THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PUBLIC HEALTH

Tekspenuh

(1)M al. ay a. RESPIRATORY HEALTH IMPACT OF HAZE EXPOSURE AND ITS FINANCIAL IMPLICATIONS. U. ni. ve. rs i. ty. of. MOHD HAFIZ BIN JAAFAR. FACULTY OF MEDICINE UNIVERSITY OF MALAYA KUALA LUMPUR 2019.

(2) M al. ay a. RESPIRATORY HEALTH IMPACT OF HAZE EXPOSURE AND ITS FINANCIAL IMPLICATIONS. of. MOHD HAFIZ BIN JAAFAR. U. ni. ve. rs i. ty. THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PUBLIC HEALTH. FACULTY OF MEDICINE UNIVERSITY OF MALAYA KUALA LUMPUR. 2019.

(3) UNIVERSITY OF MALAYA ORIGINAL LITERARY WORK DECLARATION. Name of Candidate: Mohd Hafiz Bin Jaafar Matric No: MWA 160004 Name of Degree: Doctor of Public Health Title of Project Paper/Research Report/Dissertation/Thesis: Respiratory Health Impact of Haze Exposure and Its Financial Implications. I do solemnly and sincerely declare that:. ay a. Field of Study: Public Health, Environmental Health, Health Economics. ve. rs i. ty. of. M al. (1) I am the sole author/writer of this Work; (2) This Work is original; (3) Any use of any work in which copyright exists was done by way of fair dealing and for permitted purposes and any excerpt or extract from, or reference to or reproduction of any copyright work has been disclosed expressly and sufficiently and the title of the Work and its authorship have been acknowledged in this Work; (4) I do not have any actual knowledge nor do I ought reasonably to know that the making of this work constitutes an infringement of any copyright work; (5) I hereby assign all and every rights in the copyright to this Work to the University of Malaya (“UM”), who henceforth shall be owner of the copyright in this Work and that any reproduction or use in any form or by any means whatsoever is prohibited without the written consent of UM having been first had and obtained; (6) I am fully aware that if in the course of making this Work I have infringed any copyright whether intentionally or otherwise, I may be subject to legal action or any other action as may be determined by UM. Date:. U. ni. Candidate’s Signature. Subscribed and solemnly declared before, Witness’s Signature. Date:. Name: Designation:. ii.

(4) RESPIRATROY HEALTH IMPACT OF HAZE EXPOSURE AND ITS FINANCIAL IMPLICATIONS ABSTRACT Haze imposes a substantial disease burden on populations, especially those in the Southeast Asian region. In Malaysia, haze episodes have been primarily associated with transboundary sources following biomass burning of peat, soil and forests in. ay a. neighbouring countries. Local anthropogenic sources, especially from domestic waste burning, motor vehicle emissions and industrial production, also contributed. M al. significantly to the occurrence of haze episodes in this rapidly developing country. It is important to determine the trends in haze episodes and in associated healthcare utilisation and healthcare costs, particularly for respiratory illnesses. This study aimed. of. to describe the haze situation and factors associated with haze episodes in Selangor. It also sought to determine healthcare utilisation and healthcare costs incurred by public. ty. healthcare providers. A cross-sectional study was conducted using secondary data from. rs i. 2012 to 2015 on air pollutants and ecological factors obtained from the Department of. ve. Environment and healthcare utilisation data retrieved from the Ministry of Health Malaysia and the University Malaya Medical Centre. The calculation of costs associated. ni. with managing haze-related respiratory illnesses was based on the unit cost derived. U. from other studies. This was used to compare the direct medical cost for treatment of haze-related respiratory illnesses during haze and non-haze episodes. In this study, haze episodes were defined based on average monthly PM10 readings. Between 2012 and 2015, a total of 129 (67.2%) haze episodes were recorded at four selected stations in Selangor. The haze episodes were associated with higher healthcare utilisation due to haze-related respiratory illnesses in general and for both inpatient and outpatient visits (p<0.05). The numbers of inpatient and outpatient visits during haze episodes were 74 iii.

(5) (± 62.1) and 320 (± 650.1), respectively, compared to 34 (± 16.5) and 146 (± 170.5), respectively, during non-haze episodes. The four-year cumulative direct medical cost differences between haze and non-haze episodes were RM 13.4 million for inpatient cases and RM1.4 million for outpatient cases. The total cost difference during haze and non-haze episodes was approximately RM14.9 million. Findings showed that more haze episodes than non-haze episodes were recorded between 2012 and 2015 in Selangor.. ay a. The deterioration in air quality due to haze episodes creates a significant financial burden for the public healthcare system as the result of increased utilisation. It is essential to implement a mitigation policy and to educate the population in order to. M al. reduce exposure to haze and to ensure that adequate financial and human resources are available in the relevant healthcare facilities to minimise the health and financial. of. impacts of haze.. U. ni. ve. rs i. ty. Keywords: Financial implications, haze, respiratory illnesses, Selangor. iv.

(6) IMPAK KESIHATAN RESPIRATORI DAN IMPLIKASI KEWANGAN DISEBABKAN PENDEDAHAN TERHADAP JEREBU ABSTRAK Jerebu memberikan impak kesihatan yang besar kepada penduduk di rantau Asia Tenggara. Di Malaysia, jerebu sering dikaitkan dengan sumber rentas sempadan yang berpunca daripada pembakaran biomas yang melibatkan tanah gambut dan hutan di. ay a. negara jiran. Selain itu, faktor antropogenik tempatan terutamanya daripada pembakaran sisa domestik, pelepasan asap kenderaan bermotor dan pengeluaran perindustrian turut menyumbang ke arah berlakunya jerebu di negara yang sedang pesat membangun. M al. seperti Malaysia.. Oleh itu, ianya sangat penting untuk mengenal pasti pola jerebu dan implikasinya terhadap penyakit respiratori dari segi penggunaan fasiliti kesihatan dan. of. kos perubatan. Kajian ini bertujuan untuk menerangkan situasi jerebu dan faktor yang berkait dengan jerebu di Selangor. Ia juga dijalankan untuk mengkaji penggunaan. ty. fasiliti kesihatan awam dan kos perubatan bagi penyakit respiratori berkaitan dengan. rs i. jerebu yang ditanggung oleh pembekal servis kesihatan awam di Selangor. Kajian secara rentas telah dijalankan melalui pengumpulan data sekunder melibatkan data. ve. untuk pencemaran udara dan faktor ekologi bagi tahun 2012 hingga 2015 yang. ni. diperolehi dari Jabatan Alam Sekitar dan data bagi penggunaan fasiliti kesihatan awam. U. bagi penyakit respiratori berkaitan dengan jerebu yang diperolehi daripada Kementerian Kesihatan Malaysia dan Pusat Perubatan Universiti Malaya. Data bagi kos untuk merawat penyakit respiratori berkaitan dengan jerebu pula berdasarkan kos unit yang diperolehi daripada kajian yang pernah diterbitkan sebelum ini dimana kos perubatan langsung untuk merawat penyakit respiratori berkaitan dengan jerebu sewaktu jerebu dan tiada jerebu dibandingkan, Di dalam kajian ini, episod jerebu ditentukan melalui purata bacaan PM10 di dalam satu bulan. Diantara tahun 2012 hingga 2015, terdapat 129 v.

(7) (67.2%) episod jerebu direkodkan di empat stesyen di Selangor. Jerebu mempunyai kaitan dengan penggunaan fasiliti kesihatan bagi kes penyakit respiratori berkaitan dengan jerebu yang lebih tinggi secara keseluruhannya dan juga bagi pesakit dalam dan pesakit luar (p <0.05). Jumlah pesakit dalam dan pesakit luar sewaktu jerebu adalah 74 (± 62.1) dan 320 (± 650.1) kes berbanding 34 ± (16.5) kes pesakit dalam dan 146 (± 170.5) kes pesakit luar sewaktu tiada jerebu. Perbezaan kos perubatan secara langsung. ay a. selama empat tahun sewaktu berlakunya jerebu dan tiada jerebu berjumlah RM 13.4 juta untuk kes pesakit dalam dan RM1.4 juta untuk kes pesakit luar. Jumlah keseluruhan perbezaan kos sewaktu berlakunya jerebu dan tiada jerebu adalah kira-kira RM14.9 juta.. M al. Hasil dapatan kajian ini menunjukkan lebih banyak episod jerebu direkodkan diantara tahun 2012 ke 2015 di Selangor. Kemerosotan tahap kualiti udara disebabkan jerebu. of. memberikan kesan yang besar terhadap beban kewangan kepada kesihatan awam akibat daripada peningkatan dalam penggunaan fasiliti kesihatan. Disamping memastikan. ty. langkah pencegahan melalui pelaksanaan dasar mitigasi dan mendidik masyarakat untuk perancangan untuk menyediakan. rs i. mengurangkan pendedahan terhadap jerebu,. peruntukan kewangan dan sumber manusia yang mencukupi di fasiliti kesihatan yang. ve. terlibat juga penting untuk meminimakan kesan jerebu terhadap kesihatan dan. ni. kewangan.. U. Kata kunci: Jerebu, penyakit respiratori, implikasi kewangan, Selangor. vi.

(8) ACKNOWLEDGEMENTS I would like to express my deepest gratitude and warm appreciation to my supervisors, Professor Dr. Maznah Dahlui and Dr. Marzuki Isahak for their mentorship and unwavering support throughout this journey.. Thank you for all your. encouragements, comments, assistance and contributions towards the completion of this research. All your guidance will be remembered and may all your kindness be rewarded. ay a. in the hereafter.. I would also like to thank all my friends and colleagues in the Community Health. M al. Unit, Primary Care Department, Universiti Sains Islam Malaysia and also lecturers and supporting staff in the Department of Social and Preventive Medicine, University of Malaya for all the support given. To all my batchmates, DrPH cohort 2016-2019, who. of. have shared their knowledge, excitement and most of the time nervousness, thank you. ty. for all the memories.. rs i. Last but not least, to my beloved wife, Dr. Amirah Azzeri and my two adorable children, Amani Haniya Mohd Hafiz and Afnan Hanif Mohd Hafiz, thank you for all the. ve. sacrifice, encouragement, motivation and never ending support given to smoothen my. ni. path towards success. To my loving and caring parents, Jaafar Mohd Nor and Raedah. U. Md Dahan, my siblings and in laws, and all other family members, thank you so much for your pray and assistant.. Above all, all praise to the Almighty for His kindness and generosity, without which this study would not have come forth.. vii.

(9) TABLE OF CONTENTS. Abstract ...................................................................................................................... iii Abstrak ......................................................................................................................... v Acknowledgements ..................................................................................................... vii Table of Contents ....................................................................................................... viii List of Figures ............................................................................................................ xiii. ay a. List of Tables ............................................................................................................. xiv. M al. List of Symbols and Abbreviations ............................................................................ xvi. CHAPTER 1: INTRODUCTION............................................................................... 1 Introduction ......................................................................................................... 1. 1.2. Haze Phenomenon in Malaysia ............................................................................ 3. 1.3. Health Impact and Disease Burden of Haze.......................................................... 5. 1.4. Problem Statement ............................................................................................... 8. 1.5. Rationale and Justification ................................................................................... 9. 1.6. Research Questions ............................................................................................ 10. ve. rs i. ty. of. 1.1. Study Objectives ................................................................................................ 11. 1.8. Summary and Organisation of The Thesis .......................................................... 11. U. ni. 1.7. CHAPTER 2: LITERATURE REVIEW ................................................................. 13 2.1. Introduction ....................................................................................................... 13. 2.2. CAQM station ................................................................................................... 15. 2.3. API and Specific Pollutants Assessment ............................................................ 20. 2.4. Definition of Haze Based on Malaysian and WHO Air Quality Guidelines ........ 23. 2.5. Trend in Air Pollutions and Factors Associated with Haze Episode in Malaysia . 26 viii.

(10) 2.6. Healthcare Utilisation for Haze-related Illnesses ................................................ 28. 2.7. Financial Implications of Haze-related Respiratory Illnesses .............................. 36 2.7.1 Cost Centre and Cost Component ........................................................... 38 2.7.2 Health-related Financial Implications of Haze ........................................ 38. 2.8. Non-health-related Financial Impact of Haze ..................................................... 50 2.8.1 Tourism.................................................................................................. 50. ay a. 2.8.2 Transportation ........................................................................................ 51 2.8.3 Environmental ........................................................................................ 52 Conceptual Framework ...................................................................................... 55. M al. 2.9. CHAPTER 3: METHODOLOGY ........................................................................... 57 Introduction ....................................................................................................... 57. 3.2. Study Design ..................................................................................................... 57. 3.3. Study Period ...................................................................................................... 58. 3.4. Study Area ......................................................................................................... 58. 3.5. Study Population................................................................................................ 64. 3.6. Sampling Method .............................................................................................. 64. 3.7. Exposures and Outcomes ................................................................................... 65. ni. ve. rs i. ty. of. 3.1. Data Collection .................................................................................................. 66 3.8.1 Data collection for air pollutants and ecological parameters.................... 66. U. 3.8. 3.8.2 Data collection for healthcare utilisation of haze-related respiratory illnesses ................................................................................................ 67 3.8.3 Data collection for financial implications of haze-related respiratory illnesses ................................................................................................ 69. ix.

(11) 3.9. Data Management .............................................................................................. 73 3.9.1 Air pollutants and ecological factors ....................................................... 73 3.9.1.1 Data cleaning .......................................................................... 73 3.9.1.2 Management of missing data .................................................. 73 3.9.2 Healthcare utilisation of haze-related respiratory illnesses ...................... 74 3.9.3 Financial implications of haze-related respiratory illnesses ..................... 75. ay a. 3.10 Data Analysis .................................................................................................... 75 3.10.1 Air pollutants and ecological factors ..................................................... 76 3.10.2 Healthcare utilisation of haze-related respiratory illnesses ..................... 78. M al. 3.10.3 Determination of treatment cost of respiratory diseases ......................... 81 3.11 Ethical consideration .......................................................................................... 85. of. 3.12 Operational Definitions ...................................................................................... 86. ty. 3.13 Flow Chart ......................................................................................................... 88. rs i. CHAPTER 4: TREND AND FACTORS ASSOCIATED WITH HAZE EPISODE IN SELANGOR ...................................................................................................... 89 Introduction ....................................................................................................... 89. 4.2. Methodology ..................................................................................................... 90. ni. ve. 4.1. 4.2.1 Management of missing data .................................................................. 90. U. 4.2.2 Data analysis and presentation ................................................................ 92. 4.3. Results .............................................................................................................. 93 4.3.1 Air pollutants and ecological factors levels ............................................. 93 4.3.2 Distribution and pattern of air pollutants and ecological factors levels .... 96 4.3.3 Association between air pollutants and ecological factors with haze episodes .............................................................................................. 102 x.

(12) 4.3.4 Multivariate analysis: Factors associated with haze episode .................. 103 4.4. Discussion ....................................................................................................... 106 4.4.1 Pattern and distribution of air pollutant and ecological factor readings.. 106 4.4.2 Factors associated with haze episodes................................................... 110 4.4.3 Respond to haze situation ..................................................................... 113 Summary ......................................................................................................... 115. CHAPTER. 5:. HEALTHCARE. ay a. 4.5. UTILISATION. AND. FINANCIAL. IMPLICATIONS OF HAZE-RELATED RESPIRATORY ILLNESSES IN .................................................................................................... 117. M al. SELANGOR. Introduction ..................................................................................................... 117. 5.2. Methodology ................................................................................................... 118. 5.3. Results ............................................................................................................ 120. of. 5.1. ty. 5.3.1 Healthcare utilisation of haze-related respiratory illnesses .................... 120. 5.4. rs i. 5.3.2 Financial implications of haze-related respiratory illnesses ................... 134 Discussion ....................................................................................................... 137. ve. 5.4.1 Healthcare utilisation for haze-related respiratory illnesses ................... 137. ni. 5.4.2 Financial implications of haze-related respiratory illnesses ................... 142 Summary ......................................................................................................... 148. U. 5.5. CHAPTER 6: CONCLUSION ............................................................................... 149 References ................................................................................................................ 153 List of Publications and Papers Presented ................................................................. 163 APPENDIX A: KEYWORDS AND SEARCHING STRATEGIES .......................... 165 APPENDIX B: CHEC-LIST FOR QUALITY ASSESSMENT OF ECONOMIC EVALUATION STUDY .......................................................................................... 166 xi.

(13) APPENDIX C: COMPARISON OF COSTS FROM MULTIPLE CENTRES (INPATIENT)........................................................................................................... 167 APPENDIX D: COMPARISON OF COSTS FROM MULTIPLE CENTRES (OUTPATIENT) ....................................................................................................... 168 APPENDIX E: ETHICS APPROVAL LETTERS ..................................................... 169. U. ni. ve. rs i. ty. of. M al. ay a. APPENDIX F: CHRONOLOGY OF HAZE EPISODES IN MALAYSIA ................ 172. xii.

(14) LIST OF FIGURES. Figure 2.1: Process of calculating the API value from CAQM stations. ....................... 22 Figure 3.1: Map of Selangor and its districts. .............................................................. 60 Figure 3.2: Location of Selangor, Sumatra and Straits of Malacca and the direction of wind during Southwest monsoon season). ................................................................... 61 Figure 3.3: Location of hospitals under the MOH in Selangor ..................................... 62. ay a. Figure 3.4: Location of CAQM stations in Selangor .................................................... 63 Figure 3.6: Study flow chart. ....................................................................................... 88. M al. Figure 4.1: Trend of wind speed level in Selangor 2012-2015. .................................... 98 Figure 4.2: Trend of temperature level in Selangor 2012-2015. ................................... 98 Figure 4.3: Trend of humidity level in Selangor 2012-2015......................................... 99. of. Figure 4.4: Trend of PM10 (µg/m3) level in Selangor 2012-2015. ................................. 99. ty. Figure 4.5: Trend of CO (µg/m3) level in Selangor 2012-2015. ................................. 100. rs i. Figure 4.6: Trend of SO2 (µg/m3) level in Selangor 2012-2015.................................. 100 Figure 4.7: Trend of NO2 (µg/m3) level in Selangor 2012-2015. ................................ 101. U. ni. ve. Figure 4.8: Trend of O3 (µg/m3) level in Selangor 2012-2015. ................................... 101. xiii.

(15) LIST OF TABLES. Table 2.1: Instruments used to measure parameters at CAQM station ......................... 19 Table 2.2: New Malaysia Ambient Air Quality Standard ............................................. 24 Table 2.3: Updated WHO Air Quality Guidelines values ............................................ 25 Table 2.4: Summary of systematic reviews and meta-analyses on healthcare utilisation associated with PM levels ........................................................................................... 33. ay a. Table 2.5: Summary of studies included in the systematic review: Health-related financial implications of air pollution in Asia .............................................................. 42. M al. Table 3.1: Source of studies for costing analysis ......................................................... 72 Table 4.1 Summary of missing data for air pollutants and ecological factors ............... 92. of. Table 4.2: Levels of daily air pollutants and ecological factors levels in Selangor 20122015............................................................................................................................ 95. ty. Table 4.3: Association between air pollutants and ecological factors with haze episode ................................................................................................................................. 104. rs i. Table 4.4: Crude and adjusted odds ratios in simple binary and multiple logistic regression analyses ................................................................................................... 105. ve. Table 4.5: Final model of multiple logistic regression analysis on factors associated with haze episode ............................................................................................................. 106. ni. Table 5.1: Healthcare utilisation for AEBA and AECOPD in public healthcare facilities for all studied districts 2012-2015 ............................................................................. 121. U. Table 5.2: Number of haze episodes and socio-demographic characteristics of AEBA and AECOPD cases 2012-2015 (from UMMC data) ................................................. 122 Table 5.3: Association between healthcare utilisation and haze episodes ................... 124 Table 5.4: Correlations between healthcare utilisation and other pollutants and ecological factors ...................................................................................................... 126 Table 5.5: Association between healthcare utilisation and gender (inpatient) ............. 128 Table 5.6: Association between healthcare utilisation and gender (outpatient) ........... 128 xiv.

(16) Table 5.7: Association between healthcare utilisation and ethnicity (inpatient) .......... 129 Table 5.8: Association between healthcare utilisation and ethnicity (outpatient) ........ 129 Table 5.9: Association between healthcare utilisation and age ................................... 130 Table 5.10: Crude and adjusted odds ratios in univariate and multivariate regression analyses (inpatient) ................................................................................................... 132 Table 5.11: Crude and adjusted odds ratios in univariate and multivariate regression analyses (outpatient) ................................................................................................. 133. U. ni. ve. rs i. ty. of. M al. ay a. Table 5.12: Direct medical cost of haze-related respiratory illnesses and cost difference during haze and non-haze episodes ........................................................................... 136. xv.

(17) LIST OF SYMBOLS AND ABBREVIATIONS. :. Acute Exacerbation of Bronchial Asthma Acute Exacerbation of Chronic Obstructive Pulmonary Disease. AHC. :. Amended Human Capital. AQG. :. Air Quality Guideline. ASEAN. :. Association of Southeast Asian Nations. CAQM. :. Continuous Air Quality Monitoring. CHEC. :. Consensus Health Economic Criteria. CO. :. Carbon Monoxide. CVA. :. Cerebro Vascular Accident. CVS. :. Cardio Vascular System. DOE. :. Department of Environmental. HC. :. Human Capital. HIC. :. Health Informatics Centre. IHD. M al. of. ty. :. Ischaemic Heart Disease. :. Inter-quartile range. :. Kolmogorov-Smirnov. ve. IQR. ay a. AECOPD :. rs i. AEBA. ni. KS. :. Ministry of Education. MOH. :. Ministry of Health. NO2. :. Nitrogen Dioxide. O3. :. Ozone. PM10. :. Particulate matter 10. PM2.5. :. Particulate matter 2.5. SO2. :. Sulphur dioxide. U. MOE. xvi.

(18) :. Standard deviation. SW. :. Shapiro-Wilk. UMMC. :. University Malaya Medical Centre. VOSL. :. Value of Statistical Life. WHO. :. World Health Organization. WTP. :. Willingness to Pay. U. ni. ve. rs i. ty. of. M al. ay a. SD. xvii.

(19) CHAPTER 1: INTRODUCTION 1.1. Introduction. Haze is a condition associated with the disruption of visibility, clarity and transparency in an area due to the presence of fine suspended particles (Cheng et al., 2013; Othman, Sahani, Mahmud, & Sheikh Ahmad, 2014). Specifically, it is defined as “an aggregation in the atmosphere of very fine, widely dispersed, solid or liquid. ay a. particles, or both, giving the air an opalescent appearance that subdues colours’’ (Hyslop, 2009, p. 182).. M al. There are two major components of haze, particles and gaseous pollutants. Particles are formed by suspended liquid or solid elements and are commonly referred to as. of. particulate matter (PM) (Hyslop, 2009). PM consists of “nitrates, sulphates, elemental and organic carbon, organic compounds (polycyclic aromatic hydrocarbons), biological. ty. compounds (endotoxin, cell fragments), and a variety of metals (iron, copper, nickel,. rs i. zinc, and vanadium)” (Brook et al., 2004, p. 2656). Primary particles are those released directly into the environment, while secondary particles are formed as a product of. ve. physicochemical transformation of gaseous pollutants (Brook et al., 2004).. ni. The other component, gaseous pollutants, are made up of carbon monoxide (CO),. U. sulphur dioxide (SO2), nitrogen dioxide (NO2) and ozone (O3). NO2 is one of the compounds of the family of nitrogen oxides (NOx), along with nitric oxide (NO), nitrogen trioxide, nitrogen tetroxide, and di-nitrogen pentoxide. Fossil fuel combustion results in the formation of atmospheric nitrogen (N 2). The oxidation process of atmospheric N2 produces NO and subsequently NO2. CO is produced through incomplete combustion of carbonic materials and is a good indicator of the presence of pollution associated with combustion. SO2 can combine with water to produce 1.

(20) sulphurous acid, which causes considerable irritation, especially in the eyes, skin and mucous membrane. The main source of SO2 is the combustion and roasting of sulphurcontaining materials such as metal sulphide ores. Finally, O 3 is formed in the troposphere through the reaction between reactive hydrocarbon (from combustion), UV light and NO2 (Brook et al., 2004).. There are three major sources of haze— mobile sources, stationary sources and open. ay a. burning sources (Afroz, Hassan, & Akma, 2003). Mobile sources involve land transport and the use of motor vehicles. These are the major contributors and account for almost. M al. 70% of total air pollution in Malaysia (Afroz et al., 2003). Rapid economic growth and greater purchasing power have led to an increase in the number of individual motor vehicle owners. In 2009, there were almost 19.02 million registered motor vehicles in. of. Malaysia, the majority of which were concentrated in the Klang Valley (Abdullah, Abu Samah, & Jun, 2012), where traffic congestion is a daily occurrence, especially during. ty. morning and evening peak hours. The emissions from motor vehicle exhaust. rs i. significantly reduce air quality in the affected area (Afroz et al., 2003).. ve. Stationary sources of air pollution include factories, power stations and industrial. ni. fuel burning activities and contribute to approximately 20% of total air pollution in. U. Malaysia (Afroz et al., 2003). Industrial development and economic growth have led to the development of new commercial activities. Urbanisation has created demand for products related, for instance, to construction, machinery and other forms of heavy industry, which in turn encourages manufacturers and suppliers to increase their production rate. More factories have opened, resulting in more emission of pollutants to the environment (Abdullah et al., 2012).. 2.

(21) The final source of haze, open burning activities, is due to either forest fires or solid waste burning and accounts for approximately 10% of total air pollution in Malaysia. It can originate from a local source or migrate from a transboundary source. Haze that occurs as a result of open burning from a transboundary source usually produces a greater burden on affected countries (Afroz et al., 2003). Although it is not a constant occurrence and accounts for only a small percentage of total haze, transboundary haze. ay a. from the neighbouring country of Indonesia has resulted in more harmful conditions and severe complications than other sources due to the intensity of the air pollutants it. 1.2. M al. produces (Afroz et al., 2003).. Haze Phenomenon in Malaysia. of. Over recent decades, Malaysia’s transformation from an agricultural-based to a more industrial-based country has led to rapid urbanisation and an increase in manufacturing. ty. and industrial processing activities (Abdullah et al., 2012). Migration to urban areas. rs i. such as Klang Valley has resulted in a dramatic increase in population density and traffic congestion. Exhaust emissions from motor vehicles and pollutants from industrial. ve. activities, in that order, are the main domestic sources of haze in Malaysia (Abdullah et. ni. al., 2012; Afroz et al., 2003; Mohd Shahwahid, 2016). Transboundary air pollutants. U. from neighbouring countries also contribute significantly to the occurrence of haze episode in Malaysia. Along with other countries in the Southeast Asian region, Malaysia has been affected several times by haze episodes due to open forest burning in Indonesia (Mohd Shahwahid, 2016; Othman et al., 2014).. The first transboundary haze episode in Malaysia occurred in April 1981. The concentration of PM10 recorded was 430g/m3, which was nine times higher than the limit set by the World Health Organization (WHO) at that time. It was followed by a 3.

(22) second episode of haze in September 1982, which affected Petaling Jaya, Selangor. The recorded PM10 was lower (244g/m3) than during the first episode. Both of the incidents were related to the open burning of agricultural waste in Indonesia (Mohd Shahwahid, 2016).. Subsequently, haze episodes were recorded in August 1990, June and October 1991, and from August to October 1994. The worst episode occurred in 1997 when the whole. ay a. country was covered with thick smoke from Kalimantan and Sumatra. During that time, the Malaysian government declared an emergency in some states, including Sarawak. M al. and Johor, due to the hazardous air pollution index (API) reading which was greater than 300 (PM10 > 420 µg/m3) and led to the closure of schools in the affected area (Mohd Shahwahid, 2016; Othman et al., 2014). Several minor haze episodes were. of. recorded in 2005, 2006 and 2010 followed by severe episodes in 2013, with Muar in. ty. Johor recording the highest API reading of 641 (Mohd Shahwahid, 2016).. rs i. The latest severe haze episode occurred from September to November 2015. It was initiated in late June 2015 by forest fires resulting from illegal slash-and-burn activities. ve. conducted by firms and farmers in Indonesia. By the end of August 2015 it had spread. ni. to the neighbouring countries of Malaysia, Singapore, Thailand, Vietnam and Brunei. U. (‘Haze chokes Indonesia, Malaysia and Singapore’, 2015). During this haze episode, almost all states were affected, especially in the west-coast area of Peninsular Malaysia. Among the states that were badly affected were Selangor, Kuala Lumpur, Putrajaya, Negeri Sembilan, Malacca and some parts of Sarawak. The highest API level was recorded in Shah Alam, Selangor, with an API reading of 308. The conditions were aggravated by a dry season and the occurrence of an El Nino phenomenon during that time (‘Haze worsens nationwide, API in Shah Alam hits “hazardous” level’, 2015). 4.

(23) As in the 1997 event, the Ministry of Education decided to close schools once the API readings exceeded 200. On 15 September 2015, all schools in Selangor, Kuala Lumpur, Putrajaya, Malacca, Negeri Sembilan and Sarawak were closed due to these very unhealthy API readings (>200). Schools were ordered to close again for two days on 4 October 2015 in all states except Kelantan, Sabah and Sarawak. Almost 2,696,110 students and 4,778 schools were affected by the haze episode in 2015 (Chapman, 2015;. 1.3. ay a. ‘Ministry orders closure of schools in three states’, 2015).. Health Impact and Disease Burden of Haze. M al. Haze poses serious and recurring medical problems, especially among susceptible individuals such as those suffering from chronic diseases, children, the elderly and. of. pregnant women. The burden of illness has been shown to be directly proportionate to the intensity of haze due to higher healthcare utilisation and related healthcare costs. ty. during the haze episode.. rs i. Haze can result in both local and systemic health complications. Local effects are. ve. caused by irritant and allergic reactions triggered by the pollutants and include eye inflammations such as conjunctivitis and scleritis. These pollutants can irritate the nasal. ni. and throat areas and lead to increased mucous production by the cell lining. Eventually,. U. the excessive mucous secretion causes a blockage in the respiratory tract. Haze can also cause skin irritation, especially in those who have underlying eczema or other skin conditions. The local effects of haze are usually short-term and self-limiting (Gao et al., 2015; Othman et al., 2014; Wan Mahiyuddin et al., 2012).. Apart from these local effects, haze is commonly associated with cardiovascular (CVS) and respiratory system problems. The pathophysiology of haze-related 5.

(24) cardiovascular diseases is related to the systemic inflammatory process. This process triggers the formation of atherosclerosis plaque within the coronary blood vessels through the action of Interleukins-6 (IL-6) and C-reactive protein (CRP). On the other hand, exposure to PM can also activate a coagulation cascade, which in turn leads to thrombus formation. Thrombus causes blockage of the blood vessels and disrupted blood flow to the heart. Both factors can produce subsequent exacerbation of ischaemic. ay a. heart disease and heart failure (Shah et al., 2012).. In the respiratory system, the pathophysiology of disease is mainly due to the. M al. inflammatory reaction and formation of pulmonary oxidative stress. Exposure of the epithelial lining along the respiratory tract to PM triggers the release of inflammatory cytokines and reactive oxygen species (ROS). This process results in pulmonary. of. damage, thus affecting overall lung function. PM can also aggravate the immunoglobulin E (IgE) responses that lead to asthmatic attacks and atopy. Other. ty. effects of PM on the respiratory system include “increased airway responsiveness to. rs i. methacholine, increased neutrophil numbers in bronchial lavage, decreased CO. ve. diffusion capacity, and decreased maximum mid-expiratory flow” (Anderson,. ni. Thundiyil, & Stolbach, 2012, p. 170).. U. The health impacts of air pollution are more pronounced on the respiratory system. than on the CVS. Healthcare utilisation related to CVS illnesses mainly involves inpatient rather than outpatient cases, as most of these conditions require patients to be admitted for further monitoring and assessment. In contrast, exposure to air pollution has an immediate impact on the respiratory system, which makes it easier for researchers to attribute the episode to air pollution (Adar, Filigrana, Clements, & Peel, 2014; Sahani et al., 2014). The main reasons for respiratory-related outpatient visits and 6.

(25) hospital admissions are asthmatic attack, acute exacerbation of chronic obstructive pulmonary disease (AECOPD), acute bronchitis, pneumonia and bronchiolitis (in infants), with exacerbation of asthma and chronic obstructive pulmonary disease accounting for the majority of cases (Anderson et al., 2012; Laumbach & Kipen, 2014; Mehta, Shin, Burnett, North, & Cohen, 2013; Peacock et al., 2011).. Apart from the acute health effects, many epidemiological studies have been. ay a. conducted to determine the association between haze exposure and long-term respiratory health disorders such as lung cancer. A study conducted during an episode of. M al. peat fires in Kalimantan in 2009 found that “4 or 5 individuals out of 1000 exposed to smoke haze can be affected by cancer after prolonged exposure to high concentrations of carcinogenic metals in PM2.5 emitted from peat fires” (Betha et al., 2013, p. 577).. of. Findings from the European Study of Cohorts for Air Pollution Effects (ESCAPE) indicated that exposure to air pollution, particularly PM, increased the risk of. ty. developing lung cancer, with a hazard ratio of 1.22 for PM 10 and 1.18 for PM2.5. rs i. (Raaschou-nielsen et al., 2013). A study in Japan also found that long-term exposure to. ve. ambient air pollution could increase the risk of developing lung cancer, with hazard ratios ranging from 1.17 for NO2 to 1.24 and 1.26 for PM2.5 and SO2, respectively. U. ni. (Katanoda et al., 2011).. Premature mortality can also be attributed to haze exposure. In 2012, approximately. 3 million deaths worldwide were attributable to ambient air pollution. The WHO Western Pacific (1.1 million cases) and South East Asian (799 000 cases) regions contributed the majority of these cases (Gumy & Pruss-Ustun, 2016). Premature mortality can be due to either respiratory causes or natural causes. Premature mortality due to respiratory causes occurs one to two days after exposure to air pollution, usually 7.

(26) among those with underlying respiratory illnesses such as asthma and COPD (Sahani et al., 2014). Deaths from natural causes usually occur two to five days following exposure to air pollution among vulnerable groups such as young children, the elderly, pregnant women and those from low socioeconomic status backgrounds (Sahani et al., 2014).. Health complications due to haze-related illnesses are significantly associated with increased healthcare utilisation and reduced productivity due to work absenteeism. Both. ay a. transboundary and local sources that result in haze episodes have been found to play a significant role in the increased use of healthcare facilities (Brauer & Jamal, 1998;. M al. Othman et al., 2014) and, hence, to increased healthcare cost and expenditure. Together with the loss of productivity due to complications from haze-related illnesses, the health impact of haze episodes produces a significant financial burden for both healthcare. of. providers and patients (Kochi, Donovan, Champ, & Loomis, 2010; Othman et al., 2014). The healthcare utilisation rate and financial implications of haze-related illnesses. Problem Statement. ve. 1.4. rs i. ty. are discussed in detail in Chapter 2.. Malaysia faces periodic intense exposure to haze from both domestic sources (such. ni. as motor vehicles and industrial activities) and international sources (such as open forest. U. fires from neighbouring countries). Despite the implementation of various precautionary measures, the problem persists, with the latest severe haze episode being recorded in 2015.. The respiratory system accounts for the highest proportion of haze-related illnesses. The two main diseases of the respiratory system—acute exacerbation of bronchial asthma and chronic obstructive pulmonary disease—are significantly associated with 8.

(27) increased utilisation of public healthcare facilities, in comparison to other types of illnesses. During the 1997 haze crisis, Sarawak, Kuala Lumpur and Selangor recorded similar trends of outpatient visits to public healthcare facilities due to haze-related respiratory illnesses. In Kuching, a two- to three-fold increase in outpatient visits was observed during that period. Similarly, Kuala Lumpur recorded an increase in respiratory disease-related outpatient visits from 250 to 800 patients per day, while. ay a. Selangor recorded increases in asthma patients from 912 to 5,000 and in patients with acute respiratory infection from 6,000 to 30,000 (Afroz et al., 2003; Othman et al.,. M al. 2014).. The economic losses and increased healthcare cost and expenditure associated with haze-related illnesses are significant. In Malaysia, the total cost of haze-related illness. of. incurred by the government of Malaysia during 1997 and 2013 haze episodes were RM 19.1 million and RM 1.5 billion, which comprised approximately 0.3% and 0.48%,. ty. respectively, of the country’s total gross domestic product (Mohd Shahwahid, 2016;. Rationale and Justification. ve. 1.5. rs i. Othman & Mohd Shahwahid, 1999).. ni. It is important for Malaysia to scientifically document the trends in haze incidence. U. and healthcare utilisation in Malaysia so that steps can be taken to combat haze and allocate appropriate resources to meet the healthcare demands associated with haze episodes.. As respiratory illness constitutes the major medical burden during haze, data on the financial burden of respiratory illness will facilitate decision-making and resource allocation for healthcare services in Malaysia. However, no previous study has 9.

(28) investigated the economic burden of haze-related respiratory illnesses, including the costs of both inpatient and outpatient cases, from the provider’s perspective. Nor has any previous study estimated the actual treatment cost of haze-related respiratory illnesses in public healthcare facilities in Malaysia.. Research by Othman et al. (2014) described the economic burden of haze-related illnesses among patients admitted to Klang Valley hospital. However, the economic. ay a. value measured was productivity loss due to hospital admission. Actual treatment costs were not calculated. Another study conducted in Malaysia described the economic. M al. burden of haze-related illnesses only from the patient’s perspective but failed to evaluate the cost incurred from provider’s perspective (Mohd Shahwahid, 2016).. of. The present study addressed this gap in knowledge of the issue by investigating the disease and economic burden of haze-related respiratory illnesses from the provider’s. ty. perspective, taking account of the actual cost of treatment of both inpatients and. rs i. outpatients in public healthcare facilities. The findings are expected to inform policy. ve. making and resource allocation to manage future impacts of haze-related illnesses.. 1.6. Research Questions. ni. 1. What factors were associated with haze episodes in the study area?. U. 2. What was the utilisation of public healthcare facilities due to haze-related respiratory illnesses in the study area? 3. Is there any association between haze episodes and public healthcare facilities utilisation due to haze-related respiratory illnesses in the study area? 4. What were the healthcare costs of managing haze-related respiratory illnesses in public healthcare facilities from the provider’s perspective?. 10.

(29) 1.7. Study Objectives. General objective To determine the respiratory health impact of haze exposure and its financial implications. Specific objectives 1. To determine the factor/s associated with haze episodes.. respiratory illnesses.. ay a. 2. To determine the utilisation of public healthcare facilities due to haze-related. 3. To determine the association between haze episode and public healthcare facilities. M al. utilisation for haze-related respiratory illnesses.. 4. To calculate the direct medical cost of haze-related respiratory illnesses in public. of. healthcare facilities from the providers’ perspective.. 1.8. Summary and Organisation of The Thesis. rs i. ty. In broad terms, this study has three main components:. 1. Distribution, trends and factors associated with haze episodes.. ve. 2. Public healthcare facilities resource utilisation due to haze-related respiratory. ni. illnesses.. U. 3. Financial implications of haze-related respiratory illnesses.. This chapter has briefly introduced each component. It has provided basic. information about haze (definition, components, sources and chronology of haze episodes in Malaysia) and summarised what is known about the health impact and disease burden related to haze, including health-related complications, healthcare utilisation and financial burden of haze-related illnesses. The chapter also presented the. 11.

(30) study background, including problem statement, rationale and justification, research questions and objectives.. Chapter 2 presents a comprehensive review of literature relevant to the study. The first section examines trends in air pollution and factors associated with haze episodes in Malaysia. The second section focuses on the healthcare utilisation rate for hazerelated illnesses during haze episodes. The third section discusses the financial. ay a. implications of haze, both non-health-related (such as tourism, transport and the environment) and health-related.. M al. Chapter 3 presents a detailed description of the study methodology, including study design, study period, study area and study population. It also explains sampling, study. of. variables, and methods of data collection and statistical analysis.. ty. Chapter 4 presents and discusses findings for the first study component (trends and factors associated with haze episodes). These include daily trends in air pollutants and. rs i. ecological factors throughout the study period as well as factors associated with haze. ve. episodes in the study area.. ni. Chapter 5 presents results for the second and third study components (healthcare. U. utilisation and financial implications of haze-related respiratory illnesses in Selangor). First, results for healthcare utilisation and its association with haze episodes are presented. The financial implications are examined through comparisons of costs during haze and non-haze episodes. Results for both components are discussed and compared with findings from other studies.. Chapter 6 synthesises the overall findings, discusses their significance, makes recommendations and summarises the thesis. 12.

(31) CHAPTER 2: LITERATURE REVIEW 2.1. Introduction. Haze is not a rare event in Malaysia. The first haze episode was recorded in April 1981 (How & Ling, 2016; Norela, Saidah, & Mahmud, 2013). Since then, Malaysia has been affected by multiple haze episodes that caused disruption to normal daily routines.. a. The latest such incident occurred in 2015.. ay. Haze episodes in Malaysia are commonly associated with transboundary sources, namely, forest and peat fires in Indonesia. In 1997, the routine practice of agricultural. al. land clearing by burning became uncontrolled. Approximately eight million hectares of. M. land in Indonesia were burned, generating severe smoke and haze that spread to. of. neighbouring countries such as Malaysia, Singapore, Brunei, Thailand, Vietnam and the Philippines (How & Ling, 2016). It remains the worst haze episode in Malaysian. ity. history. It created national tension and became a controversial regional issue for. rs. member countries of the Association of Southeast Asian Nations (ASEAN) (Diyana,. ve. Hanan, & Othman, 2015).. Local sources from motor vehicle emissions, industrial production and oil- and gas-. ni. related activities also contribute significantly to deterioration of air quality, especially in. U. a rapidly developing state like Selangor (Abdullah et al., 2012; Rahman et al., 2015). Compared to the seasonal impact of transboundary sources, local sources are more important determinants of background air pollution levels (Abdullah et al., 2012; How & Ling, 2016).. Haze has social, financial and health impacts. Social impacts are commonly associated with transboundary haze, which usually produces short-term high-level air 13.

(32) pollution. The amount of pollutants can reach hazardous levels, thus restricting everyday activities, such as travelling to work, school or university, and attending outdoor social events (Mohd Shahwahid, 2016). The financial implications can be further categorised as health- and non-health related. Like social impacts, non-health related financial impacts are usually linked to transboundary haze. These include economic losses from tourism, transportation, and environmental and ecosystem. a. damage (Adriani, Moyer, Kendrick, Henry, & Wood, 2016; Schweithelm, Glover, &. ay. Jessup, 2006).. al. In contrast, health-related impacts are associated with both transboundary and local. M. air pollution sources. The effects of haze on health take localised form (such as skin irritants, allergic conjunctivitis and scleritis) (Wan Mahiyuddin et al., 2012),. of. cardiovascular illnesses (heart failure and ischaemic heart disease) (Shah et al., 2012) and respiratory illnesses (acute exacerbation of asthma and chronic obstructive. ity. pulmonary disease) (Anderson et al., 2012). Health complications due to haze-related. rs. illnesses are significantly associated with increased healthcare utilisation. Haze episodes. ve. related to both local and transboundary air pollution sources have been reported to result in a significant increase in utilisation of public healthcare facilities (Brauer & Jamal,. ni. 1998; Othman et al., 2014). This increase creates a considerable economic burden on. U. the Ministry of Health (MOH), which is the main provider of public healthcare facilities in Malaysia (Othman & Mohd Shahwahid, 1999; Othman et al., 2014).. This chapter presents a comprehensive review of published literature relevant to this study. The first section examines trends in air pollution and the factors associated with haze episodes in Malaysia. The second section discusses findings on healthcare utilisation for haze-related illnesses during haze episodes from both local and 14.

(33) international research. The third section explores the financial burden resulting from both health- and non-health related effects of haze.. 2.2. CAQM station. Monitoring of air quality and collection of ecological data are carried out by the DOE through CAQM stations. Currently, 51 CAQM stations measure the air quality. a. level in Malaysia. These stations are divided into five categories: industrial (26%), 57%. ay. residential (57%), traffic (2%), background (2%) and PM10 (13%) (Department of Environment, 2010). Alam Sekitar Malaysia Sdn. Bhd. (ASMA) has been given the. al. responsibility of maintaining these stations by the DOE through a privatisation. M. concession. The operation protocols and monitoring instruments used in these stations. of. are based on the guidelines provided by the United States Environmental Protection Agency (USEPA). The instruments and procedures used to monitor air quality level at. ity. each CAQM station accord with protocols and guidelines provided by USEPA. These. rs. include calibration of the instruments, methods used to collect samples, interpretation of the results and quality control assessment to ensure the correct techniques are applied. ve. throughout the process of sample collection. Collection and handling of data,. ni. interpretation and analysis are performed by ASMA according to the quality control. U. protocols. The maintenance procedures in these stations are conducted on a regular basis, following the standard protocols, to ensure the validity and quality of the data (Othman et al., 2014). The locations of CAQM stations were selected based on six main criteria (Department of Environment, 2010), as elaborated below.. 15.

(34) 1) The results of past and current monitoring. The selection of locations for CAQM stations depends on the pattern and trend of air quality levels in the particular area, especially areas with multiple sources of air pollution. Several CAQM stations are concentrated in Klang Valley due to its hourly variation in pollutant levels, which can have a significant influence on overall air quality measurement for that location. Except for the rural CAQM station in Jerantut,. ay. a. which is close to a national forest reserve, all other CAQM stations are located in areas that are easily affected by air pollution based on the results of past and current. M. 2) Representativeness of the selected area. al. monitoring.. of. Air quality monitoring from CAQM stations acts as a proxy to determine exposure. ity. levels of the population to air pollution. The location must be strategic enough to represent air quality level in that particular area. Areas with high population density,. rs. especially in urban settings or state capitals, are preferred due to their capacity to. ve. represent exposure levels of the population to air pollution.. ni. 3) Accessibility. U. CAQM stations must have good accessibility and connections. These include land. access, distance, and ability of the surrounding area to facilitate the processes of data collection, maintenance and monitoring.. 4) Availability of support services. Since the instruments used to monitor air quality levels require a power supply and communication tools, the availability of support services such as electricity and 16.

(35) telephone lines in the selected location are essential to ensure the continuity of monitoring processes.. 5) Security level. Since all the instruments used to measure air quality level are expensive, a high level of security in the surrounding area is also vital in the selection of CAQM locations. This. a. is to ensure that monitoring instruments in the CAQM station are protected from. ay. misconduct or vandalism.. al. 6) Topographical effects. M. Flat areas of land are preferred for the establishment of a CAQM station. The site. of. should be free from obstacles to ensure collected samples are similar to the atmosphere. Almost all CAQM stations are located on flat land to ensure that no specific feature of. ity. the topography can affect the results of data recording.. rs. The majority of stations are located inside school/college compounds, which. ve. generally fulfil the above-mentioned criteria for a CAQM station site, namely, accessibility, security, availability of support services and absence of specific. ni. topographical effects. These schools are located in high population density areas with. U. multiple sources of air pollution to ensure their representativeness in relation to the surrounding population.. Each CAQM station is equipped with: . Measurement instrument (to measure levels of pollutants and ecological parameters);. . Support instrument (to support gaseous and for calibration of equipment); 17.

(36) . Shelter (for protection and security of instruments); and. . Data acquisition system (for data collection and storage) (Department of Environment, 2010).. Different measurement instruments are used according to the type of parameters that are being monitored. Table 2.1 summarises the instruments used to measure those. U. ni. ve. rs. ity. of. M. al. ay. a. parameters at each CAQM station.. 18.

(37) Table 2.1: Instruments used to measure parameters at CAQM station. Parameter. PM10. a. SO2. Instrument Beta Attenuation Mass Monitor (BAM-1020), manufactured by Met One Instrument Incorporation, USA Equipped with cyclone, PM10 head particle traps, fibre glass tape, flow control and a data logger Fairly high resolution of 0.1 μgm-3 at a 16.7 L min-1 flow rate, with lower detection limits of b4.8 μgm-3and b1.0 μgm-3for 1h and 24h Teledyne API Model 100A/100E, manufactured by Teledyne Technologies Instruments, USA Based on ultraviolet fluorescent method, with 0.5% precision and lowest level detection at 0.4 ppb Teledyne API Model 200A/200E, manufactured by Teledyne Technologies Instruments, USA Based on chemiluminescene detection method, with 0.5% precision and lowest level detection at 0.4 ppb Teledyne API Model 300/300E, manufactured by Teledyne Technologies Instruments, USA Based on non-dispersive, infrared absorption method with 0.5% precision and lowest detection of 0.04 ppm Teledyne API Model 400/400E, manufactured by Teledyne Technologies Instruments, USA Ultraviolet absorption method, with 0.5% precision and lowest detection limit of 0.4 ppb Met One 010C sensor, manufactured by Met One Instrument Incorporation, USA Met One 062 sensor, manufactured by Met One Instrument Incorporation, USA Met One 083d sensor, manufactured by Met One Instrument Incorporation, USA. al. ay. NO2. M. CO. Wind Speed. ve. Humidity. rs. Temperature. ity. of. O3. U. ni. Source: Talib, Dominick, Ahamad, Khan, & Juneng (2014). 19.

(38) 2.3. API and Specific Pollutants Assessment. Different agencies provide various methods and guidelines for determining air quality level. These include the Air Quality Health Index (AQHI) in Canada (‘Air Quality Health Index Categories and Health Messages’, 2015) and Hong Kong (‘Air Quality Health Index’, 2014), the Daily Air Quality Index (DAQI) in the United Kingdom (Holgate & Ayres, 2011), the Common Air Quality Index (CAQI) in Europe. ay. a. and the Pollutant Standard Index (PSI) in the United States (Abdullah et al., 2012). In Malaysia, the monitoring of air quality and pollution levels is based on the Air Pollution. al. Index (API). This was adapted from the Pollutant Standard Index (PSI) developed by. M. the United States Environmental Protection Agency (USEPA). The PSI is also used in. of. Singapore and Brunei (Ostermann et al., 2001).. In Malaysia, API is commonly used to report air pollution levels to the public. It also. ity. serves as the reference parameter for decisions related to school closures and the. rs. declaration of an emergency (Department of Environment, 2000). The measurement of API is based on five major elements, namely, PM10, NO2, SO2, CO and O3. The index. ve. values are obtained by calculating the average of 24-hour monitoring values for PM10. ni. and SO2, 8-hour for CO and hourly readings for both O3 and NO2. These readings are. U. converted to API values based on API sub-index functions specific to each pollutant. The highest index values among the five major pollutants are taken as the API values for that particular hour (Department of Environment, 2000). Figure 2.1 summarises the process of calculating the API value from CAQM stations.. The API values can be used as the parameter for determining the health effects associated with the level of air pollutants. An API value of more than 50 is considered to have a moderate effect on health, more than 100 is considered as unhealthy, between 20.

(39) 201-300 as very unhealthy, 301-500 as hazardous and above 500 as emergency (Abdullah et al., 2012).. Although API is a good indicator of air quality level, it is a country-specific assessment and comparison across countries is difficult (How & Ling, 2016). The ideal method of measuring air pollution impact is through assessment of a true value for each parameter that contributes to the deterioration of air quality level. This helps to identify. ay. a. major air pollutants and possible sources of air pollution. It is also possible to compare air pollution implications across studies and countries using a specific parameter (for. U. ni. ve. rs. ity. of. M. al. example PM10) rather than API.. 21.

(40) Data on the five air pollutants (PM10, NO2, SO2, CO and O3) are collected according to sufficient averaging time periods. M. al. ay. during the data collection process. a. Calibration, validation, quality control and quality assurance are conducted. rs. ity. of. Average concentration of pollutant for specified time period is calculated. U. ni. ve. Sub-index value is calculated using sub-index function. The final report on hourly API is presented. Figure 2.1: Process of calculating the API value from CAQM stations. 22.

(41) 2.4. Definition of Haze Based on Malaysian and WHO Air Quality Guidelines. There are various definitions of a haze episode. In Malaysia, the Recommended Malaysia Air Quality Guidelines (RMG) were introduced by DOE in 1989. Subsequently, the first air quality index system, the Malaysian Air Quality Index (MAQI), was developed in 1993. This was revised in 1996 with the adoption of API from the US PSI system (Department of Environment, 2000). The New Malaysia. ay. a. Ambient Air Quality Standard, introduced in 2015, added PM2.5 as a new pollutant to be measured and identified interim target 1 (IT-1) in 2015, interim target 2 (IT-2) in 2018. al. and standard level in 2020 (Department of Environment, 2013). Table 2.2 summarises. M. the New Malaysia Ambient Air Quality Standard.. of. In 2005, the WHO Working Group on Air Quality Guidelines met in Bonn to update the Air Quality Guidelines (AQG) on air quality assessment and the impact of exposure. ity. to air pollution on health. The guidelines were made publicly available in October 2006. rs. as part of the Global Update of WHO Guidelines. These guidelines were first introduced in 1987 and updated in 1997. Due to new evidence from various studies on. ve. the health impact of air pollution (especially from low- and middle-income countries),. ni. expert panels conducted a review that led to the development of new AQG applicable to. U. all countries in different continents (Krzyzanowski & Cohen, 2008; World Health Organisation, 2005). Table 2.3 summarises the international WHO AQG.. 23.

(42) Table 2.2: New Malaysia Ambient Air Quality Standard. Averaging Time. μg/m3. 1 Year. 50. 45. 40. 24 Hour. 150. 120. 100. 1 Year. 35. 25. a. 15. 24 Hour. 75. 50. 35. 350. 300. 250. 105. 90. 80. 320. 300. 280. 1 Hour. 75. 75. 70. 1 Hour. 200. 200. 180. 120. 120. 100. 1 Hour. 35. 35. 30. 8 hour. 10. 10. 10. ity. 24 Hour. rs. 8 Hour. ve. CO (mg/m3). (2020). μg/m3. 24 Hour. O3. (2018). Standard. μg/m3. 1 Hour. N02. (2015). M. SO2. IT-2. ay. PM2.5. IT-1. al. PM10. Ambient Air Quality Standard. of. Pollutants. U. ni. Source: Department of Environment (2013). 24.

(43) Table 2.3: Updated WHO Air Quality Guidelines values. Averaging time. AQG value (μg/m3). PM2.5. 1 year. 10. 24 Hour. 24. 1 Year. 20. 24 Hour. 50. O3. 8 Hour, daily maximum. 100. NO2. 1 Year. PM10. M al. 1 Hour. ay a. Pollutant. SO2. 40. 200. 24 Hour. 20. 10 min. 500. of. Source: Krzyzanowski & Cohen (2008). ty. A comparison of the guidelines from Malaysia and WHO indicates some. rs i. discrepancies in target level, averaging time and type of pollutants included. The target level for CO is not included under the new WHO AQG. The averaging time for SO 2 (10. ve. min and 24 hour) and NO2 (1 hour and 1 year) also differ slightly from the Malaysian. ni. guidelines (1 hour and 24 hour for both SO2, and NO2). In addition, the targets (IT-1, IT-. U. 2 and standard) set by the Malaysian guidelines are much more lenient than the level set by WHO AQG. For instance, 1 year PM10 target levels under the Malaysian guidelines are 50μg/m3 (IT-1) and 45μg/m3 (IT-2) and 40μg/m3 (standard). Even at the standard level, the target PM10 is higher than the level set by WHO AQG (20μg/m3).. In this study, the WHO AQG was used as the main reference for air quality assessment, for three reasons. First, WHO AQG is accepted at the international level. Many studies used this guideline as their main reference for air quality assessment, thus 25.

(44) facilitating comparison of results across studies. Second, WHO AQG was developed based on extensive literature reviews of previous studies in both developed and developing countries. These guidelines therefore take account of regional variability in air pollution exposure and related health complications. Hence they are highly reliable and applicable to any country, including Malaysia. Third, target levels set by WHO AQG are stricter than the Malaysian guidelines. Use of the former guidelines establishes. ay a. a clearer target and direction for air pollution reduction strategies to achieve international air quality standard.. Trend in Air Pollutions and Factors Associated with Haze Episode in. M al. 2.5. Malaysia. of. The severe air pollution episodes that have occurred in Malaysia were associated with biomass burning, which originated from Indonesia (Diyana et al., 2015; How &. ty. Ling, 2016; Juneng, Latif, & Tangang, 2011; Norela et al., 2013; Rahman et al., 2015).. rs i. Biomass burning from forests, peat soils and plant residues released high amounts of pollutants into the atmosphere (Juneng et al., 2011; Norela et al., 2013). Burning and. ve. agricultural land clearing usually took place during the southwest monsoon and dry. ni. season (June to September), thus exacerbating the situation and resulting in the. U. migration of pollutants from biomass burning in Indonesia towards neighbouring countries, especially Malaysia, Singapore and Brunei (Azmi & Latif, 2010; Juneng et al., 2011).. Haze episodes resulting from transboundary sources are more seasonal events than those due to local sources, which are more constant and influential in determining background air pollution levels in Malaysia (Abdullah et al., 2012; How & Ling, 2016; Afroz et al., 2003; Azmi & Latif, 2010; Dominick, Talib, Zain, & Zaharin, 2012; 26.

(45) Shaadan, Jemain, Latif, & Deni, 2015). The local anthropogenic sources from motor vehicle emissions, industrial production and oil- and gas-related activities contribute significantly to the deterioration of air quality level, especially in a rapidly developing state like Selangor (Abdullah et al., 2012; Azmi & Latif, 2010; Rahman et al., 2015).. The main air pollutant during haze episodes is particulate matter 10 (PM 10). Apart from the intensity of local anthropogenic and transboundary sources, PM10 levels can. ay a. also be affected by ecological factors such as temperature, humidity and wind speed (Abdullah et al., 2012; How & Ling, 2016; Rahman et al., 2015). The influence of. M al. ecological factors on PM10 levels results from complex interactions involving chemical transformation and reactions between these factors and the structure of the pollutants (Juneng et al., 2011). These factors are largely influenced by the geographical location. of. of Malaysia, which is near the equator. As a tropical country, Malaysia’s climate is hot and humid throughout the year, with two types of monsoon season – the north-east and. ty. the south-west. The north-east monsoon generates the wet season from November to. rs i. March with heavy rainfall, especially in the east coast region of Peninsular Malaysia,. ve. while the south-west monsoon is associated with the dry season from June to September. ni. (Rahman et al., 2015; Wong, Venneker, Uhlenbrook, Jamil, & Zhou, 2009). U. Variations in PM10 levels have been observed between the dry season (south-west. monsoon) and wet season (north-east monsoon) (Juneng et al., 2011; Rahman et al., 2015). A previous study showed that PM10 had a negative correlation with relative humidity but a positive correlation with ambient temperature (Rahman et al., 2015). These correlations explain why the PM10 levels were slightly higher during the dry season compared to the wet season. During the dry season, the humidity level is low and less rainfall is recorded. Attenuated by high ambient temperature, the particulate matter 27.

(46) remains suspended in the air, resulting in a high concentration of PM 10. In contrast, during the wet season the humidity level is high and there is a higher chance of rainfall. The water vapour washes away all the suspended particles from the atmosphere, thereby reducing the PM10 level (Azmi & Latif, 2010; Rahman et al., 2015).. As for wind speed, results from different studies are varied. Some studies reported that high wind speed had a positive correlation with PM10 levels, especially during the. ay a. south-west monsoon, which carried all the pollutants from Sumatra to Peninsular Malaysia (Azmi & Latif, 2010; Shaadan et al., 2015). Some other studies, however,. M al. concluded that wind speed was negatively correlated with PM10 levels and played a part in dispersing the particles away from the affected area (Azid et al., 2015; Rahman. of. et al., 2015).. Nevertheless, the reported variations in PM10 level associated with ecological factors. ty. are small and, without the influence of mobile, stationary or open burning sources, the. rs i. PM10 level would remain below the recommended standard. Although the direct effect of ecological variations, such as wind speed, humidity and temperature, on PM10 level. ve. are minimal, these factors have a significant impact on forest fire burning, which. ni. indirectly increases the PM10 level. This effect was observed during previous haze. U. episodes when the El Nino phenomenon was associated with more extensive spread of forest fires in Indonesia (Tangang, Latif, & Juneng, 2010).. 2.6. Healthcare Utilisation for Haze-related Illnesses. For cardiovascular illnesses, the two main reasons for hospital admission associated with haze exposure are ischaemic heart disease (IHD) and heart failure. Numerous studies showed that an increase in PM level was associated with increased hospital 28.

(47) admissions and cardiovascular mortality. For a 10µg/m3 increase in PM10, there were 0.8% and 0.7% increases in admissions for heart failure and IHD, respectively. Similarly, for a 10µg/m3 increase in PM2.5, the admission rate increased 0.44% for IHD, 1.28% for heart failure and 4.5% for acute ischaemic attack (Anderson et al., 2012). In another study, admissions due to heart failure were associated with increased levels of CO, NO2 and SO2, although not for O3. There was also a significant association between. ay a. PM level and hospital admission (2.12% increase per 10µg/m3 increase in PM2.5 and 1.63% increase per10µg/m3 increase in PM10) (Shah et al., 2012). Overall, however, the available literature indicates that healthcare utilisation for cardiovascular illnesses was. M al. mainly related to inpatient rather than outpatient services, as most of these conditions required patients to be admitted for further monitoring and assessment. In contrast, the. of. local health effects of haze, such as conjunctivitis and eczema, were mainly managed and treated as outpatient cases. These kinds of cases rarely required admission to. ty. hospital ( Gao et al., 2015; Othman et al., 2014; Wan Mahiyuddin et al., 2012).. rs i. Compared to both cardiovascular and local health impacts of haze, utilisation of. ve. healthcare facilities due to respiratory illnesses involved both inpatient and outpatient cases. The main reasons for hospital admissions were acute exacerbation of bronchial. ni. asthma (AEBA), acute exacerbation of chronic obstructive pulmonary disease. U. (AECOPD), acute bronchitis, pneumonia and bronchiolitis (in infants), with AEBA and AECOPD accounting for the majority of cases (Anderson et al., 2012; Laumbach & Kipen, 2014; Mehta et al., 2013; Peacock et al., 2011). During the haze episode in 1997, a 100% increase in casualty visits was recorded in Kuching and Kuala Lumpur, the majority of which were due to asthma and acute respiratory infection (Brauer & Jamal, 1998). In Singapore, a similar pattern was observed in that year, with an increase of 30% in outpatient cases due to haze-related respiratory illnesses (Emmanuel, 2000). 29.

(48) From a systematic review and meta-analysis on the effects of PM on human health, Adar et al. (2014) found that increased healthcare utilisation due to respiratory illnesses was significantly associated with increased PM levels. The random-effects summary estimated 1.0% (95% CI: 0.1-1.8) higher rates of hospital admissions associated with respiratory illnesses with every 10µg/m3 increase in PM10-2.5. The authors also found that respiratory illnesses (mainly asthma and COPD) had a stronger association with. ay a. healthcare utilisation compared to cardiovascular outcomes. Another systematic review and meta-analysis similarly reported that the random-effects summary of respiratory hospital admission was higher, 0.96% (95% CI: −0.63-2.58) compared to cardiovascular. M al. hospital admissions, 0.90% (95% CI: 0.26-1.53) per 10 mg/m3 increase in PM2.5 (Atkinson, Kang, Anderson, Mills, & Walton, 2014).. of. Another review on healthcare utilisation for respiratory illnesses (Anderson et al., 2012) reported 1.61% (95% CI: 0.56-2.66) and 1.47% (95% CI: 0.93-2.01) increases in. ty. COPD admissions for every 10µg/m3 increase in PM2.5 and PM10, respectively. The. rs i. effect was higher in paediatric respiratory admissions, which showed a 17% (95% CI: 6-. ve. 29) increase for every 6.5µg/m3 increase in PM10-2.5. Among adults, the results varied according to the size of PM. The effects of PM on respiratory admission were: 2.07%. ni. (95% CI: 1.2-2.95) increase for every 10µg/m3 increase in PM2.5, 0.33% (95% CI: -0.21-. U. 0.86) increase for every 10µg/m3 increase in PM10-2.5 and 3.37% (2.34-4.40) increase for every 14.8µg/m3 increase in PM10. The main reasons for respiratory admissions and emergency department (ED) visits were asthma, COPD and pneumonia, with children and the elderly comprising the population most vulnerable to the effects of exposure to PM (Anderson et al., 2012).. 30.