Air Pollutants and Daily Hospital Admissions for Psychiatric Care: A Review

Air pollution is well known to cause disease and is of significant public health concern in both developed and developing countries (1). In 2016, ambient (i.e., outdoor) air pollution in both cities and rural areas was estimated to cause 4.2 million premature deaths worldwide. In 2016, 91% of the global population was living in locations where the World Health Organization (WHO) Air Quality Guidelines levels were not observed. The WHO guidelines estimate health effects of air pollution and provide thresholds for pollution levels harmful to health. These guidelines are based on expert assessments of current scientific evidence for the health effects of particulate matter (PM) (PM₁₀, diameter between >2.5 and 10 micrometers [μm]; PM_2.5, ≤2.5 μm), nitrogen oxide (NO), nitrogen dioxide (NO₂), sulfur dioxide (SO₂), ozone (O₃), and carbon monoxide (CO) in all WHO regions.

Air pollution affects health and increases the rates of morbidity and mortality due to cardiovascular and respiratory diseases. By reducing air pollution levels, countries can decrease the burden of disease from stroke, heart disease, lung cancer, and both chronic and acute respiratory diseases, including asthma. Evidence is now emerging that the aforementioned harmful compounds may reach all organs and even penetrate the blood-brain barrier. As a result, the central nervous system (CNS) may be affected in many ways, resulting in neuropathic inflammatory processes, activation of the immune system, oxidative stress, damage to blood vessels, and alterations in cerebral neurotransmitter concentrations, which can lead to mental or behavioral alterations (2, 3).

Air pollution is regarded as the largest single environmental risk to health, and interest in air pollution as a potential environmental risk factor for mental disorders has grown in recent years. However, in our opinion, research in this field remains insufficient. The aim of this review was to briefly summarize original research articles investigating the short-term effect of air pollution exposure on mental health. For this purpose, we chose to focus on articles analyzing associations between daily levels of air pollutant concentrations and daily hospital admissions (HAs) or emergency department (ED) visits for psychiatric disorders and symptoms.

Methods

A systematic search of the online databases PubMed, Scopus, and Web of Science (from any time until March 4, 2019) was independently conducted by two of the authors (F.B. and L.A.). Boolean search queries included the following terms: (“air pollution” or “air pollutants”) combined with (“mental health” or “psychiatry” or “psychiatric disorder” or “psychiatric disorders”) and (“hospitalization” or “admission” or “admissions” or “visit” or “visits”). We compiled publications that were original peer-reviewed articles, were published in any language, provided empirical data (as opposed to reviews or commentaries), and reported data on associations between daily air pollution levels and daily HAs or ED visits for mental disorders. The titles and abstracts of all identified citations were screened on the basis of the defined inclusion criteria. All references of included studies and those of any previous pertinent reviews were carefully reviewed to identify additional relevant studies. Consensus was then derived, resulting in a final set of publications to be reviewed and summarized. On the basis of the aforementioned search terms and criteria, 127 articles were initially identified (a PRISMA diagram is included in an online supplement to this article). From this list, we selected 19 original research reports as a final set of publications to be reviewed (2, 4–21). Two independent reviewers (F.B. and L.A.) abstracted data elements (sample characteristics, period considered, country, air pollutants considered, and main findings) and then jointly reached consensus about any discrepant abstracted information. Any articles published after the search date that became otherwise known to us were also included. Institutional review board approval was waived for this review article.

Results

Among the 19 studies of the association between daily air pollution levels and daily HAs or ED visits for mental disorders, all but three considered multiple air pollutants (Table 1). The pollutants considered in the studies were the following: NO₂ (N=17 studies), SO₂ (N=15), PM₁₀ (N=16), O₃ (N=13), CO (N=13), PM_2.5 (N=12), NO (N=1), coefficient of haze (N=1), total oxidants (N=1), and total hydrocarbons (N=1). Most of the studies were conducted in China (N=7) and Canada (N=6) and the remainder in Korea (N=3), Italy (N=1), Sweden (N=1), and the United States (N=1). The selected studies analyzed the associations between air pollution levels and daily HAs or ED admissions for mental disorders in general (N=9) and for depression (N=5), schizophrenia (N=2), panic attacks (N=1), substance abuse (N=1), and suicide attempt or ideation (N=1). Characteristics and main findings of the reviewed articles are summarized in Table 1.

The 19 studies we considered included different pollutants in their econometric frameworks and used different statistical identification strategies. Almost half of the 19 reviewed studies (N=9) analyzed associations between daily air pollution levels and ED visits and HAs for mental disorders in general and did not present data on specific mental disorder subtypes. However, the findings in all studies revealed significant associations between daily HAs and ED admissions for mental disorders and daily levels of some or all air pollutants. The strongest and more consistent evidence across studies was found for PM pollutants: 13 of the 16 studies that included PM₁₀ in their analyses reported significant associations with daily HAs and ED admissions, as did eight of the 12 studies that included PM_2.5. Some evidence was found for CO (in six of 13 studies including CO), NO₂ (in seven of 17 studies), and SO₂ (in six of 15 studies), whereas the evidence for O₃ appeared somewhat weaker (four of 13 studies found a significant association of this pollutant with HAs and ED admissions).

Daily HAs and ED admissions for generic mental disorders were significantly associated with daily levels of PM₁₀ (N=7 studies), PM_2.5 (N=5), CO (N=2), NO₂ (N=1), NO (N=1), O₃ (N=1), and SO₂ (N=1). Admissions for depression were examined in five studies and were associated with daily levels of PM₁₀ (N=3 studies), SO₂ (N=3), PM_2.5 (N=2), CO (N=2), NO₂ (N=2), and O₃ (N=2). The two articles studying admissions for schizophrenia reported significant associations of these admissions with daily levels of NO₂ (in both studies), PM₁₀ (N=1), and SO₂ (N=1). Another three studies analyzed the associations between pollutant levels and panic attacks (indicating association with O₃), substance abuse disorders (associated with PM₁₀, PM_2.5, CO, and NO₂), and suicide attempts (associated with PM₁₀, CO, SO₂, and NO₂)_.

Table 2 presents details of each study by outcome of interest.

Discussion

In this article, we reviewed the literature investigating associations between elevated daily levels of air pollution with psychiatric ED visits and HAs. The potential role of air pollutants in the etiology of CNS disorders is receiving considerable attention (1). The outcome considered in the studies we reviewed was the daily number of HAs and ED admissions for generic or specific mental disorders, a broad marker of worsening mental health. Because almost half of the reviewed studies did not present data on the association between ED visits and HAs for specific mental disorder subtypes and daily air pollution levels, we could not conduct a comprehensive analysis of the impact of air pollution exposure on HAs or ED visits for specific diagnoses or symptoms subgroups.

Although the pollutants examined, the psychiatric conditions studied, and study period investigated were variable among the studies, the results suggest that air pollution is associated with a wide range of psychiatric disorders in the short term. The studies found significant associations between daily HAs and ED admissions for mental disorders and daily levels of various air pollutants, with greater evidence for an association of PMs (PM₁₀ in particular) with these disorders. The results of the reviewed studies varied with respect to which pollutant had the strongest impact on daily admissions; demographic characteristics of participants, such as age and sex; patterns of seasonality (i.e., associations with pollutants year-round or only in the cold or warm seasons); and study period. Differences across the studies’ findings may be attributed to differences in study designs, sociodemographic and clinical characteristics of the sample populations, air pollutant mixtures, and the diagnostic criteria used. Moreover, because statistical significance is determined in part by sample size, statistical significance to quantify the relative strengths of associations approximates the real effects of air pollutants only partially. In this regard, several cross-validations enlarging the sample size and using alternative statistical techniques would be needed.

Nevertheless, during the past decade, a consistent body of evidence has revealed the potentially harmful effects of inhaled air pollutants on the CNS (22), indicating that air pollutants might particularly target cerebral white matter, cortical gray matter, and basal ganglia (23). Although the physiopathological mechanisms remain unclear, the association between daily air pollution levels and daily HAs and ED visits for mental disorders may be somewhat biologically plausible. A 2017 literature review of air pollution as a risk factor for psychotic disorders indicated that exposure to xenobiotic heavy metals, such as lead and cadmium; constituents of air pollution, such as PM, nitrogen and sulfur oxides, and organic solvents; and other environmental pollutants could be component causes for schizophrenia and other psychotic disorders (24). Urbanicity (being born or raised in cities) has been commonly described as an environmental factor associated with increased risk for schizophrenia and other nonaffective psychoses, so air pollution exposure may mediate this association, at least in part (25).

The results summarized in this review extend the evidence that air pollution may cause or exacerbate a large group of mental diseases and conditions, including severe psychiatric disorders, such as schizophrenia and depression. If air pollution is an independent risk factor for mental health conditions, policy measures should be developed and expanded to address air pollution, which also increases risk for several other diseases. Regulatory interventions are needed to reduce environmental pollution levels and their negative effects on health.

Studies that analyze the associations of daily levels of pollutants with HAs and ED visits for mental disorders have become feasible because data about admissions are often readily available from hospital registers, and measures of air pollutants are increasingly used worldwide. Nevertheless, this approach presents important limitations, which warrant further discussion. The ecological nature of time series–based analysis can capture only acute effects of air pollution levels on the increase of HAs for mental disorders and may make it difficult to assess personal factors related to air pollution exposure. For instance, the reviewed studies took into account only outdoor pollution levels by using measurements from fixed-site monitor stations as a proxy for the pollution exposure of the population, which may not fully reflect personal exposure, because pollution exposure also may occur in other settings, such as workplaces and homes. For example, domestic wood smoke, as well as indoor cooking, may have health effects similar to those of particles stemming from automobile traffic and should be taken into account as an additional risk factor for mental health conditions (26). Seasonal and individual features (e.g., sex and age) might also modulate the health effects of air pollutant levels (27). The reviewed studies reported that various pollutants have different effects on different mental conditions, with different lag times. Also, the results of the reviewed studies indicated different effects on distinct subpopulations, with inconsistent results regarding which sex and age groups are more affected. Because each pollutant may operate through different pathways and mechanisms and may have different effects on the CNS, future studies should focus on the exact mechanisms underlying the effects of different pollutants on increased HAs for mental disorders. They should also explore the delayed effects of air pollution on mental health among different subpopulations, taking into consideration data on sociodemographic and clinical characteristics, including medical diagnosis and reason for admission.

Conclusions

The results summarized in this review extend existing evidence indicating that air pollution may contribute either to worsening of psychiatric symptoms or to seek treatment for a large group of mental diseases and conditions, including severe psychiatric disorders such as schizophrenia and depression. Future research in this field is strongly encouraged to increase statistical power of the available data and the generalizability of the results discussed here. One focus of this future research will likely be on the neurotoxic components of air and on environmental pollutants, revealing their pathological mechanisms on the CNS and investigating the effects of specific pollution components on the risks for mental illness and neurodevelopmental and neurodegenerative diseases. A large research program requiring collaboration among researchers and scientists from different fields appears to be required to understand this important human health issue (22).