Philippines Environmental Health Assessment : Volume One

RA 566.6 .P6 M27 1996 v.1 Martins, Jo. M., 1936- Philippines, environmental health assessment Philippines Environmental Health Assessment Volume One LO (0 0 T- O) 0 0 _J Cf) ' *** RA 566.6 .P6 Human Resources Operating Division M27 Country Department I 1996 v.1 East Asia and Pacific ~egion  Philippines Environmental Health Assessment Volume One  Acknowledgments This report supports a briefer synthesis entitled "Philippines: Health and the Environment". This report was prepared with information from the Philippines' national government agencies and a multitude of other national and international sources. Toe World Bank is grateful to the Government of the Philippines for its assistance in the collection of data and interpretation of the analysis, and the active participation of the Inter-agency Committee on Environmental Health, chaired by the Secretary of Health, in the discussion of terms of reference, interpretation of findings and formulation of recommendations. The work received financial support from the Canadian International Development Agency. The report was prepared by a team led by Jo. M. Martins (Task Manager). Clyde Hertzman (Consultant) developed an epidemiological framework and worked with Elma Torres-(Consultant) and Ronald Subida (Consultant) on the epidemiological analysis and findings. Serge Villard (Consultant) carried out the assessment of environmental health services. Angela Cortez (Consultant) undertook the economic valuation of the burden of mortality. Maria Barroetavena, Sheila Losen:ada, and Brenda Santillan provided research assistance. Gbangi Kimboko and Denise West provided assistance with document processing. Stanley Scheyer· (ASTHR) and Richard Ackerman (ENVPE) were peer reviewers. Toe report was prepared under the initial direction of J. Shivakumar, and then Sven Burmester (EAlHR); the Lead Economist was William McCleary (EAIDR); and the Department Director was Callisto Madavo (EAlDR).  - 1- PHILIPPINES ENVIRONMENTAL HEALTH ASSESSMENT Table of Contents Page 1. Health and the Environment in the Philippines 1 Purpose I Scope 2 Methodological Approach 3 Report Content 5 2. Determinants of Health and Environmental Health Risks 7 A Framework for Determinants of Health 7 Regional Variations in Health Status in the Philippines 9 - The Interface Between the Socioeconomic Environment and Pollution \ 3. Environmental Exposure Pathways - and the Burden of Mortality and Morbidity 15 The Principal Environmental Exposure Pathways 15 Urban Air Pollution Pathway 15 Water Supply Pathway 21 Fertilizers, Pesticides, Food, & Rural Waterways Pathway 23 Urban Solid Waste Pathway 27 Rural Point Source Pathway 29 The Burden of Morbidity and Mortality -Attributable to Environmental Exposures 32 Dust-Related Disease 32 Diarrhea and Skin Conditions 36 Lead Poisoning 38 Pesticides and Morbidity 39 Overall Impact of Pollution on Morbidity and Mortality 40  -11- PHILIPPINES ENVIRONMENTAL HEALTH ASSESSMENT Table of Contents (cont'd) Page 4. Knowledge and Attitudes, Monitoring and Regulation 45 Knowledge and Attitudes 45 Dust-Related Disease 45 Diarrheal and Skin Conditions 46 Lead Poisoning 46 Pesticides and Morbidity 48 Agency Responsibilities along Exposure Pathways: Organization Issues 48 Diversity and Focus on Health 48 Dispersion and Dilution 49 Local and Regional Services 49 Agency Responsibilities: Monitoring and Laboratory Analytical Capacity 50 Air Quality 50 Drinking Water and Sanitation 50 Capability and Capacity 51 Skills, Training and Quality 51 Equipment and Buildings 52 Organization 52 Assessment and Monitoring 52 5. Findings and Agenda for Action 55 Summary of Major Findings 55 Health Outcomes 55 Knowledge and Attitudes 57 Agency Responsibilities 57  -111- PHILIPPINES ENVIRONMENTAL HEALTH ASSESSMENT Table of Contents (cont'd) Page Agenda for Action 59 Priority Areas 59 Proposals for Short-Term Action 60 Development of Medium-Tenn Interventions 60 A Medium-Tenn Agenda 61 Institutional Strengthening 64 References 67  1. HEALTH AND THE ENVIRONMENT IN THE PHILIPPINES 1.1 Purpose. The aim of this report is to provide the Department of Health (DOH) and Government of the Philippines (GOP) with an assessment of the nature of the impact of environmental pollution on human health and its dimensions. And to promote the preparation of an agenda for policy and service development, to diminish the impact of these risks on human health. This assessment is concerned with elements of sustained development, such as the impact of environmental degradation on people, and alleviation of poverty through improvement in the health status of the poor, the people most affected by environmental health risks. 1.2 Air, water and soil pollution and food contamination are serious health hazards in the Philippines. Increasing industrialization and urbanization have caused pollution and placed undue pressure on the existing infrastructure, while use of pesticides and fertilizers in agriculture, deforestation and mining have created additional health risks. Evidence of environmental health risks to the Filipino population continues to accumulate, from studies concerned with specific risks in given geographical areas and fields of economic activity. Child exposure to lead has had deleterious effects on their intellectual and physical development, and air pollution has caused high levels of chronic obstructive pulmonary disease among transport workers. The mishandling of pesticides has severely affected the health of fann workers. Dust from varied sources has caused respiratory problems to people in urban and rural areas. Water pollution and poor sanitation continue to be major causes of diarrheal disease outbreaks. 1.3 Although DOH has responded to individual outbreaks of cholera, diarrheal and other diseases and has dealt with the sequelae of air pollution and poisonings, the Government has lacked a comprehensive view of the health risks involved, and of its role in the prevention and amelioration process. This has been particularly difficult because the responsibility for the "Environment" is shared by many agencies and levels of government. DOH can play the role of "Health Advocate" in Government, by bringing to bear the strength and influence of the of the entire Government on other relevant agencies, only if it has a clear understanding of the nature of the problem, its dimensions and sources. The Government and DOH need an agenda for action that distinguishes between actions for which it is directly responsible, and those in which it must work through and with other Government agencies and the private sector. Prior to this assessment relevant studies had been undertaken in the Philippines. However, they had not been brought together and given a human focus, into a cohesive and policy-oriented agenda. I .4 At the outset, it 1s important to note that the Government's concern has been translated into a range of purposeful efforts to deal with environmental pollution, prior to its  -2- request for this assessment. The following examples of efforts being made are evidence of useful experience of relevance to future development. The establishment of the Inter-agency Committee on Environmental Health was an important step in bringing the different government agencies together for the common purpose of improving environmental health. This has been complemented by a greater focus on environmental health in the Department of Health, and the establishment of a health-risk assessment unit there. With the support of WHO and UNDP, the Government has prepared guidelines for the integration of health and environment issues in national plans for sustainable development. In Cebu, with the support of GTZ, an innovative process was established to manage industrial pollution. In addition, the Government has passed regulations to prevent pollution and contamination, such as the formulation of drinking water and food quality standards, and air quality guidelines. Further, the Government has taken steps for the introduction of lead-free gasoline, and the enforcement of regulations against "smoke-belchers". In a more direct way, it has diminished the potential for air pollution from power generation by an increased use of gas- turbines and geothermal sources, and better emission control. In Manila, the establishment of sanitary landfill sites has reduced pollution sources from solid waste disposal. A project concerned with the promotion of the health of urban "squatters" provides among other things for safe water and sanitation. Another project is improving rural water supply and sanitation in some 37 provinces. A number of well designed studies have also provided valuable information on the impact of environmental pollution on people's health. 1.5 This report emphasizes health problems caused by environmental pollution or exacerbated by it, and which are amenable to primary prevention, through actions that are feasible and reasonably well understood. . The report provides a coherent framework for understanding these issues; and identifies a manageable set of priority environmental health issues for DOH and other responsible agencies to consider. By its very nature, it provides a working agenda rather than a definite plan. 1.6 Scope. The ·concept of health problems related to environmental pollution used in this report is narrower than "all environmental health problems". The problems dealt with in this report are the diseases or conditions which are caused by human exposure to polluted air, water, soil, food, or (liquid or solid) waste from residential, commercial, transportation, or industnal sources. Two types of environmental health problems are left out of this definition. The first group omitted includes those problems which are caused by naturally occurring toxics. For example, health problems related to dust exposures from volcanic eruptions are important environmental health problems, but they are preventable by interventions of public or private agencies. Another example is Red Tide, which although related to water pollution, appears to be driven by factors which are not well known and we do not know how to control. The most effective way to manage the situation is through emergency measures: to recognize affected seafood as soon as possible and then control the catching, sale and distribution of it. These measures are of vital importance to public health, but are outside the ambit of this report, which focus is on primary prevention. The second group comprises those health problems which are primarily mediated by individual human behavior and not by environmental pollution. Problems related to personal hygiene are the main among these. In many situations, exposure to polluted water, food soil, and even air may be modified, by  -3- personal habits and living conditions in the home environment. For example, diarrheal diseases which are caused by polluted drinking water are relevant to this study and those that result from poor personal hygiene (for instance, from fecal-oral spread) are not. But in practice, the influence of personal habits is important in determining whether or not those who are exposed will develop disease. If personal habits are ignored, problems will be misunderstood and subsequent efforts may be misdirected. For the purposes of this assessment, all conditions have been included in which environmental pollution is a significant component of the pathway to human disease, despite the relative importance of personal habits. Nonetheless, this still leaves out certain conditions of concern in public health, such as measles, which does not have an environmental pollution component along its causal pathway, or malaria, which has a pollution component so tangential to the main causal pathway as to place it outside the boundaries of this assessment. 1.7 Methodological Approach. To succ_ essfully undertake this study, it was necessary to establish a free and open flow of information with a wide variety of agencies, institutions, and individuals which had knowledge of significant exposures, agency responsibilities, populations exposed and health outcomes from the universe of potential environmental health hazards. The findings in this appraisal are based on the information made available by the range of national sources. The assessment was constrained by the availability of relevant information in the Philippines and the ability of those with such information to provide it in a timely fashion. A further limitation was the size and character of the world's knowledge base on environmental health, which is huge and often makes no distinctions between serious and trivial problems from the stand point of human health. In face of these limitations, there was a risk that the assessment, in trying to be comprehensive, would be merely superficial, and \ would be ineffective as a policy instrument. To avoid this pitfall, a problem-oriented approach based on _ the concept of "Principal Environmental Exposure Pathways" (PEEPs) was developed. 1.8 Environmental Exposure Pathways begin with an inventory of human health conditions which fit the definition of being "related to environmental pollution" (Table 1.1), and work backwards to their source, to identify the steps in the environment transmission pathway which give rise to human exposure. In other words, they extend the concept of a causal pathway backwards from outcome and exposure to the origins of the environmental pollution that affects humans. This approach reveals that health conditions related to environmental pollution fall into groups which have certain exposure pathways in common. These have been defined as the Principal Environmental Exposure Pathways. PEEPs are not planning models. They mix together various environmental media which are ordinarily managed separately (e.g. domestic garbage with industrial solid waste, drinking water with liquid industrial waste) because human exposures actually occur in these mixed forms. The characterization of PEEPs involved tracking down information as it applied to each PEEP and collating what was available. It included the identification of responsible agencies at each point in each pathway; information on the volume/type of industrial chemicals used, and emissions/effluents from point sources and diffuse sources; ambient concentrations of key toxics in workroom air, community air, drinking water, food, and soil; estimates of the size of exposed populations; studies of health impacts on people, including biological exposures,  -4- epidemiological studies, and notifiable disease and registry data; and demographic data such as housing types, water supply, income, and employment. Table 1.1: Environmental Health Outcomes and Exposure Pathways Health Problem Most Relevant PEEPs Diarrhea Commun Water, Pest/Rural/Food Bronchitis UrbAir, RurSource, UrbWaste Asthma UrbAir, RurSource Malignant Neoplasms: All pathways could contribute lung, liver, nasal, bladder, leukemia, lymphoma lschemic Heart Disease UrbAir, RurSource Hepatitis Commun Water, Pest/Rural/Food, UrbWaste Pneumoconiosis, occupational cancer, chronic chemical UrbAir, RurSource poisoning syndromes Typhoid/paratyphoid/salmonella & others Commun Water, Pest/Rural/Food, UrbWaste H-fever/dengue/Schistosomiasis Pest/Rural/Food Cholera CommunWater, Pest/Rural/Food Lead poisoning (acute & chronic) UrbAir, RurSource, Pest/Rural/Food Pesticide poisoning (acute & chronic) Pest/Rural/Food Other heavy metal and chemical poisoning UrbAir, RurSource Upper respiratory tract conditions UrbAir, RurSource Skin rashes/ear infections Comm Water, UrbAir, UrbWaste, RurSource 1.9 Five PEEPs have been identified for the purposes of this assessment: • Urban Air Pollution Pathway (UrbAir) • Urban Water Supply Pathway (CommunWater) • Pathway where fertilizers and pesticides affect food, workers health, and rural water supplies (Pest/Rural/Food) • Urban Solid Waste Pathway (UrbWaste) • Rural "Point Source" Pathway (RurSource)  -5- 1.10 The characterization of the PEEPs aims to pinpoint: • populations at risk, to estimate the burden of morbidity and mortality and identify target populations; • health outcomes they experience or could experience (based upon direct epidemiological evidence or epidemiological studies of similarly exposed populations elsewhere); • health outcome information adequate availability; • exposure sources adequate identification; • agency responsibilities significant gaps; • data flow needs (and accompanying equipment and skills); • improvement possibilities through inter-agency coordination; • priorities for action including remediation and regulation . . ~ -::O 1. 11 Report Content. In addition to this introduction, the report is in four other major parts. The first examines the determinants of health and the potential importance of environmental health hazards in the Philippines. The second reviews the exposure of populations along the five Principal Environmental Exposure Pathways, their health implications, and estimates of their attributable contribution to the burden of morbidity and mortality in the Philippines, and also its economic cost in terms of costs of lost earnings due to premature death. The third discusses the institutional framework in the Philippines to address these problems including agency responsibility and monitoring capacity, as well as knowledge, attitudes and practices related to Environmental Health. Finally, it puts forward an agenda for action arising from findings and discussions by the Inter-agency Committee on Environmental Health (IACEH). -6-  2. DETERMINANTS OF HEALTH AND ENVIRONMENTAL HEALTH RISKS 2.1 Concerns with pollution are widespread in the Philippines. According to the National Health Survey of 1992, about half of Filipinos perceive that there is a pollution problem in their vicinity: 95 percent in the urban areas and 33 percent in the rural (Department of Health 1992a). In addition to these perceptions, to assess the relative importance of the threat to health from environmental pollution, the impact needs to be examined in a broader framework. This should encompass the range of determinants of health, so that the contribution of environmental pollution can be appraised in its proper context. FRAMEWORK FOR DETERMINANTS OF HEALTH 2.2 The factors that determine the health status of populations can be divided into five broad categories: the socioeconomic/psychosocial environment; environmental pollution; individual habits and behaviors; health care services; and genetic endowment. The socioeconomic/psychosocial environment includes a wide range of factors related to income, education, occupation, social support, the lifelong impact of early childhood experiences, demand/control relationships at work and home, the influence of the political system, and the capacity of "civil society" to buffer stress. Environmental pollution includes all physical, \ chemical, and biological aspects, but does not include those aspects of the built environment (e.g. motor vehicle safety) which affect health. Individual habits and behaviors include smoking, drinking, personal hygiene, food choices, risk-taking activities, and all other health risk factors that are commonly thought to be based upon individual choice. Health care services involve all aspects of human agency specifically designed to protect health, and prevent and cure disease. Thus, activities as_diverse as vaccination, workplace health and safety inspection, and trauma surgery are all health care services. 2.3 The boundary lines between these factors are not clear and distinct. There are problems of overlap and hierarchy. For example, in the relationship between the socioeconomic environment and individual health habits, the ability to eat healthy food depends upon its availability and affordability. In a country where healthy food is readily available and affordable to the population, dietary patterns could be construed as a function of "individual choice" (despite the fact that there are systematic socioeconomic differences in diet even in the richest countries). In the Philippines, the availability and affordability factors will be most important, and individual health habits and behaviors are best construed as responses to socioeconomic conditions. In other words choosing healthy food is as much a socioeconomic factor as it is an individual health habit. 2.4 The quality, availability, and affordability of health care services will also be largely determined by socioeconomic conditions. In poor societies, effective services of various  -8- sorts are lacking; so health care services may be a more important limiting factor for health status in poor than in wealthier societies. In 1990, the Philippines spent approximately 2 percent of GDP on health care, this places it in the lower end of the range of 2 - 7 percent by countries in its income bracket (World Bank 1993). Moreover, the public share of total expenditure is a low 30 percent, whereas the mean for its income bracket is approximately 45 percent (World Bank 1993). Spending on environmental and public health services is affected by these patterns, and may be lower than necessary to protect the public from the effects of environmental pollution. As this report will show, investments in environmental health infrastructure could have a significant impact on the population's health status. The same could also be said about other health care services, but evaluating the level of provision of these services is outside the scope of this report. 2.5 Income is an important determinant of health (World Bank 1993). There is a positive correlation between GNP per capita and life expectancy. However, the relationship between life expectancy and national income needs qualification. In 1990, among countries with GNP per capita of more than US$ I 0,000, the correlation was weak, while it was stronger among the lower income countries (World Bank 1993). Nonetheless, there are large variations in health status among countries like the Philippines, and the health status of its people might be better compared with that of other lower-middle-income countries. 2.6 In general, the Philippines is "in the middle of the pack" so far as health status among developing countries is concerned. There is a 25 year range in life expectancy among countries in the -income bracket US$500 - US$1,380 per capita. Among other countries, Sri Lanka has a life expectancy 6 years greater than the Philippines, despite lower income. On the other hand, at 65 years, the Philippines is much ahead of its nearest income neighbors / (Senegal and Papua New Guinea) in life expectancy (Table 2.1). Nevertheless, the Philippines' gain in life expectancy in recent years has been modest. Between 1960 and expectancy of Filipinos rose by only 5 years (from 59 to 64 years), compared with 1990, life _ an average of 17 years in the demographically developing group and 12 years in the world as a whole (World Bank 1993). 2.7 In recent years, there has been increasing evidence that a well-educated and independent female population is a positive determinant of health in low income countries (Caldwell 1986). Although personal independence is hard to measure, the illiteracy rate offers a measure of education for international comparisons. The Philippines has achieved an admirably low illiteracy rate among adult females - among the lowest in its income bracket (Table 2.1). One of the reasons •given for the protective effect of female education on health status is that it translates into more effective care of children from the standpoint of nutrition, sanitation, and stimulation. Yet, powerful predictors of future health and well-being among the population, such as the proportion of low birth weight newborns and the rate of malnutrition among young children are unacceptably high in the Philippines.  -9- Table 2.1: Health Status in the Philippines Compared to Other Selected Countries Life Adult Female Low Birth Prevalence of Country GNP/capita Expectancy Illiteracy Weight Babies Malnutrition US$ (1991) (1990) (1985) <5 Yrs (1990) (Years) % % % Sri Lanka 500 71 17 28 45 Senegal 720 48 75 10 22 Philippines 730 65 11 18 19 PapuaN. G. 830 56 62 25 - Jamaica 1,380 73 1 8 8 Source: World Bank 1993. 2.8 The latter factor, in particular, complicates the interpretation of environmental health data. Although its validity may be challenged, it is in accordance with data from several sample surveys from around the Philippines. They show that in excess of 10 percent of Filipino school children are below the third percentile in height for age, and a similar proportion are below expected weight for age (Department of Science and Technology 1982- 1990). Malnutrition may be both a risk factor for environmental disease and a result of it. For instance, it is well known that malnourished children often have poorly functioning immune systems (Chandra 1988; Susl4R,9 _ 1977), . ~~- compromised intestinal function (Lifschitz 1990), which increase the dur~iion and severity of pollution-related infectious diseases (Lifschitz 1990). Even the effects of lead pollution on learning and behavior in young children can be exacerbated .by the adverse effects of economic deprivation and malnutrition on the developing brain (Bellinger 1989; Wasserman 1994). On the other hand, pollution-related diarrheal diseases can be a cause of malnutrition in children who are already near the borderline (Gracey 1985). REGIONAL VARIATIONS IN HEALTH STATUS IN THE PHILIPPINES: THE INTERFACE BETWEEN THE SOCIOECONOMIC ENVIRONMENT AND POLLUTION 2.9 There are large differences in infant mortality and life expectancy across the Philippines. In _1990, infant mortality by province ranged from 44 per thousand live births to 81. Similarly, the range of average life expectancy was 51 to 73 years. When provinces are aggregated into regions, the variation is not as large but is still considerable (Table 2.2). Infant mortality and life expectancy (corrected for under-reporting) markedly improve with better sanitary conditions. The correlation of decreasing infant mortality with increasing fractions of households with sanitary toilets is O.73 (and the proportion of variance explained is 53 percent). Similarly, the correlation of increasing life expectancy with increasing fractions of households with sanitary toilets is 0.83 (and the proportion of variance explained is 69 percent). A question that ·will be examined further in this assessment is whether these high correlations are due to socioeconomic differences, symbolized by the presence or absence of sanitary toilets, or directly related to the role of sanitary facilities in controlling disease transmission. On the other hapd, the provision of community water (as an index of urban services) is a much weaker correlate of each measure of health status. The correlation  - 10- with decreasing infant mortality is 0.37 (proportion of variance explained is 14 percent) and with life expectancy is 0.35 (proportion of variance explained is 13 percent). Table 2.2: Regional Differences in Health Status, Sanitation, and Water Supply Philippines, 1990 Infant Mortality * Life Households with Households with Region (Per thousand) Expectancy* Sanitary Commun. Water** (live births) (Years) Toilets** % % NCR 45.8 67 96 85 Region 1 55.6 66 91 33 Region 2 61.6 63 89 26 Region 3 44.7 68 84 38 Region 11 55.7 64 83 41 Region 10 57.2 63 79 53 Region 4 53.1 66 78 49 Region 12 56.4 63 77 32 Region 7 55.1 67 71 37 CAR 63.0 62 66 66 Region 8 76.2 61 66 47 Region 9 63.5 62 65 32 Region 6 60.7 64 62 32 Region 5 63.6 64 61 41 ARMM. 74.0 53 21 22 * Source: Flteger 1994. * * Source: Department of Health 1992a. 2.10 These patterns point to a fundamental problem in interpreting routinely collected data for environmental health purposes in the Philippines. Those regions that are most subject to certain forms of environmental pollution, especially urban air pollution, are also the most "developed," in the traditional sense. Therefore, the health effects of environmental pollution will tend to be masked by the greater prosperity and access to sanitary services in the more developed regions of the country. It will be difficult to use comparisons of environmental disease rates between polluted and unpolluted regions of the country, as evidence of the burden of morbidity and mortality due to pollution. Such comparisons will tend to underestimate the effects of pollution, since the impact of pollution will be partially offset by the health protective effects of prosperity and development. 2.11 There is a further problem that concerns the relationship between the quality of registration of health data and the socioeconomic character of a region. A careful study of death registration in the Philippines (Flieger 1994) showed that about half of infant deaths are not routinely registered in the country. Under-registration falls rapidly with increasing age at death. More than 80 percent of deaths over the age of 14 are registered. In addition, under-  - 11 - registration is not equally distributed across the country. In the National Capital Region (NCR) under-registration is negligible, but in some other regions overall death registration is as low as 20-40 percent. The completeness of death registration is higher among higher socioeconomic status regions than lower ones· (Table 2.3). Accordingly, it is reasonable to assume that registration of notifiable disease, too, will decline as socioeconomic status declines. This effect is demonstrated by three of the five variables listed in Table 2.3. Two of these are worth direct comment because they clearly illustrate the methodological problem: the percentage of total income spent on food and the percentage of households without sanitary toilets. As the proportion of total income spent on food rises (and, presumably, the level of malnutrition rises) the completeness ofreporting of malnutrition will tend to fall. With respect to toilets, the implication is that, as the proportion of the population without sanitary toilets rises (and, presumably, waterborne diseases such as diarrhea rise) the completeness of reporting of diarrhea will tend to fall. This effect will tend to bias notifiable disease data against revealing the regional correlations between environmental variables and health outcomes. Table 2.3: Socioeconomic Status Variables by Region and their Correlation with Regional Variations in the Level of Under-reporting of Death Registration %of Variable ,. Correlation Variance t (r) Explained (p) % of families with incomes <15,000 pesos, 1991 * - 0.19 0.04 0.51 % of total income spent on food, 1991 * -0.74 0.55 0.002 % of households with strong roofing materials, 1990** 0.47 0.22 0.09 % of individuals over 10 who are literate, 1990** 0.79 0.63 0.001 % of households without sanitary toilets, 1990** - 0.65 0.42 0.01 % of households with level 3 water supply, 1990** 0.29 0.09 0.31 .. * Source: National Stat1st1cs Office 1994a ** Source: National Statistics Office 1990b. 2.12 The methodological problem described above is illustrated by the data in Table 2.4. It shows that regional differences in notifiable disease rates for environmentally sensitive diseases do not correlate strongly with the proportion of households with sanitary toilets or with access to community water supplies. This is in contrast to infant mortality and life expectancy outcomes in Table 2.2, which have been corrected for under-reporting. When not corrected data are used, the only correlations of interest that emerge are: sanitary toilets with typhoid/paratyphoid/ salmonella (r= -.56; r2= .31); access to community water supply with diarrhea (r= -.25; r2= .06); and access to community water supply with hepatitis (r= .36; r2= .13).  - 12 - Table 2.4: Sanitary Conditions, Access to Community Water, and Selected Notifiable Diseases. Philippines, by Region - 1990. Case Rates per 100,000 People Households Households Typhoid, Region w Sanitary wCommun Paratyph, Diarrhea H- Infectious Cholera Bronchitis Toilets Water Salmon el. fever Hepatitis % % NCR 96 85 17.3 705.1 15.9 22.0 0.7 634.7 Region 1 91 33 33.1 979.5 7.6 22.7 0.7 815.2 Region 2 89 26 2.7 1508.0 5.4 17.9 1.2 2493.6 Region 3 84 38 4.4 3585.3 9.7 10.7 1.3 1083.9 Region 11 83 41 21.9 1383.8 45.9 22.6 1.2 2570.4 Region 10 79 53 36.1 1596.0 25.2 41.8 6.7 2573.1 Region4 78 49 10.8 635.1 2.2 12.4 0.5 478.8 Region 12 77 32 20.4 2049.1 12.5 8.1 6.3 1021.9 Region 7 71 37 22.2 1889.7 28.2 13.6 1.6 896.1 CAR 66 66 36.1 2252.9 2.2 40.4 . 2.3 2892.5 Region 8 66 47 22.7 1631.7 71.8 19.3 7.2 2256.5 Region 9 65 32 28.9 2089.6 6.0 21.9 1.0 2089.0 Region 6 62 32 40.7 1261.0 23.4 29.9 1.8 1524.9 Region 5 61 41 34.2 1507.7 2.9 26.7 0.9 2048.0 ARMM 21 22 - - - - - - Source: Department of Health 1992a, 1992c. The Contribution of Environmental Pollution 2.13 This report will provide estimates of the impact of environmental pollution on the health status of Filipinos, despite the obvious problems in the quality of notifiable disease data, and the complex relationship between socioeconomic and pollution-related determinants of health. An estimate of the days of healthy life lost (DHLL) due to a variety of conditions that have a significant environmental pollution component, based upon the experience in the Philippines is contained in Table 2.5. Comparison with Table 1.1 shows that not all of the environmental health conditions of interest have been evaluated, but the list includes the most common ones for which routinely collected data are available. 2.14 The total number of the estimated days of healthy life lost associated with environmental-related health conditions is 45 percent the total for all conditions from all causes (Table 2.5). This should not be interpreted that "45 percent of the morbidity in the Philippines is related to conditions associated with pollution of the environment." It is important to note, this is far different from a claim that 45 percent of Filipino morbidity is directly due to pollution. The critical factor is the proportion of the incident cases of each condition that might be attributable to pollution. This is examined in the following section of this report (Chapter 3), on the basis of the data presented to characterize the PEEPs.  - 13 - Table 2.5: Days of Healthy Life Lost in the Philippines, 1990 Environmental-Related and Other Conditions per 100,000 People Condition Days of Healthy Life Lost (DHLL) per 100,000 People Primarily Environmental: Diarrhea 368,012.6 Schistosomiasis 31 ,672.3 Typhoid Fever 29,260.4 H-Fever 9,342.7 Bronchitis 1,528,118.5 Hepatitis A 4,877.1 Total Primarily Environmental 1,971,283.6 Secondarily Environmental: Lung Cancer 37,387.7 Liver Cancer 1,066.1 Leukemia ,, ~·~ 3,194.5 Heart Diseases 555,935.5 Protein Calorie Malnutrition. 241,044.6 ., Total Secondarily Environmental 838,628.4 Total Environmental-Related 2,809,912.0 Total DHLL from All Causes 6,250,386.8 Source: Saro! et al 1994. - 14 -  - 15 - 3. ENVIRONMENTAL EXPOSURE PATHWAYS AND THE BURDEN OF MORTALITY AND MORBIDITY THE PRINCIPAL ENVIRONMENTAL EXPOSURE PATHWAYS Urban Air Pollution Pathway (UrbAir) 3 .1 The Urban Air Pollution Pathway covers those communities where household, motor vehicle and multiple industrial emission sources·mix together in an urban air shed. Emissions from road vehicles combine with industrial sources to pollute the air, which leads to three principal exposure scenarios for the population. The first affects transport workers, who are exposed to air pollution at the roadside level on a daily basis. The second affects children, who are exposed directly to airborne emissions through inhalation and also, indirectly, to soil and road dust ,which is contaminated by deposition from the air. Young children are known to passively ingest dust through hand-to-mouth spread, at a much greater rate than adults (La Goy 1987). In the urban environments of North America this may be as much as 100 milligrams per day (La Goy 1987). In urban environments such as Manila, where there is less ground cover, more open soil by the roadside, and a large proportion of people living close to the ground, the average daily ingestion of soil by children is likely to be much higher. The final exposed group is the general population, who are exposed to both urban air \, pollution and, in some instances, poorly ventilated indoor cooking fires. The UrbAir pathway is illustrated in Figure 3.1. Figure 3.1 Urban Air Pollution Pathway Worker exposure ,------, Cooking fires & Jeepneys - - - ~ Production cigarette smoking Air Exp sure to -------1- -Population Industrial Point sources _ _ _ _ _ _ __. - - - - - - - - . . . Transport workers Worker exposure Direct ingestion (mostly children) 3 .2 Air Quality Standards. Selected air quality guidelines for the Philippines and Canada (Table 3.1) will be used to assist in the interpretation of air quality information. The Canadian guidelines are considered to be "maximum acceptable" concentrations (Statistics Canada 1994). Despite efforts to assess the air pollution situation in developed countries and  - 16 - plot their progress, international standards for air quality do not generally exist (OECD 1991). The guidelines in Table 3.1 emphasize long term exposures, which are most useful, because their time frames tend to match the available monitoring data better than short-term excursion limits. The guidelines are not widely different. The Canadian sulfur dioxide limit is lower, as is the dust limit when it is multiplied by 0.55 to convert to PM10 equivalents (Dockery and Pope 1994). However, the Filipino limits for nitrogen dioxide, carbon monoxide, and ozone are lower than Canada's. Table 3.1: Air Quality Guidelines in Canada and the Philippines Canadian Guideline Filipino Pollutant Monitoring Period * Guideiine ** 1. Sulfur Dioxide Annual 13 ppb 28ppb 2a. Suspended Dust Annual 23 ppb 90 µglm 3 2b.PM10 Annual 70 µg!m3 60 µg/m3 3. Nitrogen Dioxide Annual/Daily -------/55 ppb ------/80 ppb 4. Ozone Annual/Hourly 106 ppb/15 ppb ------/70 ppb 5. Carbon Monoxide 8-hour 82ppb l0ppb .. * Source: Stat1st1cs Canada 1994. ** Source: DENR 1994. I 3.3 Worker Exposure. The Urban Air Pollution Pathway begins with worker exposures in industrial facilities. As expected, hazards in the workplace do not derive from airborne exposures alone (Appendix A contains census and survey data on hazardous occupations and occupational injuries), but industrial facilities,. in combination with motor vehicle emissions, are significant contributors to pollution of the urban air shed. Thus, industrial facilities belong at the beginning of UrbAir, and the exposures which occur within them are accounted for in this pathway. 3.4 The available data on working conditions in industrial facilities in the National Capital Region (Metro Manila) are of relevance to the review of this PEEP .. The data from Region IV are also included because they focus on some issues not studied in the urban environment (Department of Health 1991a, 1991b). In general, the proportion of work places experiencing a variety of different types of hazards is high. For instance, 97 percent of workplaces surveyed in the National Capital Region, and 80 percent in Region IV were found to have significant dust exposures; and respectively 76 percent and 46 percent were found to have excessive noise exposure. Occupational hygiene control measures were uncommon, especially those requiring investment in equipment. This impression was reinforced by  - 17 - evidence that the use of personal protective equipment was not as widespread as would be expected on the basis of the hazard frequencies (Department of Health 1991a, 1991b). 3.5 When all the available strategies for protection are added together, the impression is created that virtually all workplaces had some modality of protection in place (Department of Health 1991a, 1991b). However, actual protection at individual worker level was low. An important question is the effectiveness of specific strategies, and the degree of compliance with them. There is a scarcity of data available on these key issues. Yet, workers' health outcome studies carried out by the Occupational Health. and Safety Center show that in the workplaces studied overexposures to dust, lead, noise and organic vapors were common. Of particular interest are the estimates of the prevalence of pneumoconiosis in four workplaces with different types of dust exposure (8 - 18 percent) and the high proportion of those with the complication of silicotuberculosis (14 - 21 percent). A study of metal working establishments addresses the questions raised above. It showed that safety devices were uncommon, in particular adequate machine guards (Occupational Safety and Health Center 1993). 3.6 Airborne Emissions. The PEEP diagram for UrbAir (Figure 3.1) shows that the working conditions of those who make Jeepneys and other vehicles are relevant. In practice, no data is available to characterize this aspect of the PEEP. However, data does exist to characterize some of the principal emissioµs which lead to air pollution in urban areas of the Philippines. In 1994, the total number of vehicles in Metro Manila exceeded 959,000. In the four-years 1990-1994 consumption of gasoline increased by more than 30 percent, and diesel by more than 60 percent (Department of Transportation and Communication 1995). Information available provides a basis for the estimation of total motor vehicle emissions (Baker et al 1993) for Metro Manila in 1990: about 497,100 (metric) tones of carbon monoxide, 80,900 tones of hydrocarbons, 65,200 tones of oxides of nitrogen, 1,750 tones of sulfur dioxide, 650 tones of lead, and 11,100 tones of suspended particulates. These • emission levels increased by approximately 50 percent over the following four years. 3.7 The emissions from po.wer stations in Metro Manila (Baker et al 1993) are low compared to motor vehicle emissions. The emissions of carbon monoxide from the power plants are only 0.1 percent of motor vehicles; oxides of nitrogen are 13 percent of motor vehicles; and total suspended particulates are 40 percent of motor vehicles, after the difference in units has been taken into account. On the other hand, the power plants are the largest source of sulfur dioxide, generating 35 times more than motor vehicles.· 3.8 Ambient Air Quality. The average annual measurements of concentrations of total suspended particulate and sulfur dioxide, from the routine air-quality monitoring stations in Metro Manila (DENR 1988-1992), show that particulate (dust) is found at levels which could affect human health. The dust levels are in the same range as those in the most polluted regions of Central and Eastern Europe (Hertzman 1995). On the other hand, sulfur dioxide ievels are similar to those found in many major OECD cities, which means they are relatively less increased than dust levels. Airborne lead measurements in Metro Manila are much less complete than for dust or sulfur dioxide (DENR 1988-1992). The levels tend to be well ' ' '· ·i".f',: ,Y ~\ t,: ;(.• •: •• ~-~ "  - 18 - above cities in other countries where leaded gasoline has been eliminated, but lower than other cities with large numbers of leaded gasoline vehicles, such as Budapest. Nonetheless, the airborne lead levels in Manila are of public health significance. 3 .9 Levels of ozone, oxides of nitrogen, and carbon monoxide are not routinely monitored in Manila. Monthly air quality data from two monitoring stations (DENR 1994) reinforce the impression that dust (along with lead) is a principal concern as a long-term exposure factor. Levels of ozone, oxides of nitrogen, and carbon monoxide are all comparatively low by international standards, and below guideline values. Yet, the impression created is modified by a review of air monitoring data for shorter time periods (DENR 1988-1992, 1993), which shows that monthly sulfur dioxide levels sometimes reach over 100 parts per billion at specific monitoring stations. This is well above a guideline of 30 parts per billion, and similar to coal-using regions of Central and Eastern Europe. Short term excursions of carbon monoxide reach 30 to 100 times the long term average levels (DENR 1993), reaching levels where human health could be affected. The same is true for nitrogen dioxide (DENR 1993). 3.10 Exposures of Defined Populations. Health studies conducted on samples of targeted populations in Manila give a measure of air pollution exposures of individuals at risk of pollution-related health problems (Subida and Torres 1991, 1994). These data give six-hour personal exposure estimates for jeepney drivers, air-conditioned bus drivers, and commuters to dust, sulfur dioxide, carbon monoxide, and lead (as well as total oxidants). In every case, the levels of exposure are considerably higher than the long-term ambient airborne measurements from the monitoring stations mentioned earlier. For all three groups, especially for jeepney drivers, the dust levels are of public health significance. The same is true for sulfur dioxide and lead. Carbon monoxide levels, in contrast, are of significance only for jeepney drivers. Assuming that these data are reliable, the measurements taken in the breathing spaces of those who work or travel in heavy traffic, risks are much greater than exposure estimates based upon area monitoring. This has been noted quite consistently in other contexts (Donaldson 1980; International Agency for Research on Cancer 1984). Accordingly, the personal exposure values for dust are approximately twice the monitoring station values for school children and child street vendors (Subida and Torres 1994). Similarly, exposures to sulfur dioxide and nitrogen dioxide appear to be of concern for child street vendors based upon personal sampling. 3.11 Health Outcomes among the Urban Population as a Whole. Data is scarce with regard to the impact of air pollution upon the health of the general urban population. •There are approximately 6-7,000 visits to emergency departments per month for respiratory diseases in Manila (Torres and Subida 1993). However, these have not been studied in association with fluctuations in air pollution in a way which would help determine its contribution. Similarly, there are approximately 1,500 emergency visits for cardiovascular disease each month in Manila (Torres and Subida 1993). Undoubtedly, air pollution plays a more limited role in these events, but the barriers to interpretation are similar to those for respiratory disease. Information available (Subida and Torres 1991) is of great value because it indicates that the risk of "airways obstruction" in the urban population increases by approximately 31 percent for each 50 microgram per cubic meter increase in sulfur dioxide  - 19 - concentration in the air 1. Airways obstruction is important because it is part of the constellation of factors which make up chronic bronchitis, and it can have a direct effect upon well-being and the ability to perform physical activities. The effect of air pollution on airways obstruction is exacerbated by smoking behavior (Subida and Torres 1991), as would be expected from the public health literature on the subject (Morin 1992). 3.12 Health Outcomes among Transport Workers. Epidemiological studies have been specifically designed to assess the respiratory health of jeepney drivers and other transport workers in Manila. The quality of related information on health outcomes is comparatively high. Chronic obstructive pulmonary disease and its components (chronic cough, phlegm, wheeze, and shortness of breath) are more than twice as prevalent among jeepney drivers as among commuters or air conditioned bus drivers (Subida and Torres 1991). Pulmonary tuberculosis is also more prevalent among jeepney drivers: an outcome which may reflect either socioeconomic differences within the study sample or an effect of pollution. The symptom and disease reported are validated by evidence that the proportion of jeepney drivers with "pulmonary function" below. normal limits is higher than the two comparison groups. These differences persist after taking account of smoking in the comparison groups. Being a Manila jeepney driver increases the risk of abnormal pulmonary function by approximately 50 percent and the risk of chronic obstructive pulmonary disease by more than double (Subida and Torres 1991). These results have direct relevance to estimating the burden of disease associated with urban air pollution, and are consistent with the estimates produced later in this section of the report. 3.13 Other health outcomes associated with exposures to air pollution are less well studied among transport workers. Blood lead levels among jeepney drivers average 19.9 micrograms per deciliter, with 4.7 percent above 30 micrograms per deciliter. These levels are above unexposed populations, but much lower than populations which work directly with lead. The principal concern here would be whether or not lead levels in this range would have an impact on heart disease risk due to its effect upon blood pressure. Another risk factor for heart disease is the level of carboxyhemoglobin in the blood, which is associated with exposure to carbon monoxide. The levels of carboxyhemglobin in the blood of jeepney drivers show no difference between smokers and non-smokers (Subida and Torres 1991). This is because the air pollution effect has overwhelmed the smoking effect. Carboxyhemoglobin levels in smokers not exposed to Manila-style air pollution range from 0.9 to 10.4 percent saturation, with a median of 3.2 to 6.2 (Stewart 1974). These medians are only one-fifth of what is found injeepney drivers (Subida and Torres 1991). The researchers who did the Manila study tried to find out whether or not higher carboxyhemoglobin levels were associated with higher rates of heart disease and high blood pressure, but the study sample was too small to tell. 3.14 Health Outcomes among Urban Children. In light of the problems with air quality in the urban environment, it would be expected that children in Manila would have more 1 The tenn "pulmonary function" will be used throughout this report for any of a series of breathing tests which are designed to ascertain whether or not the bronchial passages are obstructing airflow to and from the lungs.  -20- health problems associated with air pollution than children in rural areas. However, this difference may be difficult to detect because of socioeconomic factors, and the presence in the rural environment of agents which can affect respiratory health. There are few studies available to assess urban-rural differences in respiratory health or other outcomes. Comparisons of pulmonary function between children in Manila and Baybay, Leyte, (Subida and Torres 1994) suggest that pulmonary function is poorer among Manila children. 3.15 Most of the available information involves comparisons between defined groups of children in the Manila area, mostly between child vendors and school children. According to this comparison,· child vendors should be at greatest risk, because of the roadside environments they work in and their low socioeconomic status, and the risks which attend it. However, the health outcome studies are not consistent on this point. Child vendors do seem to have higher prevalence of cough, phlegm, wheeze, and shortness of breath than school children (Subida and Torres 1994). But the data suggest that school children have poorer pulmonary function than child vendors, and also report a higher prevalence of respiratory disease. The problem here might be explained by the data which do not show clear trends to declining pulmonary function or respiratory disease with increasing level of exposure to particulates. The reason seems to be that virtually all the children are exposed to particulate levels in excess of 100 micrograms per cubic meters. Thus, there may be no unexposed children in the urban area with which to compare targeted populations. 3.16 Respiratory health among Manila children has also been studied in relation to parental smoking, crowding, and the use of cooking fuels. These data are inconsistent, too. Maternal and paternal smoking is not consistently associated with declining pulmonary function or the prevalence of respiratory diseases. Neither is crowding. These data (Subida and Torres ( " 1994) show that crowding appears to be a risk factor, a protective factor, or neither, depending upon which outcome is under consideration. Similarly, pulmonary function appears to be unaffected by the choice of home cooking fuel. This result is different from what is found in studies of rural children, as .will be shown in the Rural Point Source pathway. 3.17 The situation regarding lead exposures among Manila children is much clearer, because children's blood lead levels can be interpreted directly, and do not require an "unexposed" comparison group to set a benchmark. Data available by neighborhood in Manila show a clear tendency to higher blood lead levels in the inner city areas, where traffic flows are greatest (Subida and Torres 1994). With the only exception of Muntinlupa, all areas have mean blood lead levels that are at or above those of lead smelter towns in the most developed societies (Hertzman 1991). Among child vendors, the distribution of blood lead levels is higher still. One-third of them have blood lead levels above twenty micrograms per deciliter (Subida and Torres 1994). The range of lead exposures found in Manila school children and street vendors is of public health significance, because lead's toxic effects on brain development affect behavior and learning in ways which can interfere with success in school, and have potential for lifelong impact.  - 21 - Water Supply Pathway (CommunWater) 3.18 The health issues surrounding water supply begin with the quality of surface waters used by the population. There is a general problem of widespread pollution of surface waters with coliform bacteria. This is clearly shown by reference to recreational waters at beaches and rivers around Manila (DENR 1993 and 1994). The implications for the drinking of surface water are illustrated by the large number and considerable discharge volume of sources of chemical oxygen demand into Laguna Lake (DENR 1992a). In addition, the water quality issues facing Laguna de Bay, as a drinking water source, include the full range of biological and chemical issues: coliforms, nitrates, metals, industrial and agricultural waste, salinity, and eutrophication (Laguna Lake Development Authority 1995). Figure 3.2 Water Supply Pathway Toilet Toil.e t household household sanitation sanitation Industrial Agricultural effluents run-off Level3 Fonnal .;;;~• i.:.:;.:.:.''_·---!:==~·.·'·'· ·-__,..-=======-----11.! Infonnal water users Industrial Sewage users supply effluents Sewage Level 1 or 2 water 1 - - - - - - - - Fonnal users supply Toilet household sanitation 3.19 Urban dwellers in the Philippines may get their water from Level I, II, or III supply types2. All three types could potentially be polluted with sewage and/or industrial liquid waste. If the water source is Level III, it is likely to be chlorinated, adding a measure of protection not usually enjoyed by smaller sources. Nonetheless, levels of chlorine will decline as the water spreads to the far ends of the system, so that effective chlorination cannot necessarily be guaranteed. Moreover, chlorination of water which is polluted with industrial substances may produce increased concentrations of chlorinated hydrocarbons~ some of which are toxic to humans. In Manila, squatter communities are known to tap, informally, into water pipes, creating opportunities for further exposure to sewage and industrial waste 2 Level I water supplies include ope~ dug wells, developed springs, public wells, private wells with or without faucets, and tubed/piped/improved dug wells. Level II water systems are community public taps, and Level III are community supplies piped into residences or yards. Undeveloped springs and rain water are unclassified.  - 22- along the way. Regardless of water supply type, household sanitation and the presence of •toilet facilities are important mediators of the risk of waterborne disease. However, septic tanks are a problem because of lack of service. They tend to leak and overflow during heavy rains, becoming a source of contamination of the water supply. 3.20 Domestic Water Quality. Generally, the poor quality of potential sources of domestic water underscores the importance of supplying Level III water to as large a share of the population as possible. By 1990, about 42 percent of the Filipino population was being supplied with Level III water: 61 percent in urban and 23 percent in rural areas (National Statistics Office 1990a). The Metro Manila had the highest degree of Level III water supply: 82 percent. This would appear to confer an advantage on the residents of Manila compounded by other factors. For instance, households in cities with Level III water supplies boil their water more readily than rural dwellers. Moreover, the prevalence of boiling water declines between households with Level III water and all the rest. This is associated with levels of education and income, which are lower in rural than urban areas (Department of Health 1992a). 3.21 Research in the Philippines and elsewhere has shown that access to sanitary toilets is also an important protective factor against the spread of waterborne disease. The 1992 target for toilet facility coverage was 98 - 100 percent of Filipino households. By 1990 coverage was still under 70 percent (Health Safety and Environmental Management Consultancy, Inc. 1993). In the previous eight years the increase in the proportion of households with toilet facilities was less than 10 percent. Toilet bowls and chlorine jars are distributed throughout the country, but this program has not met need. Based upon an estimate of 3.8 million households in the Philippines without a sanitary toilet (National Statistics Office 1990a, 1994a), the 25,200 toilet bowls handed out in 1994 represent just 0.7 percent of present need. The 5,000 chlorine jars handed out in 1992 represent 0.2 percent of the households with less than Level III water supply. Once again, urban communities have benefited more than people in rural areas: 84 percent of urban households had sanitary toilets by 1990, compared with 52 percent of rural households. Metro Manila was in the forefront with 93 -96 percent coverage by the early 1990s (National Statistics Office 1990a; Department of Health 1992a). In both urban and rural areas, access to a sanitary toilet follows a pronounced socioeconomic gradient, with more affluent households more likely to have a toilet (Department of Health 1992a). 3.22 Health Outcome, Water Supply, and Access to Toilets. The report on the International Drinking Water and Sanitation Decade in the Philippines gives useful information about the magnitude of the risks associated with different types of water supply and access to toilet facilities (Health Safety and Environmental Management Consultancy, Inc 1993). In general, inside faucets or rainwater sources are associated with the lowest prevalence of disease, and public faucets are associated with the highest. The variation in the prevalence of diarrheal disease is approximately 20 percent, lower than for skin diseases, which exceeds I 00 percent from lowest to highest. Skin diseases and diarrhea were more common among households (and individuals) without toilets, which is consistent with the observations for water supply type. When these relationships are subjected to more careful  - 23 - analysis, taking account of confounding variables which may have threatened their validity3, they provide estimates of "magnitude of risk, 11 which can be used for estimating the burden of disease due to poor water supply or sanitary facilities. At the level of the individual, the data shows that using a public or neighbor's water supply increased the risk of diarrheal disease by 25 percent compared with using a household source. Similarly, living in a household without a toilet increases risk by 42 percent compared to households with a toilet. The use of a public or neighbor's water supply increased the risk of skin disease by 71 percent, and the use of rainwater or other sources increased risk by 54 percent, compared with a household source. Lack of a toilet increased risk by 28 percent. The risks change when the various factors are analyzed in combination. For both diarrhea and skin diseases, the worst combination is a household without either a toilet or a water supply. For both outcomes, the risk is increased by 76 percent compared with households that have both. After the effects of toilets and water supply are considered, a supplementary risk associated with household crowding emerges. When households are matched for water supply and toilet facility, a relatively crowded house will have more than double the prevalence of diarrhea, and 69 percent more skin disease, than the uncrowded house (Health Safety and Environmental Management Consultancy, Inc. 1993). 3.23 Acute Outbreaks. In addition to the day-to-day risks of water-and-sanitation-related disease, water can also be a source of acute epidemics, which are best evaluated through traditional outbreak investigations. The results of investigation reports from 1989 to 1994 have been collated by the Disease Surveillance Unit of the Department of Health. This source reveals that, indeed, drinking water is the most common source of reported outbreaks since 1989. Eight of the episodes involved cholera, one involved hepatitis A, one involved typhoid, and three were recorded as diarrhea (not otherwise specified). Most of these outbreaks were small, but one involved more than 1,000 cases and 12 deaths and another involved more than 10,000 cases and 80 deaths (Department of Health 1989-1994). It is unclear how many other acute outbreaks of waterborne disease go undetected or unreported in the Philippines. Fertilizer, Pesticides, Food, & Rural Waterways Pathway (Pest/Rural/Food) 3.24 This is the most complex environmental exposure pathway. It begins with the importation of pesticides and fertilizers, and, separately, with their formulation in the Philippines. On the farms .and plantations, exposures occur to farm workers and their families as a result of application. Crops are affected two ways: first by direct application, and second through contamination of water in the fields. Water may also be contaminated by sewage. Fish in fish-rice operations, and those in adjacent waterways, are exposed by the waterborne route too. The fish and crops end up at the marketplace, where the conditions of handling, cleaning, storage, and cooking will further mediate their hygienic quality. At the 3 The confounding variables include father's education, mother's education, crowding index, ownership of appliances, awareness of a water and sanitation program, house ownership/land tenure, type of housing (house material/construction and current condition of house), type of refuse collection and disposal, and . KAP (knowledge, attitude, and practices on health).  - 24- farm, the contamination of water by fertilizers, pesticides, and sewage may also affect the drinking water of those living in and around farms . . Figure 3.3 Fertilizers, Pesticides, Food and Rural Waterways Pathway Pesticide lmpor:tation - - . . . . i Formulation Pest/Fen Fertilizerlmpor:tation _ __ Farms& 1-T--- Runoffto Plantations wataerways Worker/family Worker exposure exposure Level 1&2 water supply Vendor hygiene Market Place Fish Sleds 3.25 Fertilizer and Pesticide Importation. In the case of pesticides, importation represents virtually all that is used in the Philippines. Information on imports of fertilizers and pesticides show that the level of importation for both fertilizers and pesticides has leveled off and even fallen in some cases, in recent years (Fertilizer and Pesticide Authority 1990- 1993; Department of Trade and Industry 1987-1993). As far as fertilizers are concerned, the most immediate public health concern is their contribution to the nitrate load in food and water. In the case of pesticides, each has a specific pattern of toxicity which may or may not be of concern (Tejada et al 1993; Cagauan 1990). 3.26 Fertilizers, Pesticides, and Farmers. From a series of sample studies done in the context of agricultural research, a picture emerges of the use, storage, and handling of pesticides among small rice farmers in two areas in the Philippines. Although it is not clear if the conditions found in these studies are applicable to the rest of the country or to larger farming operations, they are, nonetheless, useful for identifying important issues. The use of pesticides is very widespread and the concentration of fertilizer used by different farmers varies by more than five-fold (Bhuiyan and Castaneda 1994). These studies also show that unsafe practices in the use of pesticides are the rule (Warburton et al 1994). Few farmers wear full protective clothing when they spray and spraying against the wind is not uncommon. Most storage practices are unsafe, and disposal of empty pesticide bottles often involves selling them, or alternatively dumping into the paddy ecosystem, rather than safe disposal. These studies reveal that recognition and understanding of hazard rankings is almost non-existent among small farmers, despite the fact that most of them report relying upon government technicians or pesticide labels for information. The results of a "ranking  -25 - game" experiment showed that more than 80 percent of farmers and their spouses were unable to match hazard rankings with the color codes on pesticide labels (Appendix A contains a summary of hazardous pesticide spraying techniques and other practices by farmers). 3.27 Health Problems among Pesticide Users. The low level of knowledge and practice regarding pesticides use should be a contributing factor in the findings that individual exposures are of public health significance, and that pesticide-related health problems are common (Tejada 1994). In two pesticide using areas, nearly all of the farmers reported symptoms which are consistent with pesticide poisoning (Warburton et al 1994). However, controlled studies provide more solid evidence because pesticide-related symptoms are not specific and can be imitated by other conditions. Such a study (Rola and Pingali 1993) compared the prevalence of various health impairments for 57 exposed farmers and 39 controls and showed that pesticide users had higher prevalence of eye, skin, respiratory, heart, digestive, and blood-related conditions than controls. Pesticide users tended to have multiple health problems at one time. The modal number of conditions among the users is four, whereas for the controls it is one. Moreover, the risk of developing pesticide related symptoms increases with the frequency of use; doing two or three applications of pesticide per season increases the risk for conditions in each of five body-organ systems. The sample studies show that the majority of pesticide users have health problems associated with their use. Since more than 8.5 million Filipinos are employed in agriculture, this would suggest that there are actually millions of pesticide poisoning episodes per year in the Philippines. In 1993, only 100 pesticide poisonings were reported to the National Poison Control Information System: 79 in-patients, 21 out-patients, including 9 deaths (National Poison Control Center 1993). This indicates that most pesticide poisoning cases are not recorded. 3.28 Pesticides as Environmental Pollution. A major opportunity for pesticides to adversely affect human health occurs when farm workers come in contact with them. Another opportunity comes as a result of the pollution of the environment where they are used. The media which may be affected include not only groundwater and well water, but also soil and air (Tejada et al 1992). There has not been a comprehensive evaluation of the fate of pesticides in environmental media in the Philippines, but there are some surveys which give useful indications of the dimensions of the issue (Bhuiyan and Castaneda 1994; Tejada 1992; Tejada 1994). Such a survey showed that well water in irrigated areas not only yielded the pesticides that were in current use, but also pesticides which were not reported as being used. In the Nueva Ecija area, pesticides which were not reported as being used were detected in well water more frequently than those which were reported to be in use. The list of unused, yet still detectable, pesticides was longer than those in use and the concentrations found in groundwater were of the same order of magnitude as those in use (Bhuiyan and Castaneda 1994). 3.29 These observations raise a number of questions about the persistence of pesticides which are not currently in production or imported, and the accuracy of reporting of pesticides currently in use, and also about the knowledge of users of pesticides used. These concerns are reinforced by the data from the Laguna Lake ecosystem which shows measurable levels  - 26 - of DDT, DDE, and DDD in several locations (Tejada 1994). But the problem goes further. If pesticides are circulating for long periods of time in the environment regardless of their level of use, then pesticides may be found in places where they are not currently being applied. Indeed, there may be no relationship· between current pesticide practices and the levels of pesticide residue which are found in environmental media or crops. This problem is illustrated with reference to pesticide residues in water, soil, rice stalks, and rice leaves at two farms (Tejada et al 1992): one which is a high user of pesticides and another which is a low user. The data show that the concentration of chlorpyrifos in rice found in the low user farm is actually higher than the high user farm. It is thought that this may have been influenced by differential rainfall at the time of sampling of high and low user farms (Tejada 1995). 3.30 Pesticides and Food Crops. There is no data routinely collected or standardized on pesticide residues in crops in the field, storage, and in produce ready for sale, in ways that would allow a useful interpretation from a public health standpoint. Small sample surveys show evidence that residues are reaching the population in the market place (Rola et al 1992; Tejada 1994). This deserves more systematic investigation, regarding both the levels and types of pesticide in residues and the practices which determine the level of residue found. 3.31 Pollution and Food. As food products converge upon the marketplace, various forms of pollution converge there too. Fish are for sale which have been concentrating heavy metals from point and non-point sources (DENR 1992b; DENR 1992d); bacteria and fungi multiply under market conditions to reach unacceptable levels (Rodriguez et al 1994); and a wide range of pesticide residues persist (Tejada 1994). The public health effects of daily exposure to food contaminated with metals, microbes, and pesticides is uncertain and has not been systematically studied in the Philippines. Several acute outbreaks of foodbome disease have been reported in the past five years, all of which are related to microbiological contamination of food (Department of Health 1989-1994). 3.32 These reports, however, do not tell about small scale episodes, or the effects of low dose, long-term exposure to substances which accumulate in the body (e. g. lead) or have cumulative pathologic effects (e. g. organophosphate pesticides). For instance, at least one study suggests that blood lead levels in rural populations in the Philippines are comparable to urban •populations. Furthermore, the study seems to show that closeness to dusty roads, exposure to leaded house paint, having a parent in a high risk occupation, having a tendency to eat dirt (pica), and drinking breast milk are all not risk factors for high lead in rural areas (Fagela-Domingo et al 1994). These results pose important questions given what is known about the usual sources of lead exposure. They_ underscore the need for further investigation of the determinants of high lead in rural areas. 3.33 At the same time, little is known about individual practices which increase or decrease risk. For instance, the majority of Filipinos report that they purchase cooked food from kitchenettes, vendors, or at restaurants (Department of Health 1992a). This is important because cooking can either increase or decrease the risk of microbiological food poisoning, depending upon how it is done. Are there systematic differences in the microbiological quality of the food from these different locations? Are Filipinos with certain purchasing  - 27 - practices at greater risk than others? Individual practices may also be important in relation to pesticide residues. Cooking can reduce the residues of certain pesticides by 40 - 80 percent and washing can reduce them by 15 - 70 percent (Tejada 1994). What proportion of Filipinos follow these practices? Urban Solid Waste Pathway (UrbWaste) 3.34 For the purposes of this pathway, solid wastes mean wastes that should not be poured into a sewer or waterway. They include the residential, institutional, and industrial wastes that are either collected and disposed of in an organized system, or strewn about squatter communities, roadsides, or vacant lots. In the latter case, direct exposures occur primarily among squatters. In the former case, garbage sent to sanitary landfills is reasonably well- managed. However, garbage that ends up in dumpsites, where scavengers live and work, becomes a source of exposure through direct contact, combustion, and contamination of leachate. Figure 3.4 Urban Solid Waste Pathway ... , \· Residential Collection ~ Solid waste Institutional - ~ Worker exposure ::. 1 _Disposal site ·,· I v scavengers Exposureto& their families Industrial Non-col ection / ,, Leachate to water Squatter contact 3.35 Waste Generation and Disposal. There are different estimates of the volume of residential, institutional, and industrial waste generated in Metro Manila in 1992. They differ by 27 percent: a difference of 700 tons per day (DENR 1992d). Using the higher estimate, it can be further calculated that approximately 7 percent (250 tons per day) is burned, goes into waterways and sewers, or is left lying around. Close to 71 percent is handled by the municipal collection system, and is subject to landfilling and scavenging (Department of Public Works and Highways 1989). Most of this goes to Carmona and San Mateo landfills (Metro Manila Authority 1994), which are reasonably well-engineered and have no squatter communities. The balance is either collected and disposed ofby private haulers (14 percent) or recycled (8 percent) (Department of Public Works and Highways 1989). 3.36 In the past, most of the public health attention was focused on Smokey Mountain, because of the presence of a squatter community and scavengers. Although it has been closed, the health outcomes associated with the experience at Smokey Mountain are of continuing relevance because there are squatter communities and scavengers on .other  - 28 - dumpsites in the Philippines. There are four open dumpsites still operating in Metro Manila; and a large volume of waste is not being effectively collected in many locations across the country. Thus, the conditions experienced by individuals who are directly exposed to garbage from diverse sources continue to occur. 3.37 Dumping as a Source of Exposure. A review of the methods of waste disposal indicates that situation is worse outside Metro Manila than in it (Department of Health 1992a). Individual burning is the method of choice outside Manila by 45 percent of the population. This method may carry with it risks of exposure to airborne pollutants similar to open home cooking fires (Torres at al 1991). Individual open dumping and dumping into lowland areas are also popular methods for garbage disposal, making up 18 percent of the total. These carry risks of direct contact and increases in water pollution. Flies are commonly found in association with the four most popular forms of disposal (Department of Health 1992a). Whether or not these flies become a vector for disease has not been investigated. 3.38 The Smokey Mountain site illustrates the potential impact of open dumpsites upon the quality of the water which passes through them (Torres et al 1991). Available data shows that Smokey Mountain was an important source of chemical oxygen demand, chloride ions, and lead. An even more useful comparison of water quality upstream and downstream from the Payatas dumpsite in Manila (Bacud et al 1994) indicates the dumpsite was a source of acidification, chloride ions, nitrates, and coliform bacteria (including E. coli). Nitrates and coliforms are of public health significance among populations who use the water for drinking or recreational purposes. Open dumpsites, if they smoke, can have a powerful impact on local air quality. Two studies were made of the particulate, carbon monoxide, and airborne lead levels at Smokey Mountain while the dumpsite was still open (Torres et al 1991). The particulate levels were widely different between the two studies, but both estimates were very high, even by Manila standards. The carbon monoxide levels ranged from high normal to dangerously high, as did the lead levels. 3.39 Smokey Mountain as a Model of Health in a Scavenger Community. The health status of children who scavenged on Smokey Mountain in 1991 was worse than that of child street vendors and school children described earlier in this report (Torres 1992; Subida 1994). The prevalence of bronchitis was 31 percent and the prevalence of asthma was 29 percent. Pulmonary function was below normal in more than half the children, and blood lead levels in the scavenger children were also higher than others in Manila (Torres et al 1991). The range of mean blood lead levels among school children by neighborhood in Manila was 7 - 19 micrograms per deciliter. At Smokey Mountain it was 28. About 73 percent of children in Smokey Mountain had blood lead levels of over 20 micrograms per deciliter, compared with 10 percent of school children and 33 percent of child vendors. 3.40 Other health problems encountered among scavenger children are difficult to compare directly with children in Manila, but they are important because they exemplify the health impact of exposures such as those found among scavengers. The prevalence of underweight at rates of 2 - 5 perent was low by Filipino norms, but rates of stunting of 10 - 18 percent  -29- were similar to other studies of children around the country. Similarly, the level of anemia was not high by Filipino experience (Torres et al 1991). In general, it would seem that the nutritional status of the child scavengers was not below their peers elsewhere. This may be evidence of the success of the economic endeavor of scavenging. At the same time, intestinal parasitism was very common among scavenger children: more than half were found to be infested with ascaris or trichuris (Torres et al 1991). This is not surprising, because the biological activity of a dumpsite is high. Ninety percent of drinking water samples from the dumpsite were positive for bacterial contamination. Rural Point Source Pathway (RurSource) 3.41 In rural areas of the Philippines there are many locations where mines, discrete industrial point sources, or small groups of industrial sources may be significant sources of contamination of air, recreational or drinking water, and soil. The industrial sites are a potential source of risk to the workers and the local populations, as are mercury and arsenic emissions from mines and runoff from mine tailings. Risks from industrial airborne exposures will be mediated by airborne road dust, which is ubiquitous in rural environments, and also by in-home cooking methods. A few investigations have been made of environmental health issues around point sources of pollution in rural areas of the Philippines. At present, there is no systematic evaluation of workplaces, which would be likely to present environmental health risks and help to identify what are the most important problems. This section will briefly review three issues which have been investigated. They serve as illustrations only, and cannot be construed as a thorough assessment of the rural point source pathway. Figure 3.5 Rural Point Source Pathway (ooking, road dust Air • -------------~; One or more Water Local mining/industrial population source in one area Soil Worker exposure 3.42 Power Stations in Rural Areas. Data is available on air quality adjacent to three power plants in rural areas, where the air quality is unlikely to be significantly affected by motor vehicle traffic or other point sources (DENR 1994). The ozone, oxides of nitrogen, and sulfur dioxide levels are, in each case, low enough to be of no immediate public health concern. Particulate levels, measw-ed as respirable dust, are comparable to urban environments in OECD cities. This suggests that the power plants are imposing a significant  - 30 - dust load on the surrounding areas, but it is also possible that road dust is playing an important role. At the Tiwi plant, low levels of hydrogen sulfide were measured as well (DENR 1994). 3.43 Gold Processing in Davao del Norte. Mercury exposures to workers and nearby residents has been an issue in the gold producing areas in Davao del Norte. Studies undertaken in this area showed that mercury levels in workplace air frequently exceeded the American NIOSH standards (Torres 1989). The maximum recorded values from the four plants were seven times or more above the NIOSH standards, whereas the median values tend to be at or below the standard. Among the four areas, Tagum had the worst air quality. Even the median values were four times the NIOSH standard. Physical and biochemical studies showed that workers had indeed been overexposed to mercury (Maramba 1989). One to three quarters of the workers from the four locations had evidence of mercury deposits in their mouths. Much smaller numbers had elevated blood or urine levels; a surprising finding, given the levels of mercury in workplace air. 3.44 As would be expected, the ambient air in the surrounding communities was not as high in mercury as the workroom air. However, exposures in the community affected pregnant women and young children. This is important because mercury is toxic and affects the development of the nervous system. The maximum recorded airborne mercury level in each of the communities exceeded the WHO permissible limit by two- to one hundred-fold (Torres 1989). Greater detail is available for Tagum, which showed enormous variability between monitoring stations, and over time. Mercury had also gotten into the fish stocks in the area and the average mercury levels exceeded FAO standards in most locations sampled (Torres 1989). Once in the marine food chain, mercury concentrates in aquatic plants, and is likely to persist for decades. 3.45 Exposures through air and fish stocks, and the possibility of exposures through water, soil, and dust constitute a major threat to health. Children in Tagum showed evidence of personal-social delay and gross motor delay compared with control children, though not language or fine motor delay (Gapas et al 1991). These effects were buffered by maternal education and augmented by large family size, as would be expected. The effects were also more pronounced in younger children, which begs the question of whether or not intellectual functions which usually emerge among older children have been affected too. 3.46 Industrial Estate at Isabel, Leyte. Studies showed that the mean air pollutant levels in the communities surrounding the industrial estate at Isabel, Leyte, had levels of dust and sulfur dioxide which exceed Philippine standards. Lead and nitrogen dioxide levels also exceeded standards in one or two communities. Sulfur dioxide, lead, and oxides of nitrogen require an industrial or commercial source, and, undoubtedly, are a product of the industries in the local area. However, dust can come from "natural" sources. A comparison between the communities surrounding the industrial area and a control provided useful insights into the relative importance of industrial sources in the levels of sulfur dioxide and dust in rural communities (Torres and Subida 1992). The comparison showed that sulfur dioxide levels were low in the control community, as one would expect, but the dust levels in some areas of the control community were as high as in the industrial area, and exceeded the Philippine  - 31 - standard by as much as four-fold. Similarly, there was overlap in the indoor dust levels between the exposed and control groups, but the range of values in the exposed communities was higher overall (Torres and Subida 1992). 3.47 Findings of the study comparing the respiratory health of children in the industrial and control areas had a wider importance. The two groups of children were comparable in terms of height for age, level of household crowding, and maternal education; although the fathers in the study area tended to have higher levels of education than in the control area. Even more important, the fathers, and to a lesser degree, the mothers in the control area smoked more than the study area and more frequently used wood as a cooking fuel. Children in the industrial area had much higher rates of cough, phlegm, and bronchitis than children in the control area, despite the fact that the smoking and cooking fuel characteristics would have biased the results toward higher symptom reports among the controls. A closer examination of the determinants of below normal pulmonary function provide additional insights (Torres and Subida 1992). As expected, father's smoking was a risk factor, but residence in the study area (and its implications for outdoor and indoor air quality) was a stronger factor. Type of cooking fuel turned out to be the most important factor of all. The risk of below normal pulmonary function was more than 16 times higher among children in households with wood or coal fires compared with the alternatives (Torres and Subida 1992). 3.48 This latter result is consistent with findings in· other developing countries, but is inconsistent with findings from Manila. It is not clear why this should be so. It might be noted that, in the study and control areas, over 90 perent of households used wood or charcoal. Thus, the houses using other fuels probably differed from them in many systematic ways. In the Manila study, by contrast, less than 10 percent of the households represented used wood or charcoal, too small a number to affect the analysis much. 3.49 Other Considerations. Initial expectations could have been that environmental health risks in the countryside would be concentrated in specific areas, where point sources were found. Gradually, a different picture is being created; one which emphasizes more broadly based exposures. There are high blood lead levels among rural children not living near point sources of lead (Fagela-Domingo et al 1984), high dust levels in communities with no industry (Torres and Subida 1992), and health risks associated with cooking methods which are very common in the rural areas (Torres and Subida 1992). These findings point to the need to understand background exposures better, and a future research agenda emerges concerned with: the importance of living near the roadside, as a risk factor for lead and dust exposure; the other determinants of high blood lead levels in rural children; the contribution of burning activities such as garbage burning, burning sugar cane fields, and "smoking" fruit trees in rural air pollution and pulmonary disease; the importance of the method of cooking as a risk factor for respiratory health in rural areas but not in the city.  - 32- THE BURDEN OF MORBIDITY AND MORTALITY ATTRIBUTABLE TO ENVIRONMENTAL EXPOSURES 3.50 How significant are environmental exposures as determinants of health status in the Philippines? Although necessary information is lacking in relation to many health conditions and exposures identified along the PEEPs, there are sufficient data to make initial estimates of the impact of some environmental exposures. The methods used are described in detail in an appendix (Appendix C). In each case, data presented as part of the descriptions Principal Environmental Exposure Pathway models were used in the estimation process. Dust-Related Disease 3.51 To assess the impact of respirable dust on mortality in a variety of exposed populations in the Philippines (Table 3.2), it is necessary to estimate the volumes of PMIO (the respirable fraction of dust), because dose-response models available are generally based on PMIO. However, most of the dust measurements in the Philippines come in the form of Total Suspended Particulate (TSP) not PMI0. The estimates made in this assessment used conversion factors available in the scientific literature, based upon extensive side-by-side comparisons of TSP and PMI0 concentrations from the same samples (Dockery and Pope 1994). The exposed population figures are based on estimates of people exposed as identified along the different PEEPs, whose levels of exposure have been measured. The calculation used "excess PMIO", and estimated mortality for all causes, respiratory causes, and cardiovascular diseases. "Excess PMl 0" was used rather than the absolute value, because it should be assumed that there is a background level of dust that is not controllable by human intervention. This was set at 30 parts per billion of TSP, and it is subtracted before the conversion to PMIO is made. Table 3.2: Dust Exposures in Defined Populations Population Group (Thousands) Mean TSP Excess PM10* Metro Manila 7,907.4 177 81 Jeepney Drivers (Mnl) 159.6 570 297 Bus Drivers 1.0 320 160 School Children 2,350.0 314 156 Street Vendors 30.0 345 173 Scavengers (Sm. Mtn) 21.0 4,600 2,514 Other Urban 21,524.3 **100 39 Rural Point 15,559.5 70-2,500 22-1,359 Rural Non-point 15,559.5 30-100 0-39 Sources: National Statistics Office 1990a; Sub1da and Torres 1991; Torres, Sub1da and Rabuco 1991; Torres and Subida 1992; DENR 1994; Subida and Torres 1994; Department of Transportation and Communications, Land Transportation Office, 1995; Dockery and Pope 1994. * Excess PM10 = (TSP- 30) X 0.55; where "30" is taken as the background TSP level. **Estimated value based upon averaging Manila and background levels.  - 33 - 3.52 The estimated mortality related to dust is based upon dose-response factors for PMl0 and actual mortality. The dose-response factors are "best estimates," derived by combining dose-response information from several large-scale epidemiological surveys of dust and mortality (Dockery and Pope 1994). In general, these studies focus on daily variations in mortality in relation to daily fluctuations in airborne dust levels. In contrast, our analysis applies the dose-response estimates to average dust concentrations measured over comparatively long periods of time in order to make inferences about the stable effect of air pollution over the long term4 . 3.53 In the estimation process, the largest sources of uncertainty are in the rural areas and urban areas outside Manila. For instance, it is not clear what fraction of the rural population lives near a source of dust. Some sources, such as those described in the RurSource pathway, are well known. But data from sample surveys (presented in the RurSource pathway) suggest that dust levels in the rural areas without industrial point sources can be quite high too. Potential sources of non-industrial dust in the rural areas are road dust, burning of sugar cane fields, "smoking" fruit trees, and burning of wastes. It is estimated that up to half of the rural population may be exposed to industrial or non-industrial dust. The range of dust estimates given in Table 3.2 for "rural point sources" encompasses the full range found adjacent to the industrial estate in Leyte, which is the best characterized rural location. The "rural non-point sources" are assigned a much narrower range, which is meant to accommodate rural populations who do not live close to a ·.·dusty road, :and are not exposed to dust from agricultural burning activities. The estimate of dust levels in urban areas other than Manila is based upon the approximate midpoint between the average levels in Manila and background levels. · 3.54 The estimates of excess mortality from all causes, respiratory diseases, and cardiovascular diseases (Table 3.3) suggest that the impact of dust on mortality, especially ; respiratory disease, is significant in the population of Metro Manila and among other specially exposed populations. Manila jeepney drivers are the best characterized of these. Smokey Mountain scavengers are included despite the fact that the site is in the process of remediation, because they represent the subset of the population which may be exposed to similar circumstances at other dumpsites or in other places where incomplete combustion is a regular feature of the airshed. 3.55 Current reviews of the epidemiological literature (Dockery and Pope 1994) provide a wide array of outcomes estimated from varied sources of data, that can be used here: disease- based, service-based, and pulmonary-function-based. Once again, the estimated effect of air pollution is considerable for the heavily exposed populations for all outcomes. For instance, if the estimate of excess emergency department visits for respiratory disease in Manila is 4 This approach would be biased if the mortality being captured in studies used to make the dose-response estimates were merely a "harvest" of those who were about to· die anyway. This would be a potential problem if the follow-up periods in the studies were short, and the air pollution conditions more extreme than usual. But they are not. Because the studies involved relatively long periods of follow-up, the mortality patterns they detect represent a steady-state relationship and not a harvest of extremely vulnerable individuals.  - 34- applied to the approximate total number of visits (7,000 per month, according to Torres and Subida 1993), between 567 and 1,925 visits per month could be attributed to airborne dust. 3.56 Two estimates of excess morbidity (or mortality) are given for each condition in Tables 3.3 to 3.5. The lower value is based upon the "best" estimate of dose-response from the main reference (Dockery and Pope 1994). This is actually a very conservative estimate in the Filipino context. As a validity check, the ratio of rates of cough and lower respiratory disease, jeepney drivers to bus drivers, was compared between the data presented in the UrbAir pathway and the lower estimate in Table 3.5. The latter ratio is smaller than the former, indicating that the dose-response relationship underlying the best estimates from the international literature is conservative compared to the Filipino reality. The higher estimates help address this bias. They are generally based upon the strongest dose-response relationship between dust and disease among the individual studies listed in the main reference (Dockery and Pope 1994). Thus, the high-range estimates of burden of morbidity and mortality in the tables should be interpreted as "conservative to high". Table 3.3: Excess Mortality from Dust Exposures in Selected Populations Excess Mortality** Group Excess Total Respiratory Cardiovascular PM10* % % % Metro Manila 81 8.1 - 12.2 27.5 - 29.9 11.3 - 14.5 Jeepney Drivers 297 29.7 -44.6 101.0 - 109.9 41.6 - 53.5 Bus Drivers 160 16.0- 24.0 54.4- 59.2 22.4 - 28.8 School Children 156 15.6- 23.4 53.0 - 57.7 -- Street Vendors 173 17.3 - 26.0 58.8 - 64.0 -- Scavengers (Sm.Mtn) 2,514 251.4- 377.1 854.8 - 930.2 352.0 - 452.6 Other Urban 39 3.9 - 5.9 13.2 - 14.4 5.5 - 7.1 Rural Point 22-1,359 2.2- 203.9 7.5 - 502.9 3.1 - 244.7 Rural Non-point 0-39 0-5.9 0 - 14.5 0 - 7.1 Sources: As per Table 3.2. * Excess PM10 = (TSP - 30) X 0.55; where "30" is taken as the background TSP level. ** Based upon mortality factors for acute effects of particulates on daily mortality (Dockery and Pope 1994:107-32).  - 35 - Table 3.4: Excess Respiratory Morbidity* from Dust Exposures in Selected Populations Hospital Resp. Emerg. Dept. Group Excess PM10 * Admission Visit Asthma Attacks % % % Metro Manila 81 6.5 - 27.5 8.1 - 27.5 24.3 - 91.2 Jeepney Drivers 297 23.8 - 101.0 29.7 - 100.0 89.1 - 341.6 (Mnl) 160 12.8 - 54.4 16.0 - 54.4 48.0 - 184.0 Bus Drivers 156 12.5 - 53.0 15.6- 53.0 46.8 - 179.4 School Children 173 13.8 - 58.8 17.3-58.8 51.9 - 199.0 Street Vendors 2,514 201.1 - 854.8 251.4- 854.8 754 -2,891.0 Scavengers (SmMtn) 39 3.1 - 13.3 3.9 - 13.3 11.7 - 44.9 Other Urban 22-1,359 1.8 - 462.1 2.2 - 462.1 6.6 -1,563.0 Rural Point 0-39 0 - 13.3 0-13.3 0 - 44.9 Rural Non-point Sources: As per Table 3.2 for acute effects of particulates on daily mortality (Dockery and Pope 1994: 107- * Based upon mortality factors _ 32). - ** Pulmonary Function represented as% decline in FEV1.o- Table 3.5: Excess Respiratory Morbidity* from Dust Exposures (continued) Lower :, Upper Pulmonary Group Excess PM10* Respiratory- Respiratory · Cough Function** % % % % Metro Manila 81 24.3 - 123.1 5.7 - 56.2 9.7-69.5 1.2 - 2.8 Jeepney Drivers 297 89.1 - 451.4 20.8-205.0 , - 35.6- 255.1 4.5- 10.5 Bus Drivers 160 48.0- 243.2 11.2 - 110.4 ', 19.2 - 137.6 2.4- 5.6 School Children 156 46.8 - 237.1 10.9-107.4 18.7- 134.0 2.3 - 5.4 Street Vendors 173 51.9 - 263.0 12.1 - 119.3 20.8 - 149.1 2.6 - 6.1 Scavengers(SmMt) 2,514 754.2 - 3,821 176.0- 1,735 301.7 - 2,162 37.7 - 88.0 Other Urban 39 11.7 - 59.3 2.7 - 26.6 4.7 - 33.7 0.6 - 1.4 Rural al Point 22-1 ,359 6.6-2,065 1.5 - 937.4 2.6 - 1,169 0.3 -47.6 Rural Non-point 0-39 0- 59.3 0-26.6 0- 33.7 0-1.4 Sources: As per Table 3.2. * Based upon mortality factors for acute effects of particulates on daily mortality (Dockery and Pope 1994:107-32). ** Pulmonary Function represented as% decline in FEVt.o- 3.57 Beyond dust, only sulfur dioxide has clearly defined dose-response characteristics. However, sulfur dioxide is almost always found in association with dust, so its dose-response characteristics apply simultaneously to the same persons at risk, and thus to the same "excess morbidity/mortality." Consequently, it is biased to add the "dust effect" to the "sulfur dioxide effect,". because they do not add together in ·nature, but overlap (Dockery and Pope 1994). Moreover, dust exposures are relatively greater than sulfur dioxide exposures in most contexts in the Philippines. However, there is direct dose-response evidence available from studies in Manila on the impact of sulfur dioxide on pulmonary function (Subida and Torres 1991). In contrast to the . pulmonary function measure in Table 3.5, the Manila study outcome is an estimate of the proportion of individuals whose pulmonary function falls below an existing benchmark of normality. Combining the _ information available (Subida and Torres 1991, 1994) it is possible to estimate the increased risk of abnormal pulmonary  - 36 - function among certain _ subpopulations in Manila. The increase in the rate of abnormal pulmonary function attributed to the excess of sulfur dioxide are impressive and important. They vary in a range of 12 - 43 percent, as the excess in sulfur dioxide increases. However, there is no baseline rate with which to compare them, so the absolute magnitude of the increase in pulmonary function abnormality cannot be estimated directly. 3.58 Another special population, from the standpoint of respiratory health, is children in rural areas who live in households with wood or charcoal cooking fires. They would appear to make up as much as 90 percent of rural households: a population of approximately 13.7 million individuals under 18 years of age. By the outcome measure established in Table 3.4, the increase in the rate of abnormal pulmonary function among this group may be greater than 1,500 percent. If so, there will undoubtedly be other respiratory effects, such as cough and bronchitis. Because of the size of the affected population and the magnitude of the risk, this is a very important element of environmental risk in the Philippines which needs to be better characterized. Table 3.6: Sulfur Dioxide and the Risk of Abnormal Pulmonary Function in Manila Increase in Rate of Population Excess Sulfur Dioxide (ppb) Abnormal Pulmonary Function (%) Jeepney Drivers 75 43 Child Street Vendors 59 37 Bus Drivers 54 33 Commuters 40 25 School Children 19 12 Sources: Subida 1991; Dockery and Pope 1994. Diarrhea and Skin Conditions 3.59 The national census (National Statistics Office 1990b), the international water decade report (Health Safety and Environmental Management Consultancy, Inc 1993), the national health survey (Department of Health 1992a), and population and risk data are available and allow a simple and complete analysis of the "population attributable risk" for diarrhea and skin problems associated with less than Level III water supply and lack of toilets (Table 3.7). It is estimated that 19 percent of diarrhea cases nationally are "attributable" to the use of Level I or II water, or the lack of a toilet in the household. This translates into 42 million days of healthy life lost per year, based upon the rate of diarrhea in the notifiable disease data (Department of Health 1992e), or 1,168 million days, based upon disease prevalence data from the drinking water decade report (Health Safety and Environmental Management Consultancy, Inc. 1993).  - 37 - Table 3.7: Burden of Morbidity from Diarrhea Related to Water Supply and Sanitation Number Pop. People at Proportion of Odds Attrib. No. of Cases Preventable Risk Group Risk Total Pop* Ratio Risk** Attributable DHLL (Thous.) % (Thousands) (millions) Level II Water 19,027.7 0.314 1.25 6.3 59.l - 1,560.3 14 - 370 Level I Water 15,291.2 0.252 1.12 2.7 25.3 - 596.4 6 - 141 No San. Toilet 19,627.2 0.324 1.42 9.6 90.0 - 2,688.9 22 - 657 Total 18.6 174.3 - 4,845.6 42 -1,168 Sources: Sarol et al 1994; Health Safety and Environmental Management Consultancy, Inc, 1993; Department of Health 1990; Last 1993; World Bank estimates. * Based upon a total Philippine population of60,559,l 16 (1990). ** Population Attributable Risk= (Odds Ratio - 1) x Proportion of Total Population at Risk. 3.60 The rate of attributable risk percent should be interpreted as follows: "if all the other nutritional, microbiological, and socioeconomic influences are held constant, the fraction of Filipino diarrhea morbidity related to water supply type and lack of a toilet is approximately one-fifth." This is not the same as saying that the environmental component is only one-fifth of the morbidity. It has been suggested that water in Level III facilities is often contaminated with bacteria, which, if controlled, would likely affect the prevalence of diarrhea substantially. At present, there is no way to estimate the impact of this factor with any prec1s1on. 3.61 The fraction of skin conditions related to water supply and lack of toilet facilities is higher than for diarrhea, nearly 30 percent (Table 3.8). Because skin conditions are not part of the notifiable disease system, it is more difficult to estimate the number of attributable cases or their severity in terms of days of healthy life lost. Instead of using notifiable disease data, it is possible to estimate rates of "skin disease" (as the term implies in this context) from the water decade report (Health Safety and Environmental Management Consultancy, Inc. 1993). There are approximately 65 percent as many episodes of skin disease _as diarrhea, based on a large sample size (i.e. 4,655 diarrhea cases to 3,042 skin diseases). This ratio was applied to the number of notified diarrhea cases for the Philippines in 1991 (Department of Health 1992e), to get the low estimate of the "number of attributable cases" (Table 3 .8). The higher estimate of cases is calculated directly from the prevalence rate of skin co1,1ditions in the drinking water decade report. These estimates indicate that at least 176,000 cases of skin disease but more likely 4,157,000 are atpibutable to poor water supply and inadequate toilet sanitation.  - 38 - Table 3.8: Burden of Morbidity from Skin Conditions Attributable to Water Supply and Sanitation Number People Pop. Attrib. Number of Risk Group at Risk Proportion of Odds Ratio Risk** Attributable (Thous.) Total Pop* % Cases Level II Water 19,027.7 0.314 l.71 13.0 79,630 - 2,054,989 Level I Water 15,291.2 0.252 1.54 8.8 53,903 - 1,253,876 No San. Toilet 19,627.2 0.324 1.28 7.0 42,878 - 843,970 Total 28.8 176,411 - 4,152,835 Sources: As per Table 3.7. * Based upon a total Philippine population of 60,559,116 (1990). ** Population Attributable Risk= (Odds Ratio - l) x Proportion of Total Population at Risk. Lead Poisoning 3.62 The results of the estimation of the contribution of lead to decrements in I.Q. (Intelligence Quotient) among children in the Philippines are shown in Table 3.9. I.Q. is defined as a measure representing the whole dimension of cognitive development. Lead has also been associated with high blood pressure and through it cardiovascular disease, in large scale "ecological correlation" studies (Needleman 1991). But unlike the relationship between dust and mortality, evidence of cause and effect from high quality analytical studies is lacking to support this relationship. Therefore, the choice of l.Q. as the sole health outcome for lead is conservative but defensible. There are four principal questions with the estimates presented (Table 3.9). First, the size of the affected populations is not well defined. Second, there are clear contradictions among the available blood lead sample surveys. One survey of blood lead levels among Manila school children put the average level at 14.1 µg/dl, yet the survey of urban children quoted further down the column gives an average level of 22.9. Similarly, the rural children appear to have higher average levels than the urban children (though fewer of them are found at the high end of the-range). It is not clear to what extent these inconsistencies are due to sampling problems, variations in laboratory methods, or underlying differences in exposure which are inadequately understood. The third question relates to the uncertainty surrounding the dose-response estimates. A range of I.Q. declines per unit of blood lead has been found in studies from the "developed" world (see Appendix C). Evidence from Ea.stem European countries suggests that, in the context of socioeconomic deprivation, the dose-response relationship may be much steeper (Hertzman 1995). Therefore, a range of l.Q. decrements is given in Table 3.9. The lower value is based upon studies in the developed world (Baghurst 1992) and the higher value is based upon relatively deprived populations in Poland and Hungary (Hertzman 1995). The last question is: it is not clear what the long-term effects of childhood decrements in I.Q. really are. There is reason to believe that the neurobehavioral effects of lead which go along with I.Q. decrements (e.g. irritability and distractibility) increase the chances of school failure, and have long-term effects on well-being because of that. In the absence of other evidence, this should be taken as the principal form of "morbidity" associated with lead exposures in childhood.  - 39 - Table 3.9: I.Q. Decrements Attributable to Lead Exposures among Filipino Children Population Mean Blood Lead Meanl.Q. Highest Risk Risk Group (Thousands) (µg/dl) Decrement * Subgroup %>xµg/dl (x=20) Manila School Children (6-14) 1,400.0 14.1 2.23 -4.55 10.3 Street Vendors 30.0 17.8 3.13 - 6.40 32.7 Scavenger Child 15.0 28.4 5.73 - 11.7 72.6 (x=30) Urban Child (<14) 10,207.9 22.9 4.39- 8.95 7.8 Rural Child (<14) 12,377.1 24.1 4.68 - 9.55 1.2 Sources: Baghurst 1992; McMichael 1994; Sub1da and Torres 1994; Torres et al 1991; Fagela-Domingo et al 1984; World Bank estimates. * Mean I.Q. decrement expressed as the "best" estimate, followed by the maximum estimate. 3.63 The estimated range ofl.Q. decrements is approximately 2-12 points among different groups of Filipino children (Table 3.9). This range is of public health significance, because it is large enough to affect children's ability to learn. · ·,The effect is enough to change the average life course across large populations of children. Pesticides and Morbidity 3.64 The routinely collected data on pesticide poisonings, and the results of sample j surveys of farmers using pesticides, give different ideas of the level of pesticide-related morbidity. If only the routinely collected data are considered, it would seem that pesticide poisonings were rare and sporadic events. However, the sample survey data suggest that they are a feature of daily life among farmers who use pesticides. Given farm workers lack of personal protective equipment, their lack of knowledge about pesticide safety, the toxicity of the pesticides, and the acknowledged weaknesses of the poison control data, the sample survey data is much more likely to be close to the real situation. 3.65 Based on survey data (Rola and Pingali 1993), it can be estimated that, at any given time, agricultural workers who are using pesticides have 1.9 more health symptoms than farmers who are not. According to the Bureau of Agricultural Statistics (1993) there are 8,556,000 agricultural workers in rural' areas in the Philippines. It is not clear how many of them work with pesticides. On the assumption that at least half of them use pesticides, then as many as 4,278,000 people cotild be experiencing pesticide-related symptoms at any one time. This scenario begs three questions: how many agricultural workers are exposed? are they exposed in ways which make the sample survey data applicable to them? and what is the level of severity of these symptoms? Even if the proportion of exposed workers is less than half, it would appear that the burden of pesticide morbidity may be quite considerable in the  -40 - Philippines. Since the conditions of interest are not part of the notifiable disease system, they do not show up as measurable morbidity in the national statistics. Overall Impact of Pollution on Morbidity and Mortality 3 .66 Table 2.5 gave estimates of the days of healthy life lost due to a variety of conditions (for which routinely collected data are available) which have a significant environmental pollution component. To obtain a reasonable estimate of the proportion of Filipino morbidity associated with pollution, estimates of the proportion of risk attributable to pollution have been assigned to each condition, based upon the insights given by the principal environmental exposure pathways characterized earlier. The results are contained in Table 3.10. The table includes the same environmental-related conditions mentioned in Table 2.5 but has been rearranged to distinguish those conditions for which direct evidence is available regarding the impact of pollution, from those for which only informed estimates are available. 3 .67 Between 9 and 44 percent of morbidity from diarrhea, heart disease and bronchitis can be attributed to dust (as an index of air pollution), poor water supply and lack of toilet facilities in households. These are the conditions that can be evaluated from direct evidence. The population attributable risk for bronchitis, heart disease, and diarrhea are given in Table 3.10. The bronchitis estimates are based upon the range of estimates of excess morbidity for lower respiratory disease (Table 3.5). The heart disease estimates are based upon the range of excess mortality (Table 3.3). The lower estimate for diarrhea is based upon the best estimate of the odds ratio for each risk factor (Table 3.7), while the higher estimate is based upon the upper 95 percent confidence limit of the odds ratio (Health Safety and / Environmental Management Consultancy, Inc. 1993). 3.68 The conditions for which direct evidence is not available, and those which do not appear on the table, represent priorities for further investigation. Among the most important areas for research is the relationship between malnutrition and diarrhea, which no doubt is one of the most important ways by which sanitation and water supply affect health. Another priority is the impact of pollution on cancer risk. There is· a scarcity of data available to assess this directly in the Philippines, and so the estimates in this assessment (Tables 3.10 and 3.11) are based on general estimates from developed countries (Doll 1981 ). That the impact may be greater can be illustrated by the example of diesel dust. According to one risk model, the lifetime risk of lung cancer increases by 0.1 - 0.6 percent for each 100 µg/cu.m. of airborne exposure to diesel dust (Hattis 1994). In Manila, where diesel dust is ubiquitous, it is not inconceivable that as much as 100 µg/cu.m. might be contributed by diesel. If so, this alone could double the risk of lung cancer, since the cumulative lifetime risk at present is approximately 0.4 percent (Sarol et-al 1994).  - 41 - Table 3.10: Days-of Healthy Life Lost due to Environmental Health Conditions, 1990 Risk Attributable Condition All Causes DHLL to Attributable to Pollution per 100,000 People Pollution DHLL per 100,000 (%) People Direct Evidence: Diarrhea 368,012.6 18.6- 30.0 68,450.3 - 110,403.7 Heart Disease 555,935.5 4.0- 24.0 22,237.4 - 133,424.5 Bronchitis 1,528,118.5 8.4 - 55.2 128,362.0 - 843,521.4 Total Direct Evidence 2,452,066.6 8.9-44.3 219,049.7 -1,087,349.6 Informed Estimate: Schistosomiasis 31,672.3 100 31,672.3 Typhoid Fever 29,260.4 100 29,260.4 H-Fever 9,342.7 100 9,342.7 Hepatitis A 4,877.1 50 2,438.6 Lung Cancer 37,387.7 10 3,738.8 Liver Cancer 1,066.1 10 106.6 Leukemia 3,194.5 10 319.5 Protein Calorie 241,044.6 •, 10 ,- 24,104.5 . Malnutrition. Total Inf. Estimate 357,845.4 28 100,983.4 .·•.,_ Total Environmental- Related Conditions 2,809,912.0 11.4-42.3 320,033.1 - 1,188,333.0 ) Sources: As per Tables 3.7 and 3.8; World Bank estimates. 3.69 Based on these assumptions, it is estimated that about 28 percent of days of healthy life lost from the range of conditions involved (Table 3.10) could be attributed to pollution. In order of importance, the conditions are schistosomiasis, typhoid, malnutrition, H-fever, lung cancer, hepatitis A, leukemia and liver cancer. According to these estimates, the burden of disease for this group is substantially less than that for the conditions for which there is direct evidence. However, this lower estimate may also be the result of the poor information available. 3.70 Altogether, it is estimated that that between 11 and 42 percent of the days of healthy life lost for the range of conditions considered are attributable to pollution. Evidence available shows that the lower range is a large underestimate, for instance diarrhea. Therefore, it would be more reasonable to take at least the •mid point as a more realistic estimate of the burden of disease attributable to pollution. This is represents a considerable loss to the Philippines both in terms of costs of treatment and loss of earning by those in the workforce.  -42 - 3.71 An estimate has also been made to assess losses due to mortality alone (PYLL), rather the combination of morbidity and mortality (DHLL). There are fewer environmental conditions with clearly defined estimates for mortality than there are for morbidity. When analyzed from the standpoint of PYLL, environmental conditions have a disproportionate impact. . This is because mortality ·from diarrhea occurs, on average, at a young age, thus contributing a relatively large number of years of life lost from premature death before age 70. Nonetheless, the range of estimates of the burden of mortality is lower than morbidity because the impact of dust upon respiratory mortality is •1ess than it is on respiratory morbidity. 3. 72 Between 8 and 18 perent of years of life lost due to premature death are attributable to environmental pollution (Table 3.11). This is a slightly lower range than for DHLL (Table 3.10). Looking at individual conditions, the range of "person years of life lost" (PYLL) attributable to pollution is similar to the range of DHLLs in Table 3.10. This is because the estimates of attributable risk for diarrhea and heart disease are the same as above, while the estimate for respiratory mortality is based on the range of excess for mortality given in Table 3.4. Table 3.11: Years of Life Lost due to Premature Mortality Attributable to Pollution 1990 Premature Mortality Premature Mortality Risk Attributable Attributable to Condition from All Causes to Poliution Pollution Years of Life Lost % Years of Life Lost (Before Age of 70) (Before Age of 70) Diarrhea 342,600 18.6 - 30.0 63,724 - 102,780 Bronchitis 55,715 8.8-32.2 • 4,903 - 17,940 Heart Disease 640,287 4.0- 24.0 25,611 - 153,669 Malignant Cancer 364,826 5 18,241 Other 307,920 5 15,396 Total Environinental- Related Conditions 1,711,348 7.5 -- 18.0 127,875 - 308,026 Sources: As per Table 3. IO; World Bank estimates. 3.73 The estimated mortality burden due to exposure to pollution results in considerable economic loss. Lost earnings due to premature death are an important measure of this loss. The present value the discounted future earnings lost due to premature death in 1990, of people participating in the workforce, is estimated at about 2 percent or 4 percent of Gross Domestic Product, depending whether the mid-point or the upper-limit value in the range of mortality attributable to pollution is taken (Cortez 1995). The actual economic cost is much  -43 - higher because this considerable but conservative estimate excludes the cost of treatment of people sick from pollution - no reasonable estimate of treatment costs for the range of diseases involved is available. This estimate of economic cost also does not take into consideration losses involved in the large I.Q. decrements in children from lead exposure. - 44-  - 45 - 4. KNOWLEDGE AND ATTITUDES, MONITORING AND REGULATION 4.1 The large burden of disease arising from environmental pollution poses questions involving the ability of the private and public sectors to identify environmental health hazards and undertake preventive or abatement measures. Among others, two major issues are apparent. The first is the level of knowledge and attitudes that influence the behavior of member of households and others in the private sector related to health risks from environmental pollution. The second is the responsibilities of different government agencies in supporting preventive and containment efforts, through the identification of hazards, education, monitoring and regulation. KNOWLEDGE AND ATTITUDES 4.2 This assessment shows that changes in current practices related to environmental hazards would result in substantial health benefits. However, a review of the information available for each principal environmental exposure pathway (PEEP) reveals a paucity of data on knowledge and attitudes of the population, regarding pollution and its impact on health. Often, there is no adequate information to assess whether behavior reflects lack of knowledge or the influence of socioeconomic factors, such as general education, and income levels, in the prevention of pollution or reduction of its impact on health. Dust-Related Disease 4.3 The National Health Survey (Department of Health 1992a) indicates that the linkage between air pollution and pulmonary disease is recognized by a large proportion of the population. Nevertheless, existing information does not clearly show that the relative importance of specific risk factors is understood. For instance, although the importance of pollution from motor vehicles is often recognized, there is no obvious understanding of the effects of other factors in the prevention of pulmonary disease, such as respirable particles in the work environment, charcoal and wood as cooking fuels, agricultural and garbage burning. In all these cases, lack of knowledge is likely to be associated with economic factors, that contribute to current harmful practices. The absence of investments in the work place (Department of Health 1991a, 1991b; Occupational Safety and Health Center 1993), especially by small establishments with poor access to credit, is a major constraint in improving air conditions in the work place. In the case of road dust, improved pavement and roadside conditions require investments by national and local governments. The use of alternative cooking fuels needs an initial investment in new appliances and funds for the purchase ofless harmful fuels by the poor.  -46- Diarrheal and Skin Conditions 4.4 The same national survey (Department of Health 1992a) shows that knowledge about the association between pollution and diarrheal and other diseases is not as widespread, as it might be expected. Most people do not boil their drinking water, especially in rural areas, in spite of the substantial inadequacy of safe water supply. According to this survey, the proportion of . households that boiled their drinking water increased with education. However, the possibility of economic barriers need to be considered. This is illustrated by the increase in the proportion of households with sanitary toilets as income rises, in both urban and rural areas. The linkage between poor sanitation and skin disease is the least known (Department of Health 1992a), although the prevalence of skin disease attributable to poor sanitation is large (Health Safety and Environmental Management Consultancy Inc. 1993). Table 4.1 Households with Sanitary Toilet by Income, Philippines, 1992 Monthly Income Percentage of Households with Sanitary Toilet (Pesos) Urban Rural Philippines Less than 2,000 67 50 60 2,000 - 3,999 83 65 73 4,000 - 5,999 93 78 87 6,000 - 9,999 96 89 94 I 0,000 and over 98 91 96 Source: Department of Health, 1992a Lead Poisoning 4.5 It is difficult to ascertain how much people know about the relationship between lead pollution and health and child development outcomes, in urban and rural areas. There is perhaps little that individual households can do in relation to prevention from motor vehicle sources. However, data available point to other sources of lead poisoning (Fagela-Domingo et al 1994). This knowledge is important in the development of constituencies that will support measures that will diminish exposure to lead especially by children.  -47 - Table 4.2: Agency Responsible in the Principal Environmental Exposure Pathways Agency UrbAir CommWater Pest/R/Food UrbWaste RurPoint Department of Energy (DOE) X National Power Corp (NPC) X National Oil Coy (PNOC) X Dept Trade & Industry (DTI) Bureau Product Standards X Board oflnvestment X Bureau Domestic Trade X Dept Transport & Comm (DOTC) Land Transp. Office (LTO) X Department of Finance (DOF) Bureau of Customs X X Dept Public Works & H'ways (DPWH) X X Energy Regulation Board X Dept Int. & Local Government (DILG) Local Govt Units (LGUs) X X X X X Phil. Nat. Police (PNP) X Dept Environment & Nat Res. (DENR) X X X X X Envir. Management Bureau (EMB) X X X X X DENR Regional Offices X X X X Laguna Lake Dev. Auth. ., X Department of Health (DOH) X X X X X Env. Health Services (EHS) X X X X X Non-Comm.Diseases Service X Bureau ofFood & Drugs (BFAD) X DOH Regional Offices X X X X X Dept Labor & Employment (DOLE) Bureau Working Conditions X X X X 0cc. Saf. Health Cnt(OSHC) X X X Regional Offices X X X X Public Information Agency (PIA) X X .x Nat. Water Resources Council X Metro. Waterworks & Sewerage System (MWSS) X (Metro Manila) Loe. Water Util. Administ. (LWUA) X (Cities & Munic. outside Manila) Metro. Man. Dev. Auth.(MMDA) X Dept of Agriculture (DOA) X Fert. Pesticide Auth.(FPA) X Pest. Tech Adv. Cmt (PTAC) X Bureau of Plant Industry X Bureau Fisheries & Aquatic Res. X Presidential Task Force Solid Waste X  -48 - Pesticides and Morbidity 4.6 Poor knowledge of the toxicity of pesticides among users is clearly demonstrated by studies conducted in the Philippines. Only a small proportion of farmers and laborers are able to rank pesticides according to toxicity. Practices concerning application, storage and disposal confirm that the level of understanding of the impact of pesticides on health is extremely low (Warburton et al 1994). There is also an apparent lack of knowledge of pesticides currently in the ecosystem (Bhuiyan and Castaneda 1994). It is unclear whether farmers and laborers are aware that the symptoms of pesticide poisoning that many of them present (Warburton et al 1994; Rola and Pingali 1993) are caused by inappropriate application of pesticides. As most farmers obtain their knowledge from government technicians (Warburton et al 1994), it is apparent there is need to improve and or change the means of communication of advice to users. AGENCY RESPONSIBILITIES ALONG EXPOSURE PATHWAYS: ORGANIZATION ISSUES 4. 7 Many of the issues involved in the prevention and amelioration of environmental h~alth risks involves public goods, such as the quality of the air people breathe. Often, people affected by pollution have no ownership or control over the source of pollution. Therefore, there is a legitimate role for the public sector in the promotion of a healthy environment and regulation of harmful practices. This assessment includes the identification of agencies with responsibilities along the principal environmental exposure pathways (Table 4.2). Individual agencies may have a multi or single focus. For instance, among the single-focus agencies, the Department of Energy (DOE) (and its national agencies) and the Department of Transport (DOT) are mostly concerned with pathways related to air pollution, while the Metropolitan Waterworks and Sewerage System (MWSS) has responsibility, as the name indicates, for water and sanitation in Metro Manila alone. Three national agencies are multi-focal: Department of Environment and Natural Resources (DENR), Department of Labor and Employment (DOLE) and the Department of Health (DOH). Local Government units are hybrid in that they focus on a specific geographical area but are concerned with all the five pathways. • Diversity and Focus on Health 4.8 There are at least ten National government departments with responsibilities along the five principal environmental exposure pathways, and some thirty divisions, bureaus or authorities. The type of agencies involved in each pathway varies from about eight or nine in the case of Urban Solid Waste and the Rural Point Source Pathway to between sixteen and twenty three for the Fertilizer, Pesticides, Food, the Water Supply Pathway, and the Urban Air Pollution Pathway. This diversity of focus and numbers reflects strengths and specialization by the different agencies, as well as constraints in the attainment of a cohesive program of prevention and abatement of pollution with a focus on human health. The important role of DOH in ensuring a focus on health is apparent, as it is the only national agency that has health as its only responsibility. This also points to the continuing need for  -49- the Inter-agency Committee on Environmental -Health, as a means of bringing together the diverse interests of the different government agencies, and guide them towards cohesive programs of prevention of illness from pollution and improvements in health status and development. Dispersion and Dilution 4.9 The review of agency responsibilities also points to the potential for dispersion of effort and dilution of human skills and other resources. For instance, unless there are cooperative arrangements, it is unlikely .that the individual strengths of DENR, DOLE and DOH can be used in.a synergistic way_ in the establishment of a core mass of skillful people in different disciplines. This is a requirement to address the multidimensions of environmental health hazards described by the five principal environmental exposure pathways. The same applies in investment in equipment and facilities required for a purposeful program of prevention and improvement. There are already some cooperative ·arrangements among the three agencies that take cognizance, for instance, of the engineering skills in occupational safety in DOLE, the air pollution expertise in DENR, and the environmental health knowledge and skills in DOH. They form a useful base on which to develop greater cooperative efforts with a focus on the health impact of environmental hazards for which they are responsible. This would rely on the development of complementary skills and facilities to 'enhance the establishment of a core mass of experienced services. Local and Regional Services 4.10 An important issue that requires both legal and operational clarification is the demarcation of responsibilities for environmental health between local and national agencies. The recent devolution to Local Government units (LGUs) has created a degree of ambiguity, which has led to unclear definition of responsibility for implementation of monitoring and regulation functions. This has increased barriers to the flow of information to map the locus of national and regional environmental health hazards. It has also heightened obstacles to the evolution of cooperative operational arrangements, and deprived local populations of skillful human resources required to address most environmental health problems It is clear that most LGUs are too small to harness the range of skills and facilities to carry out their multifocal responsibilities. There are obvious benefits in the establishment of working arrangements between them and the specialized skills and resources of DENR, DOLE and DOH. To ensure that all people throughout the Philippines are served, systematic formal arrangements should be made between each individual LGU and the three national departments, on a regional basis. The question of critical mass and scale is also relevant to DENR, DOLE and DOH regional services concerned with environmental health hazards. A cohesive and complementary approach to environmental health at regional level is essential, in view of the paucity of resources available and the formidable task involved. An initiative that deserves close attention and evaluation is the program in Cebu, which has brought together the regional resources of DENR, DOLE and DOH into a joint program of  - 50 - occupational and environmental health improvement. This joint effort brings together medical and engineering skills focusing on a healthy environment and is complemented by a local committee that involves the agencies and the private sector. AGENCY RESPONSIBILITIES: MONITORING AND LABORATORY ANALYTICAL CAPACITY 4.11 Philippine legislation and regulations set out environmental standards regarding air, drinking water, food , and environmental exposure to various hazardous material. The characterization of the principal environmental exposure pathways revealed that there is a dearth of information available from environmental standard monitoring by the responsible agencies. Monitoring of environmental standards requires a network of people at local level that are trained to carry out the systematic collection of samples, and laboratory analytical capacity. A recent expert review of laboratory facilities in the Philippines of relevance to the monitoring of environmental health hazards (Villard 1995) showed that responsible agencies efforts are frustrated by the lack of adequate human resources and laboratory facilities, qualify assurance and organization. Air Quality 4.12 The Government of the Philippines has developed air quality guidelines for sulfur dioxide, suspended particles (and PMlO), nitrogen dioxide, ozone and carbon monoxide (DENR 1994). Air quality sampling is carried out by a small number of air sampling stations in Metro Manila and sites near power generation plants. Usually automated sampler analyzers can analyze contaminants including carbon monoxide, nitrogen oxides, sulfur dioxide and ozone. The older sampling stations are capable of testing sulfur dioxide and total suspended particles. The newer ones have enhanced capability including ground level ozone, but no capability for the measurement of organic carcinogens such as polycyclic aromatic hydrocarbons (PAHs). On-going air sampling in other parts of the Philippines is limited or non-existent. DENR and Laguna Lake Development Authority laboratories are extremely limited to provide these agencies with the analytical support necessary for monitoring and research purposes. Drinking Water and Sanitation ' 4.13 In support of the Public Health Act, Drinking Water Standards were rewritten in 1994. These standards require testing capacity in seven major areas: inorganic tests, such as hardness, alkalinity, cyanide and nitrate; bacterial tests such as fecal and total coliform; aesthetic tests such as color and turbidity; trace metal tests such as mercury and lead; tests for volatile organochlorines such as byproducts of chlorination; tests for organic substances such as phenolics, chlorophenolics and toxaphene; tests for pesticides and herbicides such as aldrin, chordane, 2,4-D and 2,4,5-T. An examination of the volume of laboratory analysis performed by a range of agencies showed that capability and capacity is greatly inadequate for chemical and microbiological testing., even in Metro Manila. Food Sanitation and Quality  - 51 - 4.14 The Philippines has food standards which list permissible food additives and chemicals. In addition, the Public Health Act stipulates the conditions under which food premises are maintained and milk, meat, vegetables and fish are obtained, stored and protected·from contamination. Usually, a testing regime for the surveillance of food quality for local consumption would involve: testing for a range of bacteria, such as total plate count, salmonella, cholera vibrio, clostridium and staphylococcus; physical and microscopic examination of containers and contamination with foreign matter; testing for chemical inorganic constituents such as mercury and lead; testing -for common pesticides and herbicides and banned substances; - and presence of organic industrial contaminants and solvents such pentachlorophenols, and PCBs. A review of laboratory capacity both at national and regional levels indicates that there is only a small dedicated laboratory capacity to handle chemical testing of food and relatively low volume of routine microbiological work on food. This is reflected in the dearth of information encountered during this assessment to characterize the relevant principal environmental exposure pathway. Capability and Capacity 4.15 It is apparent that the range of analytical capability is most _comprehensive in microbiology, but its is increasing limited in chemistry, heavy metals, pesticides and fertilizers, and organics testing. According to an expert "guess-estimate", the pooled capacity of different departments, at national and regional levels,· constitutes less than twenty percent of requirements for an active surveillance and monitoring system for air, water, food and hazardous materials. Capacity for food is particularly small (Villard 1995). A detailed assessment of requirements is needed to provide guidance for the development of laboratory capability and capacity. Skills, Training and Quality 4.16 In addition to lack of capacity, training on specific tests and experience in their performance is also hindering agencies' ability to perform their surveillance responsibilities. It is apparent that it would be beneficial to link training offered by national programs and donors with the testing requirements of statutory responsibilities. Need for training in laboratory management and quality assurance is also obvious (Villard 1995). During this assessment, concerns were expressed by a number of agencies regarding the quality of testing and possible improvements through accreditation processes. For an accreditation process to be useful in improving quality, a pool of technical knowledge and experience is essential. This would need to be built up from the existing meager base. Quality is also influenced by the volume of specific tests performed by staff in existing laboratories. The present fragmentation of capability and capacity places .major barriers to the development of adequate experience (Villard 1995).  - 52 - Equipment and Buildings 4.17 Visits to laboratories show that there is a wide range of equipment available but not being used. Possible reasons are poor operating conditions, inadequate laboratory facilities and lack of staff training. • The activation of idle equipment would make a significant contribution to capacity, but would not be sufficient to meet the requirements outlined. This activation is also unlikely to take place unless appropriate training is given to the staff and new facilities become available (Villard 1995) . With few exceptions such as the laboratories operated by the Bureau of Food and Drugs, existing laboratory buildings do not meet adequate standards of space, services, organization and safety . The realization of the full potential of equipment that might be acquired will be hindered by the quality of the buildings available (Villard 1995). Organization 4.18 Currently, helpful informal and formal mutual support are practised across government departments, levels of government, academic institutions and the private sector. An informal network has developed that helps sharing of the scarce expertise available. These features of practice constitute a base on which future cooperation can be built. A critical mass of experience and resources is required and involves a degree of specialization and concentration of expertise and resources in a few laboratories. These laboratories might institute explicit contractual arrangements with agencies with specific needs to discharge their responsibilities. At national level, it should be possible for different departments to develop individual capabilities for areas for which they are responsible. However, in the interest of efficiency and quality, a degree of specialization in lower volume testing is desirable. At local level, it is unlikely that LGUs can afford or are capable of developing adequate laboratory capacity. Therefore, an appropriate strategy would be to develop regional facilities (Villard 1995), to service not only LGUs in the regions but also the various national departments such as DENR, DOLE , DOH and DOA. Assessment and Monitoring 4.19 The characterization of the principal environmental exposure pathways showed that the most useful information contained data on both the source of pollution and the characteristics of people affected by it. Information from periodic and systematic monitoring by the various agencies responsible was often incomplete or unavailable in a timely manner. The findings of this assessment have relied to a great extent on the availability of population- based sample surveys. Good examples are the Manila school children survey and the Water Decade report. The most useful information on pesticides and food is also the result of sample surveys. The same applies to rural point sources. In the short term, it is possible but unlikely that the responsible agencies will be able to overcome current constraints in the collection of periodic and systematic information. Consequently, it would be useful to take advantage of the investment made in the design and experience gained in the use of population-based sample surveys, to monitor on a national basis the identified major threats to health from environmental pollution: dust, lead, water and sanitation, pesticides and food.  - 53 - The instruments developed by a number of researchers should be reviewed and improved where necessary, with a view to their use in stratified national samples. These national surveys could be repeated periodically to guide intervention programs and assess progress made in the prevention and reduction of health risks from environmental pollution. The private sector should also be encouraged to undertake self-assessment of occupational and environmental conditions in different areas of economic activity. These monitoring and assessment tools would provide the Inter-agency Committee on Environmental Health, and individual government agencies, with a factual basis on which relevant intervention programs can be developed and their impact reviewed. - 54-  - 55 - 5. FINDINGS AND AGENDA FOR ACTION 5.1 This assessment has provided considerable evidence of the impact of pollution on the health of exposed populations throughout the Philippines. In many cases, the evidence is strong and provides a reasonable basis for prioritization of areas for intervention. In some cases, information is reasonable to identify problems, their _source and give a sense of its dimensions, but further research would be helpful to provide guidance for appropriate intervention. Yet in other cases, there is a dearth of information on the dimensions of the problem and its impact on health. SUMMARY OF MAJOR FINDINGS Health Outcomes 5.2 There are significant environmental health problems identified along each of the principal environmental exposure pathways: • Dust (particulate matter) and lead are the most significant exposures from a human health perspective and respiratory diseases and LQ. decrements in children are the most important health outcomes, along the Urban Air Pollution Pathway. • Lack of Level III water supplies and in-house toilets are the most important "exposures" and diarrhea and skin conditions the most prevalent he. alth outcomes, along the Water Supply Pathway. • Pesticide exposures by agricultural workers and the pesticide-poisoning symptoms they experience appear to be the most important problem, along the Pesticides, Fertilizers, Food and Rural Waterways Pathway. This has gone greatly unrecognized by routine disease surveillance in the Philippines. • Sample studies along the Rural Point Source Pathway suggest that exposures to dust and industry-specific pollutants at a level which can affect human health do occur in the non- urban environment, these can be manifested as respiratory disease or poisoning-related conditions. • The health impact of the Urban Solid Waste Pathway is largely a problem among the people who live and work on dumpsites around the country, and depends upon whether or not there is burning of garbage or incomplete combustion occurring at the sites. Leakages from dumpsites might also affect populations nearby. Respiratory disease, lead-poisoning, and intestinal parasitism are some of the manifestations of exposure to the range of health hazards in this pathway.  - 56 - 5.3 The conservative estimates made in this assessment show a considerable burden of morbidity and disability from exposure to dust, lead, pesticides and the lack of safe water supply and toilet facilities in the Philippines. They justify a high order of priority in addressing these causes of health problems. In 1990, the human cost of environmental pollution in the Philippines could be summarized as follows: Bronchitis due to dust 294 million days of healthy life lost Heart Disease due to dust 47 million days of healthy life lost Diarrhea due to less than level III water supply and no toilet • 54 million days of healthy life lost Lead Exposure average 5 points ofl.Q. loss among all children Premature mortality from pollution: Bronchitis, Diarrhea, Heart Disease 184,000 years of life lost 5.4 The available data shows that health effects of environmental pollution along at least three pathways (CommunWater, RurSource, and Pest/Rural/Food) are mediated by household variables. In particular, home cooking fuels, the absence of a toilet, and crowding are important. In general, available data indicate that environmental factors are a greater risk than smoking, in contrast with the situation in developed countries. This is not because smoking does not have an effect, since it does, but because other factors seem to be stronger in the Filipino context. Moreover, urban air quality seems to make crowding and cooking fuels less important risk factors in'the city than in the rural areas. Similarly, there is evidence that pesticide applicators' knowledge, attitudes and practices are important mediators of certain forms of environmental risk in the Philippines. 5.5 It is clear from several studies in the Philippines that childhood -malnutrition is highly prevalent, when measured by stunting and underweight. This is important because even borderline levels of malnutrition can have a significant impact on intellectual development (Grantham-McGregor 1991). Therefore, stunting and underweight can serve as surrogate measures of intellectual development. The interaction between pollution and malnutrition is also important. In particular, two issues are worthy of further consideration. The first is the role of poor water supplies and the lack of toilets in promoting diarrhea, which leads to or exacerbates borderline malnutrition. The second is the potential synergy of lead overexposure and malnutrition in impeding intellectual development. 5.6 It is clear that working conditions in many workplaces in the Philippines carry important health risks. Risk factors often include noise, dust, airborne fumes, heat, and poor · sanitary conditions. Long working hours can add to the exposure time, and thus to risk. From non-systematic observation, the workroom air measurements from Tagum (Gapas et al 1991) could be typical of the overexposures which occur at workplaces in the Philippines.  - 57 - 5.7 The role of carboxyhemoglobin as a risk factor for heart disease risk in heavily exposed urban populations deserves further investigation. The fact that blood lead levels in rural areas is so high is surprising and disturbing. The role of lead in food, milk, and roadside dust have been subject to preliminary investigation, but more thorough study is needed. Similarly, there needs to be further investigation of waste disposal outside Manila, with particular emphasis on the living and working conditions of scavengers. It would also be important to learn more about the health implications of uncollected garbage by roadsides, in play areas, and in streams. 5.8 Risks from different food sources have not been adequately studied. For instance, are there systematic differences in the microbiological quality of food from the different locations where Filipinos buy cooked or raw food? Are Filipinos with certain purchasing practices at greater risk than others? Individual practices may also be important in relation to pesticide residues. Cooking can reduce the residues of certain pesticides by 40 - 80% and washing can reduce them by 15 - 70 percent (Tejada 1994). What proportion of Filipinos follow these practices, and are there differences among them in terms of biological measures of pesticide exposure or symptoms? Knowledge and Attitudes 5.9 Knowledge and attitudes of householders, workers, and entrepreneurs in the private sector are significant factors in the reduction of environmental hazards and their impact on health. There are two major variables associated with the recorded range of perceptions and attitudes: education and income. Consequently, intervelitions aimed at improving the present situation need to be targeted to relevant populations and take into consideration the economic barriers involved. 5.10 In the case of dust-related diseases, the health hazards of respirable particles from transport are reasonably perceived, but the hazards presented by dust in the work place, from charcoal and wood as cooking fuels, and from agricultural and garbage burning are not clearly understood. In spite of its importance, knowledge of the effects of lead exposure on child development seems to be confined to a relatively small and better educated group. Knowledge of the impact of poor sanitation and water quality on diarrheal disease is not as widespread as might be expected. The proportion of households that boil their drinking water decreases with lower education, and the availability of sanitary toilets falls with lower incomes levels, in both urban and rural areas. From the sample surveys available, the knowledge of and attitudes among farm workers towards the toxicity of pesticides, their storage, application, and disposal is highly inadequate and is associated with a high prevalence of symptoms alike pesticide poisoning. Agency Responsibilities 5.11 Environmental health is concerned in most cases with "public goods", such as the quality of air that people breathe. Also in many cases, the people affected by pollution do not own · or control the source of pollution. Consequently, there is a legitimate role for  - 58 - government and the public sector in the promotion of an environment that is not damaging to health and in the regulation of harmful practices. The multi-facet nature of economic and other activities affecting pollution, and the multitude of sources, has inevitably led to the involvement of a large number of agencies in monitoring and regulation of pollution affecting human health. The agencies involved have either a multi or single focus. Their focus might be on an economic activity, such as transport, or a geographical area, such as Manila. Local government units are a hybrid type, that has a single geographical focus and responsibilities over a multitude of economic activities and household sources of pollution. 5.12 Three relevant national agencies have a multi focus on a number of economic activities and geographical areas: Department of Environment and Natural Resources, Department of Labor and Employment and the Department of Health. All have relevant specialization and major roles to play in environmental pollution control. The Department of Health is the agency with health specialization and interest to give a health focus to government and private sector activities. In addition to these three national multi focus agencies, the array of single-focus agencies has specialized knowledge, experience, and statutory power that add to national capability. The Inter-agency Committee on Environmental Health chaired by the Secretary of Health has the potential to integrate the varied interests, responsibilities, skills and resources into a national environmental health program, with shared responsibilities. 5.13 The spread of the scarce human resources for the improvement of environmental health, among the numerous agencies, has led to effort dilution. This has been only partly offset by laudable cooperative efforts, involving the relevant agencies and the mobilization of the private sector, especially in time of major emergencies. Cebu offers an example of cooperative efforts worthwhile evaluating and possible replication. DENR, DOLE and DOH have pooled their human resources and complementary skills in a program aimed at ameliorating environmental and occupational health problems. The private sector and local government are also involved in this joint effort. 5.14 A major barrier to cooperative efforts is the present ambiguity in the legal and resourcing responsibilities of national and local government agencies, arising from efforts to devolve responsibilities and resources to local government in the Philippines. This has hindered possible arrangements for the sharing of the paltry human resources available. It has also reduced information flows required for the identification of national trends and pnont1es. In addition, the scale of local government units is an impediment to the organization of affordable and effective technical services in monitoring and supporting remedial action in environmental health. 5.15 This assessment shows that there is a gap between the statutory responsibilities of national agencies and local government and the information available to assess compliance with formulated standards. Relevant agencies are frustrated by the lean spread of the network of trained people at central, regional and local levels, and the facilities, equipment and organization for laboratory analysis. The monitoring of air quality standards is insufficient in range and geographical coverage in Metro Manila, and there is virtually no monitoring  - 59- capacity outside Manila, if monitoring associated with power-generating plants is excluded. There is also limited capacity for the monitoring of microbiological contamination of water, and even lesser capacity for the monitoring of chemical and heavy metals pollution. The capacity dedicated to monitoring of food contamination is also relatively small. An initial estimate indicates that present laboratory capacity and capability to meet statutory requirements might be only twenty percent of requirement. 5.16 The lack of analytical capacity is accompanied by a paucity of information on the determinants of cancer, hepatitis, H-fever, and other infectious diseases of relevance to environmental health. Access to data on pesticide exposure and contamination of the food- chain is limited to small sample surveys. Information on occupational exposures and diseases is also lacking. Data collected indicate the National Poison Control Center receives only a relatively small proportion of poisoning cases "in the field". It is apparent that the present system of reportable diseases to Department of Health greatly underestimates the total burden of disease; and the recorded geographical distribution could be misleading if used for the determination of regional priorities. In addition, much of the information collected by different agencies and academic institutions of relevance to environmental health does not flow to the Department of Health. Consequently, it is not organized with the human health focus given in this assessment. The present strength, that should be built upon, is the experience with a number of well designed population-based surveys that provided most of the useful and reliable information used in this assessment. AGENDA FOR ACTION Priority Areas 5.17 The preparation of this assessment used a participatory approach. This involved the participation of expert national researchers, discussions and feed back from local and national agencies, and the private sector in Manila, Cebu and Mindanao. In addition, the Inter-agency Committee on Environmental Health (IACEH) of the Philippines chaired by the Secretary of Health was party to the discussion and adoption of objectives, methodological approaches . used in the assessment, and the interpretation of findings, and formulation of recommendations. In discussions of the IACEH there was agreement that the findings point to five priority areas for prevention or reduction of the health impact of environmental pollution: • Dust (particulate matter) from industrial, transport and household sources affecting respiratory diseases. • Lead exposure from transport and other sources affecting child development. • Contamination of drinking water and poor sanitation leading to diarrheal and skin diseases. :.,_::  - 60- • Pesticide exposure from poor use affecting the health of farm and plantation workers: • Food contamination leading to diarrheal diseases. Proposals for Short-Term Action 5.18 During a workshop in June 1995, the IACEH formulated proposals for short-term activities aimed at either preventing or diminishing the impact of exposure to the five priority areas identified. These proposals contain three useful elements in relation to each of the five areas: measures for the prevention or reductions in exposure; mechanisms for relevant environmental surveillance; measures to improve behavior of government, industry and community to reduce the impact of the environmental health hazards. The IACEH also prepared research proposals for each of the five priority areas. They show the objectives of the research, the locus of responsibility, research collaborators, timing and source of funds. The proposals also indicate how, when and by whom these activities should be undertaken. 5.19 At a follow-up workshop also in June 1995, discussions took place and proposals were made to deal with: specific issues identified; mechanisms to strengthen linkages between agencies; specific interventions and research activities. Some of these proposals have already been acted on. The proposals should be reviewed by each relevant agency, to ensure that they are capable of integration in the work program of the agencies responsible, and are appropriately resourced. These two sets of proposals are included as appendices to this report. Development of Medium-Term Interventions 5.20 In the development of a program of interventions, the application of a number of principles would be useful to enhance success. Some of these principles are already being applied to some programs in the Philippines (GTZ 1995). They should be adopted and followed in a consistent manner to improve the behavior of the major players, including government, at national and local levels, industry, NGOs and households, in their contributions to improvements in environmental health. These principles involve: • Focusing on health, and integration of the implementation of national environmental policies. • Sharing responsibilities, among different levels of government, among environmental, labor and health agencies, between public and private sectors, and different areas of economic activity. • Public participation to enhance informed behavior change and sharing of responsibilities. • Prioritization of problems and activities.  - 61 - • Costing of action plans. These principles have wide acceptance in the Philippines and in other countries(WHO 1995). The work-up to the completion and implementation of the IACEH proposals would benefit from the systematic application of these principles. A Medium-Term Agenda 5 .21 In the development of policy and its implementation, the Philippines has a number of options that involve the use of self-improvement by the private sector, formulation and use of guidelines, provision of technical advice and support, use of economic tools, supply of information to consumers and interested groups, creation of incentives for behavior modification, and regulation. Some of these tools will be more appropriate .than others depending on the problems being faced and their possible remediation. The following agenda for medium-term action is concerned with the five major priority areas identified, and points to a range of possible measures in response to the health hazards identified. 5.22 Dust-Related Diseases. A major source of respirable particles and harmful gases causing pulmonary and other diseases are emissions from motor-vehicle transport. A long- term agenda would include the development of alternative and more efficient means of mass transport less harmful to health. In the medium term, in addition to proposed efforts in reducing "smoke-belching", the agenda for action should include a combination of incentives for change to less harmful fuels, more efficient engines and use of emission abatement devices. All these are being considered in the Philippines. The medium term agenda should make use of a mixture of tools available, including advice on the advantages of L:rG fuel for mass transport vehicles and incentives •for its use. This would eliminate the emission of respirable particles and lead from this source, and reduce considerably the emissions of carbon monoxide and reactive hydrocarbons (Homer 1993). The "polluters pay" principle could also be applied, by increases in taxes on more polluting fuels and exemption or lower taxes for LPG. This would provide added incentive for the desired substitution. The availability of credit would also be an incentive for the owners of jeepneys, taxis and others, to finance the modification of existing vehicles and for substitution of existing inefficient engines by more efficient ones, less harmful to health. Modeling exercises have found that a combination of targeted taxes and substitution are likely to achieve best results (Anderson and Cavendish 1992). 5.23 In rural areas, the impact of cooking fuels on dust-related diseases is a major agenda item. In the longer term, the generic household energy ladder indicates there should be a gradual substitution of wood and charcoal by less harmful and more efficient fuels (Brandon and Ramankutty 1993). In the medium term, the agenda should include efforts to create demand for stoves with higher energy efficiency and lower emissions (Barnes et al 1994). As females are the major users of biomass fuels in households and also those likely to be most affected, together with children, the availability of small loans to poor women would help them to purchase more efficient and less harmful stoves. NGOs should be mobilized in this drive. The supply of more efficient stoves may require technical assistance by relevant  - 62- authorities to local producers of biomass stoves (Barnes et al 1994). It may also depend on the availability of credit to private entrepreneurs considering the local manufacture of low cost and efficient stoves. Mechanisms to improve ventilation of houses in rural and poor urban areas is another item that requires research and consideration as an effective intervention. 5.24 Other agenda items to respond to the hazards caused by respirable particles in rural areas are road dust, and soot from the burning of sugar cane and household solid wastes. In the longer term, advice and incentives should be ·given, to avoid the location of houses in the immediate proximity of dusty rural roads and the placement of entrances and windows in a fashion that will reduce exposure to road dust. In the medium term, research should take place on methods that would allow households and local government to reduce road dust. In the longer term, mechanical· harvesting might reduce the burning of sugar cane foliage. In the medium term, growers should be encouraged to •use sugar cane foliage as a source of energy, rather than following the practice of burning it before harvesting. Advice on the harmful effects of burning of solid wastes and the benefits of composting should also be used to reduce exposure to respirable particles in rural areas. NGOs and relevant specialized agencies of the national government would need to work together towards these ends. 5.25 Dust in the work place is another problem identified. Of major concern are conditions in small industrial establishments with poor access to credit facilities. In addition to social marketing techniques to promote awareness of the health hazards found in the work place, incentives might be given through technical advice and improved accessibility to credit, that will help the owners of small establishments to improve both their efficiency and working conditions. 5.26 Lead Poisoning. · There is no doubt that exposure to lead is a major hazard to child development in the Philippines, even •though there are questions that need further investigation. In the longer term, the change from leaded to unleaded fuels for motor vehicles should make a major impact on lead exposure of children and adults. In the medium term, the suggested incentives for changes in fuel use and engines of mass- transport vehicles would also make a significant contribution to lessen human exposure. However, the high levels of lead found in rural areas is a signal that alternative sources of lead exposure might exist. Among others, an area that should be reviewed as a potential source of lead exposure is lead in the glazing of containers for cooking and storing of food, and lead in metals_ used in cooking, drinking and eating utensils. 5.27 Water and Sanitation Related Diseases. The evidence presented in this assessment shows that the absence or quality of sanitation and water supply cause large burdens of diarrheal and skin diseases throughout the Philippines. One of the estimates indicates that as much as 54 million days of healthy life are from this source. It is apparent that household size and crowding are important factors that have a major impact on the prevalence of these diseases. They will need to be •addressed over a longer term. In the medium tetm, the provision of household toilets and safe water supply to the poor are appropriate objects of poverty alleviation and social policy because of the externalities involved in the spread of  - 63 - disease. As there is a positive association between income of households and the availability of sanitary facilities (Table 4.1 ), the targeted provision of sanitation to the poor could be also considered as a measure of social equity. The present program concerned with the provision of basic sanitary facilities will not meet the challenge presented and reduce substantially the burden of disease. It needs to be reviewed to make a larger and realistic impact. This review should consider the use of local NGOs in the mobilization of local resources and promotion of self-help when appropriate. In addition, social marketing techniques should also be used to encourage drinking-water boiling as a preventive measure. This should be targeted to householders with low levels of education, as there is an association between levels of education and water boiling practices. 5.28 Pesticide Poisoning. To gain a better understanding of the full dimensions of the burden of disease posed by pesticide poisoning a program of relevant research should be organized and systematically implemented. Nevertheless, information available points to the need for a number of behavioral changes. In the first instance, the level of knowledge and attitudes of users needs to be improved. In this respect, the means of conveying information to people with low levels of education in sparsely populated rural areas should be examined and alternatives evaluated. . In the past, government agencies have been the main source of information on pesticide use and toxicity for rural workers. If the level of knowledge found in the sample surveys available is used as a guide, it is apparent this source or the means used has been ineffective. The behavior of suppliers also needs to be improved to make them more responsible for the disposal and retrieval of containers used in the distribution of pesticides. Another item of concern is protective clothing. Research should be undertaken to develop protective clothing and devices that are affordable and comfortable in the tropical climate of the Philippines. 5.29 Food-Contamination Related Diseases. There is need to increase knowledge on the biological and chemical contamination of food in the Philippines and its impact on health. Nonetheless, data available from surveys points to contamination of food reaching the market place and its further deterioration due to exposure after its arrival in the market. In many cases, the facilities available and their management are contributing factors. It is not clear whether municipalities and other local government have the resources to improve local market places. The management of these markets by NGOs or other private entities is an alternative. These private entities would be given the management on a contractual basis, including the maintenance of hygienic conditions and standards that reduce exposure and contamination. The investment and running costs would be recovered from the rent of selling areas. To improve the quality of ready-to-eat food from commercial sources and households, it will be necessary to change their behavior to increase the adequacy of washing and cooking, as a means of reducing both biological and chemical contamination. A private voluntary rating system (NGO-based) could also provide information to consumers to favor rated ready-to-eat food establishments and vendors, and give incentives to providers to be rated, in accordance with adequate standards.  - 64- Institutional Strengthening 5.30 In support of efforts to reduce the level of disease burden caused by environmental pollution, the medium term agenda should include the fostering of a predisposing and enabling social and political milieu, and institutional strengthening. 5.31 Public participation is a helpful step in this direction. This involves communication with and the involvement of interest groups, professionals, politicians, at national and local levels, to raise their awareness and create helpful constituencies that will support resource mobilization and activities for change. It will include dissemination and sometimes appropriate dramatization of information on the sources of environmental health hazards, their impact on specific populations, and their costs in terms of mortality and disability. The engagement of the various groups will require effective approaches, including the concerted use of mass media in given geographical areas, communication with NGOs, use of opinion leaders, and support with relevant material and technical advice; and the use of professional, industrial and commercial organizations in the promotion of environmental health standards and practices. The formal education system could be used further, to build helpful attitudes and hands-on academic experience on the nature of environmental health hazards, which will reinforce child and adolescent understanding of appropriate responses to the environmental health challenge. Further, the involvement of the population at large in the process of change and self help needs to be encouraged and supported. 5.32 Sharing of responsibilities in an integrated national approach is another agenda item in institutional strengthening. To improve working relationships and appropriate division of effort, it will be important to clarify the statutory responsibilities and resourcing of national agencies and local government. It is obvious that regardless of legal statutes, there is need for cooperative efforts to use effectively the inadequate resources available to national and local agencies. There are also "natural" specializations among national agencies that should provide the basis for cooperative efforts. Although the Department of Health has statutory responsibilities and specialization in health, it needs to work with the Department of Environment and Natural Resources and the Department of Labor and Employment, among others, using their statutory powers in specified areas and their specialist skills, in the abatement of pollution and environmental hazards that affect health. The same applies to the -sharing of responsibilities between national and local government. In this case, the ability of national agencies to address inadequate responses by individual local government units should be considered by the government and clarified by appropriate regulation. The present practice of formulating memoranda of understanding (MOUs) between partner agencies should be used more systematically, and adopted as a basis to review the progress made in the accomplishment of shared responsibilities. 5.33 A greater focus on environmental health and enhanced monitoring should improve the quality of information from routine collections in the longer term. In the medium term, training of health professionals in the recognition of conditions arising from environmental and occupational hazards would improve monitoring. The network of environmental health  - 65 - workers and other health providers, as well as local government, are a developing basis for the monitoring of environmental hazards and their impact on health. 5.34 At present, this network is hindered by the lack of appropriate analytical laboratory capability and capacity. Improvements need to take place in skills, equipment, facilities, and organization and management, including quality assurance. The present informal arrangements to share and mobilize scarce expertise forms a basis for future development and cooperation. In the long term, central agencies should develop appropriate capability and capacity. In the medium term, the establishment of a critical mass of experience and resources will require specialization and concentration of resources in a few laboratories. Low volume tests will continue to require specialization. Contractual arrangements should run parallel to the development of such facilities. At local and •regional level, local government units are unlikely to be able to afford their own laboratories, and the regional services of central agencies would benefit from the pooling of resources in few laboratories. An appropriate strategy would be to establish regional laboratories to be used by local government and the national agencies on a cooperative basis, with appropriate contractual arrangements. 5.35 In addition, impact monitoring and policy development require appropriate information on the major areas of concern. Advantage should be taken of the national capacity for the design and organization of population-based surveys and experience gained from them, to monitor the impact of pollution threats identified: dust and lead exposure, poor water and sanitation, pesticide exposure and food contamination. Revised sample-survey instruments should be used in national stratified samples to provide periodic information of the impact of pollution on populations. These national surveillance tools should be organized either under the auspices of individual agencies or the Inter-agency Committee on Environmental Health. In either case, they should have a health focus and the findings should be reviewed by the IACEH, with a view to their integration in the development of national environmental health programs and policy advice to government. The results should also be made public through appropriate media. 5.36 The Department of Health will need to provide a focal point and institutional capacity for epidemiological, statistical and other skills required in environmental health risk assessment and impact on the population. The Environmental Health •and Health Risk Assessment Unit in the Department of Health could be developed to contribute to this capability, which should harness relevant resources in specialized agencies, academic institutions and the private sector. In addition, the development of national policy advice to the government and the formulation of integrated national programs for environmental health require a consultative and deliberative device, through which relevant agencies can work together with a health focus. Even though in some cases other players will need to be added, such as the proposed use of the taxation system, the Inter-agency Committee on Environmental Health could be energized, with appropriate support, to provide a basis for the fulfillment of this objective. - 66 -  - 67 - References Anderson, Dennis and William Cavendish. 1992. Efficiency and Substitution in Pollution Abatement: Three Case Studies. World Bank Discussion Papers 186. Washington D. C. Bacud, L., F. Sioco, and J. Majam. 1994. A Descriptive Study of the Water Quality of Drinking Wells Around the Payatas Dumpsite (Student Thesis). Manila: College of Public Health, University of the Philippines. Baghurst, P. A., A.J. McMichael, N. R. Wigg, G. V. Vimpani et al. 1992. Environmental Exposure to Lead and Children's Intelligence at the Age of Seven Years: The Port Pirie Cohort Study. The New England Journal of Medicine 327(18): 1279-1284. Baker, J. , R. Santiago, et al. 1993. Vehicular Emission Control in Metro Manila. Barnes, Douglas F., et al. 1994. What Makes People Cook with Improved Biomass Stoves: A Comparative International Review of Stove Programs. World Bank Technical Paper 242. Energy Series. Washington D. C. Bellinger, D., A. Leviton, C. Waternaux, H. Needleman and M. Rabinowitz. 1989. Low- Level Lead Exposure, Social Class, and Infant Development. Neurotoxicology and Teratology 10: 497-503. Bhuiyan, S. and A. Castaneda. 1994. Impact of Ricefield Pesticides on the Quality of Freshwater Resource. From: Impact of Pesticides on Farmer's Health and the Rice Envirol1Il)ent. Pingali, P. and P. Roger, eds. (Draft). Bradley, David et al. 1992. A Review of Environmental Health Impacts in Developing Country Cities. Urban Management Program Discussion Paper 6. World Bank. Washington D.C. Brandon, Carter and Ramesh Ramankutty. 1993. Toward an Environmental Strategy for Asia. World Bank Discussion Papers 224. Washington D. C. Cagauan, A., 1990. Fish Toxicity, Degradation Period and Residues of Selected Pesticides in Rice-Fish Culture. Freshwater Aquaculture Center, Central Luzon State University, Nueva Ecija. Paper presented in the Workshop on Environmental and Health Impacts of Pesticide Use in Rice Culture, International Rice Research Institute, Los Baiios. Caldwell, J. C. 1986. Routes to Low Mortality in Poor Countries. Population and Development Review 12: 171-220. Calumpang, S. 1993. Human Exposure to Pesticides. Annual Report. National Crop Protection Center (NCPC), UP Los Baiios.  -68 - _ 1994: 77(3). The Philippine Agriculturist. Fate of Niclosamide in Various Components of a Fish and Prawn Pond Ecosystem. Chandra, R., ed. 1988. Nutrition and Immunology. Contemporary Issues m Clinical Nutrition, Vol. 11. New York: Alan R. Liss, Inc. Cortez, A. M. 1995. Economic Evaluation of Mortality. Quezon City: University of the Philippines. Dayrit, M., M. White, et. al. 1990a. An Outbreak of Typhoid Fever in Minglanilla, Cebu Caused by Water-borne Infection. Field Epidemiology Training Program. Department of Health. . 1990b. Water-borne Cholera Outbreak in Barangay Bato, Lamitan, Basilan. Field Epidemiology Training Program. Department of Health. Department of Environment and Natural Resources (DENR), 1988 to 1993. National Capital Region. Air Quality Monitoring Reports. 1990a. Environmental Management Bureau. A Feasibility Study on the Pasig River Rehabilitation Project. _ 1990b. Environmental Management Bureau. Manila Bay Monitoring Program. _ 1992a. Report on Industrial Efficiency and Pollution Control Program. _ 1992b. Environmental Management Bureau. Ecological Indicators of Pollution Report. _ 1992c. Environmental Management Bureau. Drinking Water Quality of Metro Manila Project. 1992d. Environmental Management Bureau. Environmental Management Strategy. _ 1993. Metropolitan Environmental Improvement Program. Urban Air Quality Management in Asia. _ 1993 and 1994. Regional Reports. _ 1994. Environmental Management Bureau. Air Quality Monitoring Reports Near Power Stations. Department of Health, 1989 to 1994. Disease Surveillance Alert. 1991a. Non-Communicable Disease Control Service. Highlights of the National Occupational Health Survey Phase I (NCR).  - 69- 1991b. Non-Communicable Disease Control Service. Highlights of the National Occupational Health Survey Phase II (Region IV). _ 1992a. National Health Survey. Manila. _ 1992b. Environmental Health Service. Environmental Health Program for 1993 to 1998. Department of Health. 1992c. Field Epidemiology Training Program. Cholera Outbreak. in Masbate, Masbate. _ 1992d. Field Epidemiology Training Program. Dengue Fever (Hemorrhagic) Fever Outbreak. in Ifugao. 1992e. Health Intelligence Service. Philippine Health Statistics. 1993a.. Field Epidemiology Training Program. Cholera Cases in San Lazaro Hospital. _ 1993b. Field Epidemiology Training Program. Cholera Admission Rate by City and Municipality. 1993c. Field Epidemiology Training Program. Hepatitis A Cases m San Lazaro Hospital. 1993d. Field Epidemiology Training Program. Preliminary Report on Cholera Investigation in Balenciagao, Pasil, Kalinga-Apayao. _ 1993e. Field Epidemiology Training Program. Investigation of Reported Diarrhea Cases in El Nido, Palawan. _ 1993£ Field Epidemiology Training Program. Cholera Outbreak. Investigation in Pasil, Kalinga-Apayao. _ 1993g. Field Epidemiology Training Program. Pre.liminary Report on Diarrhea Outbreak. in Isabela. _ 1993h. Field Epidemiology Training Program. Food Poisoning Outbreak., Iba, Zambales. _ 1994a. Field Epidemiology Training Program. Rapid Assessment Survey Concerning the Cholera Outbreak. in Camarines Sur. 1994b. Field Epidemiology Training Program. Update on Cholera Outbreaks. 1994c. Field Epidemiology Training Program. Summary of Recent Cholera Outbreak. Investigations. Department of Labor and Employment. 1992. Bureau of Working Conditions and Bureau of Labor and Employment Statistics. Occupational Injuries and Work Accidents.  - 70 - Department of Public Works and Highways. 1989. Feasibility Study on Solid Waste for Metropolitan Manila, Philippines. Department of Science and Technology. 1982.'.1990. Food and Nutrition Research Institute Survey Reports. Department of Trade and Industry. 1995a. Philippine Merchandise Imports of Organic Chemicals FOB Value in US Dollars. _ 1995b. Philippine Merchandise Imports of Chemical Materials and Products, N.E.S. FOB Value in US Dollars. _ 1987-1993. Foreign Trade Statistics. Department of Transportation and Communications, Land Transportation Office. 1995. Number of Motor Vehicles Registered by Region by Classification & by Type of Fuel Used 1990-1994. Dockery D. W. and C.A. Pope III. 1994. Acute Respiratory Effects of Particulate Air Pollution. Annual Review of Public Health 15:107-132. Doll, R. and R. Peto. 1981. The Causes of Cancer: Quantitative Estimates of Avoidable Risks of Cancer in the United States Today. Oxford- New York: Oxford University Press. Donaldson, H.M., and W. T. Stringer. 1980. Beryllhnn •sampling methods. American Industrial Hygiene Association Journal 41(2), 85-90. Employees Compensation Commission. 1994. Total Benefits Awarded, Total Number of Claimants and Average Benefits Paid Per Claim 1977-1994. Fagela-Domingo, C., B. Alavanza, et. al. 1994. Milk, Lead Levels and Mental Development. National Research Council of the Philippines. 39(3); 515-532 Philippine Council for Health and Research Development- Department of Science and Technology. Fertilizer and Pesticide Authority. 1987-1991 reports. Pesticide Importations. 1990-1993 reports. Fertilizer Importations by Grade, By Month, in Metric Tons. Flieger, W., and J. Cabigon. 1994. Life Table Estimates For the Philippines, Its Regions and Provinces, By Sex: 1970, 1980 & 1990. Health Finance Development Project. Monograph No. 5. Manila. Gapas, J., R. Subida, and E. Torres. 1991. Neuro-developmental Profile of 3-6 y.o. Children Residing in a Mercury Exposed Gold Mining District in Tagum, Davao del Norte, Philippines. Manila: German Agency for Technical Cooperation/ College of Public Health, University of the Philippines.  - 71 - Gracey, M., ed. 1985. Diarrheal Disease and Malnutrition: A Clinical Update. New York: Churchill Livingstone. Grantham S.M., C. A. . Powell, S. P. Walker and J. H. Himes. 1991. Nutritional supplementation, psychosocial stimulation, and mental development of stunted children: the Jamaican Study. The Lancet 338: 1-5. GTZ 1995. The Philippines-German Project: Industrial Pollution Control - Cebu. Cebu. Hattis D. and K. Silver. 1994. Use of Mechanistic Data in Occupational Health Risk Assessment: The Example of Diesel Particulates. In Chemical Risk Assessment and Occupational Health. Westport: Auburn House. Health Safety and Environmental Management Consultancy, Inc., for the Department of Health. 1993. Evaluation of the International Drinking Water Supply and Sanitation Decade in the Philippines (1980-1990). Heitzman, C. 1995. Environment and Health in Central and Eastern Europe. A Report For The Environmental Action Programme For Central And Eastern Europe. World Bank. Washington, D.C. Hertzman, C., H. Ward, N.Ames, S. Kelly and C. Yates. 1991. Childhood Lead Exposure in Trail Revisited. Canadian Journal of Public Health 82: 385-391. Homer, John 1993. Natural Gas in Developing Countries: Evaluating the Benefits to the Environment. World Bank Discussion Papers. Washington D; C. International Agency for Research on Cancer. 1984. "Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans: Polynuclear Aromatic Compounds, Part 3, Industrial Exposures in Aluminium Production, Coal Gasification, Coke Production, and Iron and Steel Founding". Vol. 34. Geneva: World Health Organization. Jin, A., C. Hertzman, S. H. S. Peck, and G. Lockitch. 1995. Blood Lead Levels in Children Aged 24 to 36 Months in Vancouver. Canadian Medical Association Journal 152: 1077-1087. Kingley, G. Thomas et al. 1994. Managing Urban Environmental Quality in Asia. World Bank Technical Paper Number 220. Asia Technical Department Series. Washington D. C. La Goy, E.T. 1987. Estimated soil ingestion rates for use in risk assessment. Risk Analysis 7: 355-359. Laguna Lake Development Authority. 1995. Laguna de Bay Master Plan (Draft) Last, J. 1993. A dictionary of epidemiology. Toronto: Oxford University Press. Laudico, A., C. Esteban, et.al. 1993. Cancer in the Philippines. Vol. 2.  - 72- Lifschitz, CH. and B.L. Nichols., eds. 1990. Malnutrition in Chronic Diet-Associated Infantile Diarrhea: Diagnosis and Management. San Diego, California: Academic Press, Inc. Local Water Utilities Administration. 1994. Bacteriological Quality Reports by Area. Maramba, N. 1989. A Cross-sectional Study of the Health Impact of Mercury Exposure Among Workers in Gold Processing in Davao Del Norte. German Agency for Technical Cooperation/College of Medicine, University of the Philippines, Manila. McMichael, A. J. 1994. Lead in the Human Environment: Problems and Risks. Personal Communication. Medina, J., S. Calumpang, and M. J. Medina. 1991. 74(2) Insecticide Residues in Selected Well Waters in Calamba and Calauan, Laguna. Metro Manila Authority. 1994. Report on Carmona and San Mateo Sanitary Landfill. Metropolitan Waterworks and Sewerage System. 1993. Wastewater Management for Metro Manila. _ 1994. Manila Second Sewerage Project, Project Executive Summary. Mink, Stephen D. 1993. Poverty, Population, and the Environment. World Bank Discussion Papers 189. Washington D. C. Morin, M., M. J. Kaiserman and K. Leahy. 1992. Regional Mortality Attributable to Tobacco Use in Canada, 1989. Chronic Diseases in Canada 13(4): 64-67. National Electrification Administration. 1994. Status of Energization, by Region. National Poison Control Center. 1993. University of the Philippines-Philippine General Hospital. National Statistics Office. 1990a. Census of Population and Housing Report No. 3: Socio- Economic and Demographic Characteristic for 1990. _ 1990b. Census of Population and Housing Report No. 4: Housing Statistics. _ 1994a. Family Income and Expenditures Survey for 1991. _ 1994b. Annual Survey of Establishment: Volume III for 1990. Needleman, H. L. and D. Bellinger. 1991. The Health Effects of Low Level Exposure to Lead. Annual Review of Public Health 12: 111-140. Occupational Safety and Health Center. 1993. Abstracts/Summaries of Research, Studies and Surveys from 1988 to 1992.  - 73 - Organisation for Economic Co-operation and Development (OECD). 1991. The State of the Environment.' Paris. Pingali, P., C. Marquez, et. al. 1994. Pesticides and Philippine Rice Farmer Health: A Medical and Economic Analysis: 76(587-592) American J. Agr. Econ. Roces, Ma. C., R. Rayray, et. al. 1990. Water-borne Diseases in Boracay, Malay, Aldan. Rodriguez, I., P. Ibarra,. et. al. 1994. Microbiology of Meat from Los Bafios Public Markets. The Philippine Agriculturist. 77(2) 261-269. Rola, A. , and P. Pingali. 1993. Pesticides, Rice Productivity, and Farmers' Health (an economic assessment). International Rice Research Institute and World Resources Institute. Rola, A. 1989. Socio-Economic Implications of Pesticide Use in Philippine Agriculture. Center for Policy and Development Studies, UP at Los Bafios. Rola, A., A. Chupungco, et.al. 1992. Pesticide and Pest Management in the Philippines: A Policy Perspective, Draft Report for International Development Research Center, University of the Philippines at Los Banos, Laguna, Philippines. Sarol, J., M. Mella, and J. Baltazar. 1994. Epidemiologic Trends of Selected Diseases in the Philippines ( 1965-1990). Draft Report. Statistics Canada. 1994. Human Activity and the Environment 1994. Cat. No. ll-509E, Ottawa. Stewart, R. D., E. D. Baretta, L. R. Platte et .al. 1974. Carboxyhemoglobin Levels m American Blood Donors. JAMA 229(9), 1187-1195. Subida, R., and E. Torres. 1991. Epidemiology of Chronic Respiratory Symptoms and Illnesses Among Jeepney Drivers, Air-conditioned Bus Drivers and Commuters in Metro Manila World Health Organization/College of Public Health University of the Philippines, Manila. _ 1994. Impact of Vehicular Emissions on Vulnerable Populations in Metro Manila. World Health Organization/College of Public Health University of the Philippines, Manila. Suskind, R. M., ed. 1977. Malnutrition and the Immune Response. Kroc Foundation Series, Vol. 7. New York: Raven Press. Tejada, A. 1994. Residues of Pesticides in Foods and the Environment as a Consequence of Crop Protection. Professional Chair Lecture, Metro Manila Commission, May 25, 1994. 1995 . . Personal communication with Elma Torres.  - 74- Tejada, A., C. Bajet, et. al. 1993. Toxicity and Toxicity Indices of Pesticides to Some Fauna of the Lowland Rice-Fish Ecosystem. The Philippine Agriculturist, 76(4), 372-382. Tejada, A., S. Calumpang and N. Gambalan. 1990. Effect of Processing on Residues of Chlorpyrifos in Stored Corn and Rice. The Philippine Agriculturist 73(4). Tejada, A., L. Varca, et. al. 1992. Fate and Residues of Pesticides in Paddy Rice Production in the Philippines. Proceedings, first Meeting on Environmental Toxicology in Southeast Asia. ed. Widiananko, B. et. al., 1994. Torres, E. 1989. Environmental Levels of Mercury in Small-scale Gold Processing Areas in Davao Del Norte. German Agency for Technical Cooperation/College of Public Health- University of the Philippines. Manila. Torres, E., and R. Subida. 1992. Environmental Health Surveillance in the Surrounding Communities of an Industrial Estate in Leyte, Philippines. German Agency for Technical Cooperation/ College of Public Health University of the Philippines. Manila. _ 1993. Urban Air Quality and Health for the URBAIR Manila Project. Torres, E., R. Subida, and L. Rabuco. 1991. Health Profile of Child Scavengers in Smokey Mountain, Balut, Tondo, Manila. German Agency for Technical Cooperation/International Labor Organization/ College of Public Health University of the Philippines. Manila. Villard, S. 1995. Assessment of Environmental and Public Health Laboratories in the Philippines. Consultant's Report. World Bank. Wallace, N.D. and G. L. Davis. 1974. Smoking and Carboxyhemoglobin in the St. Louis Metropolitan Population: Theoretical and Empirical Considerations. Archives of Environmental Health 29:136-142. • Warburton, H., F. Palis, and P. Pingali. 1994. Fanner Perceptions, Knowledge and Pesticide Use Practices from Impact of Pesticides on Fanner Health and the Rice Environment. Pingali, P. and P. Roger eds. (Draft). Wasserman, G.A., J.·H. Graziano, P. Factor-Litvak, D. Popovac, et al. 1994. Consequences of Lead Exposure and Iron Supplementation on Childhood Development at Age 4 Years. Neurotoxicology and Teratology 16(3): 233-240. World Bank. 1993. World Development Report 1993. Investing in Health. New York: Oxford University Press. 1996. Contry Assistance Strategy of the World Bank for the Republic of the Philippines. Washington D. C. World Health Organization. 1993. Draft WHO Global Strategy for Health and Environment. Document A46. l 1. Geneva.  - 75 - . 1994. Environmental Health Action Plan for Europe. Regional Office for Europe. Copenhagen. _ . 1995. Developµient of National Environmental Health Action Plans in Europe. Regional Office for Europe. Copenhagen. Zacarias, N. 1990. Outbreak ofEnteric Water-borne Disease in Bagong Buhat, Sapang Palay, Bulacan.