Impacts of Climate Change and Air Pollution on Neurologic Health, Disease, and Practice

Abstract

Background and Objectives

Although the international community collectively seeks to reduce global temperature rise to less than 1.5°C before 2100, irreversible environmental changes have already occurred, and as the planet warms, these changes will continue to occur. As we witness the effects of a warming planet on human health, it is imperative that neurologists anticipate how the epidemiology and incidence of neurologic disease may change. In this review, we organized our analysis around 3 key themes related to climate change and neurologic health: extreme weather events and temperature fluctuations, emerging neuroinfectious diseases, and pollutant impacts. Across each of these themes, we appraised and reviewed recent literature relevant to neurologic disease and practice.

Methods

Studies were identified using search terms relating to climate change, pollutants, and neurologic disease in PubMed, OVID MEDLINE, EMBASE, PsycInfo, and gray literature. Studies published between 1990 and 2022 were included if they pertained to human incidence or prevalence of disease, were in English, and were relevant to neurologic disease.

Results

We identified a total of 364 articles, grouped into the 3 key themes of our study: extreme weather events and temperature fluctuations (38 studies), emerging neuroinfectious diseases (37 studies), and pollutant impacts (289 studies). The included studies highlighted the relationships between neurologic symptom exacerbation and temperature variability, tick-borne infections and warming climates, and airborne pollutants and cerebrovascular disease incidence and severity.

Discussion

Temperature extremes and variability both associated with stroke incidence and severity, migraine headaches, hospitalization in patients with dementia, and multiple sclerosis exacerbations. Exposure to airborne pollutants, especially PM2.5 and nitrates, associated with stroke incidence and severity, headaches, dementia risk, Parkinson disease, and MS exacerbation. Climate change has demonstrably expanded favorable conditions for zoonotic diseases beyond traditional borders and poses the risk of disease in new, susceptible populations. Articles were biased toward resource-rich regions, suggesting a discordance between where research occurs and where changes are most acute. As such, 3 key priorities emerged for further study: neuroinfectious disease risk mitigation, understanding the pathophysiology of airborne pollutants on the nervous system, and methods to improve delivery of neurologic care in the face of climate-related disruptions.

The effects of climate change on health are only beginning to be understood. In 2012, the Global Climate and Health Alliance drafted the Doha Declaration, a call to prioritize global policies to protect health as it is affected by climate change.¹ International calls for political advocacy and environmental justice surrounding the threat of climate change to human health have since followed.² In September 2021, in the face of increasing extreme weather events attributed to climate change, over 220 medical journals published a joint editorial calling for the “urgent action to keep average global temperature increases below 1.5°C, halt the destruction of nature, and protect health.”³

Climate impacts on human health are well-documented,^4-7 but the influence on patients with neurologic disease and how these impacts relate to risk of neurologic disease are less well-characterized. Factors associated with climate change that affect neurologic health and disease are incredibly wide-ranging, including effects not only related to temperature increases and ecosystem collapse, but also related to exposure to air pollution, food insecurity, changing patterns of infections, neurodevelopment, and mental health. These pleiotropic effects of a changing climate are unequally distributed and will disproportionately affect those in developing nations that have historically polluted less than wealthy nations.^8,9 Climate change is also inextricably linked with airborne pollutant emissions produced by the combustion of fossil fuels, and studies on the topic have brought attention to the effects not only on the respiratory and cardiovascular systems,^10-12 but also on neurologic disorders.^13-15

As the warming of our planet becomes increasingly apparent, there is an urgency to understand the impact of increasing temperatures on neurologic health to mitigate the effects on morbidity, mortality, and the burden on healthcare workers and health systems. Neurologists and neuroscientists have a duty not only to critically examine these potential changes, but also to quantify their impact to better prepare patients and healthcare systems. In this study, we present a scoping review of the literature pertaining to neurologic disease in adults associated with climate change and airborne pollutants and identify avenues for further research in this increasingly important field.

Methods/Literature Search Strategy

The key question we sought to answer through this scoping review was “what is known about the relationship between neurologic diseases (incidence, prevalence, morbidity, and mortality) and climate change?” Studies were identified by literature search of the PubMed, OVID MEDLINE, EMBASE, and PsycInfo databases for publications from January 1st 1990 to March 30th 2022 using the following search terms (alone or in combination with): “climate,” “climate change,” “global warming,” “air pollution/adverse effects,” “drought,” “flood,” “health systems,” “health policy,” “neurological disorders,” “neurology,” “meningitis,” “encephalitis,” “multiple sclerosis,” “dementia,” “stroke,” “seizures,” “epilepsy,” “headache,” “motor neuron disease,” “amyotrophic lateral sclerosis/ALS,” “Parkinson disease,” or “migraine.” Search terms relating to specific disease entities were chosen based on the Global Burden of Disease Study 2016 data, which identified stroke, Alzheimer disease and other dementias, meningitis/encephalitis, migraine/tension-type headache, epilepsy, Parkinson disease (PD), multiple sclerosis (MS), and motor neuron disease as the leading causes of disability-adjusted life years because of neurologic illness.¹⁶ Search strategy was developed in consultation with Loren Hackett, research librarian at the Cleveland Clinic. Our study was not registered with PROSPERO or any other review database.

Our focus was on adult neurologic diseases, and studies of neuropediatric conditions and outcomes were excluded. Gray literature was examined by a direct internet search for relevant articles, as well as Global Electronic Theses and Dissertations, Grey Matters, Government Documents Round Table, and Open Grey. No additional articles were identified by the addition of a general internet search or gray literature search. The identified articles were imported into the Covidence systematic review software where studies were deduplicated and screened.¹⁷ Study inclusion and exclusion criteria are provided in the Table. Of note, traumatic brain injury, spinal cord injury, and neurologic malignancies were excluded from our review. One study with human data included future modeling and was included. A PRISMA diagram depicting the studies included is presented in Figure 1. All included studies are listed in eAppendix 1 (links.lww.com/WNL/C482).

Table.

Inclusion and Exclusion Criteria for Included Studies

Figure 1. PRISMA Diagram Depicting the Selection Process for Studies Included in This Scoping Review.

S. Louis and A. Dhawan independently screened the remaining titles and study abstracts. Included studies examined the incidence, prevalence, morbidity, or mortality of a neurologic condition or the underlying neurobiology for a disease process, as it related to climate change or climate change–associated variables (e.g., meteorologic indices).

Studies were excluded if the publications were without English translation, had primarily animal subjects (apart from neuroinfectious diseases wherein studies relating to insect or animal reservoirs of pathogens were included), or if the study topic was not relevant to the above-listed neurologic conditions. Review articles, case reports, editorials, and opinion pieces were excluded from this scoping review. The remaining studies were included for full-text review, categorization into the themes of the manuscript, and data extraction. There was no assessment of study quality performed because this is a scoping review. Data extraction was performed in duplicate by S. Louis and A. Dhawan. Elements extracted from each manuscript included the general theme under which it was classified, study location, study population, intervention, comparator, outcome measured, study design, whether any multivariate or demographic adjustment was made, and result summary. For articles specifically related to exposures, the pollutants studied and how those pollutants were measured were also extracted. Any discrepancies were addressed by consensus after discussion. Manuscripts were grouped into one of 3 categories: the effects of extreme weather and temperature changes on neurologic conditions, emerging neuroinfectious diseases, and pollutant impacts. The authors intended to classify articles into categories before obtaining studies, but the categories themselves were not predefined and were chosen after literature search was completed. When classifying studies, all authors were able to reach consensus about which category each study fell into.

The map depicting neuroinfectious diseases was created using the rnaturalearth v0.1,¹⁸ rnaturalearthdata v0.1,¹⁹ and ggplot2²⁰ packages in R v4.1.1.²¹ A Sankey diagram was created using the networkD3 v0.4²² package in R v4.1.1.

Results

Our search revealed a robust body of literature describing the effects of climate change in relation to neurologic disease (Figure 1). The included publications consisted of 364 studies, with 149 studies from Asia, 105 studies from North America, 93 studies from Europe, 8 studies from Africa, 3 studies from Australia and Oceania, and 6 studies from South America. No relevant society guidelines, position statements, or governmental reports were identified.

Two independent reviewers (S.L. and A.D.) grouped manuscripts into one of 3 categories: the effects of extreme weather and temperature changes on neurologic conditions, emerging neuroinfectious diseases, and pollutant impacts. Relevant data including study population, intervention, comparators, and outcomes were extracted from all studies and synthesized for this scoping review. A subsection focusing on neuroinfectious diseases was created because of the high number of relevant articles identified. In each subsection, we grouped findings by disease and presented these in order of disability-adjusted life years attributable to each condition per the Global Burden of Disease report.

The Effects of Temperature Fluctuations on Neurologic Conditions

Stroke

Our search identified 24 studies relating climate to ischemic stroke incidence, with evidence favoring increased stroke risk at temperature extremes. There was no consensus on mechanism relating changes in climate to ischemic stroke events. Six studies showed an increase in the incidence of ischemic stroke events with increasing temperature and extremes of relative humidity.^23-28 These results were in contrast to 2 studies with increased stroke admissions at lower temperatures, perhaps because cold temperatures induce vasoconstriction and increase blood viscosity.^29,30 The longest term study came from Bai et al.,²⁹ who estimated the attributable fraction of stroke admissions in Ontario to cold temperature as 1.71%, with the burden attributable to moderate changes in temperatures as opposed to extreme changes. Bai et al. examined data over a 17-year period encompassing 1.4 million hospitalizations. Temperature variability may also contribute to stroke incidence, as Lei et al.³¹ examined over 140,000 first-time strokes in Shenzhen, China, and attributed 2%–4% of strokes to an increased temperature range (greater than 5.5–8°C in a 24-hour period, seasonally dependent), although temperature was only measured from a single location. The burden of disability-adjusted life years related to heat waves in a study by Yoon et al.³² in South Korea was driven by cerebrovascular diseases, estimated at 72.1% of the total burden of disease, although no adjustment for income level, age, sex, or other population levels was made in this study. Two studies using climate projections into the mid-late 21st century showed an increase in years of life lost due to stroke while accounting for population change, fertility, greenhouse gas emissions, and physical inactivity levels.^33,34

Dementia

Two studies examined dementia-related hospital admissions and meteorologic variables. The largest included 3,069,816 Medicare patients in New England over a 10-year period and estimated the association between hospital admissions for dementia and temperature variability. In the summer months, mean temperature increases of 1.5°C increased risk of admission by 12%, after adjustment for sex, race, age, socioeconomic status, zip code, and state of residence.³⁵ Two studies showed that patients with dementia are often at higher risk of injury or death because of extreme heat events such as bushfires, and a third suggested that heat and particulate matter exposure may increase the risk of dementia hospitalizations.^36-38

Headache

Two studies examined headache in association with meteorologic variables. A study of over 22,000 headache visits to the emergency department (ED) showed that an increase in temperature by 5°C was associated with a relative risk of headache presentation of 1.042 (95% CI 1.009–1.076), but did not specify International Classification of Headache Disorders diagnosis.³⁹ Mukamal et al.⁴⁰ examined headache incidence at a single center using a case-crossover design, examining temperature, barometric pressure, relative humidity, and pollutant levels in the 24–72 hours preceding presentation. In this study, higher temperatures and lower barometric pressure were associated with greater risk of presentation with any headache, especially nonmigraine headache.⁴⁰

Epilepsy

We identified 3 studies relating seizure frequency to meteorologic changes. Unstable weather, defined as a change in barometric pressure by 10 hectopascals and temperature change of 5°C in 48 hours, in a series of 30 patients with epilepsy was associated with a seasonal difference in seizure frequency. Unstable weather was associated with increased seizure frequency among 40% of participants in spring, autumn, and winter, but only 7% of participants in the summer.⁴¹ A single-center case-control crossover study from Germany involved 604 adult patients with epilepsy and found that low atmospheric pressure and high relative humidity increased seizure risk, and higher ambient temperatures associated with reduced seizure risk.⁴² Chiang et al.⁴³ examined outpatient and inpatient visits in patients with epilepsy over a 4-year period and showed an association with temperature and various air pollutants, with more visits during the colder winter months.

MS

Four studies of patients with MS met the inclusion criteria. It is well-known that high temperatures may exacerbate symptom severity in patients with MS. A time-stratified case-crossover study of ED presentations for MS disease exacerbation (1,265 patients) showed that exacerbations were associated with increased temperature variability on the preceding day. There was an 8.81% increase in presentations per 1°C increase in temperature range (95% CI 3.46%–14.44%).⁴⁴ Elser et al.⁴⁵ examined 106,225 individuals with MS over 15 years and found a positive association between anomalously warm weather and ED visits. A single-center analysis of 260 MS admissions from Serbia showed a reduction in relapses during the period of high vitamin D exposure and an increased number of relapses in spring compared with other seasons.⁴⁶ A randomized crossover study of 60 patients with MS undergoing physical therapy demonstrated differential responses in a warm climate (Spain) as opposed to a cold climate (Norway), with statistically significant changes favoring warmer climates and persisting up to 6 months post-treatment.⁴⁷

Emerging Neuroinfectious Diseases

37 studies were related to climate patterns and neuroinfectious diseases. Most studies on the topic of climate associations with neuroinfectious diseases were observational, retrospective, and used vector incidence or disease prevalence with historical geospatial climactic data to establish associations between weather and infection risk. Diseases included West Nile virus (WNV, 11 studies), meningococcal meningitis (6 studies), Japanese encephalitis virus (JEV, 5 studies), TBE (13 studies), unspecified viral meningitis (1 study), and coccidioidomycosis (1 study). There were no studies relating to neurohelminthic infections, although these are an important subset of neuroinfectious diseases affected by climate change. Figure 2 highlights the geographic skew of evidence toward Europe and the paucity of studies in Africa and South America. Several existing reviews also highlighted the impact of extreme weather, such as floods, on the transmission of mosquito-borne and rodent-borne diseases.

Figure 2. Neuroinfectious Diseases Studied by Location.

Map depicts the number and types of reports examining neuroinfectious diseases stratified by location. 10 reports were published from North America, 14 reports from Europe, 5 reports from Africa, 8 reports from Asia, and 1 report from Australia. Abbreviations: JEV = Japanese encephalitis virus; TBE = tick-borne encephalitis.

The ecologic interactions leading to the emergence of neuroinfectious diseases are complex, and a wide range of factors including human population density, host-human interactions, land use patterns, pollutants, and disease adaptability all affect disease incidence.^48-51 For instance, in examining why JEV incidence increased in the Himalayas, Baylis et al.⁵² noted that incidence may be explained by improved detection and inhomogeneous vaccine distribution, an increased incidence of JEV due to climate change, or a combination of these factors. In addition, regional factors beyond temperature alone may facilitate disease transmission. In the African ‘meningitis belt,’ aerosolized matter blown by the Harmattan winds increases meningitis transmission, whereas in the Czech Republic, flooding associates with the incidence of TBE.^52-55 Similarly, reasons for the increase of TBE in Europe remain controversial, with some arguing that socioeconomic factors such as the fall of the Soviet Union caused changes in human behavior leading to increased exposure and others arguing that climactic change better accounts for the increased incidence.^56,57

Predictive modeling of WNV incidence in North America with simulations from 2021 to 2080 suggest an expansion of suitable conditions for WNV in the southern United States because of higher temperatures, lower rainfall, a lengthening of mosquito season, and an increase in droughts.^58-60 Transmission periods for TBE in Europe are predicted to lengthen and new foci in Scandinavia may emerge, as suggested by analyses of epidemiologic data between 1969 and 2018 from Sweden, Germany, Czechia, Austria, Slovenia, and Italy.^e1,e2

Pollutant Impacts

Studies relating pollutant impacts to neurologic disease were subgrouped by disease: MS, headaches, ischemic stroke and TIA, intracerebral hemorrhage (ICH), amyotrophic lateral sclerosis (ALS), PD, and dementia. Pollutants have also been shown to increase mortality, with one study finding that periods of high air pollution due to heat waves and fires tripled the relative risk of death due to neurologic disease.^e3 Figure 3 demonstrates the numerous inter-relationships between pollutant exposures and neurologic disease.

Figure 3. Relationships Between Pollutants and Identified Neurologic Links.

Sankey diagram depicting inter-relationships between pollutant exposure (left) and neurologic outcome (right). PET refers to positron emission tomography. TIA refers to transient ischemic attack. Abbreviations: MS = multiple sclerosis; PD = Parkinson disease.

Ischemic Stroke/Transient Ischemic Attack

Long-term and short-term exposures to airborne pollutants had substantial support for association with ischemic stroke incidence and mortality. Of the 166 relevant studies we identified, 92 were from Asia, and 60 of these were from China. There were 28 studies from North America, 41 studies from Europe, 1 study from Australia, 3 studies from South America, and 1 study from Africa. Studies examined a wide range of pollutants including particulate matter less than 1 μm diameter (PM1, 2 studies), particulate matter less than 2.5 μm diameter (PM2.5, 92 studies), particulate matter less than 10 μm diameter (PM10, 80 studies), particulate matter 2.5–10 μm in diameter (PM(coarse), 4 studies), carbon monoxide (CO, 37 studies), sulfur dioxide (SO₂, 61 studies), ozone (O₃, 65 studies), and nitrous oxides (NO₂/NOx, 93 studies).

An analysis of the Global Burden of Disease study concluded that 9% of stroke disability-adjusted life years and 8.5% of stroke deaths could be attributed to PM2.5 exposure.^e4 After adjusting for sociodemographic factors, a study of 3,287 participants with incident stroke living within 100 m of a major roadway (associated with increased airborne pollutant exposure) was associated with a hazard ratio of 1.42 (95% CI 1.01–2.02) for ischemic stroke, although imaging-based biomarkers of small vessel disease in a subset of these patients did not show significant association.^e5,e6 A Chinese study of 12,291 ischemic strokes revealed a statistically significant association with PM1 and PM2.5 exposures during the 3 years before the stroke event, but not with NO2 or PM10 exposure.^e7 Long-term exposures in a German study showed elevated risk of ischemic stroke, regardless of particle size, examining PM2.5 abs, PM2.5, and PM10.^e8 Long-term PM2.5 exposure examined in a study of 39,054 participants in China demonstrated an association with increased stroke mortality (HR 1.3, 95% CI 1.04–1.65).^e9

Studies examining short-term pollutant exposure and stroke risk generally used a case-crossover design. Results were conflicting depending on the population studied and the geographic precision with which pollutants were examined. Stroke mortality may be higher on days with greater PM1, PM2.5, PM (coarse), and PM10 exposures.^e10-e12 Short-term PM2.5 exposure was more controversial in its association with the incidence of stroke and small vessel disease. We identified 3 clinical studies and one imaging-based study that did not reveal any association^e13-e15 and 3 clinical studies and one imaging-based study did show an association between stroke, small vessel disease incidence, and PM2.5 exposure.^e16-e19 Similar studies examining first-time strokes and ozone exposure have shown mixed associations, with a single study showing a positive association^e20 and 4 studies without association.^{30,e14,e20-22} NO2 exposure may be associated with a greater incidence of stroke-related admissions and stroke mortality in the short term,^e23,e24 although 2 additional studies did not find this.^30,e25 Shen et al. examined short-term SO2 exposure in relation to the rates of ischemic and hemorrhagic strokes and found a positive association,^e26 but 2 other studies did not find any association^30,e25 Individual factors, such as ethnic or genetic factors or degree of exposure, may contribute to the variability in these associations.^e14,e18,e21

Five studies examining TIAs were identified. A variety of air pollutants (PM2.5, PM10, CO, NO2, O3, and SO2) were examined, but studies showed conflicting associations, which, in some cases, were also modified by daily temperature.

Intracerebral Hemorrhage

Forty-one studies examined risk of ICH attributable to environmental pollutants. In one study of 368 cases of small vessel-attributed ICH, short-term exposure to increased PM2.5 was greater on the 3 days preceding the hemorrhage compared with a control period 15–17 days before the index event.^e27 A second study of 517 patients did not identify any association between PM2.5, black carbon, or nitrogen dioxide and ICH and used a similar design comparing exposures 1–7 days before the hemorrhage to a referent set of days in the same month.^e28 PM2.5 and PM10 were not shown to be associated with hemorrhagic stroke in a large Chinese study involving 69,399 participants.^e16

Dementia

We identified 51 studies relating dementia to airborne pollutants (27 from North America, 11 from Europe, and 13 from Asia). Studies examined PM2.5 (41 studies), PM (coarse) (3 studies), PM10 (18 studies), NO2 (24 studies), SO2 (5 studies), and O3 (14 studies). 6.1% of cases of incident dementia are estimated as attributable to PM2.5 and NO2 exposures.^e29 Clinical diagnosis of dementia^e29,e30 and measures of cognitive decline such as the MMSE^e31,e32, semantic fluency, and word recall^e33 associated with prior years' exposure to PM2.5. Amyloid PET scans were more likely to be positive if the participant lived in an area with greater PM2.5 in the preceding 13–15 years.^e34 In patients with preexisting Alzheimer disease, PM2.5 associated with risk of hospitalization.^e35 Healthy patients with serial MRI scans have greater white matter loss in those with higher PM2.5 exposure.^e36-e40 PM(coarse) is less well-studied, and only a single report linking exposure in women to a decreased Mini-Mental Status Examination (MMSE) score has been published.^e41 Exposure to PM10, particularly in patients harboring the APOE4 allele, associated with dementia incidence, increased rate of cognitive decline, and frontotemporal thinning on MRI.^{30,e5,e32,e40,e42} The effects of exposure to ground-level ozone remain less clear because there have not been as strong associations with dementia incidence^e42, odds of amyloid PET positivity^e34, or MMSE score decline.^e32 Nitrogenous gas exposure associated with vascular dementia^e43, global brain atrophy^e40, and a faster rate of cognitive decline in those with the APOE4 allele, with differing results for MMSE-scored cognitive decline and all-cause dementia.^{e29,e30,e32,e41} Sulfur dioxide (SO2) exposure also associated with MMSE score decline.^30,e32 Ten-year exposure to black carbon did not show any association with all-cause dementia.^e30

Headaches

Fourteen studies examined headaches and pollutant exposures. The largest used a case-control design to examine 89,575 cases of migraine in California, finding an association between higher frequency of migraine-specific urgent care visits and elevated average annual PM2.5 and NO2 levels.^e44 Two smaller studies, each with at least 18,000 patients, examined short-term pollutant exposures and visits for migraine with statistically significant effects for PM2.5, PM10, NO2, O3, and CO exposures in adults.^e45,e46 One study examined 143 patients with tension-type headaches and showed that O3 exposure was greater on days when these patients presented to hospital with an exacerbation, although this finding has not been replicated.^e47

PD

PD incidence as it relates to particulate matter exposure was examined by 19 studies (4 from Asia, 1 from Australia, 6 from Europe, and 8 from North America). In addition to PM2.5 (15 studies), PM10 (10 studies), PM (coarse) (4 studies), NO2 (8 studies), SO2 (1 study), CO (2 studies), and O3 (4 studies), 3 studies examined airborne metals, with one specifically examining airborne manganese. A French study of 111,378 patients with PD examined airborne copper and showed that there was a slightly increased risk of developing PD in regions with airborne copper in the highest quintile.^e48 Studies generally showed varying associations with each of the airborne pollutants examined. Among the largest studies was a case-control study involving 38,475 patients with PD from Ontario where 2-year exposures for PM2.5, NO2, and O3, at a resolution of 1 × 1 km, were shown to be associated with PD incidence (HR ranged from 1.03 to 1.04).^e49 NO2 exposure was examined in 7 additional studies, and its impact remains controversial. Increased concentrations of NO2 are noted in PD hotspots.^e50 Two studies (combined 11,525 patients) suggested an association between NO2 concentration and PD incidence^e51,e52, although these were in opposition to 2 smaller studies (combined 1,496 patients) that did not show any association between NO2 concentration and PD incidence.^e53,e54

ALS

Three studies related ALS incidence to air pollutant exposure, with varying results. One study involving 917 patients with ALS from a Dutch national registry showed an increased risk of ALS in patients with long-term exposure to PM2.5 and NO2, but another series of 52 patients that examined PM10 exposure did not replicate these findings.^e55,e56 The risk of disease aggravation in association with short-term exposure to PM2.5 for patients with ALS and PD is less controversial.^e57

MS

Nineteen studies examined the association between MS incidence and pollutant exposures, of which 5 were from Asia, 11 were from Europe, and 3 were from North America. Pollutants studied included PM2.5 (7 studies), PM10 (8 studies), PM (coarse) (1 study), SO2 (2 studies), NO2 (6 studies), O3 (3 studies), and benzene (2 studies). Studies examined the short-term and long-term risks of exposure to airborne pollutants on MS incidence and exacerbation-related hospitalization.^e58,e59 Generally, airborne pollutants were associated with exacerbations of disease activity with short-term exposure.^e60-e62 However, definitive associations with long-term airborne pollutant exposure and incidence of MS were not well-established, and many studies did not correct for known risk factors, such as vitamin D exposure or family history.

Discussion

Our scoping review has identified a wide body of literature pertaining to the impacts of climate change on neurologic disease. The results highlight our current knowledge on the relationship between the incidence and severity of neurologic disease and the effects of climate change and pollution.

Studies of outdoor airborne pollutant exposure were the predominant literature identified by our review. The majority pertained to pollutant exposure and stroke or TIA incidence and implicated particulate matter, especially PM2.5 and NO2, in increasing risk and severity of ischemic and hemorrhagic strokes. Airborne pollutants also had strong epidemiologic associations with a broad range of neurologic diseases, including dementia, headache, PD, and MS. Ambient temperature and temperature variability also was associated with stroke incidence, and it is estimated that up to 4% of stroke risk may be attributable to temperature and its variability alone.^29,31 Headaches and MS exacerbations requiring hospital admission increased with greater diurnal temperature variability and temperature, respectively. In patients with dementia, higher temperatures increased the risk of hospitalization. Taken together, these findings suggest that the physiologic changes occurring due to extremes of temperature, high temperature variability, and specific airborne pollutants have broad consequences for the nervous system through several potential mechanisms including a phenomenon of accelerated aging.

Studies relating to neuroinfectious diseases documented the wider geographic range of tick-borne and mosquito-borne illnesses over the past century were concentrated in Northern Europe and North America. These findings remain controversial because of potential confounding by the increased diagnosis and changing human behavior and land use patterns. Despite this, these studies clearly outline the altered geographic extent of hosts, reservoirs, and vectors of zoonotic diseases due to a changing climate, increasing disease risk for previously unaffected populations.

Three key areas emerged for expedited study owing to their potential for rapid change, large-scale impact on neurologic health, and potential for risk mitigation. Foremost, there is a priority to develop an understanding of emergent neurotropic infectious diseases. Diseases such as Zika virus, WNV, and TBE have both the potential and precedence for rapid spread across susceptible populations with only a rudimentary understanding of their long-term effects on the nervous system. Clarifying how climate and human activity affects the incidence of zoonotic disease is crucial to informing risk mitigation strategies, such as vaccination programs, surveillance systems, and ecological defenses (e.g., Wolbachia bacteria against Zika virus).^e63,e64 Improving surveillance through international organizations such as the European Center for Disease Control will also be key to understanding and mitigating risk of these diseases. Data produced by large-scale disease surveillance projects will allow for strategic and pre-emptive planning, such as the distribution of vaccinations to newly at-risk populations. Neurologists, informed by these data, can advocate for vaccination programs to mitigate risk among susceptible populations.

Second, there is an opportunity to enhance the understanding of how neurologic disease mechanistically relates to airborne pollutant exposure. Our review has highlighted how specific components of air pollution, such as particulate matter, preferentially associate with increased stroke and neurodegenerative disease risk. This may represent a shared pathophysiology, such as small vessel disease. Dense population centers in developing nations are the regions with the highest concentration of airborne pollutants and, therefore, put a growing young population at risk of these potentially preventable chronic neurologic diseases. In recognizing this risk, the World Health Organization (WHO) has initiated monitoring and interventions to improve health outcomes relating to airborne pollutants. Notably, the WHO interventions also encompass indoor air pollutants, not discussed in our review, which are an additional key source of airborne pollutants affecting health.^e65-e68 Third, there is an unmet need in planning for the robust delivery of neurologic care in the face of ecologic instability with natural disasters limiting access to health services. Teleneurology can serve as a solution to the disruptions in neurologic care occurring because of climate change. For patients living in areas prone to climate change–related weather events, a single extreme weather event could limit physical access to health care for weeks to months at a time. Telemedicine could deliver care to patients in areas made geographically inaccessible because of extreme weather events. In the short term after a natural disaster, there may be the need for specialty neurologic consultation, and telemedicine could facilitate delivery of this care at temporary clinics. The COVID-19 pandemic has highlighted how telemedicine can help patients obtain healthcare access but could not otherwise physically access it and how this has the added benefit of dampening air pollution through reduction in transport to receive care.^e69-e71

Our study has several limitations. We were limited to studies published in English or with suitable English translation. A risk of bias assessment for this scoping review was not performed given the retrospective, observational nature of nearly all studies. Studies often relied on multivariate modeling and were, therefore, biased by choice and measurement of the variables included, and differing modeling approaches did not allow for direct comparison of the results. There were also challenges unique to meteorologic literature, including limitations on spatial resolution of climate data, missing historical data, instrument-dependent effects, and limited data on true individual exposures. Examining climactic variables' effects on health outcomes often involves the selection of a lag time between exposure and outcome, which can be arbitrary, and we observed variability in lag times for the included studies. Studies were geographically biased toward high-income countries, whereas climate change disproportionately affects those in developing nations. For example, our search strategy did not identify any studies on infectious diseases from South America and only 5 studies from Africa. We attempted to correct for this by searching the gray literature, but there remains a high risk of bias because climate studies generally rely on data obtained in resource-rich settings. Finally, studies relating to headache often did not distinguish between types of headaches (e.g., tension headache vs migraine headache) or use the International Classification of Headache Disorders. Moreover, these studies only included patients experiencing headache severe enough to present to their healthcare provider, thereby excluding patients with mild-moderate headache syndromes.

Climate change poses many challenges for humanity, some of which are known but not well-studied and others that have yet to declare themselves. For instance, while our review did not identify any articles related to effects on neurologic health from food and water insecurity, these are clearly linked to neurologic health and climate change. Akin to social determinants of health, a changing climate and environmental pollutants cannot be overlooked as mediators of disease burden. An elderly patient with dysautonomia or MS and lacking access to air conditioning, a patient with a small-vessel stroke in a city with increasing air pollution, or a patient with TBE due to a vector not previously seen in their region are all vignettes that will undoubtedly become more common. Their neurologic conditions are directly caused or exacerbated by the effects of human-induced climate change, and we must acknowledge the burden of illness on those who bear little responsibility for pollutant and greenhouse gas emissions. In doing so, our goal is to inspire action through a collective voice among neurologists and healthcare workers to elicit true change in mitigating the effects of the climate crisis.

Glossary

ALS

amyotrophic lateral sclerosis

ED

emergency department

ICH

intracerebral hemorrhage

JEV

Japanese encephalitis virus

MS

multiple sclerosis

PD

Parkinson disease

TBE

tick-borne encephalitis

WHO

World Health Organization

Appendix. Authors

Study Funding

The authors report no targeted funding.

Disclosure

The authors report no disclosures relevant to the manuscript. Go to Neurology.org/N for full disclosures.