Introduction

Long-term exposure to particulate matter (PM2.5) mainly affects the respiratory and cardiovascular systems, resulting in millions of premature deaths per year1,2. Globally, high PM2.5 concentrations have become a major health threat, due to high urban density and significant use of conventional fuel sources3,4. Meanwhile, traffic congestion has greatly increased PM2.5 levels, due to idling and the release of pollutants from accelerating vehicles5. Additionally, the physical structure of cities, especially the so-called “street canyons,” have concentrated PM2.5 by limiting its dispersion6. All of these factors have contributed to high levels of PM2.5 pollution in urban areas, impacting the well-being of residents7.

In Thailand has tracked its ambient air quality, including PM2.5. The Pollution Control Department (PCD) monitors PM2.5 on a daily basis, which varies seasonally and regionally across the country8,9. Between 1996 and 2016, the long-term average PM2.5 levels across Thai provinces varied from 20.5 to 37.4 µg/m³, with Bangkok and the northern regions experiencing the highest concentrations8,9,10. In this regard, urban areas in Chiang Mai frequently experience air pollution issues9 due to topography, wind patterns, pollution from neighboring countries, and agricultural fires, while Bangkok’s main pollution season (November–February) might be mainly due to traffic congestion8,10. Thus, it is clear that both urban areas have significant air quality concerns.

In this regard, how individuals perceive the risk of PM2.5 is a result of public understanding, awareness of its health effects, and their own experiences with air pollution11. For example, individuals are more likely to use masks or limit outdoor activities when there are high PM2.5 levels12. In addition to identifying health risks, research should also address public perceptions, behavioral responses, and preventive actions13. There is also a need to understand PM2.5 risk perception levels, since limited research has examined how air quality perception influences health decisions and behaviors14, how PM2.5 perception shapes health risk beliefs and preventive behaviors15, how certain factors (e.g., personal, cultural, and social) influence people’s willingness to understand PM2.516, and how people obtain PM2.5 risk information17. Therefore, this study examines PM2.5 risk perception and associated sociodemographic factors in urban Thailand. For this purpose, it performs the following: (1) gathers baseline information on how the population in urban Thailand perceives PM2.5 risk; (2) explores the association between PM2.5 risk perception and sociodemographic factors; and (3) investigates the association between health risk perception levels and preventive behaviors. The findings from this study are expected to indicate a significant positive association between higher PM2.5 health risk perception levels and the adoption of more frequent or effective preventive behaviors in urban Thailand.

Methods

Study design

This study conducted a cross-sectional survey over a three-month period (March–May 2024). Ethical approval was obtained from the Ethics Committee of the Faculty of Medicine Vajira Hospital, Navamindradhiraj University, Bangkok, Thailand (Approval No. COE: 057/2567). (See supplementary file 1 for details.)

Participants

The study included participants aged 18 years and above who consented to participate. All individuals in the study population resided in either Bangkok or Chiang Mai, Thailand. As for the sample size, it was calculated by using the estimation of an infinite population proportion19, a 95% confidence level, 5% margin of error, 50% assumed proportion, a design effect of 2, and a 20% addition for incomplete responses, resulting in a total of 921 participants. (1) a person who has resided in either Bangkok or Chiang Mai for over a year and has also experienced PM2.5 exposure for more than one year.; and (2) aged 18 years or older. The exclusion criteria included individuals with a diagnosed psychological condition.

Questionnaire

The questionnaire comprised demographic and individual characteristics such as gender, age, marital status, education, underlying disease, occupation, income, address, and duration of residence. It also covered various aspects related to PM2.5 exposure, including news tracking of PM2.5 factors and preventive behaviors, which consisted of six items (yes/no). Example questions included: “Avoid leaving the house or engaging in outdoor activities”; “Wearing a generic mask”; “Wearing an N95 mask”; and “Using a mobile air purifier”.

Perceptions of health risks associated with PM2.5 were measured by using a 13-item scale comprising two subscales: perceived susceptibility (8 items; e.g., “Outdoor activities currently involve higher PM2.5 exposure,” “PM2.5 exposure is elevated during outdoor activities”) and perceived severity (5 items; e.g., “PM2.5 is more damaging to health than general dust,” “Outdoor activities with high PM2.5 levels can readily cause illness”). The questionnaire items were adapted from18 and supported by a review of relevant literature, (see supplementary file 2 for details). The responses were based on a five-point Likert scale, ranging from 1 (strongly disagree) to 5 (Strongly agree). The total scores were then categorized into five levels of perceived PM2.5 risk, with 4.51–5.00 indicating the highest level, 3.51–4.50 indicating a high level, 2.51–3.50 indicating a moderate level, 1.51–2.50 indicating a low level, and 1.00–1.50 indicating the lowest level.

Regarding content validity, it was assessed by a panel of three experts in public and environmental health, health behavioral science, and psychology, using the Index of Item-Objective Congruence (IOC), with IOC values for each item ranging from 0.67 to 1.00. To evaluate reliability, a pilot study was conducted with 25 participants drawn from a population similar to the target group. As for the internal consistency of the instrument, it was measured by using Cronbach’s alpha. In this case, the coefficient was 0.824, indicating high reliability.

Data collection

The data was collected online by using Google Forms. In this regard, the survey link and QR code were distributed via multiple social media platforms based on snowball sampling including Facebook, Line, Instagram, and X(Twitter). Detailed information about the study was provided on the first page of the form, while informed consent was obtained on the subsequent page. Only participants who provided consent and whose willingness to participate was expected to continue were included in the survey. Those who declined to provide consent were automatically excluded.

Statistical analysis

In addition to the descriptive statistics were used. Binary logistic regression was also employed to test the association between influencing factors and PM2.5 risk perception (highest vs. lower) as well as the association between risk perception and preventive behaviors (yes vs. no). To investigate the demographic and individual characteristics associated with risk perception, univariate analysis was conducted. Meanwhile, factors with p < 0.05 were included in the multivariable model. In addition, the forward selection method was used to determine the final model, while statistical significance was set at p < 0.05. All the analyses were performed by using SPSS version 28 software (IBM Corp., Armonk, NY, USA).

Results

Participants’ characteristics

In this study, a total of 921 questionnaire responses were analyzed, with the resulting PM2.5 risk perception levels and frequency scores in urban areas of Thailand presented in Table 2. The majority of the participants were female (65.8%), age 25–44 years (52.2%), most single, divorced, widowed, and separated (66.3%). Bachelor’s degree (50.8%), don’t have underlying disease (65.5%), government or private employee (31.9), and income 15,001–30,000 Thai-Bath/month (39.1%). In addition, participants’ responses living in Bangkok (54.3%) and in Chiang Mai (45.7%). News tracking of PM2.5 in the area, most via Facebook (72.5%).

Frequency of PM2.5 risk perception in urban areas of Thailand

The frequency of PM2.5 risk perception items is presented in Table 1. Based on the findings, the majority of the participants showed either strong agreement or agreement regarding the perceived susceptibility and severity of PM2.5, demonstrating a strong sense of PM2.5 risk.

Table 1 Frequency scores for perceived PM2.5 risk.
Full size table

Association between sociodemographic factors and PM2.5 risk perception levels

Table 2 presents the association between the sociodemographic factors and the PM2.5 risk perception levels among the population in urban Thailand based on binary logistic regression and univariate analysis. Sociodemographic factors such as being female, being in the 18–24 and over 60 age groups, and having a bachelor’s degree or higher were significantly associated with an increased perception of PM2.5 risk (p < 0.05). Both employment status (including roles like government or private employee, student, or retired) and a monthly income over 30,000 THB were significantly linked to how people perceived PM2.5 risks (p < 0.05). Additionally, both living in Chiang Mai and a longer duration of residence (more than 20 years) were associated with a heightened perception of PM2.5 risk (p < 0.05). In terms of news tracking of PM2.5 factors, we found that those who received news via Facebook, LINE, X (Twitter), television, radio, general websites, and newspapers were associated with PM2.5 risk perception levels (p < 0.05).

Table 2 Univariable analysis of sociodemographic factors and risk perceptions of PM2.5 levels (n = 921).
Full size table

Table 3 shows the association between the sociodemographic factors and the PM2.5 risk perception levels among the population in urban Thailand based on binary logistic regression and multivariable analysis. According to the findings, the female participants reported the highest PM2.5 risk perception levels, compared to the males (p = 0.039; OR = 1.39, 95% CI 1.02–1.90). Meanwhile, individuals with a bachelor’s degree or higher were more likely to report the highest PM2.5 risk perception levels, compared to those with lower educational attainment (bachelor’s degree: p = 0.033, OR = 1.55, 95% CI: 1.04–2.31; higher than bachelor’s degree: p = 0.003, OR = 2.15, 95% CI: 1.30–3.54). Similarly, individuals with a monthly income greater than 30,000 THB were more likely to report the highest PM2.5 risk perception levels, compared to those with an income of 15,000 THB or less (p < 0.001; OR = 3.51, 95% CI: 2.17–5.66). Specifically, the residents in Chiang Mai reported higher PM2.5 risk perception levels than those in Bangkok (p < 0.001; OR = 4.06, 95% CI 2.94–5.60).

Moreover, those who received information through various news channels were associated with the highest PM2.5 risk perception levels. In this regard, the details are as follows: LINE (p = 0.007; OR = 1.52, 95% CI: 1.12–2.06), X (p < 0.001; OR = 2.17, 95% CI: 1.48–3.17), television/radio (p = 0.012; OR = 0.67, 95% CI: 0.50–0.92), general websites (p = 0.023; OR = 1.43, 95% CI: 1.05–1.95), and newspapers (p = 0.017; OR = 0.43, 95% CI: 0.22–0.86).

Table 3 Multivariable analysis of sociodemographic factors and risk perceptions of PM2.5 levels.
Full size table

Association between health risk perception levels and preventive behaviors

Table 4 summarizes the association between health risk perception levels and preventive behaviors among the population in urban Thailand based on binary logistic regression and univariate analysis. Based on the findings, those with the highest PM2.5 risk perception levels avoided leaving the house, engaged in outdoor activities, and used air purifiers more than those with moderate to high PM2.5 risk perception levels (p < 0.001; OR = 2.59, 95% CI: 1.97–3.39 and p < 0.001; OR = 2.44, 95% CI: 1.87–3.19, respectively). In addition, those with the highest PM2.5 risk perception levels wore N95 masks more than those with moderate to high PM2.5 risk perception levels (p < 0.001; OR = 3.07, 95% CI 2.28–4.15). Meanwhile, those with the highest PM2.5 risk perception levels wore generic masks less than those with moderate to high PM2.5 risk perception levels (p < 0.001; OR = 0.49, 95% CI 0.35–0.68). However, the use of mobile air purifiers and nose sprays was not associated with PM2.5 risk perception levels.

Table 4 Univariate analysis of health risk perception and preventive behavior.
Full size table

Discussion

This study investigated PM2.5 risk perception and its associated factors among the general population in urban Bangkok and Chiang Mai, Thailand. A total of 921 participants, all with more than one year of documented PM2.5 exposure, were merged for analysis. Their perceptions of PM2.5-related health risks were measured using a 13-item scale. The findings revealed that the majority of participants expressed strong agreement or agreement regarding their perceived susceptibility to and severity of PM2.5, indicating a pronounced sense of PM2.5 risk.

As for perceived susceptibility, the majority of the participants reported that PM2.5 exposure increases the risk of respiratory and lung diseases, especially among children, women, older adults, and lung disease patients. Regarding perceived severity, the participants believed that PM2.5 is more harmful than general dust and that outdoor activities during periods with high PM2.5 levels can easily cause illness.

Associations between sociodemographic factors and PM2.5 risk perception levels

According to the results, the residents in urban Chiang Mai were more than four times as likely to report the highest PM2.5 risk perception levels than those in Bangkok, supporting previous research showing that Chiang Mai has been affected by air pollution for many years9. In addition, Bangkok’s main pollution period is November–February, due to traffic congestion, whereas Chiang Mai’s main pollution period is January–April, due to topography, wind, cross-border pollution, and agricultural burning8,10. Despite both cities experiencing significant air quality issues, our findings indicate that the residents in Chiang Mai perceived a greater risk from PM2.5 pollution than those in Bangkok. The government and health agencies utilize various channels to educate the public on PM2.5 health risks, preventive measures (like mask usage), and proper waste management.

Our finding that females were 1.4 times more likely to report the highest levels of PM2.5 risk perception than males is consistent with previous research indicating that females may be more susceptible to its health impacts20. In this regard, there is a need to address this higher perceived PM2.5 risk in this demographic, which is possibly linked to trust in the government and pollution source attribution21. Thus, we suggest employing trusted female figures to disseminate accurate information and encourage protective behaviors.

Meanwhile, individuals with a bachelor’s degree or higher were about twice as likely to report the highest levels of PM2.5 risk perception compared to those with less than a bachelor’s degree. Specifically, educated individuals were more aware of air pollution and its risks, which is consistent with previous research22. Moreover, a related study found that higher education can heighten concerns regarding the health impacts of air pollution, consequently increasing the perceived risk23. Conversely, this increased awareness may lead to a higher perception of risk, even if the actual risk levels among different groups are not directly comparable.

We also found that those with a monthly income of more than 30,000 THB were 3.5 times more likely to report the highest levels of PM2.5 risk perception than those with a monthly income of 15,000 THB or less. Previous studies have supported the idea that higher income may correlate with greater access to air pollution information and thus a higher perceived threat11,25. Meanwhile, the ability of higher income individuals to afford and use protective measures, such as air purifiers and masks, or move to relatively cleaner areas, might also affect their perceived risk24. In sum, this study found a complex association between income, risk perception, and air pollution, while higher income may increase awareness and concerns about PM2.5, highlighting the broader socioeconomic disparities regarding exposure and vulnerability25. Thus, policies should be implemented to reduce pollution, especially in disproportionately affected low-income communities.

Finally, this study found that individuals who received information through various news channels—especially LINE, X, and general websites—were more likely to report the highest levels of PM2.5 risk perception, with X showing the highest odds ratio (OR = 2.2). In contrast, receiving information from newspapers and television/radio appeared to be protective, being associated with a 33% and 57% decrease in reporting the highest levels of PM2.5 risk perception, respectively. These channels, which frequently include air quality updates, most likely heighten awareness and concerns regarding PM2.5 health risks26. In this regard, trustworthy information from television, newspapers, radio, magazines, online news, etc., is crucial, since they are the primary sources for the public regarding natural hazards27. Based on this finding, frequent news exposure about PM2.5 most likely increases PM2.5 risk perception levels.

Association between health risk perception levels and preventive behaviors

A key finding was that the highest PM2.5 risk perception levels were associated with a greater adoption of preventive behaviors (e.g., staying indoors, limiting outdoor activities, etc.) and air purifier use, compared with the moderate to high PM2.5 risk perception levels. Previous research has demonstrated that high PM2.5 risk perception levels cause individuals to limit both the time and duration/effort of outdoor activities28. In addition, such levels increase the likelihood of using air purifiers, especially high-efficiency particulate air models, for indoor air filtration and exposure reduction29. Thus, we suggest that beyond staying indoors and using air purifiers, other potential responses to PM2.5 risk perception levels should include wearing masks, closing windows, and avoiding burning.

Our findings also showed those with the highest PM2.5 risk perception levels wore N95 masks more than those with moderate to high PM2.5 risk perception levels. However, even with an N95 mask, long-term exposure to high PM2.5 levels can still be risky, since filter effectiveness may decrease and breathing resistance can cause discomfort/strain30,31. Hence, clean air policies should be adopted to encourage local and national efforts to reduce PM2.5 pollution, as evidenced by studies demonstrating their effectiveness in various contexts, including China and Europe, where targeted policies have led to significant reductions in PM2.5 concentrations and associated health impacts36,37.

Moreover, our study showed that those with the highest PM2.5 risk perception levels wore generic masks less than those with moderate to high PM2.5 risk perception levels. Counterintuitively, understanding PM2.5 risks and wearing masks might lead to a moderate level of perceived risk32. Previous studies have indicated that air pollution awareness, instead of perceived PM2.5 risk, is correlated with mask use39. When individuals prioritize indoor air quality to avoid outdoor PM2.5 exposure, especially in low-income areas, cost-effective strategies are crucial33. These include utilizing materials like weatherstripping, caulk, or even simple tape to seal cracks around windows, doors, and other openings (e.g., around air conditioners and exhaust fans), thereby reducing the infiltration of outdoor pollutants.

Conversely, the use of mobile air purifiers and nose sprays was not associated with PM2.5 risk perception levels. Specifically, these strategies reduced the respiratory impact of PM2.5, but not the perceived danger or risk34,35, even though these methods appeared to be effective enough to warrant their use. Thus, it is important to encourage the use of reliable air quality applications and devices to help individuals make informed decisions about their outdoor activities and preventive measures.

Meanwhile, the strong agreement among the majority of the participants regarding their perceived susceptibility and the severity of PM2.5 demonstrates a significant sense of PM2.5 risk. For example, in urban Chiang Mai, the females with higher education and income, and whose information sources were various news channels, were more likely to have their PM2.5 risk perceptions influenced. Furthermore, the best ways to prevent PM2.5 exposure included staying indoors, limiting outdoor activities, using air purifiers, and wearing N95 masks.

The extended experience with air pollution likely compelled individuals with the strongest PM2.5 risk perception to seek out and implement more effective self-protection strategies. This brings forth a pressing concern: How long can these individuals be reasonably expected to tolerate continuous PM2.5 or air pollution exposure? This phenomenon highlights the urgent need for a paradigm shift: from individual self-protection to comprehensive, equitable, and sustained policy interventions addressing the root causes of PM2.5 pollution. Relying indefinitely on individuals, especially vulnerable low-income communities with limited mitigation resources, to endure chronic air pollution is unsustainable and unjust.

Limitations

Due to its cross-sectional design and specific focus on urban settings, this study’s findings may be subject to selection bias and may not be generalizable to the broader population. The use of snowball sampling disseminated through social media platforms may have led to an overrepresentation of younger, more digitally literate individuals, thereby affecting the representativeness of the sample and potentially variable the study outcomes. Moreover, the questionnaire in this study only captured the perceptions and experiences of a specific group in their current situations (for more than one year experience with PM2.5 exposure). Our study focuses on participants living in urban Bangkok and Chiang Mai, not specific areas or districts, but based on population having experience with PM2.5 exposure more than one year. This study lacks data on participants’ baseline health status and actual PM2.5 exposure levels, both of which could confound perceptions and behaviors. The study also did not account for individual PM2.5 exposure levels, or daily frequency of device and platform use. Furthermore, other potential contributors to anxiety, such as exposure to additional air pollutants, were not considered.

Conclusion

This study investigated PM2.5 risk perception and related factors among 921 participants in urban Bangkok and Chiang Mai using a 13-item scale and binary logistic regression. Findings showed that PM2.5 risk perception was significantly higher among females and those in the 18–24 and over 60 age groups. This was also true for people with a bachelor’s degree or higher, a monthly income over 30,000 THB, and various employment statuses. A heightened perception of risk was also linked to living in Chiang Mai, a longer duration of residence more than 20 years and getting news from a wide range of sources, including Facebook, LINE, X (Twitter), television, radio, general websites, and newspapers. Common prevention strategies included staying indoors, limiting outdoor activities, using air purifiers, and wearing N95 masks. Recognizing these heightened perceptions and the ongoing need for effective self-protection, the study recommends implementing more education and information campaigns to drive policies, raise awareness, and eliminate PM2.5 pollution sources.