ࡱ> z|wxy  bjbj 2xxK   aa   4AAAhAAt^(^A`A`A`A`A`A`A/DF`A `AadHuA...  ^A.^A..~>|f?pA'>JAA0A ?ZcGz,ncG$f?cG f?.`A`A.AcG ':  Globalization, Distance and Disease: Spatial Health Disparities in Rural India Anirudh Krishna Professor of Public Policy and Political Science Duke University 212 Sanford School Durham, NC 27708-0245, USA (919) 613-7337 HYPERLINK "mailto:ak30@duke.edu"ak30@duke.edu and Kripa Ananthpur Assistant Professor Madras Institute of Development Studies 79, 2nd Main Road,Gandhinagar, Adyar Chennai 600020, India +91 44 2441 1574(Ext 329) kripa@mids.ac.in Abstract More than 50 percent of the Indian population lives in villages that are located more than five kilometers from the nearest town. This half of India is more likely to experience illnesses of different kinds and simultaneously less likely to get qualified medical treatment. The incidence of premature deaths, infant and child mortality, and malnutrition are all significantly higher within villages located further from towns. In consequence, such villagers are more susceptible than others to being overcome by the medical poverty trap. Poverty has increased within villages located more than five kilometers from towns, even as the national economy was surging ahead. Globalization privileges cities, disadvantaging locations at greater distances from towns. Public policy is required to compensate. Efforts to limit spatial inequalities must take precedence in future health policies. Keywords: spatial health disparities, globalization, rural India, distance from town Introduction: Investigating Spatial Inequality Spatial inequalities have widened during the period of post-liberalization economic growth in India. Globalization produces effects that privilege cities. Prior analyses have shown how spatial inequalities in India have become more pronounced in relation to per capita incomes and household assets. Cities, together with a small group of villages located close to cities, have acquired greater economic potential, moving further ahead. Villages located at greater distances from towns have fallen further behind. This article examines spatial inequalities in the realm of health. It is not clear that disparities in income will automatically find reflection in similar inequalities in health. At least insofar as government-run medical services and incentives are concerned, the effects of public policy should be to minimize, rather than reinforce, inequalities of different kinds; that is, after all, a guiding objective of public provision. And yet, as the evidence advanced below demonstrates, as you go deeper into rural areas, health outcomes become progressively worse. Simultaneously, qualified care becomes harder to access. A vast majority of Indians continues to live in rural areas, with this share falling only marginally, from 71 to 69 percent, over the decade prior to the Indian census of 2011. It would be foolhardy to expect that this share will fall to Western proportions at any time within the foreseeable future. Policies intended to serve rural India will be required for a long time and such policies need to be designed bearing in mind the growing importance of different degrees of rural-ness. Villages in India can be segmented according to their distance from the nearest town: 22 percent of the rural population lives within 5 kilometers from the nearest town, constituting an inner belt of villagers; 28 percent are situated between 5 and 10 kilometers of a town; while the remaining 50 percent of the rural population lives more than 10 kilometers from the nearest town. Thus, a total of 78 percent of the rural population amounting to more than 50 percent of all Indians lives in settlements that are located 5 or more kilometers from the nearest town. The results presented below show that it is among this half of the Indian population that multiple health disparities are clearly visible. In general, the more rustic ones existence the further one lives from towns the greater are the odds of disease, malnourishment, weakness, and premature death. Section 2 reviews the general proposition concerning how spatial disparities within nations have arisen together with advancing globalization. Section 3 interrogates the available national data, uncovering evidence of significant spatial inequalities in relation to a variety of health outcomes. In an effort to understand better the processes giving rise to such inequalities and how these effects are experienced among different households, Section 4 probes primary data collected by the authors in one rural part of India. Section 5 concludes by bringing together supply-driven and demand-based explanations for rising spatial inequalities, offering suggestions for policy reform. 2. Globalization, Geography and Growth Alongside advancing globalization, economic opportunity has become concentrated within cities, especially larger ones. Sassen (2001: 3) notes how a combination of spatial dispersion and global integration has created a new strategic role for major cities [which] now function in four new ways: first, as highly concentrated command points in the organization of the economy; second, as key locations for finance and for specialized service firms, which have replaced manufacturing as the leading economic sectors; third, as sites of production; and fourth, as markets for the products and innovations produced. These changes in the functioning of cities have had a massive impact Cities concentrate control over vast resources. Another commentator has similarly noted how economic activity in the era of globalization has become concentrated within city-based clusters of highly specialized skills and knowledge, institutions, rivals, related businesses, and sophisticated customers in a particular nation or region. Proximity in geographic, cultural, and institutional terms allows special access, special relationships, better information, powerful incentives, and other advantages in productivity and productivity growth that are difficult to tap from a distance (Porter 2000: 32; emphasis added). The effects of living at a distance from a city or town are experienced in terms of differences in economic opportunity. While larger cities advance economically, remote rural communities lag behind. Remarking upon the spiky nature of current-day economic growth, Florida (2008: 19) notes how the tallest spikes the cities and regions [concentrated around cities] are growing ever higher, while the valleys mostly languish. Such spatial clustering of economic opportunity has become acute within many parts of the developing world. Chinas remarkable economic growth, for example, is a result of only a handful ofspiky centers such as Shanghai, Shenzhen, and Beijing, each of which is a world apart from its vast impoverished rural areas In 2006, average household incomes in urban China were two and a half times those in rural areas [where]17 percent of Chinas population lives on less than a dollar a day, almost half lives on less than two dollars a day The prospects for bridging these gaps are weak But all that pales in comparison with the growing pains felt by Indias poor. Indias growing economic spikes city regions such as Bangalore, Hyderabad, Mumbai, and parts of New Delhi are also pulling away from the rest of that crowded country (Florida 2008: 35-36). Analysts examining the rise of inequality in India have noted how income differentials are widening between urban areas (which still account for no more than 30 percent of the countrys population), and the vast rural countryside (Deaton and Dreze 2002; Dev and Ravi 2007; and Sen and Himanshu 2004). The biggest Indian towns have the largest concentrations of assets. In towns with populations of more than five million (home to six percent of the Indian population), 24 percent of all households possessed cars in 2005, 82 percent had color TVs, 64 percent had refrigerators, and 54 percent had mobile phones. The corresponding percentages in towns with fewer than 50,000 people were 7 percent, 51 percent, 26 percent, and 21 percent. In rural India, these percentages were lower still, respectively, 3 percent, 24 percent, 8 percent, and 7 percent less than half the corresponding proportions within the smallest towns. The potential for upward mobility is significantly implicated with geography. How well you do depends to a considerable extent upon where you happen to live. A close observer of these trends concludes that despite all the hype about the death of distance and the flat world, where you live matters more than ever (Moretti 2012). These observations are borne out by recent trends in India. No matter what ones level of education or training, earnings are higher if one lives within a large town compared to a small town and in a small town compared to a rural village. Individuals who have only a primary education earned up to 68 percent more by living in a metro city (one that has more than five million people) compared to a smaller town (with fewer than 500,000 people). Among people with college degrees, the corresponding income differential is smaller though still substantial: 38 percent (Shukla 2010). Spatial economic differences have intensified over time in India. During the period 1993-2005, for example, when Indias economy grew rapidly, the largest cities experienced the largest average income gains. Smaller towns also gained but not by as much. Beyond towns, the benefits from economic growth were radially dissipated: Inflation-adjusted per capita incomes grew in villages located within five kilometers of towns. But outside this inner circle of villages, inflation-adjusted per capita incomes have fallen, with the deepest reductions occurring in villages located at greater distances from towns - which had, to begin with, lower per capita incomes. To make matters worse, the poorest income groups within such, more remote, villages have suffered the largest cuts in purchasing power; evidence of widening income inequalities simultaneously along both spatial and socioeconomic dimensions (Krishna and Bajpai 2011). Concurrently, poverty has grown. In villages located between 5 and 10 kilometers from towns, the percentage of households below the official poverty line increased from 35.8 percent to 41.4 percent, a gain of 5.6 percentage points over this 12-year period, widely regarded as a period of unprecedented high-speed growth. In villages located more than 10 kilometers from towns, the increase in poverty was even larger: 6.2 percentage points. That half of the Indian population which lives more than 5 kilometers from the nearest town is thus faced with a grimmer set of prospects. In the upward direction its movement is restricted; for many, their already bad situations are becoming worse. One would hope and expect that in the realm of health care, at least, such disparities would be minimal or actively reduced. Bad health and medical expenses can bankrupt families. A slew of recent studies show how the largest numbers of people fall into poverty and remain poor on account of ill health and high health care costs (Krishna 2010; Whitehead, et al. 2001; Xu, et al. 2001). As many as 3.7 percent of the entire Indian population falls below the poverty line each year on account of unaffordable medical expenses, often incurring unbearable burdens of debt (EQUITAP 2005; Garg and Karan 2005). Spatial disparities in health, therefore, need to be carefully examined. If the burden of disease were higher and access to qualified care simultaneously lower among villagers located more than 5 kilometers from towns, then they would be cumulatively disadvantaged. Coupled with the lower chances that they have, compared to villages located closer to cities, of gaining higher incomes and accumulating assets, the existence of a greater danger of descent, of falling into poverty, would tend to make spatial inequalities wider still, difficult to surmount without sustained external assistance. Both parts of the analysis presented below help demonstrate that such, indeed, is the case. We look first at the national picture, focusing on the supply side of the explanation. Section 4 considers a micro-level view, helping understand better some demand-related aspects of disparate outcomes and health-seeking behaviors. 3. Spatial Health Disparities in National Context. Prior examinations of health disparities in India have identified a number of factors, significantly associated with diverse outcomes and behaviors among different population segments. Analysts have examined differences arising on account of caste and wealth, finding significantly poorer outcomes among scheduled castes (SCs) and scheduled tribes (STs) and between richer and poorer Indians (Balarajan, Selvaraj, and Subramanian 2011; Gaudin and Yazbeck 2006; Mohindra, Haddad, and Narayana 2006; Subramanian, et al. 2004). Differences between men and women have also been found to be salient, particularly when seen alongside socioeconomic inequalities (Iyer, Sen and George 2007; Iyer, Sen, and stlin 2008). Substantial regional differences across states of India have been uncovered (Pande and Yazbeck 2003); and gaps between rural and urban areas found to be persistent and large (Baru, et al. 2010; Duggal 2005). For reasons examined in the previous section, it is important additionally to examine differences arising within rural areas, particularly among villages located close to towns and others situated more remotely. Only one previous study has examined spatial differences of this kind. An examination of data collected in the early 1990s, during the initial phase of globalization-driven economic growth in India, found that inequality in health indicators is very high both infant and child mortality rates increase sharply [among villages located at greater distances from the nearest town] short-term morbidity also shows a positive relationship with distance (Kundu, Pradhan, and Subramanian 2002: 5042-3). In the 20 years since these data were collected, this stream of explanation has not been followed up. While scholars have investigated diverse aspects of the relationship between globalization and health care, finding both positive and negative features, the spatial dimension of health disparities has not attracted further examination within India. In other developing countries, researchers have looked at the effects of distance to nearest health facility, concluding variously how distance, so measured, does or does not correlate with diverse health outcomes and disparate care-seeking behaviors. A study conducted in the late-1980s in Ghana found, for example, that distance was an important deterring factor in seeking institutional health care; the cost of care was less important in comparison to distance (Lavy and Germain 1994). A study undertaken in rural parts of one state of Nigeria came to a similar conclusion (Awoyemi, Obayelu and Opaluwa 2011). Other studies have, however, arrived at the opposite conclusion, finding that distance makes either no or relatively little impact (Acharya and Cleland 2000; Kesterton, et al. 2010; and Moisi, et al. 2010). It is timely and important, therefore, to examine the recent evidence for India, considering whether and how in the phase of advancing globalization spatial health disparities have become larger or less significant. We present below the results from analyses undertaken using recent nationally-representative data sets. District-Level Household and Facility Surveys (DLHS) were launched by the Government of India in 1996-1997 to provide important indicators on maternal and child health. These surveys were conducted by the Mumbai-based Indian Institute for Population Sciences (IIPS) and published by the Ministry of Health and Family Welfare of the Government of India. The second such survey (DLHS-2) was conducted in 2002-2004. A total of 620,000 households were interviewed from 593 districts and 26 states in India using a systematic, multi-stage stratified sampling design. DLHS-3, the third in this series of surveys, was conducted from 2007 to 2008. It provides information related to 720,320 households from 28 States and 6 Union Territories of India. A total of 78 percent of the surveyed households (559, 663 households in all) lived in rural areas, and we focused upon this part of the DLHS-3 sample, also consulting data from DLHS-2. We looked, first, at patterns of institutional delivery in national context, paralleling the micro-level examinations presented in the next section. Table 1 presents these figures at the national level. Table 1 about here - These results show that villages located at greater distances from towns have consistently lower proportions of institutional deliveries. Among inhabitants of villages located within two kilometers of towns, nearly 45 percent of all deliveries were conducted within institutions, but in the group of villages located more than 10 kilometers from towns, a much smaller proportion (33 percent) of deliveries were conducted within institutions of any kind. The proportion of institutional deliveries does not, however, provide the only indication that villages located at greater distances have health behaviors and outcomes of different kinds. Distance to town matters across a broad spectrum. People in villages located further from towns are significantly more likely to have cholera, malaria, and other diseases. Health problems during pregnancy are more frequent in such villages, resulting in significantly higher infant and child mortality. More children below the age of 5 years are severely or moderately malnourished within these more remote villages. Table 2 reports a sample of these results, reporting statistically significant differences between villages located within 5 kilometers and others located further away from towns. The 5-kilometer dividing line, separating inner-belt villages from more distant ones, remains salient across analyses considering a variety of outcomes, not only in the realm of health but as well in relation to per capita incomes, asset holdings, and educational trends, as prior analyses, discussed above, have also consistently found. Table 2 about here - The 50 percent of India that lives beyond a radius of 5 kilometers from the nearest town faces much greater odds of disease, malnourishment, weakness, and premature death. Health problems experienced during pregnancy occur in greater frequency. Mortality and morbidity among mothers and newborns are higher in further-away villages. Among adults as well, the likelihood of illnesses is higher within villages located at greater distances from towns. Malnutrition, a problem across a large swathe of the Indian population (Chatterjee 2007; Deaton and Dreze 2009), is more prevalent within more distant villages. The body-mass index (BMI) of children below the age of 5 is lower; many more children are mildly to severely malnourished and these disparities have grown larger with the passage of time (Table 3). Table 3 about here - In 1993-94, the percentage of mildly-to-severely malnourished children was not widely different among villages at different distance ranges, being, in fact, somewhat higher in villages within 5 kilometers of towns (81 percent) and somewhat lower in villages at the greatest distances, more than 10 kilometers, from towns (78.4 percent). But over the ensuing 12-year period, this difference reversed itself and grew larger. The incidence of malnourishment fell in villages within 5 kilometers of towns, but in the more remote villages (more than 10 kilometers away) the incidence of malnourishment went up, becoming a little over 80 percent in 2004-05, which is the highest proportion among villages of the three distance-related groups. How does one explain these reinforcing distance-related disparities? It seems likely that both supply-related and demand-based factors are to account. We will look at aspects of health-seeking behaviors in the next section. Meanwhile, let us look at some recent trends in the supply of health care, examining particularly the locations of hospitals, dispensaries, and clinics in both the public and private sectors, each of which has grown at a rapid clip particularly in rural locations closer to cities. With the advent of globalization, towns have acquired greater importance, as discussed above, and governments, national and state, have followed where markets have led, preferentially locating public infrastructure, including medical facilities, within urban and close-to-urban rural locations. Our analysis of data from the Indian census of 2001 showed that while more than two-thirds of all villages within five km of towns have been provided with paved roads, fewer than half of villages beyond 20 km from towns were similarly endowed through public provision. The corresponding proportions for electric power supply are 85 percent and 64 percent. A similar position obtains with respect to the location of health-care facilities. Table 4 shows how all types of medical facilities, including both more sophisticated and less sophisticated ones, are more likely to be situated in villages located within 5 kilometers of a town. Table 4 about here - Public as well as private provision has been disproportionately located within an inner belt of villages. While villages located within 5 kilometers of towns account collectively for less than one-fourth of the rural population, they nearly three times as likely compared to villagers further away to have a government or a private hospital and more than twice as likely to have a PHC or Block PHC. In every respect, despite their much smaller share of the rural population, inner-belt villages are far better endowed with medical infrastructure. Hospitals, with their sophisticated equipment and teams of specialists, can reasonably be expected to be centralized, being better located at centers of concentrated population, within or close by larger towns. However, the rationale behind locating even the simpler facilities public clinics and government dispensaries more abundantly in villages inside the 5-kilometer inner circle seems both inexplicable and counterproductive. However, the chances of finding a Block PHC or government dispensary are all higher in villages closer to towns, and lower in villages beyond the 5-kilometer inner circle. Partly as a result of these supply-side factors, people of further-away villages, who are more often sick, are less able to obtain treatment close to hand and because their illnesses remain untreated for longer experience worse health outcomes compared to inhabitants of villages nearer to towns. The prospects of falling into poverty are consequently larger within more remote villages; one reason why poverty has grown within more-distant villages, as we saw above for the period between 1993 and 2005. 4. Examining care-seeking behaviors and health outcomes in rural Karnataka Supply-related factors are not all that matter. Peoples health-seeking behaviors factors influencing their demand for qualified health care can also make a difference, helping produce spatial disparities of different kinds. A small-scale investigation undertaken in two parts of rural Karnataka helped understand better the effects of distance from town when seen in conjunction with several other factors including caste, wealth, gender and education, which, according to previous analyses, influence peoples demands for qualified health care. Instead of ranging widely across all types of health outcomes, we selected to focus in these investigations upon a specific subset births and deliveries examining the nature of factors associated with peoples decisions to select institutional deliveries in preference to deliveries at home. Two districts, Gulbarga and Raichur, were selected for this part of the inquiry, which count among the more backward districts in Karnataka in relation to various socio-economic indicators. Both districts are also included within the coverage of the Janani Suraksha Yojana (JSY), part of the Indian governments National Rural Health Mission, launched in 2005, which is intended to boost institutional deliveries through providing cash incentives to delivering mothers (Campbell and Graham 2006; Lim, et al. 2010). As we will see below, financial incentives do not suffice to remove the handicaps that are associated with living at a greater distance. A combination of qualitative and quantitative methods was utilized for this study. Twelve villages were selected in the following manner. Villages of the two selected districts were initially categorized into four distance categories: within 2 kilometers of nearest town; from 2 to 5 kilometers; from 5 to 10 kilometers; and beyond 10 kilometers. Random sampling helped select 12 villages, four from each of these distance categories. An average of 306 households live within these villages, and average household size is just below 5.5, making for average village population of 1,700. A census of all households was carried out in each village. Basic information, related to caste group and relative wealth, especially house type kaccha (mud) v. pukka (brick) and land holdings was collected at the same time. Households were categorized on the basis of their land holding into four groups: landless, small and marginal, middle, and large. Random sampling was employed to pick 20 percent of households from within each of these four categories. A total of 772 households selected in this manner were interviewed using a pretested questionnaire. Particularly detailed information was elicited about all pregnancies and births occurring over the ten-year period preceding the survey. A subset of 47 women (selected from among the list of pregnant women maintained by local health workers) was selected for further detailed process documentation. Field investigators visited each of these women on a monthly basis, discussing their ante-natal care (ANC) practices and their interactions with health care providers of different kinds. Discussions were also held with focus groups, which included village elders, grama panchayat (elected village council) members, youth leaders, members of women self-help groups, village health professionals, and NGO workers. Separate interviews with held with doctors and nurses at block, district, and state capitals and with village-level health workers. These results show that during the ten-year period preceding our survey, the 772 households selected for interviews had a total of 2,261 deliveries, of which 45 percent occurred within institutional settings 18 percent in a government hospital, 3 percent at a Primary Health Centre (PHC), and 24 percent in private hospitals. The remaining 55 percent of all deliveries occurred in non-institutional settings: 24 percent at the home of the delivering mother, another 30 percent at the home of her parents, and a further one percent on the way to a hospital. Untrained dais (traditional birth attendants) supervised the majority of these non-institutional deliveries, with relatives or friends attending the remaining part. These proportions of institutional and non-institutional deliveries are not significantly different from those observed above for all of rural India (Table 1). In order to examine household characteristics associated with institutional deliveries, we created a household-level variable, dividing households into the following mutually-exclusive categories: Never Institutional: Households in which all deliveries over the past 10 years took place outside institutions. Sometimes Institutional: Household that had at least one institutional delivery and at least one non-institutional delivery over this ten-year period. Always Institutional: Household in which all deliveries over the past 10 years took place within institutional settings. Table 5 gives the breakdown of households by these categories. Notice how roughly equal proportions of households fall within the never institutional, sometimes institutional, and always institutional groups. Table 5 about here - Using these categories, we examined the effects upon demand for qualified health care of diverse household- and individual-level factors that prior studies have identified. We began our analysis by looking at diverse indicators of household wealth. In line with what prior analyses have also found, higher wealth is positively associated with decisions to have institutional deliveries. Richer households (those who live in pukka homes) are nearly twice as likely compared to poorer ones (with kaccha homes) 60 percent v. 32 percent to always have institutional deliveries. A similar result obtained when we looked instead at land or asset holdings as alternative indicators of wealth. Interestingly, households who have added to their asset holdings in the ten years preceding the survey were also more likely to have institutional deliveries compared to households whose assets holdings have declined or remained the same, suggesting that as people become richer the proportion of institutional deliveries should rise. However, the relationship between wealth and care-seeking is far from determinate. Even among the richest households in these villages those who have the largest land holdings, better houses, televisions, and in a few cases, also motor cars as many as 31 percent of households are in the never institutional category. Other factors need to be considered; for instance, higher education should go together with higher demand for institutional deliveries. Table 6 breaks down these categories always, sometimes, and never institutional in relation to the highest level of education in the household. Table 6 about here - Notice how the proportion of never institutional households is as high as 48 percent among households where the highest level of education is 1-4 years (and it is nearly 44 percent among households whose highest education is 5-7 years), but this share falls to 36 percent among households whose highest education level is 8-10 years (and further to just over 21 percent in households who have pre-university (PUC) qualification or a university degree). Simultaneously, the proportion of always institutional households rises from 15 percent to 53 percent. Education is rising rapidly in Indian villages, and the effects of rising education should be felt in terms of increasing institutional deliveries over time. But even within households where the highest education level is a university degree, as many as 21 percent fall within the never institutional category, and another 25 percent fall within the sometimes institutional category indicating that, just as in the case of rising wealth, the effects of rising education are not assured or predictable in all cases. Something else matters in addition to education and wealth. We examine below the effects of caste and religion, looking later at several characteristics considered together. A variety of caste and religious groups live within these villages, some with relatively few members. We looked at the major caste categories, each with more than 20 households in all, considering especially, Scheduled Castes (SCs), Scheduled Tribes (STs), and Muslims. We also looked at two specific caste groups, Lingayat and Kuruba, who constitute substantial shares of village populations and have been dominant, numerically and in terms of economic and political power, within these villages. Examining dominant castes specifically is especially important, as advised by Srinivas (1987). Table 7 provides these results. Table 7 about here - All caste and religious groups exhibit a spread of care-seeking behaviors. Every group has a significant proportion of always institutional households, but every group also has a fair proportion of never institutional households. The proportion of always institutional households is lowest among STs, followed by Muslims, in line with what investigations undertaken in other parts of India have earlier revealed (e.g., Subramanian, et al. 2004; Subramanian, Smith, and Subramanyam 2006). As remarked above, the pattern of institutional deliveries for SC households does not differ a great deal from the average for all households. Further, the fact that the proportion of sometimes institutional households is high among both Muslims and STs in fact, higher among STs than the average for the population suggests that caste- and religion-specific norms are not determinative but pliable. Experiences of discrimination can reduce the motivation for institutional deliveries; service providers have been known to treat poor villagers, SCs, Muslims, and other disadvantaged groups with disfavor, presenting attitudes of disdain, and some times, outright rudeness (Malhotra and Do 2012). Members of the study team heard, and on occasion, witnessed, examples of such discriminatory behaviors. Still, our focus group interviews showed how a belief has taken root that institutional deliveries are safer and better, and a rising proportion of households prefer to take this road. Younger households across all social groups have opted for institutional deliveries in much higher proportions compared to their older counterparts. Compared to deliveries occurring between 2001 and 2004, a greater proportion of deliveries occurring between 2007 and 2010 took place within institutional settings and this rise in proportion was experienced by all social groups. Despite this overall rising trend in demand for qualified care, physical access continues to represent a problem. Care-seeking behaviors within villages located at greater distances from towns are substantially different from those in villages located closer by. On average, 31.7 percent of all households are always institutional, i.e., all of their deliveries in the past ten years have occurred within institutional settings, but in Delari Village, located 4 kilometers from the nearest town, this proportion is as high as 60 percent, while at the other extreme, in Rakami Village (15 kilometers from the nearest town), it is only 17 percent. In general, villages located outside the 5-kilometer inner circle have uniformly lower shares of institutional deliveries. Figure 1 presents a visual depiction of this relationship. Figure 1 about here - Villages within the 5-kilometer inner circle have variously high and low percentages of always-institutional households, on average performing better than villages further away. But villages located beyond the 5-kilometer radius have consistently lower-than-average shares of institutional deliveries. It could be that villages located at greater distances have higher shares of STs and lower average incomes, thus making the effects of distance nothing more than an artifact of other and more basic effects. However, even when seen alongside other factors, including caste, religion, family education, and relative wealth, distance to town continues to make a significant difference to peoples health-seeking behaviors. Logistic regression was carried out (coding always institutional households as one and all other households as zero). Separate analyses considering never institutional as the response category did not produce qualitatively different results in terms of which independent variables gained significance. Table 8 reports these results. Table 8 about here - This analysis shows that several factors influence the demand for qualified medical care, in this case, institutional delivery. In multiple specifications of the regression model, however, distance continued to exert a significant influence. Among household-level factors, wealth and education are significant for this analysis. Two particular social groups (Lingayat and ST) also gain significance. The coefficient for SC is positive, but it is not statistically significant. Similarly, the coefficient for Muslim is negative but not statistically significant. Two separate distance variables are significant. Distance to nearest town is consistently significant. Distance to paved road makes a separate difference. In substantive terms, distance to town outweighs distance to paved road. While distance to town varies from two to 31 kilometers, distance to paved road varies within a much narrower range. Greater distance from town is reflected at the household level in terms both of higher transportation costs and sheer physical difficulty of gaining access. Distance to paved road compounds the difficulty of gaining access. Qualitative investigations uncovered the main reasons for why distance exerts this nature of influence. We asked each of more than 700 respondents about the main barriers they have faced while seeking institutional health care. The largest number, 41 percent, mentioned lack of transport as the most important constraint. A further 21 percent and 26 percent, respectively, mentioned distance and time taken, making for a total of 88 percent for whom problems of physical access figured among the top three constraints. In comparison, many fewer households, 43 percent in all, mentioned costs or quality of service among their top three constraints, showing how financial incentives can go only a part of the way toward altering care-seeking behaviors and improving health outcomes. The introduction of cash incentives through the governments Janani Suraksha Yojana (JSY) program has reduced the financial disincentive to institutional deliveries. However, our interviews revealed that considerable obstacles still remain. Although JSY incentives have been available in these villages since 2005, no more than eight percent of all deliveries had been assisted either by this program or by any other government program in force. Admittedly, our inquiries spanned a longer period of time, ranging over the ten years between 2000 and 2010. A little fewer than half of all 2,261 deliveries occurring over this ten-year period took place after the coming into force of JSY. Still, eight percent is a very small proportion and this proportion is smaller still within villages located outside the 5-kilometer inner circle. In order to probe this issue further, we looked at the cases of those 176 households that had at least one delivery both before and after the introduction of JSY. How many of these households increased their institutional delivery percentage after the introduction of JSY, how many decreased their institutional delivery percentage after the introduction of JSY? We found that in 64 percent of these households the introduction of JSY did not change the households behavior; in another 26 percent of cases, a household that elected non-institutional delivery before JSY chose, instead, to have an institutional delivery after 2005; but in the remaining ten percent of cases, the opposite behavior pattern was discerned: households that had one or more institutional deliveries before 2005 elected to have one or more non-institutional deliveries after JSY incentives were put in place. Households in villages located further from towns were considerably were less likely to alter their care-seeking behaviors. Cash incentives cannot alone resolve the problem of distance. Liabilities associated with physical access on account of distance to town continue to make a separate and substantial difference. Further, these investigations revealed how the birth of a child is not the only occasion when residents of more remote villages suffer handicaps while seeking and obtaining professional health care. Instances of diseases left untreated were more often cited in villages located further away. Morbidity and mortality are higher across the board within these further-away villages among newborns, infants, children, and adults as we have saw above while looking at national results. 5. Conclusion: Distance and Disease Contrary to what might once have been true, the rustic life is not necessarily healthier. In fact, the more rustic ones existence the further one lives from towns the greater are the odds of disease and malnourishment, and the smaller is the incidence of institutional treatment. In situations where physical access remains a challenge, many families either select to forego treatment, or they incur huge costs in order to come to the hospital, a decision that is often made at a very late stage, requiring specialized medical attention, which further raises costs. Ironically, it is not just medical emergencies; even ordinary events, such the birth of a child, can drive rural families into a medical poverty trap (Ensor and Cooper 2004; Mavalankar, et al. 2009; Whitehead, et al. 2001). On the one hand, people who live in villages more than 5 kilometers from the nearest town have seen their real per capita incomes drop. On the other hand, their likelihood of having a health episode (with expenses to boot) is higher. Thus, even as their prospects of moving up are limited, the risks of moving down, falling into poverty, are larger. Especially in a context, such as rural India, where out-of-pocket costs account for the vast bulk of health care expenditure, such overlapping disparities can become progressively hard to surmount, resulting in impoverishment and immiserization on a large scale. The greater the distance to town, the higher are these risks. Not a great deal is being done at present to address spatial disabilities. A few noteworthy examples exist of interventions by NGOs and private foundations, but these are mainly spots of light in an otherwise dark landscape, much like a map of India at night. Unless corrective measures are put in place through public policies, spatial disparities will likely become worse in years to come. Why have such spatial differences arisen, and how can they be rectified? We looked above at a range of factors. No simple or mono-causal explanation will suffice. Several elements must be considered at the same time, including factors related, respectively, to the demand for and supply of better health care. On the supply side, the provision of health care facilities has been concentrated disproportionately within Indias cities and within villages located less than 5 kilometers distant from towns (Table 4). Residents of more distant villages have to cross higher hurdles just in order to have their loved ones attended to by a qualified doctor or nurse. A second supply-side problem, which quantitative examinations cannot easily uncover, but which our qualitative investigations in Karnataka helped illustrate, relates to the scant supervision provided to the health personnel who are deployed by the government at the village level. A vast army of such barefoot or basic health practitioners, including ANMs and ASHAs, paid for by the Indian taxpayer, has been assembled by the Indian government, and put in place in villages, including the furthest-out ones. The quality of services provided by these individuals is, however, highly variable. In some villages, ASHAs and ANMs are highly-valued resources, being regularly available and serving diligently. In other villages, these employees are hard to find, discriminatory, careless or callous about their work, and otherwise negligent. Villagers themselves have little control over the health staffs deployed to serve them. Beyond complaining to higher officials, most often located in towns, villagers can do little or nothing to check absenteeism, enforce accountability, or insist upon norms of professional behavior. Greater distance to towns reduces the frequency of complaint, lowering one motivation for providing a higher quality of service. Focus group interviews revealed how instances of uncaring behavior and absenteeism on the part of health staff occurred more often in villages further from towns; out-of-sight and often out of mind for higher officials. These factors related to the supply of health care are reinforced by other factors, which in different ways influence villagers care-seeking behaviors. We saw how several factors, related to household wealth, caste, religion, gender, and education, are implicated with higher and lower demands for institutional health care. Distance matters both in addition to these other demand-related factors and as well in interaction with them. The effects of distance higher cost, more time taken, more hassle acting as deterrents, push downward the demand for qualified health care in further villages. Further, some other demand-related factors are themselves significantly related with distance. For instance, the proportion of adults with high school (or higher) education falls progressively the further one goes from towns, ranging from 48 percent in villages within two kilometers of towns to 35 percent in villages more than ten kilometers from towns. Average years of schooling are lowest in more distant rural locations; reading, writing and computation ability also become progressively lower with increasing distance. The relative impacts made by different influences acting together and in combination will need to be investigated more closely in different parts of India. Understanding the different ways in which distance influences behaviors and outcomes will help fashion better remedies against growing spatial disparities of multiple kinds. Poverty in India cannot be reduced substantially without first making affordable and accessible quality health care available to all (Gupta and Mitra 2004; Krishna 2010). Removing spatial disparities is an essential part of getting health care right. In addition to taking account of disparities arising on account of gender, caste, and wealth, efforts aimed at improving health equity in rural India must compensate for the effects of distance to town. Some recent initiatives have helped in this regard, particularly the introduction in the year 2008 of the 108-ambulance service, a public-private partnership. In Karnataka, for example, each 108-ambulance service is expected to cover a population of 100,000 over distances of no more than 30 kilometers. It seems important to note, however, that despite the introduction of the 108-ambulance service only a small proportion of all women who obtained institutional deliveries were brought to the hospital or clinic in an ambulance. The majority utilized hired means of transportation, including, most often, tractors, auto rickshaws, and trucks, and less frequently, buses and cars. Vastly expanding the network of free or low-cost ambulance services is essential to reduce the human misery and often ruinous costs associated with greater distance and its consequence, inability or reluctance to seek and obtain institutional health care. Other measures e.g., mobile health clinics can also help. Enhancing accountability and reducing absenteeism among government health personnel deployed in PHCs and at the village level is another important step. Lapses in governance as much as (and probably more than) shortages of money represent the principal obstacles that remain to be overcome. *** Acknowledgements Data collection exercises in Karnataka were partly supported by a grant (number OW2: 205) received from the International Initiative for Impact Evaluation (3ie).We thank, without in any way implicating, 3ie, and its Director, Howard White, both for the grant and for comments and advice. We also thank Gregory Schober for help with this part of the data analysis and Tulsi Patel for assistance with the DLHS data. The usual disclaimers apply. References Acharya, L.B. and Cleland, J. (2000). Maternal and Child Health Services in Rural Nepal: Does Access or Quality Matter More? Health Policy and Planning, 15 (2), 223-9. Arora, R, Neogi, A., and Misra, M. (2011). Innovative Ways to Meet Health Challenges. Available at HYPERLINK "http://iuc2011.in/sites/default/files/white-papers/Innovative%20Ways%20to%20Meet%20Health%20Challenges%20of%20Urban%20India_Radhika%20Arora_Sourav%20Neogi_Madhavi%20Misra.pdf"http://iuc2011.in/sites/default/files/white-papers/Innovative%20Ways%20to%20Meet%20Health%20Challenges%20of%20Urban%20India_Radhika%20Arora_Sourav%20Neogi_Madhavi%20Misra.pdf. Awoyemi, T.T., Obayelu, O.A. and Opaluwa, H.I. (2011). Effect of Distance on Utilization of Health Care Services in Rural Kogi State, Nigeria. Journal of Human Ecology, 35 (1), 1-9. Balarajan, Y., Selvaraj, S. and Subramanian, S. V. (2011). Health Care and Equity in India. The Lancet, 377, 505-15. Baru, R., Acharya, A., Acharya, S., Shivakumar, A.K. and Nagaraj, K. (2010). Inequities in Access to Health Services in India: Caste, Class and Region. Economic and Political Weekly, September 18, 49-58. Campbell, O. and Graham, W.J. (2006). Strategies for Reducing Maternal Mortality: Getting on with What Works. The Lancet, October 7, 1284-99. Chatterjee, P. (2007). Child Malnutrition Rises in India despite Economic Boom. The Lancet, April, 1417-8. Corbridge, S., Williams, G., Srivastava, M. and Veron, R. (2005). Seeing the State: Governance and Governmentality in India. New York: Cambridge University Press. Deaton, A. and Dreze, J. (2009). Food and Nutrition in India, Facts and Interpretation. Economic and Political Weekly, February 14, pp. 42-65. Dev, S.M. and Ravi, C. (2007). Poverty and Inequality: All-India and States, 1983-2005. Economic and Political Weekly, February 10, 509-21. Duggal, R. (2005), Public Expenditures, Investment and Financing under the Shadow of a Growing Private Sector, in L. Gangolli, R. Duggal, and A. Shukla, eds., Review of Health Care in India. Mumbai, Centre for Enquiry into Health and Allied Themes. Ensor, T. and Cooper, S. (2004). Overcoming Barriers to Health Service Access: Influencing the Demand Side. Health Policy and Planning, 19(2), 69-79. EQUITAP. (2005). Paying Out-of-Pocket for Health Care in Asia, Catastrophic and Poverty Impact. Equitap Project Working Paper No. 2. Available at HYPERLINK "http://www.equitap.org"www.equitap.org Fort, M., Mercer, M.A. and Gish, O., eds., (2004). Sickness and Wealth: The Corporate Assault on Global Health. Cambridge, MA: South End Press. Garg, C. and Karan, A.K. (2005). Health and Millennium Development Goal 1, Reducing Out-Of-Pocket Expenditures to Reduce Income Poverty Evidence from India. Equitap Project Working Paper No. 15. Available at www.equitap.org. Gaudin, S. and Yazbeck, A.S. (2006). Immunization in India 1993-1999, Wealth, Gender, and Regional Inequalities Revisited. Social Science and Medicine, 62, 694-706. Gupta, I. and Mitra, A. (2004). Economic Growth, Health and Poverty: An Exploratory Study for India. Development Policy Review, 22 (2), 193-206. Hazarika, I. (2010). Medical Tourism: Its Potential Impact on the Health Workforce and Health Systems in India. Health Policy and Planning, 25, 248-51. Iyer, A., Sen, G. and George, A. (2007). The Dynamics of Gender and Class in Access to Health Care, Evidence from Rural Karnataka, India. International Journal of Health Services, 37 (3), 537-54. Iyer, A, Sen, G., and stlin, P. (2008). The Intersections of Gender and Class in Health Status and Health Care. Global Public Health, 3 (S1), 13-24. Kesterton, A.J., Cleland, J., Sloggett, A. and Ronsmans, C. (2010). Institutional Delivery in Rural India: The Relative Importance of Accessibility and Economic Status. BMC Pregnancy and Childbirth, 10, 30. Krishna, A. (2010). One Illness Away, Why People Become Poor and How they Escape Poverty. Oxford, Oxford University Press. Krishna, A. and Bajpai, D. (2011). Lineal Spread and Radial Dissipation, Experiencing Growth in Rural India, 1993-2005. Economic and Political Weekly, 46, 44-51. Kruk, M.E. (2012). Globalization and Global Health Governance: Implications for Public Health. Global Public Health, 7 (S1), S54-62. Kundu, A., Pradhan, B.K. and Subramanian, A. (2002). Dichotomy or Continuum, Analysis of Impact of Urban Centres on Their Periphery. Economic and Political Weekly, 46, 5039-46. Labonte, R. and Schrecker, T. (2007). Globalization and Social Determinants of Health. Part 1: Globalization and Health, 3 (5): 1-10; Part 2, Globalization and Health, 3 (6): 1-17; Part 3, Globalization and Health, 3 (7): 1-15. Lavy, V., and Germain, J.M. (1994). Quality and Cost in Health Care Choice in Developing Countries. Living Standards Measurement Study Working Paper No. 105. Washington, DC: World Bank. Lim, S. S., Dandona, L., Hoisington, J.A., James, S.L., Hogan, M.C. and Gakidou, E. (2010.) Indias Janani Suraksha Yojana, a Conditional Cash Transfer Programme to increase Births in Health Facilities: An Impact Evaluation. The Lancet, June 5, 2009-23. Malhotra, C. and Do, Y.K. (2012). Socio-Economic Disparities in Health System Responsiveness in India. Health Policy and Planning, forthcoming. Manor, J. (2010). Local Governance, in Jayal, N.G. and Mehta, P.B. eds., Oxford Companion to Politics in India, pp. 61-79. New Delhi: Oxford University Press. Mavalankar, D., Singh, A., Sureshchandra, Patel, R., Desai, A., Singh, P.V. (2009). Saving Mothers and Newborns through an Innovative Partnership with Private Sector Obstetricians: Chiranjeevi Scheme of Gujarat, India. International Journal of Gynecology and Obstetrics, 107, 2716. Mohindra, K.S., Slim Haddad, and D Narayana. (2006). Womens health in a rural community in Kerala, India, do caste and socioeconomic position matter? Journal of Epidemiology and Community Health, 60,10201026. Mosi, J., Gatakaa, H., Noor, A., Williams, T., Bauni, E., Tsofa, B., Levine, O. and Scott, J. (2010). Geographic Access to Care is not a Determinant of Child Mortality in a Rural Kenyan Setting with High Health Facility Density. BMJ Public Health, 10, 142 Moretti, E. (2012). The New Geography of Jobs. New York: Houghton, Mifflin, Harcourt. Pande, R. and Yazbeck, A. (2003). Whats in a Country Average? Wealth, Gender, and Regional Inequalities in Immunization in India. Social Science and Medicine, 57, 2075-88. Sen, A. and Himanshu. (2004). Poverty and Inequality in India I and II. Economic and Political Weekly, September 18, pp. 4247-63, and September 25, pp. 4361-75. Srinivas, M.N. (1987). The Indian Village: Myth and Reality. In M.N. Srinivas, Dominant Caste and Other Essays, pp. 20-59. Delhi: Oxford University Press. HYPERLINK "http://beta.epw.in/haysearch/?q=caste%20health&selected_facets=author_exact:S%20V%20Subramanian,%20Shailen%20Nandy,%20Michelle%20Kelly,%20Dave%20Gordon,%20George%20Davey%20Smith"Subramanian, S.V., Nandy, S., Kelly, M., Gordon, D., and Smith, G.D. (2004). HYPERLINK "http://beta.epw.in/newsItem/comment/59336/"Health Behaviour in Context. Economic and Political Weekly, February 14, 684-93. Subramanian, S.V., Smith, G.D. and Subramanyam, M. (2006). Indigenous Health and Socioeconomic Status in India. PLoS Medicine, 3, 10. Whitehead, M., Dahlgren, G. and Evans, T. (2001). Equity and Health Sector Reforms, Can Low-Income Countries Escape the Medical Poverty Trap? The Lancet, September 8, 833-6. Xu, K, Evans, D., Kawabata, K., Zeramdini, R., Klavus, J., and Murray, C. (2003). Household Catastrophic Health Expenditure: A Multi-country Analysis. The Lancet, 12 July, pp. 111-117. Table 1: Distance to Town and Institutional Delivery: All-India Data Village distance to nearest townPlace where last delivery was conducted0-2 kilometers2-5 kilometers5-10 kilometersMore than 10 kilometersHospital14.6312.4411.5610.77Dispensary0.170.170.170.19UHC/UHPC/UFWC0.290.230.270.26CHC/Rural Hospital5.524.874.914.28PHC5.936.075.115.38Sub Center0.790.860.690.58Ayush Hospital/Clinic0.010.030.020.03NGO/Trust Clinic0.340.370.280.24Private Hospital/Clinic16.8917.0914.9611.52Private Ayush Hospital0.160.20.170.15All Institutional44.7342.3338.1433.4On way to hospital0.420.390.50.42At home50.6352.9356.7761.62At parents home3.854.024.244.2Work place0.050.060.050.07Other0.310.290.270.28All Non-Institutional55.2657.6961.8366.59 DLHS data Table 2: Worse Health Indicators in Further Villages of India Nature of problem Significant difference in odds (Villages > 5 kilometers from town compared to other villages)(A) During pregnancyConvulsions12% greaterVisual Disturbance16% greaterMalaria14% greaterHypertension22% greaterPaleness, giddiness, weakness8% greaterExcessive bleeding 6 weeks after pregnancy14% greaterWeak or no movement of fetus9% greaterExcessive bleeding during delivery13% greater(B) Among newbornsDiarrhea during last two weeks17% greaterPneumonia during last two weeks23% greaterChild died20% greater(C) Among general populationMalaria31% greaterCholera28% greaterOther communicable diseases31% greater DLHS data Table 3: Proportion of Mildly-to-Severely Malnourished Children 5 years and younger (BMI<18.4) (percent of all children) Distance to town1993-942004-05<5 kilometers81.074.15 to 10 kilometers81.577.2>10 kilometers78.480.2 NCAER data Table 4: Availability within villages of different health care facilities FACILITYPercentage of villages having this facilityWithin 5 kilometers of townsBeyond 5 kilometersGovernment hospital10.43.5Private hospital15.45.3AYUSH health facility18.210.5Primary Health Centre (PHC)23.211.8Block PHC13.46.5Government dispensary15.611.1Private clinic28.517.7 DLHS data Table 5: Household Categories and Institutional Deliveries in Karnataka Villages Category of householdNumber of householdsShare of totalNever Institutional26634%Sometimes Institutional26134%Always Institutional24532%Total772100% Table 6: Household Education and Institutional Deliveries Category of HouseholdHighest Level of Education in HouseholdPercent of totalNever InstitutionalSometimes InstitutionalAlways InstitutionalTotalIlliterate6%33%38%29%100%1-4 years13%48%37%15%100%5-7 years19%44%30%27%100%8-10 years32%36%33%31%100%Pre-university19%21%40%39%100%College degree10%22%25%53%100%Post-graduate<1%0%50%50%100% Table 7: Social Group and Institutional Deliveries Category of HouseholdSocial GroupShare in village populationNever InstitutionalSometimes InstitutionalAlways InstitutionalTotalKuruba10%38%33%29%100%Lingayat24%22%27%51%100%Muslim8%49%29%22%100%Scheduled Caste19%37%32%31%100%Scheduled Tribe19%44%44%12%100%Other20%32%37%31%100%Average34%34%32% Table 8: Factors associated with being an Always Institutional household (results of logistic regression) CoefficientStd. errorP>|z|Constant2.433*1.1380.033Household characteristicsKaccha house-0.939*0.2740.001Female-headed-0.2170.3080.480Household education (years)0.187*0.0750.013Lingayat0.619*0.2400.033Kuruba-0.0320.3270.921SC0.1030.2630.697ST-0.780*0.3290.018Muslim-0.5010.3790.186Village characteristicsDistance to town (km)-0.041*0.0170.016Distance to paved road (km)-0.099*0.0460.033Literacy rate-0.9521.6620.567Percent landless-0.8321.1510.470Population0.0000.0000.551N =768LR chi-sq. 108Prob>chi-sq.0.000Pseudo R-sq.0.22* indicates significance at 5% level or better FIGURE 1: Distance to nearest market town and share of households that always have institutional deliveries  NOTES      PAGE \* MERGEFORMAT 17  PAGE \* MERGEFORMAT 31  We use the terms town, city, and urban area interchangeably. In the empirical analysis that follows, we use the official definition of town employed by Indias government. See http://censusindia.gov.in/2011-prov-results/paper2/data_files/India2/1.%20Data%20Highlight.pdf  These data were derived from a nationally-representative sample survey the Indian Human Development Survey carried out by the National Council for Applied Economic Research (NCAER) in 2004-05.  A remarkably similar argument advanced in the specific context of contemporary urban India is provided by Chatterjee (2004: 142-47).  SCs are former untouchables, while STs are, loosely speaking, Indias indigenous people.  See, for instance, Fort, Mercer, and Gish (2004); Hazarika (2010); Labonte and Schrecker (2007); and Kruk (2012).  Out of a total of 2,343 recalled pregnancies, as many as 2,216 (94.6%) resulted in live births. Roughly equal percentages (2.7% each) resulted in still births or were aborted.  We coded all deliveries occurring at any of the following locations as institutional deliveries: government hospital or dispensary; UHC/UFWC; CHC; PHC; SHC; Ayush hospital or clinic; NGO/Trust hospital or clinic; and private hospital or clinic. The remaining deliveries, occurring at any of the following locations, were classified as non-institutional: at home, at parents home, at work, on the way to a hospital.  We separately analyzed these data considering the mothers education level and not the highest education level of any household member. But these results were not qualitatively different.  These findings from the case histories that we compiled are not presented here because of lack of space. A separate report, presenting these cases and the related findings, is available upon request.  Actual village names have been disguised to preserve confidentiality.  The results reported below were robust to alternative specifications of the regression model and diverse constructions of the social group variables. Tests for collinearity did not give reason for concern.  Zero-one dummy variables were constructed for each of these groups. High-caste Hindus serves as the comparison category. Apart from Hindus and Muslims no other religions are represented in any significant numbers.  For some instructive contemporary examples, see Arora, et al. (2011).  ANMs are auxiliary nurse midwives, and ASHAs are accredited social health activists, both deployed in rural areas by the governments of different Indian states.  Whether deepening decentralization will serve as a suitable remedy to these problems is itself a topic worthy of additional research. See, for instance, Corbridge, et al. (2005) and Manor (2010).  Author calculations from nationally-representative data collected by NCAER in 2004-05. &'QR     " 4 9 [ _ m n v   ɾɥ}yr}k}k} hh[ hhmh"JGh[hDh,9h] hJhmhhm56 hm56 hm5hmhm0JmHsH hmhmjhmhmUhmhmmHsHhmmHsHhm0J!5\^Jhmhm0J!5\^JhIhA0J!5\^J''QRb    ) = e Q R gdmgd:[$a$gdm ( ) : ? C Q     N O h l o u w  9 @ G P Q \ *tu,5FL٧ hl-jhDhXw hl-jhmhS hl-jh;(_hl-jh:[5 hXw5 h05 hm5 hm6hmhyh9@h]h~Ph"JGhDh[h,9 hhm:R KLz{stRSTU|}d e $gd;(_gd, UgdH cgd]o & Fgd:[gd:[gdmK #-!/068HJLMNZz{89  "HIJOil 5>HP`۶hSh, U6jhl-jhm0JUh,9h{hyhDh, UhShXw hl-jhmh0jhl-jhD0JUH`lqt    */3STU|}YZs}Uu;!>gd:[gdH cgd{gd;(_)**+*,********++)+2+7+=+S+X+Z+[+++++++++,#,,,+---l------,/;/ 0)0l0m0w0y00000&1'11112 22u2222222222ؼؼؼؼ hl-jh:+7h:+7h:+7h:+76h, Uhl-jhm6h~P hl-jhmhO"hm hxsh{ h9h{h{ hl-jh{G23 333G4Y4d4o4s4t4445555555666!6.6b6o6s6|666L788848688898:8m8n888888888999999;;n<v<w<}<<<䵨ݠhl-jhm6hl-jhm0JCJaJjhl-jhm0JUht9hthDhl-jhm5 hl-jhD hl-jhmh hhGKh:+76h{h:+7hm?<<<<&=??@0@9@@@@AA=AFAAAAA+B5BBCGDDWGG H H1HMHQHRHSHWHXHHHHH I I*I+I9IRI\JeJnJvJJJYK^KfKKKKK^M`MfMgMNNNNNN1NiNjN hfVaJh:hfVaJhfVh~Phm hl-jht9hl-jhm\jhl-jhm0JUh ht9 hl-jhmI>8@9@nCoCDDVHWH#I$I9IKK_M`MfOgO|O>Q?QRRR$ & Fda$gdl-j$ & Fd`a$gdl-j`gdl-jgdl-jgd:[jNN$OHOdOeOfOmOnORRRRSS$T&TTTJUSUUUUUUUUVVVVVWWWXXXXXXXX*Z}Z@[B[C[X[[}]]B^I^^^^^1_;_____Ѿůųųů hRhBhNh$ hBhB hRh$hBhmh,9hfVht9hl-j hl-jhl-j hl-jhm ht9aJh:hfVaJ hfVaJ@RUUWW)Z*ZB[C[X[|]}]__aaaa*b+bcceehhgd*gdK$ & Fda$gdBgdBgd:[gdl-j_1`9`earaaaaaaaa*b+bbbbbccc_cccccdddddd#e+e4e5e^e~eeeeeeeeeeeefff|ffffffBgXggghh"h-h2hGhKhShbhdhiƷh*h*6 h*hI@ h*h*h:h~Ph$hDh*h*5 h*hDhmhNhI@h!<hKHiZj[jfjjjjjk-k>klkrklllmm0nunnnoooooqprp%q&qrr]rors s"s5sOsPsSsTsssttt.ttt&u:uuu5v6v7vRvuvvvwv}v~v h*hP2Whdh*6 h*hdh*hP2WjhI/0JUjh*h*0JUhdhNh:hI@h*h*6 h*h*h$ h*h:?hllooSsTsttt&uuvvwvvww@yAy{{||}:;`gd*$ & Fda$gd*gd*gd*gdd~vvvww'w7w9wwwwwwxx"x,x5x;x@xAxix{xxxxxxy>y?yAy zzzzc{x{{{{{2|C|d|e|||||||&}+}789;*EFǁ΁ށ6hJh]ajh*h*0JUhd hP2WaJh:hP2WaJh*h*6hDh:hP2Wh$ h*h*I;EFgh}gh΋ϋYZ(WXN$ & Fda$gd*gd:$ & Fda$gd*gd*gd*gdd6?@MNno},0hƅׅ l}Ő΋ϋ A`ŒXՎ()8;GHTޏ <WX1Ցst(.4 hJaJh:hJaJhKNhu_"hNh[h!< h:h: h:h*h:jh*h*0JUh$hdhJ h*h*C45LMUVdTVXBC|ӗ՗֗ )3ΘИؘ٘_m%;A˛ћٛ0=.WmntϞüഭ੢ hN^Jh^hI/^J h^hI/ hNhI/hI/ h*hu_"h*hu_"6 h*hShSh .hKNhNjh*h*0JUhu_"h!<jh*hJ0JU h*h*hd7NOdWX՗֗-.#$ϤФѤҤ$a$gd .gd:[gdBgdI/gd*gdd$ & Fda$gd*`gd*ϞОҞڞxc)jIOͤϤФѤҤԤߤ:M4ļļĸİˈ}h)h2SB*phh2SB*phh2S h .5 h"JG5 h2S5 h5hDhBhmhVxhKNh4h!< h*h*hNhu_" h^hI/hI/ h*hI/ h4^Jh^hI/^J hI/^J hN^J1«}~ޮ߮\]#$ϾоgdF[gdgdtgd2Sgd .gdX4>?^qwŧɧ(Z^tz WX«EF(|}~í,9Jˮܮ߮ǯOP{ް߰оjh{0JUh'*hthTqhNjh .0JUh .hXhC1h!<6 h!<^Jh!< hLGh2Sh{h2S h)h2SD BCó״ٴ&'=?\߶neԸ߸ -@C[]WY"#ü߼+348EHξоdϿhNh] hShF[hF[ h~h'*h hl-jh h9ph2Shf h!<htjh2S0JUh'*h2Sh{E3459;LM G%@9 $xxa$gd . $a$gdgdts$a$gdts$a$gdf gdgdF[ $2349:;M #$ 8FGbcrtķ߅}hFh6 h~hhDhh .0J^JmH sH  hDhh .jhDhh .Uh .^JmH sH hDhh .^JmH sH  hG(h h6 h^hh^h5h hts5htshf h'*hF[h]h"JG1 9:Ts3afpt%./9<} +?@IKABMPh"h"6 h"h" h"ht7h^hB*phhTvh6 hTvhh&rh\h^h6 h\h^h\h h^hBPZefqrxyde ()3DFj)EKTXʻ؏ h6hmHsHh^h6mHsHh^hmHsHh^hhh"hh h6h hhh^h0JhjhUh^h6 h\h^h\ h^h89)F: &\A(CmZ $a$gd $a$gdX`dqtc$&0OZ\fgux4@Ahh5\hh6 hh h"h h6h^h0Jh^h0J6h^h\h^h6h h^hB &-3:BJMVW_u`|'(56?CPQY]dhm5 ƽֽֽֽhOh\h= hXhhTvh6^JhTvh^Jh%h^J h6^J h^Jh`h6h^h6 h^hhhw h6>!(.28@DFKMQVX\bgkp} +Flmr 0d ɽ窟 hTvhh"h"6PJ^Jh"h"PJ^J h"ht7 h6 h^h hO"hO" hO"6hO"h6KH$\ hKH$\hOhKH$\h hOhhOh\ h\5 ,.7YZ[WX]ghbqwx(6;DHSW_cmqxzվվ򷯨h"^JmH sH h"h"^JmH sH h^h6 h^hhv-h6 hv-hhT^h6hT^h0J>*B*ph hT^hjhT^hU hTvhh h6hTvh62x(*+,M $$Ifa$gdV $Ifgdl7$a$gdVgdVgdV $a$gd )*+,MNv0Y/:DH2E3=>A8CEȳȳȳȳȳȳȳϜ hXh>uh>uhhVhV^JhVhV56hVhV6 hV5hVhV5 hVhVhV h"hl7h"^JmH sH h"h"^JmH sH h"h"6^JmH sH 9MNv{rffff $$Ifa$gdV $Ifgdl7kd$$IfTl,0,"f f8 t0644 lBaytVTB9--- $$Ifa$gdV $Ifgdl7kd$$IfTl,r,"f ffff t0644 lBaytVT6- $Ifgdl7kdp$$IfTl,r,"f ffff t0644 lBaytVT $$Ifa$gdV 6- $Ifgdl7kd:$$IfTl,r,"f ffff t0644 lBaytVT $$Ifa$gdV!"56- $Ifgdl7kd$$IfTl,r,"f ffff t0644 lBaytVT $$Ifa$gdV5:?DIJ6kd$$IfTl,r,"f ffff t0644 lBaytVT $$Ifa$gdVJNSX]b $$Ifa$gdV $Ifgdl7bcnsx}B9--- $$Ifa$gdV $Ifgdl7kd$$IfTl,r,"f ffff t0644 lBaytVT}6- $Ifgdl7kdb$$IfTl,r,"f ffff t0644 lBaytVT $$Ifa$gdV6- $Ifgdl7kd,$$IfTl,r,"f ffff t0644 lBaytVT $$Ifa$gdV6- $Ifgdl7kd$$IfTl,r,"f ffff t0644 lBaytVT $$Ifa$gdV6kd$$IfTl,r,"f ffff t0644 lBaytVT $$Ifa$gdV!%*/ $$Ifa$gdV $Ifgdl7/0BHNTB9--- $$Ifa$gdV $Ifgdl7kd$$IfTl,r,"f ffff t0644 lBaytVTTYZmrw6- $Ifgdl7kdT $$IfTl,r,"f ffff t0644 lBaytVT $$Ifa$gdVw{6- $Ifgdl7kd $$IfTl,r,"f ffff t0644 lBaytVT $$Ifa$gdV6- $Ifgdl7kd $$IfTl,r,"f ffff t0644 lBaytVT $$Ifa$gdV6kd $$IfTl,r,"f ffff t0644 lBaytVT $$Ifa$gdV $$Ifa$gdV $Ifgdl7B9--- $$Ifa$gdV $Ifgdl7kd| $$IfTl,r,"f ffff t0644 lBaytVT#6- $Ifgdl7kdF $$IfTl,r,"f ffff t0644 lBaytVT $$Ifa$gdV#)/0D61gdVkd$$IfTl,r,"f ffff t0644 lBaytVT $$Ifa$gdVDEGHYkd$$IfTl0$ t0&644 laeytl7T $$Ifa$gdl7 $Ifgdl7$a$gdVgdVgdV '~ $$Ifa$gdl7 $Ifgdl7lkd|$$IfTl$& t0&644 laeytl7T'(;Gwk $$Ifa$gdl7 $Ifgdl7kd$$IfTl0$ t0&644 laeytl7TGHP\wk $$Ifa$gdl7 $Ifgdl7kd$$IfTl0$ t0&644 laeytl7T\]jvwk $$Ifa$gdl7 $Ifgdl7kdF$$IfTl0$ t0&644 laeytl7Tvwwk $$Ifa$gdl7 $Ifgdl7kd$$IfTl0$ t0&644 laeytl7Twk $$Ifa$gdl7 $Ifgdl7kd$$IfTl0$ t0&644 laeytl7Twk $$Ifa$gdl7 $Ifgdl7kd,$$IfTl0$ t0&644 laeytl7T%1wk $$Ifa$gdl7 $Ifgdl7kd$$IfTl0$ t0&644 laeytl7T12Ew $Ifgdl7kdp$$IfTl0$ t0&644 laeytl7TEFeq~ $$Ifa$gdl7 $Ifgdl7lkd$$IfTl$& t0&644 laeytl7Tqrwk $$Ifa$gdl7 $Ifgdl7kd$$IfTl0$ t0&644 laeytl7Twk $$Ifa$gdl7 $Ifgdl7kd:$$IfTl0$ t0&644 laeytl7Tw $Ifgdl7kd$$IfTl0$ t0&644 laeytl7T~ $$Ifa$gdl7 $Ifgdl7lkd~$$IfTl$& t0&644 laeytl7Twk $$Ifa$gdl7 $Ifgdl7kd$$IfTl0$ t0&644 laeytl7T'wk $$Ifa$gdl7 $Ifgdl7kd$$IfTl0$ t0&644 laeytl7T'(=>?@A{{v{{nnnn{$a$gdVgdVgdVkdH$$IfTl0$ t0&644 laeytl7T ]T $Ifgdl7kd$$IfTl?Fe QF t0    44 lBaytl7T $$Ifa$gdl7 i`TT $$Ifa$gdl7 $Ifgdl7kd$$IfTl,Fe QF t0    44 lBaytl7T%*/i`TT $$Ifa$gdl7 $Ifgdl7kd\$$IfTl,Fe QF t0    44 lBaytl7T/0CEidd\dSG $$Ifa$gdl7 $Ifgdl7$a$gdVgdVkd$$IfTl,Fe QF t0    44 lBaytl7Tymm $$Ifa$gdl7 $Ifgdl7|kd$$Ifl40> $`  t0644 laytl7 ofZZ $$Ifa$gdl7 $Ifgdl7kdn$$Ifl4F> $  x x t06    44 laytl7(-1qh\\ $$Ifa$gdl7 $Ifgdl7kd$$IflF> $ x x t06    44 laytl712HMRqh\\ $$Ifa$gdl7 $Ifgdl7kd$$IflF> $ x x t06    44 laytl7RSotyqh\\ $$Ifa$gdl7 $Ifgdl7kdh$$IflF> $ x x t06    44 laytl7yzqh\\ $$Ifa$gdl7 $Ifgdl7kd $$IflF> $ x x t06    44 laytl7qh\\ $$Ifa$gdl7 $Ifgdl7kd $$IflF> $ x x t06    44 laytl7qh\\ $$Ifa$gdl7 $Ifgdl7kdZ!$$IflF> $ x x t06    44 laytl756qllggbbZb$a$gd>ugd>ugd:[gdVkd"$$IflF> $ x x t06    44 laytl7 456p235efl&-/2367:<=A OPQZh{ü븴h;hp/5>*h;hp/5 hp/5h>uhp/ hXhJhJhJ5 hJ5hXh>u^JhXh>u6hXh>u56 hXh>u h 5hXh>u586LapqdXX $$Ifa$gdl7kd"$$IfTlFM ," e z t06    44 laytl7T $Ifgdl7mdXX $$Ifa$gdl7 $Ifgdl7kdS#$$IfTlFM ," e z t06    44 laytl7TmdXX $$Ifa$gdl7 $Ifgdl7kd$$$IfTlFM ," e z t06    44 laytl7TmdXX $$Ifa$gdl7 $Ifgdl7kd$$$IfTlFM ," e z t06    44 laytl7T3mhh`hWWK $$Ifa$gdl7 $Ifgdl7$a$gd>ugd>ukdZ%$$IfTlFM ," e z t06    44 laytl7T345]nUI== $$Ifa$gdl7 $$Ifa$gdl7kd&$$Ifl\a %"%eh t0T&644 laytl7 $Ifgdl7#kd&$$Iflֈa f%"%ehc\ t0T&644 laytl7 $Ifgdl7 $$Ifa$gdl7 $$Ifa$gdl7 $Ifgdl78// $Ifgdl7kd'$$Iflֈa f%"%ehc\ t0T&644 laytl7 $$Ifa$gdl78// $Ifgdl7kd($$Iflֈa f%"%ehc\ t0T&644 laytl7  $$Ifa$gdl7!%8// $Ifgdl7kd)$$Iflֈa f%"%ehc\ t0T&644 laytl7%)-16 $$Ifa$gdl767FJ8// $Ifgdl7kdr*$$Iflֈa f%"%ehc\ t0T&644 laytl7JNRV[ $$Ifa$gdl7[\ko8// $Ifgdl7kd]+$$Iflֈa f%"%ehc\ t0T&644 laytl7osw{ $$Ifa$gdl78// $Ifgdl7kdH,$$Iflֈa f%"%ehc\ t0T&644 laytl7 $$Ifa$gdl7833+$a$gd>ugd>ukd3-$$Iflֈa f%"%ehc\ t0T&644 laytl7Dkd.$$Ifl\% &"%<F t0T&644 laytl7 $$Ifa$gdl7 $Ifgdl7gd>u2J_e $$Ifa$gdl7 $Ifgdl7efmq8// $Ifgdl7kd.$$Iflֈ%  &"%<6/ t0T&644 laytl7quy} $Ifgdl7 $$Ifa$gdl78// $Ifgdl7kd/$$Iflֈ%  &"%<6/ t0T&644 laytl7 $Ifgdl7 $$Ifa$gdl78// $Ifgdl7kd0$$Iflֈ%  &"%<6/ t0T&644 laytl7 $Ifgdl7 $$Ifa$gdl78// $Ifgdl7kd1$$Iflֈ%  &"%<6/ t0T&644 laytl7 $Ifgdl7 $$Ifa$gdl78// $Ifgdl7kd2$$Iflֈ%  &"%<6/ t0T&644 laytl7 $Ifgdl7 $$Ifa$gdl7 8// $Ifgdl7kdt3$$Iflֈ%  &"%<6/ t0T&644 laytl7 % $Ifgdl7 $$Ifa$gdl7%&./8// $Ifgdl7kd_4$$Iflֈ%  &"%<6/ t0T&644 laytl7/37;< $Ifgdl7 $$Ifa$gdl7<=>A833+$a$gd>ugd>ukdJ5$$Iflֈ%  &"%<6/ t0T&644 laytl7Hkd56$$Ifl\h K$  t0644 laytl7 $Ifgdl7gd>u$a$gd>uIkd6$$Ifl\h K$  t0644 laytl7 $Ifgdl7 $$Ifa$gdl7 &||| $Ifgdl7$$If^a$gdl7jkd7$$Iflh$# t0644 laytl7&'5<BH^NEEE $Ifgdl7$$If^a$gdl7kd)8$$Ifl\h K$  t0644 laytl7HIelrx^NEEE $Ifgdl7$$If^a$gdl7kd8$$Ifl\h K$  t0644 laytl7xy^NEEE $Ifgdl7$$If^a$gdl7kd9$$Ifl\h K$  t0644 laytl7^NEEE $Ifgdl7$$If^a$gdl7kdE:$$Ifl\h K$  t0644 laytl7^NEEE $Ifgdl7$$If^a$gdl7kd:$$Ifl\h K$  t0644 laytl7^NEEE $Ifgdl7$$If^a$gdl7kd;$$Ifl\h K$  t0644 laytl7^NEEE $Ifgdl7$$If^a$gdl7kda<$$Ifl\h K$  t0644 laytl7^R $$Ifa$gdl7kd=$$Ifl\h K$  t0644 laytl7)17=~~~ $Ifgdl7$$If^a$gdl7hkd=$$Iflh$# t0644 laytl7=>Zbhn^NEEE $Ifgdl7$$If^a$gdl7kdK>$$Ifl\h K$  t0644 laytl7no}^NEEE $Ifgdl7$$If^a$gdl7kd>$$Ifl\h K$  t0644 laytl7^NEEE $Ifgdl7$$If^a$gdl7kd?$$Ifl\h K$  t0644 laytl7^NEEE $Ifgdl7$$If^a$gdl7kdg@$$Ifl\h K$  t0644 laytl7^RIII $Ifgdl7 $$Ifa$gdl7kdA$$Ifl\h K$  t0644 laytl7^RIII $Ifgdl7 $$Ifa$gdl7kdA$$Ifl\h K$  t0644 laytl7^RIII $Ifgdl7 $$Ifa$gdl7kdB$$Ifl\h K$  t0644 laytl7^RIII $Ifgdl7 $$Ifa$gdl7kd7C$$Ifl\h K$  t0644 laytl7L^U $Ifgdl7kdC$$Ifl\h K$  t0644 laytl7LMNOQ|z|z|zgd;(_ $^a$gdp/gd:[gd>uhkdD$$Iflh$# t0644 laytl7 #$%rsBCWcjoxÿöÿÿíÿ˞yyy h?h,9jh,90JUh:h,90JaJh:h,9aJjh:h,90JUaJh,9mHnHuh~PmHnHuh,9jh,9UhkjhkUhmhp/5 hm5hp/mHnHu#j!Ehp/hp/5UmHnHu/$rBV#k;[gd2SgdJgd:gdD$a$gd;(_01VW#$kl9;<Z[\] 흺yhmhp/5hkjh .h,90JUaJh .h,9aJh{h,9mH sH h .h,9mH sH  h .h,9jh,90JU h,9aJh:h,9aJjh:h,90JUaJhI/h,9mH sH h,9mH sH h,9 h?h,9& gd:[61h:pl7BP/ =!"#$% 21h:pl7/ =!"#$% 21h:pGf/ =!"#$% $$If!vh5 5#v #v:V l, t065 589BaytVT$$If!vh5 5555#v #v:V l, t065 59BaytVT$$If!vh5 5555#v #v:V l, t06,5 59BaytVT$$If!vh5 5555#v #v:V l, t06,5 59BaytVT$$If!vh5 5555#v #v:V l, t06,5 59BaytVT$$If!vh5 5555#v #v:V l, t06,5 59BaytVT$$If!vh5 5555#v #v:V l, t06,5 59BaytVT$$If!vh5 5555#v #v:V l, t06,5 59BaytVT$$If!vh5 5555#v #v:V l, t06,5 59BaytVT$$If!vh5 5555#v #v:V l, t06,5 59BaytVT$$If!vh5 5555#v #v:V l, t06,5 59BaytVT$$If!vh5 5555#v #v:V l, t06,5 59BaytVT$$If!vh5 5555#v #v:V l, t06,5 59BaytVT$$If!vh5 5555#v #v:V l, t06,5 59BaytVT$$If!vh5 5555#v #v:V l, t06,5 59BaytVT$$If!vh5 5555#v #v:V l, t06,5 59BaytVT$$If!vh5 5555#v #v:V l, t06,5 59BaytVT$$If!vh5 5555#v #v:V l, t06,5 59BaytVT$$If!vh5 5555#v #v:V l, t06,5 59BaytVT$$Ife!vh55#v#v:V l t0&6,55aeytl7T$$Ife!vh5&#v&:V l t0&65&aeytl7T$$Ife!vh55#v#v:V l t0&6,55aeytl7T$$Ife!vh55#v#v:V l t0&6,55aeytl7T$$Ife!vh55#v#v:V l t0&6,55aeytl7T$$Ife!vh55#v#v:V l t0&6,55aeytl7T$$Ife!vh55#v#v:V l t0&6,55aeytl7T$$Ife!vh55#v#v:V l t0&6,55aeytl7T$$Ife!vh55#v#v:V l t0&6,55aeytl7T$$Ife!vh55#v#v:V l t0&6,55aeytl7T$$Ife!vh5&#v&:V l t0&65&aeytl7T$$Ife!vh55#v#v:V l t0&6,55aeytl7T$$Ife!vh55#v#v:V l t0&6,55aeytl7T$$Ife!vh55#v#v:V l t0&6,55aeytl7T$$Ife!vh5&#v&:V l t0&65&aeytl7T$$Ife!vh55#v#v:V l t0&6,55aeytl7T$$Ife!vh55#v#v:V l t0&6,55aeytl7T$$Ife!vh55#v#v:V l t0&6,55aeytl7T$$If!vh555#v#v#v:V l? t0555Baytl7T$$If!vh555#v#v#v:V l, t0555Baytl7T$$If!vh555#v#v#v:V l, t0555Baytl7T$$If!vh555#v#v#v:V l, t0555Baytl7T$$If!vh5 5#v #v:V l4 t06+5 5aytl7$$If!vh5 5x 5x #v #vx :V l4 t06+,5 5x aytl7$$If!vh5 5x 5x #v #vx :V l t06,5 5x aytl7$$If!vh5 5x 5x #v #vx :V l t06,5 5x aytl7$$If!vh5 5x 5x #v #vx :V l t06,5 5x aytl7$$If!vh5 5x 5x #v #vx :V l t06,5 5x aytl7$$If!vh5 5x 5x #v #vx :V l t06,5 5x aytl7$$If!vh5 5x 5x #v #vx :V l t06,5 5x aytl7$$If!vh5 5x 5x #v #vx :V l t06,5 5x aytl7$$If!vh5 5e 5z #v #ve #vz :V l t065 5e 5z ytl7T$$If!vh5 5e 5z #v #ve #vz :V l t065 5e 5z ytl7T$$If!vh5 5e 5z #v #ve #vz :V l t065 5e 5z ytl7T$$If!vh5 5e 5z #v #ve #vz :V l t065 5e 5z ytl7T$$If!vh5 5e 5z #v #ve #vz :V l t065 5e 5z ytl7T$$If!vh5e5h55#ve#vh#v#v:V l t0T&65e5h55ytl7$$If!vh5e5h55c5\5#ve#vh#v#vc#v\#v:V l t0T&65e5h55c5\5ytl7$$If!vh5e5h55c5\5#ve#vh#v#vc#v\#v:V l t0T&65e5h55c5\5ytl7$$If!vh5e5h55c5\5#ve#vh#v#vc#v\#v:V l t0T&65e5h55c5\5ytl7$$If!vh5e5h55c5\5#ve#vh#v#vc#v\#v:V l t0T&65e5h55c5\5ytl7$$If!vh5e5h55c5\5#ve#vh#v#vc#v\#v:V l t0T&65e5h55c5\5ytl7$$If!vh5e5h55c5\5#ve#vh#v#vc#v\#v:V l t0T&65e5h55c5\5ytl7$$If!vh5e5h55c5\5#ve#vh#v#vc#v\#v:V l t0T&65e5h55c5\5ytl7$$If!vh5e5h55c5\5#ve#vh#v#vc#v\#v:V l t0T&65e5h55c5\5ytl7$$If!vh5555#v#v#v#v:V l t0T&65<55F5ytl7$$If!vh5555#55#v#v#v#v##v#v:V l t0T&65<55565/5ytl7$$If!vh5555#55#v#v#v#v##v#v:V l t0T&65<55565/5ytl7$$If!vh5555#55#v#v#v#v##v#v:V l t0T&65<55565/5ytl7$$If!vh5555#55#v#v#v#v##v#v:V l t0T&65<55565/5ytl7$$If!vh5555#55#v#v#v#v##v#v:V l t0T&65<55565/5ytl7$$If!vh5555#55#v#v#v#v##v#v:V l t0T&65<55565/5ytl7$$If!vh5555#55#v#v#v#v##v#v:V l t0T&65<55565/5ytl7$$If!vh5555#55#v#v#v#v##v#v:V l t0T&65<55565/5ytl7$$If!vh5 5^55#v #v^#v#v:V l t065 5aytl7$$If!vh5 5^55#v #v^#v#v:V l t065 5aytl7$$If!vh5##v#:V l t06,5#aytl7$$If!vh5 5^55#v #v^#v#v:V l t065 5aytl7$$If!vh5 5^55#v #v^#v#v:V l t065 5aytl7$$If!vh5 5^55#v #v^#v#v:V l t065 5aytl7$$If!vh5 5^55#v #v^#v#v:V l t065 5aytl7$$If!vh5 5^55#v #v^#v#v:V l t065 5aytl7$$If!vh5 5^55#v #v^#v#v:V l t065 5aytl7$$If!vh5 5^55#v #v^#v#v:V l t065 5aytl7$$If!vh5 5^55#v #v^#v#v:V l t065 5aytl7$$If!vh5##v#:V l t065#aytl7$$If!vh5 5^55#v #v^#v#v:V l t065 5aytl7$$If!vh5 5^55#v #v^#v#v:V l t065 5aytl7$$If!vh5 5^55#v #v^#v#v:V l t065 5aytl7$$If!vh5 5^55#v #v^#v#v:V l t065 5aytl7$$If!vh5 5^55#v #v^#v#v:V l t065 5aytl7$$If!vh5 5^55#v #v^#v#v:V l t065 5aytl7$$If!vh5 5^55#v #v^#v#v:V l t065 5aytl7$$If!vh5 5^55#v #v^#v#v:V l t065 5aytl7$$If!vh5 5^55#v #v^#v#v:V l t065 5aytl7$$If!vh5##v#:V l t065#aytl7 Dd yb  s >A?Picture 1"Q NM_1,z eEb/m- eE@=% NM_1,z eEb/m5 | 8N!9 xZ}ln?J@`A\v8W ~B Jf7?P^Re#d|߶A#3"Q-q8~=}o{Kji;9yuCCC-t)F$qDb2P'#&R;Y"o#Zmd:a^&B"\!2O@.0Z>M6|d^T/`J)9/= ®ʔBF?9 `(OS՟ad5h VhʽacPeꝝI9yXśݳqwA▆$uLchECv[ }r]$țL9NU2KV|B3)*v-smf a?0 u]C %"19)|}bDNB}+п~T_cV6r?b`rI‚|1[x1c;SeQY&G94p,ux}m"~Rg;h"qՃS1X<6$╼LJ([u |0^bpb/gʪǥ#Vb 0?UoK#Hh|gj&8ϗTCC3Sc]{W1{D{qg1Ci; ;ĝkZ#8GqE! kgX%#%jux$M!fw)6fxY7ٞ ; ɂK {&|녋\ʄ?7 {p/5w]!\ʄ˂']h1,/]6}^{{=^"ItY@eTvIs݈;u:)4tR,~nH}+65cQlX2=蘗?7{e[*: bi};1)Ç~;>q?퇼l_g]\_DG1&6L^s{) h34z2Otܻi#ܧܻ}v&k@G@/ⰹA^9]}et/*]s#=d'[X;?~unm,>{헲~-߭P˺xp@~)3}l~1{]fe; ϒΛgf[/_m}o!Kl~-z(_טc <9~(~^vMn]Fcsv78og:,+tmpk?3-C#!XO>Vn|c\OG~3Ǡ˜}:=Rݎ/Y~ -dlQur7 IB - |B>N׾ףߘ%dׁӎ_x6jm2^bQI>7ϵUV2a)KAl0R۔QuXvզCE}ʔjm 9Z AL&T<*nu}Tb~g2ܤ}`\Yu:['Lmn q^f/@Hx޿1&zDHG&{7 / V7aһʹ^pfp?0p3s”x/LRp򅻞/9\&b߽eȹ 띁 !˶Frw8l506l"̩듲,gгn]K(~Gbz9b)%8,]׾|c)MwExnbѳXh1=tb}Se SH={9nqNu=;͞i#jGw|ք>=C=3q)d"%kIڟqC75=܌(Qnmw6ӚDkBj~y7Np>5:lVi2\[,.j(էe.L1A,^dj" 666666666vvvvvvvvv6>6666>6666666666666666666666666666666666666666666666hH6666666666666666666666666666666666666666666666666666666666666666662 0@P`p2( 0@P`p 0@P`p 0@P`p 0@P`p 0@P`p 0@P`p8XV~ OJPJQJ_HmH nH sH tH R`R ]oNormal$a$$CJOJPJQJ_HaJmH sH tH DA`D Default Paragraph FontRi@R 0 Table Normal4 l4a (k ( 0No List Z Z ;(_0 Footnote Text$da$CJPJaJmHsHtHHoH ;(_0Footnote Text Char OJQJaJd+`d ;(_Endnote Text,Char, Char$a$CJaJmHsHtH/! ;(_&Endnote Text Char,Char Char, Char Char OJPJQJ^JaJmHsHtH>*`1> ;(_Endnote ReferenceH*@& A@ ;(_0Footnote ReferenceH*@ R@ 0Header  H$ mHsHtH<ma< ;(_01 / 1.1 / 1.1.1 FF/qF 0 Header CharCJOJPJQJ^JaJ@ `@ 0Footer  H$ mHsHtHF/F 0 Footer CharCJOJPJQJ^JaJD@D :[ List Paragraph ^m$B'`B :[0Comment ReferenceCJaJ6U`6  Hyperlink >*B*phO "JP`J 0 Body Text 2xPJaJmHsHtHH/H 0Body Text 2 CharCJOJQJaJjj V Table Grid7:V 0 8o8 mapple-style-spanPK![Content_Types].xmlj0Eжr(΢Iw},-j4 wP-t#bΙ{UTU^hd}㨫)*1P' ^W0)T9<l#$yi};~@(Hu* Dנz/0ǰ $ X3aZ,D0j~3߶b~i>3\`?/[G\!-Rk.sԻ..a濭?PK!֧6 _rels/.relsj0 }Q%v/C/}(h"O = C?hv=Ʌ%[xp{۵_Pѣ<1H0ORBdJE4b$q_6LR7`0̞O,En7Lib/SeеPK!kytheme/theme/themeManager.xml M @}w7c(EbˮCAǠҟ7՛K Y, e.|,H,lxɴIsQ}#Ր ֵ+!,^$j=GW)E+& 8PK!Ptheme/theme/theme1.xmlYOo6w toc'vuر-MniP@I}úama[إ4:lЯGRX^6؊>$ !)O^rC$y@/yH*񄴽)޵߻UDb`}"qۋJחX^)I`nEp)liV[]1M<OP6r=zgbIguSebORD۫qu gZo~ٺlAplxpT0+[}`jzAV2Fi@qv֬5\|ʜ̭NleXdsjcs7f W+Ն7`g ȘJj|h(KD- dXiJ؇(x$( :;˹! I_TS 1?E??ZBΪmU/?~xY'y5g&΋/ɋ>GMGeD3Vq%'#q$8K)fw9:ĵ x}rxwr:\TZaG*y8IjbRc|XŻǿI u3KGnD1NIBs RuK>V.EL+M2#'fi ~V vl{u8zH *:(W☕ ~JTe\O*tHGHY}KNP*ݾ˦TѼ9/#A7qZ$*c?qUnwN%Oi4 =3ڗP 1Pm \\9Mؓ2aD];Yt\[x]}Wr|]g- eW )6-rCSj id DЇAΜIqbJ#x꺃 6k#ASh&ʌt(Q%p%m&]caSl=X\P1Mh9MVdDAaVB[݈fJíP|8 քAV^f Hn- "d>znNJ ة>b&2vKyϼD:,AGm\nziÙ.uχYC6OMf3or$5NHT[XF64T,ќM0E)`#5XY`פ;%1U٥m;R>QD DcpU'&LE/pm%]8firS4d 7y\`JnίI R3U~7+׸#m qBiDi*L69mY&iHE=(K&N!V.KeLDĕ{D vEꦚdeNƟe(MN9ߜR6&3(a/DUz<{ˊYȳV)9Z[4^n5!J?Q3eBoCM m<.vpIYfZY_p[=al-Y}Nc͙ŋ4vfavl'SA8|*u{-ߟ0%M07%<ҍPK! ѐ'theme/theme/_rels/themeManager.xml.relsM 0wooӺ&݈Э5 6?$Q ,.aic21h:qm@RN;d`o7gK(M&$R(.1r'JЊT8V"AȻHu}|$b{P8g/]QAsم(#L[PK-![Content_Types].xmlPK-!֧6 +_rels/.relsPK-!kytheme/theme/themeManager.xmlPK-!Ptheme/theme/theme1.xmlPK-! ѐ' theme/theme/_rels/themeManager.xml.relsPK]  Y17qh%i7wLBWOB `0DY) I  ? 282l2 ++++++++++++JJJM `)2<jN_i~v64Ϟ4PX R $>Rh;N9M5Jb}/Tw#D'G\v1Eq'/1Ry63%6J[oeq %/<&Hx=nL      #޼dZWg XXXXX $'+CFM!!8@0(  B S  ?Ral(3<Ddijm1 5 !!% %++122 2#2)2.252<2D2R2Z22333)3/3333333346666 ::::{::::::#;*;=;F;Y;^;IIII]]|^^^^^^bbffff{pp1{9{>{D{6|>|Y~d~ ͌Ռ%+,45;t~ZhûȻͻ׻ܻ GPV^ƾ;Ӿھ 49@F18FN:CX`iq\cluAEOVCJ_efnouCPCKfn!(FKQVmr{&. ,1bf(*HQW] fl yWa ' EJ!Rab()<=det{;ǽȽT-e%EF|9YZC j:}/0[\@AdV_`BCo |+Em/0c+-ax(yQZ#&qtADUX"%jm:=Z!3333333333333333333333333333333333333333333333333333333333333333" ()?C O z { $$+%-%@@ A AJMSM\\ΐАnnntϖЖږtt£EFܦߦިިDD4 GVr!_x@' /_L8/N (h t k ']z ^`o(. ^`hH. pL^p`LhH. @ ^@ `hH. ^`hH. L^`LhH. ^`hH. ^`hH. PL^P`LhH. h^h`hH. P^`PhH.. ^`hH... x^`xhH.... ^`hH .....  X^ `XhH ......  ^ `hH.......  8^`8hH........  `^``hH.........^`OJPJQJ^Jo(-^`OJQJ^Jo(o p^p`OJQJo( @ ^@ `OJQJo(^`OJQJ^Jo(o ^`OJQJo( ^`OJQJo(^`OJQJ^Jo(o P^P`OJQJo( h^h`hH. P^`PhH.. ^`hH... x^`xhH.... ^`hH .....  X^ `XhH ......  ^ `hH.......  8^`8hH........  `^``hH......... h^h`hH. P^`PhH.. ^`hH... x^`xhH.... ^`hH .....  X^ `XhH ......  ^ `hH.......  8^`8hH........  `^``hH......... h^h`hH. P^`PhH.. ^`hH... x^`xhH.... ^`hH .....  X^ `XhH ......  ^ `hH.......  8^`8hH........  `^``hH.........x@'']zt kN (h/_L          *        sQI@k   ~N f A fVC+X+GfVT^Jv O"u_".* .I/p/:+7l7 8DL:9@_[@IhA"JGKGKJO9lP~P2SUT, UJVP2W;(_6;bH cd"gl-jek5mZoiots>uXw}DbS]!<mh=S(79VxNxiLh,9s" EF[Bk$kmKNt9t7G(2]aTq'*2ZHN>0V*t{'<::[4t{]o[[$4Ny@ x@Unknown G* Times New Roman5Symbol3. * Arial7.{ @Calibri?= * Courier NewECaecilia-RomanGTimesNewRomanPS9Garamond;WingdingsA BCambria Math"qh› gJ '"w$I}!20PV3qHX $P;(_2! xxTrinity Technical ServicesTrinity Technical Services$     Oh+'0 $0 P \ ht|Trinity Technical Services Normal.dotmTrinity Technical Services26Microsoft Office Word@bj9@DP?@֖"՜.+,D՜.+,@ hp  TTS, Duke UniversitywP  Title 8@ _PID_HLINKSATy +http://beta.epw.in/newsItem/comment/59336/U_ http://beta.epw.in/haysearch/?q=caste%20health&selected_facets=author_exact:S%20V%20Subramanian,%20Shailen%20Nandy,%20Michelle%20Kelly,%20Dave%20Gordon,%20George%20Davey%20Smithu7http://www.equitap.org/ http://iuc2011.in/sites/default/files/white-papers/Innovative Ways to Meet Health Challenges of Urban India_Radhika Arora_Sourav Neogi_Madhavi Misra.pdfTpmailto:ak30@duke.edu  !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~      !"#$%&'()*+,-./0123456789:;<=>?@ABDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefhijklmnpqrstuv{Root Entry F=}Data (P1TableCGWordDocument2SummaryInformation(gDocumentSummaryInformation8oCompObjy  F'Microsoft Office Word 97-2003 Document MSWordDocWord.Document.89q