile dementia, pre-senile dementia, persistent mental disorders, Pick’s disease, or dementia with Lewy bodies) and dementia subtypes (vascular dementia, AD, and senile dementia). This definition of incident dementia is much broader than other studies; however, they did report specifically on vascular dementia, AD, and senile dementia. The risk of first hospitalization for vascular dementia increased with 10 μg/m³ increases in PM₁₀ (HR = 1.06; 1.02 to 1.10), 5 μg/m³ increases in PM_2.5 (HR = 1.07; 1.01 to 1.12), 10 μg/m³ increases in NO₂ (HR = 1.05; 1.03 to 1.07), 20 μg/m³ increases in NOx (HR = 1.08; 1.06 to 1.10); and living within 50 meters of a major road (HR = 1.17; 1.10 to 1.24), living within 50–100 meters (HR = 1.11; 1.03 to 1.19), and living within 101–200 meters (HR = 1.10; 1.03 to 1.17) compared to living beyond 300 meters. However, the risk of first hospitalization for AD was significantly lower, in fact indicated some protection, with increased levels of PM₁₀ (HR = 0.95; 0.91 to 0.99), PM_2.5 (HR = 0.91; 0.87 to 0.94), NO₂ (HR = 0.91; 0.89 to 0.94), and NOx (HR = 0.96; 0.94 to 0.98) and, also living within 50 meters of a major road (HR = 0.97; 0.90 to 1.04). There is the possibility that diagnosis misclassification may be a potential bias in this study as a number of neurocognitive conditions cannot be definitively diagnosed prior to postmortem examination (e.g., AD), and this may contribute to the differing risk estimates reported. Also, as they were unable to control for key individual risk factors such as smoking, alcohol consumption, body mass index, and history of depression or cardiovascular conditions, there may be effects of residual confounding in the analyses [43].

In Sweden a study reported that those living in areas with higher annual levels of NO₂ (third quartile >17–26 μg/m³) compared with those living in areas with lower levels (lowest quartile 4.8–9 μg/m³) had a higher risk of incident dementia (HR = 1.49; 1.04 to 2.14); and non-significant increased risk of AD (HR = 1.51; 0.96 to 2.37) and vascular dementia (HR = 1.46; 0.83 to 2.61) [28].

In Taiwan, one retrospective cohort study reported a significant risk between lowest to highest quartile of NO₂ exposure and incident dementia (HR = 1.54; 1.34 to 1.77) [44]. Another study, using the same data, reported that 4.34 μg/m³ increases in PM_2.5 was associated with an increased risk of incident AD (HR = 2.38; 2.21 to 2.56) [45].

A recent study in Sweden explored the impact of cardiovascular disease on the risk between exposure to PM_2.5 and NOx at the residential address and incident dementia. They reported increased risk for incident dementia per 0.88 μg/m³ increases in PM_2.5 (HR = 1.54; 1.33 to 1.78), and per 8.35 μg/m³ increases in NOx (HR = 1.14; 1.17 to 1.75). Comorbid cardiovascular disease (CVD) increased the magnitude of these risks: heart failure (PM_2.5 HR = 1.93; 1.54 to 2.43; NOx HR = 1.43; 1.17 to 1.75) and ischemic heart disease (PM2.5 HR = 1.67; 1.32 to 2.12; NOx HR = 1.36; 1.07 to 1.71). A mediation analysis indicated that 49.4% of the association between PM_2.5 and dementia was explained by preceding stroke and that direct effects of PM_2.5 accounted for only a small fraction of the risks. No significant mediation relationships were identified between cardiovascular disease, NOx levels, and dementia [46]. It appears that CVD is an emerging mediator for the effects of PM_2.5 on neurodegenerative conditions.

It is becoming increasingly clear that exposure to TRAP does not affect neurological disorders in isolation from other spatially correlated environmental exposures such as traffic-related noise or exposure to natural greenness (vegetation) but disentangling the confounding, protective, or additive effects is challenging. A study in Sweden examined the independent and combined effects of TRAP and noise on the incidence of dementia. In fully adjusted models they found NOx levels between 17–26 μg/m³ was associated with increased risk of dementia compared to <9 μg/m³ (HR = 1.48; 1.03 to 2.03) but environmental noise levels ≥55 dB (24-h average) were not associated with increased risk of dementia compared with lower noise levels <55 dB (HR = 0.95; 0.57 to 1.57). Interaction models between noise and NOx reported no significant results [47]. A recent Canadian study reported that living near major roads (<50 meters) or a highway (<150 meters) was associated with increased risk for non-AD (HR = 1.14; 1.07 to 1.20), and borderline increased risk for AD (HR = 1.19; 0.95 to 1.49) and PD (HR = 1.07; 0.96 to 1.18). These risks appeared to be attenuated by higher levels of greenness in the residential locations: non-AD (HR = 1.12; 1.05 to 1.20), AD (1.02; 0.77 to 1.36), and PD (HR = 0.99; 0.88 to 1.13). Exposure to higher levels of air pollutants, except NO, were generally trending toward increased risks for non-AD and PD but not AD. An increase of 1.54 to 1.65 μg/m³ in PM_2.5 was associated with increased risk of PD (HR = 1.09; 1.02 to 1.16), and tendency to increased risk of non-AD (HR = 1.02; 0.98 to 1.05) and AD (HR = 0.90; 0.76 to 1.07) and these associations did not alter when noise or greenness were included in the modelling. These results were similar in magnitude and direction for NO₂ [48]. The results from these studies suggest that traffic-related noise does not modify associations between TRAP and neurodegenerative diseases in these settings. But these results do not rule out effects that may be occurring in settings where environmental noise levels are higher than those in this study.

It is important to note that these studies have been conducted in the northern hemisphere and in high income countries where TRAP levels are relatively low and regulated by government laws. There is currently research lacking in low to middle income countries (LMIC) where increasing traffic density is contributing to higher TRAP concentrations, air pollution controls are less well regulated and other risk factors for neurodegenerative diseases may not be well understood. With the predications that global dementia prevalence is to increase from 50 million people in 2015 to 152 million in 2050, with the majority residing in LMIC [49], any contributing environmental exposure that can be reduced or prevented could lessen the incidence and burden of these conditions on individuals, communities, health systems and countries.

TRAP EXPOSURE ASSESSMENT

Accurate assessment of TRAP exposure is essential when attempting to evaluate the impact of TRAP on health, particularly for neurodegenerative health, in epidemiological studies. To date, ambient TRAP levels have been assessed using a variety of modelling methodologies with the aim of obtaining higher-resolution and accurate exposure estimates at the residential address of study participants (Table 2). These models have ranged from 1) Interpolation approaches from measurements obtained at fixed or mobile air quality monitoring sites; 2) Land Use Regression (LUR) models or satellite-derived estimates to improve predicted levels in areas which are located far from air quality monitoring stations; 3) Emissions approach with dispersion modelling which combines atmospheric chemistry with meteorological variables; in some studies, where modelled exposure metrics are not available, distance to nearest road or major road is used as a proxy measure. Individual exposure measurement methodologies are under development but remain expensive to undertake and quite onerous for the average person.

However, as modelling approaches require greater extrapolation of the air pollution exposure data, then uncertainty in the exposure metric increases. There is likely to be exposure misclassification within these TRAP modelling approaches, as modelled estimates at a single location are unlikely to represent the true exposure of a population that is not stationary. This is particularly relevant for air pollutants that exhibit a high degree of spatial variability, such as NO₂, but less so for pollutants that exhibit a more uniform distribution such as PM₁₀ and PM_2.5. Other challenges remain associated with the difficulty in disaggregating the independent effects of individual air pollutant types and environmental exposures, such as greenspace or noise, that are moderately to highly correlated with each other.

CONCLUSION

As can be seen from the scope of papers in this review, there has been increasing interest in the effects of TRAP on the central nervous system and neurodegeneration. The exact mechanisms of air pollution on neurodegenerative diseases are largely unknown, despite the fact these diseases affect increasingly large proportions of the aging population. As it stands, the evidence is growing that higher TRAP exposure is associated with incidence of a range of neurodegenerative diseases. The majority of evidence is based on studies that have focused on PM_2.5 from vehicle tailpipe and non-tailpipe emissions. This indicates that small-spatial and long-term exposure to TRAP may influence the risk of developing neurodegenerative disorders. Neurodegenerative disorders usually develop later in life, so exposures during infancy or early life can often not be investigated. To date, there is no clear consensus on what is the most informative exposure period in which to assess the impacts of TRAP on neurological function or neurodegenerative diseases [50]. Unlike many personal risk factors for neurodegeneration, exposure to TRAP is ubiquitous and generally unavoidable during ‘normal life circumstances’ and is life-long in nature. Ongoing recognition of the risks to health that TRAP poses is needed to ensure exposure levels at the population-level are minimized. From this, there is enormous potential for health benefits (physical, cognitive, and quality of life) and health care cost-savings if the risk of neurocognitive diseases can be reduced. Presently, ongoing research is required to support robust policy and personal recommendations.

DISCLOSURE STATEMENT

Authors’ disclosures available online (https://www.j-alz.com/manuscript-disclosures/20-0813r1).

References

McGlade

, Landrigan

(2019) Five national academies call for global compact on air pollution and health. Lancet 394, 23.

World Health Organization (WHO), Air Pollution, https://www.who.int/health-topics/air-pollution#tab=tab_1.

, Nethery

, Sabath

, Braun

, Dominici

(2020) Exposure to air pollution and COVID-19 mortality in the United States: A nationwide cross-sectional study. medRxiv, 2020.04.05.20054502.

Domingo

, Rovira

(2020) Effects of air pollutants on the transmission and severity of respiratory viral infections. Environ Res 187, 109650.

Calderón-Garcidueñas

, Torres-Jardón

, Kulesza

, Mansour

, González-González

, Gónzalez-Maciel

, Reynoso-Robles

, Mukherjee

(2020) Alzheimer disease starts in childhood in polluted Metropolitan Mexico City. A major health crisis in progress. Environ Res 183, 109137.

Block

, Calderon-Garciduenas

(2009) Air pollution: Mechanisms of neuroinflammation and CNS disease. Trends Neurosci 32, 506–516.

Pant

, Harrison

(2013) Estimation of the contribution of road traffic emissions to particulate matter concentrations from field measurements: A review.. Atmos Environ 77, 78–97.

Bowatte

, Erbas

, Lodge

, Knibbs

, Gurrin

, Marks

, Thomas

, Johns

, Giles

, Hui

, Dennekamp

, Perret

, Abramson

, Walters

, Matheson

, Dharmage

(2017) Traffic-related air pollution exposure over a 5-year period is associated with increased risk of asthma and poor lung function in middle age. Eur Respir J 50, 1602357.

Karagulian

, Belis

, Dora

CFC

, Prüss-Ustün

, Bonjour

, Adair-Rohani

, Amann

(2015) Contributions to cities’ ambient particulate matter (PM): A systematic review of local source contributions at global level. Atmos Environ 120, 475–483.

10.

Rothen-Rutishauser

, Muhlfeld

, Blank

, Musso

, Gehr

(2007) Translocation of particles and inflammatory responses after exposure to fine particles and nanoparticles in an epithelial airway model. Part Fibre Toxicol 4, 9.

11.

Muhlfeld

, Rothen-Rutishauser

, Blank

, Vanhecke

, Ochs

, Gehr

(2008) Interactions of nanoparticles with pulmonary structures and cellular responses. Am J Physiol Lung Cell Mol Physiol 294, L817–829.

12.

Zwack

, Paciorek

, Spengler

, Levy

(2011) Modeling spatial patterns of traffic-related air pollutants in complex urban terrain.. Environ Health Perspect 119, 852–859.

13.

Kozawa

, Winer

, Fruin

(2012) Ultrafine particle size distributions near freeways: Effects of differing wind directions on exposure. Atmos Environ (1994) 63, 250–260.

14.

Feigin

, Roth

, Naghavi

, Parmar

, Krishnamurthi

, Chugh

, Mensah

, Norrving

, Shiue

, Ng

, Estep

, Cercy

, Murray

CJL

, Forouzanfar

(2016) Global burden of stroke and risk factors in 188 countries, during 1990-2013: A systematic analysis for the Global Burden of Disease Study 2013. Lancet Neurol 15, 913–924.

15.

Calderón-Garcidueñas

, Reynoso-Robles

, Vargas-Martínez

, Gómez-Maqueo-Chew

, Pérez-Guillé

, Mukherjee

, Torres-Jardón

, Perry

, Gónzalez-Maciel

(2016) Prefrontal white matter pathology in air pollution exposed Mexico City young urbanites and their potential impact on neurovascular unit dysfunction and the development of Alzheimer’s disease. Environ Res 146, 404–417.

16.

Brady

, Morfini

(2017) Regulation of motor proteins, axonal transport deficits and adult-onset neurodegenerative diseases. Neurobiol Dis 105, 273–282.

17.

Meyer

, Tiepolt

, Barthel

, Hesse

, Sabri

(2014) Radioligand imaging of α4β2^* nicotinic acetylcholine receptors in Alzheimer’s disease and Parkinson’s disease. Q J Nucl Med Mol Imaging 58, 376–386.

18.

Oppenheim

, Lucero

, Guyot

, Herbert

, McDonald

, Mabondzo

, Lund

(2013) Exposure to vehicle emissions results in altered blood brain barrier permeability and expression of matrix metalloproteinases and tight junction proteins in mice. Part Fibre Toxicol 10, 62.

19.

Jayaraj

, Rodriguez

, Wang

, Block

(2017) Outdoor ambient air pollution and neurodegenerative diseases: The neuroinflammation hypothesis. Curr Environ Health Rep 4, 166–179.

20.

Calderón-Garcidueñas

, Reed

, Maronpot

, Henrí-quez-Roldán

, Delgado-Chavez

, Calderón-Garcidueñas

, Dragustinovis

, Franco-Lira

, Aragón-Flores

, Solt

, Altenburg

, Torres-Jardón

, Swenberg

(2004) Brain inflammation and Alzheimer’s-like pathology in individuals exposed to severe air pollution. Toxicol Pathol 32, 650–658.

21.

Calderon-Garciduenas

, Maronpot

, Torres-Jardon

, Henriquez-Roldan

, Schoonhoven

, Acuna-Ayala

, Villarreal-Calderon

, Nakamura

, Fernando

, Reed

, Azzarelli

, Swenberg

(2003) DNA damage in nasal and brain tissues of canines exposed to air pollutants is associated with evidence of chronic brain inflammation and neurodegeneration. Toxicol Pathol 31, 524–538.

22.

Calderon-Garciduenas

, Villarreal-Calderon

, Valencia-Salazar

, Henriquez-Roldan

, Gutierrez-Castrellon

, Torres-Jardon

, Osnaya-Brizuela

, Romero

, Torres-Jardon

, Solt

, Reed

(2008) Systemic inflammation, endothelial dysfunction, and activation in clinically healthy children exposed to air pollutants.. Inhal Toxicol 20, 499–506.

23.

Hüls

, Vierkötter

, Sugiri

, Abramson

, Ranft

, Krämer

, Schikowski

(2018) The role of air pollution and lung function in cognitive impairment. Eur Respir J 51, 1701963.

24.

Saito

, Ihara

(2016) Interaction between cerebrovascular disease and Alzheimer pathology. Curr Opin Psychiatry 29, 168–173.

25.

Toledo

, Arnold

, Raible

, Brettschneider

, Xie

, Grossman

, Monsell

, Kukull

, Trojanowski

(2013) Contribution of cerebrovascular disease in autopsy confirmed neurodegenerative disease cases in the National Alzheimer’s Coordinating Centre. Brain 136, 2697–2706.

26.

Gottesman

, Schneider

, Zhou

, Coresh

, Green

, Gupta

, Knopman

, Mintz

, Rahmim

, Sharrett

, Wagenknecht

, Wong

, Mosley

(2017) Association between midlife vascular risk factors and estimated brain amyloid deposition. JAMA 317, 1443–1450.

27.

Béjot

, Reis

, Giroud

, Feigin

(2018) A review of epidemiological research on stroke and dementia and exposure to air pollution. Int J Stroke 13, 687–695.

28.

Oudin

, Forsberg

, Adolfsson

, Lind

, Modig

, Nordin

, Adolfsson

, Nilsson

(2016) Traffic-related air pollution and dementia incidence in Northern Sweden: A Longitudinal Study. Environ Health Perspect 124, 306–312.

29.

Tzivian

, Dlugaj

, Winkler

, Weinmayr

, Hennig

, Fuks

, Vossoughi

, Schikowski

, Weimar

, Erbel

, Jöckel

, Moebus

, Hoffmann

(2016) Long-term air pollution and traffic noise exposures and mild cognitive impairment in older adults: A cross-sectional analysis of the Heinz Nixdorf Recall Study. Environ Health Perspect 124, 1361–1368.

30.

Ritz

, Lee

, Hansen

, Lassen

, Ketzel

, Sorensen

, Raaschou-Nielsen

(2016) Traffic-related air pollution and Parkinson’s disease in Denmark: A case-control study. Environ Health Perspect 124, 351–356.

31.

Kirrane

, Bowman

, Davis

, Hoppin

, Blair

, Chen

, Patel

, Sandler

, Tanner

, Vinikoor-Imler

, Ward

, Luben

, Kamel

(2015) Associations of ozone and PM2.5 concentrations with Parkinson’s disease among participants in the agricultural health study. J Occup Environ Med 57, 509–517.

32.

Chen

, Hung

, Chang

, Hsu

, Muo

, Tsai

, Wu

(2017) Long-term exposure to air pollution and the incidence of Parkinson’s disease: A nested case-control study.e. PLoS One 12, 0182834.

33.

, Li

, Martini

, Hou

(2019) Association between air pollution and risk of vascular dementia: A multipollutant analysis in Taiwan. Environ Int 133, 105233.

34.

Chen

, Kwong

, Copes

, Hystad

, van Donkelaar

, Tu

, Brook

, Goldberg

, Martin

, Murray

, Wilton

, Kopp

, Burnett

(2017) Exposure to ambient air pollution and the incidence of dementia: A population-based cohort study. Environ Int 108, 271–277.

35.

Kioumourtzoglou

, Schwartz

, Weisskopf

, Melly

, Wang

, Dominici

, Zanobetti

(2016) Long-term PM2.5 exposure and neurological hospital admissions in the Northeastern United States. Environ Health Perspect 124, 23–29.

36.

Palacios

, Fitzgerald

, Hart

, Weisskopf

, Schwarzschild

, Ascherio

, Laden

(2014) Particulate matter and risk of Parkinson disease in a large prospective study of women. Environ Health 13, 80.

37.

Palacios

, Fitzgerald

, Hart

, Weisskopf

, Schwarzschild

, Ascherio

, Laden

(2017) Air pollution and risk of Parkinson’s disease in a large prospective study of men. Environ Health Perspect 125, 087011.

38.

Lee

, Schwartz

, Wang

, Dominici

, Zanobetti

(2019) Long-term effect of fine particulate matter on hospitalization with dementia. Environ Pollut 254, 112926.

39.

Cacciottolo

, Wang

, Driscoll

, Woodward

, Saffari

, Reyes

, Serre

, Vizuete

, Sioutas

, Morgan

, Gatz

, Chui

, Shumaker

, Resnick

, Espeland

, Finch

, Chen

(2017) Particulate air pollutants, APOE alleles and their contributions to cognitive impairment in older women and to amyloidogenesis in experimental models.e. Transl Psychiatry 7, 1022.

40.

Oudin

, Andersson

, Sundström

, Nordin Adolfsson

, Oudin Åström

, Adolfsson

, Forsberg

, Nordin

(2019) Traffic-related air pollution as a risk factor for dementia: No clear modifying effects of APOEɛ4 in the Betula Cohort. J Alzheimers Dis 71, 733–740.

41.

, Lin

, Yu

, Chen

, Sun

, Wen

, Yip

, Chu

, Chen

(2015) Association between air pollutants and dementia risk in the elderly. Alzheimers Dement (Amst) 1, 220–228.

42.

Carey

, Anderson

, Atkinson

, Beevers

, Cook

, Strachan

, Dajnak

, Gulliver

, Kelly

(2018) Are noise and air pollution related to the incidence of dementia? A cohort study in London, England. BMJ Open 8, e022404.

43.

Cerza

, Renzi

, Gariazzo

, Davoli

, Michelozzi

, Forastiere

, Cesaroni

(2019) Long-term exposure to air pollution and hospitalization for dementia in the Rome longitudinal study. Environ Health 18, 72.

44.

Chang

, Chang

, Muo

, Wu

, Chen

, Kao

(2014) Increased risk of dementia in patients exposed to nitrogen dioxide and carbon monoxide: A population-based retrospective cohort study. PLoS One 9, e103078.

45.

Jung

, Lin

, Hwang

(2015) Ozone, particulate matter, and newly diagnosed Alzheimer’s disease: A population-based cohort study in Taiwan. J Alzheimers Dis 44, 573–584.

46.

Grande

, Ljungman

PLS

, Eneroth

, Bellander

, Rizzuto

(2020) Association between cardiovascular disease and long-term exposure to air pollution with the risk of dementia. JAMA Neurol 77, 801–809.

47.

Andersson

, Oudin

, Sundström

, Forsberg

, Adolfsson

, Nordin

(2018) Road traffic noise, air pollution, and risk of dementia - results from the Betula project. Environ Res 166, 334–339.

48.

Yuchi

, Sbihi

, Davies

, Tamburic

, Brauer

(2020) Road proximity, air pollution, noise, green space and neurologic disease incidence: A population-based cohort study. Environ Health 19, 8.

49.

World Health Organization. Dementia. http://www.who.int/mental_health/neurology/dementia/en/en.

50.

Peters

, Ee

, Peters

, Booth

, Mudway

, Anstey

(2019) Air pollution and dementia: A systematic review. J Alzheimers Dis 70, S145–s163.