Ambient Air Pollution and Outpatient Visits for Acute Respiratory Infection at Sentinel Health Clinics in Melaka, Malaysia: A Correlation Study From 2012 to 2015

<title>Abstract</title> <p> Ambient air pollution is an important environmental health concern, particularly during haze episodes in Southeast Asia. This study examined associations occurring on the same day between daily ambient air quality indicators and outpatient visits for acute respiratory infection (ARI) at two sentinel health clinics in Melaka, namely Klinik Kesihatan Bukit Rambai (KKBR) and Klinik Kesihatan Tengkera (KKT), from 2012 to 2015. The clinics were selected because they provide primary care for acute presentations and were sentinel facilities for surveillance of illnesses related to haze, which included upper respiratory tract infection (URTI), asthma and conjunctivitis. Secondary air quality data were obtained from two Department of Environment monitoring stations located in Melaka, at Sekolah Menengah Kebangsaan Bukit Rambai (SMKBR) and Sekolah Menengah Kebangsaan Tinggi Melaka (SMKT). Daily ARI clinic visit data were obtained from routine clinic reports. Weekends and public holidays were excluded because regular outpatient services were not provided on those days. Pearson correlation coefficients were used to assess associations between ARI visits and the Air Pollutant Index (API), PM ₁₀ , O ₃ , CO, SO ₂ and NO ₂ . At KKBR, significant positive correlations were observed for API in 2012, 2013 and 2015; PM ₁₀ in 2012, 2013 and 2015; and CO in 2015. Significant negative correlations were observed for NO ₂ in 2012 and for API and PM ₁₀ in 2014. At KKT, no statistically significant correlations were observed from 2012 to 2014, whereas API and PM ₁₀ were positively correlated with ARI visits in 2015. All statistically significant correlations were small in magnitude (|r| ≤ 0.269). The findings varied by year and location and did not show a consistent relationship between air pollution and respiratory health outcomes. Because the analysis was unadjusted and ecological, the results should not be interpreted as evidence of causation. </p>