ISSN: 2047-2986

Facebook-f Twitter Youtube Linkedin Envelope
ISOGH

IMPACT: Google Scholar: H5-Index = 69; H5-Median = 102 (12th in the World in Public Health); Web of Science/Clarivate: Impact factor = 5.8 (5-year) / 5.5 (2-year); Scopus: Citescore = 6.2

Articles

Time series analysis of environmental factors and their association with emergency visits for asthma exacerbations in Petaling Jaya, Malaysia

Norita Hussein1, Wee Cheah2, Chun Lin3, Chng Saun Fong4, Nasrin Aghamohammadi5, 6, Nik Sherina Hanafi1, Yong Kek Pang7, Mohmmad Salleh Yahya8, Ahmad Tajuddin M Nor9, Hilary Pinnock3, Harry Campbell3, Ee Ming Khoo1; on behalf of the RESPIRE Collaboration

1 Department of Primary Care Medicine, Faculty of Medicine, Universiti Malaya, Malaysia

2 Institute of Ocean and Earth Sciences, Universiti Malaya, Malaysia

3 National Institute for Health Research Global Health Research Unit on Respiratory Health, Usher Institute, University of Edinburgh, UK

4 Institute for Advanced Studies, Universiti Malaya, Malaysia

5 Department of Social and Preventive Medicine, Faculty of Medicine, Universiti Malaya, Malaysia

6 Curtin University Sustainability Policy Institute, School of Design and the Built Environment, Curtin University, Perth, Australia

7 Department of Medicine, Faculty of Medicine, Universiti Malaya, Malaysia

8 Trauma and Emergency Unit, Universiti Malaya Medical Centre, Kuala Lumpur, Malaysia

9 Department of Emergency and Trauma, Hospital Tengku Ampuan Rahimah, Klang, Malaysia

DOI: 10.7189/jogh.16.04249
Cite as: Hussein N, Cheah W, Lin C, Fong CS, Aghamohammadi N, Hanafi NS, Pang YK, Yahya MS, Nor ATM, Pinnock H, Campbell H, Khoo EM. Time series analysis of environmental factors and their association with emergency visits for asthma exacerbations in Petaling Jaya, Malaysia. J Glob Health. 2026;16:04249.
Keywords: asthma, emergency visits, time-series, pollution, meteorological, PM10

(12 pages)

Download PDF (121)

Download PDF

(12 pages)

Share:

Abstract

Background

The burden of respiratory conditions, especially asthma, is likely to increase with urbanisation and pollution unless effectively managed. Petaling Jaya, a metropolitan area, faced heightened risk due to worsening outdoor air pollution, which can trigger asthma exacerbations. We aimed to examine the relationships between environmental factors and emergency visits for asthma exacerbations.

Methods

We used retrospective data over four years (1 January 2014 to 13 April 2017) comprising clinical records from a main hospital in Klang Valley and environmental data (air pollutants and meteorological data) obtained from nearby stations. We used generalised additive models to assess the associations between environmental factors and asthma emergency visits. We used distributed lag nonlinear models to address the delayed effects and nonlinearity of particulate matter at 10 μm (PM10) exposure on these visits.

Results

Asthma exacerbations accounted for 2.3% of total emergency visits. Daily mean (x̄) PM10 level exceeding Malaysian Ambient Air Quality Standard (MAAQS) threshold was associated with a 21.3% (95% confidence interval (CI) = 2.2, 44.0) increase when levels reached 200 μg/m3 and 34.0% (95% CI = 4.7, 71.7) increase when levels reached 300 μg/m3 in asthma emergency visits. The lag effect was associated with PM10 exposure levels, peaking on days two and three at 100 μg/m3 and on the same day at 200 μg/m3 and 250 μg/m3. The cumulative exposure-response curve from lag 0–7 days shows that the increased relative risk of asthma emergency visits began at PM10 levels of 150 μg/m3.

Conclusions

We provide preliminary insights into the respiratory health burden in the Klang Valley. The associations observed between PM10 and asthma emergency visits are exploratory. Our findings suggest that the risk of visits may increase when PM10 levels near or above current MAAQS thresholds, highlighting the potential benefit of aligning local air quality standards with stricter international guidelines. These results may support the value of targeted preventive measures and healthcare resource planning for asthma, though larger multi-year data sets are required to enhance statistical robustness.

Asthma, the most common chronic respiratory condition worldwide, affects about 300 million people globally [1]. The global burden of asthma is estimated to increase due to expanding urbanisation [2], especially in cities with high outdoor air pollution from economically driven activities [3]. Evidence indicates that air pollution has a detrimental effect on asthma outcomes in both adult and child populations [4]. The World Health Organization (WHO) identifies specific air pollutants that can trigger asthma exacerbations, including nitrogen dioxide (NO2) [5–7], ozone (O3) [6–8], sulphur dioxide (SO2), and fine particulate matter (PM) at 10 μm (PM10) and 2.5 μm (PM2.5) [6,7,9]. The pollutant with the most significant impact on human health is PM, which is widely used as an indicator of air quality [10]. International studies have established an association between worsening of asthma symptoms and air pollution. Gordian et al. in Alaska reported a 3–6% increase in hospital visits for asthma when PM10 concentrations rose by 10 μg/m3 [11]. A study in the USA found a significant association between asthma emergency visits among children and PM10 exposure [12]. Further evidence reported associations between exposure to carbon monoxide (CO) and NO2 and severe asthma in adults [13] and children [14]. A five-year study (2009–2013) in London primary care facilities found that even brief exposure to PM10 and NO2 was associated with increased visits for respiratory symptoms and inhaler prescriptions [15]. Air pollution and climate change are closely connected, forming an intertwined system that poses significant risks to human health [16]. A review reported that extreme changes in temperature or weather were associated with increased risks of asthma exacerbations, 1.25-fold (95% confidence interval (CI) = 1.14, 1.37) for emergency visits and 1.10-fold (95% CI = 1.04, 1.17) for hospital admissions [17]. Meteorological factors, such as air temperature, wind, rainfall, and humidity, can exacerbate air pollution by affecting the transport and dispersion of pollutants [18].

Malaysia has an estimated asthma prevalence of 5.3–7.1% among children and adolescents, and 6.2% among adults [19–21]. Selangor, one of the states in Klang Valley, Malaysia, has the highest prevalence of adults with asthma at 22.0% [22]. Klang Valley is densely populated with over seven million residents and numerous air-polluting industries, including factories, power plants, and transport activities [23]. The challenge of suboptimal asthma control persists, where 70% of adults continue to experience poor control [24]. An environmental trigger, haze, was reported as a main exacerbating factor in 50% of affected adults [24]. Local studies have identified associations between haze and increased hospital admissions for asthma [25] and other respiratory illnesses [26]. Another study using 2000–2009 pollution data reported that urbanisation and high vehicular emissions in Klang Valley have severely polluted the area with PM10, putting the population at risk of chronic respiratory problems [27]. Pollutants affect not only the respiratory system but also the cardiovascular system, with studies indicating that a 10 μg/m3 increase in NO2 concentrations was associated with a 7.32% increase in cardiovascular admissions [28].

Between 2014 and 2017, several haze episodes affected the Klang Valley. In 2014, haze occurred from February to March due to seasonal dry weather and local peatland fires. A subsequent episode from June to October 2014 was caused by transboundary haze pollution from the Southwest monsoon. Forest fires in Sumatra and Kalimantan, Indonesia, in August and September 2015, significantly worsened air quality [29]. We aimed to examine the relationships among daily pollutant concentrations, meteorological factors, and emergency visits for asthma exacerbations, using retrospective data from 2014–2017, a period that included severe haze events. We employed generalised additive models (GAMs), a flexible semi-parametric approach, to estimate the associations of emergency visits with changes in environmental factors.

METHODS

Study design and setting

We conducted a retrospective study in Petaling Jaya (PJ), a city in Klang Valley, where environmental data from the Department of Environment and Malaysia Meteorological Department monitoring stations, and records of emergency visits for asthma exacerbations from the Trauma and Emergency Unit of Universiti Malaya Medical Centre (UMMC) were all available for data collection (Appendix S1 in the Online Supplementary Document). The UMMC is a government-funded teaching hospital approximately 1.6 km from the PJ air quality station and 2 km from the PJ meteorological station. PJ is an industrial and residential urban area with high levels of road traffic congestion and pollutant emissions [30–32]. Based on patients’ addresses, 85.6% resided within 20 km of the hospital (Appendix S3 in the Online Supplementary Document).

Data collection

The available data we used span from 1 January 2014 to 13 April 2017, covering both haze and non-haze periods. Several haze events occurred in 2014 and 2015, when PM10 levels exceeded the daily mean (x̄) limit of the Malaysian Ambient Air Quality Standard (MAAQS) 2015 of 100 μg/m3 [33] at multiple locations [34,35].

Health data

We collected the health data either manually from clinical records or electronically from the Electronic Medical Records. We removed all hospital registration numbers and assigned non-identifiable identification codes before analysis. To ensure data collection was clear and systematic, we created a codebook outlining inclusion and exclusion criteria, a detailed account of the retrieved cases, and the total number of daily emergency visits. Inclusion criteria included all patients who visited the hospital’s Trauma and Emergency Unit with a documented diagnosis of asthma exacerbation by the attending physicians (referred to as ‘asthma emergency visit’ in this paper), with a documented underlying history of asthma (Appendix S2 in the Online Supplementary Document). This also included exacerbations secondary to upper respiratory tract infections, pneumonia, cold exposure, pets, and environmental changes. To ensure adherence to the established criteria, the lead investigator (NH) conducted an audit of a randomly selected sample. We collected age, daily attendance for asthma emergency visits, and home address data. We used the total number of patients visiting the Trauma and Emergency Unit during the study period as the denominator to calculate the asthma-related visit ratio. We focused on examining emergency visits for asthma exacerbations on a per-visit rather than per-patient basis. For patients who had repeated visits, we treated each visit as a separate incident. We considered this approach to reflect the short-term impact of air pollution on hospital workload.

Air quality and meteorological data

We used air pollutants (i.e. PM10, CO, NO2, SO2, and O3) and meteorological data (i.e. air temperature, humidity, atmospheric pressure, wind, and rainfall) obtained from nearby stations. We calculated daily x̄ concentrations for PM10, CO, NO2, SO2, air temperature, relative humidity, atmospheric pressure, and wind (speed and direction) from hourly measurements. For O3, we used the maximum eight-hour rolling average concentration for the day, whereas we calculated daily rainfall (duration and volume) as the sum of the day. We conducted all calculations in accordance with the WHO Air Quality Guidelines (AQG) metrics [10] for assessing short-term exposure effects. We applied a data capture threshold of 75% when calculating daily data. During the study period, only PM10 data were available for PM, whereas PM2.5 data became available only after July 2017; therefore, we excluded it from the analysis.

Data analysis

We reported age and asthma emergency visits as frequencies and percentages. We examined associations between daily visits and environmental variables using GAMs with a negative binomial distribution and log link to account for overdispersion (variance-to-x̄ ratio = 1.42) (Appendix S3 in the Online Supplementary Document). The model included parametric and nonlinear terms to control for covariates (i.e. long-term trends, seasonality, month, day of week, holidays, air pollutants, and meteorological variables). Parametric predictors included month, day, holidays, and rainfall. We modelled continuous variables (i.e. air pollutants and meteorological factors) with cubic splines for nonlinear relationships. We estimated smoothing parameters using restricted maximum likelihood. The model showed a good fit (Pearson dispersion = 1.006; adjusted R2 = 0.28; deviance = 32.1%). To explore delayed and nonlinear effects of PM10, we used a distributed lag nonlinear model within the GAM framework over a lag period of up to seven days, modelling nonlinear exposure-response and lag structure. We estimated relative risks (RRs) of asthma emergency visits associated with PM10 concentrations using 100 μg/m3 as the reference, corresponding to the MAAQS threshold [33].

We used R, version 4.5.1 (R Core Team, Vienna, Austria) for all analyses.

RESULTS

Emergency visits and environmental variables

A total of 10 411 asthma emergency visits were recorded from 1 January 2014 to 13 April 2017, representing 2.3% of all emergency visits during those four years (Table 1). The median of asthma visits per day was 8 (interquartile range = 6–11). Most patients were aged ≥18 years, with a median age of 34 years (Table 2). The daily x̄ PM10 concentration during the study period was 51.47 μg/m3, which was higher than the WHO AQG for the daily x̄ limit of 45 μg/m3 [10] but lower than the MAAQS daily x̄ limit of 100 μg/m3 [33] (Table 3).

Table 1. Emergency visits between 1 January 2014 and 13 April 2017
n (%)
Total number of all emergency visits450 607
Asthma emergency visits10 411 (2.3)
Daily number of asthma emergency visits, MD (IQR)8 (6–11)

IQR – interquartile range, MD – median

Table 2. Age distribution of the included participants (n = 10 411)
n (%)
Age in years
x̄ (SD)36.3 (24.3)
MD34
Age group distribution in years
≤51327 (12.7)
6–111084 (10.4)
12–17454 (4.4)
18–605239 (50.3)
>602292 (22.1)
Unknown15 (0.1)

MD – median, SD – standard deviation, x̄ – mean

Table 3. Average values of the recorded environmental variables’ daily concentrations
x̄ (SD)MD (IQR)
Air pollutants (μg/m3)
PM1051.47 (31.94)43.81 (34.41–57.06)
O399.54 (41.68)94.82 (69.16–126.83)
NO249.05 (12.16)48.68 (40.99–56.67)
SO211.13 (4.94)10.65 (7.55–14.22)
CO (mg/m3)1.35 (0.38)1.31 (1.11–1.55)
Meteorological data
Daily temperature (°C)28.49 (1.28)28.51 (27.57–29.49)
Daily relative humidity (%)73.88 (7.69)74.25 (68.48–79.62)
Daily wind speed (m/s)4.90 (1.14)1.12 (0.97–1.30)
Atmospheric pressure (hPa)1009.00 (1.35)1009.00 (1008.00–1010.00)
Rainfall duration (min)82.10 (1.07)30.00 (0.0–143.00)
Rainfall volume (mm)9.20 (1.76)0.20 (0.0–11.00)

IQR – interquartile range, MD – median, SD – standard deviation, x̄ – mean

Asthma emergency visits and PM10

Peak daily asthma emergency visits occurred in early and late 2014, early 2015, and mid-to-late 2015 (Figure 1). This observation appeared to follow a similar pattern with the highest recorded x̄ daily PM10 levels of haze episodes, exceeding the daily x̄ MAAQs threshold of 100 μg/m3 and reaching 300–400 μg/m3. There were 4855 (45.6%) of asthma emergency visits when PM10 concentrations were above the WHO AQG thresholds, and 631 (6.06%) when they were above the MAAQS thresholds.

Figure 1. Daily count of asthma emergency visits (light brown bars) and daily mean PM10 concentrations (black dots) observed during the study period. The blue line indicates the World Health Organization Air Quality Guidelines for the daily mean limit of 45 μg/m3 [10], and the red line represents the Malaysian Ambient Air Quality Standard for the daily mean limit of 100 μg/m3 [33]. PM10 – particulate matter at 10 μm.

Association of environmental variables with asthma emergency visits

PM10 concentration below the daily x̄ of 51 μg/m3 was not significantly associated with emergency visits (x̄ = −3.8%; 95% CI = −21.4, 17.8%) (Figure 2, Panel A). With a PM10 increase from 51 μg/m3 to 100 μg/m3, there was a 12.8% (95% CI = −20.8, −4.1) reduction in emergency visits. The association became positive at higher concentrations. A PM10 increase from 100 to 200 μg/m3 was associated with a 21.3% (95% CI = 2.2, 44.0) increase, while an increase from 100 to 300 μg/m3 corresponded to a 34.0% (95% CI = 4.7, 71.7) increase in emergency visits. The PM10 increase from 200 to 300 μg/m3 was associated with an emergency visit increase of 10.5% (95% CI = −16.3, 45.8) but was not statistically significant, most likely due to fewer observations (Figure 1, Figure 2, Panels A). Overall, the results indicate a threshold effect, in which the association between PM10 and asthma emergency visits becomes apparent only above the MAAQS daily x̄ limit of 100 μg/m3. The relationship with atmospheric pressure was nonlinear (estimated degrees of freedom (EDF) = 2.98; P = 0.01) (Figure 2, Panel B), whereas the relationship with wind speed was moderately nonlinear (estimated degrees of freedom = 2.2, P = 0.01) (Figure 2, Panel C). Across the central 90% of the observed atmospheric pressure range (1007–1011 hPa), a 4.35 hPa increase was associated with a 12.8% (95% CI = −17.7, −2.3) reduction in emergency visits. An increase in wind speed from 1 to 2 m/s was associated with a 12.0% (95% CI = −22.5, −0.1) decrease in emergency visits. For rainfall (Figure 2, Panel D), an increase in volume from 0 to 50 mm was associated with a decrease in emergency visits by 6.4% (95% CI = −12.5, 0.2). The increase in rainfall from 50 to 100 mm was associated with an increase in emergency visits of 4.11% (95% CI = −2.9, 11.5); however, it was not statistically significant.

Figure 2. Daily mean of PM10 concentrations, atmospheric pressure, wind speed, and rainfall volume and their associations with daily asthma emergency visits based on GAM analysis. Panel A. PM10. Panel B. Atmospheric pressure. Panel C. Wind speed. Panel D. Rainfall volume. The bold black line estimates the relative effect sizes for asthma emergency visits (y-axis), and the shaded area represents the 95% confidence interval. PM10 – particulate matter at 10 μm.

Patterns of asthma emergency visits during the study period

We analysed patterns of asthma emergency visits across days of the week, revealing statistically significant variations. Compared with the Friday baseline, Mondays (21.8%; P < 0.001), Tuesdays (13.3%; P = 0.001), and Sundays (9.5%; P = 0.02) showed increases in asthma visits. The public holiday effects were not statistically significant, with no evident seasonal trend observed.

The lag effect

The distributed lag nonlinear model analysis showed that at a PM10 concentration of 150 μg/m3, the RR ranged from 0.995 to 1.015, with the highest lag effects observed on days two and three (Figure 3, Panel A). When the concentration increased to 200 μg/m3, there was a 5.3% increase in emergency visits on the same day (lag 0) and persisted through day two (Figure 3, Panel B). At 250 μg/m3, the RR was 1.165 on day 0, indicating a 16.5% increase in emergency visits on the same day (Figure 3, Panel C).

Figure 3. The RR of asthma emergency visits at various daily mean PM10 concentrations compared to the daily MAAQS threshold. Panel A. PM10 100 vs. 150μg/m3. Panel B. PM10 100 vs. 200μg/m3. Panel C. PM10 100 vs. 250μg/m3. The bold line indicates the point estimate, and the horizontal line at relative risk = 1 denotes no excess risk, while values >1 represent increased risks. The shaded area is 95% confidence interval. PM10 – particulate matter at 10 μm, RR – relative risk.

Cumulative exposure-response relationship

The cumulative exposure-response curve (lag 0–7) illustrates a positive, nonlinear relationship between PM10 concentrations and RR of asthma emergency visits (Appendix S4 in the Online Supplementary Document). The risk remains relatively stable at 100 μg/m3 (MAAQS threshold) but begins to show a noticeable upward trajectory as levels exceed 150 μg/m3, reaching RR values above 1.5 at the highest observed concentrations. However, the differences were not statistically significant. We used the lag window of 0–7-day in the model to demonstrate the short-term effects of PM10 exposure. We have examined the cross-correlation between PM10 and asthma emergency visits over lags from −15 to 15 days and found significant lags within 0–7 days (Appendix S5 in the Online Supplementary Document).

DISCUSSION

We found that from 1 January 2014 to 13 April 2017, asthma exacerbations accounted for 2.3% of all emergency visits at UMMC. The highest number of visits occurred in early 2014 and mid-to-late 2015. This appeared to follow a similar pattern with the highest recorded x̄ daily PM10 levels observed in Klang Valley. Higher concentrations of PM10 above the MAAQS threshold were associated with increased risk of asthma emergency visits. Lower atmospheric pressure and wind speed were significantly associated with increased risk, while moderate rainfall volume was associated with reduced risk of emergency visits. The lag effect on asthma emergency visits was influenced by PM10 exposure level, with the risk peaking on days two and three at 100 μg/m3 and on the same day at 200 μg/m3 and 250 μg/m3. The cumulative exposure-response curve from lag 0–7 days shows that the RR of asthma emergency visits began at PM10 levels of 150 μg/m3. There was a significant increase in asthma visits on Mondays (21.8%; P < 0.001), Tuesdays (13.3%; P = 0.001), and Sundays (9.5%; P = 0.02).

Our findings on the association between PM10 concentrations and the risk of asthma emergency visits were consistent with several other studies [3,36–38]. We observed that the association was nonlinear. At 100 μg/m3, there was a minimal or decreased risk of emergency visits. However, when levels exceeded the MAAQS threshold of 100 μg/m3, there was an increase in the number of visits of 21.3% at 200 μg/m3 and of 34.0% at 300 μg/m3. Although a positive trend is suggested, the model lacks the statistical power to confirm a significant effect at concentrations above 200 μg/m3 and should be interpreted with caution. Nevertheless, the associated increase in emergency visits may represent a clinically meaningful trend and may justify proactive measures for patients with asthma and healthcare facilities.

Regarding meteorological factors, other studies have reported inconsistent findings on the relationship between wind speed and asthma exacerbations. Some studies have indicated that higher wind speeds are associated with increased asthma emergency visits [39,40]. The explanation was that high winds can disperse PM, pollen, and mould spores, increasing their presence in the air and making them more likely to be inhaled by people with asthma [39]. In contrast, we found higher wind speeds to be associated with decreased asthma emergency visits. A possible explanation is that PM10 is readily transported over long distances by wind [41]. Low wind speeds can reduce pollutant dispersion, leading to higher pollutant levels in the area [42] and potentially triggering asthma exacerbations. These inconsistencies in results could arise from variations in cohorts, age groups, settings, or geographical locations, and future studies are needed to confirm this. Regarding atmospheric pressure, our findings were similar to those from Beijing, reporting a negative correlation with daily asthma hospital admissions [43]. While the precise physiological mechanisms by which atmospheric pressure influences asthma are not fully understood, it could be hypothesised that lower pressure often coincides with extreme weather changes, such as thunderstorms, which can trigger asthma symptoms [44]. The association between rainfall volume and asthma has also been controversial; however, a moderate amount has been reported to have a protective effect, consistent with our findings, which is attributed to its removal of particulates and allergens from the atmosphere [45].

Asthma emergency visit patterns also indicated that weekdays accounted for most daily visits, specifically Monday, followed by Tuesday and Sunday. In PJ, Sunday is a designated public holiday, with Monday and Tuesday following. The surge in emergency visits on these days is likely due to primary or outpatient clinics being closed or to a backlog of cases accumulated over the weekend, when clinics are closed. Additionally, this pattern may also be influenced by the delayed physiological response to air pollution exposure, since symptoms can take a few days to worsen enough to require emergency care. These findings may also be related to increased human mobility and activity on weekends, leading to greater exposure to environmental pollutants.

The lag effect of PM10 concentrations on the risks of asthma emergency visits varies. At 150 μg/m3, which is above the MAAQS threshold, the peak effect occurred on days two and three. When PM10 exceeded 200 μg/m3, a higher risk was observed on the same day (lag 0) and persisted up to three days. The lag effect observed on the same day at higher concentrations may have been due to higher PM10 levels, which could have prompted vulnerable individuals to visit the hospital earlier, leading to fewer visits on subsequent days. This is known as a harvesting effect or short-term displacement: people seek care earlier due to earlier exposure to high concentrations, thereby decreasing the number of people seeking care in the following days [46]. Although it may appear speculative, from a public health perspective, it suggests that patients might remain indoors or take preventive medication during periods of moderate to high pollution, resulting in lower emergency visits. The harvesting effect, where elevated pollution levels at 200 μg/m3 and 300 μg/m3 accelerate health events that would have occurred shortly afterwards, could account for the increased emergency visits observed on the same day. It is important to note that asthma exacerbations can still occur up to 2–3 days after exposure. The observed pattern of lagged effects over several days post-exposure to PM10 appears biologically plausible, as inflammatory responses to air pollution may require several days to manifest as symptoms severe enough to necessitate hospital visits [47]. This insight can assist public health authorities in issuing warnings that cover a longer period of vulnerability following a pollution event. Such warnings might encourage people to stay indoors or prompt authorities to take actions like closing schools, which could subsequently reduce exposure. The cumulative exposure-response curve from lag 0–7 days shows that the increased risk of asthma emergency visits began at PM10 levels of 150 μg/m3. It remains unclear whether the daily x̄ PM10 threshold of 100 μg/m3, set by the MAAQS, is ideal or appropriate given this narrow margin. However, these findings are preliminary because the wide CI limits definitive conclusions. Further research using larger, multi-year data sets is needed to narrow CIs and strengthen the findings, which could help shape regional air quality policies and clinical planning in the Klang Valley.

This study has several limitations. We obtained the health data from a single main government hospital, which could limit the generalisability of the findings to the broader population, as it could not have detected all emergency visits within the Klang Valley. We initially considered gathering data from multiple public/private hospitals and clinics, but the lack of systematic record-keeping meant that collecting accurate retrospective daily clinical records was not possible. We selected the most reliable data source available for our time-series analysis, but acknowledge that the risk estimates for emergency visits associated with PM10 levels was limited to a single catchment area and should be interpreted with caution. Second, the retrospective nature of this study imposes certain limitations. While we used the x̄ daily concentration to estimate population exposure, it may not accurately reflect individual exposure levels due to unaccounted confounding factors, such as indoor air pollution, smoking, socioeconomic status, and pre-existing co-morbidities. These limitations could introduce bias in assessing the effects of air pollution, and therefore, the findings should be interpreted with caution. Further prospective studies are needed to strengthen the findings. Third, regarding data recording practices, we acknowledge that not using International Classification of Diseases coding may lead to variability in data entry. This also represents a limitation in retrospective studies. Additionally, concerning time-varying confounding, we incorporated only day of the week, month of the year, and public holidays within the GAM model (Appendix S3 in the Online Supplementary Document). We omitted other potential variables, such as infectious respiratory epidemics and school holidays, which could potentially introduce bias into the exposure-response relationship. Future prospective studies should consider these variables to ensure that the estimated PM10 risks are independent of temporal variations. Additionally, we made assumptions that all asthma exacerbations would present to the UMMC emergency unit, which might not be true, suggesting that the true impact of pollution could be underestimated. Our study also has several strengths. We chose emergency visits for asthma exacerbations as the outcome variable rather than hospital admissions, which better captures the possible association with pollution. GAMs, the modelling technique used in this study, have provided a strong framework for data analysis, helping to address the potential nonlinearity of air pollution exposure and better understand the association between pollution and health service use.

CONCLUSIONS

Although we used data from a single facility and may not be fully representative of the entire Klang Valley population, we provide an important ‘sentinel’ perspective, potentially contributing to a broader understanding of environmental triggers in the Klang Valley. Our findings suggest that PM10 exposure levels above MAAQS thresholds may coincide with a period of increased risk for patients with asthma, highlighting the potential value of targeted preventive measures. Given the observed association of pollution with emergency care utilisation, further consideration of the alignment between local thresholds and international WHO guidelines may help better protect vulnerable populations and inform health care resource planning. The observed narrow margin of PM10 concentrations between 100 and 150 μg/m3 (MAAQS specifies a limit of 100 μg/m3) offers preliminary insights. However, these results should be regarded as exploratory, necessitating further investigation through analysis of larger, multi-year data sets to narrow CIs and enhance the robustness of the findings to inform regional air quality policies and clinical preparedness in the Klang Valley.

Additional material

Online Supplementary Document

Acknowledgements: The Global Health Research Unit on Respiratory Health (RESPIRE) is a collaboration comprising UK and low-middle-income countries grant holders, partners, and research teams, as listed on the RESPIRE website (www.ed.ac.uk/usher/respire). We thank Professor Dr Jamal I-Ching Sam, the Department of Environment (DOE), Malaysia, the Malaysian Meteorological Department (MetMalaysia), the Institute for Medical Research, Malaysia, and the University Malaya Medical Centre for their collaboration and assistance in providing data.

Ethics statement: The Medical Research Ethics Committee of the Ministry of Health Malaysia (NMRR-19-3076-49700), Universiti Malaya Medical Centre Ethics Committee (UMMC-071119), and the sponsor (ACCORD) (AC 19115) approved this study. We obtained formal permission from the relevant authorities, the Malaysian DOE and MetMalaysia, to access environmental data. We performed all methods in this study in accordance with relevant guidelines and regulations.

Data availability: The data that support the findings of this study are available from the corresponding authors upon reasonable request. Environmental data will require the relevant governmental agencies’ authorisation and permission.

Funding: The UK National Institute for Health Research (NIHR) (RESPIRE; 16/136/109) using UK aid from the UK Government to support global health research supported this study. The funder played no role in study design, data collection, analysis, and interpretation of data, or the writing of this manuscript. The views expressed in this publication are those of the authors and not necessarily those of the NIHR or the UK Government.

Authorship contributions: All authors contributed to the work in conceptualisation, conception, and design of the work. NH, CL, and WC were involved in analysing and writing the manuscript, and all authors reviewed and critically edited the manuscript for intellectual content. All authors have read and approved the final version of the manuscript.

Disclosure of interest: The authors completed the ICMJE Disclosure of Interest Form (available upon request from the corresponding authors) and declare the following activities and relationships: EMK, HC, HP, and CL report grants from the NIHR RESPIRE. EMK reports personal fees from AstraZeneca and General Practitioners Research Institute, and serves as a board director of the International Primary Care Respiratory Group, UK and Primary Care Respiratory Group Malaysia. HC reports being the co-editor-in-chief of the Journal of Global Health. All other authors declare no further conflicts of interest.

references

  1. National Institute for Health and Care Excellence. What is the prevalence of asthma? 2022. Available: https://cks.nice.org.uk/topics/asthma/background-information/ prevalence/. Accessed: 2 October 2025.
  2. Bryant-Stephens T. Asthma disparities in urban environments. J Allergy Clin Immunol. 2009;123:1199–206. doi: 10.1016/j.jaci.2009.04.030 PMID: 19501229
  3. Guarnieri M, Balmes JR. Outdoor air pollution and asthma. Lancet. 2014;383:1581–92. doi: 10.1016/S0140-6736(14)60617-6 PMID: 24792855
  4. Tiotiu AI, Novakova P, Nedeva D, Chong-Neto HJ, Novakova S, Steiropoulos P, et al.. Impact of Air Pollution on Asthma Outcomes. Int J Environ Res Public Health. 2020;17:6212. doi: 10.3390/ijerph17176212 PMID: 32867076
  5. Air Quality Expert Group. Nitrogen Dioxide in the United Kingdom. London, England: Air Quality Expert Group; 2004. Available: https://ukair.defra.gov.uk/library/assets/documents/reports/aqeg/nitrogen_dioxide_in_the_UK-summary.pdf. Accessed: 12 November 2025.
  6. World Health Organization. WHO Air quality guidelines for particulate matter, ozone, nitrogen dioxide and sulfur dioxide: global update 2005: summary of risk assessment. Geneva, Switzerland: World Health Organization; 2006. Available: https://apps.who.int/iris/handle/10665/69477. Accessed: 12 November 2025.
  7. Orellano P, Reynoso J, Quaranta N. Short-term exposure to sulphur dioxide (SO2) and all-cause and respiratory mortality: A systematic review and meta-analysis. Environ Int. 2021;150:106434. doi: 10.1016/j.envint.2021.106434 PMID: 33601225
  8. Air Quality Expert Group. Ozone in the United Kingdom. London, England: Air Quality Expert Group; 2009. Available: https://uk-air.defra.gov.uk/library/assets/documents/reports/aqeg/aqeg-ozone-report.pdf. Accessed: 12 November 2025.
  9. Air Quality Expert Group. Particulate matter in the United Kingdom. London, England: Air Quality Expert Group; 2005. Available: https://uk-air.defra.gov.uk/library/assets/documents/reports/aqeg/Particulate_Matter_in_The_UK_2005.pdf. Accessed: 12 November 2025.
  10. World Health Organization. What are the WHO Air quality guidelines? 22 September 2021. Available: https://www.who.int/news-room/feature-stories/detail/what-are-the-who-air-quality-guidelines. Accessed: 12 November 2025.
  11. Gordian ME, Ozkaynak H, Xue J, Morris SS, Spengler JD. Particulate air pollution and respiratory disease in Anchorage, Alaska. Environ Health Perspect. 1996;104:290–7. doi: 10.1289/ehp.96104290 PMID: 8919767
  12. Norris G. YoungPong SN, Koenig JQ, Larson TV, Sheppard L, Stout JW. An association between fine particles and asthma emergency department visits for children in Seattle. Environ Health Perspect. 1999;107:489–93. doi: 10.1289/ehp.99107489 PMID: 10339450
  13. Chatkin J, Correa L, Santos U. External Environmental Pollution as a Risk Factor for Asthma. Clin Rev Allergy Immunol. 2022;62:72–89. doi: 10.1007/s12016-020-08830-5 PMID: 33433826
  14. Lin W, Brunekreef B, Gehring U. Meta-analysis of the effects of indoor nitrogen dioxide and gas cooking on asthma and wheeze in children. Int J Epidemiol. 2013;42:1724–37. doi: 10.1093/ije/dyt150 PMID: 23962958
  15. Ashworth M, Analitis A, Whitney D, Samoli E, Zafeiratou S, Atkinson R, et al.. Spatio-temporal associations of air pollutant concentrations, GP respiratory consultations and respiratory Inhaler prescriptions: a 5-year study of primary care in the borough of Lambeth, South London. Environ Health. 2021;20:54. doi: 10.1186/s12940-021-00730-1 PMID: 33962646
  16. Afifa, Arshad K, Hussain N, Ashraf MH, Saleem MZ. Air pollution and climate change as grand challenges to sustainability. Sci Total Environ. 2024;928:172370. doi: 10.1016/j.scitotenv.2024.172370 PMID: 38604367
  17. Makrufardi F, Manullang A, Rusmawatiningtyas D, Chung KF, Lin SC, Chuang HC. Extreme weather and asthma: a systematic review and meta-analysis. Eur Respir Rev. 2023;32:230019. doi: 10.1183/16000617.0019-2023 PMID: 37286218
  18. Nakyai T, Santasnachok M, Thetkathuek A, Phatrabuddha N. Influence of meteorological factors on air pollution and health risks: A comparative analysis of industrial and urban areas in Chonburi Province, Thailand. Environ Adv. 2025;19:100608. doi: 10.1016/j.envadv.2024.100608
  19. Ahad A, Ming Khoo E. Asthma control and care among Malaysian primary school children: a cross-sectional study. Asia Pac J Public Health. 2017;29:422–9. doi: 10.1177/1010539517717766 PMID: 28719788
  20. Chan YY, Teh CH, Lim KK, Lim KH, Yeo PS, Kee CC, et al.. Lifestyle, chronic diseases and self- rated health among Malaysian adults: results from the 2011 National Health and Morbidity Survey (NHMS). BMC Public Health. 2015;15:754. doi: 10.1186/s12889-015-2080-z PMID: 26246019
  21. Institute for Public Health. National Health and Morbidity Survey (NHMS) 2023: Non-Communicable Diseases and Healthcare Demand. Selangor, Malaysia: Institute for Public Health; 2024. Available: https://iku.nih.gov.my/images/nhms2023/report-nhms-2023.pdf. Accessed: 1 December 2025.
  22. Institute for Public Health. Third National Health and Morbidity Survey (NHMS III) 2006 Asthma. Selangor, Malaysia: Institute for Public Health; 2008. Available: https://iku.nih.gov.my/images/IKU/Document/REPORT/2006/Asthma.pdf. Accessed: 2 November 2025.
  23. Ramakreshnan L, Aghamohammadi N, Fong CS, Bulgiba A, Zaki RA, Wong LP, et al.. Haze and health impacts in ASEAN countries: systematic review. Environ Sci Pollut Res Int. 2018;25:2096–111. doi: 10.1007/s11356-017-0860-y PMID: 29209970
  24. Hussein N, Liew SM, Hanafi NS, Lee PY, Cheong AT, Ghazali SS, et al.. Asthma control and care among six public health clinic attenders in Malaysia: A cross-sectional study. Health Sci Rep. 2023;6:e1021. doi: 10.1002/hsr2.1021 PMID: 37152232
  25. Ismail H, Syed Hussain TPR. Hubungan jerebu dengan kes asma; kajian di beberapa buah hospital di Lembah Klang. Akademika. 2001;59:117–27.
  26. Ming CR, Ban Yu-Lin A, Abdul Hamid MF, Latif MT, Mohammad N, Hassan T. Annual Southeast Asia haze increases respiratory admissions: A 2-year large single institution experience. Respirology. 2018;23:914–20. doi: 10.1111/resp.13325 PMID: 29923364
  27. Mohd Shafie SH, Mahmud M, Mohamad S, Rameli NL, Abdullah R, Mohamed AF. Influence of urban air pollution on the population in the Klang Valley, Malaysia: A spatial approach. Ecol Process. 2022;11:3. doi: 10.1186/s13717-021-00342-0
  28. Abd Rahim N, Ahmad Zaki R, Yahya A, Wan Mahiyuddin W. Acute effects of air pollution on cardiovascular hospital admissions in the port district of Klang, Malaysia: A time-series analysis. Atmos Environ. 2024;333:120629. doi: 10.1016/j.atmosenv.2024.120629
  29. Department of Environment. Chronology of Haze Episodes in Malaysia. Putrajaya, Malaysia: Department of Environment; 2021. Available: https://www.doe.gov.my/wp-content/uploads/2021/09/Chronology-of-Haze-Episodes-in-Malaysia.pdf. Accessed: 3 July 2025.
  30. Azmi SZ, Latif MT, Ismail AS, Juneng L, Jemain AA. Trend and status of air quality at three different monitoring stations in the Klang Valley, Malaysia. Air Qual Atmos Health. 2010;3:53–64. doi: 10.1007/s11869-009-0051-1 PMID: 20376168
  31. Petaling Jaya City Council. Background. 2020. Available: https://www.mbpj.gov.my/en/mbpj/profile/background. Accessed: 2 November 2025.
  32. Raja Gopal NG, Niltanan KM. Residents’ preferences on attributes of urban air quality improvement in Petaling Jaya, Selangor, Malaysia. J Sustain Sci Manag. 2022;17:112–35. doi: 10.46754/jssm.2022.02.010
  33. Department of Environment. Malaysia Environmental Quality Report. Putrajaya, Malaysia: Ministry of Natural Resources And Environmental Sustainability; 2015. Available: https://www.doe.gov.my/en/environmental-quality-report/. Accessed: 1 December 2025.
  34. Leh OLH, Mohamed Musthafa SNA, Abdul Rasam AR. Urban environmental health: Respiratory infection and urban factors in urban growth corridor of Petaling Jaya, Shah Alam and Klang, Malaysia. Sains Malays. 2014;43:1405–14.
  35. Latif MT, Othman M, Idris N, Juneng L, Abdullah AM, Hamzah WP, et al.. Impact of regional haze towards air quality in Malaysia: A review. Atmos Environ. 2018;177:28–44. doi: 10.1016/j.atmosenv.2018.01.002
  36. Gasparrini A, Guo Y, Hashizume M, Lavigne E, Zanobetti A, Schwartz J, et al.. Mortality risk attributable to high and low ambient temperature: a multicountry observational study. Lancet. 2015;386:369–75. doi: 10.1016/S0140-6736(14)62114-0 PMID: 26003380
  37. Mészáros D, Markos J, FitzGerald DG, Walters EH, Wood-Baker R. An observational study of PM10 and hospital admissions for acute exacerbations of chronic respiratory disease in Tasmania, Australia 1992–2002. BMJ Open Respir Res. 2015;2:e000063. doi: 10.1136/bmjresp-2014-000063 PMID: 25593705
  38. Choudhury AH, Gordian ME, Morris SS. Associations between Respiratory Illness and PM10 Air Pollution. Arch Environ Health. 1997;52:113–7. doi: 10.1080/00039899709602873 PMID: 9124870
  39. Zhang H, Liu S, Chen Z, Zu B, Zhao Y. Effects of variations in meteorological factors on daily hospital visits for asthma: A time-series study. Environ Res. 2020;182:109115. doi: 10.1016/j.envres.2020.109115 PMID: 31923850
  40. Kwon JW, Han YJ, Oh MK, Lee CY, Kim JY, Kim EJ, et al.. Emergency Department Visits for Asthma Exacerbation due to Weather Conditions and Air Pollution in Chuncheon, Korea: A Case-Crossover Analysis. Allergy Asthma Immunol Res. 2016;8:512–21. doi: 10.4168/aair.2016.8.6.512 PMID: 27582402
  41. United States Environmental Protection Agency. Health and Environmental Effects of Particulate Matter (PM). 2025. Available: https://www.epa.gov/pm-pollution/health-and-environmental-effects-particulate-matter-pm. Accessed: 6 July 2025.
  42. Cichowicz R, Wielgosiński G, Fetter W. Effect of wind speed on the level of particulate matter PM10 concentration in atmospheric air during winter season in vicinity of large combustion plant. J Atmos Chem. 2020;77:35–48. doi: 10.1007/s10874-020-09401-w
  43. Fu J, Liu Y, Zhao Y, Chen Y, Chang Z, Xu KF, et al.. Increased Risk of Hospital Admission for Asthma from Short-Term Exposure to Low Air Pressure. J Asthma Allergy. 2022;15:1035–43. doi: 10.2147/JAA.S367308 PMID: 35967095
  44. D’Amato G, Annesi MI, Molino A, Vitale C, D’Amato M. Thunderstorm-related asthma attacks. J Allergy Clin Immunol. 2017;139:1786–7. doi: 10.1016/j.jaci.2017.03.003 PMID: 28342913
  45. Gu W, Xie D, Li Q, Feng H, Xue Y, Chen Y, et al.. Association of humidity and precipitation with asthma: A systematic review and meta-analysis. Front Allergy. 2024;5:1483430. doi: 10.3389/falgy.2024.1483430 PMID: 39713043
  46. Zhou H, Geng H, Dong C, Bai T. The short-term harvesting effects of ambient particulate matter on mortality in Taiyuan elderly residents: A time-series analysis with a generalized additive distributed lag model. Ecotoxicol Environ Saf. 2021;207:111235. doi: 10.1016/j.ecoenv.2020.111235 PMID: 32942099
  47. Bronte-Moreno O, González-Barcala F, Muñoz-Gall X, Pueyo-Bastida A, Ramos-González J, Urrutia-Landa I. Impact of Air Pollution on Asthma: A Scoping Review. Open Respir Arch. 2023;5:100229. doi: 10.1016/j.opresp.2022.100229 PMID: 37496874

Correspondence to:
Wee Cheah
Institute of Ocean and Earth Sciences, Universiti Malaya
Kuala Lumpur, 50603
Malaysia
[email protected]
Ee Ming Khoo
Department of Primary Care Medicine, Faculty of Medicine, Universiti Malaya
Kuala Lumpur, 50603
Malaysia
[email protected]