Differential Health Impacts of Short‐Term Particulate Matter Exposure in Arid and Semi‐Arid Regions: A Scoping Review Authors: Emmet Norris, Jennifer Bailey, Yiqun Ma, Anusha Goswami, Sarah Aarons, Tarik Benmarhnia Published: 2026 ABSTRACT AbstractParticulate matter (PM) exposure is a leading risk factor for cardiovascular and respiratory morbidity and mortality across the world. While arid and semi‐arid regions account for more than 40% of terrestrial land, there is a paucity of research investigating the relationship between PM exposure and health outcomes or how risk varies across different populations specific to these climate regions. To address this, we performed a scoping review to assess the extent of literature in arid and semi‐arid regions that consider such morbidity and mortality risks from short‐term PM exposure. Our review identified 102 studies published after 2012 located within four arid climate classifications: hot arid, cold arid, semi‐arid hot, and semi‐arid cold. The meta‐analysis found a pooled overall increased relative risk (RR) of both cardiovascular and respiratory mortality and morbidity, with an 8% higher risk of morbidity in hot arid regions (1.11, 95% confidence interval (CI): 1.05, 1.17) compared to cold semi‐arid regions (1.02, 95% CI: 1.00, 1.03) for adults (18–64). Further, we assessed effect modification (age, sex, population density and season) and found weak evidence of increased risk for both males and adults. The studies included in this review highlighted compounding risks, such as dust storms, wildfire, and extreme temperature, however there was a significant lack of studies situated in highly populated arid regions, with none identified in Africa. Given the public health burden from PM exposure and the impact of climate change and anthropogenic activities which destabilize land‐surfaces and increase dust emissions, additional focus is owed to demographic effect modifiers and the role of climate specific risks. Introduction Air pollution, and specifically particulate matter (PM) pollution, constituted the largest risk factors contributing to the global burden of disease and deaths in 2021 (Brauer et al., 2024; Health Effects Institute, 2024). It is estimated that approximately 4.7 million deaths globally could be attributed to ambient, or outdoor, fine particulate matter (PM2.5, or particles 2.5 μm or less in diameter) exposure (Health Effects Institute, 2024), with 99% of the world's population exposed to PM2.5 at concentrations higher than the 2021 World Health Organization (WHO) global air quality recommended exposure limit of 5 μg/m3 (Yu et al., 2023). PM is commonly studied in health research because it has well‐documented, severe adverse health effects, differentially impacts health due to its source and chemical composition, and is a widespread pollutant generated by both human and natural sources (WHO, 2021b). A growing body of epidemiological evidence links both short‐ (over hours or days) and long‐term (over months or years) exposure to PM with a wide range of acute and chronic health endpoints, particularly in respiratory and cardiovascular systems (e.g., Atkinson et al., 2015; Orellano et al., 2020; Rajagopalan et al., 2018). Research has consistently found that there is no amount of PM exposure that could be considered safe at the population level (Rajagopalan et al., 2018). As PM can penetrate the respiratory system when inhaled, lung tissue becomes irritated and immune responses are triggered, leading to inflammation (Thompson, 2018). Fine particles can enter the bloodstream, resulting in systemic inflammation and oxidative stress that affects blood vessels, increases blood pressure, and promotes clot formation (Araujo, 2010; Lutsey et al., 2025). These processes elevate the risk of respiratory infections, asthma exacerbation, and cardiovascular events like heart attacks and strokes. Understanding the health impacts of short‐term PM exposure is critical for effective policy implementation, as such knowledge enables the design of emergency responses that mitigate increases in hospitalizations and emergency room visits for the general public, and particularly among vulnerable populations that are more severely impacted by exposure to PM. Both exposure and vulnerability to PM and related health impacts are not equally distributed across space or populations. Although ambient PM2.5 concentrations have decreased or stabilized in many predominantly high‐income countries over recent decades, pollution levels remain high in Asian, African, and Middle Eastern countries (Health Effects Institute, 2024), and 80% of people exposed to unsafe average annual concentrations of PM2.5 live in low‐ and middle‐income countries (Rentschler & Leonova, 2023). Previous work has shown that some population subgroups are systematically exposed to higher levels of PM than others (Jbaily et al., 2022; Ma et al., 2023), and that some are more susceptible to developing adverse health outcomes from PM exposure compared to others due to both external and intrinsic risk factors (Bell et al., 2013). As more people concentrate in urban areas globally, cities tend to experience higher pollution levels due to increased human activities and anthropogenic sources of PM emissions (Health Effects Institute, 2024; Karagulian et al., 2015), with dense populations facing greater public health burdens from elevated exposure (Kim et al., 2025). Urban populations may be particularly vulnerable to PM‐related health impacts due to the combined influence of elevated exposures, exposure to pollution sources associated with particularly harmful PM compositions and co‐pollutant mixtures (Nan et al., 2023), and underlying social and structural inequities that modify both exposure and susceptibility (Bailey et al., 2017; Hajat et al., 2015). In these complex urban environments, the interaction between variable PM composition exposure and differential health outcomes is currently poorly understood, despite being critical for reducing the overall and disproportionate health burdens experienced globally. Previous literature reviews of the health impacts from short‐term PM exposure show that some populations, often termed “susceptible” or “vulnerable” have an increased risk of PM‐related health effects (Bell et al., 2013; Makri & Stilianakis, 2008; Sacks et al., 2011). Sociodemographic factors such as age, sex, ethnicity (Jbaily et al., 2022), socioeconomic status (SES, Hajat et al., 2015; Jbaily et al., 2022), and minority (Liu, Clark, et al., 2021; Yang et al., 2022) or migrant status (Fong et al., 2022; Liang et al., 2019) have been shown to increase vulnerability and modify the effects of PM exposure on mortality and morbidity risks. For example, in the United States (U.S.) non‐Hispanic Black people are 3.47 more likely to die from PM2.5‐attributable cardiovascular disease than non‐Hispanic white people (Ma et al., 2023). Such differential susceptibility arises through multiple, often interacting pathways that span individual (e.g., pre‐existing comorbidities, increased psychosocial stress) and structural (e.g., systemic policies and inequalities such as discriminatory practices dictating housing, proximity to pollution sources, and access to healthcare) risk factors (Bailey et al., 2017). Human exposure to PM varies both temporally (e.g., day of week, season) and spatially (climate, land cover, PM source, proximity to emission source, urban vs. rural; Dodman et al., 2022). Climate change is expected to increase global particulate matter emissions (Park et al., 2020) primarily through increasing droughts, with arid and semi‐arid regions especially susceptible (Safriel et al., 2005). The combination of land‐use changes (e.g., urban development, agriculture, mining, etc.) and anthropogenic emissions (e.g., fossil fuel combustion, biomass burning, etc.) alter PM concentrations and increase exposure risk for communities in arid and semi‐arid regions. However, whether the vulnerable populations that have been identified in other climate regions are the same as those in arid and semi‐arid regions warrants further investigation. Arid and semi‐arid regions cover 30.2% and 15.2% of the terrestrial surface, respectively (Beck et al., 2018; Peel et al., 2007), supporting a combined population of 2.5 billion people (Gaur & Squires, 2018; Scholes, 2020). While anthropogenic activity drives significant short‐term changes in PM emissions (Chen et al., 2018), increases in aerosol optical depth (AOD) between 2001 and 2010 are linked to long‐term meteorological variability in arid and semi‐arid regions such as the Middle East and Northern Africa (Pozzer et al., 2015), indicating the role of climate on PM emission increases. While semi‐arid regions have already expanded in areal extent by 7% from 1948–1962 to 1990–2004 (Huang et al., 2016), climate models of the 21st century indicate these regions are expected to become more arid with decreased rainfall and increased evaporation (Scholes, 2020). Between 1990 and 2020, 77.6% of the Earth's terrestrial land became drier (Vicente‐Serrano et al., 2024). In a scenario of 3°C global average temperature increases due to anthropogenic climate change, 5.2 million km2, or 3.5% of total terrestrial land, is estimated to become arid (Spinoni et al., 2021). Further contributing to PM emissions, rapid population growth in arid and semi‐arid regions has led to widespread land‐surface disturbance during urbanization, industrial and commercial operations (Tsiouri et al., 2015), and agricultural activities (Guan et al., 2016). During the period of 1982–2015, the combination of unsustainable land‐use and climate change led to desertification of 6% of global dryland area (Mirzabaev et al., 2022). Deserts, unique to arid regions, currently emit more than half of the global PM mass (Kok et al., 2017), and exposure to desert dust specifically has been associated with adverse human health impacts (Goudie, 2014; Zhang et al., 2016). In the Southwest U.S., climate change‐induced increases in aridity are projected to increase the emissions of fine dust by 57% (Achakulwisut et al., 2019). Arid regions are especially vulnerable to soil disturbances, leading to new and emerging dust sources (Guan et al., 2016). In these regions, climate change is driving more frequent and intense extreme heat events, which often coincide with pollution episodes due to shared underlying meteorological drivers (Schnell & Prather, 2017). Additionally, heat can intensify the formation of PM through chemical reactions that occur more readily in hot, sunny conditions (Alahmad et al., 2023). Together, extreme heat and PM pollution have been shown to have a synergistic effect, worsening the health impacts beyond the sum of their individual contribution (Rahman et al., 2022; Schnell & Prather, 2017). Despite this, there is a limited body of research assessing this compound exposure in arid environments. Further, both environmental and sociodemographic characteristics of semi‐arid and arid regions increase the potential for large and disparate health impacts related to PM. Roughly 72% of drylands and almost 100% of hyper‐arid land are located in developing countries (Huang et al., 2017; Hulme, 1996; Safriel et al., 2005). Semi‐arid regions are thought to be most vulnerable to destabilization and ecosystem loss as they are at the intersection between population growth and fragile land surfaces (Safriel et al., 2005). Broadly, rural communities (Singh & Chudasama, 2021; Sternberg & Edwards, 2017) and women (Yadav & Lal, 2018) are identified as being most vulnerable to climate change‐related impacts on environmental health in arid regions. This results from the intersection of multiple factors related to poverty, limited infrastructure, access to healthcare, gender norms, environmental exposures, and evidence of enhanced susceptibly to PM during pregnancy (Sapkota et al., 2012). Despite evidence that, the majority of research has focused on PM exposure in high income countries (Jaganathan et al., 2019; Yegros‐Yegros et al., 2020). Separately, there is a lack of studies in many arid regions (e.g., Imane et al., 2022), leading to gaps in knowledge and potentially inaccurate effect estimates. For example, a study in Qatar found excess mortality from fugitive particulate matter to be 11 times greater than World Health Organization models (Hassan et al., 2022), highlighting the need for additional, location‐specific studies to better understand this relationship. Given the unique sources of PM and population demographics in different climates, this study aims to conduct a scoping review to summarize existing scientific literature on the association between short‐term exposure to PM2.5 and PM10 and cardiovascular and respiratory morbidity and mortality in global cities within arid and semi‐arid regions. By doing so, we intend to address a gap in the literature regarding how PM exposure impacts population health in semi‐arid and arid climates, and, particularly, how these impacts differ across sub‐populations to identify those most vulnerable. To do this, we identify potential effect modifiers of this association that have been assessed in these regions and perform a meta‐analysis to synthesize the existing literature around vulnerable populations in semi‐arid and arid regions. Based on these analyses, we provide future directions for informed research to lessen the health burden in such regions. Methods Study Region Arid regions are classified based on an aridity index, defined as the long‐term average of the ratio of the mean annual precipitation to mean annual evaporation (arid = 0.05–0.20, and semi‐arid = 0.20–0.50) and generally low precipitation (arid <25 cm, and semi‐arid <50 cm per year; Safriel et al., 2005). This study uses the Köppen climate classification (Köppen, 1918)—using the classifications for the period 1991–2020 (Beck et al., 2023)—to identify regions that are arid or semi‐arid, divided into four distinct subcategories. The Köppen climate classification system was selected for this study because it integrates temperature, precipitation, and vegetation—factors that, both independently and in combination, are relevant to understanding regional differences in PM exposure risk. Hot arid climates (BWh) are typically low latitude deserts with average annual temperatures greater than 18°C (64°F), such as Northern Mexico, parts of the southwestern United States, and Saudi Arabia. Cold arid climates (BWk) are typically found in mid‐latitude deserts such as Central Asia or southern South America, with annual average temperatures below 18°C. Semi‐arid regions are similarly characterized by hot (BSh) and cold (BSk), and extend from the boundaries of arid deserts into vast grasslands, notably the U.S. Great Plains and Eurasian Steppe (BSk), and in Southern Africa and Australia (BSh). As city delineations can be arbitrary and many cities exist at a boundary between an arid and semi‐arid climate or a semi‐arid and continental climate, in the 1991–2020 classification (Beck et al., 2023), the most arid characterization for this period was chosen when identifying cities that fell within a boundary of two climate types. We sought studies which assessed the relationship of interest at the city scale, according to urban center delineations from the Global Human Settlement Urban Centre Database (Mari Rivero et al., 2024). We focused on the city level so as to extract the estimated health impacts for a population that was likely exposed to more similar PM sources and concentrations as compared to wider geographic areas (e.g., counties, states, countries). To enhance comparability, cities were considered as meeting our criteria if they included semi‐arid or arid climates, and larger spatial units were considered if either fully arid, semi‐arid, or some combination of the two. For example, studies quantifying the relationship of interest for the country of Kuwait could be considered as the entire country is arid. Search Strategy The goal of this study was to assess the extent of global literature considering short‐term PM‐related cardiovascular and respiratory health outcomes in arid and semi‐arid regions. We identified peer‐reviewed studies published in English between 1 January 2012 and 24 April 2024 in two separate searches. The initial search was conducted in April 2024 and updated in September 2024 so as to include any articles published during the course of the review. We followed the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses Extension for Scoping Reviews (PRISMA‐ScR) checklist to conduct this review (Tricco et al., 2018) with slight changes constructed through our own protocol. Following the framework to establish effective questions in exposure and health research (PECO; Morgan et al., 2018): our review addresses human populations in semi‐arid and arid regions (population) and the effect of short‐term exposure to PM (exposure), comparing subpopulation effect modifiers (comparator) on cardiovascular (ICD‐10: I00‐I99), respiratory (ICD‐10: J00‐J99), and all‐cause (ICD‐10: A00‐R99) mortality and morbidity (outcome). The review further aimed to identify empirical epidemiology studies investigating effect modification in such associations between PM exposure and cardiovascular and respiratory mortality and morbidity. The review is composed of two searches, where the first component serves as an update of an existing systematic review published by Bell and colleagues in 2013. The second component is an expanded search based on the Bell et al. (2013) approach, designed to capture a broader range of studies by including more comprehensive keywords and study components explicitly grouped into five categories: health outcomes, study type, effect modification, PM exposure, and exposure type. All terms in their final searchable format are reported in Text S1 of Supporting Information S1. We used these terms to search the National Library of Medicine's MEDLINE database through PubMed (National Library of Medicine, n.d.). We followed the same approach for both the initial and updated search. Titles were first manually screened for duplicates, reviews, relevance, and locations outside of our climate classification of interest (Figure 1). Titles and abstracts were then further screened and included based on the following criteria: considered short‐term exposure (i.e., same‐day or within the last 2 weeks); PM was the primary exposure; considered cardiovascular or respiratory health outcomes using objectively measured data (i.e., hospital admissions, emergency department visits, mortality counts, etc.); provided an effect estimate on the relationship between PM and such health outcomes; provided this estimate for a specific semi‐arid or arid city or area. As such, this meant that studies which provided estimates of the synergistic effect of PM with other variables, such as temperature, or only included pooled multi‐city, nationwide, or multi‐country estimates (and not city‐specific estimates) were excluded. While the inclusion of an estimate of the relationship between PM exposure and cardio‐respiratory health outcomes was necessary to be considered in this review, we regarded this criteria met for both explicitly stated estimates and those shown in exposure response curves so as to comprehensively capture research efforts that quantified this relationship, however, estimates could not be extracted (i.e., were only reported in figures). Given the nature of the studies and our inclusion criteria, a full‐text review including supplementary material if relevant was required for the majority of studies. The following information was extracted from each study: citation, study location, coordinates, climate classification of the study area, study start and end dates, study population, data source, particulate matter type (i.e., PM2.5, PM10), temporal resolution or lag structure considered, health outcomes assessed, study method (i.e., time‐series, case‐crossover), statistical model employed, effect measure estimates, any confounders considered, effect modifiers analyzed, and estimates associated with the effect modifiers if considered. This data was obtained from tables, text descriptions, and supplemental material from the respective articles. The studies which met our criteria were included in the review and synthesized in an overview table (Table S1). Studies which did not report effect estimates in writing or tables (e.g., only in a figure), could not be included in the meta‐analysis. Flow diagram detailing the study selection process and reasons for study exclusion. Meta‐Analysis For the meta‐analysis, a secondary table was created to include only studies with reported effect estimates. For each of these studies, we extracted the overall effect estimates, estimates associated with effect measure modifiers (along with all relevant details, such as subpopulation assessed, pollution unit considered, etc.), and the measure of uncertainty calculated for both types of estimates (e.g., confidence intervals). Both time‐series and case‐crossover studies were included and compared together in the meta‐analysis, as estimates of air pollution related mortality and morbidity from both study types have been found to be comparable (Fung et al., 2003; Lu & Zeger, 2007). When risk estimates for multiple lags were available, the shortest (e.g., same day, lag 0–3, lag 1) was extracted to be used in the meta‐analysis. We chose to focus on the shortest lag as same day exposure has been shown to be biologically relevant for capturing immediate physiological responses for cardiovascular and respiratory outcomes and is the most consistently reported exposure window across studies, allowing us to maximize study inclusion and ensure comparable estimates (Newell et al., 2017; Wu et al., 2022). While studies reported results using a variety of risk measures (Table S1), all estimates were converted to relative risk estimates (RR) and confidence intervals from 10 μg/m3 increase in exposure for direct comparison (see Text 2.3 in Supporting Information S1). Estimates from studies that considered specific disease as well as general outcomes were grouped into either cardiovascular (e.g., ischemic stroke ICD‐10: I63.*; and general, ICD‐10: I00‐I99), respiratory (e.g., asthma, ICD‐10: J45; and general, ICD‐10: J00‐J99), or all‐cause categories. Additionally, studies that used a range of age categories in the primary analysis or when assessing age as an effect modifier, were grouped into children (ages 0–18), adults (ages 0–65), and older adults (aged 65 and up) to allow for general comparison between studies. Studies reported effect estimates in two categories for sex (male/female) as a biological distinction, while some reported gender (male/female) which is a social distinction. For the purposes of this review, we have decided to use both estimates under the category of sex. Following previously published methods (Benmarhnia et al., 2015), meta‐analyses were considered for categories (climate, outcome, and PM exposure type) and effect modifiers (demographic or temporal) if estimates were available from more than 10 studies. The effect modifiers that were sufficiently numerous for meta‐analysis were sex, age, season, climate, and population density. Population density was calculated using the urban population and boundary of each city (GHS Urban Centre Database; Mari Rivero et al., 2024). We performed separate meta‐analyses for PM10 and PM2.5, as well as mortality and morbidity for all modifiers if sufficient studies were available. Many studies reported effect estimates for multiple outcomes (e.g., both cardiovascular and respiratory diseases), which were assessed for heterogeneity separately, leading to a greater number of comparators than studies. In analyses stratified by subpopulation, heterogeneity was examined for pairings (e.g., children vs. adults) when at least five effect estimates were available, following the threshold applied by Bell et al. (2013). First, a restricted‐maximum likelihood (REML) random‐effects model and I 2 statistic was used to create a pooled estimate for effect modifier subgroups. A heterogeneity assessment was then performed using the ratio of relative risks (RRR) for subgroups from the same study (Altman & Bland, 2003), calculated for all eligible estimates (e.g., RRMale/RRFemale). The RRR estimate and the I 2 statistic were used to assess the significance, with an I 2 threshold of >50% (Higgins & Thompson, 2002). This determines whether there was in fact statistically significant heterogeneity in the magnitude of effect between subgroups or if the variation is due to sampling error, or otherwise stated: testing if variation exceeds the amount expected under a null hypothesis of no heterogeneity (using 10% as an a priori threshold; Kaufman & MacLehose, 2013). Studies were excluded from heterogeneity assessment if two comparable subgroup estimates were not provided (e.g., both male and female). Further, a Cochran's Q test was performed to assess heterogeneity between population subgroup risk estimates across multiple studies. Equations for all meta‐analyses are available in the Text S2 of Supporting Information S1. Publication bias was assessed using both funnel plots and Egger's regression model for each subgroup RRR analysis. All analyses were performed in RStudio using the {meta} package (see Text S2 in Supporting Information S1). Results Study Locations and Types Our search identified 5,472 records, and after initial screening for inclusion criteria, 952 articles remained for consideration. After further screening for cardiovascular and respiratory studies and for sufficient geographic analysis, 102 studies were included in this analysis, including one study that was found outside of the search (Bayart et al., 2024). A detailed account of the number of studies along with the reason for exclusion is presented in Figure 1. Of the 102 studies included in this study, 70 reported morbidity and 37 mortality data, with the majority (68%) using time series analysis (Table 1). Many studies assessed both PM10 and PM2.5 exposure effects (n = 43), while 22 considered only PM10, 36 only PM2.5 and one assessed PM of all sizes (total suspended particles; TSP). Age was the most considered modifier (n = 67), followed by sex (n = 55) and season (n = 24) (Figure S1 in Supporting Information S1). Ten studies did not consider any effect modification. Only two studies considered SES, and one poverty. Additionally, five studies explored the impact of dust storms on health, and five studies the effect of wildfire. Respiratory diseases were considered in 54 studies, while cardiovascular diseases were considered in 50 studies. Many studies (n = 28) also presented estimates for all‐cause mortality and morbidity (ICD‐10: A00‐R99), which are also included for comparison. Table 1Results of the Scoping Review, Showing the Number of Studies in Several CategoriesCategoryCountOverallTotal Studies102ClimateSemi‐arid, Cool (BSk)68Semi‐arid, Hot (BSh)19Arid, Hot (BWh)19Arid, Cold (BWk)1Outcomes studiedRespiratory Disease Mortality16Respiratory Disease Morbidity38Cardiovascular Disease Mortality20Cardiovascular Disease Morbidity30All‐cause Mortality25All‐cause Morbidity3PM exposurePM2.5 78PM10 64Effect modifiers consideredSex55Age67Season24Study typeCase‐crossover24Time‐series69 Results of the Scoping Review, Showing the Number of Studies in Several Categories The location of studies is geographically inconsistent, with no studies located in Africa or Oceania (Figure 2) despite the presence of arid and semi‐arid regions. In comparison, there were 43 studies located in China, 14 in the US southwest, and 11 in Iran. Overall, the most studies were conducted in Asia (n = 49), followed by the Middle East (n = 22), North America (n = 16), South America (n = 10), and Europe (n = 8). The majority of studies were in the BSk zone (n = 68), followed by the BWh zone (n = 19), then the BSh zone (n = 19), and only one study in the BWk climate zone. Global map, showing the location and number of studies, by city, region, and climate, and the arid and semi‐arid climate zones as per the Köppen climate classification. Pooled Effect Estimates and Heterogeneity Assessments This meta‐analysis is intended to summarize the evidence of short‐term PM exposure and cardiovascular, respiratory, and all‐cause morbidity and mortality, and further, the potential effect modification of available variables from all applicable literature. After producing combined RR estimates for the entire population across all studies, we then systematically calculated pooled RR estimates by potential effect modifiers, including climate, sex, age, and population density using a restricted maximum likelihood random effects (REML) model. Using the same approach, heterogeneity between subgroups was explored and, finally, we provide a pooled ratio of relative risk (RRR) estimates. In line with our previously mentioned criteria, results are only reported for the specific combinations of PM exposure and disease categorization where there were sufficient (n = 5) studies with usable estimates for assessing pooled relative risk as well as subgroup heterogeneity and its statistical significance regarding effect heterogeneity. All Population Estimates We considered the association of PM2.5 and PM10 with all‐cause, cardiovascular, and respiratory mortality and morbidity. As seen in Figure 3, for PM2.5 exposure and morbidity, respiratory diseases had a higher pooled RR of 1.04 (95% CI = 1.04, 1.05) than cardiovascular disease (pooled RR = 1.03, (95% CI = 1.02, 1.05). However, for PM10 exposure, risk of cardiovascular morbidity (RR = 1.02, 95% CI = 1.01, 1.03) was greater than that for respiratory morbidity (RR = 1.01, 95% CI = 1.01, 1.01) (Figure S2 in Supporting Information S1). The increase in cardiovascular, respiratory, and all‐cause mortality risk related to PM2.5 and PM10 exposure were extremely similar when considering the entire population (pooled relative risks of 1.01 with varying confidence intervals at or above the null, see Figure S3 in Supporting Information S1 for more details). The only slight deviation occurred in the estimated pooled risk for all‐cause mortality in relation to PM10 exposure (RR = 1.00, 95% CI = 1.00, 1.01). Pooled morbidity risk ratios from 10 μg/m3 increase in PM2.5 exposure from included studies. Abbreviations in figure: CVD = cardiovascular disease; RD = respiratory disease. Effect Modification by Sex Of the 55 studies considering effect measure modification by sex, there are 13 which reported specific effect estimates for PM2.5 exposure and 19 for PM10 (Table S1). For PM2.5 exposure morbidity, females had a pooled relative risk of 1.04 (95% CI = 1.03, 1.06) for all respiratory diseases and 1.01 (95% CI = 1.00, 1.01) for all cardiovascular diseases, while males had a risk of 1.03 (95% CI = 1.01, 1.06) and 1.05 (95% CI = 1.03, 1.07) for the same, respectively (Figure 3). The pooled morbidity risk from PM10 exposure for women (respiratory RR = 1.01, 95% CI = 1.00, 1.02; cardiovascular RR = 1.01, 95% CI = 0.99, 1.03) and for men (respiratory RR = 1.01, 95% CI = 1.01, 1.02; cardiovascular = 1.02, 95% CI = 1.00, 1.05) is lower than that from PM2.5 exposure (Figure S3 in Supporting Information S1). While the estimated associations between short‐term exposures to PM2.5 and morbidity were generally slightly higher (and marginally statistically significant) for men compared to women, heterogeneity analyses found no statistically significant difference in respiratory (RRR = 1.00, I 2 = 33%) or cardiovascular morbidity risk (RRR = 1.00, I 2 = 0%) by sex (Table 2). PM10 exposure estimates were only sufficient (n = 6) for all‐cause mortality analysis, suggesting a 3% lower risk for females (RR = 0.97, 95% CI = 0.95, 0.99) and a 4% lower risk for males (RR = 0.96, 95% CI = 0.94, 0.97) (Figure S2 in Supporting Information S1). Table 2Summary of Meta‐Analysis of HeterogeneityMeta‐analysis# Of studiesRRR estimatesHeterogeneity statisticsREML estimateLower CIUpper CI I 2 H statisticTauP‐valueCochran's QZ statisticRespiratorySeason (RR_Cold/RR_Warm)71.020.991.0574.1%1.960.020.0023.151.88Sex (RR_Female/RR_Male)131.001.001.007.0%1.040.000.3812.90−0.14Age (RR_Adult/RR_Older Adult)81.060.971.1588.7%2.970.080.0061.681.52Age (RR_Child/RR_Adult)110.980.951.0085.8%2.650.040.0070.33−1.92Age (RR_Child/Older RR_Adult)81.000.981.0275.0%2.000.010.0027.950.15CardiovascularSex (RR_Female/RR_Male)81.000.991.014.8%1.020.010.397.35−0.98Age (RR_Adult/RR_Older Adult)61.020.981.060.0%1.000.000.782.481.07 Note. Effect modification by age, sex and season are displayed for respiratory and cardiovascular morbidity from PM2.5 exposure. RR = Relative Risk Estimate; REML = Restricted Maximum Likelihood; CI = 95% Confidence Interval. Summary of Meta‐Analysis of Heterogeneity Note. Effect modification by age, sex and season are displayed for respiratory and cardiovascular morbidity from PM2.5 exposure. RR = Relative Risk Estimate; REML = Restricted Maximum Likelihood; CI = 95% Confidence Interval. Effect Modification by Age Effect estimates for three age categories children (0–18), adults (18–65) and older adults (65+) were compared for all respiratory morbidity outcomes (Figure S4 in Supporting Information S1). From PM2.5 exposure, respiratory morbidity risk was highest for adults, with a pooled RR 1.09 (95% CI = 1.05, 1.13), followed by children 1.06 (95% CI = 1.04, 1.08), while risk for older adults was null and insignificant (RR = 1.00, 95% CI = 0.96, 1.03). The difference in risk between adults and older adults is statistically significant with a RRR of 1.06 (I 2 = 89%). Regarding cardiovascular morbidity from PM2.5 exposure, the adult pool RR was 1.05 (95% CI = 0.95, 1.06), whereas the older adult risk was 1.03 (95% CI = 0.99, 1.06). The ratio of relative risks is also statistically significant (RRR = 1.02) between adults and older adults for cardiovascular morbidity from PM2.5 exposure. Effect Modification by Climate Zone Effect modification was assessed for the three climate zones with greater than 10 studies: BSk (n = 67); BSh (n = 18); and BWh (n = 13). Cardiovascular and respiratory morbidity risk from PM2.5 exposure was highest in populations living in arid regions (RR = 1.05, 95% CI = 1.01, 1.09 and RR = 1.07, 95% CI = 1.05, 1.08, respectively). The risk of cardiovascular morbidity increased by 3% from PM2.5 exposure in cold semi‐arid regions (RR = 1.03, 95% CI = 1.02, 1.04), however, in semi‐arid, hot regions, the risk of cardiovascular morbidity increased by an insignificant 1% (RR = 1.01, 95% CI = 0.95, 1.08). Conversely, the risk of respiratory morbidity increased by 4% for populations exposed to PM2.5 in semi‐arid, hot regions (RR = 1.04, 95% CI = 1.03, 1.05), while semi‐arid, cool climates experienced a 2% insignificant increased risk of respiratory diseases (RR = 1.02, 95% CI = 0.99, 1.06). Effect Modification by Population Density We considered the effect of population density on mortality and morbidity risks by grouping studies into two categories: high (>4,500 people/mile) and low (<4,500 people/mile) density cities. For cardiovascular outcomes in relation to PM2.5 exposure, there was a 5% increased risk of morbidity (RR = 1.05, 95% CI = 1.02, 1.07) in high density cities compared to a 2% increase (RR = 1.02, 95% CI = 1.01, 1.02) in low density cities (Figure 3). For cardiovascular mortality (Figure S3 in Supporting Information S1), there was a 2% increased risk (RR = 1.02, 95% CI = 1.01, 1.02) in high density cities compared to 1% (RR = 1.01, 95% CI = 1.00, 1.02) in low density cities. Respiratory morbidity risk from PM2.5 exposure was also slightly greater in high density cities (RR = 1.05, 95% CI = 1.03, 1.07) compared to low density cities (RR = 1.04, 95% CI = 1.03, 1.05). There was no observed increased or decreased risk of all‐cause mortality from PM2.5 exposure in high density cities (RR = 1.00, 95% CI = 1.00, 1.01), while a marginal increase in risk was observed for low density cities (RR = 1.01, 95% CI = 1.00, 1.01)—see Figure S3 in Supporting Information S1. Considering PM10 exposure, both cardiovascular morbidity (RR = 1.02, 95% CI = 1.01, 1.04) and mortality (RR = 1.02, 95% CI = 1.00, 1.07) risk increased by 2% in low density cities compared to a 1% increased risk for morbidity (RR = 1.01, 95% CI = 0.99, 1.04) and mortality (RR = 1.01, 95% CI = 1.01, 1.02) in high density cities. However, it should be noted that the increase of 1% in cardiovascular morbidity risk from exposure to PM10 is non‐significant in high‐density cities. The small increase in respiratory morbidity risk was similar across low‐density (RR = 1.01, 95% CI = 1.01, 1.01) and high‐density cities (RR = 1.01, 95% CI = 1.00, 1.01). The estimate related to PM10 exposure and all‐cause and respiratory mortality in high‐density cities was 1.01 (95% CI = 1.00, 1.01) which was greater than the low‐density risk of 1.00 (95% CI = 1.00, 1.00). If further stratified by climate, there are more significant differences. For example, in high‐density cities located in BWh regions there was a significantly higher risk estimates for respiratory morbidity of 1.09 (95% CI = 1.07, 1.11) compared to 1.02 (95% CI = 1.01, 1.02) observed in low‐density, arid cities. In BSk climates the relative difference is inverted, with low‐density cities having a pooled risk of 1.04 (95% CI = 1.03, 1.05) and high‐density cities with a risk of 1.00 (95% CI = 1.00, 1.00). Effect Modification by Seasonality Studies generally explored seasonal variation on effect modification by reporting risks from the “cold” and “warm” seasons, corresponding with local climates. From available relative risk estimates (Figure 3), the pooled respiratory morbidity risk from PM2.5 exposure in the cold season was 1.05 (95% CI = 1.03, 1.08), in comparison to the warm season, which was 1.04 (95% CI = 1.01, 1.07). Heterogeneity analysis of these risks finds the cold season has a 2% increased risk (RRR = 1.02, I 2 = 74%). Pooled respiratory morbidity risk from PM10 exposure was not significant, with a risk of 1.01 (95% CI = 1.00, 1.03) in the cold season, and 1.00 (95% CI = 1.00, 1.01) in the warm season. Similarly, PM10 exposure was found to have an insignificant 1.00 (95% CI = 1.00, 1.01) pooled relative risk for cardiovascular morbidity during both periods. Event Specific and Effect Modification by Compounded Risks Multiple studies focused on the potential of PM concentration and composition modulating events to impact PM exposure‐related health outcomes, specifically dust storms and wildfire related smoke. Dust storms were considered in eight studies, one in China (Ma et al., 2016); Spain (Díaz et al., 2012); the Canary Islands (Lopez‐Villarrubia et al., 2021); and five in the Middle East (Aghababaeian et al., 2021; Al‐Taiar & Thalib, 2014; Neophytou et al., 2013; Shahsavani et al., 2020; Vodonos et al., 2015). In summary, the findings include: an increased number of ER visits for respiratory diseases during dust storms (Ma et al., 2016); increased daily mortality in one city, Ahvaz, but decreased in another, Tehran (Shahsavani et al., 2020); limited evidence for increased cardiovascular mortality during dust days (Neophytou et al., 2013); seasonally variable risk for cause specific mortality increase during dust days, though not statistically significant (Díaz et al., 2012); increased emergency department visits during dust storms, especially low altitude and long duration (>5 days) storms (Lopez‐Villarrubia et al., 2021); statistically significant excess risk of mortality (Aghababaeian et al., 2021); increase in myocardial infarction hospitalization, but only at lag >2 (Vodonos et al., 2015); no significant association between dust storm events and same‐day respiratory mortality (Al‐Taiar & Thalib, 2014). Increased morbidity and mortality from wildfire were the focus of eight studies, six from the USA (Do et al., 2024; Horne et al., 2024; Hutchinson et al., 2018; Kiser et al., 2020; Leibel et al., 2020; Resnick et al., 2015), one from Europe (Faustini et al., 2015), and one global study, of which Kuwait was the only applicable location (Chen et al., 2021). Findings included; increased risk of mortality (Chen et al., 2021); elevated respiratory diagnoses during wildfire (Hutchinson et al., 2018); increased asthma hospitalization during the wildfire season (Horne et al., 2024); increased emergency visits for asthma when smoke was present (Kiser et al., 2020); increased risk of ED visits for respiratory and cardiovascular disease, with the older adults especially susceptible (Resnick et al., 2015); increased cardiovascular mortality on smoky days (Faustini et al., 2015), and an increase in pediatric respiratory visits (Leibel et al., 2020). Overall, studies consistently found an increase in the studied health outcomes during wildfire events in comparison to non‐wildfire periods. Chemical Constituent and Source Five studies (Basagaña et al., 2015; Colonna et al., 2023; Joshi et al., 2022; Kim et al., 2012; Lu et al., 2023) considered the effect of heterogeneous PM composition on health risk. Three of these studies considered chemical components (Basagaña et al., 2015; Joshi et al., 2022; Lu et al., 2023) while two studies considered source apportionment PM by positive matrix factorization (PMF, Colonna et al., 2023; Berger et al., 2018). Considering components, in India, Joshi et al. (2022) found nitrate, ammonium, chromium, and both elemental and organic carbon showed a higher mortality impact than the total PM2.5 mass, with the above 65 age group most vulnerable. In Denver, USA Kim et al. (2012) found elemental and organic carbon generally were associated with a higher immediate risk for cardiovascular and respiratory diseases, with other species such as sulfate and nitrate being associated with higher risk for longer lags. In Pakistan, Lu et al. (2023) found the relationship to exposure to nickel in PM2.5 (5%–14% increase per interquartile range) and cardiovascular hospital admissions was most significant, which was thought to come from fossil‐fuels combustion and industrial emissions, and to a lesser extent elements aluminum, iron, and titanium which were thought to come from crustal sources, indicating anthropogenic produced PM was more harmful. Considering PM sources, dust, as opposed to traffic and regional sources, had the higher respiratory disease hospitalization risk in Kuwait (Colonna et al., 2023). Ostro et al. (2016) found vehicular emissions uniquely associated with all cardiovascular emergency department visits, while vehicular emissions, biomass burning, and soil sources were associated with all respiratory emergency department visits, asthma especially, although the soil source (including road dust) had the highest risk estimate for asthma. Discussion Summary of Main Findings This scoping review investigated literature on the cardiovascular and respiratory health outcomes related to PM exposure in semi‐arid and arid regions across the world, using a meta‐analysis to produce pooled risk ratios analysis of heterogeneity between effect modifiers. We found statistically non‐significant evidence of increased risk for males versus females, and for adults versus older adults. PM2.5 exposure in BWh regions was found to have higher risks for both cardiovascular and respiratory disease compared to semi‐arid regions. Findings between high‐ and low‐density cities across health outcomes and PM exposure types were generally statistically non‐significant, except when stratified by climate where in BWh regions there was a higher risk estimates for respiratory morbidity high‐density cities, and conversely and lower risk in BSk climates. Our review also partially serves as an update to the literature regarding short term PM exposure and differential susceptibility and vulnerability among populations, for which the most recent work was published in 2013 (Bell et al., 2013). The geographic focus of studies is strongly focused on China, the US, and Iran. Relatively high numbers of studies focused on PM exposure in China and the US have been identified in relevant previous studies (Bell et al., 2013; Yang et al., 2019; Yee et al., 2021), however never before in regards to arid regions. Of note is the complete lack of studies spanning the entire continent of Africa and the country and continent of Australia, despite the fact that these two continents combined contain more than half of global arid and semi‐arid regions (Prăvălie, 2016). In the case of Australia, 87% of the population lives 50 km from the coast (Australian Bureau of Statistics, 2020), which is mostly not arid. However, this is not the case in Africa, with some of the continents' largest cities such as Cairo (Egypt), Luanda (Angola), Khartoum (Sudan), and Dhakar (Senegal) are characterized by an arid or semi‐arid climate. Due to the criteria of this scoping review, the absence of studies in Africa may be partially due to data availability in many low and middle income countries, which is primarily in the form of surveyed health outcomes (such as the Demographic and Health Survey data) or at the national level as opposed to objectively measured hospitalization or mortality data (WHO, 2021a; Zhao et al., 2022). This corroborates previous assessments finding a dearth of studies in low‐ and middle‐income countries, and an over emphasis in high income countries where PM exposure is comparably lower (Lim et al., 2022). We conclude there is a gap in the representation of arid region low income and middle‐income countries in short‐term exposure related health outcome studies. Climate Regions Our meta‐analysis finds the highest respiratory and cardiovascular morbidity risk from exposure to PM2.5 in the most arid (BWk) regions relative to semi‐arid regions (Figure 4). While there is an absence of literature specifically assessing differential health impacts between arid and semi‐arid regions, many studies have investigated the variation in outcomes between multiple cities, in both arid and non‐arid climate regions (Franchini & Mannucci, 2007; Liu et al., 2022c; Odo et al., 2022; Yin et al., 2017) with no systematic findings that semi‐arid or arid cities have a higher total or disease specific mortality risk than non‐arid cities. However, several studies we reviewed report higher risks related to PM exposure during specific events (i.e., wildfires and dust storms) or the synergistic effects with other environmental exposures (i.e., PM and heat) pointing toward the need for additional research climate‐informed PM sources and risk. The results of this study show a slightly higher risk for cardiovascular and respiratory morbidity and mortality from PM2.5 exposure in arid climates compared to both hot and cold semi‐arid climates, which could be due to sociodemographic variables (e.g., access to air conditioning, workplace norms and location, acclimatization), compounding environmental exposures (e.g., humid heat and PM), or climate specific differences in PM source type and climate dynamics (e.g., dust storms). For example, populations in hotter, drier regions could experience enhanced dehydration or the presence of humid heat in semi‐arid regions, which have more precipitation, can hamper the body's response to heat, both increasing strain on the cardiovascular system and the risk of adverse events in the presence of PM (Ebi et al., 2021). Further, the majority of studies focused on outcomes in cold, semi‐arid regions, leading to an unequal distribution of data across climate types and temperature ranges and extremes. Comparison forest plots for cardiovascular and respiratory outcomes from PM2.5 exposure (a) arid, hot (BWh) regions, and (b) in semi‐arid, cold regions. The pooled relative risk (RR) uses a restricted maximum likelihood random effects model. Considering that half a billion people live in areas that are projected to become drylands in the 21st century in a 4°C global warming levels (GWLs) scenario (Spinoni et al., 2021) the authors of this study feel there is significant need to better constrain the impact of climate on PM exposure and health outcomes. To our knowledge, prior to this study there has only been one previous review that considered climate specific short‐term PM exposure effect estimates, which focused on cold climates (Wine et al., 2022). Wine et al. (2022) highlight that unique mechanisms of exposure due to temperature and other climate variables are poorly understood. Temperature alone has been found to exacerbate respiratory morbidity risk (Areal et al., 2022) though the synergistic relationship with PM exposure is poorly understood. This is especially of concern in arid and semi‐arid regions, some of which are projected to surpass human tolerability in the 21st century (Raymond et al., 2020), putting vulnerable populations at an even higher risk in the context of climate change. In particular, the impact of climate and human development related destabilization of land surfaces in arid and semi‐arid regions leading to more PM emission from various sources is an understudied topic (Safriel et al., 2005). Overview of Meta‐Analysis and Effect Modification Our meta‐analysis found a slight increase in the pooled relative risk of respiratory and cardiovascular morbidity and mortality across all populations included in our scoping review (Figure 3 and Figures S2, S3 in Supporting Information S1), with the greatest increase in risk for respiratory (4%; 4%–5%) and cardiovascular disease (3%; 2%–5%) related to PM2.5 exposure. We also explored a disease specific sensitivity analysis (Table S2 in Supporting Information S1). Considering specific respiratory diseases and PM2.5 exposure, pneumonia (RR = 1.10, 95% CI = 1.06, 1.14), bronchiolitis and bronchitis (RR = 1.09, 95% CI = 1.06, 1.12), asthma (RR = 1.05, 95% CI = 1.03, 1.06), and acute lower respiratory tract infection (ALRI) (RR = 1.07, 95% CI = 1.05, 1.08) had considerably higher risk compared to estimates reporting all respiratory morbidity diseases combined (RR = 1.02, 95% CI = 1.01, 1.02). For cardiovascular disease, all cerebrovascular disease (stroke) morbidity (RR = 1.03, 95% CI = 0.99, 1.08), coronary ischemic events (RR = 1.05, 95% CI = 1.02, 1.08), and heart failure (RR = 1.02, 95% CI = 1.01, 1.03) risk was higher than all‐cardiovascular disease (ICD‐10: I00‐I99, RR = 1.01, 95% CI = 1.00, 1.02). Four effect modifiers were included in multiple reviewed studies (age, sex, season, population density) and analyzed in our meta‐analysis. Both season (cold vs. warm) and population density (low vs. high) had relatively inconsistent and weak associations across all studied health outcomes and PM exposures. While sociodemographic characteristics, such as those with lower socioeconomic status and in minority populations have been shown to increase the risk for adverse health outcomes in relation to PM exposure (Bell et al., 2013; Geldsetzer et al., 2024; Ma et al., 2023), only one identified study considered SES and four considered race/ethnicity. This may be due to our study inclusion criteria, as SES and race/ethnicity are often studied at a national level (e.g., Jbaily et al., 2022; Ma et al., 2023), and thus, these publications were not included based on our criteria, despite the importance and need to assess differential health impacts across such sub‐populations. Regarding age–which was the most studied effect modifier–we observed a 9% increase in the risk of respiratory morbidity as a result of PM2.5 exposure among the adult population (those aged 18 to 64). Children (0–18 years) experienced an increase in risk of 6% (95% CI: 4%–8%), whereas older adults (ages 65 and up) experienced no increased or decreased risk in these regions. Though limited by a small number of studies (n = 8), the heterogeneity in risk between adults and older adults (RRR = 1.06%) is significant. Our findings are at odds with previous studies (e.g., Bell et al., 2013; Karimi & Samadi, 2024), which have found that older adults have a statistically higher cardiorespiratory mortality and morbidity risk from PM exposure. This could be potentially due to inconsistency in our results could be potentially due to sampling bias, or a different population composition in semi‐arid and arid regions. Further, the adult population could be exposed to higher levels of PM within the workplace, potentially pointing to an important sub‐population to consider in these regions. As for effect modification by sex, pooled relative risk estimates indicate a general trend of higher risks for males compared to females, but we did not find a statistically significant difference when applying heterogeneity tests (e.g., PM2.5 exposure respiratory RRR = 1.00, I 2 = 7%). As more thoroughly addressed elsewhere (Clougherty, 2010), discussion of a binary sex‐based effect modification is imprecise as it may be influenced by biological factors (e.g., lung size or hormonal influences on chemical transport) or societal differences in lifestyle (e.g., gendered occupations or behaviors such as smoking), leading to vastly different conclusions. Therefore, to investigate heterogeneity in risk based on differential exposure, or separately, differential susceptibility, by different genders or sexes, more specific geographically and socially qualified data is needed to reveal trends that may support activities to mitigate risks. Unique Sources of PM Multiple studies note the lack of information concerning the chemical composition of PM a weakness in results (e.g., Al‐Taiar & Thalib, 2014; Vodonos et al., 2015). While the composition of PM was only the focus of five studies, all found that some sources had greater association with health outcomes than others (see Section 3.3). Previous literature reviews concerning source‐specific PM exposure on long‐term (Wyzga & Rohr, 2015) and short‐term (Xu et., 2022) health outcomes found this focus a growing trend in global studies, with significant potential to address uncertainties regarding the most harmful PM and associated variability in health outcomes. While the identification of PM sources using chemical composition is a well‐established field of research (Colucci & Begeman, 1965; Karagulian et al., 2015), its application to qualifying PM exposure risk has recently been gaining more attention as applied to pediatric health outcomes (Huang et al., 2019) and regarding cause specific mortality (Berger et al., 2018). In some regions using chemical compositions to study PM source and associated health impacts is more accessible through the availability of national speciated PM monitoring networks in the US (Chemical Speciation Network; US EPA, 2023) and China (National Aerosol Composition Monitoring Network (NACMON); Dao et al., 2019). A related research trend concerns health risk from exposure to PM from unique events, such as dust storms (Lian et al., 2025) and wildfires (Aguilera et al., 2021; Wang et al., 2025). Thus far, there is inconclusive evidence that dust storms are associated with increased cardiovascular and respiratory health risks (see Section 3.2.6). The PM generated by dust storms may be less toxic than anthropogenic PM (Vodonos et al., 2015), although this conclusion is highly variable by location and possibly influenced by decreased healthcare utilization during dust storms. City size, frequency of storms, and composition of dust may influence PM health effects (Shahsavani et al., 2020), which we were unable to study in this review. Dust storms—which are associated with increased mortality (Zhang et al., 2016)—are an issue in arid regions, downwind environments, major urban centers (Goudie, 2014), as well as in non‐arid regions such as Korea (Lee et al., 2013) and Europe (Khaniabadi et al., 2017). Studies that met our criteria were all focused on urban populations, leaving cardiorespiratory health risks in rural communities unrepresented. Urban and rural areas differ markedly in PM sources and composition: urban settings are dominated by vehicle emissions, secondary traffic‐related pollutants, and industrial sources (Karagulian et al., 2015), whereas rural populations face greater exposure to agricultural burning, unpaved‐road dust, soil and mineral aerosols, wildfire smoke, and windblown dust events (Borghi et al., 2023; Cleland et al., 2025; Kundu & Stone, 2014; Reid et al., 2016; Tong et al., 2023). These contrasts contribute to differences in toxicity and PM‐induced biological effects between rural and urban PM mixtures, with studies reporting variation in toxicity, oxidative potential and mortality associations between rural and urban populations (Liu, Clark, et al., 2021; Mirowsky et al., 2013; Perrone et al., 2013; Singh et al., 2026; Wang et al., 2022), likely reflecting the large variability and heterogeneity in sources. In the United States, urban tracts exhibit higher PM2.5 concentrations overall, with strong racialized disparities in exposure across both urban and rural contexts (Kilpatrick et al., 2024), underscoring the importance of evaluating both social and geographic variations in exposure. Together, these findings indicate that PM‐related health risks differ across environments due to variation in composition, toxicity, and exposure dynamics, highlighting the need for more research on rural PM exposure and its distinct health impacts. All the studies reviewed that considered wildfires found increased relative risk during wildfire exposure, or increased risk in comparison to non‐wildfire periods (see Section 3.2.6). Wildfire smoke particles have been identified as being uniquely harmful for respiratory health compared to other PM (Aguilera et al., 2021), with synergistic adverse health effects with heat (Chen et al., 2023; Cleland et al., 2025; Jones‐Ngo et al., 2025) which is a concern as compound exposures are increasingly occurring in many regions due to climate change. While wildfire and dust storm exposures are not unique to arid regions–though the production of dust storms is a unique phenomenon in drylands–the unique interaction between these events and regional climates is a subject requiring greater attention. Climate change is expected to intensify extreme temperatures, and increase the frequency of droughts, wildfires, dust storms, flooding, storm surges, and hurricanes (Khraishah et al., 2022). Given the potential for compounding and exacerbated climate‐driven hazards, it is critical to understand current and future impacts of climate change on human health in different regions. Future Directions Air pollution does not abide by political borders (Schwarz et al., 2023). It is capable of traveling long distances from the primary emission source, and often disproportionately impacts those who are not responsible for the majority of emissions (Tessum et al., 2019). In review of the current literature, we identified three main gaps and potential focus areas for future research to elucidate the differential relationships that exist between PM exposure and health: rural populations, comparisons between climate types, and source specific PM risk estimates. While the vast majority of PM exposure‐related health studies are in dense urban environments, there is a need to understand this relationship in rural areas as the sources and composition of PM are different, as is the population exposed. For example, the impact of dust on human health may be exacerbated in cropland regions, areas which are understudied if not entirely left out of the literature (Guan et al., 2016). Beyond more research on the constituents and sources of PM, there is a dire need for epidemiological studies quantifying the specific relationship between health outcomes and populations at a local scale. Specifically, this is needed in understudied arid and semi‐arid regions, but also for unique populations with sociodemographic characteristics that dictate their exposure to PM and susceptibility to subsequent negative health impacts. Further, such research is crucial to project future health impacts due to increased desertification and heat from climate change (Marlier et al., 2025) and design effective policies and interventions to address this. Limitations While this work contributes to a gap in the literature on health impacts from PM exposure in arid and semi‐arid regions and how that relationship differs across sub‐populations and climate, there are some limitations to this study that must be noted. Due to the exclusion criteria of our scoping review, there were studies assessing the relationship of interest that were not written in English or using subjectively measured health outcomes (e.g., surveys) that were not eligible. Further, many national or global studies exist with robust data that analyze arid region PM health risks, however as they only reported pooled estimates, we were unable to utilize their findings in our review. As a result, estimates of the relationship of interest exist for regions and populations not represented in our findings. These criteria were chosen to make the research feasible for the authors to complete and to enable direct comparisons between study findings but may have led to the exclusion of relevant studies in some regions, notably South America, Central Asia, the Middle East, and both sub‐Saharan and Northern Africa, leading to incomplete or skewed conclusions. Additionally, the comparison of estimates from different regions, cities, and years within the 12‐year period without equivalent studies in comparator regions may have led to inaccurate conclusions regarding the effect of geography on health risks. The notable absence of studies in Africa represents a major limitation in our study's ability to estimate risk in these climate regions and may have led to an under or overestimation of global risk. Another potential source of bias is the unequal number of studies in each subgroup comparison, which ideally would be balanced to ensure the statistical methods used yield robust and unbiased estimates. We chose to focus solely on same day exposure to PM, however, as acknowledged in the literature, we suggest future work to examine lag structures in these regions given the potential to capture important delayed or cumulative effects in distributed lag patterns. Lastly, we were unable to fully capture all adverse health outcomes related to PM exposure as our focus was solely on cardiovascular and respiratory related outcomes. Conclusions Through this scoping review, we identified 102 studies across four continents that quantified the relationship between exposure to PM and cardiovascular and respiratory morbidity and mortality in semi‐arid and arid regions of the world. While the number of studies published per year generally increased over this interval, we identified significant gaps in both geographical representativity and quantitative estimation of the relationship between climate and health risks from PM exposure, especially concerning vulnerable subpopulations (e.g., migrants, lower socioeconomic status, and minority groups). The link between PM and morbidity and mortality is well established, although understanding concerning the direct relationship between specific sources of PM and health outcomes is underexplored. Through our meta‐analysis of reported effect estimates, we found that across populations exposure to PM2.5 and PM10 resulted in increased risk of respiratory and cardiovascular morbidity and mortality. There was weak evidence of effect measure modification across subgroups, including sex, age, seasonality and aridity. Notably, pooled relative risk for cardiovascular and respiratory morbidity was greater in arid regions in comparison to semi‐arid regions. Although limited, our findings highlight the existence of variation across populations and places, and the need for additional measurement of this relationship, especially in areas with limited air quality monitoring networks or a lack of existence in the literature. Given the co‐occurrence of climate change induced environmental extremes and continued anthropogenic PM emissions, we urge future work to consider compounding health risks and climate synergies across diverse populations. Conflict of Interest The authors declare no conflicts of interest relevant to this study. Supporting information Supporting Information S1 Table S1