Air pollution exposure and adverse sleep health across the life course: A systematic review (2024)

The publisher's final edited version of this article is available at Environ Pollut

2.1. Search strategy

Literature searches were conducted using PubMed, CINAHL, Cochrane, Scopus, Web of Science, and PsycInfo to identify suitable research articles published through October 2019 for this review (no publication date restriction was set). These databases were searched using the various terms for air pollution and sleep outcomes. A sample search included (“air pollut*” [Title/Abstract] OR “environment* exposure” [Title/Abstract] OR “environment* pollut*” [Title/Abstract] OR “particulate matter” [Title/Abstract] OR “PM10” [Title/Abstract] OR “PM2.5” [Title/Abstract] OR “ozone” [Title/Abstract] OR “carbon monoxide” [Title/Abstract] OR “nitrogen dioxide” [Title/Abstract] OR “sulfur dioxide” [Title/Abstract]) AND (sleep*[Title/Abstract] OR “sleep quality” [Title/Abstract] OR “sleep duration” [Title/Abstract] OR “sleep efficiency” [Title/Abstract] OR “sleep disturbance” [Title/Abstract] OR “sleep impair*” [Title/Abstract] OR “sleep disordered breathing” [Title/Abstract] OR “sleep apnea” [Title/Abstract]), Additional MeSH terms, including “Air Pollutants”, Air Pollution”, “Sleep”, “Sleep Wake Disorders”, “Actigraphy”, and “Polysomnography” were included in PubMed. The entire search strategy is provided in Appendix A.

2.2. Inclusion criteria

2.3. Study selection

Papers identified through the electronic search process were retrieved. Duplicate articles were removed, and titles and abstracts were evaluated. Full texts of relevant publications were examined based on the inclusion criteria, and references were scanned for additional articles suitable for inclusion. Duplicate screening was conducted to ensure accuracy and consistency.

2.4. Data extraction

Data from suitable publications were extracted, including year of publication, study design and location, sample characteristics of participants, air pollution and sleep measures, adjusted covariates, major findings, and study strengths and limitations. Extraction was conducted in pairs to ensure accuracy, and disagreements were resolved through discussion.

2.5. Quality assessment

Level of certainty rating for individual studies was utilized to assess the quality of included papers. A modified systematic review framework was used to rate the level of certainty for each health outcome. This framework is derived from a previously published systematic review based on established methods of systematic reviews for the medical, public health and environmental health fields (Bamber et al., 2019). These frameworks incorporate most of Bradford Hill’s criteria for causation such as studies with specificity and biological plausibility and that were temporal and consistent (Schünemann et al., 2011). These classic criteria were used to develop a meaningful scope of review and determine criteria for study certainty.

We rated study findings as having low, moderate, or high certainty that the reported result was close to that of the true effect based on the methodology established in a recent systematic review (Bamber et al., 2019). The findings were initially ranked as low certainty and were upgraded according to fourteen study evaluation questions assessing various domains. These criteria were based on established frameworks that specified the domains, questions, or study limitations used to evaluate individual studies for use in a systematic review (Guyatt et al., 2008; Higgins and Altman, 2008; Rooney et al., 2014; Wells et al., 2017; Woodruff and Sutton, 2014). We categorized the study evaluation questions into the following five categories: population and sample, exposure, health outcomes, confounders, and reporting. Two authors reviewed each study evaluation question with a yes-or-no response for each included study. Conflicting responses were resolved through discussion and additional review of the study. Studies with greater than 50% “yes” answers (i.e., 8 “yes” answers out of 14) were considered for potential upgrade of their findings to moderate certainty; and studies with greater than 75% “yes” answers (i.e., 11 “yes” answers out of 14) were considered for potential upgrade to high certainty (no study fulfilled this criterion). All findings of each study were ascribed the same level of certainty after evaluations were complete.

Table 1

Characteristics of included data-based studies on air pollution exposure and sleep outcomes in human participants, ordered by study design.

Table 2

Results, strengths, and limitations of included data-based studies on air pollution exposure and sleep outcomes in human participants, ordered by study design.

Table 3

Quality assessment scores of included studies.

3.1. Multiple Air Pollutants

While the majority of studies included in this present systematic review investigated a specific type of exposure, three large cross-sectional studies examined the general effect of ambient air pollution on sleep (Kheirandish-Gozal et al., 2014; Lawrence et al., 2018; Yu et al., 2019). Kheirandish-Gozal et al. (2014) estimated exposure to various pollutants, including inhalable particulate matter (PM₁₀), nitrogen dioxide, sulfur dioxide, carbon monoxide, and ozone, in five districts of Iran with varying levels of exposure. Among 4322 children aged 6–12 years, those living in regions with higher levels of pollutant exposure were more likely to have habitual snoring, defined as loud snoring three or more times a week. This effect remained after controlling for other risk factors, such as family history of snoring, wheezing, respiratory problems, and parental smoking status, indicating that poor air quality can lead to the development of habitual snoring and subsequently disrupt child sleep quality.

A similar study conducted on 59,754 children and adolescents aged 5–17 years (mean age 10.3 years) in China monitored exposures to the same individual pollutants as the previous study and assessed sleep quality and various sleep disorder symptoms using the Sleep Disturbance Scale for Children (SDSC) (Lawrence et al., 2018). The scale reported information on sleep-wake transition disorders (SWTD), disorders of initiating and maintaining sleep (DIMS), disorders of excessive somnolence (DOES), disorders of arousal, sleep hyperhidrosis, and sleep-breathing disorders. By stratifying the analysis by both pollutants and symptoms of sleep disorders, the authors demonstrated positive associations between all pollutants and sleep disturbances evaluated. Notably, the observed relationships were generally stronger among females than males, the strongest association was observed between PM₁ exposure and disorders of excessive somnolence, and exposure to PM₁ resulted in the greatest risk for sleep disturbances among all other pollutants.

Another study conducted in China assessed the association between ambient pollutants and sleep duration in college freshmen. Air pollution exposure was estimated by city air monitoring stations, and measures included air quality index and PM_2.5, PM₁₀, and NO₂ exposure. Sleep duration was assessed using the Chinese version of the Pittsburgh Sleep Quality Index (CPSQI). Across five cohorts with a total of 31,582 participants, exposure to higher concentration of air pollutants was associated with shorter sleep duration.

3.2. Particulate matter

One major component of air pollution is particulate matter (PM), composed of both solid and liquid particles found in the air and classified by aerodynamic diameter. The smaller sized particles, PM_2.5, have diameters of 2.5 μm or less and pose the greatest risk to health (EPA). Similarly, exposure to larger particles, such as PM₁₀, has also been shown to affect sleep.

Eight studies have examined the effects of general particulate matter, using a variety of methods to measure air pollution exposure, on sleep disturbances across the life course. In a longitudinal study on 397 mother-child pairs (mean age 27.7 and 4.8 years respectively) in Mexico, Bose et al. (2019) reported that prenatal maternal exposure to PM_2.5 has differing effects on later child sleep patterns depending on the gestational phase during which exposure occurs. Daily PM_2.5 exposure for each mother-child pair was individualized using spatiotemporal models based on the home address of the participants. By averaging these daily values over each week of gestation and assessing child sleep via actigraphy, the authors were able to identify windows of susceptibility during which the fetus is especially sensitive to air pollution exposure. While PM_2.5 exposure early in gestation (weeks 1–8) was negatively associated with child sleep efficiency, greater exposure to particulate matter with diameters of 2.5 μm or less later in the gestational period (weeks 31–53) was associated with shorter child sleep duration.

Furthermore, a similar cross-sectional study of 276 children (mean age 9.26 years) from four school districts estimated PM₁₀ exposure levels using the monitoring stations closest to each school (Abou-Khadra, 2013). Despite the highly variable levels of exposure across the districts, the author found significant positive associations between PM₁₀ exposure and both sleep hyperhidrosis and disorders of initiating and maintaining sleep.

Conversely, a large prospective cohort study conducted on 14,110 freshmen (mean age 18 years) at a Chinese university by An and Yu (2018) demonstrated that greater PM_2.5 exposure in young adults resulted in longer sleep duration both during the day and at night. This association remained significant after adjusting for covariates such as age, BMI, current smoking and drinking status, and self-assessed physical and mental health, and there was no difference in effect between males and females. However, despite the longitudinal design and large sample size, sleep measures were obtained from self-report questionnaires, possibly resulting in response bias, and the study population of young university students is not representative of the larger population, leading to a lack of generalizability of the results.

Chuang et al. (2018) estimated occupational PM_2.5 exposure in Taiwanese welding workers and office workers (mean age 46.2 years) using personal air sampling sensors. Welding workers were observed to have greater wake times during sleep, as measured by actigraphy, due to increased exposure to PM_2.5 pollutants in metal fumes.

Four cross-sectional studies on large adult populations examined the specific effect of particulate matter exposure on sleep disordered breathing (SDB) (Billings et al., 2019; Lappharat et al., 2018; Shen et al., 2018; Zanobetti et al., 2010). Using estimates of air pollutant exposure from monitoring station data, Shen et al. (2018) reported a positive association between both PM_2.5 and nitrogen dioxide exposure on SDB, measured by the apnea-hypopnea index (AHI) and oxygen desaturation index (ODI) via polysomnography, in their population of 4312 Taiwanese adults (mean age 45.8 years). These relationships were observed for both short-term (daily mean exposure estimates) and long-term exposures (annual mean exposure estimates) and were the most significant in the spring and winter seasons. A similar study by Billings et al. (2019) on an older population of 1974 US participants (mean age 68 years) confirmed the association between PM_2.5 and nitrogen dioxide exposure and SDB. Use of spatiotemporal models allowed Billings et al. (2019) to estimate individual participant exposure levels based on home address, resulting in more accurate measures of air pollution exposure. After adjusting for demographics, other comorbidities, socioeconomic status (SES), and study site, greater exposure to both PM_2.5 and nitrogen dioxide increased the risk of sleep apnea, with PM_2.5 having a stronger effect.

Furthermore, in the US, sleep in a subset of the Sleep Heart Health Study (SHHS) cohort consisting of 6441 adults over the age of 39 (mean age 63 years) was assessed via polysomnography. PM₁₀ exposure in the summer was positively associated with SDB, as measured by the respiratory disturbance index (RDI) and sleep time spent in hypoxia, and negatively associated with sleep efficiency (Zanobetti et al., 2010). Higher temperatures were also observed to be associated with higher RDI scores across all seasons. A similar study conducted in Thailand examined bedroom environmental conditions and reported that increases in PM₁₀ exposure was correlated with increased severity of obstructive sleep apnea, a common subtype of SDB, using measures of AHI, RDI, and hypoxia (Lappharat et al., 2018). However, despite these findings, two similar studies conducted in adults failed to demonstrate a relationship between particulate matter exposure and sleep disordered breathing (Cassol et al., 2012; Weinreich et al., 2015).

While most studies only investigated general particulate matter in relation to sleep outcomes, few additional studies have examined a specific particulate matter, namely ozone, in relation to sleep health. As a common air pollutant, ozone is affected by various weather-related factors such as rises in ambient temperature and relative humidity (EPA, 2018). Thus, a discussion regarding ozone, temperature, and humidity is provided in the additional following section.

3.3. Ozone, temperature, and humidity

Five studies have identified associations between ozone, temperature, and humidity and SDB in particular (Cassol et al., 2012; Cheng et al., 2019; Sánchez et al., 2019; Weinreich et al., 2015; Yıldız Gülhan et al., 2019). In a cross-sectional study on 564 children aged 5–9 (median age 6 years), Sánchez et al. (2019) estimated air pollution exposure using monitoring stations located near the participants’ schools and obtained data on sleep-related respiratory symptoms via parental self-report using the Pediatric Sleep Questionnaire (PSQ). Greater exposure to both ozone and sulfur dioxide and higher humidity levels increased the risk of wheezing-related sleep disturbances. Similarly, Weinreich et al. (2015) examined the relationships between PM₁₀, temperature, ozone levels, and relative humidity and SDB. Using AHI to measure SDB, the authors reported positive associations between AHI and both temperature and ozone levels. These relationships were strongest in warm weather and remained after adjusting for covariates. Conversely, Yıldız Gülhan et al. (2019) demonstrated that AHI levels were higher during winter months. Using polysomnography and air monitoring stations to estimate exposure, relative humidity was identified to be positively associated with rapid eye movement-related AHI.

Two similar studies examined the relationship between ambient pollutants and sleep outcomes in non-population based participants (Cassol et al., 2012; Cheng et al., 2019). Measured exposures included particulate matter, carbon monoxide, sulfur dioxide, ozone, temperature, and relative humidity for both. Similarly, analysis of retrospective polysomnographic data was used to assess AHI among all participants. By obtaining patient polysomnography data taken across all seasons, Cassol et al. (2012) observed that obstructive sleep apnea (OSA) severity varied with seasonality and was positively associated with carbon monoxide and relative humidity but negatively associated with temperature. Similarly, Cheng et al. (2019) reported that a positive association between AHI and PM₁₀, ozone, sulfur dioxide, and relative humidity was only seen in those with severe OSA during non-REM sleep.

Despite the large sample sizes of these studies, these contradicting findings, particularly for the relationship between sleep outcomes and temperature, could be due to various limitations. Firstly, pollutant exposure was measured by monitoring stations nearest to the hospital and did not account for participants’ home location or indoor exposures. Most notably, the study participants were drawn from people referred to the hospitals for a diagnostic PSG due to the possible presence of sleep disorders. However, a randomized control trial on OSA patients has demonstrated that although lower bedroom temperatures was associated with longer sleep duration, higher sleep efficiency, and increased alertness during the day, AHI was higher at lower temperatures (Valham et al., 2012). Thus, this emphasizes the complex relationship between temperature and sleep outcomes, the mechanism of which is not yet understood.

3.4. Traffic-related

Air pollution resulting from road traffic has been commonly identified as a health hazard. Common markers of traffic-related pollution are black carbon, a PM_2.5 component (Fang et al., 2015), and NO₂ (Martens et al., 2018) exposure. Using self-reported data on traffic exposure collected from a large European population (n = 12,184, mean age 51.5 years), Gislason et al. (2016) demonstrated a significant risk of daytime sleepiness due to high levels of Traffic-related pollutant exposure. In an additional study conducted in the US, exposure to black carbon was estimated using spatiotemporal models and participant home location (Fang et al., 2015). Self-report questionnaires were used to assess for sleep duration, sleep latency, and sleep apnea, and increased black carbon exposure was found to decrease sleep duration in males and those of low SES but increase sleep duration in African Americans. Surprisingly, exposure to black carbon was not associated with sleep latency or sleep apnea. This may be due to the lack of objective sleep measures considering the large sample size (n = 3821) and individualized air pollution estimates. A similar study, based in the Netherlands, also used spatiotemporal models to estimate individual-level exposure to NO₂ (Martens et al., 2018). The authors demonstrated that overall sleep quality and incidences of sleep disturbances were not only related the modeled air pollution exposure, but also to self-reported perceived exposure to traffic-related pollutants.

3.5. Indoor air quality

In addition to ambient exposures, three studies have focused specifically on the effects of indoor air quality, specifically examining the relationship between pollutants due to cooking and sleep in both children and adults. Two intervention studies conducted by one research team on Peruvian children under the age of 14 reduced household exposure to biomass pollution by replacing highly polluting stoves with reduced polluting Inkawasi stoves (Accinelli et al., 2014; Castañeda et al., 2013). Using parent reports of child sleep habits, the authors reported improvement in SDB-related symptoms, such as snoring and nighttime awakening, with decreased biomass pollution. Further, lower levels of pollutant exposure were associated with increased willingness to sleep and ease of falling asleep and waking up for both exclusive and partial use of the improved stoves. In adults, a cross-sectional study conducted on 2197 Chinese adults (mean age 37.52 years) investigated the relationship between cooking oil fume (COF) exposure and sleep (Wei et al., 2017). Cooking practices and sleep patterns were measured using self-report questionnaires. Additionally, urine samples were collected and analyzed for 1-hydroxypyrene (1-HOP), a urinary biomarker of polycyclic aromatic hydrocarbons in the COFs. Both subjectively reported and objectively measured exposure were positively associated with poor sleep quality. Furthermore, subjective report of COF exposure increased the risk for long sleep onset latency, daytime dysfunction, and sleep disturbances.

4.1. Children and adolescents

More than 10% of school-aged children and adolescents are reported to experience sleep problems (Stein et al., 2001). The published literature examined in this review reports a negative association between sleep quality and exposure to pollutants in this population. Because of their developing nervous and immune systems, children have been shown to be more susceptible to the effects of air pollution (Sánchez et al., 2019). This vulnerability begins in utero, as lower sleep efficiency and shorter sleep duration at a preschooler age has been linked to prenatal exposures to PM_2.5 at different stages of gestation (Bose et al., 2019). Despite the indirect nature of this exposure, these results suggest that air pollution can have long-lasting effects on sleep quality.

Similarly, respiratory-related sleep disturbances have been observed to be associated with air pollution exposure. Higher levels of ambient air pollution are positively associated with a wide range of disturbances that affect children’s sleep quality, including habitual snoring, wheezing, and sleep disorders symptoms, with overall stronger relationships seen in females compared to males. Notably, sulfur dioxide and ozone levels were related to habitual snoring in children and increased the risk of wheezing-related sleep disturbances (Kheirandish-Gozal et al., 2014; Sánchez et al., 2019), and lower levels of CO₂ and PM_2.5 exposure were associated with improved SDB-related symptoms (Castañeda et al., 2013). Generally, children are more prone to these disturbances than adults, possibly due to inhaling larger volumes of air per body weight on average and an increased permeability of the airway epithelium (Sánchez et al., 2019). These physiological factors result in greater relative exposure to pollutants that remain in the airways for longer periods of time, likely inducing more severe effects on sleep in children than adults.

Apart from respiratory-related sleep disturbances, several studies have also observed associations between particulate matter and a wide range of sleep disturbances (Lawrence et al., 2018). In young children, particulate matter PM_2.5 and above are associated with sleep hyperhidrosis, initiating/maintaining sleep, wheezing- and snoring-related sleep disturbances, and nighttime awakenings. In adolescents, higher levels of PM_2.5 exposure are associated with longer daytime and nighttime sleep durations.

Inconsistencies in these findings could be due to the varying participant populations assessed or the limitations in study design. All sleep measures relied on self-report from the child’s parents, rather than objective methods, and may be subject to recall bias. Additionally, the questionnaires used also differed between studies. Further, exposure estimates only included either ambient pollutants or indoor pollution and did not account for exposure from the other environment or from other sources, such as noise and diet. Thus, future studies should account for these additional confounders and use objective sleep measures to further investigate the association between sleep disturbances and air pollution exposure in children and adolescents.

4.2. Adults

It has been shown that sleep disorders are more likely to occur in older adults than a younger population (Neubauer, 1999). All except one reviewed study pertaining to this population consistently reported the association between disruptions in sleep quality and air pollution exposure. Although positive associations between sleep-disordered breathing (SDB) with PM_2.5 and PM₁₀ exposure in older adults had been reported in some studies (Billings et al., 2019; Lappharat et al., 2018; Shen et al., 2018; Zanobetti et al., 2010), but not in others (Cassol et al., 2012; Weinreich et al., 2015). Therefore, the relationships between air pollution exposure and SBD remain uncertain. Published studies also revealed associations between overall sleep quality/disturbances (increased daytime sleepiness and decreased sleep efficiency/duration) and air pollution have also been found.

Altogether, nine of the fifteen relevant studies used objective sleep measures, namely actigraphy and polysomnography, while the remainder assessed sleep duration, quality, and sleep disturbances with various self-report questionnaires. Additionally, exposure to air pollution was measured using differing methods across the literature. These diverse study designs led to varying findings of the specific relationships between pollutant exposure and sleep. For example, while increased sleep disturbances and daytime dysfunction were correlated with greater exposure to subjectively appraised traffic-related pollution (Gislason et al., 2016; Martens et al., 2018), sleep duration was associated with general air pollution exposure in Chinese university students (Yu et al., 2019) as well as black carbon exposure in males, those of lower socioeconomic status, and those of African American descent (Fang et al., 2015). Further, amount of time spent awake during time in bed was positively associated with exposure to particulate matter (Chuang et al., 2018).

Notably, contrary to related research (Bose et al., 2019; Fang et al., 2015; Scinicariello et al., 2017), An and Yu demonstrated a positive relationship between sleep duration and exposure to air pollution (An and Yu, 2018). This discrepancy could result from the use of different methods to measure exposures, as the present study estimated PM_2.5 levels via general monitoring stations while others have spatiotemporal models (Bose et al., 2019; Fang et al., 2015) to assess pollutant levels. Further, these other studies were conducted on significantly younger (Bose et al., 2019) or older (Fang et al., 2015) participants, suggesting that these findings may be age dependent. However, this relationship cannot be concluded due to the absence of similar studies.

4.3. Ambient vs. indoor exposures

Interestingly, most publications examining the associations between air pollution and sleep have been focused on ambient exposure, with double the number of available studies as compared to those assessing indoor exposures. However, these articles display diversity in methodology, resulting in more variation in study design, measures, and results reported. The majority acquired air pollution exposure data from standard government- or city-regulated monitoring stations which does not account for differences in personal exposure due to residence location, indoor pollutants, etc.

Conversely, studies specifically examining the relationship between exposure to indoor pollutants and sleep outcome demonstrate comparatively more consistent methodology. All but one assessed residential exposure, and most utilized personal air monitoring sensors to allow for individualized pollution metrics.

Regardless of the source of air pollution examined, future studies ought to account for all personal exposures including potential residential/occupational exposure as well as encounters with ambient pollutants. Use of portable individual monitoring devices may be effective in considering all pollutants encountered by an individual. On the other hand, it may not be feasible for long-term exposure assessment.

4.4. Mechanisms

Mechanisms explaining the effect of air pollution on sleep are not fully understood and have only been studied minimally. Additionally, the pathway by which exposure to pollutants impacts sleep may vary across the developmental stage. Specifically, prenatal exposure may have direct effects on the fetus by passing through the placental barrier or indirect effects via affecting maternal health (Glinianaia et al., 2004). However, initial evidence suggests two general potential mechanisms which include the biochemical effects of pollutants on the central nervous system’s regulation of sleep and changes in the physiology of the respiratory system.

The biochemistry of the central nervous system may be directly affected by air pollutants via the olfactory nerve (Abou-Khadra, 2013; Billings et al., 2019; Shen et al., 2018; Zanobetti et al., 2010), resulting in altered expression and dysregulation of neurochemicals. For example, PM_2.5 exposure was observed to be associated with lower serotonin levels, an important chemical in modulating wakefulness and circadian rhythms, implying a possible direct effect of air pollution on increased sleepiness and sleep disturbances (Chuang et al., 2018). Relatedly, a study conducted in rats observed that exposure to ozone alters the expression of 5-hydroxy-indole-acetic acid (5-HIAA), the main metabolite of serotonin. These changes were notably observed in the dorsal raphe and the hypothalamic medial preoptic area, two brain structures involved in sleep regulation (González-Piña and Alfaro-Rodriguez, 2003). Furthermore, in children, the impact of exposure on highly vulnerable nervous system structures may disrupt normal brain development and impair typical nervous system functioning, including sleep (Sears and Zierold, 2017). Specifically, exposure of the vulnerable brain to pollutants may cause irritation and breakdown of protective epithelial barriers, resulting in inflammation, oxidative stress, and degeneration of neural tissue. This damage to nerve cells could likely affect behaviors regulated by the brain, including sleep (Brockmeyer and D’Angiulli, 2016). Thus, the disturbances caused by air pollution may have a wide range of effects on sleep outcomes.

Another possible model for the relationship between pollution exposure and sleep may arise from effects on the physiology of the respiratory system. Generally, air pollutants, especially small particulate matter, are thought to deposit particles in the airways, leading to cell damage (Khafaie et al., 2016). In general, injury to respiratory cells results in disturbances in respiration by causing inflammation or edema of mucous membranes. Specifically, in the upper airways, this creates increased restriction and obstruction of normal airflow, increasing the risk of apnea and hypoxia, and thereby disrupting sleep (Abou-Khadra, 2013; Billings et al., 2019; Scinicariello et al., 2017; Shen et al., 2018; Weinreich et al., 2015). Similarly, the presence of foreign particles in the airways may result in irritation and infection, compromising sleep quality (Billings et al., 2019; Sears and Zierold, 2017; Wei et al., 2017).

4.5. Implications for future research

Despite the existing volume of research on correlations between pollutants and sleep disturbances, the examined literature demonstrates a wide variability in methodology and results that may be addressed in future studies. Many current studies include participants across large age ranges, notably encompassing all primary and secondary school grades (Lawrence et al., 2018) or adults of all ages (Shen et al., 2018). In particular, minimal research has been conducted on prenatal exposures or adolescent and elderly populations, despite the increased importance of sleep and susceptibility to disease during these life stages. Future studies focusing on narrower ranges may be able to identify more specific effects of air pollution on sleep outcomes and contribute knowledge on how the relationship develops over the life course.

In addition, more studies using objective sleep measures, such as polysomnography and actigraphy, would allow for more reliable analysis of sleep outcomes. Similarly, measuring air pollution exposures on an individual level and accounting for other confounding pollutants, such as indoor exposures and noise pollution, would help validate and confirm present knowledge. Finally, the limited availability of causal models and mechanisms linking air pollution and sleep may be ameliorated by future longitudinal studies or investigations of possible mediating factors. Therefore, we are calling for more research that investigates the potential biological mechanisms underlying the association between air pollution and sleep outcomes in addition to more epidemiological studies.

4.6. Limitations of current review

Despite the scope of this systematic review, it has several potential limitations. Given the varying study populations, methodologies, and measures used by the reviewed studies, overall causal conclusions cannot be drawn. Furthermore, due to variation in both exposure and outcome assessments, conducting a meta-analysis was not plausible. As such, the magnitude of exposure on sleep outcomes could not be determined. Additionally, as sleep was the outcome of interest for this systematic review, other research that indirectly assessed the association between air pollution exposure and sleep were not included (Peel et al., 2011). Due to a lack of data, the present review did not examine the possible differential findings across gender or race; similarly, evaluation of literature published in non-English languages may expand upon the results discussed in this present review. Moreover, the available literature may be subject to publication bias and/or selective reporting. Importantly, the lower quality ratings of the included papers demonstrate the need for further studies to improve certainty in the associations currently reported between air pollution and sleep outcomes.

PRISMA Guidelines

Appendix A: Detailed search strategy

Supplemental Tables S1-S22

Funding

This work was supported by the National Institutes of Health R25-ES021649 and the University of Pennsylvania Center of Excellence in Environmental Toxicology P30-ES013508.

^☆This paper has been recommended for acceptance by Da Chen.

Declaration of competing interest

All authors have no conflicts of interest to report.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.envpol.2020.114263.

• Abou-Khadra MK, 2013. Association between PM 10 exposure and sleep of Egyptian school children. Sleep Breath. 17, 653–657. [PubMed] [Google Scholar]
• Accinelli RA, Llanos O, López LM, Pino MI, Bravo YA, Salinas V, Lazo M, Noda JR, Sánchez-Sierra M, Zárate L, 2014. Adherence to reduced-polluting biomass fuel stoves improves respiratory and sleep symptoms in children. BMC Pediatr. 14, 12. [PMC free article] [PubMed] [Google Scholar]
• An R, Yu H, 2018. Impact of ambient fine particulate matter air pollution on health behaviors: a longitudinal study of university students in Beijing, China. Publ. Health159, 107–115. [PubMed] [Google Scholar]
• Bamber AM, Hasanali SH, Nair AS, Watkins SM, Vigil DI, Van Dyke M, McMullin TS, Richardson K, 2019. A systematic review of the epidemiologic literature assessing health outcomes in populations living near oil and natural gas operations: study quality and future recommendations. Int. J. Environ. Res. Publ. Health16, 2123. [PMC free article] [PubMed] [Google Scholar]
• Banks S, 2007. Behavioral and physiological consequences of sleep restriction. J. Clin. Sleep Med3, 519–528. [PMC free article] [PubMed] [Google Scholar]
• Billings ME, Gold D, Szpiro A, Aaron CP, Jorgensen N, Gassett A, Leary PJ, Kaufman JD, Redline SR, 2019. The association of ambient air pollution with sleep apnea: the multi-ethnic study of atherosclerosis. Ann. Am. Thoracic Soc16, 363–370. [PMC free article] [PubMed] [Google Scholar]
• Blask DE, 2009. Melatonin, sleep disturbance and cancer risk. Sleep Med. Rev13, 257–264. [PubMed] [Google Scholar]
• Bose S, Ross KR, Rosa MJ, Chiu Y-HM, Just A, Kloog I, Wilson A, Thompson J, Svensson K, Rojo MMT, 2019. Prenatal particulate air pollution exposure and sleep disruption in preschoolers: windows of susceptibility. Environ. Int124, 329–335. [PMC free article] [PubMed] [Google Scholar]
• Brockmeyer S, D’Angiulli A, 2016. How air pollution alters brain development: the role of neuroinflammation. Transl. Neurosci7, 24–30. [PMC free article] [PubMed] [Google Scholar]
• Cassol CM, Martinez D, da Silva FABS, Fischer MK, Lenz M.d.C.S., Bós ÂJG, 2012. Is sleep apnea a winter disease?: meteorologic and sleep laboratory evidence collected over 1 decade. Chest142, 1499–150 . [PubMed] [Google Scholar]
• Castañeda JL, Kheirandish-Gozal L, Gozal D, Accinelli RA, Group, P.C.I.d.I.d.l.A.R., 2013. Effect of reductions in biomass fuel exposure on symptoms of sleep apnea in children living in the peruvian andes: a preliminary field study. Pediatr. Pulmonol48, 996–999. [PubMed] [Google Scholar]
• CDC, 2017. Short sleep duration among US adults. In: Statistics D.a (Ed.), Centers for Disease Control and Prevention. [Google Scholar]
• Cheng W-J, Liang S-J, Huang C-S, Lin C-L, Pien L-C, Hang L-W, 2019. Air pollutants are associated with obstructive sleep apnea severity in non-rapid eye movement sleep. J. Clin. Sleep Med15, 831–837. [PMC free article] [PubMed] [Google Scholar]
• Chuang H-C, Su T-Y, Chuang K-J, Hsiao T-C, Lin H-L, Hsu Y-T, Pan C-H, Lee K-Y, Ho S-C, Lai C-H, 2018. Pulmonary exposure to metal fume particulate matter cause sleep disturbances in shipyard welders. Environ. Pollut232, 523–532. [PubMed] [Google Scholar]
• Colrain IM, Trinder J, Swan GE, 2004. The impact of smoking cessation on objective and subjective markers of sleep: review, synthesis, and recommendations. Nicotine Tob. Res6, 913–925. [PubMed] [Google Scholar]
• Deleanu O, Pocora D, Mihălcuţă S, Ulmeanu R, Zaharie A, Mihălţan F, 2016. Influence of smoking on sleep and obstructive sleep apnea syndrome. Pneumologia (Bucharest, Romania)65, 28–35. [PubMed] [Google Scholar]
• EPA, Particulate Matter (PM) Basics, Particulate Matter (PM) Pollution.
• EPA, 2018. Ozone Concentrations, Report on the Environment.
• Fang SC, Schwartz J, Yang M, Yaggi HK, Bliwise DL, Araujo AB, 2015. Traffic-related air pollution and sleep in the Boston area community health survey. J. Expo. Sci. Environ. Epidemiol25, 451. [PMC free article] [PubMed] [Google Scholar]
• Franklin BA, Brook R, Pope CA III., 2015. Air pollution and cardiovascular disease. Curr. Probl. Cardiol40, 207–238. [PubMed] [Google Scholar]
• Fu P, Guo X, Cheung FMH, Yung KKL, 2019. The association between PM2. 5 exposure and neurological disorders: a systematic review and meta-analysis. Sci. Total Environ655, 1240–1248. [PubMed] [Google Scholar]
• Gislason T, Bertelsen RJ, Real FG, Sigsgaard T, Franklin KA, Lindberg E, Janson C, Arnardottir ES, Hellgren J, Benediktsdottir B, 2016. Self-reported exposure to traffic pollution in relation to daytime sleepiness and habitual snoring: a questionnaire study in seven North-European cities. Sleep Med. 24, 93–99. [PubMed] [Google Scholar]
• Glinianaia SV, Rankin J, Bell R, Pless-Mulloli T, Howel D, 2004. Particulate air pollution and fetal health: a systematic review of the epidemiologic evidence. Epidemiology15, 36–45. [PubMed] [Google Scholar]
• González-Piña R, Alfaro-Rodriguez A, 2003. Ozone exposure alters 5-hydroxy-indole-acetic acid contents in dialysates from dorsal raphe and medial preoptic area in freely moving rats. Relationships with simultaneous sleep disturbances. Chem. Biol. Interact146, 147–156. [PubMed] [Google Scholar]
• Grandner MA, 2012. Sleep duration across the lifespan: implications for health. Sleep Med. Rev16, 199. [PMC free article] [PubMed] [Google Scholar]
• Gulia KK, Kumar VM, 2018. Sleep disorders in the elderly: a growing challenge. Psychogeriatrics18, 155–165. [PubMed] [Google Scholar]
• Gump BB, Gabrikova E, Bendinskas K, Dumas AK, Palmer CD, Parsons PJ, MacKenzie JA, 2014. Low-level mercury in children: associations with sleep duration and cytokines TNF-α and IL-6. Environ. Res134, 228–232. [PMC free article] [PubMed] [Google Scholar]
• Guyatt GH, Oxman AD, Kunz R, Vist GE, Falck-Ytter Y, Schunemann HJ, 2008. What is “quality of evidence" and why is it important to clinicians?BMJ336, 995–998. [PMC free article] [PubMed] [Google Scholar]
• Higgins JP, Altman DG, 2008. Assessing Risk of Bias in Included Studies. Cochrane handbook for systematic reviews of interventions: Cochrane book series, pp. 187–241. [Google Scholar]
• Irish LA, Kline CE, Gunn HE, Buysse DJ, Hall MH, 2015. The role of sleep hygiene in promoting public health: a review of empirical evidence. Sleep Med. Rev22, 23–36. [PMC free article] [PubMed] [Google Scholar]
• Khafaie MA, Yajnik CS, Salvi SS, Ojha A, 2016. Critical review of air pollution health effects with special concern on respiratory health. J. Air Pollut. Health1, 123–136. [Google Scholar]
• Kheirandish-Gozal L, Ghalebandi M, Salehi M, Salarifar MH, Gozal D, 2014. Neighbourhood air quality and snoring in school-aged children. Eur. Respir. J43, 824–832. [PubMed] [Google Scholar]
• Kurt OK, Zhang J, Pinkerton KE, 2016. Pulmonary health effects of air pollution. Curr. Opin. Pulm. Med22, 138. [PMC free article] [PubMed] [Google Scholar]
• Lappharat S, Taneepanichskul N, Reutrakul S, Chirakalwasan N, 2018. Effects of bedroom environmental conditions on the severity of obstructive sleep apnea. J. Clin. Sleep Med14, 565–573. [PMC free article] [PubMed] [Google Scholar]
• Lawrence WR, Yang M, Zhang C, Liu R-Q, Lin S, Wang S-Q, Liu Y, Ma H, Chen D-H, Zeng X-W, 2018. Association between long-term exposure to air pollution and sleep disorder in Chinese children: the Seven Northeastern Cities study. Sleep41, zsy122. [PubMed] [Google Scholar]
• Liu J, Feng R, Ji X, Cui N, Raine A, Mednick SC, 2019. Midday napping in children: associations between nap frequency and duration across cognitive, positive psychological well-being, behavioral, and metabolic health outcomes. Sleep42 (9). [PMC free article] [PubMed] [Google Scholar]
• Liu J, Liu X, Ji X, Wang Y, Zhou G, Chen X, 2016. Sleep disordered breathing symptoms and daytime sleepiness are associated with emotional problems and poor school performance in children. Psychiatr. Res242, 218–225. [PMC free article] [PubMed] [Google Scholar]
• Liu J, Zhou G, Wang Y, Ai Y, Pinto-Martin J, Liu X, 2012. Sleep problems, fatigue, and cognitive performance in Chinese kindergarten children. J. Pediatr161, 520–525. e522. [PMC free article] [PubMed] [Google Scholar]
• Martens AL, Reedijk M, Smid T, Huss A, Timmermans D, Strak M, Swart W, Lenters V, Kromhout H, Verheij R, 2018. Modeled and perceived RF-EMF, noise and air pollution and symptoms in a population cohort. Is perception key in predicting symptoms?Sci. Total Environ639, 75–83. [PubMed] [Google Scholar]
• Mindell JA, Owens JA, 2015. A Clinical Guide to Pediatric Sleep: Diagnosis and Management of Sleep Problems. Lippincott Williams & Wilkins. [Google Scholar]
• Mohammadyan M, Moosazadeh M, Borji A, Khanjani N, Moghadam SR, 2019. Exposure to lead and its effect on sleep quality and digestive problems in soldering workers. Environ. Monit. Assess191, 184. [PubMed] [Google Scholar]
• Moher D, Liberati A, Tetzlaff J, Altman DG, 2009. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann. Intern. Med151, 264–269. [PubMed] [Google Scholar]
• Neubauer DN, 1999. Sleep problems in the elderly. Am. Fam. Physician59, 2551–2558, 2559–2560. [PubMed] [Google Scholar]
• Paul KC, Haan M, Mayeda ER, Ritz BR, 2019. Ambient air pollution, noise, and late-life cognitive decline and dementia risk. Annu. Rev. Publ. Health40, 203–220. [PMC free article] [PubMed] [Google Scholar]
• Peel JL, Klein M, Flanders WD, Mulholland JA, Freed G, Tolbert PE, 2011. Ambient air pollution and apnea and bradycardia in high-risk infants on home monitors. Environ. Health Perspect119, 1321–1327. [PMC free article] [PubMed] [Google Scholar]
• Peters R, Ee N, Peters J, Booth A, Mudway I, Anstey KJ, 2019. Air pollution and dementia: a systematic review. J. Alzheim. Dis1–19. [PMC free article] [PubMed] [Google Scholar]
• Rooney AA, Boyles AL, Wolfe MS, Bucher JR, Thayer KA, 2014. Systematic review and evidence integration for literature-based environmental health science assessments. Environ. Health Perspect122, 711–718. [PMC free article] [PubMed] [Google Scholar]
• Russ TC, Reis S, van Tongeren M, 2019. Air pollution and brain health: defining the research agenda. Curr. Opin. Psychiatr32, 97–104. [PubMed] [Google Scholar]
• Sánchez T, Gozal D, Smith DL, Foncea C, Betancur C, Brockmann PE, 2019. Association between air pollution and sleep disordered breathing in children. Pediatr. Pulmonol54, 544–550. [PubMed] [Google Scholar]
• Schünemann H, Hill S, Guyatt G, Akl EA, Ahmed F, 2011. The GRADE approach and Bradford Hill’s criteria for causation. J. Epidemiol. Community Health65, 392–395. [PubMed] [Google Scholar]
• Scinicariello F, Buser MC, Feroe AG, Attanasio R, 2017. Antimony and sleep-related disorders: NHANES 2005–2008. Environ. Res156, 247–252. [PMC free article] [PubMed] [Google Scholar]
• Sears CG, Zierold KM, 2017. Health of children living near coal ash. Glob. Pediatr. Health4, 2333794X17720330. [PMC free article] [PubMed] [Google Scholar]
• Shen Y-L, Liu W-T, Lee K-Y, Chuang H-C, Chen H-W, Chuang K-J, 2018. Association of PM2.5 with sleep-disordered breathing from a population-based study in Northern Taiwan urban areas. Environ. Pollut233, 109–113. [PubMed] [Google Scholar]
• Shou Y, Huang Y, Zhu X, Liu C, Hu Y, Wang H, 2019. A review of the possible associations between ambient PM2.5 exposures and the development of Alzheimer’s disease. Ecotoxicol. Environ. Saf174, 344–352. [PubMed] [Google Scholar]
• Stein MA, Mendelsohn J, Obermeyer WH, Amromin J, Benca R, 2001. Sleep and behavior problems in school-aged children. Pediatrics107e60–e60. [PubMed] [Google Scholar]
• Strine TW, Chapman DP, 2005. Associations of frequent sleep insufficiency with health-related quality of life and health behaviors. Sleep Med. 6, 23–27. [PubMed] [Google Scholar]
• Sunyer J, Esnaola M, Alvarez-Pedrerol M, Forns J, Rivas I, Lóapez-Vicente M, Suades-González E, Foraster M, Garcia-Esteban R, Basagaña X, 2015. Association between traffic-related air pollution in schools and cognitive development in primary school children: a prospective cohort study. PLoS Med. 12, e1001792. [PMC free article] [PubMed] [Google Scholar]
• Valham F, Sahlin C, Stenlund H, Franklin KA, 2012. Ambient temperature and obstructive sleep apnea: effects on sleep, sleep apnea, and morning alertness. Sleep35, 513–517. [PMC free article] [PubMed] [Google Scholar]
• Van Dongen H, Maislin G, Mullington JM, Dinges DF, 2003. The cumulative cost of additional wakefulness: dose-response effects on neurobehavioral functions and sleep physiology from chronic sleep restriction and total sleep deprivation. Sleep26, 117–126. [PubMed] [Google Scholar]
• Wei F, Nie G, Zhou B, Wang L, Ma Y, Peng S, Ou S, Qin J, Zhang L.e., Li S, 2017. Association between Chinese cooking oil fumes and sleep quality among a middle-aged Chinese population. Environ. Pollut227, 543–551. [PubMed] [Google Scholar]
• Weinreich G, Wessendorf TE, Pundt N, Weinmayr G, Hennig F, Moebus S, Möhlenkamp S, Erbel R, Jöckel K-H, Teschler H, 2015. Association of short-term ozone and temperature with sleep disordered breathing. Eur. Respir. J46, 1361–1369. [PubMed] [Google Scholar]
• Wells G, Shea B, O’Connell D, Peterson J, Welch V, Losos M, 2017. New-Castle–Ottawa Quality Assessment Scale—Case Control Studies. [Google Scholar]
• WHO, 2018a. Ambient (Outdoor) Air Quality and Health. World Health Organization. [Google Scholar]
• WHO, 2018b. Household Air Pollution and Health. World Health Organization. [Google Scholar]
• WHO, 2019a. Ambient Air Pollution - a Major Threat to Health and Climate. Air pollution World Health Organization. [Google Scholar]
• WHO, 2019b. Ambient Air Pollution: Health Impacts, Air Pollution. World Health Organization. [Google Scholar]
• Woodruff TJ, Sutton P, 2014. The Navigation Guide systematic review methodology: a rigorous and transparent method for translating environmental health science into better health outcomes. Environ. Health Perspect122, 1007–1014. [PMC free article] [PubMed] [Google Scholar]
• Yıldız Gülhan P, Güleç Balbay E, Elverişli MF, Erçelik M, Arbak P, 2019. Do the levels of particulate matters less than 10 μm and seasons affect sleep?Aging Male1–6. [PubMed] [Google Scholar]
• Yu H, Chen P, Paige Gordon S, Yu M, Wang Y, 2019. The association between air pollution and sleep duration: a cohort study of freshmen at a university in Beijing, China. Int. J. Environ. Res. Publ. Health16, 3362. [PMC free article] [PubMed] [Google Scholar]
• Zaharna M, Guilleminault C, 2010. Sleep, noise and health. Noise Health12, 64. [PubMed] [Google Scholar]
• Zanobetti A, Redline S, Schwartz J, Rosen D, Patel S, O’Connor GT, Lebowitz M, Coull BA, Gold DR, 2010. Associations of PM10 with sleep and sleep-disordered breathing in adults from seven US urban areas. Am. J. Respir. Crit. Care Med182, 819–825. [PMC free article] [PubMed] [Google Scholar]