The effects of outdoor air pollution on University students’ mental health: a within-group repeated-measures study Authors: Erkan Durar, Serap Yildirim Published: 2026 ABSTRACT BackgroundAir pollution is regarded as one of the greatest public health threats worldwide and has been linked to mental disorders. Therefore, studies on the effects of air pollution on mental health are highly important.ObjectiveThis study aimed to examine the effects of outdoor air pollution on the mental health of university students.MethodA pretest– posttest repeated-measures within-group design study was conducted with 365 students enrolled in the faculties of a university in a region of Türkiye characterized by a low air quality index during the 2023–2024 academic year and who consented to participate. Data were collected using a demographic information form and the Brief Symptom Inventory (BSI). Descriptive statistics and paired-sample t-tests were used to analyze the differences between repeated measurements.ResultsThe mean scores of depression, anxiety, and the global severity index-the BSI subscales- differed between the initial measurement, conducted when air pollution was moderate, and the final measurement, conducted when air pollution was unhealthy and poor. The average scores at the final measurement were higher than those at the initial measurement, and the difference was statistically significant (p < 0.05).ConclusionThis study concluded that when outdoor air pollution was unhealthy, students’ depression, anxiety, and general psychological distress levels increased significantly. Therefore, long-term follow-up studies are warranted.Clinical trial numberNot applicable. Introductıon The increase in energy consumption and greenhouse gas emissions has led to air pollution reaching critical levels, adversely impacting human health and living conditions by increasing the frequency and severity of extreme weather events [1]. Air pollution is not an isolated phenomenon; rather, it is intricately linked to global processes, such as climate change, socioeconomic inequalities, and environmental risk factors, thereby exacerbating its effects on population health [2]. According to the World Health Organization (WHO), seven million individuals exposed to outdoor air pollution die prematurely each year, and 99% of the global population inhales polluted air [3]. Annually, approximately seven million deaths are attributed to air pollution, primarily associated with cardiovascular and cerebrovascular diseases, respiratory illnesses, premature births, and various types of cancer [2]. Research investigating the health implications of outdoor air pollution has demonstrated that short-term exposure to air pollutants correlates with an increase in hospital admissions for respiratory diseases [4, 5]. The literature predominantly focuses on the impact of air pollution on physical health, indicating that both short- and long-term exposure to elevated levels of air pollution heightens the risk of physical health conditions, including autoimmune disorders, lung cancer, and cardiovascular diseases [2, 4]. Air pollution is recognized for its impact on mental health, comparable to its effects on physical health; however, the volume of research in this domain is significantly less than that addressing physical health issues. Existing studies indicate that air pollution generally exerts a detrimental effect on mental health [6]. Particulate matter from air pollution infiltrates the bloodstream and activates proinflammatory cytokines. This process results in neuroinflammation and oxidative damage, potentially precipitating mental disorders and neurological diseases [6, 7]. Current research findings suggest causal links between air pollution and conditions such as depression, anxiety, psychotic symptoms, reduced cognitive function, attention deficit and hyperactivity disorder in children, and neurodegenerative diseases in adults [8–11]. Evidence from the literature suggests that investigating the impact of outdoor air pollution as a predictor of mental health outcomes necessitates longitudinal research designs. These designs should incorporate baseline assessments of psychopathology, repeated follow-up measurements, and individual-level exposure data [9, 11]. Considering the cumulative and long-term nature of these effects, the lack of longitudinal and large-scale studies constitutes a notable gap in the literature. A review of the national literature indicates a paucity of studies examining the effects of air pollution on health, with existing research predominantly addressing physical health implications (Özgenç et al., 2022; [12]). Notably, there is a lack of research on the mental health consequences of air pollution. Globally, research has predominantly concentrated on examining both the immediate and long-term impacts of air pollution on physical health. This research frequently employs data derived from adults and older populations, hospitalized patients, or the general population. However, studies specifically addressing young adults are relatively scarce [9, 13, 14]. Previous research underscores the importance of modeling daily air pollution concentrations as predictors of mental health outcomes, while accounting for baseline symptom levels and pertinent covariates such as age, sex, and time spent outdoors [9, 10]. It is, therefore, essential to investigate the relationship between exposure to outdoor air pollution and mental health status among young adults, rather than focusing on causal effects, as they are at the beginning of their lives. Notably, Iğdır Province, located in the easternmost region of Turkey, where this study was conducted, is among the provinces with the highest levels of air pollution in the nation. Although the province’s air pollution arises from multiple sources, numerous national and international reports have highlighted the severity of air pollution in Iğdır [15]. This study aimed to examine the effects of outdoor air pollution on the mental health of university students. Although previous research has linked air pollution to psychiatric outcomes, most studies have focused on long-term exposure or clinically diagnosed mental disorders in the general population. Consequently, short-term symptom fluctuations among university students- a developmentally vulnerable group- remain underexplored. Moreover, the predominant use of cross- sectional and between-subject designs has limited insight into intra-individual variability in responses to environmental stressors. By employing a within-group repeated-measures design, this study addresses this methodological gap and provides preliminary evidence on the temporal associations between outdoor air pollution exposure and changes in mental health symptoms. These findings will contribute to the emerging field of environmental psychiatry by highlighting the potential role of outdoor air pollution, as one of the adverse environmental factors, in early mental health outcomes and psychological vulnerability. Method Design The research was conducted in a pretest– posttest repeated-measures within-group design at Iğdır University, Turkey, between 2023 and 2024. Iğdır University, located in the easternmost part of Turkey, was established in 2008. While the university initially had two faculties and one vocational school, as of 2025, it has 11 faculties, one institute, four vocational schools, and 23 application and research centers (ARCs). Sample and setting The study population consisted of all undergraduate and associate degree students studying at Iğdır University during the 2023–2024 academic year (N = 12,000). In this study, students residing outside the city center of Iğdır were excluded from the research population (N = 5,500). This exclusion criterion was implemented to ensure a more homogeneous exposure to outdoor air pollution, given that air quality levels can vary significantly between the city center and surrounding rural or semi-rural areas owing to differences in traffic density, heating practices, and local environmental conditions [16]. The inclusion criteria for the study were as follows: (i) currently enrolled as a university student and (ii) providing informed consent to participate in the study. Participants who met these criteria were invited to complete the study assessment. Consequently, 5,500 students were excluded, and 6,500 students residing in the city center were identified as the research population for this study. A 95% confidence level and 5% margin of error were employed to determine the sample size, resulting in a required sample size of 364 individuals. The sample was selected using a simple random sampling method, ensuring that each individual had an equal probability of selection, thereby enhancing the representativeness of the sample. Initially, 385 students were included in the study. During data cleaning, 20 students were excluded due to incomplete survey responses, resulting in a final analytical sample of 365. No additional participant loss occurred between the initial and follow-up assessments of the study. The participation criteria included being a university student and not having a diagnosed mental illness. This was assessed using a self-report screening question inquiring whether the participant had previously received a formal mental health diagnosis from a healthcare professional. The exclusion criteria were incomplete data collection and a desire to withdraw from the study during data collection. Study instruments Data were collected using a descriptive information form and the Brief Symptom Inventory. Descriptive Information Form: The form created by the researchers consisted of nine questions about the students identifying information (age, gender, marital status, department, class, place of residence, perceived income status, time spent outdoors, time spent outdoors in the last week, and diagnosed illnesses).Brief Symptom Inventory (BSI): This scale, developed following a review of the items in the Symptom Checklist-90-Revised (SCL-90-R), is designed to assess the presence and severity of psychological symptoms in individuals and comprises 53 items. The SCL-90-R is a widely used self-report instrument intended to evaluate a broad spectrum of psychological issues and symptoms of psychopathology (Derogatis, [17]). Each item on the Likert-type scale is scored from 0 (“not at all”) to 4 (“extremely”). The scale consists of nine subscales: somatization (S), obsessive-compulsive disorder (OCD), interpersonal sensitivity (IS), depression (D), anxiety disorder (AD), hostility (H), phobic anxiety (PA), paranoid thinking (PT), and psychoticism (P). The item content in the IS subscale pertains to eating and drinking disorders, sleep disturbances, thoughts about death and dying, and feelings of guilt. Three separate indices can be calculated from the scale: the Positive Symptom Distress Index, Global Severity Index, and Symptom Distress Index. The validity and reliability of the instrument have been well established in previous studies, with strong internal consistency across subscales (Cronbach’s alpha = 0.77–0.90) and demonstrated construct validity (Derogatis, [17]). In the current sample, the scale also showed excellent internal consistency (Cronbach’s alpha = 0.96), confirming its reliability for assessing psychological symptoms in university students.Positive Symptom Distress Index (PSDI): The average score was derived by dividing the sum of the subscales by 53. The PSDI is calculated as the mean of the ratings assigned to all items, excluding those left unanswered, and is regarded as a critical indicator, with values ranging from 0.00 to 4.00. A PSDI score exceeding 1.00 suggests a propensity towards psychopathological symptoms, whereas a score of 2.00 or higher indicates the presence of psychopathology [18]. The validity and reliability of the instrument have been thoroughly established in prior research. In the present sample, the scale demonstrated excellent internal consistency, with a Cronbach’s alpha of 0.95, affirming its reliability for evaluating psychological symptoms among university students.Global Severity Index (GSI): This is the total score obtained by accepting all items except those marked as “0” (all positive values) as one. This indicates the overall level of psychological distress in an individual.Symptom Distress Index (SDI): This was calculated by dividing the sum of the subscales by the total symptom score. Higher total scores on the scale indicate an increase in an individual’s psychological symptoms. Descriptive Information Form: The form created by the researchers consisted of nine questions about the students identifying information (age, gender, marital status, department, class, place of residence, perceived income status, time spent outdoors, time spent outdoors in the last week, and diagnosed illnesses). Descriptive Information Form: The form created by the researchers consisted of nine questions about the students identifying information (age, gender, marital status, department, class, place of residence, perceived income status, time spent outdoors, time spent outdoors in the last week, and diagnosed illnesses). Brief Symptom Inventory (BSI): This scale, developed following a review of the items in the Symptom Checklist-90-Revised (SCL-90-R), is designed to assess the presence and severity of psychological symptoms in individuals and comprises 53 items. The SCL-90-R is a widely used self-report instrument intended to evaluate a broad spectrum of psychological issues and symptoms of psychopathology (Derogatis, [17]). Each item on the Likert-type scale is scored from 0 (“not at all”) to 4 (“extremely”). The scale consists of nine subscales: somatization (S), obsessive-compulsive disorder (OCD), interpersonal sensitivity (IS), depression (D), anxiety disorder (AD), hostility (H), phobic anxiety (PA), paranoid thinking (PT), and psychoticism (P). The item content in the IS subscale pertains to eating and drinking disorders, sleep disturbances, thoughts about death and dying, and feelings of guilt. Three separate indices can be calculated from the scale: the Positive Symptom Distress Index, Global Severity Index, and Symptom Distress Index. The validity and reliability of the instrument have been well established in previous studies, with strong internal consistency across subscales (Cronbach’s alpha = 0.77–0.90) and demonstrated construct validity (Derogatis, [17]). In the current sample, the scale also showed excellent internal consistency (Cronbach’s alpha = 0.96), confirming its reliability for assessing psychological symptoms in university students. Brief Symptom Inventory (BSI): This scale, developed following a review of the items in the Symptom Checklist-90-Revised (SCL-90-R), is designed to assess the presence and severity of psychological symptoms in individuals and comprises 53 items. The SCL-90-R is a widely used self-report instrument intended to evaluate a broad spectrum of psychological issues and symptoms of psychopathology (Derogatis, [17]). Each item on the Likert-type scale is scored from 0 (“not at all”) to 4 (“extremely”). The scale consists of nine subscales: somatization (S), obsessive-compulsive disorder (OCD), interpersonal sensitivity (IS), depression (D), anxiety disorder (AD), hostility (H), phobic anxiety (PA), paranoid thinking (PT), and psychoticism (P). The item content in the IS subscale pertains to eating and drinking disorders, sleep disturbances, thoughts about death and dying, and feelings of guilt. Three separate indices can be calculated from the scale: the Positive Symptom Distress Index, Global Severity Index, and Symptom Distress Index. The validity and reliability of the instrument have been well established in previous studies, with strong internal consistency across subscales (Cronbach’s alpha = 0.77–0.90) and demonstrated construct validity (Derogatis, [17]). In the current sample, the scale also showed excellent internal consistency (Cronbach’s alpha = 0.96), confirming its reliability for assessing psychological symptoms in university students. Positive Symptom Distress Index (PSDI): The average score was derived by dividing the sum of the subscales by 53. The PSDI is calculated as the mean of the ratings assigned to all items, excluding those left unanswered, and is regarded as a critical indicator, with values ranging from 0.00 to 4.00. A PSDI score exceeding 1.00 suggests a propensity towards psychopathological symptoms, whereas a score of 2.00 or higher indicates the presence of psychopathology [18]. The validity and reliability of the instrument have been thoroughly established in prior research. In the present sample, the scale demonstrated excellent internal consistency, with a Cronbach’s alpha of 0.95, affirming its reliability for evaluating psychological symptoms among university students. Positive Symptom Distress Index (PSDI): The average score was derived by dividing the sum of the subscales by 53. The PSDI is calculated as the mean of the ratings assigned to all items, excluding those left unanswered, and is regarded as a critical indicator, with values ranging from 0.00 to 4.00. A PSDI score exceeding 1.00 suggests a propensity towards psychopathological symptoms, whereas a score of 2.00 or higher indicates the presence of psychopathology [18]. The validity and reliability of the instrument have been thoroughly established in prior research. In the present sample, the scale demonstrated excellent internal consistency, with a Cronbach’s alpha of 0.95, affirming its reliability for evaluating psychological symptoms among university students. Global Severity Index (GSI): This is the total score obtained by accepting all items except those marked as “0” (all positive values) as one. This indicates the overall level of psychological distress in an individual. Global Severity Index (GSI): This is the total score obtained by accepting all items except those marked as “0” (all positive values) as one. This indicates the overall level of psychological distress in an individual. Symptom Distress Index (SDI): This was calculated by dividing the sum of the subscales by the total symptom score. Higher total scores on the scale indicate an increase in an individual’s psychological symptoms. Symptom Distress Index (SDI): This was calculated by dividing the sum of the subscales by the total symptom score. Higher total scores on the scale indicate an increase in an individual’s psychological symptoms. The total score range of the scale varies between 0 and 212, and Cronbach’s alpha internal consistency coefficient for the total scale is 0.95, whereas the coefficients obtained for the subscales range between 0.55 and 0.86 [19]. In the current study, Cronbach’s alpha coefficient for the total scale was 0.96, and between 0.74 and 0.81 for the subscales. Determination of exposure to outdoor air pollution Although the main cause of air pollution in Iğdır, one of the provinces in Turkey with the highest levels of air pollution, is the inversion phenomenon resulting from the city’s topographic structure and meteorological characteristics, it is also noted that there are many factors, such as wind erosion, scarcity of forests, rapid urbanization, insufficient spread of natural gas usage, underdeveloped public transportation, and incomplete renewal of infrastructure [15, 20]. According to a report prepared for 2022 by IQAir, a Switzerland-based air quality technology company, the microgram (㎍) rate of PM2.5 per m³ in Iğdır was 114, making it the. The most polluted city in Europe regarding air quality [21]. The recommended annual PM2.5 concentration, according to the World Health Organization (WHO), is between 5 and 10 µg/m3 [22]. According to the United States Environmental Protection Agency, a PM2.5 concentration of > 55 µg/m3 is considered unhealthy. These findings suggest that the students participating in the study were exposed to outdoor air pollution levels ranging from moderate to poor. In this study, the air quality index (AQI) was used as a measure of air pollution. The AQI monitors particulate matter (PM) with aerodynamic diameters of ≤ 2.5 μm (PM2.5) and ≤ 10 μm (PM10), as well as sulfur dioxide (SO₂) and nitrogen dioxide (NO₂) levels, thereby offering a comprehensive assessment of air quality status. Air quality data measured at the station in Iğdır were obtained from the Ministry of Environment, Urbanization, and Climate Change Air Monitoring website (https://www.havaizleme.gov.tr/). Based on data from the air monitoring website, air quality was classified into four categories: sensitive (AQI: 100–150), unhealthy (AQI: 150–200), poor (AQI: 200–300), and hazardous (AQI: 300–500). In this study, AQI values were calculated from daily average concentrations recorded over the data collection periods, rather than from a single time point, to account for diurnal variations in air pollution levels. The initial data from the students were gathered during a period with daily average AQI values in the sensitive category, whereas the final data were collected during a period when daily average AQI values were within the unhealthy and poor categories. Data collection Research data were collected through the administration of a face-to-face questionnaire to students in a classroom setting. The study was conducted from October 2023 to January 2024. The initial assessment took place in the Iğdır Province from October 21 to 25, 2023, during a period characterized by sensitive air quality, as indicated by the AQI (100–150). The follow-up assessment was conducted with the same cohort between January 13 and 17, 2024, corresponding to a period when AQI levels were classified as unhealthy to poor (AQI: 150–300). Statistical analysis The data collected in this study were analyzed using the Statistical Package for Social Sciences (version 26.0; IBM Inc., Armonk, NY, USA). Descriptive statistical methods (numbers, percentages, mean, and standard deviation) were used to evaluate the data. The distribution of the data was assessed using normality tests and distribution measures, such as histograms, Q-Q plots, skewness, kurtosis, and the coefficient of variation. The reliability of the scale was examined using Cronbach’s alpha. To compare the quantitative data, the difference between repeated measurements was analyzed using a paired-samples t-test. Ethics of research The present study was performed in accordance with the Declaration of Helsinki. Ethical approval was obtained from the Iğdır University Faculty of Health Sciences, Non-Interventional Clinical Research Ethics Committee, with the decision dated, December 62,023, and numbered 3/6. After obtaining ethical approval, institutional permission was received from Iğdır University on, December 182,023, with document number E-99147595-020-123628. Finally, the students were informed about the objective, risk, and benefit of the research. Written informed consent was obtained before the investigation. Results The mean age of the students was 21.06 ± 3.58 years. Of these students, 68.5% were female, 97.5% were unmarried, 44.7% were in their second year of study, 69.9% resided in dormitories, and 59.7% perceived their income status as low. The students reported having no diagnosed physical or mental illness (Table 1). Table 1Distribution of students’ sociodemographic characteristicsSocio-demographic Characteristics n %GenderMale11531,5Female25068,5Marital StatusMarried92,5Single35697,5GradeFirst Grade14238,9Second Grade16344,7Third Grade51,4Fourth Grade5515,1Hoousing StatusIn a Dormitory25569,9At home with my family6317,3At home with a relative51,4At home with friends359,6Other71,8Income StatusIncome Less Than Expenses21859,7Income Equals Expenses12634,5Income More Than Expenses215,8Diagnosed Disease StatusNone365100,0Age (Ort ± Sd)21,06 ± 3,58 Total 365 100,0 Table 1Distribution of students’ sociodemographic characteristicsSocio-demographic Characteristics n %GenderMale11531,5Female25068,5Marital StatusMarried92,5Single35697,5GradeFirst Grade14238,9Second Grade16344,7Third Grade51,4Fourth Grade5515,1Hoousing StatusIn a Dormitory25569,9At home with my family6317,3At home with a relative51,4At home with friends359,6Other71,8Income StatusIncome Less Than Expenses21859,7Income Equals Expenses12634,5Income More Than Expenses215,8Diagnosed Disease StatusNone365100,0Age (Ort ± Sd)21,06 ± 3,58 Total 365 100,0 Distribution of students’ sociodemographic characteristics Table 2 presents the exposure status of the students involved in the study with respect to outdoor air pollution. As indicated in the table, 55.9% of the students were outdoors between 08:00 and 12:00. Students spent an average of 6.44 ± 0.15 h per day outdoors, with a total average of 7.07 ± 0.13 h outdoors during the preceding week (Table 2). Table 2Students’ exposure status to outdoor air pollutionExposure to Outdoor Air Pollution n %The Time Range Where He/She Is Most Outdoor08:00–12:0020455,912:00–16:005615,316:00–20:008523,320:00–24.00205,5Average Time Spent Outdoors During the Day6,44 ± 0,15Average Time Spent Outdoors in the Last Week7,07 ± 0,13 Table 2Students’ exposure status to outdoor air pollutionExposure to Outdoor Air Pollution n %The Time Range Where He/She Is Most Outdoor08:00–12:0020455,912:00–16:005615,316:00–20:008523,320:00–24.00205,5Average Time Spent Outdoors During the Day6,44 ± 0,15Average Time Spent Outdoors in the Last Week7,07 ± 0,13 Students’ exposure status to outdoor air pollution Table 3Comparison of students’ brief symptom inventory and subscale score averages according to measurement timesBSIFirst Measurement±Last Measurement±t p Somatization6,35 ± 5,036,81 ± 5,55-1,1940,233Phobic Anxiety3,87 ± 3,074,28 ± 3,26-1,7310,084Obsessive-Compulsivity7,96 ± 4,758,10 ± 5,04,3780,706Interpersonal Sensitivity3,78 ± 2,834,10 ± 3,22-1,4250,155Depression6,81 ± 4,537,75 ± 5,19-2,7640,006*Anxiety5,61 ± 3,906,30 ± 4,73-2,2670,024*Hostility5,21 ± 3,925,59 ± 4,52-1,2800,201Paranoid Ideation5,76 ± 3,486,21 ± 3,93-1,6380,102Psychoticism4,01 ± 3,284,34 ± 3,76-1,3040,193Additional Provisions4,46 ± 3,244,83 ± 3,46-1,5230,129Global Severity Index1,01 ± 0,571,10 ± 0,67-1,8990,058Positive Symptom Total30,38 ± 12,2132,24 ± 13,33-1,9740,049*Positive Symptom Distress Index1,69 ± 0,491,74 ± 0,65-1,1830,238BSI: Brief Symptom Inventory, *p < 0,05 t test: paired sample Table 3Comparison of students’ brief symptom inventory and subscale score averages according to measurement timesBSIFirst Measurement±Last Measurement±t p Somatization6,35 ± 5,036,81 ± 5,55-1,1940,233Phobic Anxiety3,87 ± 3,074,28 ± 3,26-1,7310,084Obsessive-Compulsivity7,96 ± 4,758,10 ± 5,04,3780,706Interpersonal Sensitivity3,78 ± 2,834,10 ± 3,22-1,4250,155Depression6,81 ± 4,537,75 ± 5,19-2,7640,006*Anxiety5,61 ± 3,906,30 ± 4,73-2,2670,024*Hostility5,21 ± 3,925,59 ± 4,52-1,2800,201Paranoid Ideation5,76 ± 3,486,21 ± 3,93-1,6380,102Psychoticism4,01 ± 3,284,34 ± 3,76-1,3040,193Additional Provisions4,46 ± 3,244,83 ± 3,46-1,5230,129Global Severity Index1,01 ± 0,571,10 ± 0,67-1,8990,058Positive Symptom Total30,38 ± 12,2132,24 ± 13,33-1,9740,049*Positive Symptom Distress Index1,69 ± 0,491,74 ± 0,65-1,1830,238BSI: Brief Symptom Inventory, *p < 0,05 t test: paired sample Comparison of students’ brief symptom inventory and subscale score averages according to measurement times BSI: Brief Symptom Inventory, *p < 0,05 t test: paired sample In this study, the average scores of students on the BSI subscales at the final measurement during the period of unhealthy air quality were higher than those at the first measurement during the period of sensitive air quality. The observed differences in scores were statistically significant for depression (p = 0.006) and anxiety (p = 0.024) subscales. Additionally, the scores for the PSDI, GSI, and SDI of the BSI during the final measurement period, characterized by unhealthy air quality, were higher than those recorded during the initial measurement period, which coincided with sensitive air quality. This difference was statistically significant for the Total Symptom Index (p = 0.049) (Table 3). Discussion Air pollution has emerged as a significant global environmental concern since the mid-20th century and adversely affects both physical and mental health [11]. Recent research has highlighted the impact of air pollution on mental health disorders, including anxiety, depression, and stress, contributing to a growing body of literature on this topic [6, 10, 11]. In this study, it was observed that students’ BSI depression subscale scores in the final assessment, conducted during a period of unhealthy and poor air quality, were significantly higher than those from the initial assessment, which took place when air quality was sensitive. Studies have reported that particulate matter causing air pollution may play an important role in the pathophysiology of depression [6, 7]. In their research involving adults over different years in China, Xue et al. [23] investigated the effect of long-term PM2.5 exposure on depressive symptoms and found a significant relationship between air pollution and depression; as air pollution decreased over the years, depression scores also declined. A review of the literature on this subject reveals a positive relationship between air pollution and depression [6, 9, 13, 24]; it has also been determined that exposure to air pollution over average periods of two, three, and five years increases the risk of depression [6, 25, 26]. The observation that depression subscale scores were significantly elevated during periods when air quality was classified as “unhealthy and poor” compared to periods with “sensitive” levels suggests a correlation between declining air quality and heightened depressive symptoms. In particular, in groups such as students, who spend a large portion of their time outdoors (Table 2) and are highly sensitive to environmental factors, it is plausible that deterioration in air quality can have adverse effects on mental health or exacerbate existing psychological distress. The findings are consistent with similar studies in the literature and highlight the importance of the relationship between air quality and depression. Nevertheless, it is important to consider that depression may also be influenced by other factors, such as genetic, psychosocial, and lifestyle conditions, not solely environmental factors. The study found that students’ anxiety subscale scores on the BSI during the final measurement, which was conducted when air quality was unhealthy and poor, were significantly higher than their scores on the initial measurement, which was conducted when air quality was sensitive. The relationship between air pollution and anxiety is a significant field of research, supported by an extensive body of literature. Scientific studies support the finding that exposure to air pollution has significant effects on individuals’ anxiety levels [27, 28]. This perception of threat can increase stress levels, potentially contributing to the development of anxiety, depression, and other psychological disorders over time. Environmental threat perception is an important factor influencing anxiety levels associated with air pollution, as polluted air is thought to create a persistent sense of uncertainty and threat among individuals. Furthermore, air pollution can generate a state of collective anxiety, which negatively impacts not only individual psychological health but also societal psychological well-being. Research shows that this type of anxiety is especially prevalent among youth and groups that are more sensitive to environmental risks [29]. The current research findings parallel similar findings in the literature and draw attention to the relationship between air quality and anxiety. In this study, it was found that the Total Symptom Index score on the BSI at the final measurement, conducted during a period when air quality was unhealthy and poor, was significantly higher than the score at the initial measurement, conducted during a period when air quality was sensitive. The Total Symptom Index is the average of the scores obtained from all subscales of the BSI and provides a general idea of both symptom intensity (severity) and the prevalence of symptoms (how many different symptoms are experienced) [19]. In the study, this finding, observed between the final measurement of BSI across all subdimensions and the initial measurement, revealed that students’ levels of psychological distress increased during periods when air pollution was unhealthy and poor. The research provides a perspective that examines the relationship between air pollution and mental health according to changes in air quality levels over time. Previous research has established a link between air pollution and mental health problems. Braithwaite et al., [10] underscored the association between air pollution and mental disorders, such as anxiety, depression, and stress, suggesting that polluted air acts as an environmental stressor that can potentially disrupt individuals’ psychological well-being. Similarly, Bhui et al., [30] demonstrated that individuals exposed to air pollution are more susceptible to mental health issues. Furthermore, Zhang et al. [6] examined the relationship between long-term exposure to outdoor air pollution and depression in China and found that prolonged exposure was linked to more pronounced mental health issues over time. This study revealed differences in total symptom index scores according to air quality measurements. It was observed that students’ mental states were adversely affected during periods when air quality was unhealthy and poor. However, it should not be overlooked that mental health can also be influenced by biological, psychological, and various environmental factors. Strengths and limitations This study has several strengths. First, conducting repeated measurements on the same group of students according to the air quality index increased the reliability of the data and supported the consistency of the results, because repeated measurements control for individual differences and allow direct observation of changes over time. Additionally, of studies that examine the effects of outdoor air pollution on mental health, this study is expected to raise awareness. Controlling variables such as the air quality index during measurement times is another important factor that strengthens the validity of these findings. These elements are believed to enhance the scientific value of this research. This study has some limitations. First, although this study attempted to control for variables that could affect mental health, considering that mental health is influenced by many variables, there may be factors that were overlooked or could not be measured, all of which may have an impact on the results. Second, although statistically significant relationships were observed in the depression and anxiety subscales, in addition to examining the other subscales in line with the Symptom Total Index findings, it is necessary to prefer more comprehensive research methods rather than conducting only descriptive studies. Third, although repeated measurements were conducted in the same cohort, the duration of the follow-up period was limited. For the reliability of the findings obtained from this study, longer-term follow-ups with the same group are required. The fourth and final limitation concerns exposure to air pollution. In this study, air pollution was identified as the primary independent environmental exposure variable and was consistently assessed across students; however, the exposure estimates were limited to outdoor air pollution only. Each student involved in the study was evaluated based on their exposure to outdoor air pollution. However, the assessment of students’ exposure to indoor air pollution, which may also influence mental health outcomes, was not feasible and should be addressed in future studies. Conclusion and recommendations Prolonged exposure to unhealthy air quality not only leads to physiological health problems but also contributes to the development of mental health issues. This study concluded that air quality is associated with mental health problems, such as depression and anxiety, among university students, and that students’ levels of psychological distress increase during periods when air quality is unhealthy. Understanding the relationship between air quality and depression, anxiety, and overall mental health is important for raising awareness on the topic, teaching methods to protect against and cope with the effects of air quality on mental health, and providing training and counseling for the early identification of mental health problems. This study increased the reliability of the data by conducting repeated measurements of the air quality index in the same group and made valuable contributions to the literature in this field. However, the inability to fully control other variables that may affect mental health, the short duration of the follow-up period, and limiting exposure to only outdoor air pollution present limitations in terms of the generalizability of the results. Therefore, improving environmental policies aimed at reducing air pollution and expanding mental health services play critical roles in the prevention and treatment of air pollution-related mental disorders. In future studies, it is recommended that other individual and environmental variables that may affect mental health be comprehensively controlled for, longer follow-ups be conducted, and exposure to indoor air pollution be evaluated. In addition, different research methods should be used to examine the sub-dimensions of mental health other than depression and anxiety.