Air Pollution and the Risk and Progression of Multiple Sclerosis: A Systematic Review and Meta‐Analysis

ABSTRACT

Purpose

Air pollution has been linked to several neurological conditions, including stroke and neurodegenerative diseases. Evidence regarding its association with multiple sclerosis (MS) remains conflicting, limited by small sample sizes.

Methods

PubMed, Embase, Scopus, and Cochrane controlled register of trials (CENTRAL) were searched on September 9, 2025 without any language or time restrictions. The inclusion criteria were observational studies that evaluated the association between exposure to air pollutants and MS development or severity. Hazard ratios (HRs) with 95% confidence intervals (CIs) were pooled as effect estimates using the fixed or random effects model. Exposures included PM2.5, PM10, nitrogen dioxide (NO₂), carbon monoxide (CO), sulfur dioxide (SO₂), and ozone (O₃). Outcomes were MS risk and severity, including relapses, disability progression measured with the Expanded Disability Status Scale (EDSS), and contrast‐enhancing lesions (CELs) development.

Results

Twenty‐two studies comprising 16,585,206 participants were included. Long‐term exposure to PM2.5 (HR = 1.21; 95% CI: 1.07–1.38), PM10 (HR = 1.20; 95% CI: 1.02–1.42), and CO (HR = 3.85; 95% CI: 1.34–11.08) were associated with increased MS risk. Short‐term exposure to PM2.5 (HR = 1.20; 95% CI: 1.01–1.43), PM10 (HR = 1.20; 95% CI: 1.01–1.45), NO₂ (HR = 1.13; 95% CI: 1.02–1.25), and O₃ (HR = 1.15; 95% CI: 1.04–1.27) were associated with relapses. Short‐term exposure to PM2.5 (HR = 2.20; 95% CI: 1.05–4.60) and PM10 (HR = 1.02; 95% CI: 1.01–1.03) were linked to CELs, while PM10 (HR = 1.31; 95% CI: 1.04–1.65) was associated with disability progression.

Interpretation

Long‐term air pollution exposure was associated with higher MS risk, and short‐term exposure with greater disease severity. Reducing air pollution may be a key strategy to protect brain health in MS.

1. Introduction

Multiple sclerosis (MS) is a chronic neuroinflammatory disease of the central nervous system (CNS) that results in demyelination and neuronal injury [1]. It was estimated that around 2.8 million people live with MS worldwide [2]. It is considered one of the most common causes of nontraumatic disability among young adults [3]. This disability has substantial impact on social and economic levels of the individual and community [3]. For example, the annual economic burden of MS in the Unites states (US) was estimated at 85.4 billion US dollars [3].

MS classification now includes several phenotypes including relapsing–remitting MS, which is characterized by alternating periods of acute neurological deficits interspersed by periods of clinical stability, secondary progressive MS, which is characterized by periods of worsening disability without any clinical or radiological evidence of new acute inflammation, although superimposed clinical relapses and new focal MRI activity can still occur, and primary progressive MS which is characterized by slow disability progression from the onset of the disease [1].

The disability in MS regardless of the clinical phenotype is mediated through two main mechanisms [4]. The first mechanism is relapses associated with worsening disability while the second is progression independent of relapse activity (PIRA) [4]. Relapse rate (RR) is associated with disability progression and hypothetically relapses are associated with T₂‐lesions in MS [5]. T₂‐lesion formation starts with gadolinium/contrast enhancing lesions (CELs) [6]. CELs are also associated with disease progression as indicated by expanded disability status scale (EDSS) [7]. PIRA in contrast is mainly driven by neurodegeneration processes such as brain atrophy and chronic active lesions in MS [1].

Similar to other neuroinflammatory diseases, the cause of MS is unknown but epidemiological studies demonstrated that environmental, genetic and lifestyle factors such as smoking and obesity overlap [8]. For example, a previous systematic review included 132 articles and 109,626 participants with MS and 16,724,390 controls found an association between Epstein bier virus, herpes virus type 6, and varicella‐zoster virus, and higher odds of occurrence MS. Smoking was also associated with higher likelihood of MS occurrence [8].

Previous studies that investigated the association between inflammatory non‐neurological and neurological diseases and air pollution demonstrated a significant association between the two. For example, in non‐neurological diseases, eye conditions were associated with long‐ and short‐term exposure to high concentrations of particulate matter (PM). Moreover, neurodegenerative diseases such as dementia [9] and Parkinson's disease [10] were also associated with long‐term exposure to higher concentrations of PM and nitrogen dioxide (NO₂). In addition, strokes were also associated with short as well as long‐term exposure to higher concentrations of PMs, NO₂ and sulfur dioxide (SO₂) [11, 12]. However, the evidence correlating air pollution and MS is less certain. Due to the important role of neurodegeneration and brain vascularity in MS pathogenesis, it is reasonable to think that air pollution is implicated as a risk factor of developing MS and its progression. While several reviews hypothesized neuroinflammatory mechanisms correlating between the two [13, 14], previous meta‐analyses were limited by sample size, only evaluated exposure to particulate matter, did not differentiate between short‐ and long‐term exposure to air pollution, and did not evaluate the impact on MS progression.

Motivated by these premises, we conducted a systematic review and meta‐analysis aiming to evaluate the association between short‐ and long‐term exposure to air pollution and risk of developing MS and its progression. We hypothesize that long‐term exposure to air pollution is associated with the risk of developing MS while short‐term exposure is associated with MS progression and severity.

2. Methods

2.1. Registration and Protocol

This systematic review and meta‐analysis was conducted in accordance with the preferred reporting items for systematic reviews and meta‐analyses (PRISMA). The study protocol was registered in the international prospective register of systematic reviews (PROSPERO) (CRD420251142351).

2.2. Search Strategy

PubMed, Embase, Scopus, and Cochrane controlled register of trials (CENTRAL) were searched on September 9, 2025 using the following keywords (((air pollution[Title/Abstract]) OR (particulate matter[Title/Abstract]) OR (carbon monoxide[Title/Abstract]) OR (Air Pollutants[Title/Abstract]) OR (Sulfur Dioxide[Title/Abstract]) OR (nitrogen dioxide[Title/Abstract]) OR (Nitrogen Dioxide[Title/Abstract]) OR (ozone[Title/Abstract])) AND ((multiple sclerosis[Title/Abstract]))). The search was done without any language or time restrictions.

2.3. Exposures and Outcomes of Interest

The exposure of interest was long‐ and short‐term exposure to air pollution including particulate matter (PM) including PM2.5 and PM10, carbon monoxide (CO), sulfur dioxide (SO₂), nitrogen dioxide (NO₂), or ozone (O₃). Short‐term exposure was defined as exposure to air pollution for ≤ 7 days while long‐term exposure was defined as exposure to air pollution for > 7 days. The primary outcome of interest was the development of MS diagnosed using the McDonald diagnostic criteria without specification on the version as it was expected that different studies were conducted at different time points and they would use different versions of the criteria. The secondary outcomes of interest were the severity of MS which included (1) relapse risk defined as a new subjective or objective neurological symptom or worsening of an old symptom typical of MS lasting at least 24 h, in the absence of infection, fever, or other causes, with at least 30 days since the onset of a previous relapse (2) disability progression defined as an increase in expanded disability scale (EDSS) score of ≥ 1.0 point increase if baseline EDSS ≤ 5.5 or ≥ 0.5 point increase if baseline EDSS ≥ 6.0 sustained for a specified period (12 or 24 weeks), and (3) CELs development defined as lesions showing gadolinium enhancement on contrasted T₁‐weighted MRI sequences.

2.4. Study Selection

The inclusion criteria were case–control or cohort studies that evaluated the association between long‐term exposure to air pollutants and MS development or short‐term exposure to air pollutants and MS severity. The exclusion criteria were reviews, editorials, case reports, studies that evaluated the association between weather changes and MS, and studies that evaluated the association between air pollution and other neurological diseases.

The study selection process was manually done on Rayyan, a web software that is used to facilitate study selection for systematic reviews. The study selection was performed in two steps: reviewing the title and abstract which was conducted between September 10, 2025, and September 24, 2025 followed by reviewing the full‐text of the papers which was done between September 24, 2025 and September 30, 2025. The process was done by AAT and TNA independently and any discrepancies were discussed until a consensus was reached.

2.5. Data Extraction

The data extraction was done between October 1, 2025, and October 11, 2025 using a preprepared sheet that included demographic data of the article including author name, publication year, study design, mean age of the included participants, male to female ratio, sample size, and data related to the exposures and outcomes of interest.

The data extraction was done by AAT and TNA independently and any discrepancies were discussed until a consensus was reached.

2.6. Quality of Evidence Assessment

The quality assessment was done using new‐castle Ottawa scale (NOS). NOS is a quality assessment tool for observational studies that is composed of three criteria: (1) the selection of the study groups; (2) the comparability of the groups; and (3) the ascertainment of either the exposure or outcome of interest for case–control or cohort studies respectively. The total score of NOS is 9, with studies rated < 5 considered of poor quality, 5–6 considered of intermediate quality, and ≥ 7 considered of good quality. The quality assessment was done by AAT.

2.7. Data Analysis

The data analysis was done using MetaXL (EpiGear Int Pty Ltd.) version 5.3 software. The effect measures used to investigate the association between the exposures and outcomes of interest were hazard ratios (HR) and their related 95% confidence intervals (95% CIs). The conversions between effect measures were done if needed according to previously published guidelines [15]. The random effect model was used if the heterogeneity was < 50% among the included studies while the fixed effect model was used if the heterogeneity was > 50%. The heterogeneity was quantified using I ² statistics and its significance was tested using Cochrane Q statistic. Publication bias was assessed using funnel plot among the included studies.

3. Results

3.1. Search Results

The search retrieved a total of 1053 articles; of them, 456 were duplicates. The remaining 597 articles were reviewed using their titles and abstracts, and 543 articles were excluded because they were irrelevant, editorials, or reviews. The last 54 articles were screened using their full‐text form, and 32 articles were excluded because they did not include data about the exposures or outcomes of interests. Finally, 22 articles were included in the analysis. Figure 1 demonstrates the process of study selection. Supporting Information cites the included papers in the Supplementary References.

FIGURE 1.

PRISMA flow chart. The chart demonstrates the process of study selection and the reasons for exclusion in each stage of screening. Consider, if feasible to do so, reporting the number of records identified from each database or register searched (rather than the total number across all databases/registers). If automation tools were used, indicate how many records were excluded by a human and how many were excluded by automation tools. From: Page et al. [16]. For more information, visit: http://www.prisma‐statement.org/.

3.2. Characteristics of the Included Studies

A total of 16,585,206 participants from the 22 articles were included in the analysis. Of these papers, 10 articles evaluated the impact of long‐term exposure to air pollution on the risk of MS development which included 16,566,694 participants with nine articles evaluated PM2.5, three evaluated PM10, one evaluated CO, one evaluated SO₂, one evaluated O₃ and four evaluated NO₂. The mean age of participants in these papers was 30.6 ± 7.6 and the male to female ratio was 0.59. On the other hand, 12 articles evaluated the impact of short‐term exposure to air pollution on the severity of the disease, and they included a total of 18,512 PwMS. The mean age of the PwMS in these papers was 39 ± 11 and the male to female ratio was 0.39. Of these papers, seven articles evaluated PM10, six evaluated NO₂, five evaluated PM2.5, three evaluated O₃, two evaluated CO, and two evaluated SO₂. Table 1 demonstrates the characteristics of the included studies.

TABLE 1.

Characteristics of the included studies.

Last author, year	Country	Design	Mean age and (standard deviation)	Male to female ratio	Sample size	NOS score
Carmona et al., 2018	Spain	Retrospective cohort	—	—	2224	5
Vazifehdan et al., 2024	Italy	Retrospective cohort	41.5 (11.5)	0.51	383	4
Faustini et al., 2018	Italy	Retrospective cohort	—	0.51	3491	5
Jeanjean et al., 2018	France	Retrospective cohort	30.5 (10)	0.34	424	6
Laura et al., 2017	Italy	Retrospective cohort	—	0.45	8287	6
Yuchi et al., 2020	Canada	Retrospective cohort	—	—	7232	6
Lavery et al., 2018	US	Case–control	14.7 (3.4)	0.59	387	6
Elgabsi et al., 2021	Israel	Retrospective cohort	52.7 (16.7)	0.52	287	6
Bai et al., 2018	Canada	Retrospective cohort	30.8 (6.1)	0.43	2,824,478	9
Oikonen et al., 2003	Finland	Retrospective cohort	—	—	406	6
Teekaput et al., 2025	Thailand	Retrospective cohort	44.9 (14.9)	0.1	153	6
Roux et al., 2017	France	Retrospective cohort	30.7 (10.1)	0.37	536	6
Bergamaschi et al., 2020	Italy	Retrospective cohort	—	—	547,251	5
Bergamaschi et al., 2020	Italy	Retrospective cohort	37 (11)	0.41	52	5
Palacios et al., 2017	USA	Retrospective cohort	—	—	121,700	6
Mehrpour et al., 2013	Iran	Retrospective cohort	34.4 (9.5)	0.3	160	6
Gregory et al., 2008	Georgia	Retrospective cohort	—	—	9,072,576	6
Mirmosayyeb et al., 2025	Iran	Retrospective cohort	42 (10)	0.29	1,996,443	6
Hedstrom et al., 2023	Sweeden	Retrospective cohort	35 (11)	0.45	123,000	8
Januel et al., 2021	France	Retrospective cohort	40 (4)	0.41	2109	5
Tateo et al., 2018	Italy	Retrospective cohort	—	0.89	936,887	8
Scartezzini et al., 2020	Italy	Retrospective cohort	—	0.89	936,740	6

Abbreviations: NOS, New‐castle Ottawa Scale; US, United States.

3.3. Air Pollution and Risk of MS

Nine papers evaluated the association between long‐term exposure to PM2.5 and risk of developing MS. The model that pooled these studies demonstrated that PM2.5 was associated with significantly higher risk of developing MS (Figure 2a: HR = 1.21; 95% CI: 1.07–1.38); the model had significant heterogeneity (I ² = 95%, p < 0.001).

FIGURE 2.

The association between air pollution and risk of developing multiple sclerosis. The figure presents the association between long‐term exposure to PM2.5 (a) and PM10 (b) and risk of developing MS. The analysis is presented using hazard ratios (HR) and their 95% confidence intervals (95% CIs).

Three papers evaluated the impact of PM10 on the risk of developing MS. The model showed a significant association between PM10 and risk of MS (Figure 2b: HR = 1.20; 95% CI: 1.02–1.42); the model had insignificant heterogeneity (I ² = 40%, p = 0.190).

Only one paper each evaluated the association between CO [17], SO₂ [17], and O₃ [18] and risk of MS. The papers found no significant association between the exposures and the outcome (data not shown).

Four papers evaluated the association between NO₂ and the risk of developing MS (data not shown: HR = 1.07; 95% CI: 0.82–1.38); the model had significant heterogeneity (I ² = 86%, p < 0.001).

3.4. Air Pollution and Risk of MS Relapses

Four papers evaluated the association between short‐term exposure to PM2.5 and MS relapses. The model that pooled these papers found a significant association between the two variables (Figure 3a: HR = 1.20; 95% CI: 1.01–1.43); the model had significant heterogeneity (I ² = 92%, p < 0.001).

FIGURE 3.

The association between air pollution and multiple sclerosis progression. The figure presents the association between short‐term exposure to particulate matter (PM)2.5 (a), PM10 (b), and nitric oxide (NO₂) (c) and risk of relapses in multiple sclerosis (MS). In (d), we present the results of the association between short‐term exposure to PM10 and disability progression. The analysis is presented using hazard ratios (HR) and its 95% confidence intervals (95% CIs).

The model that evaluated the association between PM10 and MS relapses included seven papers and demonstrated significant association between the exposure and MS relapses (Figure 3b: HR = 1.20; 95% CI: 1.00–1.45); the model had significant heterogeneity (I ² = 98%, p < 0.001).

Five papers evaluated the association between NO₂ and MS relapses. The model that pooled these studies found a significant association between NO₂ and MS relapses (Figure 3c: HR = 1.13; 95% CI: 1.02–1.25); the model had significant heterogeneity (I ² = 92%, p < 0.001).

Two papers evaluated the association between the CO and MS relapses. The model that pooled these papers did not find an association between the two variables (data not shown: HR = 1.69; 95% CI: 0.49–5.87); the model had significant heterogeneity (I ² = 81%, p = 0.020).

One paper assessed the association between SO₂ [19] and MS relapses. The paper did not find an association between the two variables (data not shown).

The model that assessed the association between O₃ and MS relapses included three papers and found a significant association between the two variables (Figure 3d: HR = 1.15; 95% CI: 1.04–1.27); the model had significant heterogeneity (I ² = 74%, p = 0.020).

3.5. Air Pollution and Risk of MS Disability Progression

Four papers evaluated the association between PM2.5 and disability progression. The model that pooled these papers did not find a significant association between the two (data not shown: HR = 1.26; 95% CI: 0.92–1.73); the model had significant heterogeneity (I ² = 79%, p < 0.001).

Two papers evaluated the association between PM10 and disability progression. The model that pooled these papers demonstrated a significant association between the two (data not shown: HR = 1.31; 95% CI: 1.04–1.65); the model had insignificant heterogeneity (I ² = 0%, p = 0.760).

One paper [20] evaluated the association between NO₂ and CO and disability progression. The paper demonstrated a significant association between the exposures and the outcomes (data not shown).

The same paper [20] evaluated the association between SO₂ and O₃ and disability progression. The paper did not find significant associations between the variables of interest (data not shown).

3.6. Air Pollution and Risk of CEL Development

One paper [21] evaluated the association between short‐term exposure to PM2.5 and CEL. The paper found a significant association between PM2.5 and CELs development (data not shown).

Two papers evaluated the association between short‐term exposure to PM10 and CELs. The model that pooled these papers demonstrated a significant association (data not shown: HR = 1.02; 95% CI: 1.01–1.04); the model had insignificant heterogeneity (I ² = 0%, p = 0.860).

3.7. Quality Assessment of the Evidence

The median score of the included studies was 6 with a range between 4 and 9 indicating that 13.6% were of good quality, 82.6% of intermediate quality, and 3.8% of poor quality. The articles mainly lost points in the three categories which were the selection criterion due to small sample size and unrepresentativeness of the population of interest (86.4%), the comparability criterion due to lack of or incomplete adjustment for confounding variables (95.5%), and the outcome criterion due to inability to present that the exposure preceded the outcome (86.4%).

3.8. Publication Bias

Publication bias assessment revealed a symmetric funnel plot indicating the low risk of publication bias. Figure 4 demonstrates the publication bias funnel plot.

FIGURE 4.

Publication bias funnel plot. The figure presents the publication bias funnel plot among the included studies. It demonstrates a symmetrical funnel plot indicating the low risk of publication bias.

4. Discussion

This systematic review and meta‐analysis was conducted to evaluate the association between air pollution and risk of MS and its severity. Our results indicated significant associations between exposure to several air pollutants on the long‐ and the short‐term and MS. First, PM2.5 and PM10 were associated with a higher risk of developing MS. Second, short‐term exposure to PM2.5, PM10, NO₂, and O₃ was associated with a higher risk of relapses in MS. Third, short‐term exposure to PM10 was associated with disability progression as indicated by EDSS. Lastly, short‐term exposure to PM10 was associated with the risk of CEL development.

Recently, the role of air pollution in the development of autoimmune disease has attracted more attention. We found that long‐term exposure to PM2.5 and PM10 was associated with a higher risk of developing MS. As indicated in the results, evidence regarding CO and SO₂ is limited in the literature. Previous systematic reviews investigated the association between exposure to air pollution specifically PM and MS [14]. Two previous meta‐analyses that included only 6 and 10 papers demonstrated that PM10 is associated with a 5% risk of MS while PM2.5 was associated with 18% [22, 23].

Here, we expand on the literature by including 22 papers and add to the literature a specification of long‐term vs. short‐term exposure to air pollution and MS while expanding to other air pollutants. We confirm previous findings of the association between PM2.5 and PM10 and risk of MS development; however, we also show that the association is much larger than it was estimated before as both PM2.5 and PM10 were associated with higher risk of developing MS by almost 20%. We also demonstrate that NO₂ was not associated with higher risk of developing MS.

Although the effect of air pollution on the pathogenesis of MS remains not fully understood, several mechanisms have been postulated. One of the major mechanisms is the inflammatory response and subsequent oxidative stress leading to neuroinflammation and breaking the balance between immunity and tolerance [24, 25]. Inhalation of air pollution initiates chemical reactions and produces reactive oxidative species (ROS) initiating inflammatory cascades via redox‐sensitive mitogen‐activated protein kinase (MAPK) and nuclear factor kappa‐light‐chain‐enhancer of activated B cells (NF‐κB), resulting in the expression of cytokines [26]. In addition, ROS results in lipid peroxidation, protein oxidation, and disruption of deoxynucleic acid (DNA) and ribonucleic acid (RNA) [27]. Thus, accumulatively, air pollution creates a significant imbalance between ROS and antioxidants, resulting in proinflammatory cytokines such as interleukins (ILs) and tumor necrosis factor alpha (TNFα) [27]. This inflammatory response results in cell death and the release of self‐antigens that are capable of stimulating the production of auto T‐cells by increasing antigen presenting cells and facilitating their entry to the CNS [28, 29]. While all air pollutants can drive this pathophysiologic mechanism, we did not find an association between NO₂ and risk of developing MS. There are no data available to understand why PMs but not NO₂ are associated with the risk of MS. One possibility is that the magnitude of the association is small and needs a larger sample size to find. Another possibility is that we still did not completely uncover the pathophysiologic association between air pollution and risk of developing MS. There is also scarcity in the data of some pollutants such as CO, SO₂, and O₃. Thus, future immunology and cellular studies are needed to further explore the association between air pollution and risk of developing MS. In addition, further research is needed to investigate the association between CO, SO₂, and O₃ and risk of developing MS.

We also found a significant association between short‐term exposure to PM2.5, PM10, NO₂, and O₃ and risk of relapses in MS. Short‐term increases to air pollutants can re‐break the balance between the ROS and antioxidants either through systemic exposure or direct entrance to the CNS through the blood–brain barrier from the olfactory system [17, 27, 28]. Another mechanism is the ability of air pollution to compromise the healthy brain vascularity [11]. Previous studies demonstrated that T₂‐lesions and chronic blackholes are more likely to form in watershed regions of the brain indicating the importance of brain vascularity in MS [30]. No previous meta‐analysis estimated the association between air pollution and risk of relapse solely; however, a previous meta‐analysis highlighted that PM2.5 and PM10 was associated with a 28% increase in the risk of relapse and incidence of MS combined [22]. Here we add that PM2.5, PM10, NO₂, and O₃ were associated with relapses among patients with MS regardless of its association with risk of developing MS. We also add that NO₂ and O₃ are also involved in the exacerbations of MS. In the same line, we add to the current knowledge by demonstrating a significant association between PM10 and risk of developing CELs. There was scarcity in the data regarding the association between PM2.5 and risk of developing CELs with no studies in the literature exploring the association of air pollutants other than PMs and risk of developing CELs.

Moreover, we demonstrated a significant association between PM10 but not PM2.5 and risk of disability progression as indicated by EDSS. Disability progression is mainly mediated by relapses in RRMS and PIRA in SPMS and PPMS [4]. The association between air pollution and risk of relapses and developing CELs explains the association between air pollution and disability progression among RRMS. However, there are no studies in the literature that evaluated the association between PIRA and air pollution in MS. From an immunologic standpoint, when air pollutants enter the CNS, they are able to cause epigenetic changes in the glial cells and activate them [27]. Microglia are considered the main drivers of PIRA through slowly expanding lesions and chronic active lesions (CALs) [17, 28]. Another explanation is the impact of air pollution on the vasculature of the brain as it was highlighted that short exposure to air pollution can compromise the blood supply to the brain [11]. Previous studies highlighted the tendency of CALs to form in watershed areas of the brain and the possibility that CALs arise from the failure of reparative mechanisms due to low blood and oxygen supply [31]. We did not find an association between PM2.5 and disability progression. Again, there is no data in the literature to explain the reason why PM2.5 is not associated with disability progression; however, since it was associated with relapses, it is possible that the analysis was underpowered to find a significant association. No studies in the literature investigated the association between CO, O₃, and NO₂ and disability progression in MS. Hence, future studies that investigate the association between these pollutants and disability progression are needed. In addition, future studies that evaluate the association between air pollution and PIRA in MS are also needed.

In addition to the importance of environmental exposure in adulthood and its impact on MS, environmental risk factors are also fundamental in childhood and adolescence. A previous systematic review and meta‐analysis that included 87 studies across 20 countries demonstrated that EBV, exposure to smoking and chemicals, as well as obesity, were associated with higher risk of MS [32].

Few limitations should be acknowledged before concluding. First, the majority of the data comes from Europe and the Americas, which limits the generalizability of our findings. However, countries in Africa and Asia are expected to have higher rates of pollution, so exploring the association between air pollution and MS in these countries is needed. Second, the case definition of developing MS, relapses, and disability progression might be less reliable when administrative data is used. Third, our findings are affected by ecological fallacy and the presence of socioeconomic confounders such as access to health, diet, and exercise. The majority of the studies lost points in NOS due to small sample size, lack of adjustment to confounders, and poor outcome assessment; thus, future studies are encouraged to increase their sample size, adjust for confounders, and improve outcome assessment. Moreover, we were not able to specify a follow‐up duration for the included studies due to the variability of the follow‐up durations across different studies, which limits the associations observed in our paper. Another limitation stems from the fact that different studies were conducted at different years, which means they have used different McDonald's criteria versions, and there were differences in the diagnostic tools and treatments used. Our analysis was also limited by not including meta‐regression due to the relatively small number of the included articles. Future larger systematic reviews and meta‐analyses should focus on studying whether specific population‐related variables such as sex or age result in differences in the impact of air pollution on the disease course and development. Lastly, some of our models had significant heterogeneity, which might be due to the differences in the characteristics of the included studies, differences in the definition of the duration of the exposure, and differences in the assessment of the outcomes since different studies were conducted at different times, as indicated in the above limitations.

Despite these limitations, our study provides strong evidence of the association between air pollution and risk of developing MS and severity of MS including relapses, developing CELs, and disability progression. MS remains a significant burden on global health, contributing to disability and quality of life. Global efforts and policies are needed to reduce and combat air pollution levels to reduce the burden of MS among other neurological and non‐neurological diseases.

Author Contributions

A.A.T. was involved in conceptualization; A.A.T. and T.N.A.‐S. were involved in project administration; A.A.T. and T.N.A.‐S. were involved in data acquisition; A.A.T. was involved in data analysis; A.A.T. and T.N.A.‐S. were involved in manuscript writing; A.A.T. was involved in editing and reviewing the manuscript.

Funding

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1: acn370330‐sup‐0001‐Supinfo.pdf.

Data Availability Statement

The data associated with this manuscript are available from the corresponding author upon reasonable request.