Women Negotiating Healthcare in Kandy, Sri Lanka

The Privatization Imperative: Women Negotiating Healthcare in Kandy, Sri Lanka

Ramya Kumar

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Dalla Lana School of Public Health University of Toronto

The Privatization Imperative: Women Negotiating Healthcare in Kandy, Sri Lanka

Ramya Kumar

Doctor of Philosophy

Dalla Lana School of Public Health University of Toronto

2018

Abstract

Since the 1980s, poorer countries have shifted health reform efforts from strengthening public systems to increasing the private sector’s role in healthcare provision. Empirical research on healthcare access focuses on quantifying out-of-pocket payments or service utilization, making invisible both user experiences and how the dynamics of public and private provision are contingent on place and social relations. Historically a model for poorer countries, Sri Lanka’s public healthcare system is seeing steady privatization following decades of insufficient state investment alongside incentivized private expansion. However, little is known about what this restructuring means for healthcare access in Sri Lanka.

I employed a Third World Marxist feminist qualitative methodology to explore how the presence of private healthcare shapes access for women in Kandy, Sri Lanka. I asked: Where do women go for healthcare? What are their impressions of the health services they use? How do they navigate public and private systems? And how are these questions shaped by social location?

Using focus groups, interviews, meetings, and a short survey, I gathered data from 40 residents of Udawatta Division. My analysis linked macrostructures and processes of healthcare restructuring with women’s everyday experiences of utilizing healthcare.

Economic exclusion and quality concerns limited the range of private healthcare ‘choices’ available to users. Almost all women mixed public and private services, with these hybrid arrangements differing by social location. Economically disadvantaged users were compelled to consume private healthcare owing to service deficits in the public system. Middle-class women mostly used private outpatient services, and exploited ‘dual practice’ to access more responsive public inpatient care. Socially disadvantaged women, particularly ethnic minorities, relied on the same pathway to avoid neglect and/or abuse within the public system.

The state’s ‘withdrawal’ from healthcare provision, and its incentives for private expansion, has wide-ranging implications for users in Kandy. As women struggle to address service gaps in the public sector, hybrid arrangements stratify services along class and ethnic lines, creating opportunities for private accumulation. My findings interrogate the direction of Sri Lanka’s health reform and call into question global health advocacy for ‘mixed health systems’ as a path to achieving ‘Universal Health Coverage.’

iii

Acknowledgments

I would like to thank the following individuals for contributing to this work:

The research participants for generously sharing their time, opinions and experiences. Without them, this research would not have been possible.

My supervisor, Peggy McDonough, for inspiring my interest in social theory and health. The focus of my research changed considerably during the initial stages of my PhD programme primarily because of the vibrant discussions that took place in her classes. I am especially thankful to her for helping me develop my ideas and sharpen my writing by challenging me to think through every word I wrote in numerous writing projects, including this thesis.

My committee – Peggy McDonough, Anne-Emanuelle Birn, Josephine Pui-Hing Wong and Rupaleem Bhuyan – for providing critical and timely feedback that shaped the ideas contained in this thesis. My committee’s expertise spanned the disciplines of public health, medical sociology, history, nursing, social work, and anthropology. Working with such a multidisciplinary committee was both rewarding and challenging, and strengthened my thesis. I acknowledge, in particular, Peggy’s guidance on theory, Anne-Emanuelle’s expertise on history and international health, and Josephine and Rupaleem’s contributions to methodology.

Mandana Vahabi and Erica Di Ruggiero for their critical appraisals of my thesis.

Anne-Emanuelle Birn for introducing me to the radical possibilities of global health.

Sofia Gruskin for guiding me through my Master’s programme at the Harvard School of Public Health and encouraging me to pursue doctoral studies in public health.

Sumathy Sivamohan for blurring the boundaries between medicine, public health, the social sciences and humanities, and gently nudging me toward Marxist feminist thinking.

Chizoba Imoka for insisting that coloniality is enduring and central to understanding social and economic conditions in Sri Lanka and beyond.

Bandula Petangoda for demanding excellence in anaesthesiology and proving to me that the public healthcare sector in Sri Lanka can and does provide a very high standard of care.

Samah Hassan for her remote and willing assistance with locating, scanning, and emailing references from Toronto to Kandy.

Niyanthini Kadirgamar and Shamala Kumar for help with formatting this thesis.

My mother, Savitri Kumar, for never doubting my ability to succeed in any kind of endeavour, and my father, Vijaya Kumar, for inciting my passion for social medicine. My parents’ steadfast use of coconut oil, refusal to monitor cholesterol levels, and rejection of statins, forced me to question everything I learned in medical school.

My sister, Shamala Kumar, for her love, generosity and sustained friendship; for inspiring me to be fearless in my thinking and actions.

My nephew, Malin Kumaranayake, for questioning the tenets of public health and reminding me that there is life beyond a thesis.

My family in North America, especially Donald Boere, Nelun Fernando, Sriani Fernando, Chizoba Imoka, Culam Mills, Sorna Mills, Viji Murugaiyah, Meena Nallainathan, Ramya Naraharisetti, Earl Nowgesic, Deepa Panchang, and Lindsay Schubiner, for taking care of me while I was away from home.

Lastly, Ahilan Kadirgamar for his love, patience and companionship, through this journey.

Funding Acknowledgements The first phase of this research was supported by a Global Health Student Research Travel Award from the University of Toronto. A Doctoral Research Award from the International Development Research Center (IDRC), Ottawa, Canada, funded the second phase of fieldwork. My doctoral studies were also supported by the University of Toronto, the Ontario Graduate Scholarship Programme, the Margaret McNamara Memorial Fund, the Lupina Foundation and the Delta Kappa Gamma World Fellowship Programme.

Table of Contents

Abstract ...... ii Acknowledgments ...... iv Table of Contents ...... vi List of Tables ...... ix List of Figures...... x List of Appendices ...... xi 1. Introduction ...... 1 1.1 Thesis organization ...... 5 2. A Third World Marxist Feminist Lens on Healthcare Privatization ...... 8 2.1 Introduction ...... 8 2.2 The antecedents of neoliberal global capitalism ...... 9 2.3 Healthcare privatization under neoliberal global capitalism ...... 11 2.4 Healthcare privatization as class struggle ...... 13 2.5 Healthcare privatization as gendered struggles ...... 15 2.6 Summary and implications ...... 17 3. ‘Free Health’ in the Face of Privatization: Breaks and Continuities with Global Health Policy ...... 19 3.1 ‘Free health’ as national consciousness ...... 19 3.2 The beginnings of ‘free health’ in Ceylon ...... 22 3.3 ‘Free health’ in the aftermath of independence ...... 24 3.4 ‘Free health’ under early economic liberalization: 1977 to the late 1980s ...... 26 3.5 ‘Investing in health’ in the second phase of liberalization: 1990s to 2009 ...... 28 3.6 The post-civil war development agenda in the age of ‘Universal Health Coverage’: 2009 to the present ...... 30 3.7 Summary and conclusion ...... 33 4. Healthcare Privatization and its Implications for Access in LMICs: A Critical Review of the Empirical Literature ...... 35 4.1 Introduction ...... 35 4.2 Privatization and out-of-pocket spending ...... 36 4.3 Privatization and quality of care ...... 38 4.4 Privatization and (in)equity ...... 42

4.5 Summary and gaps in the literature ...... 45 5. Exploring Women’s Experiences of Healthcare in Kandy: From Theory to Methods ...... 47 5.1 Locating myself in this work ...... 47 5.2 Situating my research in the critical qualitative research paradigm ...... 48 5.3 Study setting ...... 49 5.4 Phase 1: Exploratory study ...... 53 5.4.1 Objectives ...... 53 5.4.2 Summary of Phase 1 activities and findings ...... 54 5.5 Phase 2: Community-based study ...... 58 5.5.1 Research objectives ...... 58 5.5.2 Research participants and recruitment ...... 59 5.5.3 Data generation ...... 64 5.5.4 Data analysis and writing ...... 67 5.6 Trustworthiness ...... 70 5.7 Summary ...... 72 6. Healthcare in Kandy: Health System and Healthcare Facilities ...... 74 6.1 Structure and organization of healthcare ...... 74 6.1.1 The public system ...... 74 6.1.2 The private sector ...... 77 6.1.3 Standard setting and regulation ...... 79 6.1.4 ‘Public-private partnerships’ ...... 80 6.2 A snapshot of healthcare facilities in and around Udawatta ...... 81 6.2.1 Public facilities and services ...... 81 6.2.2 Private facilities and services ...... 83 6.3 Summary ...... 86 7. Class matters in exercising ‘choice’ in healthcare ...... 87 7.1 Locating the ‘demand’ for alternatives ...... 87 7.2 An array of private ‘choices’? ...... 90 7.3 Limits on ‘choice’ ...... 95 7.3.1 Economic exclusions ...... 95 7.3.2 Quality concerns ...... 100 7.4 ‘Choices’ on the margins ...... 103 7.5 Discussion ...... 107 vii

8. Negotiating Kandy’s ‘mixed health system’: Hybrid arrangements and (in)equity in healthcare ...... 109 8.1 Compelled to consume ...... 109 8.2 Channelling privilege ...... 113 8.3 Managing marginality ...... 116 8.4 Economizing on healthcare ...... 119 8.5 Discussion ...... 122 9. Discussion and conclusion...... 124 9.1 Key contributions to the literature ...... 124 9.1.1 ‘Choice’ and its limits under privatization ...... 125 9.1.2 Strategies of survival and accumulation within mixed health systems ...... 127 9.1.3 Ethnicity and interlocking dimensions of access ...... 130 9.2 Policy implications ...... 132 9.2.1 Unsustainable mixed health systems ...... 132 9.2.2 Stronger public systems ...... 135 9.3 Limitations and directions for future research ...... 137 9.4 Conclusion ...... 139 References ...... 141 Appendices ...... 168

viii

List of Tables

Table 1. Income and health expenditure in Sri Lanka and Kandy District, 2012/13 ...... 50

Table 2. Population by ethnicity: Sri Lanka and Kandy District, 2012 ...... 51

Table 3. Educational achievements of Udawatta residents ...... 52

Table 4. Selected study areas of Udawatta by social class and ethnicity ...... 60

Table 5. Overview of focus groups ...... 62

Table 6. Attendance at final meeting ...... 63

List of Figures

Figure 1. Sri Lanka: Central Province and Kandy District...... 49

Figure 2. A map of Udawatta Division ...... 56

Figure 3. Healthcare facilities in and around Udawatta Division ...... 82

List of Appendices

Appendix A. Health administrators and healthcare providers met with in Phase 1 ...... 168

Appendix B. Script to invite community informants to participate in the study ...... 169

Appendix C. Information letter and consent form for community informants ...... 170

Appendix D. Sociodemographic profile of community informants ...... 172

Appendix E. Script to invite community residents to participate in the study ...... 173

Appendix F. Form to record contact details of community residents and their consent

to be contacted for a focus group or interview ...... 174

Appendix G. Information letter and consent form for focus groups ...... 175

Appendix H. Sociodemographic data form ...... 178

Appendix I. Composition of focus groups ...... 179

Appendix J. Overview of individual interviews ...... 182

Appendix K. Information letter and consent form for individual interviews ...... 184

Appendix L. Script to invite participants to the final meeting ...... 186

Appendix M. Information letter and consent form for the final meeting ...... 187

Appendix N. Focus group discussion guide ...... 189

Appendix O. Individual interview guide ...... 191

Appendix P. Guide for the final meeting ...... 194

Appendix Q. Handout for priority-setting exercise ...... 195

Appendix R. Coding framework…………………………………………………………………… 197

1. Introduction

Over three decades ago, in 1985, the Rockefeller Foundation, an influential United States-based philanthropic foundation, brought together a team of health policy ‘experts’ at a conference in Bellagio, Italy, to investigate the impressive health achievements of Sri Lanka, China, Costa Rica, and Kerala State in India (Balabanova, McKee and Mills 2011).1 The conference report, Good Health at Low Cost, although limited in scope and vision, concluded that sustained political commitment, a social welfare and equity orientation, community participation, and intersectoral action on health, supported by state investment in education, public health and nutrition, were fundamental to their accomplishments (Rosenfield 1985). Notably, the private sector’s2 role in these achievements was not a key focus of the report (Halstead, Walsh and Warren 1985).

The Rockefeller Foundation later abandoned the Good Health at Low Cost initiative, along with its report, which suggested redistribution was critical to population health in resource-poor settings (Birn 2009). The findings of the country-level studies conflicted with the Foundation’s ‘selective primary care’ approach, which was backed by the United States government and the World Bank as a feasible framework to guide health sector development in the Third World.3 The ‘selective’ model, a response to the 1978 Alma Ata Declaration and its pledge for ‘comprehensive primary care’ (WHO 1978), advocated low-cost solutions to public health ‘problems’ in resource-poor settings (Cueto 2004). This approach had been embraced by international health and development agencies in the aftermath of the 1970s economic recession, and dovetailed with structural adjustment programmes adopted by Third World governments under a looming debt crisis. In contrast with the findings of the Good Health at Low Cost report, the health reforms contained therein endorsed privatization (Birn, Pillay and Holtz 2017).

Healthcare privatization encompasses a range of policy directives that shifts the responsibility for healthcare from the state to its citizens (Armstrong et al. 2002). In low- and middle-income

1 Cuba was included in the initial selection of case studies, but later dropped for political reasons (Birn 2005). 2 I borrow from Mackintosh and colleagues (2016) to define the private sector as “the totality of privately-owned institutions and individuals providing healthcare, including private insurers” (p. 596). Conversely, the public sector refers to government-owned clinics, health centres, and hospitals (Basu et al. 2012). 3 Here I use the term ‘Third World’ to refer to the countries of Asia, Africa and Latin America that remained non-aligned with the West and the Communist Bloc during the Cold War. In my theoretical framing, however, I use ‘Third World’ to encompass spaces and peoples who share a history of struggle in the face of colonial domination, a definition that resonates with the present.

1 2 countries (LMICs),4 privatization has encroached upon public healthcare systems at various levels through: the implementation of user-charges; contracting out public healthcare services to the private sector; and selling off state-owned healthcare facilities to private companies. In addition, governments have promoted private healthcare expansion by granting tax concessions and other subsidies to private healthcare companies; encouraging the establishment of private health insurance; opening social insurance and pension funds for foreign investment; and embarking on public-private partnerships (Birn et al. 2017; Jasso-Aguilar, Waitzkin and Landwehr 2004; Murray 2016; Sengupta 2015; Waitzkin 2011).

These policy directives have together increased the burden of payment for users, transferred a substantial share of healthcare delivery to the private sector, and resulted in the widespread adoption of market-based approaches to the financing and delivery of healthcare (Loewenson 1993; Qadeer 1994; Qadeer and Baru 2016; Qadeer and Reddy 2006; Waitzkin 2011). While the merits of public and private provision are debated, particularly in the context of weakened public systems (Hanson et al. 2008), privatization has pushed households into poverty and debt, and exacerbated inequities in access to health services (Alvarez et al. 2011; Birn et al. 2017; Rashid, Akram and Standing 2011; Varman and Vikas 2007).

Sri Lanka is among a handful of LMICs that supports a publicly financed and delivered healthcare system. Guided by a ‘free health’ policy, in place since the early 1950s, the public system delivers services free-of-charge on a walk-in5 basis at clinics and hospitals spread throughout the country (Rannan-Eliya and Sikurajapathy 2009). It is estimated to cover over 90 per cent of inpatient admissions, about 50 per cent of outpatient visits, and the bulk6 of preventive health services (Amarasinghe et al. 2015; Govindaraj et al. 2014). The remainder is delivered by the private sector at significant cost to healthcare users. At present, private health spending accounts for about 60

4 Acknowledging its hegemonic connotations, I use the term ‘low- and middle-income countries’ (LMICs) to be consistent with the contemporary literature on healthcare access. Conceptualized by the World Bank, this group includes all countries with GNI per capita below USD 12,745 in 2013 (World Bank n.d.). 5 All curative services offered by the public healthcare system are available on a walk-in basis. However, the preventive healthcare services delivered by divisional health units are available only to the residents of each Medical Officer of Health Area. More details in Chapter 6. 6 Excepting a small minority of private healthcare users, mostly in Colombo, the public sector provides preventive services to the entire population (Govindaraj et al. 2014). I was unable to find a more precise estimate of the proportion of users who relied on the private health sector for preventive healthcare.

3 per cent of total health expenditure with over 80 per cent financed out-of-pocket (Institute for Health Policy 2015).7

As in other Third World contexts, privatization of Sri Lanka’s health sector accelerated in the 1970s under the pressures of economic liberalization. After the adoption of an ‘open economy’ in 1977, successive governments underinvested in the public system and incentivized private healthcare expansion (Baru 2003). Between 1980 and 1990, the public share of national health expenditure fell from 57 to 46 per cent (Hsiao 2000). This figure remained constant between 40 and 45 per cent in ensuing decades (Institute for Health Policy 2015), but the private share of capital health expenditure increased over the years with a rapid and visible buildout of private hospitals, a result of state subsidization. Between 1990 and 2011, the number of private hospitals, hospital beds and hospital admissions doubled, while private outpatient visits quadrupled (Amarasinghe et al. 2015; Rannan-Eliya and Kalyanaratne 2005).

This shift in healthcare financing and delivery in Sri Lanka has taken place with little scholarly engagement. Extant research on healthcare access largely focuses on the public sector. It suggests that the public system protects economically disadvantaged users from financial hardship (Pallegedara and Grimm 2017; Russell and Gilson 2006; Mackintosh et al. 2016), and draws attention to overcrowded state facilities crippled by service deficits (Perera, Gunatilleke, and Bird 2007; Salgado 2012 as cited in Govindaraj et al. 2014). Economically disadvantaged users endure significant financial hardship as they spend out-of-pocket to purchase the deficit in the private sector (Jayawardena 2016; Perera et al. 2007). While this literature brings to light key issues that shape access in the public sector, we know very little about what access looks like in Sri Lanka’s private health sector.

The scant research that tangentially addresses the question of access to private healthcare in Sri Lanka suggests that healthcare users, including those without means, access private outpatient care owing to gaps in public sector services (Dayaratne 2013; Russell and Gilson 2006; Perera et al. 2007). Secondary analyses of healthcare spending at the national level indicate that private hospitals chiefly cater to the highest wealth quintile. Those representing the lower wealth quintiles

7 These figures reported by the Institute for Health Policy in Colombo vary considerably from those reported by the World Health Organization (WHO) for the year 2012 when the private sector accounted for 61% of total health spending with 51% of this financed out-of-pocket (WHO 2015).

4 rely on private outpatient care to varying degrees with the lowest quintile consuming the smallest share of private healthcare services (Mackintosh et al. 2016; Pallegedara and Grimm 2017).

Despite this evidence, a recent World Bank-commissioned study of Sri Lanka’s private health sector (Govindaraj et al. 2014) concluded that:

The private health sector in Sri Lanka is a growing force, due both to greater investment from private players as well as to greater demand from a population that is exercising its purchasing power to enjoy the benefits of “quicker,” “cleaner,” and “more flexible” service delivery. (P. 43)

The authors recommended “a more effective partnership with the private sector” to address equity concerns that arise as economically disadvantaged users pay out-of-pocket for private healthcare owing to “gaps in the products and services offered by the public sector or due to perceived better quality of care” (p. 43). While the report does not locate these ‘gaps’ in the preceding decades of insufficient state investment in the public system, it fails to address existing links between public and private sectors and their implications for access (Dayaratne 2013; Kumar 2015a). Indeed, we know virtually nothing about how users negotiate public and private services within Sri Lanka’s ‘mixed health system’ (Mackintosh et al. 2016). Neither do we know how key dimensions of social stratification such as social class and ethnicity shape experiences of using healthcare in the context of privatization in Sri Lanka.

Kandy is a densely populated city located in the central hills of Sri Lanka. The city has shared less in the post-civil war ‘development’ drive, which has concentrated in Colombo, the commercial capital. As a result, Kandy’s private health sector is far less ‘developed’ than it is in Colombo (Dayaratne 2013), but is substantial in size and at the cusp of change. Asiri Health, a major player in the domestic private hospital industry, has embarked on a large-scale commercial hospital project, scheduled to be completed in 2018 (de Silva 2016). While Asiri Health’s private hospital venture is expected to transform the landscape of private healthcare in Kandy, exploring healthcare access at this critical turning point will provide a baseline from which to understand the dynamics of privatization that are likely to follow.

Women are key decision-makers in their families and communities in Sri Lanka (Malhotra and Mather 1997). They shoulder the burden of caregiving activities (Herath 2015), and would be intimately familiar with (and profoundly affected by) the changes taking place in healthcare. Among them, women representing socially and economically marginalized groups have been in the forefront of community-led struggles on land, water, livelihoods and against state repression

(Thiruvarangan 2017; Quist 2015; Wipulasena 2013), and may be well positioned to offer a unique and transformative perspective on healthcare privatization.

In this thesis, I explore how the presence of private healthcare shapes access for women residents of an urban administrative division in Kandy, Sri Lanka. I ask: Where do women go for healthcare when confronted with a health problem? What are their impressions of and experiences with the healthcare facilities and services they use to resolve their health concerns? How do they access and navigate public and private healthcare systems? And how are these questions shaped by social location (e.g., social class, ethnicity)? Guided by a Third World Marxist feminist methodology, I link the macro processes of healthcare restructuring with women’s day- to-day practices of using healthcare to contribute a historically and socially located, nuanced political and economic analysis of healthcare access in Kandy, Sri Lanka.

1.1 Thesis organization

This thesis is organized as follows. In Chapter 2, I lay out my theoretical framework, which profoundly shaped this research from its conceptualization and design to its implementation and completion. The chapter brings together concepts from the neo-Marxist, Marxist feminist and Third World feminist literature(s) to clarify the material underpinnings of healthcare privatization. I begin by locating capitalism in its colonial past before mapping out the historical antecedents of neoliberal global capitalism in the Third World. I then draw on the neo-Marxist literature to theorize the emergence of healthcare privatization under neoliberalism and address its class dimensions. As conventional neo-Marxist critiques do not speak to the gender dimensions of privatization, I borrow from Marxist feminist political economy and Third World feminist scholarship to deal with this critical aspect. The chapter ends with a discussion of the implications of my theoretical framework to my research.

The third chapter charts the trajectory of ‘free health’ in Sri Lanka. I begin by locating ‘free health’ in Ceylon’s anti-colonial struggle(s) and describe its ascent into the national consciousness. Next, I outline the antecedents of ‘free health’ in the late colonial period and its adoption as government policy in the aftermath of independence. Drawing parallels between shifts in global and national health policy, the remainder of the chapter analyzes the fallout of ‘free health’ in three distinct periods following the adoption of an ‘open economy’ in Sri Lanka: early economic liberalization (1977 to the late 1980s); the second phase of economic liberalization (the 1990s to 2009); and the third phase of post-civil war neoliberal ‘development’ (2009 and after).

Having outlined the history of health sector development in Sri Lanka, I move on to reviewing the public health literature on healthcare access in LMICs. Chapter 4 critically reviews relevant empirical work from LMICs (including Sri Lanka) to examine the implications of healthcare privatization for users. I present my findings in relation to three dimensions: out-of-pocket spending, quality of care, and inequity. I end by drawing attention to gaps in this body of literature, particularly the dearth of empirical studies that explore the ways in which users experience and negotiate healthcare within the ‘mixed health systems’ of LMICs.

The fifth chapter addresses methodology. I begin by locating my approach in my experiences as a physician working in the public system of Sri Lanka and a graduate student in the field of ‘global health’ in North America. I then situate my project in the critical qualitative research paradigm, before providing a detailed account of how I designed and implemented my research. In doing so, I describe the methods I used for participant recruitment, data generation, and analysis, in two phases of field research. The chapter ends with a short discussion of the trustworthiness of my research.

Chapter 6 introduces the reader to the physical organization of healthcare in Kandy, Sri Lanka. It consists of two sections. The first provides an overview of the structure and organization of the public and private healthcare sectors of Sri Lanka, which largely mirrors the healthcare set up in Kandy. The second hones in on the public and private healthcare facilities in Kandy to familiarize the reader with the healthcare services available to residents of Udawatta.

In Chapter 7, I explore the use of public and private healthcare options by Udawatta residents to understand the kinds of healthcare ‘choices’ that have opened up in the presence of private healthcare for women in Kandy. The first part of the chapter delves into the conditions prevailing in Kandy’s public sector to suggest why alternative healthcare ‘choices’ may be desirable. The rest of the chapter examines the public and private healthcare ‘choices’ available to the various social groups resident in Udawatta to highlight class and ethnic differences.

Chapter 8 complicates this picture of access by delving into the strategies of survival employed by women as the state recedes from healthcare provision. I draw attention to three specific ways in which women mixed public and private healthcare services, or used what I call ‘hybrid arrangements,’ to access healthcare. First, working class women were compelled to consume private services to address widespread service gaps within the resource constrained public system. Second, middle/upper-middle class users and economically disadvantaged Muslim

7 women (and others) consulted public sector physicians in the private sector to access a more responsive service at public healthcare facilities. Third, regardless of social location, women combined public and private services to economize on healthcare spending. The chapter ends with a discussion of the implications of these arrangements for equity of access to services.

In the final chapter, I bring together my findings and discuss them in light of my theoretical framework and the literature on healthcare privatization. I first discuss the key contributions made by this thesis before moving on to their implications for policy. I then outline the limitations of this research and propose some ways they may be addressed. I conclude with some final thoughts on the future of health sector development in Sri Lanka.

2. A Third World Marxist Feminist Lens on Healthcare Privatization

Chapter overview: I approach this research through a Marxist lens, a perspective that allows me to address the material underpinnings of healthcare privatization. I begin by locating capitalism in its colonial past before mapping out the historical antecedents of neoliberal global capitalism in the Third World. I then draw on the neo-Marxist literature to outline the emergence of healthcare privatization under neoliberalism and address its class dimensions. As conventional neo-Marxist critiques do not speak to the gender dimensions of privatization, I borrow from Marxist feminist political economy and Third World8 feminist scholarship to deal with this critical aspect. The chapter ends with a discussion of the implications of my theoretical framework for my research.

2.1 Introduction

Capitalism’s history is deeply intertwined with the histories of colonial occupation and imperialist expansion. The imposition of colonial state structures in the peripheries led to class and state formations that were structurally different from those in metropolitan centres. Although indigenous middle-classes allied themselves with powers at the centre to advance capitalist development in the colonial world, the capitalist transition was incomplete as accumulated capital was siphoned off to the West, leading to the rise of a weak indigenous capitalist class (Alavi 1972). Industrialization never took off, as colonial economies, by imperial design, remained dependent on the export of primary commodities (Patnaik 1973).

Anti-colonial mobilizations brought together diverse classes and groups of different ideological persuasions into concerted mass movements for independence. The national consciousness that emerged was, however, limited by its dominant class moorings, and failed to dismantle the colonial apparatus (Fanon [1963]1991). This story was complicated by the growing indebtedness of Third World economies in the context of capitalist crisis. As I suggest in the sections that follow, the debt crisis was critical to the rise of neoliberal global capitalism and the adoption of healthcare privatization as a policy in Third World settings, including Sri Lanka.

8 Rather than a geographic construct, I use the term Third World to designate spaces and peoples sharing a history of struggle in the face of colonial domination much like Levander and Mignolo (2011:3)’s understanding of the Global South as “an entity that has been invented in the struggle and conflicts between imperial global domination and emancipatory and decolonial forces that do not acquiesce with global designs.”

2.2 The antecedents of neoliberal global capitalism

Ross and Trachte (1990) define capitalism as:

A mode of commodity production for exchange; where labor power itself is a commodity; and where a capitalist class that owns and controls the means of production extracts surplus value from a working class by purchasing its only means for survival – their labor power – through payment of wages. (P. 23)

They identify three variants of capitalism that prevailed in the West over the course of the 20th century: competitive capitalism, monopoly capitalism, and global capitalism. These variants, they argue, emerged at specific political and economic junctures under changing capital-labour, capital-capital and capital-state relations. Competitive capitalism was still the primary mode of production by the end of the 19th century when small firms vied for markets in circumscribed geographic regions. In the face of a falling rate of profit and repeated crises of under accumulation, capitalism evolved from competitive to its monopoly variant in the early 20th century (Ross and Trachte 1990).

The introduction of large-scale production processes during the post-Second World War economic boom saw the rise of big firms. Organized labour made gains during the so-called golden age of capitalism although the social reforms implemented facilitated deeper market penetration (Panitch and Gindin 2012; Veltmeyer and Petras 2005). As multinationals, aided by technological innovation, shifted production processes to places where labour was less organized (and cheap), capitalist states adopted investment-friendly policies to support the expansion and entrenchment of global capitalism (Ross and Trachte 1990). These policies were underpinned by neoliberalism (Harvey 2005).

Defined by Harvey (2005) as “a theory of political economic practices that proposes that human well-being can best be advanced by liberating individual entrepreneurial freedoms and skills within an institutional framework characterized by strong private property rights, free markets, and free trade,” (p. 2) neoliberalism is traced back to the Mont Pelerin Society founded in 1947. Neoliberal ideology was adopted as policy doctrine in the West much later to resolve the deepening 1970s economic crisis. Led by Reagan in the United States and Thatcher in the United Kingdom, Western governments embraced neoliberalism in place of post-Keynesian welfarism (Harvey 2005). Unlike Keynesian economics, which supported state intervention to reduce unemployment

10 during an economic downturn, neoliberalism espoused a shrinking role for the state in the economy, at least, in theory. In reality, the neoliberal state strengthened its repressive apparatuses and created a favourable business environment at the cost of labour. The age of neoliberal global capitalism is characterized by the growing dominance of finance capital and the steady retrenchment of welfare (Harvey 2005; Panitch and Gindin 2012).

The global expansion of neoliberalism was facilitated by the Third World debt crisis. Although the post-Second World War boom had seen renewed investment in Third World ‘development,’ the developmental drive was short-lived as the terms of trade for primary products deteriorated in the world market by the 1950s, compelling Third World governments to expend foreign reserves and borrow from metropolitan capital to finance public spending (Lakshman 1985; Patnaik 1994). As Third World debt grew, the World Bank and International Monetary Fund (IMF), created under the 1944 Bretton Woods Agreement, began to shape fiscal and monetary management of fragile Third World economies. In newly independent Ceylon (now Sri Lanka), for example, a 1951 World Bank mission recommended trade liberalization and cuts to public spending (Lakshman 1985).

As oil prices rose in the context of the 1973 OPEC9 oil crisis, Third World governments became heavily indebted. OPEC member states invested ‘petrodollars’ in international commercial banks, which, in turn, reinvested this capital as low-interest loans in Third World ‘development.’ Mediated by the World Bank and IMF, the conditions associated with the loans helped to integrate Third World markets into the global capitalist system, while debt servicing disposed of an ever-growing share of Third World revenue. An intractable Third World debt crisis ensued in 1979 as the United States Federal Reserve raised interest rates to address rising inflation in the domestic economy (Harvey 2005).

The World Bank and IMF negotiated debt relief and issued more loans to Third World governments to prevent the spread of non-repayment and protect the interests of creditors (Patnaik 1994). Widely known as structural adjustment programmes, these loans specified conditions for economic restructuring, peddled to national governments as a pathway to ‘economic development’ (Harvey 2005; Panitch and Gindin 2012; Patnaik 1994). These conditions, which were formulated under the aegis of the Washington-based World Bank, International Monetary Fund (IMF) and US Treasury (the ‘Washington Consensus’),

9 Organization for Petroleum Exporting Countries

11 encompassed a neoliberal policy package that stipulated trade liberalization, financial deregulation, regressive forms of taxation, cuts to social spending and privatization (Harvey 2005; Radice 2005).

Structural adjustment integrated Third World economies with global financial markets and increased their dependence on trade (Patnaik 1994). Subject to competition from global markets and the internationalization of production, labour came under extreme exploitation (Bakker and Silvey 2008; Mies 1982; Walton and Seddon 1994). As governments borrowed heavily from volatile capital markets to finance deficit spending, Third World debt rose, necessitating further conditional borrowing and intensified market integration (Panitch and Gindin 2012). While these developments entrenched the dependent relationship between newly independent Third World states and the First World, some view the debt crisis as having been carefully engineered to reign in the “Third World Project” in the context of the strengthening Non-Aligned Movement and calls for a New International Economic Order (Prashad 2012:43) .

In short, Western governments adopted neoliberalism as an economic framework to resolve the 1970s economic crisis. The Third World debt crisis was crucial to the global expansion of neoliberalism and the subsequent entrenchment of neoliberal global capitalism. Structural adjustment, through its support for trade liberalization, financial deregulation, regressive forms of taxation, cuts to social spending and privatization, intensified penetration of metropolitan finance and industrial capital into the Third World. Next, I draw on neo-Marxist scholarship to outline the emergence of healthcare privatization under neoliberal global capitalism.

2.3 Healthcare privatization under neoliberal global capitalism

Modern health systems emerged with the industrial revolution when healthcare shifted from the ‘private’ realm of the family to the state’s purview. Organized healthcare was necessary to ensure healthy workers for nation building and stem social unrest (Birn et al. 2017). Labour movements made concrete gains in the context of the post-Second World War boom. Influenced by post- Keynesian welfarism, governments of industrialized nations invested substantially in healthcare during this period (Panitch and Gindin 2012; Veltmeyer and Petras 2005). These efforts were more restrained in the decolonizing world, albeit with some notable exceptions, such as Sri Lanka, Cuba, Costa Rica, Uruguay, and Kerala State in India (Birn et al. 2017).

Public healthcare systems in the West saw steady privatization after the 1980s as neoliberalism eroded post-Keynesian welfarism. Governments deployed neoliberalism to implement massive cuts to health budgets and privatize healthcare, albeit under tremendous resistance (Leys 2010; Schrecker and Bambra 2015). Panitch and Gindin (2012:15) understand these events as “a political response to the democratic gains that had been previously achieved by working classes” (emphasis in original).

In the Third World, indebted national governments supported healthcare privatization under the pressures of economic liberalization (Laurell and Aralleno 1996; Qadeer 1994). Privatization entailed introducing user-charges for public sector services, out-sourcing health services to the private sector, selling off state-owned healthcare facilities, subsidizing private healthcare expansion, establishing private health insurance, and, more recently, opening social insurance and pension funds for foreign investment, and embarking on public-private partnerships (Birn et al. 2017; Murray 2016; Qadeer and Baru 2016; Sengupta 2015; Waitzkin 2011). These policy directives have together transferred a substantial share of healthcare delivery to the private sector and resulted in the adoption of market-based approaches to healthcare financing and delivery with grave implications for access (Birn et al. 2017; Loewenson 1993; Qadeer 1994).

The healthcare reforms adopted by indebted Third World governments differed in scope and extent. In Sub-Saharan Africa, the widespread introduction of user-fees in public systems, implemented across Africa as part of the 1987 Bamako Initiative, to increase funds for public healthcare, saw a decline in health service utilization (Birn et al. 2017; Loewenson 1993). In Latin America, privatization involved contracting out various dimensions of service and expansion of private health insurance, leading to stratified access to health services (Birn et al. 2016). In South Asia, states subsidized the private health sector, diverting much-needed funds from public systems to the private sector, compromising access to public healthcare services (Baru 2003; Qadeer and Reddy 2006). In Sri Lanka, the government continued to support its ‘free health’ policy while promoting private healthcare expansion (Baru 2003; Jayasuriya 2010).

In recent decades, financialization10 has facilitated opening the health sectors of LMICs to external investment. This thrust in global health policy is reflected in the health sector reforms supported

10 Fine (2013:55) defines financialization as “the increasing scope and prevalence of interest bearing capital in the accumulations of capital.”

13 under the United Nation’s Millennium Development Project, and, more recently, the Sustainable Development agenda (Kotz 2011; Waitzkin 2011; Sengupta 2015). The United Nation’s endorsement of ‘Universal Health Coverage’ as target eight of the third Sustainable Development Goal has been accompanied by a proliferation of initiatives to introduce and/or expand private health insurance, creating new markets for the multinational healthcare and insurance industries (Qadeer and Baru 2016; Sengupta 2015). As Loeppky (2010:60-1) points out, the health sector holds tremendous potential for accumulation even under crisis as “there is neither a ceiling for how healthy societies should be, nor a shortage of medical conditions – real or contrived – that require diagnosis and (preferably prolonged) treatment.”

To summarize, healthcare privatization has been driven by neoliberal ideology both in the West and in the Third World. In the latter, neocolonial structural adjustment programmes, adopted under the pressures of a deepening debt crisis, guided healthcare privatization in the age of neoliberal global capitalism. The health reforms contained therein commenced a long-term project of dismantling public healthcare systems. I now turn to the neo-Marxist literature to theorize healthcare privatization as a product of class struggle.

2.4 Healthcare privatization as class struggle

Characterized in the work of Marx and Engels as the antagonism between the bourgeoisie and the proletariat, Mandel (1990) describes class struggle as:

[A] struggle for the division of the social product between the direct producers (the productive, exploited class) and those who appropriate what Marx calls the social surplus product, the residuum of the social product once the producers and their offspring are fed … and the initial stock of tools and raw materials is reproduced …. The ruling class functions as a ruling class essentially through the appropriation of the social surplus product. By getting possession of the social surplus product, it acquires the means to foster and maintain … superstructural activities …and by doing so, it can largely determine their function – to maintain and reproduce the given social structure, the given mode of production – and their contents. (P. 5)

This two-class model, critics assert, holds less relevance today, particularly in industrialized settings where a growing professional-managerial middle class complicates class distinctions (Aronowitz 2003). For neo-Marxist scholars, however, at the heart of capitalist accumulation is class struggle. The latter continues to be deeply entrenched in daily life and manifests in the enforcement of private property, the seizure of profits by capital, and a myriad other forms of class-based oppression (Aronowitz 2003; Goldthorpe and Marshall 1992).

The rise of neoliberalism and subsequent dismantling of welfare states are conceived of as class struggle within neo-Marxist scholarship (Harvey 2005; Kiely 2005; Navarro 2007). From this perspective, neoliberalism rose to dominance in response to the waning power of the capitalist classes. The value of assets held by the wealthiest plunged after the 1970s recession as communist and socialist regimes gained ground in many parts of the world (Harvey 2005). The threat of full employment and rising inflation in the West meant that, from capital’s perspective, labour had to be disciplined. As part of this project, Western governments cracked down on unions and cut back welfare (Panitch and Gindin 2012).

Income inequalities grew exponentially in the decades that followed with the world’s wealthiest individuals holding an ever-growing share of global wealth (Harvey 2014). The assets held by the wealthiest one per cent of the world’s population dipped to 45 per cent of global wealth in 2009 following the economic crisis, and steadily increased to stand at 51 per cent by 2016 (Credit Suisse Research Institute 2016). Harvey (2005) offers the concept of “accumulation by dispossession” to explain the extraction of wealth from workers and peasants under the contemporary economic order (p. 159). Accumulation by dispossession does not rely on production processes alone, but also on dispossessing people of their hard-won entitlements through privatization, commodification, financialization, and the management of capitalist crises. These processes have together reversed mechanisms of redistribution and undermined access to the commons, including healthcare, education, land, water, and so forth (Harvey 2005:160-5).

As the dynamics of capital-labour relations shifted in favour of capital, healthcare was reoriented along profitmaking lines (Baru 2003; Leys 2010; Qadeer and Reddy 2006), under the convergence of the healthcare industry, the medical establishment and the state (Coburn 2010). National governments across the world slashed health budgets and promoted healthcare privatization by intensifying public subsidization of private healthcare, channelling more and more wealth from labour to the transnational capitalist class (Navarro 2007; Waitzkin 2011). In LMICs, international financial institutions and growing debt-servicing obligations guided the state’s withdrawal from healthcare (Birn et al. 2017; Qadeer and Baru 2016).

In sum, healthcare privatization is understood as class struggle in the neo-Marxist literature. The political and economic transformations that took place in the 1970s favoured the channelling of an ever-growing proportion of the world’s wealth from labour to capital. States intervened to dispossess the working classes by retrenching welfare and supporting privatization of the social sector, including healthcare. While neo-Marxist theories address the material underpinnings of

15 healthcare privatization, its gender dimensions remain unaddressed in much of this work. To address this gap, I turn to Marxist feminist political economy and Third World feminist scholarship.

2.5 Healthcare privatization as gendered struggles

Marxist feminists extend critiques of privatization from the realm of production to the social conditions that enable production. In doing so, they employ the concept of social reproduction to clarify the role of women’s domestic labour in sustaining the capitalist system (Bakker 2007; Bakker and Silvey 2008; LeBaron 2010; Molyneux 1979). Social reproduction encompasses three dimensions: biological reproduction, reproduction of labour power (including subsistence, education and training), and “social practices connected to caring, socialization and the fulfilment of human needs” (Bakker and Gill 2003a:4). The last includes caring for the infirm.

Fundamental changes have occurred in the governance of social reproduction since the 1970s. These shifts, which shape women’s lives and their work, drive the renewed interest in social reproduction within Marxist feminist scholarship. As Bakker (2007:541) explains:

More recent interest in [social reproduction] reflects the increasingly privatised forms of social provisioning and risk that characterise the neoliberal moment in the global political economy. In other words, the everyday activities of maintaining life and reproducing the next generation are increasingly being realised through the unpaid and paid resources of (largely) women as states withdraw from public provisioning, with the result that capitalist market relations increasingly infiltrate social reproduction. Hence, the renewed focus on social reproduction seeks to place its costs at the centre of an analysis of the capitalist system of accumulation as well as relating it to questions of how the surplus in such an economy is distributed.

Bakker and Gill (2003b:36) suggest that a “reprivatisation” of social reproduction has occurred alongside capital’s expanding reach. Here, reprivatisation encompasses four movements: the return of caregiving activities to the household and their simultaneous commodification; social transformation from a collective sense of being to fragmented households and individuals; loss of control of/access to essential needs/services through dispossession; and erosion of social provisioning through interconnected domestic and transnational mechanisms that support privatization of the state (p. 36-7). Addressing multiple levels, Bakker and Gill theorize the links between macrostructural policy shifts and individual or household practices.

Bakker and Gill’s (2003b) theorization enables a considered critique of the gendered dimensions of healthcare privatization given its focus on social reproduction. As healthcare is privatized with

16 concomitant increases in healthcare costs for users and governments, caregiving activities are transferred to the private sphere of the home. As examples, the infirm may delay seeking healthcare (or forego healthcare altogether) to save on healthcare spending, while healthcare companies introduce measures to contain costs, such as ‘day procedures,’ which translate to lengthy recuperation at home. As women pick up the slack in their families and communities, they are themselves unduly burdened by out-of-pocket payments because they make up a larger section of the unemployed poor (Armstrong et al. 2002; Ewig 2008; Gideon 2008; Gideon 2012). Privatization also has grave consequences for a feminized healthcare workforce as restructuring often entails massive public sector lay-offs (Stinson 2004; Stinson, Pollak and Cohen 2005).

Bakker and Gill’s theory of reprivatisation does not speak to the ways in which women negotiate these shifts in the governance of social reproduction. In this vein, Sassen (2000:506) points to the proliferation of new survival strategies that rely on the exploitation of women’s productive and reproductive labour, a phenomenon she refers to as the “feminization of survival.” Using the examples of trafficking and precarious immigrant labour, Sassen highlights the consequences of intensified market integration and cross-border flows for marginalized Third World women whose productive labour is a key source of foreign exchange for indebted governments grappling with debt servicing obligations. This understanding of survival may well be extended to the myriad of strategies economically disadvantaged (and other) women adopt to fend for themselves and their families as the state recedes from healthcare.

Building on Marxist feminist theories, Third World feminisms speak to both the material and cultural dimensions of women’s (and men’s) oppression. Responding to hegemonic and colonizing ‘second-wave’ feminisms, Third World feminisms reject the notion of an all- encompassing and singular experience of women’s oppression, emphasizing instead “complex and intersecting oppressions and multiple forms of resistance” (Herr 2014:2). Critical in this respect are the histories of colonialism and ongoing neocolonial interventions that shape women’s subjectivities in the Global South (Herr 2014).

From a Third World feminist location, Mohanty (2003:229) engages the experiences of marginalized Third World women to shed light on the “the micropolitics of everyday life” vis-à-vis macrostructures and processes. For Mohanty (2003:231-2), Third World women are well positioned to offer a transformative social justice oriented perspective on power and inequity because they face the brunt of capitalist globalization:

This particular marginalized location makes the politics of knowledge and the power investments that go along with it visible so that we can then engage in work to transform the use and abuse of power….My claim is not that all marginalized locations yield crucial knowledge about power and inequity, but that within a tightly integrated capitalist system, the particular standpoint of poor indigenous and Third World/South women provides the most inclusive viewing of systemic power.

Thus, Mohanty (2003) employs a feminist historical materialist methodology to decenter neoliberal global capitalism by exploring the experiences of women who remain peripheral to it. Privileging neither the ‘macro’ nor the ‘micro’ level, she addresses the dialectic between the general and the particular, while validating subjectivity and agency. Drawing on the work of Fanon, Mohanty insists on anchoring her historical material framework in decolonization, asserting that the latter is fundamental to social change and may “only be engaged through active withdrawal of consent and resistance to structures of psychic and social domination” (p. 7).

Using a Third World Marxist feminist methodology, I investigate women’s experiences of using healthcare to understand what the presence of private healthcare means for access in Kandy, Sri Lanka. In delving into women’s experience, I draw on Bannerji’s (1995:87-8) historical materialist understanding of experience as “an interpretive relation” that does not “[valorize] any person’s or group’s experience as a repository of “truth”” but rather upholds their “direct agency and … representation as knowers and practitioners,” while “[achieving] a validated status for [their] experience which contains the potential for revolutionary knowledge.” For Bannerji, experience is not “a body of content indicative of a seamless subjectivity or psychological totalization, but rather a subject’s attempt at sense-making” (p. 88).

In sum, Marxist feminist political economy helps us theorize the shifts that have occurred in the governance of social reproduction under neoliberal capitalism. While Sassen (2000) draws attention to the survival strategies that have emerged as Third World women face the onslaught of capitalist globalization, Third World feminist Mohanty (2003) advances a feminist historical materialist methodology that engages the experiences of Third World Women. Anchoring her framework in decolonization, Mohanty explores the ‘micro’ level in light of macrostructures and processes.

2.6 Summary and implications

The aggressive privatisation taking place in health and other social sectors reflects the changing landscape of capital-labour, capital-capital, and capital-state relations under neoliberal global

18 capitalism. Privatization has been pursued by states, multilateral agencies, and the capitalist classes to expand and integrate global markets and as a strategy of welfare retrenchment. Shaped by imperial and class interests, governments of LMICs implemented (and are implementing) privatizing health reforms even as the financial consequences associated with deregulation and market integration hit the working classes, and, in particular, women. In this backdrop, how does privatization shape healthcare access for women in Kandy? How do they negotiate healthcare privatization and the erosion of public healthcare? And what can we learn about macrostructures and processes of healthcare restructuring from women’s experiences?

In addressing these questions, I employ a Third World Marxist feminist methodology to explore women’s experiences of using healthcare in Kandy, Sri Lanka. Rather than attempting to represent an all-encompassing reality about healthcare in Sri Lanka, I approach my research from the vantage point(s) of women residents of an urban administrative division in Kandy, taking into account the dynamic nature of their lives, their interactions with the health system, and the material realities of healthcare privatization. Contextualizing my analysis in the particularities of time, place and community, I try to center the experiences of women whose stories about healthcare usually do not enter the mainstream. I believe they, more than any other social group, would be intimately familiar with the transformations taking place in healthcare. Having outlined the conceptual framework that guides this research, in the next chapter, I delve into the trajectory of healthcare privatization in Sri Lanka, situating national developments in global health policy.

3. ‘Free Health’ in the Face of Privatization: Breaks and Continuities with Global Health Policy

Chapter overview: This chapter explores the evolution of ‘free health’ in Sri Lanka from its beginnings under colonialism to the present moment. First, I locate ‘free health’ in Ceylon’s anti- colonial struggle and address its ascent into the national consciousness. Next, I map out the antecedents of ‘free health’ in the late colonial period and its adoption as government policy in the aftermath of independence. The remainder of the chapter outlines the fate of ‘free health’ following the adoption of an ‘open economy.’ Drawing parallels between global and national health policy, I focus on three distinct periods: early economic liberalization (1977 to the late 1980s); the second phase of economic liberalization (the 1990s to 2009); and the third phase of post-civil war neoliberal ‘development’ (2009 and after).

3.1 ‘Free health’ as national consciousness

Ceylon (or Sri Lanka after 1972) was first occupied by the Portuguese (1505 – 1638), then the Dutch (1638 – 1796), and finally the British (1796 – 1948). The island remained under the administration of the English East India Company until 1802 when it became a crown colony (Ludowyk 1966; Wickramasinghe 2006). The British colonial government adopted a laissez-faire economic policy in supporting the development of Ceylon’s capitalist plantation economy (Hewa 1995; Jayawardena and Kurian 2015; Wickramasinghe 2006).

Under the influence of anti-colonial mobilisations in India, led by Gandhi and others, the late 19th century brought together diverse classes, ethnicities, and groups of varying ideological persuasions into a cogent Ceylonese anti-colonial movement (Jayawardena 1972; Wickramasinghe 2006). In the face of agitation, the transition to self-governance came relatively soon. The 1931 Donoughmore Constitution granted universal franchise, despite opposition from the Ceylonese elite, and established a State Council comprising fifty elected state councillors (Wickramasinghe 2006).

The dominance of Marxist parties and trade unions in Ceylon’s anti-colonial movement pushed the task of nation building into considerations of welfare. The Suriya Mal (Hibiscus Flower) Movement, 11 launched in the early 1930s to protest the sale of poppies on Remembrance Day,

11 The Suriya Mal Movement later evolved into the Trotskyite Lanka Sama Samaja Party (Wickramasinghe 2006).

19 20 actively engaged in relief work during the 1934/35 malaria epidemic, and drove the government to make substantial investments in healthcare (Silva 2014). Food subsidies came into effect in the early 1940s as the government introduced a food-rationing scheme to address food scarcity during the Second World War (Mahalingasivam 1978).12 ‘Free education,’ which originated in the Swabasha (vernacular language) Movement, became government policy in 1945, and granted universal primary and secondary education to the masses (Wickramasinghe 2006).

The calls for welfare that propelled nation building developed into a mass-based consciousness, albeit limited by its dominant class moorings. Fears of spreading communism brought Ceylonese ‘moderates’ in alliance with the ruling British to stem labour unrest, forcing sections of the Marxist leadership to flee the country in the early 1940s. Fissures in the nation’s ethnic fabric became apparent in the aftermath of independence in 1948. Fuelled by anti-immigrant sentiment and the plantation workers’ allegiance to the left, the government, still a dominion of the British Empire, disenfranchised the up-country plantation Tamils (‘Indian Tamils’)13 by introducing new requirements to qualify for citizenship (Wickramasinghe 2006).14

The left’s fragmentation paved the way for the Mahajana Eksath Peramuna (MEP), a left-of-centre Sinhala nationalist party espousing socialist objectives, to rise to dominance in the 1950s. The Sri Lanka Freedom Party (SLFP), a major stakeholder in the MEP, came to power in 1956, defeating the bourgeois United National Party (UNP), in power since independence. While supporting a ‘socialist’ platform, the government used the constitutional process to remedy the perceived disproportionate representation of Burghers15 and Ceylon Tamils in the civil service and other professions. The controversial 1956 Official Language Act, which made Sinhala the official language, effectively barred persons lacking Sinhala fluency from entering public service

12 The government’s attempt to remove the rice subsidy in 1953 – a World Bank recommendation – was thwarted by a country-wide anti-government protest (‘the Great Hartal’) led by the trade union movement, resulting in the resignation of the incumbent prime minister (Lakshman 1985; Wickramasinghe 2006). The rice subsidy was chipped away gradually in the 1960s and 70s, and eventually removed in 1979 two years after the adoption of an ‘open economy’ (De Silva [1987]1997; Mahalingasivam 1978; Wickramasinghe 2006). It was replaced by a targeted food stamp scheme, which was later abandoned in favour of a means-tested nation-wide poverty alleviation programme (Hewavitharana 2004). 13 Still referred to as ‘Indian Tamils’ in the national census, political representatives of the upcountry Tamils have rejected this designation and called for the adoption of ‘Indian Origin Malayaga Tamils’ in its place (Daily Mirror 2016). 14 Sections of plantation workers were granted citizenship in Sri Lanka or repatriated to India in subsequent years. However, it was not until the 1986 Grant of Citizenship to Stateless Persons Act No. 5 and the 1988 Grant of Citizenship (Special Provisions) Act No. 39 were passed that citizenship was granted to all disenfranchised plantation workers (Wickramasinghe 2006). 15 People of European descent, often inter-married with other ethnic groups (De Silva [1987]1997).

(Wickramasinghe 2006; Kadirgamar 1989).16 Expressions of Tamil dissent to such moves were countered by a series of anti-Tamil pogroms (Wickramasinghe 2006).

The 1972 Republican Constitution, which rejected Ceylon’s British dominion status, also reaffirmed Sinhala as the official language, and gave Buddhism, the religion of a large majority of Sinhalese, ‘the foremost place.’ Tamil mobilisations against the state escalated in the North with the build-up of consensus for succession, particularly after the state-sponsored anti-Tamil riots of 1983 (Wickramasinghe 2006). As the civil war unfolded, the Liberation of Tamil Tigers Eelam (LTTE) ascended among several militant groups to claim sole representation of the Tamil people. The rise of narrow Sinhala and Tamil nationalisms meant that concerns of other ethnic minorities, including the Muslims and upcountry Tamils, remained peripheral (Hoole et al. 1990).

The 30-year civil war ended in 2009 when the Sri Lankan state militarily defeated the LTTE. After the passage of numerous United Nations Human Rights Council resolutions, the government has yet to address the grievances of ethnic minorities (Kadirgamar 2014; Kadirgamar 2017). In the post-war context, the Muslim community has become the latest target of Sinhala-Buddhist nationalism. A series of ethnically motivated campaigns led by Sinhala chauvinist elements were embarked upon in 2013 to counter the spread of perceived ‘Islamism’ (Farook 2014). In 2014, a Muslim village in the south faced mob violence, causing fatalities and colossal damages to property (Colombo Telegraph 2014). A second spate of anti-Muslim attacks began in 2017 and is ongoing (Nazeer 2017).

In short, ‘free health’ emerged in the national consciousness in the context of anti-colonial struggle. However, this mass-based national consciousness was limited by bourgeois aspirations, and resulting inclusions and exclusions. Ceylon’s welfare state strengthened in the post- independence period even as the nation’s ethnic fabric disintegrated. The social policies embraced by the state in the aftermath of independence, including the ‘free health’ policy, may hold different meanings for marginal ‘Others.’ Having located ‘free health’ in Ceylon’s anti-colonial struggles and ensuing nationalisms, I now chart the development of Sri Lanka’s health sector starting with the late colonial period.

16 A subsequent 1958 Act gave official status to the Tamil language in the Northern and Eastern Provinces where the population was pre-dominantly Tamil and Muslim. While these provisions were removed in 1961 and then reintroduced in 1965, Tamil was recognized as a national language, alongside Sinhala, much later in 1987 (Wickramasinghe 2006).

3.2 The beginnings of ‘free health’ in Ceylon

The foundations of Ceylon’s western17 medical system were laid under British colonial rule. The Portuguese and Dutch set up hospitals in strategic naval locations during the early years of occupation. In 1706, the Dutch established a ‘leprosy asylum,’ credited to be the first civilian western medical facility in Ceylon (Uragoda 1987).18 The British extended western medical services to the urban poor in 1819 by opening the Pettah Hospital in Colombo, and missionaries aided the expansion of Western medicine thereafter. The Anglican Church’s Friend-in-Need Societies set up of ‘pauper hospitals’ in major townships, and the American Ceylon Mission focused its efforts on the northern peninsula (Jones 2009; Uragoda 1987). A key milestone in the development of Ceylon’s state health sector was the 1858 establishment of the Civil Medical Department after which medical services were transferred from military to civil administration. The government subsequently took over the Friend-in-Need Society hospitals, which would form the backbone of the public curative sector in later years (Uragoda 1987).

Efforts to expand services to rural areas began in the 1870s through the introduction of a dispensary system that placed ‘assistant medical officers’19 in rural stations (Meegama 2012; Uragoda 1987). In 1912, the government introduced public health legislation to contain a hookworm scourge on plantations,20 and subsequently invited the Rockefeller Foundation, a United States-based philanthropic foundation, to assist with hookworm control efforts. Arriving on the island in 1914, the Foundation confronted an intransigent Planters’ Association, and soon shifted its programme to non-plantation areas. A significantly restructured and more comprehensive public health programme was established by the Foundation in the suburbs of Colombo with assistance from local government authorities. This sanitation programme laid the

17 Ayurveda, Unani and Siddha medical systems functioned in pre-colonial times and ran in parallel to Western medical services under colonialism (Uragoda 1987). Today, indigenous medical services constitute a vital arm of Sri Lanka’s ‘free’ health system (Jones 2004). While privatization has encroached upon the indigenous system in different ways, my thesis focuses on the (western) allopathic system. 18 Having received royal patronage in ancient ‘Sri Lanka,’ the concept of housing the ill may be traced back to the 4th century B.C. (Uragoda 1987). 19 ‘Assistant medical officers’ were trained for a shorter period (one year) at the newly opened Colombo Medical School for placement in rural dispensaries (Meegama 2012). 20 Ordinance no. 9, Medical Wants of Labourers in Planting Districts, and Ordinance no. 10, Prevention of the Spread of Diseases Among Labourers (Hewa 1995).

23 groundwork for the health units system, which would evolve into the widely successful preventive health sector (Hewa 1995).

By the 1920s, the Western medical system had expanded considerably, designed chiefly to control communicable diseases and promote healthy reproduction (Jones 2004). The government began to regulate indigenous medical services, which still served a vast majority of the indigenous population. A training college for indigenous physicians was established, and the Ayurvedic Medical Council soon followed. While these efforts built a structure through which the state would expand Ayurvedic medicine in future, institutionalization of indigenous medicine under the patronage of Western medicine resulted in the creation of a hybrid form of indigenous medicine in colonial Ceylon (Jones 2009).

Rural expansion of allopathic care accelerated after the malaria epidemic of 1934/35, which emerged in the nexus of recession, drought, and food scarcity. While the colonial government faced criticism from various quarters for its delayed response to the humanitarian crisis (Silva 2014), State Councillors demanded action under electoral pressure. In the wake of the Suriya Mal (Hibiscus Flower) Movement, and its active involvement in relief efforts, a more responsive government issued free rice rations and school meals, and took steps to strengthen rural health services (Silva 2014; Wickramasinghe 2006). By independence, the public system comprised a network of 246 institutions spanning the entire country (Jones 2004). During this period, health sector expansion was financed chiefly through taxes on imports (Moore 2017).

Private healthcare services ran in parallel, catering to a wealthier minority. The European and Ceylonese elite enjoyed private attendance at home, and visited private ‘nursing homes,’ 21 located mostly in Colombo (Jones 2009; Uragoda 1987). The colonial government encouraged public sector specialists to engage in private practice in order to maintain low physician wages in the public sector. Dual practice created a channel through which private sector patients gained entry to government hospitals through their respective private providers. A 1950 evaluation of the health sector would attribute overcrowding in state hospitals to dual practice (Jones 2009).

To sum up, the foundations of Sri Lanka’s public healthcare system were laid under British colonialism. Initially driven by missionary and philanthropic efforts, the public system developed

21 Small-scale private hospitals with limited inpatient facilities.

24 substantially after the shift to self-governance in the context of a devastating malaria epidemic. The private health sector ran in parallel serving elites in urban settings. Now I consider health sector developments after Ceylon gained independence in 1948.

3.3 ‘Free health’ in the aftermath of independence

In the first decade of independence, health spending grew both in absolute terms and as a proportion of national income, financed mainly through trade tariffs (Rannan-Eliya and De Mel 1997; Moore 2017). Prompted by Britain’s 1946 passage of the National Health Service, the government commissioned Dr. J. H. L. Cumpston, former Australian Director-General of Health Services, to evaluate Ceylon’s health sector. The 1950 Cumpston Report that ensued set the direction for health reform through its recommendations (Jones 2009). The government endorsed ‘free health’ by eliminating user-fees from the public system,22 and the 1952 Health Services Act brought state health services under a centralized department (Government of Sri Lanka 1952; Rannan-Eliya and Sikurajapathy 2009; Jayasuriya 2010).23

Public spending on health, education, housing and food subsidies increased from 5 to 12 per cent of GDP between 1950 and 1965. Correspondingly, the number of public hospital beds and health worker cadres doubled during this period (Gunatilleke 1985). Average caloric intake among the “bottom 40 per cent” increased as the state intervened to control food prices through its food distribution system (Gunatilleke 1985:116). Land reforms released state lands for agricultural resettlement in the north central and eastern regions (Gunatilleke 1985), a controversial move given the ethnic makeup of the areas of resettlement (Hoole et al. 1990). Despite the exclusion of some groups, such as the up-country Tamil plantation workers, from these benefits (Gunatilleke 1985), the early post-independence period was crucial for the consolidation of Ceylon’s welfare state.24

22 Existing literature does not clearly map out a timeline or provide an analysis of the actors and forces behind the adoption of the ‘free health’ policy in independent Ceylon. Some sources indicate that user-fees were removed from the system in 1950 (Perera 1985; Haniffa 2006) and others 1951 (Rannan-Eliya and de Mel 1997). 23 Notably, among Cumpston’s recommendations, a ban on public sector physicians engaging in private practice was opposed by the medical establishment, and did not see implementation (Jayasuriya 2010). 24 This challenges Jones’ (2004) ‘model colony’ thesis, which attributes health gains to policy transfers between Britain and colonial Ceylon with little acknowledgement of welfare progress made in later years.

By the late 1950s, foreign exchange reserves began to decline as the terms of trade became unfavourable for Ceylon’s export-dependent economy. As tax revenue fell, the incumbent left-of- centre SLFP government adopted state-led import substitution within a closed economy, and reached out to socialist nations to finance state capitalism. A pro-West UNP government returned to power in 1965, heralding the inflow of foreign aid, including an IMF stabilisation package, which stipulated cuts to public spending, devaluation of the rupee, dual exchange rates, and import liberalization (Lakshman 1985). Still, over a third of government spending was devoted to education, health, food subsidies, transport, and welfare assistance (Jayasuriya 2010).

Reflecting global political and economic conditions, the country faced an economic recession in the 1970s (Herring 1987). The incumbent Left alliance resisted the IMF’s economic regime, and the country’s debt servicing ratio spiralled as the government borrowed heavily from international financial markets (Lakshman 1985). Notably, the government did not attempt to expand its income tax base at this critical juncture (Moore 2017), but, instead, resorted to cutting welfare subsidies. The 1971 budget introduced a user-fee in the form of a stamp duty for outpatient services (Herring 1987; Rannan-Eliya and de Mel 1997). The centralized purchasing system and national formulary designed in the 1960s to rationalize state pharmaceutical imports was extended in 1972 to the private sector (Lall and Bibile 1977).25

Critical developments took place in relation to the regulation of private practice. In the late 1950s, the left-of-centre SLFP government restricted the private practice privileges of newly qualifying public sector specialists and medical officers. Strike action by the Government Medical Officers’ Association, the public sector physicians’ union, resulted in the government removing this ban in 1964, and permitting specialists to engage in dual practice at designated government- administered centres.26 Referred to as ‘channelling centres,’ these facilities generated revenue for the government by retaining a part of the consultation fee. Later, in 1970, the Left alliance in government attempted to phase out dual practice by introducing legislation that prohibited public sector specialists from engaging in private practice in stations where specialists were available

25 The Bibile and Wickramasinghe pharmaceutical reforms, later endorsed by the United Nations Conference on Trade and Development (UNCTAD) and the WHO as a model for poorer countries, faced the wrath of transnational pharmaceutical companies, and was abandoned by the government in 1976 (Lall and Bibile 1977). 26 The government also permitted some categories of medical officers in rural stations where there were no Western- qualified private general practitioners within a five-mile radius to engage in dual practice (Jayasuriya 2017).

26 on a full-time basis in the private sector. This piece of legislation resulted in many public sector specialists retiring early from state service to work full-time in the private sector (Jayasuriya 2017).

In brief, the newly independent government endorsed its ‘free health’ policy in the early 1950s by eliminating user-fees from the public system. Significant investments in the health sector followed until the mid-1960s, after which an economic downturn placed restrictions on public spending. Successive left-leaning governments attempted to curb privatization by restricting the private practice privileges of public sector physicians, resulting in the creation of ‘channel centres,’ which exist to this day. These restrictions on private practice were short-lived as a westward-leaning UNP government came to power in 1977, heralding the era of economic liberalization.

3.4 ‘Free health’ under early economic liberalization: 1977 to the late 1980s

Signing on to a Standby Agreement and Extended Fund Facility Agreement with the IMF, the new government liberalized trade, devalued the rupee, abolished price controls, cut welfare allocations, and promoted privatization, while taking draconian measures to contain the trade union movement (Lakshman 1985; Skanthakumar 2013). Budgetary allocations to the social sector plummeted from about 40 per cent between 1970 and 1977 to 11 per cent in 1981 (Jayasuriya 2010). As revenues from trade tariffs fell steeply, the government failed to increase its tax base. As a result, the proportion of government revenue financed through direct taxation declined after 1977 with a concomitant rise in financing through indirect taxation (Moore 2017).

The food subsidy, which accounted for a large share of the government’s welfare expenditure, was slashed by half in 1979 through the introduction of means testing (Gunatilleke 1985; Jayasuriya 2010; Moore 2017). However, the government remained committed to ‘free health’ and ‘free education’ (Gunatilleke 1985), perhaps owing to the electoral repercussions of dismantling these systems. In fact, the stamp duty for outpatient care, introduced in the 1971 budget, was eliminated in 1977, a move viewed by some as a populist gesture (Jayasuriya 2010). However, the government also promoted private healthcare expansion by removing restrictions on dual practice (for both specialists and non-specialist medical officers), providing loans for medical practitioners to establish private facilities, and deregulating the pharmaceutical and health insurance industries (Baru 2003; Kelegama, Rannan-Eliya and de Mel 1997).

This shift in national health policy coincided with the rise of neoliberal thinking in the aftermath of the 1970s world economic recession. Supported by the World Bank and the United States

27 government, the Rockefeller Foundation spearheaded a campaign to discredit the 1978 Alma Ata Declaration along with its pledge for universal and comprehensive primary healthcare (Kumar et al. 2016). Having endorsed Alma Ata, the Sri Lankan government adopted a national strategy to achieve ‘Health for All,’ encompassing capacity building, decentralization of health services, strengthening rural structures for advocacy and community mobilization, and investing in rural public health infrastructure. Although Alma Ata pushed the government to strengthen maternal and child health services, the ensuing decade saw a steady decline in public health spending (Economic Review 1987; Gunatilleke 1985; Perera 1985).

The commitments made by Third World governments toward ‘Health for All’ were thwarted by a looming debt crisis. Designed under the ‘Washington Consensus,’ structural adjustment programmes arrived with sweeping reforms for the health sector, including cuts to health budgets, privatization of public services, implementation of user-fees and expansion of private health insurance (Birn et al. 2017). Indeed, structural adjustment remained at the heart of World Bank interventions in the health sector through the 1980s. Reflecting this thrust, a widely cited 1987 World Bank report proposed user-fees for government services, health insurance, “effective” use of private sector resources, and decentralization of health services (World Bank 1987:5). These reforms undermined Alma Ata and conferred a greater role to the private sector in healthcare delivery.

Although some neoliberal policy prescriptions relevant to the health sector were not adopted in Sri Lanka, the government began a long-term project of dismantling the public healthcare system by adopting a strategy of sustained underinvestment in the public health sector and incentivized private healthcare expansion. Expenditure on health as a percentage of total government expenditure dropped from 5.2 per cent in 1977 to 3.4 in 1978 to 1.9 in 1979 (Perera 1985). Healthcare was decentralized to the provinces in 1987 under the 13th Amendment to the Constitution, although not as per World Bank policy.27 The taxes devolved to the provinces were not substantial, and the provincial departments of health remained financially reliant on the central government (Hsiao 2000).

27 The 13th Amendment sought to address the grievances of Tamil-speaking peoples of northern and eastern Sri Lanka by devolving a limited set of powers to the provinces (Hoole et al. 1990).

Thus, having adopted an ‘open economy’ amidst a world recession, the Sri Lankan government continued to support the ‘free heath’ policy, while cutting public health spending and promoting private sector expansion. Although the government did not fully embrace the World Bank’s health reform agenda at this critical juncture, it laid the groundwork for healthcare privatization.

3.5 ‘Investing in health’ in the second phase of liberalization: 1990s to 2009

A key milestone in the development of Sri Lanka’s private health sector was the establishment of the Board of Investment (BOI), set up to improve the “investment climate” in the country (BOI Sri Lanka 2016). The BOI offered a range of fiscal incentives to expand private healthcare, including tax holidays, concessionary rates on corporate income tax, import duty exemptions, and concessionary lease terms on state lands (Rannan-Eliya and Kalyanaratne 2005). Several large- scale private hospital projects received a nod from the BOI in the 1990s, changing the landscape of private healthcare in Colombo (Dayaratne 2013).

Efforts to establish domestic mechanisms to attract (foreign) investment to Sri Lanka’s health sector ran in parallel with the institutionalization of neoliberal ideology within structures of global governance. After the 1991 dissolution of the Soviet Union, Western governments joined forces to expand the capitalist system by creating the World Trade Organization and adopting numerous free trade agreements. Unhampered by the Cold War, they pursued liberal social agendas, while supporting privatization, trade liberalization, deregulation, and financialization (Amin 2006). Unlike the 1980s ‘Washington Consensus,’ the economic regime of the 1990s (or the ‘Post-Washington Consensus,’) acknowledged market failures and recommended state intervention to address ‘institutional constraints’ (Harvey 2005; Saad-Filho 2005).

The health reforms supported under the ‘Post-Washington Consensus’ were brought to bear in the 1993 World Development Report, Investing in Health. Framing health as an investment opportunity, the report recommended public financing of an essential “basket” of health services, while the remainder was to be offered within a competitive market where “suppliers (both public and private) … [would] compete both to deliver clinical services” in a context where “[d]omestic suppliers [would] not be protected from international competition” (World Bank 1993:6). The prescribed “basket”’ would cover pre-defined services for mothers and children and a limited set of communicable diseases. Coverage for widely prevalent non-communicable diseases would be subject to the availability of resources (Laurell and Arellano 1996).

The market order remained unchallenged in the 2000 United Nations Millennium Project, which brought forth a range of initiatives that linked health to economic development (Amin 2006). The 2001 World Health Organization (WHO) Commission on Macroeconomics and Health aimed to clarify the role of health in economic development. Reaffirming public provisioning of ‘essential’ health services, the Commission supported the introduction of community-financing schemes to cover the remainder. Under this framework, a “close-to-client” system would bring health services closer to the people through “a mix of state and non-state health service providers, with financing guaranteed by the state” (WHO 2001:7). Multilateral agencies and national governments embraced these policies as the influence of corporate actors on global health agenda-setting rose through numerous ‘public-private partnerships’ (Kumar et al. 2016).

With a ‘low-cost’ public system covering most aspects of care (Hsiao 2000), Sri Lanka did not adopt many of the ‘new’ healthcare financing strategies supported at the global level. However, the government did establish the Board of Investment and introduced measures to encourage the consumption of private healthcare. The President’s Fund, a populist initiative established in the 1990s by the President’s Office, expanded its mandate to (partially) covering the costs of a limited set of medical procedures in the private sector (Government Information Centre 2009). Since 1997, public sector employees contribute to a health insurance scheme, which offsets the costs of a pre-defined package of private services (National Insurance Trust Fund n.d.).

Economic liberalization intensified in Sri Lanka between 2001 and 2004 under the brief tenure of a UNP government. The private share of capital expenditure reached an all-time high of 29 per cent in 2002 (Institute for Health Policy 2015). Reflecting this increase, the number of private hospitals rose from 66 to 123 and the number of private hospital beds doubled, between 1990 and 2009 (Amarasinghe et al. 2015). The relatively small private health insurance market also expanded during the same period, its contribution to private health spending increasing from 1 to 5 per cent (Institute for Health Policy 2015).

This expansion of private healthcare took place even as the state invested insufficiently in the public health sector. Government expenditure on health as a percentage of general government expenditure fell from 6.8 to 5.9 per cent between 2000 and 2009 (WHO 2012) as the government invested heavily on a military offensive against the Tamil Tigers. Although the public share of health spending remained fairly constant at just over 40 per cent, admissions per state hospital bed rose from 50 to 80 per year between 2000 and 2009, reflecting intensified use of extant public facilities (Amarasinghe et al. 2015; Institute for Health Policy 2015).

The World Bank’s interventions in Sri Lanka’s health sector increased in the aftermath of the 2002 cease-fire agreement between the government and the LTTE. Between 2003 and 2010, the Bank supported the first leg of Sri Lanka’s Health Sector Development Project, which aimed to support the health sector “adapt to the challenges resulting from the double burden28 of disease by improving equity, quality and efficiency of the health system” (World Bank 2004:3). The project proposal contained plans to assess the feasibility of alternative healthcare financing options (World Bank 2004), although evidence of such an assessment was not documented in the project’s completion report (World Bank 2011). A review of the private health sector would be embarked upon after the end of the civil war to strengthen the knowledge base for healthcare privatization (details in section 3.6 below).

In sum, the Sri Lankan government entered a second phase of economic liberalization after the 1990s as opposition to neoliberal global capitalism weakened on the world stage. The health sector was opened for private investment through the Board of Investment, which led to the commencement of several large-scale commercial hospital projects in Colombo. Despite the World Bank’s interest in exploring ‘alternative’ healthcare financing options, the government evidently did not support such measures. The ‘free health’ policy remained intact, while the use of existing public facilities intensified as the state underinvested in public healthcare.

3.6 The post-civil war development agenda in the age of ‘Universal Health Coverage’: 2009 to the present

As the civil war ended in 2009, the government embarked on a massive wave of liberalization, exploiting the inflow of foreign capital (Kadirgamar 2013). Embracing the rhetoric of post-war development, it spearheaded an ambitious programme that sought to make Sri Lanka the ‘Wonder of Asia’ under the 2010 Mahinda Chinthana policy framework (Department of National Planning 2010). In the health sector, Mahinda Chinthana aimed to expand hotel-like state-of-the- art healthcare facilities and upgrade existing health services through ‘public-private partnerships.’ These services, to be covered by health insurance, were expected to support the burgeoning medical tourism industry (Department of National Planning 2010:150-3).

28 The ‘double burden’ refers to the rising incidence of communicable and non-communicable diseases in LMICs.

The National Health Development Plan 2013-2017 (NHDP) that followed was designed in parallel with the second phase of the World Bank-supported Health Sector Development Project. The NHDP emphasized the private sector’s role in healthcare delivery:

The government encourages individuals to pay for their own health care when they are able to do so and for the private health sector to meet these needs… The capacity of the private health sector to provide quality care needs to be strengthened. Consideration also needs to be given to how best to encourage partnerships between state and private sectors to deliver quality services and contribute to the national health goals (Ministry of Health n.d.:5-6).

Yet, the health sector development strategies outlined therein primarily targeted the public system with just a few explicitly supporting private sector expansion. Among the latter were, “promoting and regulating the private sector to deliver affordable and quality services; improving public- private partnerships in providing healthcare services;…promoting medical tourism; [and] …promoting alternative financing options for healthcare” (Ministry of Health n.d.:9). Some proposals contained in the NHDP’s framework for action spelled out privatization: outsourcing cleaning, laundry, security, ambulance and other transport services (p. 306); developing sections dedicated to medical tourism in government and private sector hospitals (p. 338) and; introducing social insurance and fee-for-services (p. 344).

The second phase of the World Bank-supported Health Sector Development Project, launched in 2013 and valued at USD five billion, is to see completion in 2018. While it aims to “upgrade the standards of performance of the public health system and enable it to better respond to the challenges of malnutrition and non-communicable diseases” (World Bank 2013:17), a second component of the project focuses on “innovation, results and capacity building” (World Bank 2013:18-22). A World Bank-commissioned private health sector review, undertaken before the commencement of the second phase, addressed “significant knowledge gaps on the private health sector” and hoped to “foster a dialogue on opportunities for collaboration between the government and the private sector” (Govindaraj et al. 2014:ii). Despite such overtures, the project’s loan disbursement indicators ultimately targeted the public sector (World Bank 2017a).

Paradoxically, ‘free health’ continues to receive endorsements from high-ranking politicians. For example, the incumbent President’s 2014 Election Manifesto articulated a commitment to strengthen “free health,” and promised a “unified state service” that would “coordinate Western, Eastern and indigenous systems of medicine” and provide “all medical drugs and tests” through “appropriate state institutions” (Sirisena 2014:35). The etching of ‘free health’ in the nation’s

32 consciousness may explain such endorsements. As Hsiao (2000:57) suggests, healthcare is politically contentious in Sri Lanka “so much so that [user-fees] will not be debated in public.”

Even so, there appears to be wide consensus among politicians and policymakers on increasing the private sector’s role in healthcare delivery. In the run up to the 2015 Parliamentary Elections, the two major contenders supported the introduction of health insurance to cover out-of-pocket expenses in the private health sector, while a third, a prominent leftist party, recommended public financing of private healthcare as an intermediate solution to addressing service deficits in the public system (Deshodaya Movement 2015). Moreover, the present coalition government’s 2017 budget proposals included numerous privatization strategies relevant to the health sector, including a pledge to introduce a health insurance scheme for all school-going children. The Government Medical Officers’ Association, the public sector physicians’ union, has been vocal in its objections to the budget proposals (GMOA 2016), but the privatization agenda evidently finds support in policy quarters.

The National Strategic Framework for Development of Health Services 2016-2025, developed following national level consultations with health policymakers, health administrators, and medical professionals, contains plans to introduce a national health insurance scheme to provide financial security for “certain healthcare problems” (Ministry of Health 2016:71). Evidently not associated with a World Bank credit facility, this policy document includes a number of initiatives that seek to harness the private sector’s contribution to service delivery, particularly in primary care. Noteworthy is that the National Strategic Framework repeatedly cites ‘Universal Health Coverage’ as a guiding principle (Ministry of Health 2016).

‘Universal Health Coverage’ (UHC) made its debut on the global stage in 2005 when the World Health Assembly passed a resolution on the role of social health insurance in moving toward universal coverage (World Health Assembly 2005). The WHO formally introduced UHC in the 2010 World Health Report, Health Systems Financing: The Path to Universal Coverage. Unlike Alma Ata, UHC does not call for ‘Health for All’ and fails to specify the state’s role in healthcare provision. In fact, on introducing UHC, the WHO did not venture beyond suggesting that national governments should “ensure that all providers, public and private, operate appropriately and attend to patients’ needs cost effectively and efficiently” (WHO 2010: p. xviii). The United Nations has since adopted UHC as target eight of the third Sustainable Development Goal (Sustainable Development Knowledge Platform 2016).

The rise of UHC coincided with the 2008 recession and the IMF’s support for accelerated private health sector development in LMICs (Stuckler and Basu 2009). Although UHC’s 2015 iteration in the Sustainable Development Goals addresses equity concerns,29 critics suggest that it remains overdetermined by financial risk protection (Birn et al. 2016; Sengupta 2015; Waitzkin 2015). Indeed, country-level assessments of UHC efforts have largely focused on health insurance schemes or diversifying provision through private sector engagement (Cotlear et al. 2015; Reich et al. 2015). Health systems modelled on the ‘Health for All’ principle, like Sri Lanka’s publicly financed and delivered system, have been overlooked in such assessments.

Thus, there has been a tangible shift toward privatization in the post-war health reform agenda in Sri Lanka. While the World Bank’s efforts to foster formal partnerships between the public and private health sectors are yet to materialize, politicians and policymakers endorse the introduction of health insurance and ‘public-private partnerships.’ This policy direction finds support in the sustainable development agenda in the guise of ‘Universal Health Coverage,’ a framework designed under the exigencies of the 2008 economic recession.

3.7 Summary and conclusion

‘Free health’ entered Ceylon’s national consciousness in the context of anti-colonial struggle. A thriving plantation economy with claims on the state made by a franchised population, guided by socialist mobilizations, led to the development of a strong public health sector. By the 1950s, however, as the terms of trade for primary commodities deteriorated in the world market, Ceylon’s newly independent government could not sustain investments in welfare. In response to the economic downturn, the state adopted a strategy of intensifying the use of existing public healthcare resources, while reducing capital investment in the health sector.

The private health sector ran in parallel under colonialism and in the aftermath of independence. Catering to a privileged urban clientele, the private health sector relied on the state for its physician workforce. Attempts to regulate private practice resulted in the creation of ‘channel centres,’ which play a critical role in the delivery of healthcare to this day.

29 As Target 8 of SDG 3, ‘Universal Health Coverage’ has been framed broadly to encompass “financial risk protection, access to quality essential health-care services and access to safe, effective, quality and affordable essential medicines and vaccines for all” (Sustainable Development Knowledge Platform 2016).

The state did not actively support private healthcare expansion until the adoption of an ‘open economy’ in 1977 when the incumbent government supported private healthcare expansion by deregulating private practice and providing incentives for the establishment of private healthcare facilities. In the 1990s, the government opened the health sector for private investment through the Board of Investment, granting numerous fiscal concessions to the private healthcare industry. The end of the civil war heralded a massive wave of liberalization with national policy frameworks laying out explicit plans to privatize healthcare.

These health sector developments in Sri Lanka have run in parallel with the shift toward privatization in the age of neoliberal global capitalism. In this context, the durability of its ‘free health’ policy is remarkable and may perhaps be explained by its ascent to the national consciousness. Although politicians and policymakers have been reluctant to dismantle the public system, state-supported private market expansion continues with the health sector identified as a key area for foreign investment. Health insurance and ‘public-private partnerships’ enjoys wide consensus within policy circles, dovetailing with the interests of the medical establishment, healthcare industry, and transnational capitalist classes. As we shall see in Chapter 4, however, little evidence exists in support of healthcare privatization in the quest for universal access.

4. Healthcare Privatization and its Implications for Access in LMICs: A Critical Review of the Empirical Literature

Chapter overview: In this chapter, I review the empirical literature on healthcare access in LMICs (including Sri Lanka) to examine the implications of healthcare privatization for users. I find that much of the literature on healthcare access focuses on estimating out-of-pocket spending, assessing quality of public and/or private services, and measuring (in)equity in access to healthcare. I end by drawing attention to the gaps in this literature, particularly the dearth of empirical studies that explore the ways in which users experience and negotiate healthcare within the ‘mixed health systems’ of LMICs.

4.1 Introduction

The notion of a ‘mixed health system’ has gained acceptance in the global health arena in parallel to the adoption of ‘Universal Health Coverage’ as a guiding framework for health sector development in LMICs (Barnett and Hort 2013; Nishtar 2010; The Lancet 2016). Nishtar (2010:74) defines a mixed health system as “a health system in which out-of-pocket payments and market provision of services predominate as a means of financing and providing services in an environment where publicly-financed government health delivery coexists with privately-financed market delivery.” A direct consequence of welfare retrenchment, weakening public healthcare systems and concomitant private sector expansion, mixed health systems of LMICs are fragmented with the lines between public and private, blurred (McPake and Hanson 2016).

A multitude of public and private actors are involved in healthcare delivery within the mixed health systems of LMICs, including public facilities (clinics, health centres, and hospitals), private individual for-profit providers (formal and informal), private non-profit providers, large-scale commercial hospitals, including multinationals and domestic healthcare companies, and public- private partnerships. Financing arrangements that involve varying degrees of public and private spending add complexity to this picture. Moreover, users in LMICs frequently spend out-of-pocket in both public and private sectors, and dual practice by healthcare providers is common (Basu et al. 2012; McPake and Hanson 2016; McPake et al. 2016).

Acknowledging this diversity of mixed health systems, here I review recent empirical work to understand what healthcare privatization means for access in LMICs. Considering their shared histories of colonialism, I limit my review to studies located in Asia, Africa and Latin America, and

35 36 present my findings in relation to three key dimensions: 1) out-of-pocket spending; 2) quality of care; and 3) inequity.

4.2 Privatization and out-of-pocket spending

Much of the empirical literature on healthcare access focuses on out-of-pocket spending at the household level. Used widely in the health economics literature, out-of-pocket spending encompasses direct non-reimbursable payments, which include user-fees, gratuities or in-kind payments to healthcare providers, and payments for pharmaceuticals, diagnostics, medical devices, and other health or health-promoting services. These payments may go toward public or private healthcare services and may include transport costs and over-the counter purchases (Measure Evaluation n.d.). Although a component of private spending, and a burden on users, health insurance premiums, copayments, and deductibles are not included in the World Bank’s definition of out-of-pocket spending (World Bank 2017b),30 dovetailing with its promotion of private health insurance as an alternative to direct out-of-pocket payments (World Bank 1993).

User-fees are routinely charged by private healthcare facilities, and widely implemented within the public systems of LMICs (McIntyre et al. 2006). Third World governments introduced user- fees for public services to recover costs and curb public spending under structural adjustment. User-fees became a legitimate healthcare financing strategy under the 1987 Bamako Initiative, backed by African ministers of health to increase government revenue for healthcare financing on the continent (Birn et al. 2017). User charges for public services remain in place today in many LMICs, and constitute a significant burden for economically disadvantaged users (Akinkugbe et al. 2012; Leive and Xu 2008; Perkins et al. 2009; Saksena et al. 2012).

Reflecting the emphasis placed on maternal and reproductive health by donor agencies, numerous studies estimate the out-of-pocket spending associated with accessing these services. In Bangladesh and in China, prohibitive user charges impeded access to essential obstetric services (Afsana 2004; Gao et al. 2010; Kaufman and Jing 2002), while those for perinatal procedures burdened users in other settings, including Benin, Ghana, Madagascar (Honda,

30The World Bank defines out-of-pocket spending as “any direct outlay by households, including gratuities and in-kind payments, to health practitioners and suppliers of pharmaceuticals, therapeutic appliances, and other goods and services whose primary intent is to contribute to the restoration or enhancement of the health status of individuals or population groups” (World Bank 2017b).

Randaoharison and Matsui 2011), Burkina Faso, Kenya and Tanzania (Perkins et al. 2009; Storeng et al. 2008), Sierra Leone (Oyerinde et al. 2012), Indonesia (Quayyum et al. 2010), and Nepal (Borghi et al. 2006). In other words, resource-intensive maternal and child health services remain economically inaccessible to many in low- and middle-income settings.

Catastrophic health expenditure, defined variously as out-of-pocket spending on healthcare ranging anywhere from 5 to 40 per cent of household income (Xu et al. 2003),31 is widely used in the health economics literature as an indicator of the extent to which households are burdened by out-of-pocket spending. National level studies and cross-national comparisons suggest that catastrophic spending incurred in both public and private sectors pushes users into poverty and debt (Blas and Limbambala 2001; Kruk, Goldmann and Galea 2009; McIntyre et al. 2006; Storeng et al. 2008; Van Doorslaer et al. 2006; Wagner et al. 2011).

Household contributions to health insurance schemes constitute a substantial share of private spending. Although national insurance schemes expand service coverage and reduce vulnerability to poverty and debt (Danese-de los Santos, Sosa-Rubi and Valencia-Mendoza 2011; Knaul et al. 2012; Liu et al. 2002; Meng et al. 2012; Plaza, Barona and Hearst 2001; Ruiz et al. 2007), premiums and copayments pose barriers, particularly for lower-income households (Alvarez, Salmon and Swartzman 2011; Ekman 2007; Erus and Aktakke 2012; Gao et al. 2001; Meng et al. 2012; Onwujekwe, Hanson and Uzochukwu et al. 2012; Patel et al. 2015).

Strong public healthcare systems alleviate the financial burden of healthcare spending for households (Mackintosh et al. 2016; Wagner et al. 2011). This is best exemplified by Cuba’s free public system in which out-of-pocket spending is virtually non-existent (Gericke 2005). Thailand, a second example, has recorded declines in out-of-pocket spending since the introduction of universal health insurance in 2001 (Damrongplasit and Melnick 2009). Its success is attributed to a publicly financed single-payer model preceded by decades of state investment in rural health care infrastructure (Kongsri et al. 2011; Patcharanarumol et al. 2011).

Sri Lanka is often cited as an example of a poorer country that possesses a strong public healthcare system (Mackintosh et al. 2016). Out-of-pocket spending in the public health sector is minimal, particularly for maternity services (Ensor and Ronoh 2005). However, users are

31 There is no standard definition for catastrophic health expenditure (Xu et al. 2003).

38 compelled to pay for private services owing to crippling resource constraints in the public system (Perera et al. 2007). Nevertheless, catastrophic expenditure is comparatively low because hospitalization and other resource-intensive services are mostly accessed in the public sector (Govindaraj et al. 2014). Catastrophic spending – when defined as over 40 per cent of non-food expenditures – tends to burden wealthier households in Sri Lanka, perhaps because they can afford to dispose of a larger share of income on health. However, low-income households are more vulnerable to impoverishment through out-of-pocket spending (Cavagnero and Govindaraj 2012 as cited in Govindaraj et al. 2014). On the other hand, private spending on health insurance premiums and co-payments is minimal owing to limited penetration of the private health insurance market (Institute for Health Policy 2015).

Taken together, user-charges are a significant source of out-of-pocket spending on healthcare in most LMICs. Health insurance may alleviate household spending on healthcare in some settings, particularly where universal coverage is financed by the public purse. However, where private health insurance predominates, private spending constitutes a significant financial burden for users. The public system in Sri Lanka is associated with less out-of-pocket spending, although users incur expenses owing to deficits in public healthcare services. For the most part, this research estimates out-of-pocket spending without distinguishing between public and private sectors, making comparisons difficult. Moreover, the use of quantitative measures tends to make invisible the implications of out-of-pocket payments for users in their everyday experiences of healthcare. In the next section, I review the evidence on quality of care.

4.3 Privatization and quality of care

Broadly speaking, quality of care has been considered in two domains in the health services literature: service quality and technical quality of care. Service quality generally addresses staff responsiveness and patient satisfaction, while technical quality encompasses competency of healthcare providers and adherence to clinical guidelines (Morgan, Ensor and Waters 2016). While WHO (2000) includes fairness and equity in its framework for evaluating the performance of health systems, this aspect is largely neglected in the literature on quality of care.

Two widely cited systematic reviews, which brought together evidence on quality of care in public and private health sectors in low- and middle-income settings, yielded different results (Berendes et al. 2011; Basu et al. 2012). Berendes and colleagues concluded that the private sector performs better in relation to “drug supply, responsiveness and effort,” but found no difference between

39 patient satisfaction and the competence of healthcare providers in the two sectors (p. e10000433). In contrast, Basu and colleagues found no evidence in support of the widely held belief that the private sector is “more efficient, accountable or medically effective than the public sector,” although their study did suggest the public sector “appears frequently to lack timeliness and hospitality towards patients” (p. e1001244). These divergent findings were to be expected as the two studies assessed different domains of quality of care. Berendes and colleagues limited their review to outpatient care and did not consider fairness and equity as a criterion, while Basu and colleagues included studies of both ambulatory and hospital-based care, and addressed fairness and equity in their assessment.

Lengthier wait-times and frequent delays have been highlighted as a feature of public healthcare systems in several LMICs (Huff-Rousselle and Pickering 2001; Lindelow and Serneels 2006; McLane et al. 2015; Ngo and Hill 2011; Paphassarang et al. 2002). Indeed, the time costs associated with public healthcare are a key reason for users to switch to the private sector. In Vietnam and Lao PDR, for example, women accessed private healthcare to avoid the long wait times and short consultations at public facilities (Ngo and Hill 2011; Paphassarang et al. 2002). Similarly, Cambodian women preferred to access reproductive health services at subsidized NGO facilities where wait times were shorter compared with the public sector (Huff-Rousselle and Pickering 2001).

Unpleasant interactions with public sector healthcare providers are reported from a variety of contexts, including Uganda (Amooti-Kaguna and Nuwaha 2000), Kenya (Keesara et al. 2015), Lao PDR (Paphassarang et al. 2002), Thailand (Pongsupap and Lerberghe 2006), Pakistan (Mumtaz et al. 2014; Siddiqi et al. 2002) and Turkey (Turan et al. 2006). In Lao PDR, unwelcoming health workers compelled users, including those representing economically disadvantaged groups, to seek private healthcare (Paphassarang et al. 2002). Women in Kenya opted for private family planning services for the same reason, despite believing that public sector services were technically superior (Keesara et al. 2015). These findings are not surprising as public systems in most LMICs are grossly underfunded comprising under-staffed facilities run by over-worked providers receiving insufficient remuneration (Basu et al. 2012; Birn et al. 2017).

Despite the numerous deficits in public service, studies of technical quality of care suggest that the public sector performs better in various contexts, including Mexico (Barber 2006), Vietnam (Tuan et al. 2005), and Pakistan (Siddiqi et al. 2002). A tendency to cut costs and maximize profits results in poor adherence to clinical guidelines within privatized systems (Hoa et al. 2011; Jacobs

40 et al. 2004; Pongsupap and Lerberghe 2006; Schneider et al. 2001; Siddiqi et al. 2002). Moreover, the mechanisms in place for surveillance and follow-up are weak in the private sector, at times, leading to inferior treatment outcomes. In Thailand, for example, tuberculosis treatment outcomes were worse in the private sector (Chengsorn et al. 2009). And, in Botswana, the management and follow-up of people living with HIV was substandard at private clinics (Bisson et al. 2006).

On the other hand, the availability of medicines and medical supplies is reportedly superior at private facilities (Basu et al. 2012; Berendes et al. 2011; Shayo et al 2016), although it remains unclear whether such availability yields better health outcomes (Basu et al. 2012). A key concern is over-prescribing in private facilities as demonstrated in Vietnam (Hoa et al. 2007), Thailand (Pongsupap and Lerberghe 2006), Pakistan (Siddiqi et al. 2002), and India (Gupta et al. 2009). An accompanying concern is the poor quality of drugs and widespread availability of counterfeit medicines at private pharmacies (Onwujekwe et al. 2009; Newton et al. 2011).

These differences in quality of care may partly be explained by the adoption of market-based financing strategies alongside privatization. The separation of purchasing and providing functions has necessitated fee-for-service remuneration systems, in turn, leading to overtreatment and unnecessary medical procedures. Caesarean section rates in a variety of LMICs, particularly in Latin America, are higher in the private sector for this reason (Arrieta 2011; Gonzalez-Perez et al. 2001; Murray and Elston 2005; Potter et al. 2001; Triunfo and Rossi 2009; Villar et al. 2006). In Gujarat, India, caesarean section rates rose sharply after a fully state-funded ‘public-private partnership’ was introduced to expand delivery care through the engagement of private providers (De Costa et al. 2014). In South Korea, market-based health reforms saw a rise in contracted healthcare workers and an increase in the number of inpatients per healthcare worker with adverse implications for quality of care (Oh et al. 2011).

Weak or absent regulatory mechanisms compromise quality of care in both public and private sectors in LMICs (Bhate-Deosthali, Khatri and Wagle 2011; Rashid et al. 2011; Varman and Vikas 2007). Weak regulation has led to the emergence of questionable modes of income generation on the part of healthcare providers, such as widely prevalent physicians’ dual practice. Some forms of dual practice are legislated, and have contributed toward retaining physicians within public systems, but potential conflicts of interest remain unaddressed (Ferrinho et al. 2004; McPake et al. 2016). In addition, dual practice has resulted in rising levels of absenteeism among public sector physicians, while functioning as a lucrative source of private income (Hipgrave and Hort 2014; Lindelow and Serneels 2006; Moghri et al. 2017).

Despite such broad trends, it would seem that quality of care varies substantially by context. As McPake and Hanson (2016) observe, any assessment of quality of care in the public and private health sectors of LMICs would necessarily depend on their relative strengths and weaknesses. Where privatization has been more aggressive, the quality of public healthcare may be poorer. In Brazil, for example, public sector obstetric services are inferior in technical quality compared with those available in the private sector (Victora et al. 2010). While public sector healthcare providers perform worse in all aspects of quality in relation to maternity and prenatal care in Tanzania (Boller et al. 2003), prescription practices in Uganda are inappropriate in both public and private sectors with the former performing significantly worse (Ogwal-Okeng et al. 2004).

Notwithstanding this large body of work from LMICs, quality of care remains understudied in Sri Lanka. The single study that compared quality of care in public and private sectors yielded trends similar to those found in other LMICs. By observing patient-provider interactions and conducting exit polls, Rannan-Eliya and colleagues (2015a and 2015b) found that the private sector performs better in all non-clinical aspects of quality of care. These findings are consistent with those of two qualitative studies in which users reported lengthy wait-times and delays in the public sector (Perera et al. 2007; Russell and Gilson 2006). Rannan-Eliya and colleagues (2015a and 2015b) also found that the public system outdoes the private sector in most clinical dimensions of inpatient care, but did not find any difference in the technical quality of outpatient care. Indeed, a study of health seeking behaviour in two low-income settlements in Colombo found that even the poorest households trusted the public system (Russell 2005).

Broadly speaking, the literature on quality of care suggests that some aspects of service quality are ‘superior’ in the private sector in LMICs. However, the public sector generally performs better in technical quality, although this may not be the case where public systems have been dismantled by aggressive privatization. While findings vary substantially by context, most researchers fail to locate the relative performance of public and private health sectors in the neoliberal cutbacks on healthcare spending that have critically influenced health sector development in LMICs. Neglecting this critical aspect may support widespread acceptance of the view that the public health sector is ‘inefficient’ and a waste of public resources. In the next section, I examine the implications of healthcare privatization for (in)equity of access.

4.4 Privatization and (in)equity

The WHO (2017a) defines equity as “the absence of avoidable or remediable differences among groups of people, whether those groups are defined socially, economically, demographically, or geographically.” Equity of access to healthcare has been studied primarily in relation to income inequality. This research establishes that low-income households are disproportionately burdened by private healthcare spending, inclusive of user-fees, insurance premiums and copayments (Akazili et al. 2012; Borghi et al. 2003; Cissé, Luchini and Moatti 2007; Danese-de los Santos et al. 2011; Gao et al. 2010; Lopez-Cevallos and Chi 2010; Mills et al. 2012; Mtei et al. 2012; Zere et al. 2007).

Health insurance coverage is grossly inequitable in LMICs, particularly where a combination of public and private ‘risk sharing’ arrangements exists. Government-subsidized health insurance schemes for low-income welfare beneficiaries lack comprehensiveness, while contributory plans for the formally employed offer superior coverage (Birn et al. 2016). Such inequitable coverage is reported from a variety of contexts, including India (Shahrawat and Rao 2012), Mexico (Kierans et al. 2013), Colombia (Alvarez et al. 2011), and Ecuador (Lopez-Cevallos and Chi 2010) where neither universality nor comprehensiveness has been achieved. Under such conditions, those with limited means may need to sell their assets and/or borrow to purchase healthcare for illness conditions that remain uncovered (Cavagnero et al. 2006; Kierans et al. 2013; Shahrawat and Rao 2012). Thailand is an exception where expansion of insurance coverage has improved comprehensiveness and advanced equity of access within the public system. Notably, Thailand’s national health insurance scheme follows a single-payer model and relies less on private health insurance (Hanvoravongchai 2013; Patcharanarumol et al. 2011).

Patterns of utilization of private healthcare vary considerably by context and among social groups (Benova et al. 2015; Bustreo, Harding and Axelsson 2003; Hotchkiss, Godha and Do 2014; Wang, Sulzbach and De 2011). Numerous studies have established that the private sector is frequented by economically disadvantaged users (Barber 2006; Ha, Berman and Larsen 2002; Makinen et al. 2000; Rashid et al. 2011; Saksena et al. 2012; Xu et al. 2006; Yoong et al. 2010), implying that the presence of private healthcare may contribute toward bridging inequity of access. However,

43 others have found that economically disadvantaged users tend to access informal32 healthcare providers (Afsana and Rashid 2001; Basu et al. 2012; Parkhurst, Rahman and Sengooba 2006; Rashid et al. 2011; Shaikh and Hatcher 2005; Thuan et al. 2008; Varman and Vikas 2007) whose services are appealing where public services of reasonable quality are unavailable (McPake and Hanson 2016).

Several multi-country studies demonstrate that the ‘formal’ private sector is accessed mostly by better-off users (Alcock et al. 2015; Basu et al. 2012; Campbell et al. 2016; Saksena et al. 2012). This is especially the case where public systems are stronger and offer a safety net for low-income users. Under such conditions, private facilities cater to the wealthy as reflected in the usage of private hospitals by higher income quintiles in Thailand (Somkotra and Lagrada 2009) and Sri Lanka (Mackintosh et al. 2016; Pallegedara and Grimm 2017). This is also why larger-scale private facilities are located in urban areas where wealth is concentrated. In Tanzania, for example, private healthcare expansion has led to the establishment of facilities in urban areas and densely populated rural areas (Benson 2001).

Wealthier users are advantaged in ways besides their ability to pay for services. They are more likely to consult qualified practitioners (Alvarez et al. 2011; Gage 2007; Makinen et al. 2000; Perera et al. 2007; Paphassarang et al. 2002; Thuan et al. 2008), and receive specialized services, including advanced biomedical technologies and treatments, often in the private sector (Kierans et al. 2013; Liu et al. 2002). While higher-income groups consume a greater proportion of health services (Makinen et al. 2000; Thuan 2008; Zere et al. 2007), insurance enrolment and the distribution of benefits are also positively associated with household income (Akazili et al. 2012; Ataguba and McIntyre 2012; Jehu-Appiah et al. 2011; Mills et al. 2012; Mtei et al. 2012).

Class and/or wealth-based inequities in access are found to interlock with other dimensions of difference. Low-income indigenous women in rural Mexico received technically superior services in the public sector (Barber et al. 2007), while oppressed caste women from Pakistan opted for the private sector owing to their marginalization and stigmatization within public facilities (Mumtaz et al. 2014). Although women have benefitted from improved insurance coverage following health reforms in some contexts (Chankova et al. 2008; Ewig and Bello 2009), they are simultaneously

32 Sudhinaraset and colleagues (2013) propose the following criteria in their working definition of informal providers: lacking formally recognized training; receiving direct payments from users; unregistered and/or unregulated; lacking professional affiliations.

44 disadvantaged by out-of-pocket payments since they represent a greater proportion of the poor (Ewig and Bello 2009; Onah and Govender 2014). While few have drawn attention to interlocking dimensions of difference that shape access in LMICs (Lopez-Cevallos and Chi 2010; Shaokang et al. 2002), many of these studies do not distinguish between public and private health sectors.

At times studying more than one dimension of (in)equity may yield unexpected results. In China, targeted health insurance coverage for rural dwellers reduced inequity of access between rural and urban populations, while the same reforms exacerbated inequity within urban areas (Gao et al. 2001; Xiong et al. 2013; Xu and Short 2011). Moreover, children in urban areas of China had very low rates of coverage post-health reform (Xiong et al. 2013). In Colombia, national health insurance has improved coverage for the insured, while widening the gap between the insured and uninsured (Ewig and Bello 2009; Ruiz et al. 2007). Similarly, targeted health insurance increased obstetric service coverage for the lowest income group in Indonesia, but failed to cover those with marginally higher incomes (Quayyum et al. 2009).

In Sri Lanka, inequities in access have been studied primarily in relation to income level and geographic location. Economically disadvantaged households rely mostly on the public system (Agampodi and Amarasinghe 2007; Perera et al. 2007; Russell and Gilson 2006), and struggle to meet the out-of-pocket payments associated with public healthcare (Perera et al. 2007; Russell and Gilson 2006). Meanwhile, rural users are compelled to travel long distances to access public healthcare or instead use private services because rural healthcare facilities are underequipped and lack basic medicines and diagnostic services (Perera et al. 2007).

Interlocking dimensions of inequity in the context of healthcare access have received very little attention in Sri Lanka. My literature search yielded only two studies on healthcare access that included ethnicity as a variable. First, Agampodi and Amarasinghe (2007) found that Tamil users in Colombo preferred to access private immunization services, while Muslim users accessed the public system. While the researchers did not speculate on the high rates of public healthcare use among Muslims, these findings need to be interpreted with caution as interaction between ethnicity and income level were not considered in the study. A second study by Rannan-Eliya and colleagues (2015a) found no difference in non-clinical dimensions of quality of (outpatient) care by ethnicity. However, they employed direct observation and exit interviews, methods that may not be conducive to eliciting sensitive concerns related to ethnicity.

In sum, equity of access has been studied primarily in relation to income. There is abundant evidence suggesting that privatization compromises access for economically disadvantaged users, while privileging wealthier users in multiple ways. While interlocking dimensions of inequity have received less attention in studies of healthcare access/privatization, exploring such layered experiences within health systems may yield critical knowledge on inequity in Sri Lanka and other LMICs. In what follows, I summarize the reviewed work and highlight gaps in this literature.

4.5 Summary and gaps in the literature

In this chapter, I critically reviewed empirical literature on healthcare access to understand the implications of healthcare privatization for users in LMICs. My review yielded three broad findings: first, users spend more on healthcare as health services are opened to the market; second, privatization compromises technical quality of care, although some aspects of service quality may improve under ‘market competition’; third, privatization widens income-based inequity and urban- rural disparities in access to health services. Not coincidentally, much of the evidence in support of privatization comes from Sub-Saharan Africa where aggressive privatization has dismantled public systems (Turshen 1999).

The literature on healthcare access in Sri Lanka is not substantial, but suggests that the public system protects economically disadvantaged users from financial hardship. Creeping privatization has increased out-of-pocket spending on healthcare as users visit private facilities to address gaps in public services. Similar to other LMICs, comparisons of quality of care suggest that the private sector performs better in service dimensions of quality, while the public system provides technically superior services. While very little is known about equity of access in the private sector, except that private hospitals are frequented by the wealthiest quintile, extant research deals with class- and geographically-based disparities in access to (public) healthcare. Interlocking dimensions of inequity remain understudied with very little attention to ethnicity, a notable omission given the protracted ethnic conflict that has ravaged Sri Lanka.

Much of the reviewed work comprises quantitative studies of catastrophic health expenditure, impoverishing health expenditures, or other quantitative measures of (in)equity. Excessive reliance on large secondary data sets and/or primary surveys makes invisible the experiences of users accessing healthcare. Although some studies of access and quality of care have their basis in qualitative research methodologies (Afsana 2004; Alvarez et al. 2011; Benson 2001; Gao et al. 2010; Kaufman and Jing 2002; Kierans 2013; Perera et al. 2007; Varman and Vikas 2007), they

46 focus narrowly on household spending and/or barriers to access. Studies that explore the implications of privatization for women are limited to addressing barriers to maternal and reproductive health services (Afsana 2004; Gao et al. 2010; Kaufman and Jing 2002).

Taken together, the reviewed work largely overlooks the ways in which healthcare privatization shapes the everyday lives of users accessing healthcare. It pays little attention to how users, and particularly women, negotiate privatization in LMICs. Strikingly few studies advance a critique of privatization as a policy direction or address its ideological basis. For the most part, the macrostructures and processes that shape health reform in LMICs (Armada and Muntaner 2004; Bakker and Gill 2003a; Birn, Zimmerman and Garfield 2000; Desclaux 2004; Ewig 2008; Paluzzi 2004; Pfeiffer 2004; Pfeiffer and Chapman 2010; Qadeer and Visvanathan 2004; Turshen 1999; Unger et al. 2008) remain unaddressed in this empirical work.

To address these gaps, this study uses a Third World Marxist feminist methodology to explore women’s experiences of healthcare in Kandy, Sri Lanka. I work with women positioned differently, in terms of social class and ethnicity, to paint a picture of healthcare access in a microcosm of society, Udawatta. I shed light on how neoliberal processes impact on women’s everyday life, drawing attention to the ways in which they struggle, both as individuals and as a collective, to access healthcare in the face of welfare retrenchment. Having located my study in the political economy of healthcare in Sri Lanka, and highlighted gaps in the empirical literature on healthcare access in LMICs (and Sri Lanka), I now describe how I carried out my research.

5. Exploring Women’s Experiences of Healthcare in Kandy: From Theory to Methods

Chapter overview: This chapter provides an account of how I conceptualized, planned, and implemented my research. I begin by locating my methodological approach in my experiences as a physician in the public health sector in Sri Lanka and graduate student in the field of ‘global health’ in North America. I then situate my project in the critical qualitative research paradigm, before outlining the methods I used for participant recruitment, data generation, and analysis. I end with a brief discussion of the trustworthiness of this work.

5.1 Locating myself in this work

My graduate studies took me from my position as a Medical Officer33 at the Department of Anaesthesiology and Intensive Care at the Provincial General Hospital Badulla, located in the heart of the (tea) plantation region in southeastern Sri Lanka, to Boston, and then, Toronto. Spanning a period of 13 years, this journey profoundly influenced the conceptualization of my PhD research.

As a middle-class doctor with feminist sensibilities, I was struck by the challenges women encountered within Sri Lanka’s public healthcare system. Perhaps owing to my mixed-ethnic heritage and disconnection from dominant narratives of Sinhala and Tamil nationalisms, I was sensitive to the plight of ‘other’ ethnic minorities, particularly Muslim and up-country Tamil women who represented a large proportion of healthcare users the hospital served. Hoping to explore my interests in women’s health and equity in healthcare, I decided to pursue a Master’s in public health, which promised the right mix of health and social sciences.

I entered a two-year Master’s programme in Global Health and Population at the Harvard School of Public Health.34 Knowing nothing about the fraught histories of ‘global health’ or ‘population’ (Birn et al. 2017; Kumar et al. 2016), I anticipated learning about the functioning of successful health systems in the West, which I presumed delivered superior (women’s) health services. While the medical curriculum in Sri Lanka had taught me very little about the legacies of colonial

33 The term used to designate a non-specialist medical doctor in the public health sector of Sri Lanka. 34 At the time, equivalent programmes in public health were not available in Sri Lanka. Some of the relevant content was addressed in the curriculum of the Master of Science in Community Medicine, offered by the Post-Graduate Institute of Medicine, University of Colombo. However, this program did not sufficiently speak to my interest in combining public health with the social sciences.

47 48 and imperial medicine (Birn et al. 2017), I had till then uncritically accepted global health constructs like ‘tropical medicine,’ ‘maternal and child health’ and ‘developing countries.’

At Harvard, the neocolonial architecture of global health was soon laid bare in a curriculum that devoted a substantial share of credits to health economics and demography. I was dismayed to learn about ‘cost-efficiency,’ ‘disability-adjusted life years’ and other concepts that normalized different standards for resource allocation in LMICs. Ironically, I had to leave Sri Lanka to understand that the ‘free’ public healthcare system I had worked in upended health reform models advocated by global health ‘experts’ for ‘developing countries.’

From Harvard I came to the University of Toronto where I entered a PhD program in social and behavioural health sciences. My interest in women’s health/healthcare and Third World feminisms led me to explore broader questions of inequality, which in turn, brought to the fore concerns of class, gender, race, and colonialism. Having immersed myself in the literature on healthcare access in LMICs, I began to see parallels between the neoliberal transformations taking place in healthcare in Sri Lanka and other Third World contexts. In my PhD research, I decided to explore access in the context of healthcare privatization in Sri Lanka. I now locate this work in the critical qualitative paradigm.

5.2 Situating my research in the critical qualitative research paradigm

The critical interpretive tradition encompasses a diverse set of approaches ranging from feminist emancipatory methodologies and critical ethnography to deconstruction, discourse analysis, and action research (Giacomini 2010; Guba and Lincoln 2005). This study employs a Third World Marxist feminist methodology, which locates the emancipatory demands of historical materialism and its aspirations for social change within a decolonizing feminist framework. Bringing together Western Marxist feminist and Third World feminist approaches allows for a more complex understanding of women’s oppression as shaped by class, gender, race/ethnicity, and the enduring colonial condition (Herr 2014).

Drawing on the work of Mohanty (2003:9), I delve into women’s practices of using healthcare in Kandy to shed light on “the specificities of global capitalism” by “[naming] and [demystifying] its effects in everyday life.” Conceptualizing healthcare access as historically specific, and shaped by social relations, I employ McIntyre, Thiede, and Birch’s (2009) multidimensional understanding of access, which addresses three health system features that enable individuals, households, and

49 communities, to attend to their healthcare needs. These are: availability (physical access; includes quality of care), affordability (financial access), and acceptability (cultural access) of health services (p. 183-4). By focusing on women’s stories of accessing healthcare, I seek to transcend conventional health economics measures of access/equity (e.g., catastrophic health expenditure, impoverishing health expenditure, Kakwani Index, etc.), which are at once economistic and dehumanizing. By focusing on healthcare access as an overarching concept, I try to subvert the analytic reduction of women’s healthcare to maternal and reproductive health (Kumar et al. 2016).

5.3 Study setting

Sri Lanka is an island-nation located off the southern coast of India, seven degrees north of the equator. Approximately 65,000 square kilometers in size, the country is made up of 9 provinces and 25 districts, one of which is the centrally located mountainous Kandy District, where the study is set (Figure 1).

Figure 1. Sri Lanka: Central Province and Kandy District. Source: Wikipedia 2017

In 2012, when the last national census was carried out, the country’s population was just over 20 million. Less than a fifth (18%) of the population reside in urban areas. The rural and estate

50 sectors,35 where agriculture is the primary livelihood, comprise 77 and 4 per cent of the population, respectively (Department of Census and Statistics 2015a).

Kandy District is one of three districts that make up the Central Province. About 7 per cent of the country’s population lives in the Kandy District where the urban sector comprises about 12 per cent of the District’s population (Department of Census and Statistics 2015a).36 Average household incomes in the Kandy District are substantially less than in the Colombo District, and similar to aggregate national figures (Department of Census and Statistics 2015b). Income inequality, as represented by the proportion of monthly household income shared by the ‘poorest’ and ‘richest’ 20 per cent of households, is widespread. Household spending on ‘personal care and health’ is less in the Kandy District compared to Colombo District, perhaps reflecting the lower rates of consumption of private healthcare services in Kandy (Table 1).

Table 1. Income and health expenditure in Sri Lanka and Kandy District, 2012/13

Kandy Colombo Sri Lanka

Median monthly household 30,371 [253] 50,071 [417] 30,814 [257] income - SLR [~CAD]

Share of income received by 4.9 5 4.5 the poorest 20% - %

Share of income received by 50.8 53.9 52.9 the richest 20% - %

Average monthly household 2,044 [17] 3,448 [29] 2,181 [18] expenditure on personal care and health - SLR [~CAD]

Notes. SLR = Sri Lankan Rupee; CAD = Canadian Dollars; Source: Household Income and Expenditure Survey 2012/13 (Department of Census and Statistics 2015b)

The ethnic breakdown in Kandy District roughly corresponds to that of the country, although Sri Lankan Moors (or Muslims) make up the largest minority community in the Kandy District (Table

35 The 2012 Census defines the three sectors as follows: “Urban sector is made up of Municipal council and Town council areas. Estate sector is comprised of commercial lands that are in extent [sic] of 20 acres or above and where more than 10 labourers are employed. All such commercial lands are defined as estates and those form the estate sector in Sri Lanka. Rural sector is comprised of all other areas that do not come under the above two sectors” (Department of Census and Statistics 2015a:50). 36 The urban sector includes all municipal and town council areas. However, the Census suggests that these categories need redefinition owing to “the existence of areas with urbanized characteristics in the country that do not fall within the Municipal or Town council areas” (Department of Census and Statistics 2015a:50-1).

2). Together, about a quarter of the population in Sri Lanka and the Kandy District comprise Tamil- speaking minorities, i.e., Sri Lankan Tamils, Sri Lankan Moors (Muslims), and ‘Indian Tamils’ (Department of Census and Statistics 2015a).

Table 2. Population of Sri Lanka and Kandy District by ethnicity, 2012

Ethnicity Proportion of the population Proportion of the population in Kandy District (%) in Sri Lanka (%)

Sinhala 74 75 Sri Lankan Moor (Muslim) 14 9 Sri Lankan Tamil 5 11 ‘Indian Tamil’ 6 4 Malay <1 <1 Burgher <1 <1 Other <1 <1

Source: 2012 Census (Department of Census and Statistics 2015a)

Kandy District is divided into 20 administrative divisional secretariat divisions, one of which is the Kandy Four Gravets Division, home to the city of Kandy (Ministry of Public Administration and Home Affairs 2011a). This Division is made up of 65 grama niladhari divisions or village-level administrative areas, including the research site: Udawatta Grama Niladhari Division37 (Ministry of Public Administration and Home Affairs 2011b).

The 2012 Census indicates that 4,499 individuals and 1135 households reside in the Udawatta Division. Forty-three per cent of residents 15 years or over are employed with a majority having reached at least secondary school (~ 85%) (Table 3). Over three-quarters (76%) of households live on land owned by a household member, while 17 per cent rent, and the remainder report encroachment (Department of Census and Statistics n.d.).

A majority of the households in the Division have access to pipe-borne water (94%) or a well (1.4%) within their premises with about 3 per cent relying on an external water source. Ninety- three per cent of households have a water-sealed toilet connected to a piped sewage system and

37 The name of the division is changed to protect the anonymity of participants.

98 per cent are connected to the national electricity grid (Department of Census and Statistics n.d.). While these figures roughly correspond to data reported at the national level (Department of Census and Statistics 2015a), the widespread availability of pipe-borne water, sewage systems, and electricity, also reflects the urban setting of the Division.

Table 3. Educational achievements of Udawatta residents38

Educational level Number (%) % Never been to school 57 1.4 Primary school 584 14.0 Secondary school 1158 28.0 Ordinary level 928 22.0 Advanced level 1026 25.0 Degree 418 10.0 Total population (5 years and older) 4171 100.4

Source: Census of Population and Housing 2012 (Department of Census and Statistics n.d.)

About 3 km from the city centre, the Udawatta Division is located in close proximity to several public and private healthcare facilities. Within the Division is a maternal and child health clinic, two public sector Ayurveda dispensaries, and several private general practitioner clinics. A public sector primary care facility offers walk-in services about a kilometre from the northwestern boundary of the Division. Three public sector tertiary care institutions and a facility dedicated to respiratory diseases are located within a four-kilometre radius. Numerous private providers operate from independent clinics, specialist consultation centres, and private hospitals just a short distance away from the Division.

The selection of the urban setting of Udawatta as my research site comes with some caveats. First, rural and plantation communities, who make up the most underserved social groups in the country, were excluded. However, I believe an urban division was best suited to explore my research topic given the concentration of public and private facilities. In contrast, a rural or plantation area would have had, at the most, a divisional hospital or primary medical care unit,39

38 Data on employment or occupation are not reported at the Grama Niladhari Division level (Department of Census and Statistics n.d.). 39 Divisional hospitals and primary medical care units deliver curative primary care services in the public system.

53 a maternal and child health clinic, and an Ayurveda dispensary representing the public sector with one or two allopathic private GP clinics and perhaps a private Ayurveda practitioner comprising the private sector (Weerasinghe and Fernando 2011).

Second, I have resided in Udawatta Division for over three decades. Being a resident granted a series of logistical advantages. I was already familiar with the geographical terrain and was ‘known’ to residents, particularly on the southern side of the Division. Where I was not known, an introduction from another resident permitted easy access as I was ‘from the area.’ While being ‘known’ facilitated the study’s smooth implementation and completion, my relative ‘insider’ status meant that I needed to interrogate my assumptions and alliances within the research site (Herr and Anderson 2005).

My research spanned two phases of fieldwork. Phase 1, an exploratory inquiry undertaken in Kandy between February and June 2014, contributed to the development and design of Phase 2, a community-based study in the Udawatta Division, conducted between January and September 2015. Both studies received ethics approval from the Research Ethics Board of the University of Toronto.40

5.4 Phase 1: Exploratory study

5.4.1 Objectives

The specific objectives of Phase 1 were to:

1) Better contextualize and develop my research proposal by accessing the knowledge and experience of Udawatta residents and a range of other stakeholders, including healthcare providers, administrators, government officials and policymakers; 2) Network and build trust with women from Udawatta and other stakeholders listed in 1);

40 I tried to apply for ethics approval from a local body – the Institutional Ethical Clearance Committee at the Faculty of Medicine, University of Peradeniya – but found that a ‘local supervisor’ was needed in order to do so. Since a ‘local’ ethics review was not required by the University of Toronto, and I was familiar with the context, I proceeded with my research. However, in my opinion, the latter should be a requirement for global health research as remote ethics review committees are unlikely to understand or be familiar with the social, economic, and cultural context in which the research is conducted. The absence of such a requirement at the University of Toronto reflects North-South power relations as Northern researchers generally encounter few challenges in conducting research in the South, while researchers based in the South struggle with visa requirements and other institutional barriers.

3) Become more familiar with the research site and define geographical boundaries for Phase 2; 4) Identify women residents of Udawatta willing to serve as members of a community-based advisory panel to advise me on the design and implementation of Phase 2; and 5) Identify other stakeholders (healthcare providers, health administrators, government officials and policymakers) willing to participate in Phase 2.

5.4.2 Summary of Phase 1 activities and findings

By the end of Phase 1, I achieved objectives 1 to 3 of my exploratory study. I accessed the expertise and experience of Udawatta residents and a range of stakeholders as planned (objective 1), and established relationships and engaged in trust building with these stakeholders (objective 2). From the outset, my residence in the Division offered a series of advantages. I was familiar with its geographic terrain and had little difficulty navigating the various neighbourhoods therein. Moreover, as a resident, I was able to access the administrative center of the Division – the Grama Niladhari Office – with ease. The Grama Niladhari Officer, the administrative head of the Division, welcomed me to his Office and supported my research activities. He introduced me to his staff as well as residents active in the Udawatta Village Development Society. The Samurdhi41 Officer, whose desk was located in the same premises, turned out to be a key resource in Phase 1.

I met with Udawatta residents and government officials serving the Division during Phase 1. I learned about the demographic makeup of the Division, its geographical boundaries, and the services available to residents. Village development activities took place through two Samurdhi societies. The Samurdhi Officer invited me to attend the combined monthly meetings42 of the two Samurdhi societies, and these interactions led to my establishing relationships with Samurdhi members.43 Although the membership represented low-income households, very few men

41 The Samurdhi Programme distributes government subsidies to low-income families through village-based Samurdhi societies (Samurdhi Authority 2016). 42 Regular attendance at these meetings is required for members to receive Samurdhi benefits. 43On the request of the Samurdhi Officer, I organized a health education seminar in the division. Although this was independent of my research activities, it gave me additional insights on the daily challenges negotiated by economically disadvantaged women accessing healthcare.

55 attended the meetings. These gatherings became an important point of access to economically disadvantaged women from Neluwa and Jagoda.44

I concurrently held meetings with regional health administrators at the Office of the Provincial Director of Health Services - Central Province, the Office of the Regional Director of Health Services - Kandy, and the Kandy Municipal Council, as well as public sector healthcare providers serving Udawatta Division (Appendix A). These encounters and the policy documents they led me to provided opportunities to understand the inner workings of the public healthcare system, and the challenges faced by regional administrators. At this stage, I obtained contact details from all participants with their consent to be contacted at a later date, if needed.

I gained familiarity with the research site and selected four neighbourhoods in which I planned to carry out my research activities (objective 3).45 Udawatta Division comprises a mishmash of villages and neighbourhoods divided on social class, ethnicity and landownership. The primary villages/areas are Udawatta, Neluwa, Medawatta, Dehioya, and Kirideniya (Figure 2). Residents of Udawatta represent diverse social groups, including: 1) working class families (primarily Neluwa and Jagoda); 2) land-owning middle class families (primarily Udawatta, Medawatta and Kirideniya); 3) middle/upper-middle class university academics and health professionals who moved into the area more recently to reside close to the University of Peradeniya and three tertiary care centres (primarily Kirideniya); 4) assorted families resettled in ‘housing schemes’ following the expansion of urban infrastructure (primarily Kirideniya); and 5) squatters46 (primarily Kirideniya). The Division includes several minority enclaves, including Bogoda, a Muslim neighbourhood in Medawatta, and Jagoda, a Muslim/Malay neighbourhood adjacent to/in47 Neluwa.48 For my purposes, I selected Neluwa, Jagoda, Bogoda, Dehioya and Kirideniya as my study areas.

44 Not all those eligible for Samurdhi are enrolled as Samurdhi recipients as there are caps on enrolment. The eligibility of low-income families for Samurdhi benefits must be established by the Samurdhi Officer. While there are allegations of political patronage through the Samurdhi programme(Kumarasinghe 2003), I decided to engage with the Samurdhi societies because alternative points of access, such as the village (Buddhist) temple, would have limited my interaction with minority ethnic groups. 45 I describe the selected neighbourhoods under Phase 2 as I made further changes after commencing the latter. 46 “A person who unlawfully occupies an uninhabited building or unused land” (Oxford University Press 2017). 47 Some participants spoke of Jagoda as a subarea of Neluwa, while others referred to Jagoda as a distinct neighbourhood in itself. 48 According to the Grama Niladhari Officer, Udawatta is home to very few Tamil residents.

Neluwa and Jagoda are working class neighbourhoods located adjacent to one another on the northern side of the Division. Neluwa is predominantly Sinhala, while residents of Jagoda mostly represent the Malay and Muslim communities. The living conditions in these two neighbourhoods were starkly different from those in Bogoda, Dehioya and Kirideniya. They were located on a hill where many homes lacked road access. Many homes were still under construction, although most residents had access to safe water and sanitary facilities. I expected users from this area to have significantly different experiences with healthcare compared with residents of other neighbourhoods as they resided a greater distance away from healthcare facilities and experienced greater social and economic disadvantage.

Figure 2. A map of Udawatta Division49

49 This map was obtained from the Udawatta Grama Niladhari Office with permission from the Grama Niladhari Officer. The neighbourhood demarcations and names were inserted by me.

Situated on the northeastern side of the Division, the Bogoda area is an ethnically diverse neighbourhood, comprising Sinhala, Muslim, and Malay families. The Muslim households were concentrated in an area within this neighbourhood. Although I initially designated Bogoda as a lower-middle class area with a mixed ethnic composition, I later selected the Muslim neighbourhood within Bogoda to represent the Bogoda study area. Road access was better in Bogoda compared with Neluwa and Jagoda, although roads were not paved. I observed modest dwellings with three-wheelers50 parked outside several homes. Many of the Muslim women I interacted with did not engage in income-generating activities outside their homes, although they had done so previously. I anticipated their healthcare experiences and/or practices would be shaped by the marginal location they occupied as Muslim healthcare users.

Dehioya shared similarities with Bogoda in terms of socioeconomic status. A notable difference was that many of the women from Dehioya worked outside their homes. I anticipated that they would enjoy some advantage in navigating healthcare given their proximity to Kirideniya where several medical professionals lived. Although Kirideniya was home to a wide social spectrum, it was clear that wealth was concentrated in this neighbourhood. I observed spacious and elaborately designed homes with well-maintained yards lining the sides of paved roads. Many of these homes had one or more cars or SUVs parked in private garages.

I did not achieve the fourth and fifth objectives as I made two substantial changes to my project at the end of Phase 1. First, I realised that coordinating meetings with a community-based advisory panel would prove difficult (see objective 4). The economically disadvantaged women I met with were very busy, many working informally,51 while caring for their families. Those who could contribute more time to my project were generally middle to upper-middle class women who had domestic aid at home. I abandoned the idea of an advisory panel, and, instead, identified five women I hoped to invite as community informants. I made this selection based on two criteria: interest expressed in my research and neighbourhood connectedness.

50 These vehicles are open on either side with a driver’s seat in front, and a seat behind for passengers. They are a common mode of transport in Sri Lanka although risky given their flimsy and open structure. The classed nature of ownership is reflected in the fact that I did not observe three-wheelers parked outside the homes of participants from Kirideniya, the middle/upper-middle class neighbourhood. 51 Informal employment refers to a “diversified set of economic activities, enterprises, jobs, and workers that are not regulated or protected by the state” (Women in Informal Employment: Globalizing and Organizing (WIEGO) 2017).

I identified five community informants: 1) an informally employed Sinhala woman from Neluwa, one of the two working class neighbourhoods, who was also the President of the Neluwa Samurdhi Society; 2) a Muslim homemaker from Jagoda, the other working class neighbourhood, home to Muslim/Malay ethnic minorities; 3) a Sinhala nurse from Bogoda, the lower-middle class and ethnically diverse village in Medawatta; 4) a self-employed Sinhala seamstress residing in Kirideniya, the middle-class and ethnically diverse neighbourhood; and 5) an academic of mixed ethnicity, also from Kirideniya. I selected two informants from Kirideniya to represent two different groups of middle-class households. The seamstress hailed from a Sinhala-speaking business/entrepreneurial background, while the academic belonged to an upper-middle class English-speaking professional constituency. The ability to use English in a social setting is a classed phenomenon, and I felt this group would contribute a distinct perspective to my study.

A second change in my protocol was to limit this research to exploring experiences of healthcare users (see objective 5). I could not recruit public sector health administrators and healthcare providers without prior authorization from the Provincial Department of Health Services – Central Province, in turn, contingent on ethics approval from the Ethical Clearance Committee of the Faculty of Medicine, University of Peradeniya. As I could not submit an ethics application without a co-supervisor at the University of Peradeniya, and owing to feasibility issues (i.e. completing the research in a timely manner), I decided to limit recruitment to healthcare users. Healthcare providers, health administrators, government officials and policymakers could form the core sample of a future research project that would build on the findings of the current study.

5.5 Phase 2: Community-based study

5.5.1 Research objectives

The primary objective of Phase 2 was to understand how the presence of private healthcare shapes access for women residents of Udawatta Division. Phase 2 was designed to achieve the following specific objectives:

1) To investigate where women go for healthcare when confronted with a health problem; 2) To explore women’s impressions of and experiences with the healthcare facilities and services they use to resolve their health concerns; 3) To explore the ways in which women access and navigate public and private healthcare systems; and

4) To examine how the above dimensions are shaped by social location (e.g., class, ethnicity).

5.5.2 Research participants and recruitment

On commencing Phase 2, I made changes to my selection of study areas. I merged the working class neighbourhoods of Neluwa and Jagoda into a single study area after residents questioned my motives for separating them by ethnicity. I restricted the Bogoda group to lower-middle class Muslim women to widen representation from ethnic minorities, and included Dehioya as a new study area to ensure representation from lower-middle class Sinhala women. I retained the two groups from Kirideniya as planned in Phase 1, one made up of Sinhala-speaking women and the other comprising bi-lingual women who communicated with me in English.

I stratified the study areas by social class and ethnicity. In assessing class location, I used a stratification approach that considered “individual attributes and conditions” (Wright 2009:102). Such an approach is widely used in epidemiological research, and according to Wright (2009:109), may be integrated into Marxist methodologies, which foreground “exploitation and domination within the fundamental class divisions in capitalist society.” In reality, none of the selected study areas were restricted to a particular social class (or ethnicity). However, I recruited women representing the designated class location from each neighbourhood. In doing so, I assessed class location based on insights from community informants and/or information shared by prospective participants regarding education or source of household income, and my observation of dwellings and household wealth. In some instances, I considered spousal vocation, and, as mentioned earlier, the ability to speak English.

I did not intend to limit participation to any particular ethnicity, except in the Bogoda area. I restricted the latter group to Muslim women because I expected ethnicity to add a unique dimension to their experience, and assumed they would feel more at ease addressing ethnicity- related questions among others who may have shared their experience. The Sinhala-speaking group from Kirideniya comprised only ethnic Sinhalese women because I did not meet residents representing ethnic minorities who ‘fit’ the designated class location and were available to participate. Table 4 lists the final study areas by social class and ethnicity.

I constructed two categories of participants for the purposes of my research: community informants and community residents.

Community informants: I made changes to my initial selection of community informants based on their availability and advice from my thesis supervisory committee. Of the five originally identified women, only three – the (Sinhala) President of the Neluwa Samurdhi Society, the (Muslim) homemaker from Jagoda, and the (Sinhala) seamstress from Kirideniya – retained their roles. Although Neluwa/Jagoda now constituted a single study area, I enlisted two informants to represent the two ethnically disparate neighbourhoods. I also recruited an informant from among the Muslim women from Bogoda, but did not in Dehioya as the latter was located adjacent to Kirideniya and I was ‘known’ to residents there. Given my social location and residence in Kirideniya, I did not recruit an informant to represent the professional constituency from Kirideniya.

I used the contact details collected in Phase 1 to invite three (of four) informants to participate in the study (Appendices B and C). The fourth from Bogoda was introduced to me by the Samurdhi Officer in Phase 2. The community informants were all middle-aged women who were well connected in their respective neighbourhoods and communities (Appendix D). I relied on them to introduce me to potential participants in the initial stages of the study. With time, however, I found that other residents also helped me connect with prospective participants. As the line between informants and residents became blurred, I decided not to distinguish between the two categories in my analysis.

Table 4. Selected study areas of Udawatta by social class and ethnicity

No. Study area Social class Ethnicity

1 Jagoda/Neluwa Working class Malay, Muslim, Sinhala 2 Bogoda Lower-middle class Muslim 3 Dehioya Lower-middle class Sinhala, Muslim 4 Kirideniya – Sinhala Middle to upper-middle class Sinhala 5 Kirideniya – English Upper-middle class Sinhala, Sri Lankan Tamil, Mixed

Community residents: I met women, older than 18 years, who had resided in Udawatta for at least 2 years, via introductions from informants and/or other residents.52 In Kirideniya and Dehioya, I

52 I anticipated needing help from a Tamil-speaking research assistant for the recruitment of Muslim women, but soon discovered that the Muslim women I met were fluent in Sinhala. As I had expected, owing to the unavailability of Malay instruction in schools, the Malay women were also fluent in Sinhala.

61 used my social networks to access potential participants. Apart from class and ethnic considerations, I tried to cover a diverse range of healthcare experiences in my selection. I visited women at their homes, alone, or accompanied by an informant or resident. During these visits, I described the study and tried to gauge their interest and potential contribution. If they were interested, I obtained their contact details and signed consent to be contacted later (Appendices E and F). In this way, I obtained introductions to a larger group from which I would recruit participants for the focus groups and interviews. In total, 36 community residents participated in the study. As recruitment procedures were different for focus groups, individual interviews, and the final meeting, I describe each separately below.

Focus groups. I selected focus group members purposively from the larger pool of prospective participants. My primary intention was to construct focus groups representing diverse class and ethnic backgrounds, each of which was sufficiently homogenous to foster debate and discussion (Hennink 2007). I did not restrict participation to any particular ethnicity, except in Bogoda.

Most women expressed interest in joining a focus group, but scheduling a time that suited all was challenging given their busy schedules at home and at work. Thus, participation was limited to 30 women and over-represented by those who identified as homemakers, ‘unemployed,’53 or retired. Most women were reluctant to spend more than two hours at a focus group. To devote maximum time to discussion, I conducted the informed consent procedure on the preceding day at the homes of prospective participants. Accompanied by a research assistant,54 I visited each woman in her home and provided a copy of the information and consent letter (Appendix G) and sociodemographic data form (Appendix H) in her preferred language. I explained the contents of the informed consent letter, and asked each of them to bring the signed consent form along with the completed data form to the focus group. Travel to the community centre was reimbursed or transport provided for women living beyond walking distance and for all with mobility issues.

Each focus group comprised four to seven community residents. The final makeup of the five focus groups reflected the selected study areas: 1) working class Sinhala/Muslim/Malay residents from the Neluwa/Jagoda area; 2) lower-middle class Muslim women from Bogoda; 3) lower-

53 Many women who reported they were unemployed in the survey were informally engaged in various income- generating activities within their homes and devoted much time to social engagements in their respective communities. 54 I recruited a research assistant for Phase 2. She was a recent graduate from the University of Peradeniya with prior experience in conducting focus groups.

62 middle class Sinhala women from Dehioya; 4) middle to upper-middle class Sinhala-speaking women from business/entrepreneurial backgrounds from Kirideniya; and 5) upper-middle class Sinhala/Sri Lankan Tamil English-speaking women representing professional backgrounds from Kirideniya (Table 5).

Table 5. Overview of focus groups55

Study area Duration56 No. of Age Ethnicity Education57 Occupation (hours: participants range mins) (years)

Neluwa/ 1:20 7 35-66 Muslim – 1 Primary to Homemakers Jagoda Malay – 1 Ordinary Level and self- Sinhala – 5 (Grade 5 to 10) employed women Bogoda 1:00 4 33-55 Muslim – 4 Advanced Level Homemakers (Grade 12) Dehioya 1:25 5 33-56 Sinhala – 5 Ordinary Level Homemakers, (Grade 10) to self-employed Advanced Level women and (Grade 12) NGO workers Kirideniya 1:45 7 38-84 Sinhala – 7 Ordinary Level Homemakers, (S)58 (Grade 10) to self-employed Advanced Level women, NGO (Grade 12) workers and retired women Kirideniya 1:00 7 37-75 Sri Lankan Advanced Level Homemakers, (E) Tamil – 1 (Grade 12) to teachers, Sinhala – 5 PhD managers, Mixed – 1 business women, and

academics

In total, seven focus group participants hailed from the most disadvantaged study area and eight identified as ethnic minorities. Appendix I details the composition of each focus group.

55 In this table and others, I do not include reported household income for two reasons. First, several participants chose not to disclose their household incomes. Second, many who did underreported their incomes, that is reported incomes did not correspond to the reported occupations of household members. 56 This column represents the duration of the audio-recordings (rounded off to the nearest five). 57 North American equivalents: Primary = Grade 5; Ordinary Level = Grade 10; Advanced Level = Grade 12. 58 The Kirideniya (S) focus group was conducted in Sinhala and the Kirideniya (E) in English.

Individual interviews. In total, 23 women participated in the interviews. Of them, 13 were also focus group participants (Appendix J). I invited focus group members to an individual interview when I felt it was important to obtain a deeper understanding of a particular experience or idea shared during a group discussion, or when an experience I hoped would be discussed at the focus group was not (Hesse-Biber, Nagy and Leavy 2011). I also recruited women who were interested in taking part in the study but were not available to take part in a focus group.

The informed consent procedure (Appendix K) and sociodemographic data form (Appendix H) were completed before each interview. Recruitment continued until I was convinced I had data from a sufficiently wide spectrum of social locations, and sensed that I was hearing the themes that emerged for a second or third time. Six individual interviews were held in the most disadvantaged study area and eight women reported non-Sinhala ethnicities (Appendix J).

Final meeting. I invited all participants to a final meeting at the end of the data collection period. I reconnected with those who had consented to being contacted – by phone or in person – to invite them to attend the meeting (Appendix L). I visited each participant who had committed to attend at her home and conducted the informed consent procedure on the day prior to the meeting. I asked the women to bring the signed consent form to the meeting (Appendix M). Twenty-seven (out of 40) women attended the meeting with 100 per cent attendance from Bogoda and Dehioya (Table 6).

Table 6. Attendance at final meeting

Neighbourhood Total no. participants Attendance at final meeting from study area

Neluwa/Jagoda 11 7 Bogoda 4 4 Dehioya 6 6 Kirideniya (S) 9 5 Kirideniya (E) 10 5

Total 40 27

To encourage participation, I arranged transport for all women residing outside Bogoda, where the meeting was held.

5.5.3 Data generation

Rather than seeking to generate ‘objective’ data about healthcare or create empathetic rapport with participants to encourage unbiased confessions, I explored my research questions recognizing that data generation would be shaped by my worldview and social location (Herr and Anderson 2005; Roulston 2010). I had accessed the knowledge and experience of Udawatta residents (and others) in developing the interview guides and sociodemographic data form, but the tools were eventually created by me under the guidance of members of my thesis supervisory committee. In my interviews, I employed what Roulston (2010) refers to as a transformative approach, a method that seeks to harness the transformative potential of the research-participant interaction. This involved engaging women in dialog, sharing my own perspective, and addressing issues outside my frame of thinking (Roulston 2010).

Focus groups. I began with the Kirideniya (English) focus group, which represented women from my own neighbourhood of residence. I was already acquainted with most prospective participants and felt more at ease with this group, in light of my amateur moderating skills. The Dehioya, Neluwa/Jagoda, Kirideniya (Sinhala), and Bogoda focus groups followed, in that order.

All but one focus group was held at the Kirideniya Community Centre. The Bogoda focus group was conducted in the newly opened Udawatta Village Development Society Hall in Bogoda. Except for the Kirideniya focus group representing upper-middle class/middle class English- speaking women, all others were conducted in Sinhala. I facilitated all the focus groups; the research assistant was present at all but the Bogoda focus group. Her role was limited to welcoming focus group members, offering light refreshments, and ensuring that participants handed over their completed informed consent forms and sociodemographic data forms.

The duration of the focus groups ranged from approximately an hour to an hour and 45 minutes (Table 5). I audio-recorded the focus group discussions and made hand-written notes with the participants’ consent. Using a guide (Appendix N) helped me to facilitate the discussion in ways that balanced my interests as the researcher and what the women expressed as relevant to their experiences. This format did not allow me to cover all areas of the guide within the allotted two hours. As I progressed with the focus groups, I fine-tuned and changed the order of the questions.

The focus groups proceeded fairly smoothly as my moderating skills developed and I was able to draw in silent members. However, I later regretted having not held separate focus groups in

Neluwa and Jagoda as the Muslim/Malay women from Jagoda were relatively quiet during the focus group. Like others before me (Roulston 2010; Wilkinson 1998), I observed that the direction in which the discussion progressed was frequently beyond my control (unlike with the individual interviews). As a data generation method, then, focus groups did help to shift the balance of power away from myself, the researcher (Wilkinson 1998).

I compensated all women who participated in the focus groups for their time in kind. A token of appreciation consisting of household consumables valued at more than a single day’s labour for a female waged worker in Kandy (~CAD 5) was gifted to each member after the focus group.

Individual interviews. All but one interview took place in the women’s homes; I interviewed Sonali, a director of a family-owned business from Kirideniya, in her office. With one exception, Fathima, a retired clerk from Bogoda, the interviews took place after the relevant focus group.

The interviews lasted from 55 minutes to 2 hours and 20 minutes with the majority being approximately an hour and a half in duration. I audio-recorded the interviews and made hand- written notes with the participants’ consent. Sixteen of 22 interviews were conducted in Sinhala, and the rest in English (Appendix J). I used a discussion guide (Appendix O), and, in most instances, covered all questions contained in the guide. I fine-tuned and changed the order of the questions as necessary, and based on the interviewee’s focus group contributions, when applicable. As with focus groups, I compensated participants for their time in kind.

In the course of the interviews, I was very much aware of the power asymmetry between myself the researcher/physician and some women, particularly those from the working class study area. My interactions with them emphasized my social identity as a physician with some even addressing me as “doctor.” This was less discernible in my interviews with women among whom I had interacted with more closely at the Samurdhi meetings in Phase 1, indicating that spending extended time building trust with research participants may help to alleviate the power differential. These interviews contrasted with those held in the Kirideniya area where the playing field was more level and my interactions with women took the form of dialogue.

Final meeting. This was held at the Udawatta Village Development Society Hall, adjacent to the Grama Niladhari Office in Bogoda. The objectives of the meeting were to: 1) bring participants together to learn about the preliminary findings of the study; 2) promote reflection and dialogue on healthcare access issues faced by women belonging to different social groups; 3) engage

66 participants in identifying priority issues based on the study findings; and 4) facilitate discussion on possible ways to use the study findings to achieve positive change.

As rain poured down heavily throughout the morning on the day of the meeting, I arranged transport for the women who had agreed to attend the meeting that afternoon. Providing transport may explain the attendance at the meeting; 27 out of 40 women were present (Table 8). Visiting each woman at home the day prior to the meeting may also explain this attendance.

I did not audio record the meeting proceedings. Instead, the research assistant and I took hand- written notes. I began with a 15-minute presentation in which I shared some preliminary findings, and an hour-long discussion followed (Appendix P). I then moved on to a priority-setting exercise through which I had hoped to formulate an agenda for action to improve health services. I had originally planned to compile a list of suggestions on a white board, and request each participant to mark their priorities (1 to 3) against the list. As time was short owing to the inclement weather conditions, I proceeded with a backup plan. I distributed a handout with a list of recommendations drawn from the data (Appendix Q). The handout listed all recommendations for change cited by women during the focus groups and interviews, and I requested each participant to check their priorities (1 to 3).

Moderating the meeting was far more challenging than a focus group given the number of women and their diverse class and ethnic backgrounds. In the end, I achieved only the first two objectives of the meeting. I brought together women from different social backgrounds together and presented some preliminary findings (objective 1). We engaged in dialogue regarding our concerns around accessing healthcare (objective 2), which, I hoped, would have led to some level of critical engagement with the changes taking place in healthcare in Kandy. The priority-setting exercise was not very successful and I did not have time to discuss ways in which my findings could be used to achieve positive change. I attribute these shortcomings to both the weather conditions and my still-developing moderating skills. In retrospect, I should have allocated more time for the meeting, although this may have limited attendance. Despite its shortcomings, the meeting provided an opportunity for me to express my gratitude to the study participants for their time and generosity.

5.5.4 Data analysis and writing

My field notes, the transcriptions, and the data forms constituted the raw data. Analysis took place in the order specified below, but the stages blended with one another and were part of an iterative process that began with data collection and developed with my writing.

Transcription. The audio recordings were transcribed in full. The research assistant transcribed the focus groups and individual interviews held in Sinhala, while I transcribed those in English.59

Transcription is recognized as an act of representation in critical qualitative research (Oliver, Serowich and Mason 2005). I decided what was worthy or not of inclusion in my transcripts based on “practical and theoretical necessity” (Davidson 2009: 38). I coached the research assistant to ensure that we both aimed for a similar level of detail. We first transcribed as much speech as was discernible in the audio-recordings, including some idiosyncratic elements (e.g., pauses, ‘mms’ and ‘ahs,’ and selected non-vocal expressions like ‘laughing’, ‘sad’, ‘breaking down,’ etc.). We tried to retain the original language that was used and thus did not revise grammar or sentence structure (Oliver et al. 2005). I then carefully re-read the transcripts while listening to the audio recordings, and added missing text and non-vocal expression, as necessary. During this reading, I removed any potentially identifiable information.

Sections of audio recording were inaudible and could not be transcribed. Like others (Tilley 2003), the research assistant and I found transcribing the focus groups a challenging and time- consuming process as two or more members would often speak simultaneously, a socially acceptable practice in Sri Lanka. The individual interviews were easier to transcribe, although, in some instances, part of the audio recording was unclear, for example, when it rained heavily.

Content analysis and theoretical reading. In the second phase, I analysed the field notes and the transcribed data for content. I looked for descriptive patterns across cases (Patton 2015), paying attention to health problems, healthcare facilities, providers, health-seeking practices, healthcare decision-making, changes in healthcare, and suggestions for improvement.

Next, I undertook a theoretical reading of the data guided by sensitizing concepts (e.g., public/private sector, healthcare spending, affordability, availability, social class, ethnicity, gender,

59 I did not transcribe the Sinhala focus groups and interviews as my Sinhala typing speed was/is slow.

68 inequity, quality, etc.). These were drawn from theoretical assumptions, my reading of the literature, and my experiences as a healthcare provider (Brinkmann and Kvale 2015).

I then drew out concepts from the data (in vivo concepts), which I used to build on existing theory (e.g. public-private mix, connections, women doing, convenience, etc.). In other words, rather than using a purely inductive approach that began with open coding and proceeded to conceptualization and theorization, I moved from analytic deduction to an inductive approach (Brinkmann and Kvale 2015; Patton 2015).

Coding and summarization. In the third phase, I coded the data based on the descriptive patterns identified in my content analysis and the concepts that guided/emerged from my theoretical reading. I used the HyperRESEARCH (version 3.7.3) qualitative analysis tool to break down and reorganize my data to facilitate analysis (Appendix R).

I tabulated the contents of the data forms, and compiled a summary of each focus group and individual interview, every one of which constituted a ‘case’ on its own, bringing together the contents of my field notes, transcripts, and sociodemographic data. This process allowed me to better contextualize my data and perform cross-case analysis more effectively.

Thematic analysis, theorization and writing. In the next phase, I performed a thematic analysis (Roulston 2010) in relation to my specific objectives. I then delved into each theme to understand how women’s experiences of using healthcare differed by social class and ethnicity. By this stage, it was clear that mixing public and private services was common practice, and that the specifics of such mixing differed by social location. With the assistance of the HyperRESEARCH software, I considered overlap among the codes, ‘public-private mix,’ ‘class’ and ‘ethnicity’ to understand the ways in which different social groups ‘mixed’ services. I then tried to link these healthcare practices with institutional procedures and state policies.

Like transcription, writing is an act of representation (Alcoff 1991). In line with my methodology, I attempted to ground my findings in the “micropolitics of everyday life” (Mohanty 2003:229) by weaving women’s voices into my text. As the researcher, I selected the quotes, translated them (when applicable), and edited them for length, before inserting them into the text. Recognizing that researchers often appropriate research subjects to further other agendas (Spivak 1988), I questioned my intentions at every stage of my writing, and tried to ensure that my critique of privatization did not overshadow women’s experiences.

Translation is an interpretive process, shaped by social relations and the subjectivity of the translator (Temple and Young 2004; Van Nes et al. 2010; Wong and Poon 2010). The same text may be translated from one language to another in multiple ways (Wong and Poon 2010). In translating from Sinhala to English, I confronted both practical and epistemological challenges. On numerous occasions, I was unable to find words or phrases that sufficiently conveyed what was being said (Van Nes et al. 2010). In such instances, I avoided direct translation and instead translated the idea/concept with attention to the context in which the words were uttered (Temple and Young 2004). My middle-class social location would have necessarily shaped my understanding of Sinhala as spoken by other social and ethnic groups (Wong and Poon 2010).

Having spent years immersed in North American academia, I mulled over whether or not to change the original sentence structure as spoken in Sinhala, in my translation. When transcribing the focus group and interviews held in English, I soon recognized that the syntax used by English- speaking women corresponded to the sentence structure of word-to-word translations from Sinhala to English. For example, an English-speaking participant described visits from the midwife in her area in this way: “When you are about 8, 9 months pregnant only they come.” In this sentence, the two phrases (‘When you are about 8, 9 months pregnant only’ and ‘they come’) are arranged in the order that they would be spoken in Sinhala. I retained this sentence structure in my translation rather than changing it to: “They come only when you are about 8, 9 months pregnant.” However, this syntax is not without problems as I imposed a classed indigenous English syntax on to translations of views expressed by working class and lower-middle class Malay, Muslim and Sinhala women. On the other hand, if I did not go with the original sentence structure, I would have naturalized another form of English, likely North American or British, through translation. I also decided to retain the original syntax to render visible the politics of translation (Temple and Young 2004).

In reporting my research, I used numbers and letters within parentheses to signal the study area and ethnic background of participants after each piece of quoted text. The study areas were numbered as follows: 1 – Neluwa/Jagoda (working class women of mixed ethnicity); 2 – Bogoda (lower-middle class Muslim women); 3 – Dehioya (lower-middle class women of mixed ethnicity); 4 – Kirideniya (middle class Sinhala women from entrepreneur/business backgrounds); 5 – Kirideniya (upper-middle class English speaking women of mixed ethnicity from business/professional backgrounds). I used letters to denote reported ethnicity: ‘S’ for Sinhala; ‘Ma’ for Malay; ‘Mu’ for Muslim/Moor; and ‘T’ for Sri Lankan Tamil. Having outlined my methods, I now briefly discuss the trustworthiness of this research.

5.6 Trustworthiness

As a concept, ‘trustworthiness’ is favoured over ‘rigour’ in evaluating qualitative research (Baillie 2015; Finlay 2006). While earlier frameworks of evaluation tend to replicate criteria used in ‘positivist’ appraisals (e.g., validity, reliability, etc.), more recent approaches employ different criteria to evaluate qualitative research (e.g., credibility, transferability, etc.) (Baillie 2015; Long and Johnson 2000; Morse et al. 2002; Schwandt, Lincoln and Guba 2007; Tracy 2010). Others reject pre-defined assessment criteria all together given the scope and diversity of qualitative research methodologies (Rolfe 2004; Thomas and Magilvy 2011).

In this context, Patton (2015) outlines two approaches to assessing trustworthiness: an “out-put oriented approach” and a “process-oriented approach” (p. 678). While the former evaluates qualitative research findings based on criteria largely derived from positivist frameworks, the latter focuses on the research process. Adopting a process-oriented approach, Ballinger (2006) presents an appealing framework that encompasses the following ‘considerations’: “coherence of study aims and methodology; evidence of systematic and careful research conduct; convincing and relevant interpretation and; role of the researcher accounted for in a way that is consistent with research orientation” (pp. 241-2). Borrowing from Ballinger (2006), I address the trustworthiness of this research by adopting a process-oriented approach, focusing on reflexivity, methodological coherence and quality of data.

Reflexivity is “the process of reflecting critically on the self as researcher” (Guba and Lincoln 2005:210). Reinharz (1997:5) contends that researchers bring many ‘selves’ into research: researcher-based selves, brought selves (elements that historically, socially and personally construct the researcher standpoint) and situationally created selves. Each self, Reinharz argues, gives a distinct voice to the research. While my researcher-based self is grounded in a (middle class) North American critical research enterprise, my experiences as a Sri Lankan middle-class feminist physician of mixed ethnic heritage and a ‘visa student’ exposed to North American discourses of ‘global health’ profoundly shaped this work.

I was aware at the outset that my social location raised questions of representation, particularly in relation to ‘re-presenting’ subaltern women’s concerns (Spivak 1988). My residence in the Kirideniya area constituted a double-edged sword, providing easy access to the research site but with accompanying assumptions and alliances. Indeed, my claim for membership in the ‘community’ under study was fraught and begged definition (Frisby and Creese 2011). I was an

‘insider’ in the bi-lingual professional group from Kirideniya and, to some extent, in the neighbouring Dehioya group. However, I was very much an ‘outsider’ among women from the working class neighbourhood, Neluwa/Jagoda, and the Muslim women from Bogoda. Although being a physician opened doors to public healthcare institutions and gave legitimacy to this research, my social positioning resulted in research encounters that, at times, took the form of patient-provider interactions. Devoting time toward establishing relationships in Neluwa/Jagoda and Bogoda in Phase 1 helped to alleviate this power asymmetry, albeit to a limited degree.

Yet, my positionality came with some obvious caveats. Apart from my training in medicine, which would have necessarily influenced my understanding of the data, my choice of research questions, research setting, participants, and methods, were explicitly political acts. I hoped to interpose a counter-hegemonic narrative of healthcare access by engaging women whom I believed could contribute an informed social justice perspective on privatization. I recognized the tensions between my research agenda with its basis in a (Western) critical research enterprise and my desire to create ‘bottom up’ knowledge on healthcare access. The very disconnection I felt from the disciplines of medicine and global health helped me be attentive to counter- narratives. I tried to stay close to my data and authentic to women’s stories. I repeatedly asked myself: Am I interjecting too much in my research engagement to fulfil my objectives? Am I overlooking something important as a result of my research agenda? And for whom do I speak and write? I was very much aware of the power I held in shaping every stage of this research.

Methodological coherence. Grounding my research questions in a critique of healthcare privatization reflected my critical orientation. A Marxist approach demanded that I explore the material implications of privatization through a class lens (Patton 2015). Bringing a Marxist feminist perspective to my research, I gave precedence to women’s experiences and struggles in the context of healthcare privatization, and a Third World feminist standpoint meant that I employed colonialism as a foundational concept (Mohanty 2003).

My theoretical orientation influenced the selection of methods for data generation and analysis. I tried to harness the transformative potential of the research-participant interaction by engaging in critical dialogue with research participants (Roulston 2010). I attempted to alleviate the researcher-participant power differential by using focus groups (Wilkinson 1998). And, in my analysis and writing, I linked women’s healthcare experiences at the ‘micro level’ with macrostructures and process of healthcare restructuring to make explicit the ways in which

72 women navigated and accessed healthcare as they confronted neoliberal capitalist globalization. In doing so, I gave precedence to the perspective of marginalized social groups.

Quality of data. I used several techniques to improve the quality of my data. I spent an extended period of time – one and half years – in the research site. Undertaking two phases of fieldwork allowed me to better contextualize my research questions and strengthen my research protocol.

I used multiple data generation methods to improve the quality of my data. Rather than triangulation (Patton 2015), these methods served different purposes. Group interviews made way for the expression of disagreement and dissent, and created space for reinforcement of agreement and consensus. These expressions allowed me to locate what was being said in social context (Hennink 2007; Kitzinger 1994; Kitzinger 1995). Individual interviews provided space to explore experiences in more depth and clarify contradictions in a private setting (Hennink 2007; Hesse-Biber et al. 2011; Roulston 2010).

I pursued the ‘negative case’ (Patton 2015) when a contradictory or dissenting opinion was expressed. More often than not, such ‘dissenting’ views were expressed by middle-class residents, likely owing to my amateur interview skills and the degree of comfort this group of women felt in their interactions with me. For this reason and also because middle class women inevitably had more experience with private healthcare, they were over-represented in the individual interviews conflicting with my original intention to foreground this research in the experiences of marginalized women, a concern I tried to address in my writing.

Maintaining extensive field notes allowed me to recall finer details and compelled me to think through my assumptions and the dynamics of my research engagement. While only a small section of data made their way into this thesis (Patton 2015), I have outlined the various shifts that took place in my thinking during the writing phase to indicate how I arrived at my findings.

5.7 Summary

Shaped by my experiences as a physician and graduate student, and guided by a Third World Marxist feminist approach, I conceptualized this project at the University of Toronto, and undertook my field research in two phases in Kandy, Sri Lanka. The first, an exploratory phase, informed the design and implementation of the second community-based phase of research. Using focus groups and individual interviews as my primary data generation methods, I explored how the presence of private healthcare shaped access through the experiences of women

73 residents of Udawatta Division in Kandy. In the next chapter, I draw on multiple sources, including data generated in this study, to introduce the reader to the structure and organization of healthcare in Kandy and describe the healthcare facilities that were available to participants.

6. Healthcare in Kandy: Health System and Healthcare Facilities

Chapter overview: This chapter provides an introduction to the physical organization of healthcare in Kandy. It consists of two sections. The first is an overview of the structure and organization of public and private healthcare sectors in Sri Lanka, which mirrors the set up in Kandy. The second hones in on public and private healthcare facilities in Kandy to map out the services available to residents of Udawatta Division. The chapter draws on a combination of sources, including the study data, policy documents, other texts, and online sources.

6.1 Structure and organization of healthcare

A two-tiered healthcare system operates in Sri Lanka. The public system delivers healthcare free- of-charge at points of delivery, while the private system runs in parallel, operating on a fee-for- service basis (Govindaraj et al. 2014; Rannan-Eliya and Sikurajapathy 2009). Most public and private healthcare services may be used on a walk-in basis without requirements for proof of identity, residence or citizenship.60

6.1.1 The public system

The public sector is estimated to cover over 90 per cent of inpatient admissions, about 50 per cent of outpatient visits, and the bulk of preventive and public health services (Govindaraj et al. 2014). It accounts for about 45 per cent of health expenditure, financed chiefly through indirect taxation.61 In 2013, 43 per cent of total health expenditure was funded by the government, while overseas grants accounted for just one per cent.62 The remainder was financed through private sources, for the most part, out-of-pocket payments (Institute for Health Policy 2015).

The central Ministry of Health and nine provincial health departments coordinate and deliver public healthcare services. The central Ministry oversees the National Hospital of Sri Lanka (the largest healthcare facility in the country located in the commercial capital, Colombo), all teaching

60 Community-based preventive services target residents of Medical Officer of Health Areas (details ahead). 61As the Ministry of Health and the provincial health departments receive funds directly or indirectly from the Treasury, it is difficult to surmise the percentage of health spending that comes from indirect taxation. In 2016, 86% of total government revenue came from taxes. Of this, taxes on goods and services contributed 60 per cent, while income tax comprised only 16 per cent (Ministry of Finance 2017). 62 This estimation places development ‘assistance’ in the form of loans under public expenditure because repayment is required (Institute for Health Policy 2015).

74 75 hospitals, all specialized hospitals, selected provincial and district general hospitals,63 and all vertical programmes (e.g., Family Health Programme, Anti-Malaria Campaign, Dengue Control Programme, National Programme for Tuberculosis Control and Chest Diseases, National STD/AIDS Control Programme, Non-Communicable Disease Unit, etc.). It is responsible for policymaking, developing guidelines, human resource training, recruiting health worker cadres, and monitoring, purchasing and distributing drugs and other medical supplies. With the exception of a small share of services delivered by municipal authorities,64 the provincial departments of health administer all other healthcare facilities (Govindaraj et al. 2014; World Bank 2013).

The public system is made up of curative and preventive sectors. In 2014, the curative sector comprised 1085 facilities, including primary, secondary and tertiary care centres, spread across the country (Ministry of Health, Nutrition and Indigenous Medicine 2016a). Notably, however, the public sector lacks an organized system of primary curative care. As a result, users are compelled to access a primary medical care unit or the outpatient department of a public hospital for day-today health concerns (e.g. sore throat, rash, fever), or, alternatively, visit a private healthcare facility (Mudiyanse 2014).

The delivery of preventive services is decentralized to the provincial departments of health through 26 health regions,65 made up of 338 divisional health units (or Medical Officer of Health Areas). A Medical Officer of Health runs each divisional health unit with support from a public health team consisting of a public health nursing sister, public health nurses, public health inspectors, and public health midwives. Divisional units are further divided into Public Health Inspector Areas, which are, in turn, divided into Public Health Midwife areas, each covering a demarcated population (~3000 to 5000) served by a field midwife (Family Health Bureau 2014; Ministry of Health, Nutrition and Indigenous Medicine 2016a).

Preventive services are accessible only to residents of each divisional health unit or Medical Officer of Health Area (Family Health Bureau 2014; Ministry of Health, Nutrition and Indigenous

63 The public curative sector is made up of the following tiered facilities spanning primary, secondary, and tertiary care levels (in ascending order): primary medical care units, divisional hospitals, base hospitals, district general hospitals, provincial general hospitals, specialized hospitals, teaching hospitals, and the National Hospital of Sri Lanka (Ministry of Health, Nutrition and Indigenous Medicine 2016a; Provincial Director of Health Services - Central Province n.d.). 64 Selected MCH services, dengue control/prevention activities, and primary curative services are coordinated and delivered by some (not all) municipal authorities, including the Kandy Municipal Council. 65 Each district roughly corresponds to a health region with the exception of Amparai District in the Eastern Province, which is divided into two health regions.

Medicine 2016a). Among the services delivered by divisional health units, maternal and child services are targeted to women of reproductive age and children below five years. A couple must either be “legally married or living together where the woman is between 15 to 49 years” or a household must have “a child under 5 years” to be eligible (Family Health Bureau 2014:11), raising accessibility concerns for women (and men) positioned outside traditional family structures.

Divisional health units deliver the widely popular field midwifery service, spanning pre-conception, antenatal, delivery and post-natal care, and family planning (Daniel 2016; Pathmanathan et al. 2003). The field midwives offer domiciliary care and collaborate with divisional public health teams to coordinate and provide services within their respective areas. An effective referral system links field midwifery services with hospital-based specialist obstetric care (Family Health Bureau 2014; Pathmanathan et al. 2003). About 95 per cent of deliveries take place in public facilities (Ministry of Health, Nutrition and Indigenous Medicine 2016a) where women’s experiences are largely positive (Jayasuriya 2014; Senarath, Fernando and Rodrigo 2006).

The effectiveness of the preventive sector is partly reflected in Sri Lanka’s impressive maternal and child health indicators. The maternal mortality ratio is relatively low at 30 per 100,000 live births, while antenatal coverage and skilled attendance at birth are high at 99 per cent. The under- five mortality rate is also fairly low at 10 per 1000 live births (WHO 2017b). 66 However, these achievements are tainted by regional disparities with rural, plantation, and war-effected districts faring much worse (D’Almeida 2014; Family Health Bureau 2014).

The divisional health units also coordinate communicable disease control activities. A series of successes, including WHO certification for having eliminated malaria in 2016, are credited to these efforts (WHO SEARO 2016). In addition, divisional health units deliver School Medical Inspections (through the School Health Programme, a collaboration with the Ministry of Education), Well Woman Clinics,67 and Healthy Lifestyle Clinics,68 in each Medical Officer of Health Area (Family Health Bureau 2014; Mallawaarachchi 2016). Despite this gargantuan role

66 To place these figures in context, the maternal mortality ratios of other countries in the South Asian region are (per 100,000 live births): Afghanistan (396), Bangladesh (176), Bhutan (148), India (174), Maldives (68), Nepal (258) and Pakistan (178); and the under five mortality rates in the South Asian region are (per 1000 live births): Afghanistan (91), Bangladesh (38), Bhutan (33), India (48), Maldives (9), Nepal (36), and Pakistan (81) (WHO 2017b). 67 These clinics target women 35 years and older for screening for chronic diseases (e.g. Pap smear, breast examination, blood pressure and blood glucose checks) (Family Health Bureau 2014). 68 Introduced in 2011, Healthy Lifestyle Clinics provide ‘risk assessment’ and screening for non-communicable diseases (WHO 2014; Mallawaarachchi 2016).

77 played by the preventive sector, financial allocations for primary care have declined overtime as more and more funds are channelled toward tertiary care (Institute for Health Policy 2015).

The Ministry of Health trains and recruits public sector health workers, including nurses, midwives, and ancillary medical professionals, at no direct cost to the trainees. The non-fee levying public university system trains non-specialist doctors, while the Ministry of Health subsidizes post- graduate training for its physician cadre. Trained doctors (medical officers and specialists), nurses, midwives, and most categories of paramedical staff, are guaranteed public sector employment on completing training. They are required to serve in remote and rural areas in order to climb the ranks of the public sector, ensuring a widely dispersed healthcare workforce, although with considerable regional inequities (Ministry of Health 2015). This system of training and recruitment has allowed the public health sector to run at low cost, although ‘brain drain’ from public to private raises concerns (Dayaratne 2013).

6.1.2 The private sector

The private sector comprises an ad hoc range of health facilities from independent general practice clinics and specialist consultation centers to small nursing homes69 and larger private hospitals (Amarasinghe et al. 2015; Dayaratne 2013). These facilities are supported by numerous private pharmacies and diagnostic centres of varying standards (Dayaratne 2012).

The NGO sector plays a very small role in healthcare delivery in Sri Lanka. NGOs contribute to some extent to contraceptive service provision (Department of Census and Statistics and Ministry of Healthcare and Nutrition 2009), and offer services for people with disabilities and those in need of mental health services, particular in war-affected regions (Jaffna Jaipur Centre for Disability Rehabilitation 2012; Shanthiham 2017). A handful of cooperative hospitals and faith-based organizations provide a limited set of services, particularly in Colombo and the war-affected North (WHO 2017c; Friends of Manipay Hospital n.d.).

Private healthcare is financed chiefly by households with out-of-pocket payments accounting for over 80 per cent of private health expenditure (Institute for Health Policy 2015). User-fees are not adjusted by income and a greater share of private spending goes toward outpatient consultations, except in the highest wealth quintile where a considerable portion is spent on inpatient care

69 A term used to designate smaller private healthcare facilities possessing limited inpatient services.

(Mackintosh et al. 2016). The remaining private expenditure comes from employers, private insurance schemes and non-governmental organizations (Institute for Health Policy 2015). Cost comparisons suggest that healthcare services, inclusive of drugs and diagnostics, are overpriced in the private sector (Balasubramaniam 2012; Dayaratne 2012). In 2009, 48 per cent of the volume of medicines was purchased in the private sector, accounting for over 75 per cent of spending on pharmaceuticals (Amarasinghe et al. 2015).

Social health insurance does not operate in Sri Lanka, although a contributory health insurance scheme covers a specified set of health benefits for public sector employees (National Insurance Trust Fund n.d.).70 A small but growing domestic private health insurance industry covers mostly private sector employees with pre-paid insurance plans accounting for about 5 per cent of private expenditure (Institute for Health Policy 2015; Rannan-Eliya and Sikurajapathy 2009). Some corporations implement reimbursement schemes or maintain their own healthcare facilities (Ministry of Health, Nutrition and Indigenous Medicine 2016b). In 2013, together employers covered about 7 per cent of private spending (Institute for Health Policy 2015), although the proportion may have been higher in Colombo (De Silva et al. n.d.).

General and specialist services are available on a walk-in basis in the private sector. General practitioners usually operate from independent clinics, and a specialist opinion may be obtained fairly easily at a specialist consultation center, a procedure that is referred to as ‘channeling’ (De Silva et al. n.d.). As the majority of general practitioners and specialists in the private sector are employees of the Ministry of Health, users move quite easily between sectors, opting for private outpatient care, while turning to government hospitals for (more expensive) inpatient treatment (Govindaraj et al. 2014). However, this trend is fast changing with more users in urban areas turning to the private sector for inpatient care (Amarasinghe et al. 2015), a shift that has been attributed to deficits in public sector services and the rapid expansion of private healthcare (Dayaratne 2013; Govindaraj et al. 2014).

The growing dominance of the private sector is evident in its contemporary role in the introduction of (high-cost) biomedical technologies, previously the purview of the Ministry of Health. Today, the country’s most advanced neurosurgical unit and bone marrow transplant facility are located

70 These benefits cover expenses incurred from room charges, consultations, tests, medicines, and a specified set of medical and surgical procedures in both public and private sectors. Claimable amounts differ based on the category of expenses and sector (National Insurance Trust Fund n.d.).

79 in private hospitals (Dayaratne 2013; Daily FT 2014). A major proportion of coronary artery bypass grafting is undertaken in the private sector with public sector wait lists exceeding several thousand patients (Jayamanne 2013). However, the availability of resource-intensive biomedical technologies is largely confined to Colombo-based private hospitals. In 2011, all private sector MRI scanners, blood banks, lithotripters, and over half of all private sector operating theaters, ambulances, x-ray facilities, CT scanners and mammography units were located in the Western Province (Amarasinghe et al. 2015).71 Revenue from private hospitals is also heavily Colombo- centered with more than 80 per cent attributed to the Western Province. Indeed, about three- fourths of the private market share is concentrated among four major healthcare firms operating out of Colombo (Dayaratne 2013).

This Colombo-centric development of private hospitals coincides with numerous state-promoted medical tourism ventures commenced to attract foreign investment (Oxford Business Group 2017). While the merits of medicalization with accompanying market expansion for biomedical technologies is debatable (Waitzkin 2011), these trends have important implications the public system as public sector specialists increasingly opt to work fulltime in the private sector causing a dearth of specialists in remote and disadvantaged areas (Dayaratne 2013).

6.1.3 Standard setting and regulation

The central Ministry of Health develops policies and sets standards for the public healthcare system. In contrast, the private sector has no designated body to regulate, coordinate and/or standardize facilities and services (Amarasinghe et al. 2015; De Silva et al. n.d.; Govindaraj et al. 2014). In 2006, the Private Medical Institutions (Registration) Act transferred authority for private health sector regulation from the Ministry of Health to an ‘independent’ body, the Private Health Services Regulatory Council. With wide representation from the private healthcare industry, the Council has been unable to implement the most basic requirements for registration (Amarasinghe et al. 2015).

The maintenance of professionalism, discipline and ethnical practice among health professionals is under the purview of the Sri Lanka Medical Council (SLMC) (SLMC 2007-2008). However, the mechanisms in place to redress medical malpractice are weak with available options time-

71 More recent data is not available; this situation is fast changing as a number of large-scale private hospital projects have since commenced operation in other cities and bigger towns (Fernandopulle 2016).

80 consuming, expensive and lacking transparency. Many individuals have resorted to action but with frustrating results and little or no compensation (Rupasinghe 2015). According to a study commissioned by Transparency International and Friedrich Ebert Stiftung (2009:16), “petty corruption, bribery and nepotism” are widespread in government hospitals. Lacking regulation also manifests in unethical forms of private practice and escalating costs of private healthcare (Dayaratne 2013; De Silva 2015; Jayamanne and Palihawadana 2014). A recent kidney transplant racket involving Indian donors being trafficked to Sri Lanka for organ harvesting was allegedly based at four commercial private hospitals (Thomas 2016).

6.1.4 ‘Public-private partnerships’

Privatization has encroached upon the public system through various ‘formal’ and ‘informal’ public-private partnerships. Fee-levying sections exist in some tertiary care centers, including the National Hospital of Sri Lanka (Daily FT 2016; Mudalige 2012). The President’s Fund72 offers financial assistance for patients to undergo a predefined set of procedures, including cardiac surgery, cancer therapy, and renal transplant, in the private sector (Dissanayake 2015; Government Information Centre 2009; Kumar 2015b).

Various facility services (e.g. security, laundry services) have been outsourced to private companies (Fernando 2013; Govindaraj et al. 2014; Narangoda and Khathibi 2014). Until recently, the government did not rely on such partnerships or contractual arrangements for clinical services.73 Since 2016, however, a not-for-profit ambulance service – a ‘public-private partnership’ between the Government of Sri Lanka and Indian conglomerate GVK EMRI – has operated with financial assistance from the Indian government (Economynext 2016).

The public and private sectors are linked in other ‘informal’ ways. Users are directed by public healthcare providers to the private sector for tests and medicines unavailable or in short supply at public hospitals (Dayaratne 2013; Fernando 2013). This situation is compounded by dual practice as private healthcare facilities engage the services of physicians and other health professionals employed full-time in the public sector (De Silva et al. n.d.; Govindaraj et al. 2014;

72 The President’s Fund, established under the President’s Secretariat, extended its mandate to healthcare coverage in 1995. A healthcare user, irrespective of income level, may apply once in her/his lifetime to the President’s Fund (Kumar 2015b). 73 In the past, the Ministry of Health has turned to the private sector when government services were disrupted by trade union action (Fernando 2013).

Narangoda and Khathibi 2014). Having provided an overview of the organization of healthcare, I now introduce the public and private healthcare facilities available to Udawatta residents in Kandy.

6.2 A snapshot of healthcare facilities in and around Udawatta

6.2.1 Public facilities and services

Udawatta Division is home to and surrounded by an array of public healthcare facilities (Figure 3). Within the Division, the public sector is represented by the Kirideniya Maternal and Child Health (MCH) Clinic, which provides maternity, infant and child (preventive) healthcare services to a larger area, inclusive of Udawatta Division.

Part of the community-based public MCH service that functions across the country, the Kirideniya MCH Clinic is centrally located in close proximity to the village cooperative store, post office, and community centre. The two public health midwives74 assigned to the Division use the Clinic as a base, teaming up with a medical officer,75 a public health nursing sister, a public health nurse, and other midwives, to provide MCH services to the residents of several Public Health Midwife Areas within the Kandy Municipal Council Area.76 As the Clinic serves a specified residential population, it offers some degree of continuity of care.

Outside the Division and within the Kandy city limits, the public sector comprises a primary medical care unit, four tertiary care centres and a number of special units77 (Provincial Director of Health Services – Central Province n.d.). Suduhumpola Primary Medical Care Unit, 78 the only primary curative facility located in the Kandy Municipal Council Area, is situated about a kilometre off the northwestern boundary of Udawatta Division. Colloquially referred to as ‘the Dispensary,’ the Primary Medical Care Unit in Suduhumpola functions as a walk-in facility on designated days

74 Field midwifery services are provided by two midwives as the border of two PHM areas cuts across the Udawatta Division (PHM areas are larger than grama niladhari divisions). 75 A non-specialist physician in the public sector. 76 MCH services generally come under the purview of the provincial departments of health. However, this clinic is run by the Kandy Municipal Council owing to its location within the Kandy Municipal Council area. 77 Special units offer services for specific conditions through their respective vertical programmes. For example, the Bogambara Chest Clinic depicted in Figure 3 is the district-level branch clinic of the National Programme for Tuberculosis Control and Chest Diseases. 78 The Primary Medical Care Unit in Suduhumpola is administered by the Department of Health – Central Province (Provincial Director of Health Services – Central Province n.d.).

82 of the week and, on others, as a ‘clinic’79 for the follow-up of people with longer-term health problems like diabetes or high blood pressure.

Figure 3. Healthcare facilities in and around Udawatta Division. Source: Google 201680

The Division is located close to four public tertiary care centres – Teaching Hospital Peradeniya (Peradeniya Hospital, hereafter), Teaching Hospital Kandy (Kandy Hospital, hereafter), Peradeniya Dental Hospital and Sirimavo Bandaranaike Children’s Hospital (Children’s Hospital,

79 The term “clinic” is used colloquially to refer to a public outpatient facility that offers long-term follow-up services; clinics are held across specialties and subspecialties in both preventive (e.g. MHC) and curative (e.g. medicine, surgery, cardiology etc.) sectors. 80 Map downloaded from Google Maps; names and locations of healthcare facilities were inserted by me.

83 hereafter).81 The Kandy Hospital is the largest tertiary care centre in the Central Province. Apart from services in the basic specialties – medicine, surgery, obstetrics and gynaecology, paediatrics, and psychiatry – the hospital offers services in various subspecialties and advanced treatment options like cardiac catheterization, neurosurgery, renal transplant, dialysis, and so forth (Teaching Hospital Kandy 2012-2017).

Affiliated to the University of Peradeniya, the Peradeniya Hospital delivers a range of specialist services, although in fewer subspecialties compared with the Kandy Hospital (Teaching Hospital Peradeniya n.d.). The Children’s Hospital is a newer addition to the public infrastructure of Kandy, and has expanded over the years, featuring a range of highly specialized services (Official Government News Portal of Sri Lanka 2017). All public curative facilities function on a walk-in basis. However, the specialized services at tertiary care centres may only be accessed via a referral from the outpatient department or through an in-hospital referral, unless a user is directed by a private sector specialist to their public sector unit.

Kandy District is home to several public Ayurveda facilities. Apart from the two Ayurveda dispensaries located within the Division82 (Figure 3), numerous Ayurveda hospitals and dispensaries are located in the District (Department of Ayurveda Central Province n.d.).83 Among them, participants shared experiences at Pallekelle Provincial Ayurvedic Hospital, Doluwa Ayurvedic Hospital, and Gelioya Ayurvedic Hospital, all situated between 10 and 20 km off Kandy. The indigenous system is, however, beyond the scope of this thesis.

6.2.2 Private facilities and services

The private sector in Kandy comprises numerous general practitioner clinics, specialist consultation centres, and private inpatient facilities (Figure 3).84 Most private general practitioners – colloquially referred to as ‘family doctors’ – also work full-time as medical officers with the Ministry of Health, and thus (legally) engage in private practice only outside of official work

81 The tertiary care centres in Kandy come under the administration of the central Ministry of Health (Ministry of Health, Nutrition and Indigenous Medicine 2016a; Provincial Director of Health Services - Central Province n.d.). 82 The two Ayurveda dispensaries located in Neluwa and Kirideniya are staffed by one practitioner whose time is divided between the two dispensaries. The Kirideniya Ayurveda dispensary shares its space with the MCH Clinic, giving way to the latter on scheduled days. 83 None of the Ayurveda hospitals are located within the Kandy city limits, explaining their absence in Figure 3. 84 In Kandy, the NGO sector is largely absent from healthcare delivery.

84 hours.85 They offer private consultations on a walk-in basis and dispense a limited selection of drugs through their respective facilities (Mudiyanse 2014). To my knowledge, at least five private family doctors practiced in the Division, and several others practiced at walking distance, during the study period.86

General practitioner services cost roughly between SLR87 200 to 400 (CAD ~1.50 to 3.50) per visit, less than the daily pay (~ SLR. 600 or ~CAD 5) of a female wage labourer in Kandy. While this charge generally covers consultation and (basic) treatment, out-of-pocket expenses could amount to more if additional diagnostics or drugs need to be purchased from a private pharmacy or laboratory. These expenses are, at times, offset by family doctors who arrange for their ‘clients’ to access these services at no cost at the public sector facilities in which they work.

Specialist consultation centres, popularly referred to as ‘channelling centres,’ offer direct access to specialists with no requirements for referral. Like family doctors, most private sector specialists in Kandy hold full-time positions with the Ministry of Health, and engage in ‘channel practice’ outside work hours.88 ‘Channelling’ generally involves booking a consultation by phone or in person, paying the fee upfront, and meeting with the specialist at the channelling centre.

In the absence of regulation, a channelled consultation may last from two to three minutes to half an hour or more, and may cost anywhere between SLR 800 (~CAD 7) and SLR 2000 (~CAD 17) in Kandy. While a portion of this fee is directed to the channelling centre, these fees do not include diagnostics or pharmaceuticals. Thus, together with tests and medicines, a specialist consultation could amount to several thousand rupees. Even so, channelling is standard practice for some who see it as a pathway to accessing speedier services in the public sector.

Numerous channelling centres dot the landscape of Kandy (Figure 3). Among them, the private Amaya Medical Centre (or ‘Amaya’) is located just a five-minute walk away from the northwestern side of the Division. Apart from Amaya, several channelling centres operate a short distance from

85 Private family doctors are usually available at private facilities from 7 to 8 am and after 4 pm, although some engage in unlawful private practice during work hours (my experience). 86 I requested the participants to avoid disclosing identifiable information about individual healthcare providers. However, by observation, I was aware of five such clinics operating within the Division. 87 Sri Lankan Rupees 88 This situation is changing, especially in Colombo, where specialists are increasingly resigning from the Ministry of Health to work full-time in the private sector (Dayaratne 2013).

85 the Division. They include the Kandy Channelling Centre (<5 km) and the channelling centres housed at the Suwasevana Hospital (<5 km), the Lakeside Adventist Hospital (<10 km) and the Kandy Nursing Home (<10 km).

Private facilities with the capacity to offer inpatient care are few in Kandy. Suwasevana Hospital (or ‘Suwasevana’) is the largest private facility in Kandy, offering a range of facilities and services, including a diagnostic centre, a pharmacy, a channelling centre, and inpatient services inclusive of intensive and emergency care (Suwasevana Hospitals Pvt. Ltd. 2015). While Suwasevana is located closest to the Division, the Lakeside Adventist Hospital, Kandy Nursing Home, and Kandy Private Hospital are also reasonably close (<10 km).

Wealthier residents of Kandy travel to Colombo to access inpatient services at large commercial hospitals. Among them, Nawaloka Hospital, Lanka Hospital, Asiri Central Hospital, Durdans Hospital, and Hemas Hospital are major players in the private hospital industry (Dayaratne 2013). While the private hospitals in Kandy do not match the services on offer at these Colombo-based commercial hospitals, this situation is likely to change as Asiri Health, the largest private healthcare company operating in Sri Lanka, plans to open a tertiary care centre in Kandy in 2018 (De Silva 2016).

Healthcare facilities, whether public or private, are surrounded by an assortment of private pharmacies and diagnostic centres. While the cost of medicines is known to be comparatively low in Sri Lanka (Mendis et al. 2007), drugs may be obtained at economical rates at Osu Sala pharmacies run by the State Pharmaceutical Corporation (State Pharmaceutical Corporation of Sri Lanka 2015). Pharmacists are prohibited from dispensing prescription-only medicines without a valid prescription, but it is quite easy to obtain such drugs over-the-counter from a private pharmacy (Wijesinghe et al. 2012). Diagnostic services may similarly be accessed ‘over-the- counter’ without a requisition form. Charges for these services vary. As examples, a fasting blood sugar costs about SLR 250 (~CAD 2) and a lipid profile, a little over SLR 1000 (~CAD 8). Figure 3 depicts only the two Asiri89 laboratories, which featured prominently in my data, and the pharmacies and diagnostic centres located adjacent to the Division.

89 Asiri Health is a key player in the domestic private hospital industry, and has captured a major share of the diagnostics market through regional laboratories and collection centres (Asiri Health 2016; Dayaratne 2013).

The Bowatta Dispensary, a reputed private over-the-counter Ayurveda facility among many in Kandy, is also depicted in Figure 3. While numerous commercial Ayurveda spas are located in and around Kandy (TripAdvisor 2017), they mostly cater to foreign tourists. As stated earlier, the indigenous system remains beyond the scope of this thesis.

6.3 Summary

The public system plays a pivotal role in healthcare provision in Sri Lanka. It is centrally coordinated, although a substantial share of delivery is devolved to the provinces. A relatively smaller private sector covers a considerable share of outpatient care and accounts for a growing proportion of health expenditure. Healthcare services may be accessed on a walk-in basis through public and private sectors. While the private health sector has no system in place to coordinate or standardize the delivery of services, mechanisms to redress medical malpractice are weak in both sectors. In this lax regulatory environment, various informal arrangements have emerged between public and private healthcare sectors with the private sector increasingly encroaching upon the public system.

The health system in Kandy mirrors the structure and organization of healthcare in (urban) Sri Lanka. Apart from the maternal and child health clinic located within the Division, residents may access a range of public healthcare facilities, including the primary medical care unit and four tertiary care centres. A comparable public sector presence may be seen in some areas of Colombo and perhaps adjacent to other faculties of medicine in Sri Lanka. The private sector in Kandy is far less ‘developed’ than it is in Colombo, although it is at the cusp of change.

Having introduced the reader to the health system and healthcare facilities in Kandy, in the next two chapters, I explore women’s experiences to understand how the presence of private healthcare shapes access for residents of Udawatta. The first chapter considers patterns of public and private healthcare use to understand what kinds of healthcare ‘choices’ open up for women representing various social groups in the context of privatization. The second examines how women, as classed and ethnic subjects, negotiate public and private services within Kandy’s ‘mixed health system.’

7. Class matters in exercising ‘choice’ in healthcare

Chapter overview: Having introduced the diverse healthcare options located within and in close proximity to Udawatta Division, in this chapter, I consider where women went for healthcare (research objective 1), their impressions of the various facilities and services they used (research objective 2), and how these dimensions differed by social location (research objective 4), to examine the range of healthcare ‘choices’ that have opened up for women in the presence of private healthcare. The first part of the chapter delves into the conditions prevailing in Kandy’s public sector to indicate why alternative healthcare ‘choices’ may have been desirable for Udawatta residents. The rest of the chapter examines the healthcare ‘choices’ available to various social groups within Kandy’s ‘mixed health system’ to highlight class and ethnic differences in access to healthcare.

7.1 Locating the ‘demand’ for alternatives

Neglected by the state for decades, the public system in Kandy is under-resourced and overcrowded (Hsiao 2000). From the field midwifery service and public dispensary to both out- and inpatient services at tertiary care centres, women shared stories of lengthy wait-times, congested conditions, and numerous other challenges associated with accessing public healthcare. As we shall see, residents of the working class neighbourhood, Neluwa/Jagoda (1),90 were disproportionately burdened by such hardship, suggesting that others may have overcome at least some of these challenges by accessing ‘alternative’ private options.

The field midwifery service functioned under human resource constraints. Chulani (1/S),91 a mother of four, observed that the MCH Clinic was always understaffed and overcrowded:

There are about four or five midwives… These five misses (midwives) have to manage the crowd at the clinic, the mothers and children all together; check pressure, check urine, weigh, and write

90 The numbers in parentheses designate the social class represented by each study area (1 – working class; 5 – upper-middle class): 1 – Neluwa/Jagoda (working class women of mixed ethnicity); 2 – Bogoda (lower-middle class Muslim women); 3 – Dehioya (lower-middle class women of mixed ethnicity); 4 – Kirideniya (middle class Sinhala women from entrepreneur/business backgrounds); 5 – Kirideniya (upper-middle class English speaking women of mixed ethnicity from business/professional backgrounds). 91 The letter denotes reported ethnicity (S – Sinhala; Ma – Malay; Mu – Muslim/Moor; T – Sri Lankan Tamil).

on the mothers’ cards, write on the children’s cards,92 vaccinate the children, everything is done there.

She compared the domiciliary services offered by the midwives during her first pregnancy to what was on offer today: “Those days [the midwife] used to come soon after the delivery to check on us…. until the stitches dried up she came, she checked on the baby’s cord. Now we don’t get that service.” Premalatha (1/S), the President of the Neluwa Samurdhi Society, explained this decline by drawing attention to the field midwife’s ever-increasing workload: “[The midwife] has to cover the area right up to Siyambalapitiya93…the Division is too much for her.” Namali (4/S), a middle class bank executive who also relied on the field midwifery service, pointed to the wide range of clinic-based services offered by the midwives: “They have a lot of programmes here and there, [they] must prepare for them.” According to her, the field midwives did not have time to make house calls because of these activities. Although residents across study areas, including Kirideniya (4, 5) used the midwifery service, the congestion and hurried consultations drove some like Himali (5/S) and Kishani (5/S) away from the service all together.

In contrast to the well-developed community-based (preventive) maternal and child healthcare system, there was no structure in place to address the curative care needs of Udawatta residents. The Suduhumpola Primary Medical Care Unit (or ‘the Dispensary’), the only one of its kind in the Kandy Municipal Council Area, offered walk-in services only on three days of the week. Damayanthi (1/S), a retired janitor, highlighted the challenges users confronted because of this limited service: “On clinic days, patients [who come for walk-in services] can’t get treatment.” On such days, she travelled by bus to the outpatient department (OPD) at Peradeniya Hospital, while others like Nafeesa (1/Mu) visited a private family doctor.

The OPDs were generally congested and involved lengthy wait-times at several places. Dilrukshi (3/S), a homemaker, spoke of multiple delays at the OPD: “[You]94 need to wait in the queue, first [to see the doctor] and get the prescription, and, wait again in another queue to get the medicines.”

92 Here Chulani refers to the Pregnancy Record and the Child Health and Development Record issued to pregnant women and children, respectively, at public healthcare facilities. 93 A neighbouring village outside Udawatta Division. One of the two midwives providing services in the Udawatta Division also covered Siyambalapitiya as public health midwife (PHM) areas tend to be larger than grama niladhari divisions and the border of two PHM areas cuts across the division. 94 Pronouns are not always used in colloquial Sinhala. When they are, they are usually gender neutral. I inserted pronouns to improve readability, and used “they” when the gender of the person referred to by the participant was not mentioned.

Consultations took place in a crowded room in the presence of other users as on-shift medical officers,95 several in each room, carried out consultations. As Badra (3/S) explained: “Now in one room there are about four doctors.” Apart from the lack of privacy, consultation times were inevitably short and continuity of care minimal. This meant that women (and men) with sensitive health concerns or those who needed more time to discuss a complex issue would likely seek ‘alternative’ healthcare options in the private sector. Indeed, for varying combinations of these reasons, the vast majority of women residing outside Neluwa/Jagoda (1) favoured private outpatient options over both the Dispensary and the OPD.

The public clinics at tertiary care centres served thousands of users on a daily basis. Nuzrath (1/Mu), a mother of three, had arrived at the Kandy Hospital at 6 o’clock in the morning to ensure she got a ‘number’96 for her son at the crowded ophthalmology clinic: “We waited in line for a long time and managed to get his eyes examined with the greatest difficulty.” Fara (3/Mu), a homemaker with high blood pressure, cholesterol and diabetes, had spent a good five hours seeing a doctor and refilling the prescription at her most recent clinic visit at the Peradeniya Hospital. Pushpa (3/S) had recently survived a heart attack, and reported similar experiences: “Even if [we] go at 6 in the morning, [we] don’t get the medicines till 1 or 2 (pm).”

Waiting facilities were grossly insufficient at the public clinics. Dilrukshi (3/S) described the waiting area at her mother-in-law’s medical clinic: “The people [at the clinic] suffer, they have to stand because there are no seats… there are about 5 chairs … for about 1000 people.”97 Nafeesa (1/Mu) portrayed a similar situation at the Kandy Hospital’s pharmacy where she spent several hours waiting in line to refill her monthly prescription: “[We] stand and wait in the sun, [some] feel dizzy and fall.” It was not surprising that Mala (5/S), a retired teacher, preferred the private sector for outpatient care: “We don’t have time to spend… and no energy to stay there for a long time.” Yet, economically disadvantaged women had no ‘choice’ but to spend half a day at a clinic as private care for chronic long-term illness was beyond their means.

95 Non-specialist physician in the public health sector. 96 Some clinics have caps on the number of patients seen per day. Thus, patients generally arrive at the clinic in the early hours of the morning to ensure that they are seen (my experience). 97 Dilrukshi may have exaggerated this number. Even so, several hundreds of people with chronic illnesses attend medical clinics on any given day at tertiary care centres (my experience).

Public inpatient facilities were often available under critical circumstances despite resource constraints. Leela (1/S), who had a chronic kidney problem that required frequent dialysis, observed that the dialysis unit was always full, although the nursing staff maintained a vacant bed for “emergencies.” She observed that the staff worked tirelessly into the night to serve the large number of patients who needed dialysis: “[The last dialysis session] is at 7 [pm] and it is about 12 or 1 [at night] when it ends.” However, wait-lists for some inpatient procedures could extend from several months to even years. Fara (3/M)’s husband had been on a wait-list for his “bypass” (coronary artery bypass grafting) for over two years. In her opinion, not enough doctors who could perform the procedure were available at the Kandy Hospital: “The [specialist] in charge is always abroad….there should be another doctor to do the work when [they] go away.” Her husband’s “bypass” had been postponed several times during the past two years.

Public facilities lacked basic amenities, including sanitary facilities. Damayanthi (1/S), a retired janitor, spoke plainly: “Going to a toilet in a government hospital is like going to hell.” Rifana (2/Mu) concurred: “When [we] go to hospital, [we] need to cover our noses, the smell! It’s so unclean!” As a result, (5/S), previously a manager in the apparel industry, was adamant she would never enter a public hospital: “It’s so dirty, smelly… the toilets are waaaaa, and even the beds are like very close, there’s no privacy, everyone can see what is happening there.”

Given that the public healthcare system was under resourced and involved lengthy wait-times and numerous other inconveniences, it is reasonable that women viewed the private healthcare system as appealing and desirable. However, women’s experiences suggest that private ‘choices’ were not available to those with limited economic resources. These women were compelled to stay with the public system and spend excessive time obtaining basic healthcare for themselves and their families. In other words, the gendered character of privatization manifested differently across social class with economically disadvantaged women disproportionately burdening the time costs of accessing public healthcare. As we shall see next, others overcame some of these challenges by accessing the ‘alternative’ private system.

7.2 An array of private ‘choices’?

The private sector expanded the range of ‘choices’ available to a circumscribed group of middle to upper-middle class women, and, to a lesser extent, others. They visited private family doctors and specialists on a walk-in basis, at their convenience, purchasing laboratory tests and medicines, which were readily accessible, although for a price, at private diagnostic centres and

91 pharmacies. Among them, a minority enjoyed superior inpatient amenities at private hospitals with a privileged few traveling to Colombo to access luxury hotel-like services. Notably, many women representing the upper-middle class group could access private consultations and services because of their social ‘connections.’

Women with work- or family-related demands on their time valued the shorter wait-times and ‘flexibility’ associated with private family doctors. Amali (3/S), who ran a breakfast bar, always preferred to see her family doctor: “If we go private, we can get things done quickly.” This was also why Mala (5/S), a retired teacher, chose to see her family doctor: “If you go to the hospital you have to wait for a long time, you have to stay in the queue.” Namali (4/S), a bank executive who worked late hours, consulted her family doctor after she came home from work: “[Our family doctor] is available till about 8 [pm], and [he’s] close by.” Meanwhile, Muditha (4/S), a self- employed seamstress, often dropped in at her family doctor’s office on her way home from the supermarket. Visiting a family doctor was as appealing for busy homemakers like Thushari (4/S), a mother of young twins, who appreciated the short wait-times at her family doctor’s clinic: “With my little ones, waiting at the [government] hospital [is difficult].”

Private specialist consultation centres or ‘channelling centres’ offered access to a wide range of specialist ‘choices’ for women who afford the consultation fee. Indeed, specialists were the first point of contact within the healthcare system for many women representing the upper-middle class group. Kishani (5/S), who worked part-time in a family-owned business, believed channelling was the “normal procedure.” She always consulted a specialist when she had a health problem: “You see a doctor, you channel that person, you see that person, and do according to what he says.”

These women drew upon their experience and social networks in selecting a specialist in ways that best suited their needs. Shirani (5/S), a librarian on pension, had recently returned to Kandy after retiring from her job in Australia. She usually consulted a professor attached to the University: “I thought he’s well qualified…and has a lot of experience, so I chose him.” If she was not satisfied, she would get a second opinion. When Shirani developed laryngitis after a course of antibiotics prescribed by the Professor, she called up her “doctor friend” who recommended seeing an ENT surgeon. She simply called the channelling centre and booked an appointment with the ENT surgeon at the Suwasevana Hospital.

The private health sector in Colombo delivered creative channelling options that made the outpatient experience more ‘convenient’ for middle and upper-middle class working women. Himali (5/S), who had quit her job and returned home to Kandy after the arrival of her second baby, had used ‘e-channelling’ – a mobile service that allowed clients to reserve appointments – to schedule meetings with her obstetrician. Although more expensive, she felt it was worth the extra charge as she could avoid an additional traffic-congested trip to the hospital, generally required to reserve a specialist consultation:

I just have to call over the phone and book [the appointment] …they will charge the bill to my phone bill… 1800 (~CAD 15) … and they will give me a reference number …so I know my number, and usually they give a time on that… it’s very convenient…. whenever the doctor comes into the hospital we get a message saying the doctor is in the hospital, he came at 8.10 (pm)… so according to that we adjust our time and then we go to the hospital.

Many within the upper-middle class group saw more than one specialist in parallel. While Shirani consulted the Professor when she was in Kandy, for routine “check-ups,” she visited an internist at the (private) Lanka Hospital in Colombo. In her opinion, the physician at the Lanka Hospital was more “thorough”: “Even though you don’t have symptoms, she generally do [sic] a general check-up.” Sakuntala (5/T), a degree holder and homemaker, also consulted two specialists, one for her day-to-day health problems and the other for her diabetes. The first, a close friend of her “doctor son,” lived around the corner. She called him up whenever she or her spouse, who was homebound with a disabling condition, had a health issue. The second was an experienced internist she channelled once every three or four months at the (private) Lakeside Hospital in Kandy: “I’ve been going to [the doctor] for the last 15 years… [a doctor relative] put me on to him before [my son’s friend] came into the picture really…so I continue to see him.”

The ability to choose among specialists was appealing to Mala (5/S), a retired teacher. She compared the channelling system to organized family practice in other parts of the world:

[In] Australia, also, I think, States also [the] same, first you have to go to the GP and then only [you can see a specialist]….Here we can choose any consultant (specialist) and go…get a second opinion, third opinion…. you can go to several doctors and get the opinion…it’s easy…and we have confidence also...whereas outside you can’t do like that no, in other countries.

Mala preferred to decide for herself whether to see her private family doctor or specialist. Visiting her family doctor was easier because he practiced close by and wait-times were minimal. However, when she felt her problem required specialist attention, she channelled an internist,

93 who was “known” to her husband, at the (private) Amaya Medical Centre. Mala also saw a cardiologist at the Kandy Channelling Centre every three to four months. Although seeing two specialists was a waste of time in her opinion, Mala was keen to maintain a private ‘connection’ with a cardiologist to ensure timely access to the Cardiology Unit at the Kandy Hospital in the event of an emergency. The cardiologist would admit Mala directly to the Cardiology Unit at the Kandy Hospital, if and when necessary. Mala could thus avoid the regular (tedious) route to the Cardiology Unit via the OPD. Simply stated, such ‘connections’ allowed women like Mala to ‘jump the queue’ at public facilities, constituting a commodified form of social capital. 98

Unlike at a public facility where users were referred via the OPD to the specialist who happened to be ‘on call’ on that particular day, channelling centres offered direct access to a specialist of one’s choosing. Thushari (4/S) described how she ‘channelled’ a reputed obstetrician in the private sector when she was pregnant with twins: “He doesn’t do Caesars [unnecessarily], [he] somehow does it normal (normal delivery).” Furthermore, channelled consultations took place in the privacy of a comfortable consultation room rather than amidst the hustle and bustle of a crowded public clinic. Muditha (4/S), a seamstress, generally avoided public clinics because she felt she would not be able to talk to the specialist: “The big doctors are not there, that’s the problem, they don’t see [the patients] no. They are not in the hospitals… only a small one (junior doctor) will see [you].”99 Moreover, as Kishani (5/S) pointed out, waiting facilities at channelling centres were more comfortable: “I went and saw the government [neurology] clinic [at Kandy Hospital]; it was terrible ... if I had taken [my husband] there he wouldn’t have had a place to sit even.” Thus, channelling stratified healthcare experiences for those with and without means.

Interestingly, the private sector allowed a smaller group of highly literate upper-middle class women to manage their health affairs in ways that minimized encounters with healthcare providers. Shirani (5/S), a retired librarian, usually did some preparatory blood work before she channelled her specialist to avoid an unnecessary second consultation. Most recently, when she had a sore throat, she had some tests done on the morning of her appointment:

98 I examine this practice of ‘double dipping’ in more detail in Chapter 8. 99 Here, Muditha subscribes to the widely held perception that specialists are not available at government clinics. However, they are often physically present at the clinic/hospital, although given the large numbers attending public clinics, users meet specialists only on their first visit and when problems requiring specialist opinion crop up (my experience).

I did a full blood count and I found my ESR100 was high. So I knew I had an infection, so it was convenient and quicker. I took that to the doctor and he saw that my ESR was high ….I did the test in the morning... I picked up the result when I went to see [the doctor], and I showed the doctor, so he had solid information there.

Similarly, Mala (5/S) spoke of monitoring her blood pressure at home and testing her blood sugar and cholesterol levels at a private diagnostic centre when she felt it was necessary. Although her cardiologist had advised Mala to visit the channel centre at least once every three to four months for follow-up, she did not: “I haven’t been [to my cardiologist] for one year now {laughs} but I still use the same prescription and get the same medicine. I go for tests….if my report is bad only I feel like going [to see my cardiologist].” The widespread availability of private laboratories and pharmacies combined with lax regulation enabled such practices for this group.

The hotel-like inpatient amenities that were on offer at private hospitals were accessible only to a small group of wealthier users. With one exception,101 all who shared recent experiences of private inpatient care were residents of Kirideniya (4, 5). Rather than driven by expectation of technical quality, these women accessed these services chiefly to enjoy the amenities on offer. As Suwini (5/S), a medical officer, explained: “In the private sector, it’s more comfortable to the patient. You see the [public] wards, and go see the private hospitals, especially in Colombo, the rooms are more comfortable to the patient, that’s the main thing in the private sector.” Himali (5/S), retired from the apparel industry, had recently delivered at the Nawaloka Hospital in Colombo, and described the hotel-like services that were part of an all-inclusive 3-day caesarean section “package”: “We had a bell in the room so whenever you press the bell someone will be there. Either a nurse or an attendant…immediately will come.”

The commercial hospitals in Colombo were perceived by many to provide a superior standard of (private) care than was available in the private sector in Kandy. As Suwini (4/S), the medical officer, explained: “The infrastructure is better in Colombo … even the quality of like lab tests and, yeah, the ICUs and stuff, they are better in Colombo.” These hospitals offered access to technologically advanced procedures that were not available in the public sector. Kishani and her

100 Erythrocyte Sedimentation Rate, a non-specific blood test that detects the presence of inflammation. 101 The exception was Fathima (2/Mu) who spoke of having had her womb removed at Suwasevana Hospital. Fathima’s class location did not quite “fit” with others in the Bogoda group; she had access to more economic resources as she was a retired clerk, and her husband ran a business in Colombo. Even so, the primary reason Fathima used private inpatient care was the inaccessible washrooms at public hospitals (see 7.4).

95 husband went to the Asiri Central Hospital in Colombo after they learned from a “doctor friend” that minimally invasive neurosurgical technologies were available there: “[Our friend] said, you know, if you go to do it in Kandy [Hospital]…the recovery [time] is going to be more, so you get it done from Colombo.” Kishani expressed satisfaction with the services her husband received during his stay at the private hospital:

That hospital is one of the best hospitals in Colombo …and [neurosurgery] ward was fantastic. They were fully trained, very professional … the way they cleaned him on the bed… from day one he was given physiotherapy, all that was done, so it was done very very [emphatic] professionally, and those nurses were fully trained. The medicines were given on time, the doctor came right on time, everything was done in [a] proper way. I didn’t have to do anything …it was a pleasant [experience].

In short, the private health sector provided a range of ‘convenient’ services to women who could afford to pay for healthcare. Its presence enabled these women to better organize their healthcare seeking activities around household chores and work, and widened their selection of providers. A small group of middle to upper-middle class women accessed private inpatient services to enjoy hotel-like amenities and technologically advanced treatment options that were not available in the public system. Thus, the presence of private healthcare enabled some users to upgrade their healthcare experiences, although, as we will see next, the ability to take advantage of these ‘choices’ was limited by financial constraints and quality concerns.

7.3 Limits on ‘choice’

7.3.1 Economic exclusions

Despite the availability of numerous private family doctors and a channelling centre at walking distance, the private sector did not extend the ‘choices’ available to economically disadvantaged women in any significant way. A recurring theme that emerged in the focus groups and interviews with Neluwa/Jagoda (1) residents was that the public system was the only feasible option available to them. As Damayanthi (1/S), a retired janitor, pointed out, even visiting a private family doctor was out of the question for her: “I never go [private] actually…I mean when [you] consider our income {laughs in embarrassment}.”

Using the public Dispensary set the women from Neluwa/Jagoda (1) apart from their neighbours.102 Its location was a critical consideration for Damayanthi (1/S) who saved her meagre household earnings by walking: “[We] don’t need to spend money on the bus.” When the Dispensary’s walk-in service was not available, most Neluwa/Jagoda residents accessed an outpatient department (OPD) at a public tertiary care centre, resorting to private family doctors only when public avenues were unavailable. Nafeesa (1/Mu), a homemaker, saw a family doctor when the Dispensary was closed because, at such times, visiting a private clinic was more convenient than traveling all the way to a public hospital. Only two women, Chamila (1/S) and Shereen (1/Ma), in the working class group regularly visited a private family doctor. Chamila, a seamstress, consulted her family doctor as the consultation charges were reasonable: “[The doctor in our area] charges about 200 rupees (~CAD 2)103... an amount that a poor person can afford.” Meanwhile, Shereen preferred to see the family doctor who practiced close by as traveling to the Dispensary was difficult owing to a debilitating physical disability.

The working class women who visited private family doctors could not do so on a long-term basis. Chamila (1/S) consulted her family doctor only when she expected the health issue to require no more than a single visit. She favoured the OPD at a public hospital if she anticipated additional visits or treatment. As Chulani (1/S), a food supplier to a local cafeteria, explained: “We can go once, private…after that if [we] keep going, [we] face problems getting things done.” She recalled that she did not bother to do the tests prescribed at the Dispensary on her last visit:104 “I couldn’t afford to spend that much]. [I] have four [children] to send to school.” Besides, Damayanthi (1/S) preferred to go directly to the OPD over wasting money moving between public and private sectors since family doctors inevitably directed patients to a public hospital in the event of a complication: “When things get bad everyone ends up at a government hospital.”

Needless to say, women from Neluwa/Jagoda (1) accessed the public sector for chronic health problems because they simply could not afford private care on the long-term. Nafeesa (1/Mu) brought her son to the Kandy Hospital’s Eye Clinic to monitor an eye problem, and attended three

102 Only one participant residing outside Neluwa/Jagoda (1) reported frequent use of the public dispensary. This was Ruwani (4/S) from Kirideniya, who, on principle, accessed the public system, whenever possible. Others who used public primary curative care from Dehioya (3) and Kirideniya (4, 5) preferred the outpatient department (OPD) of the Peradeniya Hospital, which was closer than the Dispensary to the southern side of the Division where they resided. 103 The daily wage of a female wage labourer in Kandy is about SLR 600 (~CAD 5). 104 Primary medical care units generally do not offer diagnostic services. Healthcare users are referred to a laboratory at a public facility or the private sector (see Chapter 8).

97 clinics herself for the follow-up of a genetic disorder. She emphasized that she could not afford this treatment in the private sector. Similarly, Chulani (1/S), a mother of four, had relied completely on the public MCH service during her pregnancies, and still consulted the field midwives when she had questions. Unlike her neighbours from Bogoda (2), Dehioya (3), and Kirideniya (4, 5), she did not consult a private specialist in parallel (details in Chapter 8).

Many economically disadvantaged women from Bogoda (2) and Dehioya (3) did use the private sector for outpatient care, although anything more than consultations with private family doctors, and perhaps an occasional visit to a channel centre, was beyond their reach. Fara (3/Mu), a homemaker, channelled a dermatologist at the (private) Amaya Medical Centre for a rash that did not go away despite treatment from a private family doctor. Yet, she could not afford to have her diabetes, blood pressure, and cholesterol followed up in the private sector: “[They] take 1000 [rupees] (~CAD 8) for channelling … [they] charge for the medicines separately105… [and] if there is a serious problem and [they] ask you to get admitted, [they] take lakhs,106 it seems.” As a result, Fara used the public system: “People with money will go [private], those without money won’t.” Similarly, Badra (3/S), a busy NGO worker, had no time to attend a public clinic for her diabetes, and, instead, consulted her private family doctor who practiced just a few steps away from home. However, she made sure to attend her husband’s neurosurgical clinic to obtain an expensive drug through the hospital’s pharmacy: “If [I] buy that [medicine] private, [I] would have to pay a big amount, 2700 rupees (~CAD 23) each…[he] needs four for one month.”

The private sector did not present an economically feasible inpatient alternative for most participants, particularly Neluwa/Jagoda (1) residents. As Shereen (1/Ma) explained, a state hospital was really the only option for inpatient care for most residents of her neighbourhood: “We may have to buy medicines from outside, but at least we can stay at the hospital for free.” Working class women (and others) shared experiences at public hospitals where they received complex medical interventions, spending little out-of-pocket. Nafeesa (1/Mu) described her lengthy stay at the Kandy Hospital where she was treated for a brain tumour: “I bought a few medicines from [a private pharmacy]…and [they] sent me from Kandy to Peradeniya [Hospital] for the MRI; everything else was done [in the Kandy Hospital].” Similarly, Nuzrath (1/Mu) entered the

105 Fara is probably more familiar with private family doctors who charge a single consultation fee that includes treatment/medicines. 106 One lakh = SLR 100, 000 = ~CAD 850.

Peradeniya Hospital with excessive menstrual bleeding and had an urgent blood transfusion followed by a surgical procedure with no direct costs associated with her treatment.

The availability of ‘free’ public inpatient services was a relief for those with complex health issues. Chamila (1/S), primary caregiver to her mother who had a chronic kidney problem, highlighted the dire situation they would face if dialysis were not available free-of-charge at the Kandy Hospital: “At Lakeside (private hospital), it costs 12, 000 rupees [~CAD 100] per day for four hours.” Two sessions of dialysis would have cost Chamila her reported monthly income. For the same reason, Fara’s husband was on the waitlist for “bypass” at the Kandy Hospital as they could not afford to have it in the private sector where the procedure was available at a hefty cost:

If [we] had money, we could have done [the bypass] somewhere else (private sector)… If we spent 8 lakhs (~CAD 6500) [on a bypass], I have two daughters, they need to be married. [We] can’t commit everything to this, no… so we have use the government [service].

Indeed, the cost of a “bypass” would have likely exceeded their annual income.107

A few lower-middle class women had explored private options, and abandoned them for economic reasons. When Badra’s husband was diagnosed with a brain tumour, the neurosurgeon at the Kandy Hospital informed them that the risks of surgery were too high for the procedure to be undertaken in the public sector.108 He directed them to a reputed neurosurgeon at the Asiri Central Hospital in Colombo. After meeting with the neurosurgeon, Badra realized they simply could not afford the procedure, which would have cost them well over their annual household income. In the absence of alternatives, Badra’s husband returned to the Kandy Hospital where the neurosurgeon eventually undertook the procedure (successfully). Meanwhile, Fathima (2/Mu), who had her arthritis followed up in the private sector, abandoned the latter when she needed joint replacement: “[Doctors] look at us and think we have five lakhs [~CAD 4150] in hand…although [the doctor] said five lakhs, [the procedure] costs more than seven lakhs [~CAD 5800].” While this amounted to over half her reported annual household income, by the end of the

107 Fara did not disclose her monthly household income. Her spouse, an electrician, may have earned a monthly salary of anywhere between SLR 30,000 to 50,000 (~CAD 250-420). 108 Badra was not informed of the reason. However, the most advanced neurosurgical services in Sri Lanka are known to be located at the Asiri Central Hospital in Colombo (Hafi 2017).

99 second phase of my fieldwork, Fathima was awaiting joint replacement at Sri Jayawardenapura General Hospital, a state-subsidized, government-owned fee-levying facility.109

Even Kirideniya residents (4, 5) expressed concerns about the costs of private healthcare. Ruwani (4/S), a retired bank clerk, complained about the soaring channeling fees in Kandy: “Thousand [rupees] (~CAD 8) for the doctor [for a single consultation)…actually doctors earn way too much no? It’s a bit unfair.” Shirani (5/S) observed that the situation was worse in Colombo where channeling fees could amount to double those in Kandy: “The best ones (doctors) are taking 2000 rupees (~ CAD 17) … which includes may be 300 or 400 [rupees] (~ CAD 2.50-3.50) to the hospital.”110

Despite the availability of hotel-like amenities and state-of-the-art biomedical technologies at commercial hospitals in Colombo, most Kirideniya residents (4, 5) could not afford these services. Kishani (5/S) and her husband struggled to pay their bill for a neurosurgical procedure at the Asiri Central Hospital, which amounted to 4.5 lakhs (~CAD 3750): “It was really too much for us, we took one year to recover from that.” Notably, the only two participants who regularly accessed private inpatient care in Colombo – Himali (5/S) and Shirani (5/S) – were covered by employer- based health benefits. Himali’s 3-day caesarean section “package” cost SLR 200, 000 (~CAD 1650), but she claimed SLR 120,000 (~CAD 1000) from her employer, and paid the remainder, which amounted to much less than her reported monthly household income. Meanwhile, Shirani (5/S) had comprehensive health insurance coverage through her spouse’s retirement package from an international organization. 111

Thus, access to existing private ‘choices’ was very limited for working class residents because of their (in)ability to pay for services. Although some lower-middle class users accessed private outpatient services on a regular basis for minor health concerns, they too used the public system when they anticipated resource-intensive services would be needed. Although middle/upper- middle class women routinely used private outpatient care, the out-of-pocket spending associated with inpatient care at private hospitals remained a challenge for many. As we shall see next, the

109 Sri Jayawardenapura General Hospital (SJGH), located in the suburbs of Colombo, was established in 1984, a few years after the adoption of an ‘open economy,’ with assistance from Japan. The hospital adopted a new model of healthcare delivery by providing government subsidized services on a fee-levying basis (SJGH 2015). It is still the only hospital of its kind in Sri Lanka. 110 The channelling fee includes the consultation fee and a facility fee. 111 Shirani (5/S) was the only participant who reported having insurance coverage.

100

‘choices’ that opened up in the presence of private healthcare for those who could afford to pay for services were limited by their quality.

7.3.2 Quality concerns

The perceived poor quality of private services in Kandy deterred wealthier women from using them. In fact, none of the women who used the private sector to a greater degree suggested that it offered technically superior services. Conversely, many among them emphasized that the public sector provided a ‘safer’ service. As Suwini (5/S), a medical officer who frequently used private healthcare, explained: “I think the service [in the public sector] is much more better… it can be improved…[but] the service in terms of quality, like safety, it’s better… and it’s certainly a lot less expensive.” Kishani (5/S), a seasoned private healthcare user, concurred: “The procedure is long …still the needful will be done.”

Private outpatient services were poorly organized and referral systems almost non-existent. Very few women relied on their private family doctors for specialist referral. Namali (4/S) had visited her family doctor when she found out she was pregnant. He recommended she see an obstetrician in the private sector whom she subsequently channelled. Muditha (4/S) also depended on her family doctor for advice on when to see a specialist. She had recently met a cardiologist on the suggestion of her family doctor as her high cholesterol levels had been resistant to treatment. However, for the most part, women went directly to a channel centre.

Despite the widespread use of channelling, women from Kirideniya (4, 5) highlighted several concerns. Wait-times were often lengthy owing to the lack of an effective system to schedule appointments. For Sakuntala (5/T), this was the “biggest drawback” in the channelling system:

When you channel doctors, they should have the courtesy to give you fixed appointments…when you call and get a number, and they say [come at 4 pm], then we know that they are not going to, so we go at 4.30, and then you have to wait and wait and wait and wait …and the doctor doesn’t put in an appearance till 2 hours later, because they go to several places.

Most specialists in Kandy operated from several channel centres,112 moving from one to the next after seeing all patients with appointments at each centre. This essentially meant that users waited an indefinite period until the relevant specialist turned up at the channelling centre. When

112 Some specialists consult at one channel centre, which may explain the varying wait-times experienced by users.

101

Ruwani’s (4/S) husband developed a hearing problem, they visited the (private) Amaya Medical Centre because getting a “number” at the Kandy Hospital’s ENT Clinic involved too much hassle. However, they had to leave before the specialist arrived because they had another engagement later that evening:

[They] asked us to come at six o’clock [pm]. We got there before six, about five minutes [before], we [live] close by no…so we went five minutes early, paid, and waited. At 7.15, no doctor, we waited till 7.30 … and left. Why did we have to suffer even after paying?

Indeed, Mala (5/S) preferred to see a private family doctor for her ‘minor’ problems as she felt ‘channelling’ was inconvenient: “First we have to go and get a number (reserve an appointment) in the morning… and then again to see [the doctor], we have to wait for about 2 hours.”

Women worried about the financial motivations that drove specialists to engage in channel practice and complained about the short consultation times that resulted. As Anushka (5/S), the manager of a private school, remarked: “It’s rarely a [specialist] will be attentive to you for about 15 to 20 minutes…they will just make you sit, put the seal, the signature, won’t ask us anything, out with the medicine.” Mala (5/S) voiced similar sentiments: “Some of the doctors… they are concerned of money … they will get us [to come] three four times … [we don’t have] a lot of confidence.” Ruwani (4/S) believed the generous incentives provided by pharmaceutical companies influenced prescription practices: “[We] know how reps (representatives of pharmaceutical companies) convince specialists to prescribe their drugs.”

The quality of inpatient care at private hospitals in Kandy were evidently of questionable quality. The development of private hospitals had stagnated, perhaps owing to the limited number of users who could pay for services in Kandy (Dayaratne 2013). As Suwini (5/S), a medical officer, observed: “There are no new hospitals [in Kandy]…when we were schooling there was Lakeside, KPH (Kandy Private Hospital) and Suwasevana in Kandy, now [also] only those three are still there.”113 These hospitals relied on the public sector for its physician workforce, which meant that specialists were frequently not available to their private inpatients. As Mala (5/S) explained: “The consultant (specialist) comes [to see their private inpatients] either very early in the morning … or they come very late, after they finish their [channelling].”

113 This list is not comprehensive; Suwini overlooked a couple of private facilities that offer inpatient care.

102

When the attending consultant was not present, private hospitals did not offer back-up specialist cover. Muditha (4/S) experienced this first-hand at the (private) Suwasevana Hospital when she brought her mother with symptoms of a stroke: “It was a [public holiday]… we were there from morning, but they did not give any medicine. [They] said the doctor (specialist) had to come.” This was also why Namali (4/S), in her eighth month of pregnancy, felt delivering at the Peradeniya Hospital would be safer: “I can’t be sure [the obstetrician] will come [on time] if I go private… my dates are too close to the New Year holidays.” In this vein, Suwini (5/S), the medical officer in the group, suggested private inpatient care was unsafe and less accountable as it was organized around individual specialists rather than a team of healthcare providers:

[In a public hospital], there is always a doctor assigned to a patient. There will be better monitoring of patients, and there is someone accountable to the patient all the time. We don’t see that in the private sector [in Kandy] …there is usually a single consultant (specialist) looking after a patient, and he’s a very busy person, so he wouldn’t always be monitoring the patients.

Although private facilities employed nursing staff on a permanent basis, women highlighted the poor quality of nursing care in the private sector (in Kandy). According to Namali (4/S), a bank executive, private facilities employed less-skilled workers as a cost-containment strategy: “We can see that the nurses and assistants are not knowledgeable, they are mostly untrained young people from distant areas who need jobs.” Kishani’s (5/S) experience at the Suwasevana Hospital following caesarean section reflects the nursing standards at private facilities:

[The nurse] didn’t come to check for anything. She didn’t. In the night… they have to check pressure or whatever, whatever. They didn’t do any of that. She was just seated and waiting, she didn’t come to the room at all… so the baby was inside the room… they had cut me a lot and I couldn’t move. If my mother was not there [to help], [the nurse] was not going to anyway help, so it would have been a difficult situation, no, if I didn’t have help.

For these reasons, the private sector was not perceived to be a ‘safe’ alternative under critical circumstances. Sonali (4/S), a director of a family-owned business, insisted the Kandy Hospital was safest for her mother who had a “serious” heart problem: “Things are done methodically there… there are problems with the toilet, at times [you] may not even get a bed … but the attention she gets is better there.” Mala (5/S) also felt the Cardiology Unit at the Kandy Hospital was the safest option in the event of a “heart problem”: “Cardiology, I think [at the Kandy Hospital]

103 is better… than going to a nursing home114 …. [A public] hospital is the better place, if you have a big problem.”

This was why Suwini (5/S) generally preferred the public system for surgical procedures: “I would feel comfortable getting [them] done from a government (public) hospital because, like I said, there’s always someone to look after you, there is more accountability.” Along these lines, Namali (4/S), in her eighth month of pregnancy, believed the public system was a better option for her delivery as it was better set up to handle an emergency: “If something happens… the nurse will talk with the [junior] doctor. There is always one, two or more nurses in the ward, there’s a doctor, so [even if the specialist is not available], they can decide what is to be done.” Recalling the experience of a colleague, she felt the private sector in Kandy did not have the capacity to deal with an emergency: “[They] transferred [my colleague] at the last minute [to a public hospital].”

Thus, the healthcare ‘choices’ available in the private sector were limited not only by associated out-of-pocket payments, but also their perceived quality. Given the unavailability of round-the- clock specialist care combined with poorly trained nursing staff and lack of an organization structure that could deal with an emergency, the private sector did not present a safe ‘choice’ for those who could afford to pay for private care. In the next section, I consider the healthcare ‘choices’ available to users on the margins to illustrate how their experiences make impossible a straightforward class-based interpretation of healthcare ‘choices’ under privatization.

7.4 ‘Choices’ on the margins

Rather than exercising ‘choice,’ opting for private healthcare was a matter of necessity for some users. Economically disadvantaged Muslim women reported neglect, abuse, and/or stigmatization in the hands of healthcare providers in the public sector. Women with disabilities experienced challenges accessing public facilities, which did not have basic infrastructure in place to serve their needs. As a result, women representing these groups had no ‘choice’ but to opt for alternatives, when available and accessible. Many among them reported turning to the private sector to address deficits in public healthcare.

114 A term used by healthcare users in Sri Lanka to refer to small private hospitals with limited inpatient facilities.

104

Tamil-speaking Muslim women visited the private sector whenever possible to avoid abusive healthcare workers who could not speak their language. Feroza (2/Mu) highlighted the unavailability of bi-lingual healthcare workers in the public sector: “If a person from Trinco (Trincomalee district115) is transferred to Kandy [Hospital], [they] are very scared. Only Tamil is used [in Trinco], so when they come here they face difficulties as they don’t know the language.” Nuzrath (1/Mu) had not experienced the “language problem” herself as she could converse fluently in Sinhala, but she observed that others were less fortunate: “If they (patients) cannot understand Sinhala, they get scolded [by the staff].” Fathima (2/Mu) emphasized that the unavailability of bi-lingual healthcare workers violated the “language policy,” which stipulates that public servants must be competent in Sinhala and Tamil: “All government workers, including minor employees.” According to Sabina (2/Mu), private providers tried to communicate with users in their mother tongue: “The [private] doctors will somehow manage to speak in Tamil.” 116

Apart from language concerns, Sabina (2/Mu), who wore a hijab,117 suggested that public sector healthcare providers showed prejudice toward markers of ethnicity: “I don’t know why, [the shawl (hijab)] is such a big issue, as soon as [the staff] see someone in a shawl, it’s as if [they] get possessed.” She was routinely asked to remove her “shawl” at public facilities, including the Kirideniya MCH Clinic: “The doctors and even some of the midwives sort of don’t like [us wearing the shawl]…“Take it off” [they say].” According to Sabina, her friend who wore niqab118 frequently encountered abuse in the public system. She recalled visiting her friend who had given birth at the Kandy Hospital where she witnessed a nurse in the maternity unit ordering her friend to remove her niqab: ““Remove [it]!” [the nurse] said, “Everyone has to [dress] the same here… what are you doing… wearing that thing and fooling around,” she scolded [her].”

Since the public sector lacked mechanisms to hold healthcare providers accountable, Muslim users were helpless in such circumstances. As Fathima (2/Mu) explained, “In the government [sector], [you] cannot [talk back to staff], they will simply say they don’t know [what you are talking

115 A Muslim/Tamil dominant district in eastern Sri Lanka. 116 Muslims in Kandy generally speak Tamil at home, and Malays speak Malay. The Muslim and Malay women who participated in the study may have received Sinhala medium instruction in school, explaining why they were conversant in Sinhala. 117 “A head covering worn in public by some Muslim women” (Oxford University Press 2017). 118 “A veil worn by some Muslim women in public, covering all of the face apart from the eyes” (Oxford University Press 2017).

105 about].” Sabina (2/Mu) believed the repercussions of taking any action could adversely impact healthcare services for others as public sector workers frequently resorted to trade union action: “[We] cannot speak with the management, [we] cannot speak with anyone, the minor staff will not accept a wrong was done, [they] will go on strike instead.” Meanwhile, Feroza (2/Mu) feared the consequences of confrontations with healthcare workers: “Say I tell them off, when I am in hospital for an operation, what if they give me a big cut for something that needs a small one?” Notably, Sinhala women did not express such anxieties and fears.

These varied concerns impelled economically disadvantaged Muslim women to access the private sector, which they felt was more welcoming of ethnic difference. Unlike their lower-middle class counterparts in Dehioya (3), who used a mix of public and private outpatient care, all but one participant in the Bogoda (2) group routinely consulted private family doctors or channelled specialists for outpatient care.119 However, this ‘choice’ of accessing private healthcare was not available to Muslim and Malay women from Neluwa/Jagoda (1). For example, Siththy (1/Ma) would have preferred to consult a private family doctor as she felt the drugs dispensed by them were of superior quality.120 However, this was not an economically feasible option for her. Noteworthy is that none of the women representing ethnic minorities from Neluwa/Jagoda (1) expressed concerns about their marginalization within the public system. In fact, Nafeesa (1/Mu), claimed otherwise: “We get all the facilities, there are no ethnic divisions, there is nothing like that.” While these women may not have felt comfortable sharing their experiences with me given my social location, their expressions may also reflect their resignation toward accessing the healthcare ‘choices’ that were financially accessible to them.

Although not an objective of this study, the stories shared by women with disabilities depicted lacking social services and a public system poorly set up to attend to people with disabilities. Shereen (1/Ma) preferred to see a private family doctor for “something like a fever or a cold” as traveling to a distant public facility was difficult owing to a debilitating physical disability. Any savings on consultation charges at the public Dispensary would invariably go toward hiring a 3-

119 Fathima’s reported household income was admittedly higher than others in this group, and Sabina’s spouse worked in the Middle East. However, even Feroza who reported a very low household income, routinely accessed private healthcare for her chronic arthritis, spending about SLR 5000 (~CAD 40) at each visit. 120 Concerns about the quality of medicines prescribed in the public sector were expressed by many, and draw on a long and contentious public debate on pharmaceutical reforms, which began with the Bibile and Wickramasinghe reforms of 1972. While the latter were abandoned in 1976 under pressure from Big Pharma (Lall and Bibile 1977), resistance to the reintroduction of these reforms manifest in widely propagated myths about the poor quality of lower cost medicines supplied to the public system by the State Pharmaceutical Corporation (Balasubramaniam 2012).

106 wheeler since she could not walk to the Dispensary. Shereen had stopped attending her physiotherapy sessions at the Peradeniya Hospital some years ago because travelling back and forth between the physiotherapy unit and home was difficult: “With time, I was unable to go [to the hospital]. [I] would fall on the road, and I was scared to go alone …[I] did not have anyone to help.” 121

On the other hand, Siththy (1/Ma), who had a less incapacitating physical disability, walked all the way to the public Dispensary. Needless to say, she could not afford regular consultations with the private family doctor who practiced close by. The Dispensary’s location at walking distance was an important consideration for Siththy since public transport was not disability accessible: “I can walk bit by bit to the Suduhumpola [Dispensary]; I can’t walk all the way to Peradeniya.” While the experiences of women like Shereen and Siththy reflect the absence of accessible transport for people with disabilities, they really felt the absence of a community-based system of ‘free’ curative care. As Siththy remarked: “[We] don’t have a dispensary (allopathic) close by, [we have] only an Ayurveda [dispensary]…it would be good if we had a village-level [allopathic] dispensary.”

Public healthcare facilities lacked accessibility services for people with disabilities. Muditha (4/S) felt she had no alternative but to bring her mother to a private hospital after she had a stroke: “[We] couldn’t take her to the [public] hospital in her condition, toilets, this and that, so [we] went to [Suwasevana].” This was also why Fathima (2/Mu), a woman with chronic arthritis, avoided the public system: “I can’t squat…and there are no commodes in the hospital. Can I go?” She complained that no special consideration was given to users like herself who experienced difficulty navigating the queues and wait-times at public hospitals. As a result, she opted for private outpatient care.

In sum, the experiences of economically disadvantaged Muslim women and people with disabilities complicate class-based interpretations of healthcare ‘choices’ under privatization. These groups were marginalized by the public system in obvious ways. Some of them could address at least some of these concerns in the private sector. However, these private ‘choices’ were not up for grabs for all users, particularly those who could not afford to pay for them. While

121 Shereen lived with her mother who worked as a security guard and was the breadwinner in the family.

107 their experiences suggest that interlocking dimensions may crucially shape healthcare access in Kandy, they also signal the limits of ‘choice’ under privatization.

7.5 Discussion

The findings of this chapter interrogate the premise of exercising ‘choice’ in healthcare. Often depicted as an expression of individual freedom, in reality, the ‘choices’ available to ‘consumers’ are always constrained by social relations. As Brown (2016:3) explains, neoliberalism produces inequality and a certain degree of choicelessness:

as de-regulation eliminates a range of public goods and social security provisions, unleashes the powers of corporate and finance capital, and dismantles classical twentieth century solidarities among workers, consumers, and electorates….[generating] intensely unprotected individuals, persistently in peril of deracination and deprivation of basic life supports, [and] wholly vulnerable to capital’s vicissitudes.

The language of ‘choice’ renders invisible the impact of neoliberal health reforms at the ‘micro level.’ As the findings of this study suggest, the very ‘demand’ for alternative healthcare options has arisen in the context of weakening public healthcare systems under neoliberalism. In Kandy, the state’s failure to invest in the public health sector had increased the time costs associated with public healthcare, and pushed women to seek alternatives, creating niches for private market expansion. Yet, the extent to which women could take advantage of the available ‘choices’ was shaped by class and ethnic relations.

The ‘choices’ that opened up for working class and lower-middle class users were restricted by their (in)ability to pay for services. In reality, only middle/upper-middle class users could exercise ‘choice’ in accessing private healthcare. They picked and chose among private family doctors, specialists, and various private healthcare facilities, in ways that best suited their needs. On the other hand, the masses who contributed more to the subsidization of private healthcare through indirect taxation were largely excluded from its so-called ‘benefits.’

A straightforward class analysis was disrupted at two levels. For one, middle/upper-middle class users relied extensively on the public sector for inpatient care. They struggled with out-of-pocket payments with only a small minority in this group routinely accessing inpatient care in the private sector. The inpatient ‘choices’ available to these women were limited by the perceived (poor) quality of private healthcare in Kandy. Private facilities lacked round-the-clock specialist care and a competent non-physician workforce, and evidently did not have the capacity to deal with

108 emergencies, deterring many from accessing Kandy’s private health sector in critical situations. However, a small group of women travelled to Colombo to take advantages of the hotel-like amenities and services on offer at commercial hospitals.

A class-based interpretation of public and private healthcare use was also troubled by the experiences of socially and economically disadvantaged users for whom the public system remained an unappealing option. Economically disadvantaged Muslim women and users with disabilities shared experiences of marginalization within the public sector. They accessed private healthcare, whenever possible, to address at least some of these concerns, although the ability to do so was contingent on class location. While these findings suggest that public systems may not be uniformly liberatory as depicted in the critical public health literature, such marginal experiences imply that exploring interlocking dimensions of inequity may yield critical and hitherto unexplored aspects of healthcare access under privatization.

A crosscutting theme that emerged from my findings was the gendered character of healthcare access in Kandy. The state’s ‘withdrawal’ from healthcare affected women as primary caregivers in distinct ways, a concern highlighted within Marxist feminist scholarship (Bakker and Gill 2003a). Yet, the burden experienced by women representing various social groups was very different under privatization. The time costs associated with accessing healthcare were higher for economically disadvantaged women who relied chiefly on the public system. Women with means accessed private healthcare to reduce these time costs, which allowed them to better organize their lives around household chores and income-generating activities.

Taken together, my findings suggest that the ‘choices’ that opened up for women in the presence of private healthcare were constrained by social relations. Rather than expanding ‘choices,’ healthcare privatization benefitted the middle classes in limited ways, while compromising existing public healthcare options for swathes of the population that were excluded from the private health sector. However, for some users on the margins, the private sector constituted the only appealing alternative given the public system’s multiple exclusions. Although private healthcare may have been less marginalizing for these groups, being compelled to spend out-of-pocket to receive a respectful service raises broader questions of equity vis-à-vis the public system. In what follows, I complicate this binary narrative of access by exploring the ways in which women mixed public and private healthcare as they struggled to access healthcare for themselves and their families within Kandy’s ‘mixed health system.’

109

8. Negotiating Kandy’s ‘mixed health system’: Hybrid arrangements and (in)equity in healthcare

Chapter overview: In the last chapter, we saw that access to the public and private healthcare options available to Udawatta residents varied by social location. Here, I complicate this picture by delving into the strategies of survival women employed to access health services as the state receded from healthcare provision. In doing so, I consider where women went for healthcare (research objective 1), their impressions of the various facilities and services they used (research objective 2), the ways in which women navigated public and private healthcare (research objective 3) and how their pathways differed by social location (research objective 4). The chapter sheds light on three hybrid arrangements122 that operated within Kandy’s ‘mixed health system.’ First, working class women were compelled to consume private services within a resource constrained public system. Second, middle/upper-middle class users, economically disadvantaged Muslim women, and marginal others, consulted public sector physicians in the private sector to upgrade the service they received at public facilities. Third, regardless of social location, women combined public and private services to economize on healthcare.

8.1 Compelled to consume

Insufficient state investment in the public sector had left it bereft of basic facilities and services. Healthcare users were forced to purchase the deficit of services in the private sector. They were directed by public sector providers to private facilities to obtain diagnostic tests, medicines, and other medical supplies that were in short supply at public facilities. This practice was especially burdensome for working class residents of Udawatta. As Damayanthi (1/S) observed: “We go to a government hospital, and we have to attend to our needs outside (private sector).”

The laboratories at tertiary care centres did not have the capacity to provide comprehensive diagnostic services. Women paid for private laboratory services while they/their family members were admitted to a public facility. Damayanthi (1/S) spent over SLR 11,000 (~ CAD 90) – more than half her reported monthly household income – at the Asiri Laboratory, located adjacent to the Kandy Hospital, when her son was admitted at Kandy Hospital with dengue fever. She carried

122 I use the term ‘hybrid arrangements’ to refer to the varying practice of combining public and private healthcare services within Kandy’s ‘mixed health system.’

110 blood samples between Asiri and the hospital at all hours: “They sent us [to Asiri] at 12 and 1 at night.” Damayanthi and her husband scraped together the necessary funds despite their paltry household income: “We somehow did the tests because we wanted to save our child’s life.” Sabina (2/Mu) spent over two thousand rupees (~CAD 17) on blood tests at Asiri after her father was rushed to the Kandy Hospital with chest pain. She had to borrow from a friend as she usually did not carry this much money with her. Getting the required tests involved two trips between the Kandy Hospital and the Asiri Laboratory for Sabina: “[The Asiri staff] gave me a syringe to take back to the [hospital] ward. [The nurse at the hospital] drew out the blood, and [I] had to take it back again to [Asiri]… [it was] such an inconvenience.” While running back and forth between public and private healthcare facilities added to women’s caregiving activities, the associated costs were a strain on household incomes, particularly for those who did not have funds to make ends meet.

The gaps in laboratory services were especially burdensome for users with chronic illnesses that required close monitoring. Premalatha (1/S), who had diabetes, high blood pressure, and elevated cholesterol levels, remarked that the laboratories at tertiary care centres did not even have the capacity to perform a lipid profile. In her experience, physicians at public facilities routinely ordered this test, which had to be performed in the private sector: “It costs one thousand odd (~CAD 8) at Asiri.” This charge amounted to more than Premalatha’s daily household income. Chamila (1/S), primary caregiver to her mother who had a chronic kidney problem, also struggled to meet the costs of tests ordered by doctors at the Kandy Hospital’s Kidney Unit: “Sometimes everything falls on my head….last time I paid over 2000 rupees (~CAD 17) for a blood test, and then again [they] sent me for a test that cost over 1000 rupees (~CAD 8).” She was in debt, having pawned her jewellery and borrowed from friends, to finance her mother’s healthcare expenses.

Public hospital pharmacies frequently ran through their drug stocks, compelling users to visit a private pharmacy or Osu Sala – an outlet of the State Pharmaceutical Corporation that makes available medicines at lower cost to users. Indrani (3/S), a retired clerk with diabetes and heart disease, usually obtained the greater part of her monthly prescription at Osu Sala. At her most recent clinic visit, several medicines that made up her prescription were not available at the Kandy Hospital’s pharmacy: “There wasn’t even metformin (blood sugar-lowering drug)…there was only losartan (blood pressuring-lowering drug) and insulin (blood sugar-lowering drug).” The availability of insulin was, however, a relief for Indrani: “It’s a great thing that [we] get insulin, otherwise that’s also about 1700 or 1800 [~ CAD 15] no, for a bottle.” For similar reasons, Aisha (1/Ma), a homemaker whose spouse was a daily waged worker, frequently purchased her

111 diabetes medication from a private pharmacy: “No sugar medicine last month, no sugar pills this month, [I] had to pay for them.”

Some essential medical supplies needed to be purchased in the private sector as they were not available at public facilities. The nurses at the Peradeniya Hospital’s maternity unit had asked Rifana’s (2/Mu) husband to purchase a list of items before her delivery. Sabina (2/Mu), who had recently delivered at the Kandy Hospital, shared the same experience: “[They asked us to bring] syringes, a lot of syringes, and some tube, I don’t know what it was.” In some instances, such spending made a huge dent on the household economy. When Indrani (3/S) needed angioplasty and stenting,123 her cardiologist directed her to a pharmaceutical company as cardiac stents were not issued at public facilities: “When we got a date for the procedure, we informed the rep (pharmaceutical representative) and [they] told us that the equipment would be ready at the hospital on that day.” Her son purchased three stents and accompanying insertion equipment, which amounted to just under five lakhs (~ CAD 4500) from the “rep” who brought the items to the cardiac catheterization lab at the Kandy Hospital. Hasty financial arrangements were made on the day of the procedure as Indrani’s condition required unforeseen equipment: “There were two other small blocks that needed to be burst…that’s why the cost increased to four [lakhs] and 92 [thousand] (~CAD 4100).”

The President’s Fund, a humanitarian undertaking by the President’s Office, provides financial assistance to cover the costs of some medical and surgical procedures, including cardiac stenting (Government Information Centre 2009). Users like Indrani may partially reimburse their healthcare costs through the President’s Fund. However, Indrani had to take a loan and reach out to family and friends to finance the remainder: “About one third was covered by the President’s Fund, that is about 1.5 lakhs [~CAD 1250], but the rest we had to pay… we’re still paying off the loan we took from the bank.” Fathima (2/Mu), who was awaiting joint replacement at a state hospital in Colombo, faced a similar situation as prosthetic devices are not issued at public facilities. She had already applied for partial reimbursement to the President’s Fund, and her family was saving

123 Angioplasty is a procedure used to treat blocked coronary arteries. It involves inserting and inflating a small balloon in the narrowed area to widen the artery. It is often combined with stenting, which involves permanently placing a small tube (stent) to keep the narrowed area open and reduce its chance of narrowing in future (Mayo Foundation for Medical Education and Research 2017).

112 up to purchase a hip prosthesis from a private company recommended by her surgeon in Colombo.124

At times, economically disadvantaged women ‘opted’ out of the public system because the time costs associated with accessing public healthcare were too high. Indrani (3/S), who had a cardiac ailment, tested her blood sugar at a private laboratory before her monthly visit to the Cardiology Clinic because using outpatient diagnostic services at the Kandy Hospital exhausted her. “The queue to the lab extends out of the OPD [building],” she complained. Premalatha (1/S), who had high blood pressure, diabetes, and high cholesterol, attended a medical clinic at the Kandy Hospital. However, because the lines at the hospital’s pharmacy were too long, and required several hours to navigate, she purchased her prescription, which she claimed was not too expensive, at a private pharmacy.

The financial and time costs associated with public healthcare drove some economically disadvantaged users to forego treatment all together. Leela (1/S), who had a chronic kidney ailment, had decided that she would not explore the option of a transplant, against the advice of her nephrologist at the Kandy Hospital. As Chamila, her daughter and primary caregiver, explained, they just could not afford the additional expenses associated with having a transplant procedure at the Kandy Hospital:

There are a lot of tests to do outside (private)… [we] have to go to Narahenpita (in Colombo) for some tests it seems, [and] a lot of other things…travel costs, and two or three tests outside (private) for 7 or 8000 (~ CAD 55-65) each, there are things like that [we] heard.

In summary, healthcare users were compelled to consume private healthcare even when they accessed care at public facilities. The need to run back and forth between public and private healthcare facilities to address gaps in public sector services inconvenienced caregivers, who were generally women. Spending on private healthcare was especially burdensome for economically disadvantaged users, who, at times, forewent public healthcare given the associated costs. In effect, the state’s ‘withdrawal’ from healthcare expanded the private market for diagnostic services, pharmaceuticals, and other medical products, while deepening inequities in access. Next, I draw attention to a second form of ‘mixing’ public and private healthcare, the

124 In January 2017, the Ministry of Health began to supply cardiac stents in the public sector (The Island 2017). However, prosthetic devices are still not issued free-of-charge at public facilities.

113 practice of consulting public sector physicians in the private sector, an arrangement used mostly by middle/upper-middle class users to upgrade the service they received at public facilities.

8.2 Channelling privilege

The public and private sectors in Sri Lanka are linked by a shared physician workforce who engage in dual practice (Dayaratne 2013). By consulting a public sector physician in the private sector, it is possible to circumvent delays and lengthy wait-lists in the congested public system. Although a few women spoke of accessing public sector services through their respective private family doctors, this section focuses on the experiences of women who used private specialist consultations or ‘channeling’ to enter tertiary care centres in Kandy. While the reasons for doing so varied by social location, middle/upper-middle class women exploited this pathway, often expecting benefits and privileges in the resource-constrained public system.

Unlike working class women, who used private healthcare out of necessity, middle/upper-middle class users combined public and private healthcare services on their own volition. As Ruwani (4/S), a retired bank clerk, explained, it really depended on each situation: “For some things we go private… if we think the [public] hospital is better, we go there.” These women generally used the private sector for outpatient care, but deferred to the public system for inpatient services. In doing so, they channeled a specialist who could connect them to the required service in the public system, unencumbered by the challenges economically disadvantaged users confronted as they accessed public healthcare.

Channeling not only allowed for one-to-one consultations in the private sector, but also guaranteed continuity of care in the public system. As Ruwani (4/S) pointed out: “The advantage of private [channelling] is that [we] can target [a specialist’s] attention to [our] problem directly. We can talk to [the specialist] and get information and then go [through the specialist] to the public service.” This is what she did when her husband experienced hearing loss. Rather than trying to obtain a referral to an ENT surgeon through the outpatient department (OPD) – the “normal procedure” that would invariably involve delays at the OPD and/or ENT clinic – they booked an appointment with an ENT surgeon at the (private) Amaya Medical Centre. If the problem required further investigation or treatment, they would enter the public system through the ENT surgeon. While Ruwani did not always channel to enter a public facility, she felt it was the best option available to them in this instance because getting a “number” at the Kandy Hospital’s congested ENT clinic was difficult.

114

Channelling expedited access to resource-intensive procedures in the public system. Anoja (4/S), a retired clerk whose husband had recently passed away, felt that consulting a specialist at the (private) Amaya Medical Centre had speeded up care for her husband. Having immediately recognized the significance of her husband’s memory loss, the specialist quickly admitted her husband to Peradeniya Hospital for an urgent CT scan. According to Anoja, accessing this service via the OPD would have invariably delayed the diagnosis of his brain tumour. Channelling also allowed for bypassing lengthy wait-lists for procedures at public facilities. By channelling her cardiologist at the Kandy Channelling Centre, Mala (5/S) circumvented the queue for echocardiography at the Kandy Hospital’s Cardiology Unit where her cardiologist also worked. She only needed to bring the referral letter provided by her cardiologist to the Kandy Hospital’s Cardiology Unit, and the nursing staff would send her directly to the cardiologist, who would attend to Mala’s needs first.

Gaining admission to a public inpatient unit via channelling came with numerous benefits. As Muditha (4/S), whose husband had recently used this pathway to enter the Peradeniya Hospital, explained: “When [the nurses] realised we were the doctor’s [private] patient, we got a bed and special attention.” This was critical in an overcrowded public hospital setting where beds were hard to come by. Moreover, according to Muditha, her husband received prompt attention as a result of the private ‘connection:’ “The [nurses] quickly gave the medicines because [we] had the letter with instructions [from the specialist].” She sensed that their ‘connection’ also influenced the way hospital staff interacted with them: “[I] felt they were more compassionate, [they] didn’t yell at us… the doctor also recognized [us]…and gave us special attention.”

A channelled consultation was not always necessary for middle/upper-middle class users to enjoy such privileges in the public system. Striking in their stories were the references to various social ‘connections’ they had with physicians. Sakuntala (5/T), a relative of a professor at the University of Peradeniya, was attended to immediately at the Ophthalmology Unit because the specialist, whom she also consulted in the private sector, was a past student of a professor relative: “I don’t even sit and wait [at the clinic]… There’s like 100s [of people there]. [He] says come at 9 o’clock, I go at 9 o’clock … [The nurse] will put me in the doctor’s room and check the eye pressure and I’m out in 10 minutes flat.” While Sakuntala attended the public clinic because the Amaya Medical Centre did not have the equipment to monitor glaucoma, through the same ‘connection,’ her husband, a retired professor, entered the Ophthalmology Unit for cataract surgery where, according to Sakuntala, he received “VIP” treatment: “There is a VIP room or something, it’s like a large hotel room, large, bigger than our bedrooms… with attached bathroom, A/C (air

115 conditioning)…nobody can pay and get it, the doctor has the freedom to put you in that room.” In this way, Sakuntala’s husband enjoyed amenities at the Kandy Hospital that were evidently on par with those on offer in the private sector, but were not available to less privileged others.

Women representing the upper-middle class group often called up a physician directly or used a social “connection” to obtain prompt medical attention. Sharmini (5/mixed), a university teacher, consulted a paediatrician “known” to her sister whenever her son had a health problem: “If you go …through someone you know [to a public hospital]… you can definitely get better service.” All she had to do was contact her sister – a medical doctor – who would arrange an appointment with her “friend,” the paediatrician. For Sakuntala (5/T), such a connection was essential to receive a decent standard of care in the public system: “You have to know somebody if you want to get something done in the [public] hospital … if you have a family member or a close friend, you’re there.” Given the numerous connections she had with doctors, Sakuntala felt she could not speak for the “average” user who did not have such “connections.”

Less commonly, economically disadvantaged women shared experiences of accessing similar privileges via ‘connections.’ Indrani (3/S) had benefitted from such an arrangement given her long- standing friendship with a professor at the University of Peradeniya, who happened to be a neighbour from Kirideniya. When Indrani shared the news of a recent episode of chest pain with her friend, the Professor immediately called up her cardiologist “friend,” and arranged for Indrani to undergo urgent angiography, bypassing the six-month wait-list at the Kandy Hospital:

I went to the [Kandy] Channelling Centre the next day and saw [the cardiologist]. [She] gave me a letter, and told me that she would take me soon for the angiogram. “I want to do this for you quickly because “madam” (the professor) asked me,” she said. So I got the chance to do [the angiogram] quickly.

Economically disadvantaged women who accessed the public system without private or social ‘connections’ felt the consequences of this ‘double dipping’ between public and private by privileged others. Damayanthi (1/S), previously a janitor, had been delayed many times at public clinics as she had no such ‘connections’: “Even if we go very early in the morning [to the clinic], the patients who channel get preference. These patients arrive late and get what they need and leave quickly.” Nafeesa (1/Mu), a homemaker whose spouse was a daily waged worker, was held up several hours as she waited for an ultrasound scan at the Kandy Hospital’s Radiology Unit: “There were [private] patients with letters from the [specialist]. Those [scans] were done quickly and [they] were sent off.” In other words, channelling allowed privileged women to bypass wait-

116 lists at public facilities, while working class women and marginal others without such ‘connections’ stood around for hours waiting their turn.

Thus, middle to upper-middle class women channelled public sector specialists in the private sector, an arrangement that allowed them to access numerous privileges in the public system. This particular hybrid arrangement paved the way for socially connected ‘private patients’ to directly access services in the specialist’s public sector unit, while economically disadvantaged users were pushed to the margins. Mixing public and private healthcare presented a commodified form of social capital for privileged women who remained largely unaffected by the resource- constraints in the public system. However, this class-based interpretation of channelling was disrupted by Malay and Muslim women who used the same pathway to manage their marginality within the public system.

8.3 Managing marginality

Channelling was not always a practice associated with privilege. While allowing socially connected middle/upper-middle class users to better their lot in the public system, channelling was also used by socially and economically marginalized women to ensure they received a minimum standard of care. Thus, some disadvantaged users saw channelling to be an essential component of public healthcare, making this social stratum vulnerable to exploitation by physicians and the vagaries of the healthcare market.

Ironically, two of the most socially (and economically) disadvantaged participants from Neluwa/Jagoda (1) reported having recently used this pathway to overcome barriers to accessing public healthcare. Shereen (1/Ma) and Siththy (1/Ma), both Malay women with physical disabilities, had paid for private specialist consultations because they could not enter the public system through the outpatient department (OPD). When Shereen (1/Ma) initially began to trip and fall – early signs of a debilitating neuromuscular condition – she visited the Kandy Hospital’s OPD, but was repeatedly turned away by doctors who did not recognize her problem: “[I] knew my problem was getting worse because I was tripping up and falling more and more.” The doctors at the OPD disregarded her complaints, and turned her away with vitamins. In the absence of alternatives, Shereen channelled a neurologist who worked at the Kandy Hospital. During the private consultation, Shereen explained that she simply could not afford treatment in the private sector. The specialist readily provided her a referral letter that enabled her to bypass earlier

117 barriers at the OPD to enter the specialist’s public sector unit where she was subsequently investigated.

With similar intentions, Siththy (1/Ma), a homemaker, brought her daughter to the (private) Amaya Medical Centre for a channelled consultation. Her daughter’s psychiatric problem had not been recognized or treated at the Peradeniya Hospital’s Medical Unit where she had been admitted days earlier. Paying no heed to Siththy’s view of her daughter’s condition, the doctors at the Medical Unit had investigated her daughter for a suspected kidney problem, and subsequently discharged her with a clean bill of health. As a last resort, Siththy channelled a specialist at ‘Amaya.’ However, unlike Shereen who knew she needed to see a neurologist, Siththy did not know how to select among the various specialists practicing at Amaya. On the recommendation of the “girl” at the reception, she saw a specialist who, once again, did not recognize her daughter’s psychiatric condition. Instead, he ordered a battery of tests and asked them to return for a second consultation. Since Siththy could afford neither the tests nor another consultation, she did not bring her daughter for a second visit. In this instance, Siththy had pocketed out the channelling fee, which she could barely afford, to no avail.

Neither Shereen nor Siththy attributed their OPD experiences to their respective ethnicities or disabilities. A combination of factors, including poverty, ethnicity, disability, and physician incompetence, may have influenced their experiences. These women used channelling under desperate circumstances, accepting the associated ‘risk’ as there was no guarantee that this pathway would offer a solution to their health concerns. Although Shereen successfully entered the public system through the specialist she channelled, Siththy was less fortunate. In the end, Siththy contacted a doctor at whose private clinic her daughter had dispensed medicines in the distant past, and asked him to help her daughter access the services she needed. Not surprisingly, most women from Neluwa/Jagoda (1) did not speak of having such ‘connections.’

In contrast to the working class women from Neluwa/Jagoda (1) among whom only two shared channelling experiences, Muslim women from Bogoda (2) routinely channelled specialists. Sabina made sure to see a specialist before entering a public hospital for inpatient care: “Everyone says [you] have to channel before [going to the hospital], if you go without [channelling] you have to wait and wait.” Recalling an experience she had at the maternity unit of the Kandy Hospital when admitted for her second delivery, Sabina emphasized that such delays could critically influence health outcomes. When Sabina informed a midwife at the maternity unit that her baby was inactive, the midwife dismissed her concerns: “She told me, “Once every hour if the baby moves

118 is enough, you wait without complaining and mind your business.”” The next morning, the specialist stopped by at Sabina’s bedside as he recognized her from his private practice. He patted her arm to reassure her and discovered she was running a temperature: “The doctor yelled at the nurses to bring a thermometer – to see my temperature was over 100!” Sabina had an emergency caesarean section, and her baby remained in the “baby room” for 11 days. Had she not channelled the specialist, she was convinced, the day would have ended in disaster: “Only because the doctor came and saw me, God was there, I thought about what a narrow escape both my baby and I had [that day].”

According to the Bogoda women, Muslim users, particularly those who did not speak Sinhala, regularly channelled specialists before entering the public system to avoid unpleasant encounters with healthcare providers who could not communicate in Tamil. As Rifana (2/Mu) explained: “People from Akurana and places like that125 go first to Suwasevana [private hospital]…they will even go in to debt and go private.” According to Feroza (2/Mu), a private sector ‘connection’ ensured respectful and timely treatment at the specialist’s public sector unit: “When [you] channel, the doctor will give a date … then [you] can go to the government (public) hospital because [the doctor] will attend.” Sabina (2/Mu) agreed that a private ‘connection’ was essential: “Yes, that is the most important point.” After all, she had experienced this first hand during her deliveries at both Peradeniya and Kandy Hospital: “If we have the doctor’s seal from the private sector, [they say,] “Ah this is that doctor’s patient, send her quickly [to the labour room].”

Yet, the poorly regulated environment in which channeling centres operated meant that users were vulnerable to exploitation. Dilrukshi’s (3/S) niece, in the early stages of pregnancy, decided to see her public sector specialist at the (private) Amaya Medical Centre to ensure she received a decent standard of care at the Peradeniya Hospital. Unlike middle/upper-class users, who established relationships with their specialists to their advantage, Dilrukshi’s niece’s encounters with her obstetrician were disappointing. At the most recent visit to ‘Amaya,’ her niece, accompanied by Dilrukshi, paid the channeling fee, and waited several hours, first for the specialist to arrive, and then for their number to be called. When they finally met the specialist, their consultation was very brief much to Dilrukshi’s disappointment:

We didn’t get a chance to speak with [the doctor]… [At her last visit he] asked [my niece] to bring a blood report (blood test)…something was low in her blood, something or the other. So I asked

125 Muslim dominant towns within the Kandy District where Tamil-speaking Muslims may not speak Sinhala.

119

[the doctor] about this. “No, that’s no problem, it will be okay if you take this medicine,” that’s all [the doctor] said. And we came out [of the room]… not even five minutes… as soon as we went in, we came out. I told my niece, “It would have been much better if we went to a government (public) hospital.”

In short, socially disadvantaged women used hybrid arrangements to manage their marginality within the public system. Coincidentally or otherwise, most women who reported doing so belonged to ethnic minorities. Although these women believed a private consultation conferred some protection to them within the public system, channelling involved some ‘risk’ and came with no guarantees. Moreover, these women essentially paid an additional consultation fee to access basic facilities in the public system, available free-of-charge to (privileged) others. At the same time, with such a broad-based clientele, channel practice shifted a substantial share of outpatient consultations to the private sector, creating avenues for private accumulation.

8.4 Economizing on healthcare

The most widely cited reason for mixing public and private services was to reduce the costs of healthcare. Irrespective of social background, women weighed the ‘costs’ and ‘benefits’ of public and private healthcare, and combined these services in ways that allowed them to address their health concerns in affordable ways. Apart from working class women who rarely ventured to the private sector, others spoke of starting out in the private sector, and ending up in the more economical public system as and when more advanced treatment options were required. As illustrated in the examples below, the threshold at which users switched to the public system differed by social class and ethnicity. As one might expect, the threshold for transfer to public healthcare was higher for wealthier residents and lower-middle class Muslim women, who preferred the private sector for outpatient care. The latter’s pattern of private healthcare use was similar to that of middle class residents, although their social conditions were starkly different.

Badra (3/S), a busy NGO worker, consulted a private family doctor for her diabetes and refilled her prescription at a private pharmacy. However, she could not afford to perform the battery of tests her family doctor had ordered, in the private sector. Instead, Badra contacted a relative who worked at the Kandy Hospital who arranged for her to have the tests done free-of-charge at the hospital’s laboratory: “She gave me a date and got everything done in one day… But I couldn’t

120 do the cholesterol test. I still haven’t got it done.”126 By drawing on her social networks, Badra accessed a quicker service in the ‘free’ public sector instead of following the regular pathway, which would have invariably involved lengthy queues and extended wait-times.

Amali (3/S), who ran a snack bar and was always busy, usually consulted her private family doctor who practiced close by, as she could not stand around at the outpatient department of Peradeniya Hospital. When anything beyond a consultation and a few basic medicines were required, however, her family doctor – who was also a medical officer127 at the Peradeniya Hospital – arranged for Amali to access these services at the Hospital: “[He] says, “[You] don’t need to go private, come to the hospital, I’m there, I will arrange this [for you].” Thus, she rarely paid for anything beyond a consultation with her family doctor. Using this pathway, Amali received expedited services, while reducing out-of-pocket expenses.

Indrani (3/S) could not afford to purchase her monthly prescription in the private sector, and attended the Cardiology Clinic at the Kandy Hospital every month to refill her prescription. However, she also visited her public sector cardiologist every three to four months in the private sector because she could not receive his undivided attention at the busy public clinic. When she brought her private medical records to the Kandy Hospital’s Cardiology Unit, a junior doctor would order a refill based on the specialist’s instructions documented in her private medical records: “[They] don’t change Sir’s (the specialist’s) prescription so I get the medicines from the [public] hospital.” Even so, Indrani purchased a sizeable portion of her monthly prescription from Osu Sala because the Kandy Hospital’s pharmacy frequently ran out of medicines. By mixing public and private, Indrani saved on her prescription and improved the quality of her care.

Kirideniya (4, 5) residents used the private sector extensively for outpatient care, but the majority in this group transferred into the public system for inpatient care owing to both financial and quality concerns. When Anoja (4/S), a retired clerk, felt her husband’s insulin was a burden on the home economy, she asked her husband’s private sector specialist to refer them to his public clinic at the Peradeniya Hospital: “I told the doctor [the cost was too much for us].” Anoja’s husband thus consulted his specialist in the private sector, while obtaining at least a part of his prescription free- of-charge from the specialist’s medical clinic at Peradeniya Hospital.

126 As pointed out by Premalatha (1/S), lipid profile was not performed at public facilities in Kandy. 127 A non-specialist physician in the public sector.

121

Economic concerns were more pressing for Kirideniya residents (4, 5) when they required inpatient care. As Muditha (4/S) remarked: “If [we] need to be admitted for a long time, we can’t stay at a private hospital no… there will be big charges.” When Muditha’s husband first developed symptoms of cirrhosis, they went directly to the outpatient department at the Peradeniya Hospital on the advice of their private family doctor. They then channelled the specialist who had attended to Muditha’s husband at the Peradeniya Hospital instead of attending the specialist’s public sector clinic. When her husband needed albumin transfusion, however, they re-entered the public system through their specialist: “The doctor told us that if [we] go private, [one albumin bottle] would cost 3500 (~CAD 30). We thought this would be a waste.” During her husband’s hospital stay, Muditha had to purchase one of his medicines and some glucose strips from a private pharmacy as they were not available at the Peradeniya Hospital: “If [we] went private, it would have been wasteful spending no.... they even charge for cotton wool at Suwasevana (private hospital).”

For similar reasons, Nalini (5/S), retired from the hospitality industry, had her womb removed at the Peradeniya Hospital. She turned to the public system not because she could not afford private services but because she “did not want to spend money.” Nalini requested the gynaecologist she consulted at the (private) Suwasevana Hospital to transfer her to his public sector unit for surgery: “Of course I was little stingy…when doctor said 45,000 (~CAD 375) [for the operation] I thought, oh god, I can buy two saris.”128 With a letter from the gynaecologist, she went to his clinic at the Peradeniya Hospital where she bypassed the wait list and received priority services.

Women from Bogoda (2) avoided the public system as far as possible, but did transfer to public hospitals when they anticipated inpatient care. Sabina (2/Mu) channelled three different obstetricians during her three pregnancies, but delivered at Peradeniya or Kandy Hospital. Feroza (2/Mu) accessed private outpatient care for her mother who had chronic constipation, but admitted her to the Peradeniya Hospital when her condition worsened and she required urgent inpatient treatment. While Fathima (2/Mu) consistently channelled private sector specialists for the follow up of her arthritis, she turned to the public system for knee replacement as she could not afford the steep charges associated with the procedure in the private sector.

128 Saris worn on special occasions may easily cost over CAD 200 each.

122

In sum, women switched from private to public sector at varying thresholds to economize on healthcare. Kirideniya (4, 5) and Bogoda (2) residents opted out of the private sector at a higher threshold, generally for inpatient care. Meanwhile, women from Dehioya (3) did so at a lower level, preferring the public sector for longer-term outpatient care. It would seem that lower-middle class Muslim users spend more on healthcare than do their Sinhala counterparts. While reducing household spending on healthcare, mixing private with public frequently allowed users to access speedier public services. Notably, these arrangements allowed the state to push a substantial proportion of consultations to the private sector, while providing a high level of universal ‘financial risk protection’ for inpatient care.

8.5 Discussion

The findings of this chapter complicate the picture of access to public and private healthcare services presented in the previous chapter. Rather than choosing between public and private healthcare, it would seem that most Udawatta residents used a mix of public and private healthcare services or, what I call, hybrid arrangements. However, the forms of hybrid arrangements employed by women differed by social location.

Economically disadvantaged women had no ‘choice’ but to access private care to address deficits in the public system, while others opted for private care to varying degrees for different reasons. Middle/upper-middle class women, lower-middle class women, among the latter group especially Muslim women, and a few representing the working class group, accessed the public system via channel centres to upgrade public service. While middle/upper-middle class women used this pathway to gain priority in the struggle for scarce public resources, Malay and Muslim women, and potentially others on the margins, established private ‘connections’ to access a standard of public care. Irrespective of social location, women mixed public and private healthcare services to alleviate the costs of healthcare.

Hybrid arrangements widened inequities in access at multiple levels. First, out-of-pocket payments associated with public healthcare disproportionately burdened economically disadvantaged women. Second, middle to upper-middle class women remained largely unaffected by the state’s failure to invest in public facilities as channelling gave them priority access to scarce public resources. Although channelling provided some leeway to socially and economically disadvantaged ethnic minorities (and others) to upgrade their marginal status within

123 the public system, they essentially paid for a private specialist consultation to access public services, which were accessible free-of-charge to others.

These hybrid arrangements reflect the class warfare that takes place at the ‘micro level’ as neoliberalism erodes welfare. Although the public system provides a semblance of universal ‘financial risk protection’ by acting as a back-up for private healthcare users, it functions under tremendous resource constraints where the healthcare needs of socially and economically disadvantaged users are rendered secondary to those of privileged users. Exploring interlocking dimensions of inequity under privatization may help us better understand the latter’s implications for users on the margins, a concern that has received little attention in the literature on healthcare access in LMICs.

The gendered character of healthcare privatization manifests in the various hybrid arrangements described in this study. They suggest that women possess critical knowledge on healthcare, and that they make crucial decisions on which services to access, when, and to what purpose. They spend inordinate amounts of time securing the best options available within existing constraints. However, the ‘benefits’ secured through their reproductive labour look very different by social location. Socially and economically disadvantaged women invest more time and work to obtain a bare minimum, while wealthier women with social connections navigate public and private healthcare systems to their advantage, accruing more benefits and privileges.

Lastly, hybrid arrangements have created niches for private accumulation, a matter I return to in the next chapter. Briefly, the state’s receding role in healthcare provision, and resulting deficits in public healthcare services, have expanded markets for diagnostics, pharmaceuticals, and medical supplies, extending their reach to even the most economically disadvantaged sections of society. The majority of users, excepting the desperately poor, have opted out of the public system for at least some aspects of outpatient care, shifting a considerable share of outpatient services to the private sector. On the other hand, the state’s authorisation of dual practice has created avenues for accumulation for individual providers and the private healthcare industry. In the next and final chapter, I bring together and discuss my findings in light of my theoretical framework, and reflect on their policy implications.

124

9. Discussion and conclusion

Chapter overview: In this chapter, I present my findings and discuss them in light of my theoretical framework and the literature on healthcare privatization. I first discuss the key contributions made by this thesis before moving on to their policy implications. I then outline the limitations of this research and propose some ways they may be addressed. I conclude with some final thoughts on the future of health sector development in Sri Lanka and LMICs.

9.1 Key contributions to the literature

As we saw in Chapter 4, much of the empirical research on healthcare access uses large secondary data sets or primary surveys to estimate household spending on healthcare, quantify utilization of health services, assess quality of care, and measure equity of access. While this work contributes valuably to the knowledge base on healthcare access in the context of privatization, it makes invisible the experiences of users and fails to engage privatization as a driver of inequity. Although healthcare privatization and its implications have been critically analysed in the political economy literature (Armada et al. 2001; Birn et al. 2017; Castro and Singer 2004; Navarro 2007; Qadeer and Baru 2016; Sengupta 2015; Waitzkin 2011), these critiques do not shed light on how privatization is experienced ‘on the ground’ in LMICs.

A small body of critical qualitative research addresses these gaps by exploring user experiences under privatization in low- and middle-income settings (Alvarez et al. 2011; Gao et al. 2010; Kaufman and Jing 2002; Kierans et al. 2013; Paphassarang et al. 2002; Perera et al. 2007; Varman and Vikas 2007). While these studies primarily focus on disparities in access based on income and, less often, the urban-rural divide, they miss the gendered character of healthcare privatization and its specific implications for women on the margins. Moreover, the strategies of survival employed by users as they struggle to access care in the context of privatization remain unaddressed in the literature.

Building on extant qualitative work, the present study used a Third World Marxist feminist methodology to link women’s experiences of using healthcare in Kandy with the macrostructures and processes shaping healthcare in Sri Lanka and other LMICs. This work makes three key contributions to the literature on healthcare privatization. First, it draws attention to the limited ‘choices’ that open up under privatization by examining the healthcare practices and experiences of various social groups in Kandy. Second, it interrogates the concept of a mixed health system

125 by shedding light on the strategies of survival that have emerged within Kandy’s ‘mixed health system.’ Third, it highlights the ways in which ethnicity and other dimensions of difference intersect with social class to complicate the dominant class narrative of healthcare access under privatization in LMICs.

9.1.1 ‘Choice’ and its limits under privatization

As we saw in Chapter 7, the public system presented the only feasible ‘choice’ for economically disadvantaged users in Kandy. Even the nominal ‘choices’ that opened up in the presence of private healthcare, such as the ability to choose a healthcare provider/facility or obtain diagnostic tests or medicines ‘over-the-counter,’ were not available to these women who simply could not afford private healthcare. Such class-based inequities in access to public and private healthcare services are widely reported from LMICs (Campbell et al. 2016; Saksena et al. 2012). In Sri Lanka, quantitative assessments of household income and expenditures confirm that economically disadvantaged users rely largely on the public system, while wealthier users consume more private healthcare (Mackintosh et al. 2016; Pallegedara and Grimm 2017).

Yet, most women in this study, including those representing the working class group, were compelled to visit the private sector as the public healthcare options available to them were not comprehensive. Widespread use of private healthcare by low-income households is reported from other LMICs (Barber 2006; Ha, Berman and Larsen 2002; Makinen et al. 2000; Rashid et al. 2011; Saksena et al. 2012; Xu et al. 2006; Yoong et al. 2010). My findings are also consistent with those of Russell and Gilson (2006) who report that over 40 per cent of residents representing the poorest quartile in two low-income settlements in Colombo accessed private outpatient care for ‘acute’ problems to avoid the ‘inconveniences’ associated with public care. Similarly, Perera and colleagues (2007) find that users with diabetes from rural Sri Lanka were compelled to seek private care owing to deficits in lower-tier public hospitals, including the incapacity to provide basic follow-up services and treatment for diabetes.

Arguably, the private health sector did expand healthcare ‘choices’ for middle to upper-middle class women in the present study. As Govindaraj and colleagues (2014:43) suggest in their review of Sri Lanka’s private health sector, these women did enjoy the benefits of “quicker,” “cleaner,” and “more flexible” outpatient care. Indeed, according to a between-sector comparison of outpatient care undertaken by Rannan-Eliya and colleagues (2015a), the private sector performs better in all non-clinical aspects of quality of care. Such positive attributes of private healthcare

126 are noted by others in LMICs, and are a key reason for users to access private facilities (Huff- Rousselle and Pickering 2001; Lindelow and Serneels 2006; McLane et al. 2015; Ngo and Hill 2011). However, the findings of the present study suggest that these private ‘choices’ served as an alternative to the public system.

Rather than an expression of ‘choice,’ women in Kandy tried to circumvent the challenges they faced accessing care in the public system by using private healthcare. They had to do so as a result of neoliberal cutbacks implemented by the state in response to the pressures of economic liberalization. Participants depicted a public system crippled by resource constraints, with specific reference to: a field midwifery service lacking human resources; a walk-in service at the public Dispensary that was frequently unavailable; crowded outpatient departments involving lengthy wait times; congested public clinics where consulting a specialist was near impossible; and public facilities lacking basic infrastructure and services. It is in this context that the ‘demand’ for alternative (private) healthcare ‘choices’ had arisen.

The private ‘choices’ that were available to women in Kandy were, however, limited by their costs and their technical quality. Indeed, even middle to upper-middle class women in the study struggled with out-of-pocket payments and expressed concerns about the private sector’s capacity to deal with an emergency. As a result, they often opted into the public system for resource-intensive care. These findings contradict Rannan-Eliya and Sikurajapathy’s (2009) claim that the public system in Sri Lanka is ‘progressive’ despite a greater share of healthcare financing coming from indirect taxation. Although the poor contribute more relative to income toward financing public healthcare, they argue, the wealthy opt out of public healthcare, enabling poorer users to avail themselves of public services. This is clearly not the case in Kandy where wealthier middle to upper-middle class users not only accessed resource-intensive services in the public sector but also received priority access to them.

How do we understand the private healthcare ‘choices’ that enabled women on the margins to avoid neglect, abuse or stigmatization in the public system? Or the disability friendly services which were evidently available at private healthcare facilities? Lower-middle class Muslim women spoke vocally about their discrimination in the hands public healthcare workers in Kandy, expressing preference for private (outpatient) care. The private sector offered at least some disability accessible services, although economically disadvantaged users were largely excluded from these ‘benefits.’ The marginalization experienced by these social groups in the public system had material consequences as many paid to access the private sector where they felt more

127 welcome. While these findings highlight the state’s failure to provide for citizens on the margins, they suggest that such deficits in public sector services created niches for accumulation as private healthcare facilities adapted to “optimise its returns” (Loeppky 2010:59).

Valourizing the ideals of individual autonomy and freedom, neoliberal discourses construct privatization as a cure for the multi-faceted ailments afflicting public healthcare systems (Brown 2006; Ferguson 2010). Within privatized systems, citizens are but consumers who make rational choices on when and what to purchase from a range of competing suppliers in the market (McGregor 2001). The International Finance Corporation (2017), the private sector investment arm of the World Bank, asserts that expanding ‘consumer choice’ improves quality of care under market competition. It is this same understanding of ‘choice’ that Govindaraj and colleagues (2014:43) evoke in their World Bank-commissioned review of Sri Lanka’s private health sector where they suggest that healthcare consumers exercise their “purchasing power” to enjoy the ‘benefits’ of private healthcare. However, such analyses overlook the fate of those without ‘purchasing power’ as healthcare undergoes privatization, and fail to locate ailing public systems in their broader ideological context.

Then, the healthcare ‘choices’ that open up in the presence of private healthcare are always constrained by social relations. For different reasons, the private sector did not present a convincing ‘choice’ for women across social classes, although it remained the preferred ‘choice’ for users who felt marginalized by the public system. In promoting healthcare privatization as a viable policy direction for LMICs, the World Bank, its intermediaries, and national governments overlook the economic exclusions associated with private healthcare. They also fail to acknowledge the quality concerns that arise as healthcare is opened to the market. I now turn to the strategies women used to access healthcare under privatization to interrogate the concept of a mixed health system.

9.1.2 Strategies of survival and accumulation within mixed health systems

As primary caregivers in their families and communities, women shoulder the burden of healthcare privatization (Armstrong et al. 2002; Ewig 2008; Gideon 2008; Gideon 2012). A largely neglected dimension in the literature on healthcare access in Sri Lanka, only Perera and colleagues (2007) highlight the impact of privatization for women as caregivers in their families and communities. The present study brings attention to the invisible and unacknowledged work women undertake as the state recedes from healthcare provision.

128

As we saw, women of Udawatta devoted much time and effort to learning about, exploring, and navigating public and private healthcare systems. They devised numerous strategies to access healthcare, straddling public and private health systems, to attend to their varied health concerns. Notably, these women, the majority married, rarely spoke of their spouses or significant others as contributors to caregiving activities in their homes. While Gideon (2008) highlights the invisible work that is shifted to the home, and, by default, to women, in the context of healthcare privatization, the differing degrees to which women representing various social groups are burdened by privatization has received less attention in the critical feminist literature.

With their basis in hybrid arrangements, women’s survival strategies in Kandy differed markedly by social location. Economically disadvantaged women borrowed and scraped together funds to purchase deficit public services from private facilities, while others opted out of the public system to reduce the time costs associated with accessing healthcare. Middle to upper-middle class users exploited dual practice to secure privileges in the public system, and marginal others did the same to obtain a very basic standard of care. Striking in their practices were the varying degrees to which different groups of women had to labour to obtain healthcare, and the widely divergent outcomes of engaging in this form of reproductive labour. The presence of private healthcare reinforced social divisions as women struggled, by whatever means, to access healthcare for themselves and their families. The study findings affirm McGregor’s (2001:86) assertion that privatization individualizes the responsibility of healthcare:

[M]arket choice leads to fewer people seeing themselves as citizens who have a right to health care paid for from tax dollars. Then, the survival of the fittest principle sets in and people no longer feel it their responsibility to lobby for health care for everyone.

The strategies of survival employed by healthcare users in a particular setting are contingent on historical and social relations. In India, for example, where the private health sector predominates, the rising costs of private healthcare force economically disadvantaged users to access the ‘informal’ private sector with “disastrous consequences” (Varman and Vikas 2007:168). The tendency for users to shift from ‘formal’ to ‘informal’ providers as healthcare is opened to the market is also reported from Colombia where economically disadvantaged users self-medicated or delayed care owing to the imposition of new financial barriers following neoliberal health reforms (Alvarez et al. 2011). Meanwhile, in Sri Lanka where ‘free health’ has endured, users have devised ways to overcome resource constraints in the public system by combining public and private healthcare services.

129

These hybrid arrangements have created avenues for accumulation in the context of the state’s ‘withdrawal’ from healthcare. They channel meagre household resources to individual healthcare providers and the healthcare industry, dispossessing working classes (and marginal others) of their entitlements to healthcare (Harvey 2005). Indeed, crippling resource constraints in the public system have expanded markets for the private healthcare industry, reflected in over half of outpatient visits and purchases of medicines taking place in the private sector in Sri Lanka (Amarasinghe et al. 2015). Such shortages together with extant links between public and private sectors encourage users to visit channel centres before they enter public facilities to ensure they receive a decent standard of care. Having the option of transferring private patients to the public system means that specialists can maintain a broad-based clientele and substantially increase their personal incomes with the promise of admission to the more economical and “safer” public system.

Hybrid arrangements intensify public subsidization of the private health sector with implications for equity of access. First, dual practice allows private healthcare facilities to function without employing a permanent salaried workforce. Not only state sector specialists and medical officers, but also various categories of ancillary health professionals, work on a fee-per-service basis in the private sector, while drawing salaries from the Ministry of Health. However, the ‘benefits’ of this subsidy are enjoyed only by users who can afford to pay for private healthcare. Second, the public system provides in-patient care for private healthcare users who cannot afford resource- intensive care in the private sector. That is, the public system functions as a backup, covering (unaffordable) resource-intensive inpatient care for private healthcare users, reflected in over 90 per cent of inpatient admissions occurring in the public system (Amarasinghe et al. 2015). Thus, the state’s ‘open-door’ policy enables private healthcare facilities to operate with minimal investment in resource-intensive care, and, instead, focus on delivering outpatient services, more profitable given the limited ‘purchasing power’ in Kandy.

These informal subsidies add to the state’s direct subsidization of private healthcare. Through direct subsidization, the state increases the profit margins of private healthcare companies, while incurring substantial financial losses. In their study of fiscal incentives provided by the state to the private health sector, Rannan-Eliya and Kalyanaratne (2005) suggest that tax exemptions on imports, corporate income tax reductions, subsidised rates on state lands, and so forth, that have been granted to large-scale commercial hospitals, are not offset by anticipated savings when users transfer to the private sector. As we saw in Kandy, commercial hospitals remain virtually inaccessible to the vast majority of the public who subsidizes them.

130

The strategies of survival women used interrogate contemporary support for mixed health systems where “out-of-pocket payments and market provision of services predominate as a means of financing and providing services in an environment where publicly-financed government health delivery coexists with privately-financed market delivery “ (Nishtar 2010:74). The policy initiatives that advocate for the strengthening of mixed health systems in LMICs fail to acknowledge the varied sanctioned and unsanctioned accumulation strategies that have emerged with the expansion of private healthcare in LMICs. In fact, it is this very model of a mixed health system that is supported under the ‘Universal Health Coverage’ framework (Sengupta 2015).

In sum, the gendered survival strategies that have emerged in Kandy as the state recedes from healthcare provision dovetail with a range of accumulation schemes that channel wealth from healthcare users to the medical establishment and the healthcare industry. These informal arrangements demand that we scrutinize more closely the functioning of mixed health systems, particularly the deeply interconnected character of public and private health sectors in LMICs. In the next section, I discuss how ethnicity and other dimensions of difference may complicate the dominant class narrative of healthcare privatization in LMICs.

9.1.3 Ethnicity and interlocking dimensions of access

Inequalities in access to healthcare have been examined chiefly in relation to income, often used as a proxy for social class. The ways in which other dimensions of difference intersect with social class have received scarce attention in the literature on healthcare access in LMICs. A few quantitative studies examine how race or ethnicity interact with poverty to shape healthcare access in resource-poor settings. Barber and colleagues (2007) used a composite index that integrated ‘household socioeconomic characteristics,’ including household wealth, education, access to piped water, and numerous other variables, including indigenous status, to assess quality of prenatal care in rural Mexico. They found that indigenous women receive a higher standard of care in the public sector than they do at private facilities. Using ethnographic methods, Mumtaz and colleagues (2014) explored societal determinants shaping access to maternal care in Pakistan to find that economically disadvantaged oppressed caste women prefer to access ‘private’ services from ‘traditional birth attendants’ owing to their marginalization within the public system. However, these studies did not seek to understand the potential role of privatization in the healthcare experiences of marginalized social groups. As suggested earlier, such experiences, whether in Mexico, Pakistan, or Sri Lanka, may be better understood if contextualized in the relative strengths and weaknesses of public and private healthcare.

131

Despite a troubled history of ethnic conflict, very little is known about the ways in which ethnicity shapes healthcare access in Sri Lanka. Existing empirical work on access focuses on class-based disparities (Perera et al. 2007; Russell and Gilson 2006; Pallegedara and Grimm 2017) and the urban-rural divide (Perera et al. 2007), conspicuously overlooking ethnicity as a determinant of access. I found only two studies that considered ethnic difference in the context of accessing healthcare. Even then, ethnicity was one of many variables rather than their focus. Rannan-Eliya and colleagues (2015a) reported no evidence of ethnic difference in their between-sector comparison of quality of outpatient care. However, they gathered data by observing patient- provider interactions and conducting exit surveys, hardly the best methods to explore sensitive concerns of ethnicity. Meanwhile, Agampodi and Amarasinghe (2007) found a higher prevalence of public sector use for immunization among Muslim households compared with Tamil and Sinhala users in Colombo. Since the researchers did not consider interaction between social class and ethnicity, these findings must be interpreted with caution.

The present study sheds light on the experiences of economically disadvantaged Muslim and Malay women accessing healthcare in Kandy. My findings suggest that these social groups experience neglect, abuse, and stigmatization within the public system, and, as a result, often avoid public healthcare. When lower-middle class Muslim women anticipated entering the public system, many channelled a specialist to ensure they received a respectful service. These women perceived the private sector to confer a degree of ‘protection’ to them, and thus consumed more private healthcare than did their lower-middle class Sinhala counterparts. Rather than enjoying an extended range of healthcare ‘choices,’ these women had fewer ‘alternatives’ since the public system – the only economically accessible option for those without means – did not present a convincing alternative. Notably, however, working class Muslim and Malay women stayed with the public sector, perhaps in the absence of economically feasible ‘alternatives.’ It would seem that ethnic minorities in Sri Lanka might be doubly disadvantaged as they face discrimination in the public sector and are thus rendered more vulnerable to the vagaries of the healthcare market.

The ‘Othering’ of Muslim women within the public healthcare system may be understood in relation to the Sinhala majoritarian post-colonial state and its failure to address the grievances of minorities (Kadirgamar 2013). The lack of Tamil/Sinhala competent bi-lingual healthcare providers is unsurprising given the deliberate education policies that have effectively separated Tamil and Sinhala speakers by language (and ethnicity) within primary and secondary schools for decades. Such divisions would necessarily give rise to learned prejudice against the ethnic ‘Other,’ in turn, translating to experiences of marginalization for minority ethnic groups (Davis

132

2011), particularly within public institutions where mechanisms of accountability are weak or absent. Such exclusions have material consequences as Muslim users are pushed toward private healthcare as the size of the proverbial ‘public healthcare pie’ shrinks under privatization.

Although not an objective of this study, women’s experiences in Kandy suggest that users with disabilities encounter significant challenges in accessing healthcare at public facilities. The public sector does not have basic accessibility services to attend to this group. Those with means access private healthcare where physical accessibility is evidently superior. Yet, these services are not available to economically disadvantaged users with physically disabilities. While the experiences of people with disabilities highlight the general absence of support services for people with disabilities in the country, Peiris-John and colleagues (2014) highlight major gaps in the disability literature in Sri Lanka, including on healthcare access. The state’s failure to invest in disability services may be guided by resource allocation policies that undervalue people with disabilities based on their future productivity, an approach reinforced by measures such as the disability- adjusted life year (DALY), currently used by international health and development agencies in health priority setting (Birn et al. 2017).

This study contributes new knowledge on the ways in which women, as classed, ethnic, and gendered subjects, navigated healthcare in the face of privatization. The layered experiences of oppression encountered by economically disadvantaged Muslim women and those with disabilities in the public system complicates a straightforward class analysis, and raises critical questions about the accessibility of Sri Lanka’s public system for marginal users. Having discussed the key contributions made by this thesis, I now turn to their implications for policy.

9.2 Policy implications

Despite evidence of inequitable outcomes, compromised quality and low cost-efficiency, the push for healthcare privatization continues, both at national and global levels. In Sri Lanka, the government plans to introduce national health insurance, a move that would surely weaken the existing publicly financed and delivered healthcare system. In this section, I discuss my findings in light of health reform in the contemporary neoliberal moment in Sri Lanka and other LMICs.

9.2.1 Unsustainable mixed health systems

The patterns of private healthcare use in Kandy reflect the inaccessibility of a mixed health system in the absence of public financing to support the consumption of private healthcare services. A

133 large majority of Udawatta residents could not address their health concerns comprehensively in the private sector owing to the unaffordability of private healthcare. Private hospitals were poorly developed in Kandy, providing services of questionable quality, perhaps owing to the absence of a sizeable market for private inpatient care. For the private health sector to offer economically accessible services of reasonable quality, the government needs to subsidize private healthcare through a national health insurance scheme or some other form of ‘public-private partnership.’ By partially covering a pre-defined set of procedures in the private sector, the President’s Fund currently does so, albeit to a very limited degree.

Since the 1980s, health reforms designed for and adopted by LMICs have supported the development of “pluralistic” health systems, combining public and private services to varying degrees (Balabanova, Conteh and McKee 2011:281). Among them, the more ‘successful’ health systems that made their way into the second Good Health at Low Cost initiative, sponsored by the Rockefeller Foundation in 2011, “[embraced] innovation” in the face of limited “financial and human resources,” while “[combining] resilience with flexibility” (Balabanova, McKee and Mills 2011:ix). Drawing on such discourses of privatization, international health and development agencies have long supported the introduction of ‘alternative’ healthcare financing strategies in LMICs (WHO 2010; World Bank 1993). In this context, health insurance has risen to prominence in the age of ‘Universal Health Coverage’ (Sengupta 2015).

Drawing on neo-classical economics, health insurance has its basis in separating the purchasing and providing functions of health systems, aimed at making health systems more ‘cost-efficient.’ While the ‘purchaser-provider split’ facilitates opening the health sector to the market (Birn et al. 2017; Sengupta 2015), where governments have committed to universal coverage, the profits to be made are limitless (Leys 2010; Loeppky 2010; Waitzkin 2011). National health insurance schemes have expanded access to a limited set of health services for specified populations (Knaul et al. 2012; Patel 2015), but have failed to bridge inequities in access in LMICs, owing to the excessive reliance on the private health insurance industry and multi-payer systems (Birn et al. 2016). State-subsidized insurances packages for economically disadvantaged users lack comprehensiveness, while wealthier formal sector workers receive superior health insurance benefits through contributory plans. Moreover, the health insurance industry is notorious for cherry-picking healthier clients for coverage (Barrientos and Lloyd-Sherlock 2000; Birn et al. 2016; Unger et al. 2008). Importantly, the implementation of health insurance schemes has diverted much-needed funds from the public sector to the private healthcare industry, resulting in

134 escalating government health spending across LMICs (De Groote et al. 2005; Sengupta 2015; Waitzkin et al. 2007).

In Sri Lanka, health insurance entered the national health policy agenda in the National Health Development Plan 2013-2017, implemented in parallel with the World Bank-supported Second Health Sector Development Project (Ministry of Health n.d.; World Bank 2013). While a national health insurance scheme was to be introduced by 2017 under this plan, the government has taken initial steps to implement a health insurance plan that will cover all school-going children (Ministry of Finance 2016). While the purchasing function of this nation-wide insurance scheme has been out-sourced to Sri Lanka Insurance, a state-owned corporation, the latter anticipates expanding coverage to the entire population (Kumar 2017).

The introduction of national health insurance in Sri Lanka would remove existing impediments to private healthcare expansion by providing (limited) ‘financial risk protection’ to those who cannot afford to pay for health services. As healthcare becomes increasingly commodified, the (manufactured) demand for health services will increase, expanding markets for biomedical technologies, and creating more opportunities for capital accumulation. As career opportunities increase in the private sector, private healthcare companies will entice physicians and other healthcare workers away from the public system, a trend that has already begun with specialists resigning from the public system to work full-time with commercial hospital ventures in Colombo (Dayaratne 2013). In other words, the introduction of national insurance will substantially weaken the public healthcare system.

In brief, contemporary health reforms in Sri Lanka support the concept of a mixed health system. To achieve equitable coverage, however, such a system requires public financing of private healthcare consumption. The government would need to subsidize a substantial share of private services for low-income groups to achieve even a semblance of universality. While increasing opportunities for private market expansion (and accumulation), this model will likely raise healthcare costs for the government. Given the prevailing ‘fiscal constraints’ in Sri Lanka’s domestic economy, a health sector development model with its basis in privatization does not offer a feasible policy direction. As discussed below, strengthening the existing publicly financed and delivered healthcare system will be more equitable and ‘efficient’ in the long-term.

135

9.2.2 Stronger public systems

Health systems that have achieved equitable ‘coverage’ have been built on decades of public investment. Sri Lanka, Kerala State (in India), China, Costa Rica, and Cuba,129 selected as case studies for the Rockefeller Foundation’s first Good Health at Low Cost study, at one time or another, emphasized welfare provision through redistribution under some form of socialist or social democratic government (Halstead et al. 1985). Even Thailand, today’s ‘poster child’ for health insurance among LMICs, achieved universal coverage through a single-payer model, preceded by decades of public investment in rural health infrastructure (Kongsri et al. 2011; Patcharanarumol et al. 2011). Among them, states that supported private healthcare expansion, including Sri Lanka, have fallen back on achievements in health equity (Thresia 2013).

Besides advancing equity in healthcare, strengthening Sri Lanka’s public system will prove more ‘cost-efficient’ in the long-term. The per capita health expenditure in Sri Lanka for the year 2014 was USD 369 (WHO 2017d), and of this 40 per cent was spent by the government (~ per capita USD 150), while the rest was financed through private sources (Institute for Health Policy 2015). The government of Sri Lanka spends less than 2 per cent of GDP on health (World Bank 2017c). While these conditions are far from ideal, the country may still be benefitting from ‘returns’ of welfare investments made under late colonialism and early independence (Birn et al. 2017). Still, the public system holds potential, if only it received the substantial state investment needed to improve equitable ‘coverage.’

As reflected in the experiences of Udawatta residents, there are significant deficits in public services, including the absence of a community-based system of primary curative care. Indeed, strengthening primary care has been identified as a critical need in the National Strategic Framework for Development of Health Services 2016-2025. The latter includes proposals for a “cluster system” to facilitate shared care with stronger links between public primary care services (primary care units and divisional hospitals) and specialist services at higher-tier public facilities (Ministry of Health 2016:13). Such a system would complement existing community-based preventive services, which recently expanded their mandate to include the prevention of non- communicable diseases (Senanayake et al. 2017). A clinic integrating preventive and curative

129 Cuba was included in the initial selection and left out for political reasons (Birn 2005).

136 services in each Division would improve access, provide continuity of care, and form the basis of a referral system that may relieve congestion at tertiary facilities.

However, strengthening primary care comes with financial disincentives for a range of powerful stakeholders, including the state, medical establishment, and healthcare industry. For one, such a system would require substantial state investment, particularly in human resources. If public primary care services strengthen and referral systems improve, private healthcare users may opt into the public system, requiring the state to invest more and more resources. A stronger public system would translate to reduced opportunities for private investment in the health sector, limiting its potential to attract foreign investment, again a deterrent for a government desperate for foreign exchange to address a looming balance-of-payment crisis. Meanwhile, both the healthcare industry and medical establishment would lose its clientele; public primary care services would make the services of private family doctors redundant and channelling centres may run out of business if congestion at tertiary care centres was reduced as a result of an effective referral system.

Strengthening a publicly financed and delivered healthcare system may not align ideologically with the agendas of international financial institutions. The push for healthcare privatization in Sri Lanka has intensified in the post-civil war context under the mediation of the IMF. In June 2016, the government received a USD 1.5 billion Extended Fund Facility from the IMF with accompanying fiscal conditions (Kotelawala 2017; Mudugamuwa 2016). The 2017 Budget, unveiled five months later, included a series of measures to narrow the budget deficit, including cuts to health spending (Moramudali and Riza 2016; Doole 2016). Noteworthy is that neoliberal health reforms in most LMICs have been adopted under the pressures of economic crisis (Mehrotra and Delamonica 2005; Roberts et al. 2003).

Any campaign to strengthen public healthcare would need to go hand in hand with a wider programme of public engagement. There is potential for such as women in Kandy demonstrated a critical awareness about the transformations taking place in healthcare. They naturally made distinctions between “the government service” and the “private” in describing their experiences of healthcare. But this space for resistance may diminish as more users view healthcare as a commodity, and take for granted the necessity for private healthcare.

Privatization individualizes the experience of accessing healthcare (McGregor 2001). As we saw in Kandy, privileged users felt less the impact of healthcare privatization as they exploited hybrid

137 arrangements to receive priority public services even as their practices pushed others to the margins. Confirming Bakker and Gill’s (2003b:36) contention that ‘reprivatization’ of social reproduction involves “social transformation from a collective sense of being to fragmented households and individuals,” this study suggests that the need to campaign for public healthcare in Sri Lanka is urgent.

9.3 Limitations and directions for future research

As with all research efforts, this study has its limitations. I drew on a liberal economic understanding of public/private (Weintraub 1997). From this perspective, public represented the state and private, the market. I adopted this definition because my intention was to interrogate the shift of healthcare from the state to the market. However, a more nuanced understanding of ‘public’ might have encompassed people-centred approaches, and a broader definition of ‘private’ might have compelled me to address the potential role of non-state actors, including NGOs and civil society groups, in achieving health systems change.

I conceptualized health services narrowly by focusing on institutionalized forms of allopathic healthcare. In doing so, I missed the health-related activities women undertake at home and did not address the social determinants of health. I left out indigenous medical services, which are seeing privatization, albeit in different forms. Notably, women constructed ‘public’ and ‘private’ in different ways in relation to allopathic and indigenous systems. They spoke of reputed independent Ayurveda practitioners who charged minimally for their services, suggesting that ‘private’ indigenous medical services may not be as commercially oriented. On the other hand, they did not share experiences at commercialized Ayurveda spas, an emerging focus of the medical tourism industry in Kandy.

Employing a qualitative research methodology, I engaged a limited number of women from selected areas of Udawatta in my research. Had I undertaken a survey of a large number of women, the knowledge I produced would have been generalizable. However, I did not intend to generalize women’s perspectives, but rather hoped to gain a deeper and more socially situated understanding of the implications of privatization for healthcare access in Kandy, Sri Lanka.

Selecting Kandy as my research setting placed considerable limits on my exploration of healthcare access in Sri Lanka. For one, I excluded socially and economically disadvantaged communities who are known to experience a greater degree of disadvantage in relation to

138 healthcare, in particular, rural dwellers, the up-country Tamil community (tea plantations workers), and ethnic minorities residing in war-torn northern and eastern parts of the country (Kottegoda et al. 2008; Perera et al. 2007; D’Almeida 2014). The dynamics of healthcare access and privatization in rural or plantation settings may look substantially different as the spread of private healthcare is limited in such settings. Moreover, exploring access in plantation regions and the war-torn north and east, where ethnic minorities constitute the ‘majority,’ may allow for a more nuanced understanding of the ethnic dimensions of healthcare privatization. Studies of healthcare access in various geographical locations where class and ethnic structures are different may yield critical knowledge on healthcare privatization in Sri Lanka.

The social class categories I conceptualized came out of an exploratory study, accepted practice in qualitative research. However, this means that any comparisons between class or income categories employed by others may not be possible (Wright 2009), reducing the relevance of my findings for policymakers. Furthermore, constructing Udawatta Division as a socially and ethnically divided space may have neglected the experiences of residents who did not fit into the selected class and/or ethnic categories. For example, Kirideniya was designated the middle/upper-middle class study area, but was also home to users from economically disadvantaged backgrounds, including squatters, who were not included in the study sample.

I missed the perspective of at least two groups of women in Udawatta. Although I recruited several middle/upper middle class women who worked outside their homes, very few employed women representing economically disadvantaged backgrounds participated in the study. Indeed, most women I met from the economically disadvantaged neighbourhoods were not formally employed, perhaps reflecting the low labour participation rate among women in Sri Lanka (World Economic Forum 2015). However, their perspective is important as they may have used private healthcare more extensively or foregone healthcare given the excessive demands on their time.

I also neglected the experiences of middle-class/upper-middle class Muslim women. The two upper-middle class women representing ethnic minorities who participated in this study reported being Tamil and of mixed ethnic heritage, respectively. They were extremely well connected and experienced numerous privileges at public and private healthcare facilities. Since there was no middle/upper-middle class group of Muslim women for comparison, I could not explore how class intersected with ethnicity in defining the experiences of economically disadvantaged Muslim women.

139

This study sought to explore class and ethnic dimensions of healthcare access under privatization. While my findings suggest that people with disabilities face distinct challenges in accessing healthcare, their experiences could form the basis of a study on access in future. Moreover, the experiences of other marginalized groups such as lesbian, gay, bi-sexual, transgender, queer and intersex groups may yield critical information about healthcare access in Sri Lanka (Human Rights Watch 2016; Rutnam 2016).

In limiting my study to the experiences of women, I missed the perspective of men accessing healthcare. I decided to focus on women for two reasons. First, my intention was to employ a critical feminist methodology that foregrounded women’s concerns in exploring privatization as a gendered process. Second, women are known to carry a greater burden of responsibility for caregiving in Sri Lanka (as elsewhere). I assumed that women, as a group, would have an intimate and experiential understanding of the transformations taking place in healthcare. Including men as study participants may have yielded interesting insights on sex-based differences in healthcare access under privatization. Having outlined the limitations of this study and some questions for future research, I move on to the conclusion of my thesis.

9.4 Conclusion

In this thesis, I explored the experiences of women residents of an urban administrative division to understand how the presence of private healthcare shapes access for diverse social groups in Kandy. I found that the range of healthcare ‘choices’ available to women were limited in the presence of private healthcare. While women across the social spectrum mixed public and private healthcare services to attend to their needs, the outcome of their reproductive labour was markedly different. Working class women were compelled to consume within a resource- constrained public system, while middle/upper-middle class women enjoyed the best of both worlds. This class-based narrative was troubled by the experiences of marginal others who opted out of the public system to receive a respectful service. Relying largely on women’s invisible reproductive labour, the hybrid arrangements that have emerged in the context of spreading privatization have individualized the responsibility for healthcare, while creating avenues for capital accumulation.

Yet, the public system still provides a semblance of universal ‘coverage.’ As we saw, even the most economically disadvantaged users in Kandy could receive ‘free’ treatment on a walk-in basis at public healthcare facilities. While strengthening this system would involve substantial public

140 investment, urgent policy interventions to address equity of access within the public system are needed. Among them, instituting mechanisms of accountability, implementing the language policy, investing in disability accessible services, and providing diversity training for healthcare providers, are urgent. Rather than investing in the public system, which has a proven record of accomplishment, however, successive governments in Sri Lanka have supported private healthcare expansion in their health reform efforts.

State investment in the private health sector finds support with the medical establishment, private healthcare industry and transnational capitalist classes who all stand to benefit from privatization. However, the accessibility and quality concerns related to private healthcare raised in this study, together with the health reform experiences of other LMICs, question the feasibility of private healthcare expansion in the quest for universal ‘coverage.’ Moreover, contemporary health reforms in Sri Lanka, designed under the tutelage of the IMF and World Bank, overlook the deeply interconnected nature of public and private healthcare sectors. Any separation of the purchasing and providing functions of Sri Lanka’s public healthcare system would invariably lead to its weakening with large sections of the population remaining ‘uncovered.’

As the incumbent government in Sri Lanka prepares for the introduction of national health insurance, my findings raise critical questions about the call to strengthen mixed health systems toward achieving ‘Universal Health Coverage’ (UHC) in LMICs. Although global health gurus may dismiss the health gains made by Sri Lanka on feasibility grounds, universality may not be possible in the absence of strong public healthcare systems. Critically oriented research that takes into account the interconnections between public and private, and the intersecting oppressions that shape healthcare access, is needed to better understand the implications of healthcare privatization for users. Such research would contribute toward challenging the ideological claims made by hegemonic actors and may support efforts to reclaim public systems in the quest for universal access. The wealth of knowledge possessed by women on the transformations taking place in the health sector in Sri Lanka suggests that possibilities for social change may lie with them.

141

References

Afsana, Kaosar. 2004. "The Tremendous Cost of Seeking Hospital Obstetric Care in Bangladesh." Reproductive Health Matters 12 (24):171-80.

Afsana, Kaosar and Sabina Faiz Rashid. 2001. “The Challenges of Meeting Rural Bangladeshi Women’s Needs in Delivery Care.” Reproductive Health Matters 9 (18):79-89.

Agampodi, S., and D. A. C. L. Amarasinghe. 2007. "Private Sector Contribution to Childhood Immunization: Sri Lankan Experience." Indian Journal of Medical Sciences 61 (4):192-200.

Akazili, James, Bertha Garshong, Moses Aikins, John Gyapong and Di McIntyre. 2012. “Progressivity of Health Care Financing and Incidence of Service Benefits in Ghana.” Health Policy and Planning 27 (Suppl 1):i13–i22.

Akinkugbe, Oluyele, Chitalu Mirriam Chama-Chiliba and Naomi Tlotlego. 2012. “Health Financing and Catastrophic Payments for Health Care: Evidence from Household-level Survey Data in Botswana and Lesotho.” African Development Review 24 (4):358-70.

Alavi, Hamza. 1972. "The State in Post-Colonial Societies: Pakistan and Bangladesh." New Left Review 74:59-81.

Alcock, Glyn, Sushmita Das, Neena Shah More, Ketaki Hate, Sharda More, Shanti Pantvaidya, David Osrin, and Tanja AJ Houweling. 2015. "Examining Inequalities in Uptake of Maternal Health Care and Choice of Provider in Underserved Urban Areas of Mumbai, India: A Mixed Methods Study." BMC Pregnancy and Childbirth 15 (1):231. DOI: 10.1186/s12884-015-0661-6.

Alcoff, Linda. 1991. "The Problem of Speaking for Others." Cultural critique 20:5-32.

Alvarez, Luz Stella, J. Warren Salmon, and Dan Swartzman. 2011. "The Colombian Health Insurance System and its Effect on Access to Health Care." International Journal of Health Services 41 (2):355-70.

Amarasinghe, Sarasi, Sanil De Alwis, Shanaz Saleem, Ravi P. Rannan-Eliya and Shanti Dalpatadu. 2015. Private Health Sector Review 2012. Retrieved June 17, 2017 (http://www.ihp.lk/publications/docs/PHSR2012.pdf).

Amin, Samir. 2006. "The Millennium Development Goals: A Critique from the South." Monthly Review 57 (10):1-15.

Amooti-Kaguna, Bwera and Fred Nuwaha. 2000. "Factors Influencing Choice of Delivery Sites in Rakai District of Uganda." Social Science & Medicine 50 (2):203-13.

Armada, Francisco and Carles Muntaner. 2004. “The Visible Fist of the Market: Health Reforms in Latin America.” Pp. 29-42 in Unhealthy Health Policy: A Critical Anthropological Examination edited by A. Castro and M. Singer. Lanham, MD: AltaMira Press.

Armstrong, Pat, Carol Amaratunga, Jocelyne Bernier, Karen Grant, Ann Pederson and Kay Willson. 2002. Exposing Privatization: Women and Health Care Reform in Canada. Aurora, ON: Garamond Press Ltd.

Aronowitz, Stanley. 2003. How Class Works. New Haven, CT and London, United Kingdom: Yale University Press.

142

Arrieta, Alejandro. 2011. “Health Reform and Caesarean Sections in the Private Sector: The Experience of Peru.” Health Policy 99:124-30.

Asiri Health. 2016. “About Laboratories.” Retrieved June 27, 2017 (http://www.asirihospitals.com/laboratory-services/our-services/about-laboratories).

Ataguba, John E. and Di McIntyre. 2012. “Paying For and Receiving Benefits from Health Services in South Africa: Is the Health System Equitable?” Health Policy and Planning 27 (Suppl. 1):i35–i45.

Baillie, Lesley. 2015. "Promoting and Evaluating Scientific Rigour in Qualitative Research." Nursing Standard 29 (46):36-42.

Bakker, Isabella. 2007. "Social Reproduction and the Constitution of a Gendered Political Economy." New Political Economy 12(4): 541-556.

Bakker, Isabella and Stephen Gill. 2003a. “Global Political Economy and Social Reproduction.” Pp. 3-16 in Power, Production and Social Reproduction edited by I. Bakker and S. Gill. Hampshire, United Kingdom and New York: Palgrave Macmillan.

Bakker, Isabella and Stephen Gill. 2003b. “Ontology, Method, and Hypotheses.” Pp. 17-41 in Power, Production and Social Reproduction edited by I. Bakker and S. Gill. Hampshire, United Kingdom and New York: Palgrave Macmillan.

Bakker, Isabella and Rachel Silvey. 2008. "Introduction: Social Reproduction and Global Transformations–From the Everyday to the Global." Pp. 1-15 in Beyond States and Markets: The Challenge of Social Reproduction edited by I. Bakker and R. Silvey. New York: Routledge.

Balabanova, Dina, Lesong Conteh and Martin McKee. 2011. “The Contribution of Health Systems to Good Health.” Pp. 269-306 in ‘Good Health at Low Cost’ 25 Years On edited by D. Balabanova, M. McKee and A. Mills. United Kingdom: London School of Hygiene and Tropical Medicine. Retrieved June 13, 2017 (http://ghlc.lshtm.ac.uk/files/2011/10/GHLC-book.pdf).

Balabanova, Dina, Martin McKee and Anne Mills, eds. 2011. ‘Good Health at Low Cost’ 25 Years On. London, United Kingdom: London School of Hygiene and Tropical Medicine. Retrieved June 13, 2017 (http://ghlc.lshtm.ac.uk/files/2011/10/GHLC-book.pdf).

Balasubramaniam, K. 2012. “Revolutionary Approach to Correcting Market Distortions in Access to and Delivery of Pharmaceuticals: Bibile Reforms 1972-1976.” Pp. 527-52 in Economic and Social Development under a Market Economy Regime in Sri Lanka edited by S. Kelegama and D. Gunewardena. Colombo Sri Lanka: Vijitha Yapa.

Ballinger Claire. 2006. “Demonstrating Rigour and Quality?” Pp. 235-46 in Qualitative Research for Allied Health Professionals: Challenging Choices edited by L. Finlay and C. Ballinger. Chichester, UK: John Wiley and Sons.

Bannerji, Himani. 1995. Thinking Through: Essays on Feminism, Marxism, and Anti-Racism. Toronto, ON: Women's Press.

Barber, Sarah L. 2006. "Public and Private Prenatal Care Providers in Urban Mexico: How does their Quality Compare?" International Journal for Quality in Health Care 18 (4): 306-13.

Barber, Sarah L., Stefano M. Bertozzi, and Paul J. Gertler. 2007. "Variations in Prenatal Care Quality for the Rural Poor in Mexico." Health Affairs 26 (3):w310-w323.

143

Barnett, Clare and Krishna Hort. 2013. “Approaches to Regulating the Quality of Hospital Services in Low- and Middle-Income Countries with Mixed Health Systems: A Review of their Effectiveness, Context of Operation and Feasibility.” Retrieved August 17, 2017 (http://ni.unimelb.edu.au/__data/assets/pdf_file/0010/834418/WP_32.pdf).

Barrientos, Armando, and Peter Lloyd-Sherlock. 2000. "Reforming Health Insurance in Argentina and Chile." Health Policy and Planning 15 (4):417-23.

Baru, Rama V. 2003. “The Privatization of Health Services: A South Asian Perspective.” Economic and Political Weekly 38(42):4433-37.

Basu, Sanjay, Jason Andrews, Sandeep Kishore, Rajesh Panjabi and David Stuckler. 2012. “Comparative Performance of Private and Public Healthcare Systems in Low- and Middle-Income Countries: A Systematic Review.” PLoS Med 9(6): e1001244. doi:10.1371/journal.pmed.1001244.

Benova, Lenka, David Macleod, Katharine Footman, Francesca Cavallaro, Caroline A. Lynch, and Oona MR Campbell. 2015. "Role of the Private Sector in Childbirth Care: Cross‐Sectional Survey Evidence from 57 Low‐ and Middle‐Income Countries using Demographic and Health Surveys." Tropical Medicine & International Health 20 (12):1657-73.

Benson, John S. 2001. "The Impact of Privatization on Access in Tanzania." Social Science & Medicine 52 (12):1903-15.

Berendes, Sima, Peter Heywood, Sandy Oliver and Paul Garner. 2011. "Quality of Private and Public Ambulatory Health Care in Low and Middle Income Countries: Systematic Review of Comparative Studies." PLoS Med 8(4):e1000433.

Bhate-Deosthali, Padma, Ritu Khatri, and Suchitra Wagle. 2011. "Poor Standards of Care in Small, Private Hospitals in Maharashtra, India: Implications for Public–Private Partnerships for Maternity Care." Reproductive Health Matters 19 (37):32-41.

Birn, Anne-Emanuelle. 2005. "Gates's Grandest Challenge: Transcending Technology as Public Health Ideology." The Lancet 366 (9484):514.

Birn, Anne-Emanuelle. 2009. "The Stages of International (Global) Health: Histories of Success or Successes of History?" Global Public Health 4 (1):50-68.

Birn, Anne‐Emanuelle, Laura Nervi, and Eduardo Siqueira. 2016. "Neoliberalism Redux: The Global Health Policy Agenda and the Politics of Cooptation in Latin America and Beyond." Development and Change 47 (4):734–759.

Birn, Anne-Emanuelle, Yogan Pillay, and Timothy H. Holtz. 2017. Textbook of Global Health. 4th ed. New York: Oxford University Press Inc.

Birn, Anne-Emanuelle, Sarah Zimmerman, and Richard Garfield. 2000. "To Decentralize or Not to Decentralize, is that the Question? Nicaraguan Health Policy under Structural Adjustment in the 1990s." International Journal of Health Services 30 (1):111-28.

Bisson, Gregory P., Ian Frank, Robert Gross, Vincent Lo Re III, Jordan B. Strom, Xingmei Wang, Mpho Mogorosi, Tendani Gaolathe, Ndwapi, Harvey Friedman, Brian L. Strom and Diana Dickinson. 2006. "Out-of-Pocket Costs of HAART Limit HIV Treatment Responses in Botswana's Private Sector." Aids 20 (9):1333-6.

Blas, Erik, and Mumbuwa E. Limbambala. 2001. "User-Payment, Decentralization and Health Service Utilization in Zambia." Health Policy and Planning 16 (Suppl 2):19-28.

144

BOI Sri Lanka. 2016. “About Us.” Retrieved June 17, 2017 (http://www.investsrilanka.com/about_us).

Boller, Christoph, Kaspar Wyss, Deo Mtasiwa, and Marcel Tanner. 2003. "Quality and Comparison of Antenatal Care in Public and Private Providers in the United Republic of Tanzania." Bulletin of the World Health Organization 81(2):116-22.

Borghi, Josephine, Tim Ensor, Basu Dev Neupane and Suresh Tiwari. 2006. “Financial Implications of Skilled Attendance at Delivery in Nepal.” Tropical Medicine and International Health 11 (2):228- 37.

Borghi, Josephine, Kara Hanson, C. Adjei Acquah, Gatien Ekanmian, Veronique Filippi, Carine Ronsmans, Ruari Brugha, Edmund Browne, and Eusebe Alihonou. 2003. "Costs of Near-Miss Obstetric Complications for Women and their Families in Benin and Ghana." Health Policy and Planning 18 (4):383-90.

Brinkmann, Svend and Steinar Kvale. 2015. InterViews: Learning the Craft of Qualitative Research Interviewing. 3d ed. Los Angeles, CA, London, UK, New Delhi, India, Washington, D.C. and Singapore: SAGE.

Brown, Wendy. 2016. “Sacrificial Citizenship: Neoliberalism, Human Capital, and Austerity Politics.” Constellations 23 (1):3-14.

Bustreo, Flavia, April Harding, and Henrik Axelsson. 2003. "Can Developing Countries Achieve Adequate Improvements in Child Health Outcomes without Engaging the Private Sector?" Bulletin of the World Health Organization 81 (12):886-95.

Campbell, Oona MR, Lenka Benova, David MacLeod, Rebecca F. Baggaley, Laura C. Rodrigues, Kara Hanson, Timothy Powell‐Jackson, Loveday Penn-Kekana, Reen Polonsky, Katharine Footman, Alice Vahanian, Shreya K. Pereira, Andreia Costa Santos, Veronique G.A. Filippi, Caroline A. Lynch and Catherine Goodman. 2016. "Family Planning, Antenatal and Delivery Care: Cross‐ Sectional Survey Evidence on Levels of Coverage and Inequalities by Public and Private Sector in 57 Low‐ and Middle‐Income Countries." Tropical Medicine & International Health 21 (4):486- 503.

Castro, Arachu, and Merrill Singer. 2004. Unhealthy Health Policy: A Critical Anthropological Examination. Lanham, MD: AltaMira Press.

Cavagnero, Eleonora, Guy Carrin, Ke Xu, and Ana Mylena Aguilar-Rivera. 2006. "Health Financing in Argentina: An Empirical Study of Health Care Expenditure and Utilization." Retrieved November 19, 2016 (http://www.who.int/health_financing/documents/argentina_cavagnero.pdf).

Chankova, Slavea, Sara Sulzbach, and Francois Diop. 2008. "Impact of Mutual Health Organizations: Evidence from West Africa." Health Policy and Planning 23 (4):264-276.

Chengsorn, N., E. Bloss, R. Anekvorapong, A. Anuwatnonthakate, W. Wattanaamornkiat, S. Komsakorn, S. Moolphate, P. Limsomboon, S. Kaewsa-ard, S. Nateniyom, A. Kanphukiew and J. K. Varma. 2009. "Tuberculosis Services and Treatment Outcomes in Private and Public Health Care Facilities in Thailand, 2004–2006." The International Journal of Tuberculosis and Lung Disease 13 (7):888-94.

Cissé, Boubou, Stéphane Luchini, and Jean Paul Moatti. 2007. "Progressivity and Horizontal Equity in Health Care Finance and Delivery: What about Africa?" Health Policy 80 (1):51-68.

Coburn, David. 2010. “Inequality and Health.” Socialist Register 46:39-58.

145

Colombo Telegraph. 2014. ““Sri Lankan Government: Take Urgent Action to Stop Attacks on Muslims” Say 300 Sri Lankans.” Colombo Telegraph, June 19. Retrieved June 15, 2017 (https://www.colombotelegraph.com/index.php/sri-lankan-government-take-urgent-action-to-stop- attacks-on-muslims-say-300-sri-lankans/).

Cotlear, Daniel, Somil Nagpal, Owen Smith, Ajay Tandon, and Rafael Cortez. 2015. Going Universal: How 24 Developing Countries are Implementing Universal Health Coverage Reforms from the Bottom Up.” Retrieved June 17, 2017 (http://documents.worldbank.org/curated/en/936881467992465464/pdf/99455-PUB-Box393200B- OUO-9-PUBDATE-9-28-15-DOI-10-1596-978-1-4648-0610-0-EPI-210610.pdf).

Credit Suisse Research Institute. 2016. “Global Wealth Report 2016.” Retrieved April 8, 2017 (http://publications.credit-suisse.com/tasks/render/file/index.cfm?fileid=AD783798-ED07-E8C2- 4405996B5B02A32E).

Cueto, Marcos. 2004. "The Origins of Primary Health Care and Selective Primary Health Care." American Journal of Public Health 94 (11):1864-1874.

Daily FT. 2014. “First-Ever Bone Marrow Transplant Unit Opens at Central Hospital.” Daily FT, June 9. Retrieved June 18, 2017 (http://www.ft.lk/2014/06/09/first-ever-bone-marrow-transplant-unit- opens-at-central-hospital/).

Daily FT. 2016. “Paying Wards in State Hospitals for Public Servants.” Daily FT, January 10. Retrieved June 18, 2017 (http://www.ft.lk/2012/01/10/pay-wards-in-state-hospitals-for-public-servants/).

Daily Mirror. 2016. “Don’t Label us as Indian Tamils: TPA.” Daily Mirror, January 27. Retrieved September 30, 2017 (http://www.dailymirror.lk/104301/mils-tpa).

D’Almeida, Kanya. 2014. “On Sri Lanka’s Tea Estates, Maternal Health Leaves a Lot to be Desired.” Daily FT, September 27. Retrieved August 16, 2017 (http://www.ft.lk/article/356889/On-Sri-Lanka-s- tea-estates--maternal-health-leaves-a-lot-to-be-desired).

Damrongplasit, Kannika, and Glenn A. Melnick. 2009. "Early Results from Thailand’s 30 Baht Health Reform: Something to Smile About." Health Affairs 28 (3):w457-w466.

Danese-De los Santos, L., A. Valencia-Mendoza, and S. Sosa-Rubí. 2011. "Analysis of Changes in Health-Care Service Access and Choice of Provider in Mexico: The Utilization of Curative Health Services 2000-2006." BMC Public Health 11:771 doi:10.1186/1471-2458-11-771.

Daniel, Smriti. 2016. “Why Sri Lanka Beats Indian in Maternal Mortality Ratios.” Al Jazeera, March 14. Retrieved June 18, 2017 (http://www.aljazeera.com/indepth/features/2016/03/sri-lanka-beats- india-maternal-mortality-ratios-160308105127735.html).

Davidson, Christina. 2009. "Transcription: Imperatives for Qualitative Research." International Journal of Qualitative Methods 8 (2):35-52.

Davis, Christina Parks. 2011. “Education in the Language of Conflict: Linguistic and Social Practice among Sri Lankan Ethnic Minority Youth.” PhD Dissertation, Department of Anthropology, University of Michigan.

Dayaratne, G.D. 2012. ”Private Expenditure in the Health Sector under the Free Market Regime.” Pp. 477-504 in Economic and Social Development under a Market Economy Regime in Sri Lanka: Buddhadasa Hewavitharana Felicitation Volume 2 edited by Saman Kelegama and Dileni Gunewardena. Colombo, Sri Lanka: Vijitha Yapa Publications.

146

Dayaratne, G. D. 2013. Private Hospital Health Care Delivery in Sri Lanka: Some Issues on Equity, Fairness and Regulation. Colombo, Sri Lanka: Institute of Policy Studies.

De Costa, Ayesha, Kranti S. Vora, Kayleigh Ryan, Parvathy Sankara Raman, Michele Santacatterina and Dileep Mavalankar. 2014. "The State-Led Large Scale Public Private Partnership ‘Chiranjeevi Program’ to Increase Access to Institutional Delivery among Poor Women in Gujarat, India: How has it done? What can we Learn?" PLoS One 9 (5):e95704. doi.org/10.1371/journal.pone.0095704.

De Groote, Tony, Pierre De Paepe, and Jean-Pierre Unger. 2005. "Colombia: In vivo Test of Health Sector Privatization in the Developing World." International Journal of Health Services 35 (1):125- 41.

Department of Ayurveda Central Province. n.d. “Hospitals and Central Dispensaries.” Retrieved June 27, 2017 (http://www.ayurveda.cp.gov.lk/en/contact-us/hospitals-and-central-dispensaries.html).

Department of Census and Statistics. n.d. “Census of Population and Housing 2012 (New) – GN Level Tables.” Retrieved August 16, 2017 (http://www.statistics.gov.lk/PopHouSat/CPH2011/index.php?fileName=Activities/TentativelistofP ublications).

Department of Census and Statistics. 2015a. Census of Population and Housing 2012. Retrieved November 16, 2016 (http://www.statistics.gov.lk/PopHouSat/CPH2011/Pages/Activities/Reports/FinalReport/FinalRep ortE.pdf)

Department of Census and Statistics. 2015b. Household Income and Expenditure Survey 2012/13: Final Report. Retrieved August 16, 2017 (http://www.statistics.gov.lk/HIES/HIES2012_13FinalReport.pdf).

Department of Census and Statistics and Ministry of Healthcare and Nutrition. 2009. Sri Lanka Demographic and Health Survey 2006-07. Colombo, Sri Lanka: Department of Census and Statistics and Ministry of Healthcare and Nutrition.

Department of National Planning. 2010. Mahinda Chinthana Vision for the Future: The Development Policy Framework Government of Sri Lanka. Colombo, Sri Lanka: Department of National Planning.

Desclaux, Alice. 2004. “Equity in Access to AIDS Treatment in Africa: Pitfalls among Achievements. Pp. 115-32 in Unhealthy Health Policy: A Critical Anthropological Examination edited by A. Castro and M. Singer. Lanham, MD: AltaMira Press.

Deshodaya Movement. 2015. “Summary of UNFGG, UPFA, JVP & TNA manifestos: Parliamentary elections 2015.” Colombo Telegraph, August 13. Retrieved June 17, 2017 (https://www.colombotelegraph.com/index.php/summary-of-unfgg-upfa-jvp-tna-manifestos- parliamentary-elections-2015/).

De Silva, Amala, Shanthi Dalpatadu, Ananda Mohan Das, Reggie Perera, and Sarath Samarage. n.d. Public Sector Consultants in Private Practice in Sri Lanka: Potential and Prospects. Colombo, Sri Lanka: Unknown.

De Silva, Charumini. 2016. “Asiri Invests Rs. 5b for hospital in Kandy.” Daily FT, December 8. Retrieved June 25, 2017 (http://www.ft.lk/article/584732/Asiri-invests-Rs--5-b-for-hospital-in-Kandy).

De Silva, Chandra Richard. [1987] 1997. Sri Lanka: A History. New Delhi, India: Vikas Publishing House Pvt. Ltd.

147

De Silva, Indrawansa. 2015. “Ven. Sobitha’s Untimely Death.” The Island, November 15. Retrieved February 16, 2016 (http://www.island.lk/index.php?page_cat=article-details&page=article- details&code_title=135409).

Dissanayake, Chathuri. 2015. “Macabre Trade in Medical Devices.” The Sunday Times, June 14. Retrieved June 18, 2017 (http://www.sundaytimes.lk/150614/news/macabre-trade-in-medical- devices-153450.html).

Doole, Cassandra. 2016. “Budget 2017: Education Down, President’s Expenditure Up.” Roar Reports, November 8. Retrieved https://roar.media/english/reports/reports/budget-2017-education- presidents-expenditure/).

Economic Review (1987). Health and Medical Services. Economic Review, 12(12), 3-4, 6-13.

Economynext. 2016. “Sri Lanka Emergency Ambulance Paramedic Service with 1990 Toll Free Number.” economynext, July 29. Retrieved June 18, 2017 (http://www.economynext.com/Sri_Lanka_emergency_ambulance_paramedic_service_with_1990 _toll_free_number-3-5721-7.html).

Ekman, Bjorn. 2007. “Catastrophic Health Payments and Health Insurance: Some Counterintuitive Evidence from One Low-Income Country.” Health Policy 83 (2-3):304-13.

Ensor, Tim and Jeptepkeny Ronoh. 2005. "Effective Financing of Maternal Health Services: A Review of the Literature." Health Policy 75 (1):49-58.

Erus, Burcay and Nazli Aktakke. 2012. “Impact of Healthcare Reforms on Out-of-Pocket Health Expenditures in Turkey for Public Insurees.” European Journal of Health Economics 13:337-46.

Ewig, Christina. 2008. "Reproduction, Reform and the Reconfigured State: Feminists and Neoliberal Health Reforms in Chile.” Pp. 143-58 in Beyond States and Markets: The Challenge of Social Reproduction edited by I. Bakker and R. Silvey. New York: Routledge.

Ewig, Christina and Amparo Hernandez Bello. 2009. “Gender Equity and Health Sector Reform in Colombia: Mixed State-market Model Yields Mixed Results.” Social Science & Medicine 68:1145– 52.

Family Health Bureau. 2014. Annual Report on Family Health 2013. Colombo, Sri Lanka: Family Health Bureau, Ministry of Health.

Fanon, Frantz. [1963]1991. The Wretched of the Earth. New York: Grove Weidenfeld.

Farook, Latheef. 2014. Muslim of Sri Lanka: Under Siege. Colombo, Sri Lanka: Latheef Farook.

Ferguson, James. 2010. "The Uses of Neoliberalism." Antipode 41 (Suppl 1):166-84.

Fernando, Joel. 2013. “Health Care Financing in Sri Lanka – Retaining Welfare and Promoting Markets.” OPA Journal 28:12-19.

Fernandopulle, Lalin. 2016. “Asiri to Build Rs. 5 Billion Hospital in Kandy.” Sunday Observer, December 11. Retrieved June 18, 2017 (http://www.sundayobserver.lk/2016/12/11/asiri-build-rs-5-billion- hospital-kandy).

Ferrinho, Paulo, Wim Van Lerberghe, Inês Fronteira, Fátima Hipólito, and André Biscaia. 2004. "Dual Practice in the Health Sector: Review of the Evidence." Human Resources for Health 2 (1):14. doi: 10.1186/1478-4491-2-14.

148

Fine, Ben. 2013. “Financialization from a Marxist Perspective.” International Journal of Political Economy 42 (4):47-66.

Finlay, Linda. 2006. "‘Rigour,’ ‘Ethical Integrity’ or ‘Artistry’? Reflexively Reviewing Criteria for Evaluating Qualitative Research." The British Journal of Occupational Therapy 69 (7): 319-326.

Friends of Manipay. N.d. “Home.” Retrieved October 19, 2017 (https://manipayhospital.org/).

Frisby, Wendy and Gillian Creese. 2011. “Unpacking Relationships in Feminist Community Research: Crosscutting Themes.” Pp. 1-15 in Feminist Community Research: Case Studies and Methodologies edited by G. Creese and W. Frisby. Vancouver, BC and Toronto, ON: UBC Press.

Gage, Anastasia J. 2007. "Barriers to the Utilization of Maternal Health Care in Rural Mali." Social Science & Medicine 65 (8):1666-82.

Gao, Jun, Shenglan Tang, Rachel Tolhurst and Keqing Rao. 2001. “Changing Access to Health Services in Urban China: Implications for Equity.” Health Policy and Planning 16 (3):302-12.

Gao, Yu, Lesley Barclay, Sue Kildea, Min Hao, and Suzanne Belton. 2010. "Barriers to Increasing Hospital Birth Rates in Rural Shanxi Province, China." Reproductive Health Matters 18 (36):35- 45.

Gericke, C. A. 2005. “Comparison of Health Care Financing in Egypt and Cuba: Lessons for Health Reform in Egypt.” Eastern Mediterranean Health Journal 11 (5/6):1073-86.

Giacomini, Mita. 2010. “Theory Matters in Qualitative Health Research.” Pp. 125-55 in The Sage Handbook of Qualitative Methods in Health Research edited by I. Bourgeault, R. Dingwall and R. de Vries. London: SAGE.

Gideon, Jasmine. 2008. “Counting the Cost of Privatised Provision: Women, Rights and Neoliberal Health Reforms in Chile.” IDS Bulletin 39 (6):75-82.

Gideon, Jasmine. 2012. "Engendering the Health Agenda? Reflections on the Chilean Case, 2000– 2010." Social Politics 19 (3):333-60.

GMOA. 2016. “Budget 2017 – Issues.” Retrieved June 17, 2017 (http://www.gmoa.lk/index.php/2016/11/budget-2017-issues/).

Goldthorpe, John H., and Gordon Marshall. 1992. “The Promising Future of Class Analysis: A Response to Recent Critiques.” Sociology 26 (3):381-400.

Gonzalez-Perez, Guillermo J., Maria G. Vega-Lopez, Carlos Cabrera-Pivaral, Armando Muñoz, and Ana Valle. 2001. "Caesarean Sections in Mexico: Are There Too Many?" Health Policy and Planning 16 (1): 62-7.

Google. 2016. “Map data.” Retrieved June 17, 2017 (https://www.google.lk/maps/@7.2892655,80.6106746,13z).

Government Information Centre. 2009. “Health Assistance.” Retrieved June 17, 2017 (http://www.gic.gov.lk/gic/index.php?option=com_info&id=39&catid=7&task=subcat&lang=en).

Government of Sri Lanka. 1952. “Health Services Act.” Retrieved November 16, 2016 (http://www.commonlii.org/lk/legis/consol_act/hs550204.pdf).

149

Govindaraj, Ramesh, Kumari Navaratne, Eleonora Cavagnero and Shreelata Rao Seshadri. 2014. Health Care in Sri Lanka: What Can the Private Health Sector Offer? Retrieved June 14, 2017 (https://openknowledge.worldbank.org/bitstream/handle/10986/20018/899540WP0Box380th0Car e0in0Sri0Lanka.pdf).

Guba, Egon G. and Yvonna S. Lincoln. 2005. “Paradigmatic Controversies, Contradictions, and Emerging Confluences.” Pp 191-215 in Handbook of Qualitative Research edited by Norman K. Denzin & Yvonna S. Lincoln. Thousand Oaks, CA: Sage.

Gunatilleke, Godfrey. 1985. “Health and Development in Sri Lanka – An Overview.” Pp. 111-24 in Good Health at Low Cost edited by S. B. Halstead, J. A. Walsh and K. S. Warren. New York: The Rockefeller Foundation.

Gupta, Sangeeta, James T. Gunter, Robert J. Novak, and James L. Regens. 2009. "Patterns of Plasmodium Vivax and Plasmodium Falciparum Malaria underscore Importance of Data Collection from Private Health Care Facilities in India." Malaria Journal 8 (1): 227. https://doi.org/10.1186/1475-2875-8-227.

Ha, Nguyen Thi Hong, Peter Berman and Ulla Larsen. 2002. “Household Utilization and Expenditure on Private and Public Services in Vietnam.” Health Policy and Planning 17 (1):61-70.

Hafi, Fathima Riznaz. 2017. “Asiri Central’s neuro-interventional treatment: Minimising physical and psychological trauma.” Daily FT, January 30. Retrieved June 27, 2017 (http://www.ft.lk/article/594473/Asiri-Central-s-neuro-interventional-treatment--Minimising- physical-and-psychological-trauma).

Halstead, Scott B., Julia A. Walsh and Kenneth S. Warren, eds. 1985. Good Health at Low Cost. New York: The Rockefeller Foundation.

Haniffa, Ruvaiz. 2006. “Is Health a Right or Commodity? – Part II, Why Sri Lanka Needs a National Health Policy!” The Island, April 19. Retrieved June 11, 2017 (http://www.island.lk/2006/04/19/features3.html).

Hanson, Kara, Lucy Gilson, Catherine Goodman, Anne Mills, Richard Smith, Richard Feachem, Neelam Sekhri Feachem, Tracey Perez Koehlmoos, and Heather Kinlaw. 2008. "Is Private Health Care the Answer to the Health Problems of the World's Poor?" PLoS Medicine 5 (11): e233. doi.org/10.1371/journal.pmed.0050233

Hanvoravongchai, Piya. 2013. Health Financing Reform in Thailand: Toward Universal Coverage under Fiscal Constraints. Washington, D.C.: The World Bank. Retrieved June 27, 2017 (https://openknowledge.worldbank.org/bitstream/handle/10986/13298/75000.pdf?sequence=1&is Allowed=y).

Harvey, David. 2005. A Brief History of Neoliberalism. Oxford, England: Oxford University Press.

Harvey, David. 2014. Seventeen Contradictions and the End of Capitalism. London, UK: Profile Books.

Hennink, Monique M. 2007. International Focus Group Research: A Handbook for the Health and Social Sciences. Cambridge, UK, New York, Melbourne, Australia, Madrid, Spain, Cape Town, South Africa, Singapore and Sao Paulo, Brazil: Cambridge University Press.

Herath, H. M. A. 2015. “Place of Women in Sri Lankan Society: Measures for Their Empowerment for Development and Good Governance.” VJM 1 (1): 1. Retrieved June 15, 2017 (http://mgt.sjp.ac.lk/vjm/wp-content/uploads/2015/11/Place-of-Women-in-Sri-Lankan-Society...- HMA-Herath.pdf).

150

Herr, Kathryn and Gary L. Anderson. 2005. The Action Research Dissertation: A Guide for Students and Faculty. London, UK, Thousand Oaks, CA and New Delhi, India: SAGE.

Herr, Ranjoo Seodu. 2014. "Reclaiming Third World Feminism: Or Why Transnational Feminism Needs Third World Feminism." Meridians: feminism, race, transnationalism 12 (1):1-30.

Herring, Ronald J. 1987. “Economic Liberalisation Policies in Sri Lanka: International Pressures, Constraints and Supports.” Economic and Political Weekly 22 (8):325-33.

Hesse-Biber, Sharlene Nagy and Patricia Leavy. 2011. The Practice of Qualitative Research. Thousand Oaks, CA, London, United Kingdom, New Delhi, India and Singapore: SAGE.

Hewa, Soma. 1995. Colonialism, Tropical Disease and Imperial Medicine: Rockefeller Philanthropy in Sri Lanka. Lanham, MD: University Press of America.

Hewavitharana, Buddhadasa. 2004. “Poverty Alleviation.” Pp. 467-496 in Economic Policy in Sri Lanka: Issues and Debates, edited by S. Kelegama. New Delhi, India: Sage Publications India Pvt. Ltd.

Hipgrave, David Barry and Krishna Hort. 2014. “Dual Practice by Doctors Working in South and East Asia: A Review of its Origins, Scope, Impact and the Options for Regulation.” Health Policy and Planning 29 (6):703-16.

Hoa, Nguyen B., Frank G.J. Cobelens, Dinh N. Sy, Nguyen V. Nhung, Martien W. Borgdorff, and Edine W. Tiemersma. 2011. "Diagnosis and Treatment of Tuberculosis in the Private Sector, Vietnam." Emerging Infectious Diseases 17 (3):562-3.

Honda, Ayako, Pierana Gabriel Randaoharison and Mitsuaki Matsui. 2011. “Affordability of Emergency Obstetric and Neonatal Care at Public Hospitals in Madagascar.” Reproductive Health Matters 19(37):10-20.

Hoole, Rajan, Daya Somasundaram, K. Sritharan and Rajani Thiranagama. 1990. The Tamil Crisis in Sri Lanka – An Inside Account. Claremont, CA: The Sri Lanka Studies Institute.

Hotchkiss, David R., Deepali Godha, and Mai Do. 2014. "Expansion in the Private Sector Provision of Institutional Delivery Services and Horizontal Equity: Evidence from Nepal and Bangladesh." Health Policy and Planning 29 (Suppl 1):i12-i19.

Hsiao, William. 2000. A Preliminary Assessment of Sri Lanka’s Health Sector and Steps Forward. Cambridge, MA: Harvard University.

Huff-Rousselle, Maggie and Helen Pickering. 2001. “Crossing the Public-Private Sector Divide with Reproductive Health in Cambodia: Out-Patient Services in a Local NGO and the National MCH Clinic.” International Journal of Health Planning and Management 16 (1):33-46.

Human Rights Watch. 2016. “Sri Lanka: Challenging ‘Gender Norms’ Brings Abuse.” Retrieved June 27, 2017 (https://www.hrw.org/news/2016/08/15/sri-lanka-challenging-gender-norms-brings-abuse).

Institute for Health Policy. 2015. Sri Lanka Health Accounts: National Health Expenditure 1990-2014. Retrieved June 14, 2017 (http://www.ihp.lk/publications/docs/HES1504.pdf).

International Finance Corporation. 2017. “Improving Emerging Markets Healthcare through Private Provision.” Retrieved September 30, 2017 (https://www.ifc.org/wps/wcm/connect/b23322f4-ff5a- 405b-b13d-5e1165335057/EMCompass+Note+31+PSD+Healthcare+MB+1- 30.pdf?MOD=AJPERES).

151

Jacobs, Bart, James Whitworth, Fred Kambugu, and Robert Pool. 2004. "Sexually Transmitted Disease Management in Uganda’s Private-for-Profit Formal and Informal Sector and Compliance with Treatment." Sexually Transmitted Diseases 31 (11):650-54.

Jaffna Jaipur Centre for Disability Rehabilitation. 2012. “About us.” Retrieved October 18, 2017 (http://www.jjcdr.com/?page_id=2).

Jasso-Aguilar, Rebeca, Howard Waitzkin, and Angela Landwehr. 2004. “Multinational Corporations and Health Care in the United States and Latin America: Strategies, Actions and Effects." Journal of Health and Social Behavior 45(Suppl):136-157.

Jayamanne, Dilanthi. 2013. “Over 5,000 on Waiting List for Heart Surgery.” The Island, March 5. Retrieved June 18, 2017 (http://www.island.lk/index.php?page_cat=article-details&page=article- details&code_title=74821).

Jayamanne, Dilanthi and Norman Palihawadana. 2014. “Lady Doctor’s Death Jolts Authorities into Inspecting Beauty Parlours.” The Island, August 13. Retrieved June 18, 2017 (http://www.island.lk/index.php?page_cat=article-details&page=article- details&code_title=108444).

Jayasuriya, Laksiri. 2010. Taking Social Development Seriously: The Experience of Sri Lanka. New Delhi, India: SAGE Publications India Pvt. Ltd.

Jayasuriya, Lucian. 2017. Email Communication, July 25, 2017.

Jayasuriya, V. 2014. "Utility of Qualitative Methods in a Clinical Setting: Perinatal Care in the Western Province." Ceylon Medical Journal 57 (1):10-13.

Jayawardena, Priyanka. 2016. “Burning Health Costs for Poor in Sri Lanka.” Colombo, Sri Lanka: IPS Talking Economics. Retrieved June 14, 2017 (http://www.ips.lk/talkingeconomics/2016/11/16/burning-health-costs-for-poor-in-sri-lanka/).

Jayawardena, Kumari and Rachel Kurian. 2015. Class, Patriarchy and Ethnicity on Sri Lankan Plantations: Two Centuries of Power and Protest. New Delhi, India: Orient BlackSwan.

Jayawardena, Visakha Kumari. 1972. The Rise of the Labor Movement in Ceylon. Durham, NC: Duke University Press.

Jehu-Appiah, Caroline, Genevieve Aryeetey, Ernst Spaan, Thomas de Hoop, Irene Agyepong and Rob Baltussen. 2011. “Equity Aspects of the National Health Insurance Scheme in Ghana: Who is Enrolling, Who is Not and Why?” Social Science & Medicine 72(2):157-65.

Jones, Margaret. 2004. Health Policy in Britain’s Model Colony: Ceylon (1900-1948). New Delhi, India: Orient Longman.

Jones, Margaret. 2009. The Hospital System and Health Care: Sri Lanka, 1815-1960. New Delhi, India: Orient BlackSwan.

Kadirgamar, Ahilan. 2013. “Second Wave of Neoliberalism: Financialisation and Crisis in Post-War Sri Lanka.” Economic and Political Weekly (Web Exclusive), August 31. Retrieved June 17, 2017 (http://www.epw.in/node/128305/pdf).

Kadirgamar, Ahilan. 2014. “Reconstructing the North and Democratising Sri Lanka.” Himal South Asian, February 7. Retrieved June 15, 2017 (http://old.himalmag.com/vacancy/5232-reconstructing-the- north.html).

152

Kadirgamar, Ahilan. 2017. “Whither Human Rights in Sri Lanka?” The Hindu, March 29. Retrieved June 15, 2017 (http://www.thehindu.com/opinion/lead/whither-human-rights-in-sri- lanka/article17708475.ece).

Kadirgamar, Santasilan. 1989. “Lanka: Nationalism, Self-Determination and Conflict.” Pp. 181-212 in Ethnicity Identity Conflict Crisis edited by K. David and S. Kadirgamar. Kowloon, Hong Kong: ARENA Press.

Kaufman, Joan and Fang Jing. 2002. "Privatisation of Health Services and the Reproductive Health of Rural Chinese Women." Reproductive Health Matters 10 (20):108-16.

Keesara, Sirina R., Pamela A. Juma and Cynthia C. Harper. 2015. "Why do Women Choose Private over Public Facilities for Family Planning Services? A Qualitative Study of Post-Partum Women in an Informal Urban Settlement in Kenya." BMC Health Services Research 15 (1):335.

Kelegama, Saman, Ravi Rannan-Eliya and Nishan de Mel. 1997. “The Evolution of the Insurance Services and Private Health Insurance Market in Sri Lanka.” Economic Review (February):17-23.

Kiely, Ray. 2005. The Clash of Globalisations: Neoliberalism, the Third Way and Anti-Globalisation. Leiden, the Netherlands and Boston, MA: Brill.

Kierans, Ciara, Cesar Padilla-Altamira, Guillermo Garcia-Garcia, Margarita Ibarra-Hernandez, and Francisco J. Mercado. 2013. "When Health Systems are Barriers to Health Care: Challenges Faced by Uninsured Mexican Kidney Patients." PloS One 8 (1):e54380.

Kitzinger, Jenny. 1994. “The Methodology of Focus Groups: The Importance of Interaction between Research Participants.” Sociology of Health and Illness 16 (1):103-21.

Kitzinger, Jenny. 1995. “Introducing Focus Groups.” BMJ 311 (7000):299-302.

Knaul, Felicia Marie, Eduardo González-Pier, Octavio Gómez-Dantés, David García-Junco, Héctor Arreola-Ornelas, Mariana Barraza-Lloréns, Rosa Sandoval, Francisco Caballero, Mauricio Hernández-Avila, Mercedes Juan, David Kershenobich, Gustavo Nigenda, Enrique Ruelas, Jaime Sepúlveda, Roberto Tapia, Guillermo Soberón, Salomón Chertorivski and Julio Frenk. 2012 "The Quest for Universal Health Coverage: Achieving Social Protection for All in Mexico." The Lancet 380 (9849):1259-79.

Knight, Lucia and Pranitha Maharaj. 2009. “Use of Public and Private Health Services in KwaZulu-Natal, South Africa.” Development Southern Africa 26 (1):17-28.

Kongsri, Suratchada, Supon Limwattananon, Supakit Sirilak, Phusit Prakongsai, and Viroj Tangcharoensathien. 2011. "Equity of Access to and Utilization of Reproductive Health Services in Thailand: National Reproductive Health Survey Data, 2006 and 2009." Reproductive Health Matters 19 (37):86-97.

Kotelawala, Himal. 2017. “Is Sri Lanka Facing an Economic Crisis?” Roar Reports, February 28. Retrieved June 27, 2017 (https://roar.media/english/reports/editorial/is-sri-lanka-facing-an- economic-crisis/).

Kottegoda, Sepali, Kumudini Samuel, and Sarala Emmanuel. 2008. "Reproductive Health Concerns in Six Conflict-Affected Areas of Sri Lanka." Reproductive Health Matters 16 (31):75-82.

Kotz, David M. 2011. “Financialization and Neoliberalism.” Pp. 1-18 in Relations of Global Power: Neoliberal Order and Disorder, edited by G. Teeple and S. McBride. Toronto, ON: University of Toronto Press.

153

Kruk, Margaret E., Emily Goldmann, and Sandro Galea. 2009. "Borrowing and Selling to Pay for Health Care in Low-and Middle-Income Countries." Health Affairs 28 (4):1056-1066.

Kumar, Ramya. 2015a. “Preserving “free health” under Sri Lanka’s privatisation policy.” The Sunday Times, February 8. Retrieved June 18, 2017 (http://www.sundaytimes.lk/150208/business- times/preserving-free-health-under-sri-lankas-privatisation-policy-134353.html).

Kumar, Ramya. 2015b. “Unequal access to health care in Sri Lanka”. The Sunday Times, September 27. Retrieved June 18, 2017 (http://www.sundaytimes.lk/150927/sunday-times-2/unequal-access-to- healthcare-in-sri-lanka-165845.html).

Kumar, Ramya. 2017. “‘Free’ Health Insurance for School-Goers? Channelling Public Funds for Private Profit.” Daily Mirror, August 29. Retrieved September 30, 2017 (http://www.dailymirror.lk/article/- Free-Health-Insurance-for-school-goers-Channelling-public-funds-for-private-profit--135549.html).

Kumar, Ramya, Anne-Emanuelle Birn, and Peggy McDonough. 2016. “Agenda-Setting in Women’s Health: Critical Analysis of a Quarter Century of Paradigm Shifts in International and Global Health.” Pp 25-44 in Handbook on Gender and Health edited by J. Gideon. Cheltenham, UK and Northampton, MA: Edward Elgar.

Kumarasinghe, Uditha. 2003. “Depoliticisation of Samurdhi, a Dire Need.” Daily News, July 16. Retrieved June 25, 2017 (http://archives.dailynews.lk/2003/07/16/fea05.html).

Lakshman, W. D. 1985. "The IMF-World Bank Intervention in Sri Lankan Economic Policy: Historical Trends and Patterns." Social Scientist 13 (2):3-29.

Lall, Sanjaya and Senaka Bibile. 1977. “Political Economy of Controlling Transnationals: Pharmaceutical Industry in Sri Lanka, 1972-1976.” Economic and Political Weekly 12 (33/34):1419-1421+1423- 1429+1431-1433+1435-1436.

Laurell, Asa Cristina and Oliva López Arellano. 1996. “Market Commodities and Poor Relief: The World Bank Proposal for Health.” International Journal of Health Services 26(1):1-18.

LeBaron, Genevieve. 2010. "The Political Economy of the Household: Neoliberal Restructuring, Enclosures, and Daily Life." Review of International Political Economy 17(5): 889-912.

Leive, Adam, and Ke Xu. 2008. "Coping with Out-Of-Pocket Health Payments: Empirical Evidence from 15 African Countries." Bulletin of the World Health Organization 86 (11):849-56.

Levander, Caroline, and Walter Mignolo. 2011. "Introduction: the Global South and World Dis/Order." The Global South 5 (1):1-11.

Leys, Colin. 2010. “Health, Health Care and Capitalism.” Socialist Register 46:7-28.

Lindelow, Magnus, and Pieter Serneels. 2006. "The Performance of Health Workers in Ethiopia: Results from Qualitative Research." Social Science & Medicine 62 (9): 2225-35.

Liu, Gordon G., Zhongyun Zhao, Renhua Cai, Tetsuji Yamada and Tadashi Yamada. 2002. “Equity in Health Care Access to: Assessing the Urban Health Insurance Reform in China.” Social Science & Medicine 55:1779–94.

Loeppky, Rodney. 2010. “Certain Wealth: Accumulation in the Health Industry." Socialist Register 46:59- 83.

154

Loewenson, Rene. 1993. "Structural Adjustment and Health Policy in Africa." International Journal of Health Services 23 (4):717-30.

Long, Tim, and Martin Johnson. 2000. “Rigour, Reliability and Validity in Qualitative Research.” Clinical Effectiveness in Nursing 4 (1):30-7.

Lopez-Cevallos, Daniel F. and Chunhuei Chi. 2010. “Health Care Utilization in Ecuador: A Multilevel Analysis of Socio-Economic Determinants and Inequality issues.” Health Policy and Planning 25:209-18.

Ludowyk, E.F.C. 1966. The Modern History of Ceylon. London, Fakenham and Reading: Cox and Wyman Ltd.

Mackintosh, Maureen, Amos Channon, Anup Karan, Sakthivel Selvaraj, Eleonora Cavagnero, and Hongwen Zhao. 2016. "What is the Private Sector? Understanding Private Provision in the Health Systems of Low-Income and Middle-Income Countries." The Lancet 388 (10044):596-605.

Mahalingasivam, R. 1978. “Food Subsidy in Sri Lanka.” Sri Lanka Journal of Social Sciences 1 (1):75-94.

Makinen, M., H. Waters, M. Rauch, N. Almagambetova, R. Bitrán, L. Gilson, D. McIntyre, S. Pannarunothai, A.L. Prieto, G. Ubilla and S. Ram. 2000. "Inequalities in Health Care Use and Expenditures: Empirical Data from Eight Developing Countries and Countries in Transition." Bulletin of the World Health Organization 78 (1):55-65.

Malhotra, Anju and Mark Mather. 1997. “Do Schooling and Work Empower Women in Developing Countries? Gender and Domestic Decisions in Sri Lanka.” Sociological Forum 12 (4):599-630.

Mallawaarachchi, D. S. Virginie, Shiranee C. Wickremasinghe, Lakshmi C. Somatunga, Vithanage T. S. K. Siriwardena, Nalika S. Gunawardena. 2016. “Healthy Lifestyle Centres: A Service for Screening Noncommunicable Diseases through Primary Health-Care Institutions in Sri Lanka.” WHO South-East Asia Journal of Public Health 5 (2):89-95.

Mandel, Ernest. 1990. “Karl Marx.” Pp. 1-38 in Marxian Economics edited by J. Eatwell, M. Milgate and P. Newman. New York and London, United Kingdom: The Macmillan Press.

Mayo Foundation for Medical Education and Research. 2017. “Coronary Angioplasty and Stents.” Retrieved September 30, 2017 (http://www.mayoclinic.org/tests-procedures/coronary- angioplasty/home/ovc-20241582).

McGregor, Sue. 2001. "Neoliberalism and Health Care." International Journal of Consumer Studies 25 (2):82-9.

McIntyre, Di, Michael Thiede, and Stephen Birch. 2009. “Access as a Policy-Relevant Concept in Low- and Middle-Income Countries.” Health Economics, Policy and Law 4:179–93.

McIntyre, Di, Michael Thiede, Göran Dahlgren, and Margaret Whitehead. 2006. "What are the Economic Consequences for Households of Illness and of Paying for Health Care in Low- and Middle- Income Country Contexts?" Social Science & Medicine 62 (4):858-65.

McLane, Hannah C., Aaron L. Berkowitz, Bryan N. Patenaude, Erica D. McKenzie, Emma Wolper, Sarah Wahlster, Günther Fink, and Farrah J. Mateen. 2015. "Availability, Accessibility, and Affordability of Neurodiagnostic Tests in 37 Countries." Neurology 85 (18):1614-22.

McPake, Barbara, and Kara Hanson. 2016. "Managing the Public–Private Mix to Achieve Universal Health Coverage." The Lancet 388 (10044):622-30.

155

McPake, Barbara, Giuliano Russo, David Hipgrave, Krishna Hort, and James Campbell. 2016. "Implications of Dual Practice for Universal Health Coverage." Bulletin of the World Health Organization 94:142-146.

Measure Evaluation. n.d. “The Ratio of Household Out-Of-Pocket Payments for Healthcare to Household Income.” Retrieved June 18, 2017 (https://www.measureevaluation.org/prh/rh_indicators/crosscutting/hss/the-ratio-of-household- out-of-pocket-payments-for).

Meegama, S.A. 2012. Famine, Fevers and Fear: The State and Disease in British Colonial Sri Lanka. Dehiwela, Sri Lanka: Sridevi Publications.

Mehrotra, Santosh and Enrique Delamonica. 2005. “The Private Sector and Privatization in Social Services.” Global Social Policy 5(2):141-74.

Mendis, Shanti, Keiko Fukino, Alexandra Cameron, Richard Laing, Anthonio Filipe Jr, Oussama Khatib, Jerzy Leowski, and Margaret Ewen. 2007. "The Availability and Affordability of Selected Essential Medicines for Chronic Diseases in Six Low- and Middle-Income Countries." Bulletin of the World Health Organization 85 (4):279-88.

Meng, Qun, Ling Xu, Yaoguang Zhang, Juncheng Qian, Min Cai, Ying Xin, Jun Gao, Ke Xu, J. Ties Boerma, and Sarah L. Barber. 2012. "Trends in Access to Health Services and Financial Protection in China between 2003 and 2011: A Cross-Sectional Study." The Lancet 379 (9818): 805-14.

Mies, Maria. 1982. The Lace Makers of Narsapur: Indian Housewives Produce for the World Market. London, UK: Zed Press.

Mills, Anne, John E. Ataguba, James Akazili, Jo Borghi, Bertha Garshong, Suzan Makawia and Gemini Mtei. 2012. "Equity in Financing and Use of Health Care in Ghana, South Africa, and Tanzania: Implications for Paths to Universal Coverage." The Lancet 380 (9837):126-133.

Ministry of Finance. 2016. “Budget Speech 2017.” Retrieved June 17, 2017 (http://www.treasury.gov.lk/article/-/article-viewer-portlet/render/view/budget-speech-2017).

Ministry of Finance. 2017. Annual Report 2016: Ministry of Finance Sri Lanka. Retrieved September 30, 2017 (http://www.treasury.gov.lk/documents/10181/12870/2016/c36d6610-d6e7-4b1c-ab35- 238a4db56b88).

Ministry of Health. 2016. National Strategic Framework for Development of Health Services 2016 – 2025. Retrieved June 15, 2017 (http://www.health.gov.lk/enWeb/HMP2016- 2025/National%20%20Strategic%20%20Framework%20%20.pdf).

Ministry of Health. 2015. Human Resources Profile: Ministry of Health. Retrieved October 20, 2017 (http://www.health.gov.lk/moh_final/english/public/elfinder/files/publications/HRMprofile2016.pdf).

Ministry of Health. n.d. National Health Development Plan 2013-2017. Colombo, Sri Lanka: Ministry of Health.

Ministry of Health, Nutrition and Indigenous Medicine. 2016a. Annual Health Bulletin 2014. Retrieved June 15, 2017 (http://www.health.gov.lk/enWeb/publication/AHB2014/AHB2014.pdf).

Ministry of Health, Nutrition and Indigenous Medicine. 2016b. Sri Lanka National Health Accounts 2013. Retrieved October 22, 2017

156

(http://www.health.gov.lk/moh_final/english/public/elfinder/files/publications/NHA/Sri%20Lanka%2 0National%20Health%20Accounts%202013.pdf).

Ministry of Public Administration and Home Affairs. 2011a. “Kandy District Secretariat.” Retrieved June 25, 2017 (http://www.kandy.dist.gov.lk/).

Ministry of Public Administration and Home Affairs. 2011b. “Kandy Four Gravets Divisional Secretariat: Grama Niladhari Divisions.” Retrieved June 25, 2017 (http://www.kandy.ds.gov.lk/index.php?option=com_content&view=article&id=35&Itemid=71&lang =en).

Moghri, Javad, Arash Rashidian, Mohammad Arab, and Ali Akbari Sari. 2017. "Implications of Dual Practice among Health Workers: A Systematic Review." Iranian Journal of Public Health 46 (2):153-64.

Mohanty, Chandra Talpade. 2003. Feminism without Borders. Durham, NC and London, UK: Duke University Press.

Molyneux, Maxine. 1979. "Beyond the Domestic Labour Debate." New Left Review 116:3.

Moore, Mick. 2017. The Political Economy of Long-Term Revenue Decline in Sri Lanka. Brighton, United Kingdom: Institute for Development Studies.

Moramudali, Umesh and Shaahidah Riza. 2016. “Health Budget Slashed Rs 175 B in 2016 down to Rs 160 B in 2017.” Ceylon Today, November 27. Retrieved June 27, 2017 (http://www.ceylontoday.lk/print20161101CT20161231.php?id=10070).

Morgan, Rosemary, Tim Ensor, and Hugh Waters. 2016. "Performance of Private Sector Health Care: Implications for Universal Health Coverage." The Lancet 388 (10044):606-12.

Morse, Janice M., Michael Barrett, Maria Mayan, Karin Olson, and Jude Spiers. 2002. "Verification Strategies for Establishing Reliability and Validity in Qualitative Research." International Journal of Qualitative Methods 1 (2):13-22.

Mtei, Gemini, Suzan Makawia, Mariam Ally, August Kuwawenaruwa, Filip Meheus and Josephine Borghi. 2012. “Who Pays and Who Benefits from Health Care? An Assessment of Equity in Health Care Financing and Benefit Distribution in Tanzania.” Health Policy and Planning 27 (Suppl. 1):i23–i34.

Mudalige, Disna. 2012. “Paying Wards at Govt Hospitals.” Daily News, January 9. Retrieved June 18, 2017 (http://archives.dailynews.lk/2012/01/09/news01.asp).

Mudiyanse, Rasnayaka M. 2014. "Family Physicians without a Defined Target Population in Sri Lanka." Journal of General Practice 2:178. doi:10.4172/2329-9126.10001784.

Mudugamuwa, Maheesha. 2016. “AED Warns Country about Bitter Medicine Prescribed by IMF.” The Island, July 29. Retrieved June 27, 2017 (http://www.island.lk/index.php?page_cat=article- details&page=article-details&code_title=149451).

Mumtaz, Zubia, 2014. “Improving Maternal Health in Pakistan: Toward a Deeper Understanding of the Social Determinants of Poor Women’s Access to Maternal Health Services.” American Journal of Public Health 104(S1):S17-S24.

Murray, Susan F. 2016. “Commercialization in Maternity Care: Uncovering Trends in the Contemporary Health Care Economy.” Pp 309-26 Handbook on Gender and Health edited by J. Gideon. Cheltenham, UK and Northampton, MA: Edward Elgar.

157

Murray, Susan F., and Mary Ann Elston. 2005. "The Promotion of Private Health Insurance and its Implications for the Social Organisation of Healthcare: A Case Study of Private Sector Obstetric Practice in Chile." Sociology of Health & Illness 27 (6):701-21.

Narangoda, Himalika B. and A. Khathibi. 2014. “Public Private Partnership in Healthcare Industry in Sri Lanka as an Alternative to Privatization.” International Journal of Management and Business Research 4 (2):95-106.

National Insurance Trust Fund. n.d. “Agrahara Insurance.” Retrieved June 17, 2017 (http://www.nitf.lk/Agrahara.html).

Nazeer, Tasnim. 2017. “Sri Lanka: Buddhist Leader Stokes Anti-Muslim Tension.” Al Jazeera, May 27. Retrieved July 3, 2017 (http://www.aljazeera.com/news/2017/05/sri-lanka-buddhist-leader-stokes- anti-muslim-tension-170526211713093.html).

Newton, Paul N., Michael D. Green, Dallas C. Mildenhall, Aline Plançon, Henry Nettey, Leonard Nyadong, Dana M. Hostetler, Isabel Swamidoss, Glenn A Harris, Kristen Powell, Ans E Timmermans, Abdinasir A Amin, Stephen K Opuni, Serge Barbereau, Claude Faurant, Ray CW Soong, Kevin Faure, Jonarthan Thevanayagam, Peter Fernandes, Harparkash Kaur, Brian Angus, Kasia Stepniewska, Philippe J Guerin and Facundo M Fernández. 2011. "Poor Quality Vital Anti-Malarials in Africa – An Urgent Neglected Public Health Priority." Malaria Journal 10 (1):352. DOI: 10.1186/1475-2875-10-352.

Ngo, Anh D. and Peter S. Hill. 2011. "Quality of Reproductive Health Services at Commune Health Stations in Viet Nam: Implications for National Reproductive Health Care Strategy." Reproductive Health Matters 19 (37):52-61.

Nishtar, Sania. 2010. “The Mixed Health Systems Syndrome.” Bulletin of the World Health Organization 88:74–5. doi:10.2471/BLT.09.067868.

Official Government News Portal of Sri Lanka. 2017. “Children’s Cardiac Treatment Unit for Sirimavo Hospital in Peradeniya.” Retrieved June 25, 2017 (https://www.news.lk/news/business/item/9486- children-s-cardiac-treatment-unit-for-sirimavo-hospital-in-peradeniya).

Ogwal-Okeng, J. W., C. Obua, P. Waako, O. Aupont, and D. Ross-Degnan. 2004. "A Comparison of Prescribing Practices between Public and Private Sector Physicians in Uganda." East African Medical Journal :S12-6.

Oh, Juhwan, Jin-Seok Lee, Yong-Jun Choi, Hyeung-Keun Park, Young Kyung Do, and Sang-Jun Eun. 2011. "Struggle against Privatization: A Case History in the Use of Comparative Performance Evaluation of Public Hospitals." International Journal of Health Services 41 (2):371-388.

Oliver, Daniel G., Julianne M. Serovich, and Tina L. Mason. 2005. "Constraints and Opportunities with Interview Transcription: Towards Reflection in Qualitative Research." Social Forces 84 (2):1273- 89.

Onah, Michael N. and Veloshnee Govender. 2014. "Out-of-Pocket Payments, Health Care Access and Utilisation in South-Eastern Nigeria: A Gender Perspective." PLoS One 9 (4):e93887.

Onwujekwe, Obinna, Kara Hanson, and Benjamin Uzochukwu. 2012. "Examining Inequities in Incidence of Catastrophic Health Expenditures on Different Healthcare Services and Health Facilities in Nigeria." PLoS One 7 (7):e40811.https://doi.org/10.1371/journal.pone.0040811

Onwujekwe, Obinna, Harparkash Kaur, Nkem Dike, Elvis Shu, Benjamin Uzochukwu, Kara Hanson, Viola Okoye and Paul Okonkwo. 2009. "Quality of Anti-Malarial Drugs Provided by Public and Private

158

Healthcare Providers in South-East Nigeria." Malaria Journal 8 (1):22. DOI: 10.1186/1475-2875- 8-22.

Oxford Business Group. 2017. “Sri Lanka Seeks to Carve Out Niche in Medical Tourism.” Retrieved June 18, 2017 (http://www.oxfordbusinessgroup.com/news/sri-lanka-seeks-carve-out-niche-medical- tourism).

Oxford University Press. 2017. “Oxford Living Dictionaries.” Retrieved September 30, 2017 (https://en.oxforddictionaries.com/).

Oyerinde, Koyejo, Yvonne Harding, Philip Amara, Nana Garbrah-Aidoo, Rugiatu Kanu, Macoura Oulare, Rumishael Shoo and Kizito Daoh. 2012. "Barriers to Uptake of Emergency Obstetric and Newborn Care Services in Sierra Leone: A Qualitative Study." Journal of Community Medicine & Health Education 2:149. doi:10.4172/2161-0711.1000149.

Pallegedara, Asankha, and Michael Grimm. 2017. "Demand for Private Healthcare in a Universal Public Healthcare System: Empirical Evidence from Sri Lanka." Health Policy and Planning: czx085. https://doi.org/10.1093/heapol/czx085.

Paluzzi, Joan E. 2004. “Primary Health Care since Alma Ata: Lost in the Bretton Woods?” Pp. 63-77 in Unhealthy Health Policy: A Critical Anthropological Examination edited by A. Castro and M. Singer. Lanham, MD: AltaMira Press.

Panitch, Leo and Sam Gindin. 2012. The Making of Global Capitalism: The Political Economy of American Empire. London, UK and Brooklyn: Verso.

Paphassarang, Chanthakhath, Khampienne Philavong, Boungnong Boupha, and E. Blas. 2002. "Equity, Privatization and Cost Recovery in Urban Health Care: The Case of Lao PDR." Health Policy and Planning 17 (Suppl 1):72-84.

Parkhurst, Justin O., Syed Azizur Rahman and Freddie Sengooba. 2006. “Overcoming Access Barriers for Facility-Based Delivery in Low-Income Settings: Insights from Bangladesh and Uganda.” Journal of Health, Population and Nutrition 24 (4):438-45.

Patcharanarumol, Walaiporn, Viroj Tangcharoensathien, Supon Limwattananon, Warisa Panichkriangkrai, Kumaree Pachanee, Waraporn Poungkantha, Lucy Gilson and Anne Mills. 2011. “Why and How did Thailand Achieve Good Health at Low Cost?” Pp. 193-233 in ‘Good Health at Low Cost’ 25 Years On edited by D. Balabanova, M. McKee and A. Mills. London, United Kingdom: London School of Hygiene and Tropical Medicine. Retrieved June 13, 2017 (http://ghlc.lshtm.ac.uk/files/2011/10/GHLC-book.pdf).

Patel, Vikram, Rachana Parikh, Sunil Nandraj, Priya Balasubramaniam, Kavita Narayan, Vinod K. Paul, AK Shiva Kumar, Mirai Chatterjee, and K. Srinath Reddy. 2015. “Assuring Health Coverage for All in India.” The Lancet 386 (10011):2422-35.

Pathmanathan, Indra, Jerker Liljestrand, Jo M. Martins, Lalini C. Rajapaksa, Craig Lissner, Amala de Silva, Swarna Selvaraju and Prabha Joginder Singh. 2003. Investing in Maternal Health: Learning from Malaysia and Sri Lanka. Washington, DC: World Bank.

Patnaik, Prabhat. 1973. “On the Political Economy of Underdevelopment.” Economic and Political Weekly 8 (4/6):197-212.

Patnaik, Prabhat. 1994. "Notes on the Political Economy of Structural Adjustment." Social Scientist 22 (9/12):4-17.

159

Patton, Michael Quinn. 2015. Qualitative Research & Evaluation Methods Integrating Theory and Practice. 4th ed. Los Angeles, CA, London, United Kingdom, New Delhi, India, Washington, D.C. and Singapore: SAGE.

Peiris-John, R. J., S. Attanayake, L. Daskon, A. R. Wickremasinghe, and S. Ameratunga. 2014. "Disability Studies in Sri Lanka: Priorities for Action." Disability and rehabilitation 36 (20):1742-48.

Perera, Myrtle, Godfrey Gunatilleke, and Philippa Bird. 2007. "Falling into the Medical Poverty Trap in Sri Lanka: What Can Be Done?" International Journal of Health Services 37(2):379-98.

Perera, P. D. A. 1985. “Health Care Systems of Sri Lanka.” Pp. 93-110 in Good Health at Low Cost edited by Scott B. Halstead, Julia A. Walsh and Kenneth S. Warren. New York: The Rockefeller Foundation.

Perkins, Margaret, Ellen Brazier, Ellen Themmen, Brahima Bassane, Djeneba Diallo, Angeline Mutunga, Tuntufye Mwakajonga and Olipa Ngobola. 2009. “Out-of-Pocket Costs for Facility-Based Maternity Care in Three African Countries.” Health Policy and Planning 24 (4):289-300.

Pfeiffer, James. 2004. “International NGOs in the Mozambique Health Sector: The “Velvet Glove” of Privatization.” Pp. 43-62 in Unhealthy Health Policy: A Critical Anthropological Examination edited by A. Castro and M. Singer. Lanham, MD: AltaMira Press.

Pfeiffer, James, and Rachel Chapman. 2010. "Anthropological Perspectives on Structural Adjustment and Public Health." Annual Review of Anthropology 39:149-65.

Plaza, Beatriz, Ana Beatriz Barona, and Norman Hearst. 2001. "Managed Competition for the Poor or Poorly Managed Competition? Lessons from the Colombian Health Reform Experience." Health Policy and Planning 16 (suppl 2):44-51.

Pongsupap, Yongyuth, and Wim Van Lerberghe. 2006. "Choosing between Public and Private or between Hospital and Primary Care: Responsiveness, Patient‐Centredness and Prescribing Patterns in Outpatient Consultations in Bangkok." Tropical Medicine & International Health 11 (1):81-9.

Potter, Joseph E., Elza Berquó, Ignez H.O. Perpétuo, Ondina Fachel Leal, Kristine Hopkins, Marta Rovery Souza, and Maria Célia de Carvalho Formiga. 2001. "Unwanted Caesarean Sections among Public and Private Patients in Brazil: Prospective Study." BMJ 323 (7322):1155-8.

Prashad, Vijay. 2012. "Dream History of the Global South." Interface: A Journal for and about Social Movements 4:43-53.

Provincial Director of Health Services – Central Province. n.d. “Annual Health Bulletin 2014.” Retrieved June 27, 2017 (http://www.healthcpc.org/downloads/annual_health_bulletin_2014.pdf).

Qadeer, Imrana. 1994. "The World Development Report 1993: The Brave New World of Primary Health Care." Social Scientist 22 (9/12): 27-39.

Qadeer, Imrana, and Rama Baru. 2016. "Shrinking Spaces for the ‘Public’ in Contemporary Public Health." Development and Change 47 (4):760-81.

Qadeer, Imrana, and Sunita Reddy. 2006. "Medical Care in the Shadow of Public Private Partnership." Social Scientist 34(9/10):4-20.

Qadeer, Imrana and Nalini Visvanathan. 2004. “How Healthy are Health and Population Policies? The Indian Experience.” Pp. 145-62 in Unhealthy Health Policy: A Critical Anthropological Examination edited by A. Castro and M. Singer. Lanham, MD: AltaMira Press.

160

Quayyum, Zahidul, Mardiati Nadjib, Tim Ensor and Purwa Kurnia Sucahya. 2009. “Expenditure on Obstetric Care and the Protective Effect of Insurance on the Poor: Lessons from Two Indonesian Districts.” Health Policy and Planning 25 (3):237–47.

Quist, Cornelie. 2015. “Widows’ Struggles in Post-War Sri Lanka.” International Oceanographic Data and Information Exchange. Oostende, Belgium: Aquatic Commons. Retrieved June 15, 2017 (http://aquaticcommons.org/19652/1/Widows%20struggles%20in%20post- war%20Sri%20Lanka_%20Yemaya%2050.pdf).

Radice, Hugo. 2005. “Neoliberal Globalization: Imperialism with Empires?” Pp 91-8 in Neoliberalism: A Critical Reader edited by A. Saad-Filho and D. Johnston. London, England and Ann Arbor, MI: Pluto Press.

Rannan-Eliya, Ravi P. and Nishan de Mel. 1997. “Resource Mobilization in Sri Lanka’s Health Sector.” Colombo, Sri Lanka: Research International (Pvt.) Ltd.

Rannan-Eliya, Ravi P. and Ajantha Kalyanaratne. 2005. Fiscal Incentives for the Development of Health Services in Sri Lanka: Evaluation of BOI Incentives for Private Health Sector Investments. Colombo, Sri Lanka: Institute for Health Policy.

Rannan-Eliya, Ravi P. and Lankani Sikurajapathy. 2009. Sri Lanka: “Good Practice” in Expanding Health Care Coverage. Colombo, Sri Lanka: Institute for Health Policy.

Rannan-Eliya, Ravindra P., Nilmini Wijemanne, Isuru K. Liyanage, Janaki Jayanthan, Shanthi Dalpadatu, Sarasi Amarasinghe and Chamara Anuranga. 2015a. “The quality of outpatient primary care in public and private sectors in Sri Lanka—how well do patient perceptions match reality and what are the implications?” Health Policy and Planning 30 (Suppl 1):i59-i74.

Rannan-Eliya, Ravindra P., Nilmini Wijemanne, Isurujith K. Liyanage, Shanti Dalpadatu, Sanil de Alwis, Sarasi Amarasinghe and Shivanthan Shanthikumar. 2015b. “Quality of Inpatient Care in Public and Private Hospitals in Sri Lanka.” Health Policy and Planning 30 (Supp. 1):i46-i58.

Rashid, Sabina Faiz, Owasim Akram, and Hilary Standing. 2011. "The Sexual and Reproductive Health Care Market in Bangladesh: Where do Poor Women Go?" Reproductive Health Matters 19 (37):21-31.

Reich, Michael R., Joseph Harris, Naoki Ikegami, Akiko Maeda, Keizo Takemi and Timothy G. Evans. 2015. “Moving Towards Universal Health Coverage: Lessons from 11 Country Studies.” The Lancet 387 (10020):811-16.

Reinharz, Shulamit. 1997. “Who Am I? The Need for a Variety of Selves in the Field.” Pp 3-20 in Reflexivity and Voice edited by R. Hertz. Thousand Oaks, CA: Sage.

Roberts, Marc, William Hsiao, Peter Berman, and Michael Reich. 2008. Getting Health Reform Right: A Guide to Improving Performance and Equity. New York: Oxford University Press, Inc.

Rolfe, Gary. 2006. "Validity, Trustworthiness and Rigour: Quality and the Idea of Qualitative Research." Journal of Advanced Nursing 53 (3):304-10.

Rosenfield, P. L. 1985. "The Contribution of Social and Political Factors to Good Health." Pp. 173-80 in Good Health at Low Cost edited by S. B. Halstead, J. A. Walsh and K. S. Warren. New York: The Rockefeller Foundation.

Ross, Robert J.S. and Kent C. Trachte. 1990. Global Capitalism: The New Leviathan. Albany, NY: State University of New York Press.

161

Roulston, Kathryn. 2010. Reflective Interviewing: A Guide to Theory & Practice. Los Angeles, CA, London, UK, New Delhi, India, Washington, D.C. and Singapore: SAGE.

Ruiz, Fernando, Liliana Amaya and Stella Venegas. 2007. “Progressive Segmented Health Insurance: Colombian Health Reform and Access to Health Services.” Health Economics 16:3-18.

Rupasinghe, P. 2015. “Medical Negligence and Doctor’s Liability; A Critical Review in Present Legal Regime in Sri Lanka.” Kotelawala Defence University: Ratmalana, Sri Lanka. Retrieved June 18, 2017 (http://www.kdu.ac.lk/proceedings/irc2015/2015/law-043.pdf).

Russell, Steven. 2005. "Treatment-Seeking Behaviour in Urban Sri Lanka: Trusting the State, Trusting Private Providers." Social Science & Medicine 61 (7): 1396-407.

Russell, Steven, and Lucy Gilson. 2006. "Are Health Services Protecting the Livelihoods of the Urban Poor in Sri Lanka? Findings from Two Low-Income Areas of Colombo." Social Science & Medicine 63 (7): 1732-44.

Rutnam, Easwaran. 2016. “A Push to Decriminalize Homosexuality.” The Sunday Leader, September 5. Retrieved June 27, 2017 (http://www.thesundayleader.lk/2016/09/05/a-push-to-decriminalize- homosexuality/).

Saad-Filho, Alfredo. 2005. “From Washington to Post-Washington Consensus: Neoliberal Agendas for Economic Development.” Pp. 113-19 in Neoliberalism: A Critical Reader edited by A. Saad-Filho and D. Johnston. London, England and Ann Arbor, MI: Pluto Press.

Saksena, Priyanka, Ke Xu, Riku Elovainio and Jean Perrot. 2012. “Utilization and Expenditure at Public and Private Facilities in 39 Low-Income Countries.” Tropical Medicine and International Health 17 (1): 23-35.

Samurdhi Authority. 2016. “Overview.” Retrieved June 25, 2017 (http://www.samurdhi.gov.lk/web/index.php?option=com_content&view=article&id=83&Itemid=91 &lang=en).

Sassen, Saskia. 2000. "Women's Burden: Counter-Geographies of Globalization and the Feminization of Survival." Journal of International Affairs 53 (2):503-524.

Schneider, H., D. Blaauw, E. Dartnall, D. J. Coetzee, and R. C. Ballard. 2001. "STD Care in the South African Private Health Sector." South African Medical Journal 91 (2):151-56.

Schrecker, Ted, and Clare Bambra. 2015. How Politics Makes Us Sick: Neoliberal Epidemics. Hampshire, United Kingdom and New York: Palgrave Macmillan.

Schwandt, Thomas A., Yvonna S. Lincoln, and Egon G. Guba. 2007. "Judging Interpretations: But is it Rigorous? Trustworthiness and Authenticity in Naturalistic Evaluation." New Directions for Evaluation 2007 (114):11-25.

Senanayake, Sunil, Buddhika Senanayake, Thushara Ranasinghe and Neelamani S.R. Hewageegana. 2017. “How to Strengthen Primary Health Care Services in Sri Lanka to Meet the Future Challenges.” Journal of the College of Community Physicians of Sri Lanka 23 (1):43-49.

Senarath, Upul, Dulitha N. Fernando, and Ishani Rodrigo. 2006. "Factors Determining Client Satisfaction with Hospital‐Based Perinatal Care in Sri Lanka." Tropical Medicine & International Health 11 (9):1442-51.

162

Sengupta, Amit. 2015. “Universal Health Coverage: The Rhetoric and the Substance.” Retrieved February 21, 2017 (https://www.twn.my/title2/resurgence/2015/296-297/cover01.htm).

Shahrawat, Renu and Krishna D. Rao. 2012. “Insured Yet Vulnerable: Out-of-Pocket Payments and India’s Poor.” Health Policy and Planning 27:213-21.

Shaikh, Babar T. and Juanita Hatcher. 2005. “Health Seeking Behaviour and Health Service Utilization in Pakistan.” Journal of Public Health 27 (1):49-54.

Shanthiham. 2017. “Home.” Retrieved October 19, 2017 (http://shanthiham.lk/).

Shaokang, Zhan, Sun Zhenwei, and Erik Blas. 2002. "Economic Transition and Maternal Health Care for Internal Migrants in Shanghai, China." Health Policy and Planning 17 (suppl 1):47-55.

Shayo, Elizabeth H., Kesheni P. Senkoro, Romanus Momburi, Øystein E. Olsen, Jens Byskov, Emmanuel A. Makundi, Peter Kamuzora, and Leonard EG Mboera. 2016. "Access and Utilisation of Healthcare Services in Rural Tanzania: A Comparison of Public and non-Public Facilities using Quality, Equity, and Trust Dimensions." Global Public Health 11 (4): 407-22.

Siddiqi, S., S. Hamid, G. Rafique, S. A. Chaudhry, N. Ali, S. Shahab and R. Sauerborn. 2002. "Prescription Practices of Public and Private Health Care Providers in Attock District of Pakistan." The International Journal of Health Planning and Management 17 (1):23-40.

Silva, Kalinga Tudor. 2014. Decolonisation, Development and Disease: A Social History of Malaria in Sri Lanka. New Delhi, India: Orient BlackSwan.

Sirisena, Maithripala. 2014. “A Compassionate Maithri Governance: A Stable Country.” Colombo, Sri Lanka: Maithripala Sirisena.

SJGH. 2015. “Vision and Mission.” Retrieved June 27, 2017 (http://www.sjghsrilanka.com/vision- mission/#1450908030391-e03d12b6-0be5).

Skanthakumar, Balasingham. 2013. “Crisis, Vulnerability and Poverty in South Asia People’s Struggles for Justice and Dignity Country Report 2013 Sri Lanka.” Kathmandu, Nepal: South Asia Alliance for Poverty Eradication.

Somkotra, Tewarit, and Leizel P. Lagrada. 2009. "Which Households are at Risk of Catastrophic Health Spending: Experience in Thailand after Universal Coverage." Health affairs 28 (3):w467-w478.

Spivak, Gayatri Chakravorty. 1988. “Can the Subaltern Speak?” Pp. 271-313 in Marxism and Interpretation of Culture edited by C. Nelson and L. Grossberg. Chicago, IL: University of Illinois Press.

SLMC. 2007-2008. “About Us.” Retrieved August 18, 2017 (http://www.srilankamedicalcouncil.org/aboutus.php).

State Pharmaceuticals Corporation of Sri Lanka. 2015. “About SPC.” Retrieved August 16, 2017 (http://www.spc.lk/about-spc.php).

Stinson, Jane. 2004. "Why Privatization is a Women's Issue." Canadian Woman Studies 23 (3/4):18-22.

Stinson, Jane, Nancy Pollak, and Marcy Cohen. 2005. The Pains of Privatization: How Contracting Out Hurts Health Support Workers, their Families, and Health Care. Vancouver, B.C.: Canadian Centre for Policy Alternatives. Retrieved February 17, 2016 (http://www.nnewh.org/images/upload/attach/8593pains_privatization.pdf).

163

Storeng, Katerini Tagmatarchi, Rebecca F. Baggaley, Rasmane Ganaba, Fatoumata Ouattara, Melanie S. Akoum, and Veronique Filippi. 2008. "Paying the Price: The Cost and Consequences of Emergency Obstetric Care in Burkina Faso." Social Science & Medicine 66 (3):545-57.

Stuckler, David, and Sanjay Basu. 2009. "The International Monetary Fund's Effects on Global Health: Before and After the 2008 Financial Crisis." International Journal of Health Services 39 (4):771- 81.

Sudhinaraset, May, Matthew Ingram, Heather Kinlaw Lofthouse and Dominic Montagu. 2013. “What Is the Role of Informal Healthcare Providers in Developing Countries? A Systematic Review.” PLoS ONE 8 (2):e54978. https://doi.org/10.1371/journal.pone.0054978.

Sustainable Development Knowledge Platform. 2016. “Goal 3: Ensure Healthy Lives and Promote Well- Being for All Ages at All Ages.” Retrieved February 16, 2016 (https://sustainabledevelopment.un.org/sdg3).

Suwasevana Hospitals Pvt. Ltd. 2015. “About Us.” Retrieved June 27, 2017 (http://suwasevana.lk/aboutus/).

Teaching Hospital – Kandy. 2012-2017. “Welcome to Teaching Hospital Kandy, Sri Lanka.” Retrieved June 25, 2017 (http://www.kandy-hospital.health.gov.lk/).

Teaching Hospital Peradeniya. n.d. “Services.” Retrieved June 27, 2017 (http://www.peradeniya- hospital.health.gov.lk/?page_id=72).

Temple, Bogusia, and Alys Young. 2004. "Qualitative Research and Translation Dilemmas." Qualitative Research 4 (2):161-78.

The Island. 2017. “Health Ministry Focuses on Combating Unhealthy Life-Styles.” The Island, February 12. Retrieved June 27, 2017 (http://www.island.lk/index.php?page_cat=article- details&page=article-details&code_title=160218).

The Lancet. 2016. “Series from Lancet Journals: Universal Health Coverage: Markets, Profit, and the Public Good.” Retrieved June 24, 2017 (http://www.thelancet.com/series/private-sector-health).

Thiruvarangan, Mahendran. 2017. “Trying to Understand Keppapilavu: Resistance, Solidarities & Politics.” Colombo Telegraph, February 28. Retrieved June 15, 2017 (https://www.colombotelegraph.com/index.php/trying-to-understand-keppapilavu-resistance- solidarities-politics/).

Thomas, Eileen, and Joan Kathy Magilvy. 2011. "Qualitative Rigor or Research Validity in Qualitative Research." Journal for Specialists in Pediatric Nursing 16 (2):151-55.

Thomas, Kris. 2016. “Sri Lanka: The Kidney Transplant Hub?” Roar Reports, January 29. Retrieved August 18, 2017 (https://roar.media/english/reports/reports/sri-lanka-kidney-transplant-hub/).

Thresia, C.U. 2013. “Rising Private Sector and Falling ‘Good Health at Low Cost’: Health Challenges in China, Sri Lanka, and Indian State of Kerala.” International Journal of Health Services 43 (1):31- 48.

Thuan, Nguyen, Curt Lofgren, Lars Lindholm and Nguyen Thi Kim Chuc. 2008. “Choice of Healthcare Provider following Reform in Vietnam.” BMC Health Services Research 8 (1):162 doi:10.1186/1472-6963-8-162.

164

Tilley, Susan A. 2003. "“Challenging” Research practices: Turning a Critical Lens on the Work of Transcription." Qualitative Inquiry 9 (5):750-73.

Tracy, Sarah J. 2010. "Qualitative Quality: Eight “Big-Tent” Criteria for Excellent Qualitative Research." Qualitative Inquiry 16 (10):837-51.

Transparency International and Friedrich Ebert Stiftung. 2009. Integrity in Government Hospitals in the Colombo District. Retrieved June 18, 2017 (http://www.tisrilanka.org/pub/pp/pdf/Abstract%20colombo%20hospital%20study.pdf).

TripAdvisor. 2017. “Spas & Wellness Centers in Kandy.” Retrieved June 27, 2017 (https://www.tripadvisor.com/Attractions-g304138-Activities-c40- Kandy_Kandy_District_Central_Province.html).

Triunfo, Patricia, and Máximo Rossi. 2009. "The Effect of Physicians’ Remuneration System on the Caesarean Section Rate: The Uruguayan Case." International Journal of Health Care Finance and Economics 9 (4):333-45.

Tuan, Tran, Van Thi Mai Dung, Ingo Neu, and Michael J. Dibley. 2005. "Comparative Quality of Private and Public Health Services in Rural Vietnam." Health Policy and Planning 20 (5):319-27.

Turan, Janet Molzan, Ayşen Bulut, Hacer Nalbant, Nuriye Ortayli, and A. Akalin. 2006. "The Quality of Hospital‐based Antenatal Care in Istanbul." Studies in Family Planning 37 (1): 49-60.

Turshen, Meredeth. 1999. Privatizing Health Services in Africa. New Brunswick, NJ, and London, United Kingdom: Rutgers University Press.

Unger, Jean-Pierre, Pierre De Paepe, P., Giorgio Solimano Cantuarias and Oscar Arteaga Herrera 2008. “Chile’s Neoliberal Health Reform: An Assessment and a Critique.” PLoS Medicine 5 (4): e79. doi:10.1371/journal.pmed.0050079.

Uragoda, C.G. 1987. A History of Medicine in Sri Lanka. Colombo, Sri Lanka: Sri Lanka Medical Association.

Van Doorslaer, Eddy, Owen O’Donnell, Ravi P Rannan-Eliya, Aparnaa Somanathan, Shiva Raj Adhikari, Charu C Garg, Deni Harbianto, Alejandro N Herrin, Mohammed Nazmul Huq, Shamsia Ibragimova, Anup Karan, Chiu Wan Ng, Badri Raj Pande, Rachel Racelis, Sihai Tao, Keith Tin, Kanjana Tisayaticom, Laksono Trisnantoro, Chitpranee Vasavid, Yuxin Zhao. 2006. “Effect of Payments for Health Care on Poverty Estimates in 11 Countries in Asia: An Analysis of Household Survey Data.” The Lancet 368 (9544):1357-64.

Van Nes, Fenna, Tineke Abma, Hans Jonsson, and Dorly Deeg. 2010. "Language Differences in Qualitative Research: Is Meaning Lost in Translation?" European Journal of Ageing 7 (4):313-16.

Varman, Rohit, and Ram Manohar Vikas. 2007. "Rising Markets and Failing Health: An Inquiry into Subaltern Health Care Consumption under Neoliberalism." Journal of Macromarketing 27 (2):162- 72.

Veltemeyer, Henry and James Petras. 2005. “Foreign Aid, Neoliberalism and US Imperialism.” Pp. 120-6 in Neoliberalism: A Critical Reader edited by A. Saad-Filho and D. Johnston. London, England and Ann Arbor, MI: Pluto Press.

Victora, C.G., A. Matijasevich, M.F. Silveira, I.S. Santos, A.J.D. Barros and FC Barros. 2010. “Socio- Economic and Ethnic Group Inequities in Antenatal Care Quality in the Public and Private sector in Brazil.” Health Policy and Planning 25 (4):253–61.

165

Villar, José, Eliette Valladares, Daniel Wojdyla, Nelly Zavaleta, Guillermo Carroli, Alejandro Velazco, Archana Shah, Liana Campodónico, Vicente Bataglia, Anibal Faundes, Ana Langer, Alberto Narváez, Allan Donner, Mariana Romero, Sofia Reynoso, Karla Simônia de Pádua, Daniel Giordano, Marius Kublickas and Arnaldo Acosta. 2006. "Caesarean Delivery Rates and Pregnancy Outcomes: The 2005 WHO Global Survey on Maternal and Perinatal Health in Latin America." The Lancet 367 (9525):1819-29.

Wagner, Anita K., Amy Johnson Graves, Sheila K. Reiss, Robert Le Cates, Fang Zhang, and Dennis Ross-Degnan. 2011. "Access to Care and Medicines, Burden of Health Care Expenditures, and Risk Protection: Results from the World Health Survey." Health Policy 100 (2):151-58.

Waitzkin, Howard. 2011. Medicine and Public Health at the End of Empire. Boulder, CO: Paradigm Publishers.

Waitzkin, Howard. 2015. "Universal Health Coverage: The Strange Romance of The Lancet, MEDICC, and Cuba." Social Medicine 9 (2):93-97.

Waitzkin, Howard, Rebecca Jasso-Aguilar and Celia Iriart. 2007. “Privatization of Health Services in Less Developed Countries: An Empirical Response to the Proposals of the World Bank and Wharton School.” International Journal of Health Services 37 (2):205-27.

Walton, John and David Seddon. 1994. Free Markets and Food Riots: The Politics of Global Adjustment. Cambridge, MA and Oxford, United Kingdom: Blackwell Publishers.

Wang, Wenjuan, Sara Sulzbach, and Susna De. 2011. "Utilization of HIV-related Services from the Private Health Sector: A Multi-Country Analysis." Social Science & Medicine 72 (2):216-23.

Weerasinghe, Manuj C. and Dulitha N. Fernando. 2011. “Paradox in Treatment Seeking: An Experience from Rural Sri Lanka.” Qualitative Health Research 21 (3):365–72.

Weintraub, Jeff. 1997. “The Theory and Politics of the Public/Private Distinction.” Pp. 1-42 in Public and Private in Thought and Practice: Perspectives on a Grand Dichotomy edited by Jeff Weintraub and Krishan Kumar. Chicago, IL and London, UK: The University of Chicago Press.

WHO. 1978. “The Declaration of Alma Ata.” Retrieved June 17, 2017 (http://www.who.int/publications/almaata_declaration_en.pdf?ua=1).

WHO. 2000. Health Systems: Improving Performance. Retrieved October 20, 2017 (http://www.who.int/whr/2000/en/whr00_en.pdf).

WHO. 2001. Macroeconomics and Health: Investing in Health for Economic Development. Retrieved November 16, 2016 (http://apps.who.int/iris/bitstream/10665/42435/1/924154550X.pdf).

WHO. 2010. Health Systems Financing: The Path to Universal Coverage. Retrieved August 22, 2014 (http://www.who.int/whr/2010/en/).

WHO. 2012. World Health Statistics 2012. Retrieved June 17, 2017 (http://apps.who.int/iris/bitstream/10665/44844/1/9789241564441_eng.pdf).

WHO. 2014. “Sri Lanka’s Low-Cost People-Centred Approach to Health Challenges.” Retrieved October 21, 2017 (http://www.who.int/features/2014/sri-lanka-health-challenges/en/).

WHO. 2015. World Health Statistics 2015. Retrieved February 16, 2016 (http://apps.who.int/iris/bitstream/10665/170250/1/9789240694439_eng.pdf?ua=1&ua=1).

166

WHO. 2017a. “Health Systems: Equity.” Retrieved September 30, 2017 (http://www.who.int/healthsystems/topics/equity/en/).

WHO. 2017b. World Health Statistics 2017: Monitoring Health for the SDGs, Sustainable Development Goals.” Geneva, Switzerland: World Health Organization. Retrieved June 27, 2017 (http://apps.who.int/iris/bitstream/10665/255336/1/9789241565486-eng.pdf).

WHO. 2017c. “Global Health Workforce Alliance: Moolai Hospital.” Retrieved October 19, 2017 (http://www.who.int/workforcealliance/members_partners/member_list/moolai/en/).

WHO. 2017d. “Countries: Sri Lanka.” Retrieved June 27, 2017 (http://www.who.int/countries/lka/en/).

WHO SEARO. 2016. “WHO Certifies Sri Lanka Malaria-Free.” Retrieved November 16, 2016 (http://www.searo.who.int/mediacentre/releases/2016/1631/en/).

Wickramasinghe, Nira. 2006. Sri Lanka in the Modern Age: A History of Contested Identities. London, UK: Hurst and Company.

WIEGO. 2017. “About the Informal Economy.” Retrieved June 25, 2017 (http://www.wiego.org/informal- economy/about-informal-economy).

Wijesinghe, Pushpa R., Ravindra L. Jayakody and Rohini D.A. Seneviratne. 2012. “Prevalence and Predictors of Self-Medication in a Selected Urban and Rural District of Sri Lanka.” WHO South- East Asia Journal of Public Health 1 (1):28-41.

Wikipedia. 2017. “Central Province: Sri Lanka.” Retrieved August 16, 2017 (https://en.wikipedia.org/wiki/Central_Province,_Sri_Lanka).

Wilkinson, Sue. 1998. “Focus Groups in Feminist Research: Power, Interaction, and the Co-Construction of Meaning.” Women’s Studies International Forum 21 (1):111-25.

Wipulasena, Aanya. 2013. “Rathupaswala Water Crisis and New Moves to Dam it.” The Sunday Times, October 20. Retrieved August 16, 2016 (http://www.sundaytimes.lk/131020/news/rathupaswala- water-crisis-and-new-moves-to-dam-it-66321.html).

Wong, Josephine Pui-Hing, and Maurice Kwong-Lai Poon. 2010. "Bringing Translation Out of the Shadows: Translation as an Issue of Methodological Significance in Cross-Cultural Qualitative Research." Journal of Transcultural Nursing 21 (2):151-58.

World Bank. 1987. Financing Health Services in Developing Countries: An Agenda for Reform. Washington D.C.: The World Bank.

World Bank. 1993. World Development Report 1993: Investing in Health. Washington D.C.: The World Bank.

World Bank. 2004. Project appraisal document on a proposed grant in the amount of SDR 40.2 million (US$ 60 million equivalent) to the Democratic Social Republic of Sri Lanka for a health sector development project. Retrieved June 17, 2017 (http://www- wds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2004/05/26/000160016_20040 526102053/Rendered/PDF/28915.pdf).

World Bank. 2011. Implementation completion and results report on a grant in the amount of SDR 40.2 million (US$ 60 million equivalent) and additional credit in the amount of SDR 16.3 million (US$ 24 million equivalent) to the Democratic Socialist Republic of Sri Lanka for a health sector development project. Retrieved August 22, 2014 (http://www-

167

wds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2011/09/09/000356161_20110 909010224/Rendered/PDF/ICR18420P050740000public00BOX361537B.pdf).

World Bank. 2013. International Development Association project appraisal document on a proposed credit in the amount of SDR 129.8 million (US$ 200 million equivalent) to the Democratic Socialist Republic of Sri Lanka for a second health sector development project. Retrieved August 22, 2014 (http://documents.worldbank.org/curated/en/2013/02/17425556/sri-lanka-second-health-sector- development-project).

World Bank. 2017a. “Sri Lanka – Second Health Sector Development Project.” Retrieved June 15, 2017 (http://projects.worldbank.org/P118806/second-health-sector-development-project?lang=en).

World Bank. 2017b. “Out-Of-Pocket Health Expenditure (% of Total Expenditure on Health).” Retrieved June 18, 2017 (http://data.worldbank.org/indicator/SH.XPD.OOPC.TO.ZS).

World Bank. 2017c. “Health expenditure, public (% of GDP).” Retrieved September 30, 2017 (https://data.worldbank.org/indicator/SH.XPD.PUBL.ZS).

World Bank. n.d. “World Bank Country and Lending Groups.” Retrieved June 18, 2017 (https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bank-country-and- lending-groups).

World Economic Forum. 2015. The Global Gender Gap Report. Retrieved June 15, 2017 (http://www3.weforum.org/docs/GGGR2015/cover.pdf).

World Health Assembly. 2005. “WHA58.33 Sustainable Health Financing, Universal Coverage and Social Health Insurance.” Retrieved August 17, 2017 (http://www.who.int/health_financing/HF%20Resolution%20en.pdf).

Wright, Erik Olin. 2009. "Understanding Class: Towards an Integrated Analytical Approach." New Left Review 60:101-16.

Xiong, Juyang, David Hipgrave, Karoline Myklebust, Sufang Guo, Robert W. Scherpbier, Xuetao Tong, Lan Yao and Andrew E. Moran. 2013. “Child Health Security in China: A Survey of Child Health Insurance Coverage in Diverse Areas of the Country.” Social Science & Medicine 97:15-19.

Xu, H. and Short S.E. 2011. “Health Insurance Coverage Rates in 9 Provinces in China Doubled from 1997 to 2006, with a Dramatic Rural Upswing.” Health Affairs 30 (12):2419-26.

Xu, Ke, David B. Evans, Patrick Kadama, Juliet Nabyonga, Peter Ogwang Ogwal, Pamela Nabukhonzo, and Ana Mylena Aguilar. 2006. "Understanding the Impact of Eliminating User Fees: Utilization and Catastrophic Health Expenditures in Uganda." Social Science & Medicine 62 (4):866-76.

Xu, Ke, David B. Evans, Kei Kawabata, Riadh Zeramdini, Jan Klavus, and Christopher JL Murray. 2003. "Household Catastrophic Health Expenditure: A Multicounty Analysis." The Lancet 362 (9378):111-17.

Yoong, Joanne, Nicholas Burger, Connor Spreng, and Neeraj Sood. 2010. "Private Sector Participation and Health System Performance in Sub-Saharan Africa." PLoS One 5 (10):e13243. doi.org/10.1371/journal.pone.0013243.

Zere, Eyob, Matshidiso Moeti, Joses Kirigia, Takondwa Mwase and Edward Kataika. 2007. “Equity in Health and Healthcare in Malawi: Analysis of Trends.” BMC Public Health 7:78 doi:10.1186/1471- 2458-7-78.

168

Appendices

APPENDIX A

Health administrators and healthcare providers met with in Phase 1

Category (number) Sector Place of work

Public Health Midwives (2) Public Kirideniya Maternal and Child Health Clinic

Medical Officer/Maternal and Public Kirideniya Maternal and Child Health Clinic Child Heath (1) and others in Kandy Municipal Council Area

Chief Medical Officer/Kandy Public Kandy Municipal Council Municipal Council (1)

Medical Officer of Public Medical Officer of Health Office Health/Gangawatakorale Gangawatakorale

Medical Officer/Planning (1) Public Office of the Regional Director of Health Services (RDHS) - Kandy

Medical Officer/Planning (1) Public Office of the Provincial Director of Health Services (PDHS) - Kandy

169

APPENDIX B

Script to invite community informants to participate in the study

Hello, As you may remember, I’m doing my PhD in Public Health at the University of Toronto in Canada. I’m studying women’s experiences of health care at a time when we seem to be relying more on more on services in private clinics and hospitals. I’d like to better understand how women make decisions about healthcare for themselves and their families and what they think about the public and private health services available to them I am grateful for the assistance you gave me with this project last year. This year, I’ll be conducting discussions and interviews with women living in the neighbourhoods and villages of xxxxx Grama Niladhari Division. Because of your familiarity with your neighbourhood, I think you could contribute to this study by:

1. Introducing me to or suggesting women you think might be interested in talking to me;

2. Taking part in a discussion with other women from the Division on what public and private healthcare means to all of us and our respective communities;

3. Participating in an interview, if necessary. You will not benefit directly by participating in this study. However, your assistance will help me complete my research, which may guide future health care policies in Sri Lanka. This study will also be a way for us to come together as a community to discuss the problems we face in accessing health services. Would you consider participating in this study? Participating in this study is completely voluntary and you may withdraw at any time.

170

APPENDIX C

Information letter and consent form for community informants

What is this research about? As you may remember, last year I did an exploratory study to define my PhD research on health care privatization. Your knowledge about your neighbourhood and the health services available in your area, and your understanding of the problems people face when they need health care, helped me a lot. Now I’m doing the second phase of this research. I’m trying to understand the growing use of private health care from the perspective of women living in the xxxxx Grama Niladhari Division. I am especially interested in understanding: 1) How women make decisions and what they do to access healthcare in government or private clinics and hospitals; 2) What women feel about the availability, accessibility and affordability of health services at government or private clinics and hospitals; 3) What women feel about their interactions with healthcare providers and the standard of care in government or private clinics and hospitals; and 4) How being poor or being Sinhala, Tamil or Muslim may affect experiences of accessing and receiving healthcare at government or private clinics and hospitals.

Who is participating in this study? Women who are at least 18 years of age and who have lived in the Division for more than two years. Your participation is completely voluntary. You do not have to take part in the study if you do not want to. Before you agree to participate, though, it is very important for you to understand the information on this page. If there is anything that is not clear to you, please let me know.

What will participating in this research involve? I will ask you to introduce me to or suggest women you think might be interested in talking to me about my topic. At the end of the study, I will also ask you to participate in a discussion with other women from the Division on what public and private healthcare means to all of us and our respective communities. I may also ask you to participate in an interview to discuss in more detail your personal experiences with healthcare. If there are some things you don’t want to talk about during our meetings or the discussion, that is fine. We do not need to discuss any topic that you do not wish to.

Can you withdraw from this study? If you wish to withdraw from this study, you may do so at any time by informing me. It will not be possible to withdraw completely from the study 4 weeks after I finish my data collection because I will not be able to trace all the information you provided back to you.

Will I mention your name? No. I will never mention your name. We will keep your personal information confidential. You do not have to provide any information that will identify you during the discussion or the interview. I cannot guarantee confidentiality at the discussion as there will be other participants but we will come to an agreement to keep what we discuss confidential and use measures to protect confidentiality. I will invite you to determine what you would like to share with the group, knowing that confidentiality cannot be guaranteed. I will request that you do not use your own name or refer to others in the group by their names during this discussion. I will make notes and record the group discussion on tape, but this is only to make sure that we don’t miss any of your valuable ideas and opinions. I may use some of what you say when I write reports of my research, but I will remove any information that identifies you. Eventually, the audiotapes and written notes will be destroyed.

171

How will this research be useful? There will be no direct benefits to you from participating in this research. However, the findings of this research may help shape healthcare financing policy in Sri Lanka at a time when major changes are being planned for the health sector. This research will also deepen our understanding of the impact of healthcare privatization from the perspective of users, something that has been neglected in Sri Lanka. Finally, this study will be a way for us to come together as a community to discuss the problems we face in accessing health services.

Will you receive compensation? If the focus group or interview is held during your regular working hours, I will compensate you for your work time. In addition, you will receive a pack of food items as a token of appreciation for participating in a discussion and/or interview. If you travel by 3-wheeler to the discussion, I will reimburse your travel expenses.

Who is doing the research? I, Ramya Kumar, am a PhD student in Public Health at the University of Toronto in Canada. This research is funded by the International Development Research Centre and the University of Toronto. For further information, you may contact the Office of Research Ethics at the University of Toronto (Email: [email protected]; Phone: 416-946-327). You may also contact the faculty member supervising this work: Dr. Peggy McDonough (Email: [email protected]; Phone: 416-946-7936).

Consent for participation

By writing my name, I express my willingness to contribute toward this study as described above.

Name: Date:

Consent to be contacted for a discussion/interview

By writing my name, I express my willingness to be contacted for a discussion and an interview in the future.

Name: Date:

172

APPENDIX D

Sociodemographic profile of community informants130

Community Study area Age Ethnicity Education Occupation informant

Nuzrath Jagoda (1) 52 Muslim Primary school Homemaker Premalatha Neluwa (1) 66 Sinhala Ordinary Level Self-employed; makes paper bags for sale to small businesses when paper is available. Fathima131 Bogoda (2) 55 Muslim Advanced Level Accounts clerk (retired) Muditha Kirideniya (4) 54 Sinhala Advanced Level Seamstress (self-employed); works only to keep herself occupied, as household income from family-owned business is adequate.

130 Pseudonyms are used in all tables to protect the anonymity of participants. 131 Fathima’s social background did not reflect those of other participants from Bogoda (2). She had worked as a clerk until chronic arthritis compromised her mobility. Her reported household income was also higher than others in the Bogoda group.

173

APPENDIX E

Script to invite community residents to participate in the study

Hello,

I’m doing my PhD in Public Health at the University of Toronto in Canada. I’m studying women’s experiences of health care at a time when we seem to be relying more on more on services in private clinics and hospitals. I’d like to better understand how women make decisions about healthcare for themselves and their families and what they think about the public and private health services available to them.

I’m inviting women from the Division to take part in a group discussion and, possibly, a follow-up interview. Those who participate will not benefit directly. However, they will contribute a great deal to the successful completion of my project, and the findings may help guide future policies on healthcare financing in Sri Lanka. This study will also be a way for us to come together as a community to discuss the problems we face in accessing health services.

Would you consider participating in this study? Participating in this study is completely voluntary and you may withdraw at any time.

174

APPENDIX F

Form to record contact details of community residents and their consent to be contacted to schedule a focus group or interview

I’m studying women’s experiences of using health care at a time when we seem to be relying more on more on services in private clinics and hospitals. I’d like to better understand how women make decisions about healthcare for themselves and their families and what they think about the public and private health services available to them.

These are some of the question I’m interested in: . How women make decisions and what they do to access healthcare in government or private clinics and hospitals; . What women feel about the availability, accessibility and affordability of health services at government or private clinics and hospitals; . What women feel about their interactions with healthcare providers and the standard of care in government or private clinics and hospitals; . How being poor, or being Sinhala, Tamil or Muslim may affect experiences of accessing and receiving healthcare at government or private clinics and hospitals.

To better understand these questions, I’m inviting women from the Division to take part in a discussion and, possibly, a follow-up personal interview. May I contact you to schedule these meetings?

Contact details

By providing my contact details, I agree to be contacted to participate in a discussion and/or interview.

Name: ……………………………………………………………………………………………..

Address: …………………………………………………………………………………………..

Telephone number: ……………………………………………………………………………….

175

APPENDIX G

Information letter and consent form for focus groups

What is this research about? We have a system of public and private healthcare and nowadays we seem to rely more and more on private healthcare. I’m trying to understand the growing use of private health care from the perspective of women living in the xxxxx Grama Niladhari Division. I am especially interested in understanding: 1) How women make decisions and what they do to access healthcare in government or private clinics and hospitals; 2) What women feel about the availability, accessibility and affordability of health services at government or private clinics and hospitals; 3) What women feel about their interactions with healthcare providers and the standard of care in government or private clinics and hospitals; and 4) How being poor, or being Sinhala, Tamil or Muslim may affect experiences of accessing and receiving healthcare at government or private clinics and hospitals.

What will participating in this research involve? One discussion with a group of about 5 to 7 women that will last about 1.5 to 2 hours. I will start the discussion with a few general questions to get the conversation going. I may ask for your thoughts about health care problems that people might have. If you do not want to answer a particular question or talk about a particular topic, you can remain silent or request to pass and we will move on. As you share your experiences, I will listen attentively. I may also ask questions to clarify something that has been said. I may also ask questions to refocus the conversation back to the purpose of this research. If you ever feel you want to leave the meeting, that is no problem. You can do so at any time.

I will be asking some of those who contributed to the group discussion to meet with me privately to talk further about your experiences with public and private health services, and how you make decisions about using healthcare. I may also ask for your thoughts about health care problems that people might have during the interview. This interview will take about 1.5 to 2 hours in a location of your choice. If you would be willing to be contacted for an individual interview, please write your name in the space provided at the bottom of this form. You can refuse to participate in the interview, if you change your mind later.

Will I mention your name? No. I will never mention your name. We will keep your personal information confidential. You do not have to provide any information that will identify you during the discussion. I cannot guarantee confidentiality at the discussion as there will be other participants but we will come to an agreement to keep what we discuss confidential and use measures to protect confidentiality. I will request that you do not use your own name or refer to others in the group by their names during this discussion. I will make notes and record the group discussion on tape, but this is only to make sure that we don’t miss any of your valuable ideas and opinions. I may use some of what you say when I write reports of my research, but I will remove any information that identifies you. Eventually, the audiotapes and written notes will be destroyed.

Since this study involves group activities, we must all agree to keep everything that is shared in this discussion confidential. This means respecting people’s privacy by not sharing any information you learn

176 about others in the study with anyone. To encourage the respect of everyone’s privacy, we will ask all participants to consent to a confidentiality agreement. In the end, however, we cannot provide a guarantee that participants will respect the agreement. I invite you to determine what you would like to share with the group, knowing that confidentiality cannot be guaranteed.

Can you withdraw from this research? If you wish to withdraw from this study, you may do so at any time by informing me. It will not be possible to withdraw completely from the study 4 weeks after I finish my data collection because I will not be able to trace all the information you provided back to you.

How will this research be useful? There will be no direct benefits to you from participating in this research. However, the findings of this research may help shape health financing policy in Sri Lanka at a time when major changes are being planned for the health sector. This research will also deepen our understanding of the impact of healthcare privatization from the perspective of users, something that has been neglected in Sri Lanka. Finally this study will be a way for us to come together as a community to discuss the problems we face in accessing health services.

Will you receive compensation? If the focus group is held during your regular working hours, I will compensate you for your work time. In addition, you will receive a pack of food items as a token of appreciation for participating in a discussion. If you travel by 3-wheeler to the discussion, I will reimburse your travel expenses.

Consent for participation in focus group discussion

By writing my name, I express my willingness to participate in this discussion. I also agree for this discussion to be audio recorded.

Name: Date:

Confidentiality agreement

By writing my name, I agree that I will not tell anyone about any of the information that I hear at this discussion.

Name: Date:

177

Consent to be contacted for an individual interview

By writing my name, I express my willingness to be contacted for an individual interview in the future. I may choose not to participate in this interview at a later time.

Name: Date:

178

APPENDIX H

Sociodemographic data form

I would like to understand how public and private healthcare is experienced by people from different backgrounds. To do this, I would like you to complete this form. Doing so is voluntary, and if you do not want to answer any particular section in this form, you can move on to the next. Please ask if you would like me to help you complete this form.

ID Number 1. Resident neighbourhood: 2. Age: 3. Highest level of education completed and institution: 4. Highest level of education completed by parents and institution: Mother : Level School Father: Level School 5. Occupation: 6. Occupations of parents: Mother : Father: 7. Number of members in household:

Adults Children/adolescents (<18 years)

8. Schools attended by children/adolescents (<18 year) in household: Child 1 Child 3 Child 2 Child 4 9. Where do you usually shop for groceries?

10. How do you travel when you visit your relatives who live outside Kandy?

11. Household income (Sri Lankan rupees): < 20,000 50,000 – 99,000

21,000 – 50, 000 >100, 000

12. Ethnicity: 13. Religion:

179

APPENDIX I

Composition of focus groups132

1 - Neluwa/Jagoda focus group

No. Name Age Ethnicity Education133 Present occupation Participated in (previous occupation) individual interview Yes/No

1 Damayanthi 48 Sinhala Ordinary Homemaker Yes Level (resigned/janitor) 2 Chamila 35 Sinhala Ordinary Self-employed Yes Level seamstress 3 Nuzrath 52 Muslim Primary Homemaker No School 4 Pradeepa 45 Sinhala Secondary Homemaker No School (resigned/janitor) 5 Chulani 42 Sinhala Ordinary Supplies snacks to No Level cafeteria 6 Premalatha 66 Sinhala Ordinary Homemaker/ makes No Level paper bags for sale to small businesses 7 Aisha 54 Malay Primary Homemaker No School

2 - Bogoda focus group

No. Name Age Ethnicity Education Present occupation Participated in (previous occupation) individual interview Yes/No

1 Sabina 33 Muslim Advanced Homemaker Yes Level 2 Feroza 41 Muslim Advanced Homemaker Yes Level 3 Fathima 55 Muslim Advanced Homemaker (retired Yes Level clerk)

4 Rifana 40 Muslim Advanced Homemaker No Level (resigned/teacher)

132 Pseudonyms are used in all tables to protect the anonymity of participants. 133 North American equivalent: Primary = Grade 5; Secondary school = Grade 6 to 10 (Ordinary Level Examination not completed); Ordinary Level = Grade 10; Advanced Level = Grade 12.

180

3 - Dehioya focus group

No. Name Age Ethnicity Education Present occupation Participated in (previous occupation) individual interview Yes/No

1 Amali 47 Sinhala Advanced Level Runs self-owned Yes breakfast bar and keeps lodgers 2 Dilrukshi 39 Sinhala Advanced Level Homemaker Yes (resigned/private hospital attendant) 3 Indrani 56 Sinhala Advanced Level Homemaker (retired Yes clerk) 4 Pushpa 33 Sinhala Ordinary Level Homemaker No 5 Badra 50 Sinhala Ordinary Level Field supervisor at NGO No

4 – Kirideniya (Sinhala)

No. Name Age Ethnicity Education134 Present occupation Participated in (previous occupation) individual interview Yes/No

1 Ruwani 50 Sinhala Advanced Level Homemaker Yes (resigned/bank clerk) 2 Muditha 54 Sinhala Advanced Level Homemaker/self- Yes employed seamstress 3 Prema 60 Sinhala Ordinary Level Homemaker No 4 Thushari 38 Sinhala Advanced Level Homemaker No 5 Kamala 65 Sinhala Ordinary Level Self-employed/supplies No food to cafeteria 6 Anoja 62 Sinhala Advanced Level Homemaker No (retired/clerk) 7 Pabha 84 Sinhala Ordinary Level Homemaker No (retired/sales person in family business)

134 North American equivalent: Primary = Grade 5; Secondary school = Grade 6 to 10 (Ordinary Level Examination not completed); Ordinary Level = Grade 10; Advanced Level = Grade 12.

181

5 – Kirideniya (English)

No. Name Age Ethnicity Education Present occupation Participated in (previous occupation) individual interview Yes/No

1 Shirani 73 Sinhala Undergraduate Homemaker Yes degree + Diploma (librarian) 2 Sakuntala 75 Sri Lankan Undergraduate Homemaker Yes Tamil degree 3 Kishani 37 Sinhala Advanced Level + Homemaker/assists Yes Diploma with family business 4 Ira >70 Sinhala Undergraduate English instructor No degree + Diploma 5 Anushka 46 Sinhala Master’s degree Assistant Manager of No a primary/secondary school 6 Nalini 63 Sinhala Advanced Level + Homemaker No Diploma in (hospitality hospitality management) management 7 Sharmini 42 Mixed PhD University teacher No

Appendix J. Overview of individual interviews135

No.136 Name Neighbour- Age Language Ethnicity Education137 Present occupation (previous occupation) hood

1 Shereen Jagoda 29 Sinhala Malay Ordinary Level Confined to home with a physical disability 2 Nafeesa Jagoda 42 Sinhala Muslim Primary Homemaker 3* Damayanthi Jagoda 48 Sinhala Sinhala Ordinary Level Homemaker (resigned/janitor) 4 Siththy Jagoda 55 Sinhala Malay Secondary Homemaker 5* Chamila Neluwa 35 Sinhala Sinhala Ordinary Level Self-employed seamstress 6 Leela138 Neluwa ~60 Sinhala Sinhala - Homemaker 7* Fathima Bogoda 55 Sinhala Muslim Advanced Level Confined to home with a physical disability (resigned/clerk) 8* Feroza Bogoda 41 Sinhala Muslim Advanced Level Homemaker 9* Sabina Bogoda 33 Sinhala Muslim Advanced Level Homemaker (resigned/computer assistant) 10* Indrani Dehioya 56 Sinhala Sinhala Advanced Level Homemaker (retired/clerk) 11* Amali Dehioya 47 Sinhala Sinhala Advanced Level Owns and runs snack bar and caters to boarders

135 Pseudonyms are used in all tables to protect the anonymity of participants. 136 Only 22 interviews were held as two participants (Chamila and Leela) were interviewed in one sitting. 137 North American equivalent: Primary = Grade 5; Secondary school = Grade 6 to 10 (Ordinary Level Examination not completed); Ordinary Level = Grade 10; Advanced Level = Grade 12. 138 Sociodemographic data form not completed.

182

No.136 Name Neighbour- Age Language Ethnicity Education137 Present occupation (previous occupation) hood 12* Dilrukshi Dehioya 39 Sinhala Sinhala Advanced Level Homemaker (resigned/hospital attendant)

13 Fara Dehioya 49 Sinhala Muslim Advanced Level Homemaker (resigned/office assistant) 14 Namali Kirideniya 27 Sinhala Sinhala Advanced Level + Bank executive officer (S) Diploma 15* Ruwani Kirideniya 50 Sinhala Sinhala Advanced Level Homemaker (resigned/clerk) (S) 16 Sonali Kirideniya 37 Sinhala Sinhala Advanced Level + Director of family-owned business (S) Diploma 17* Muditha Kirideniya 54 Sinhala Sinhala Advanced Level Self-employed seamstress (S) 18* Sakuntala Kirideniya 75 English Tamil Bachelors Homemaker (E) 19* Kishani Kirideniya 37 English Sinhala Advanced Level Works part-time in family-owned business (E) 20* Shirani Kirideniya 73 English Sinhala Bachelors + Homemaker (retired/librarian) (E) Diploma 21 Mala Kirideniya >70 English Sinhala Bachelors Homemaker (retired/teacher) (E) 22 Suwini Kirideniya 30 English Sinhala Medical degree Medical Officer (E) 23 Himali Kirideniya 35 English Sinhala Advanced Level + Homemaker (resigned/assistant manager) (E) Diploma

* Participated in a focus group

183

184

APPENDIX K

Information letter and consent form for individual interviews

What will participating in this research involve? One interview lasting about 1.5 to 2 hours. The interview will include questions about your experiences with public and private health services, and how you make decisions about using healthcare. I may ask for your thoughts about health care problems that people might have. As you share your experiences, I will listen attentively. I may also ask questions to clarify something that has been said. I may also ask questions to refocus the conversation back to the purpose of this research. If there are some things you don’t want to talk about, that is fine. We do not need to discuss any topic that you do not wish to. If you ever feel you want to end the interview, that too is no problem. I may also need to contact you again for clarification about what you said in this interview. If that’s okay, please write your name in the space provided at the end of this form.

Will I mention your name? No. I will never mention your name. I will make notes and record the group discussion on tape, but this is only to make sure that we don’t miss any of your valuable ideas and opinions. Please do not use your own name or refer to others in the group by their names during the interview. I may use some of what you say when I write reports of my research, but I will remove any information that identifies you. Eventually, the audiotapes and written notes will be destroyed.

Can you withdraw from this research? If you wish to withdraw from this study, you may do so at any time by informing me. It will not be possible to withdraw completely from the study 4 weeks after I finish data collection because I will not be able to trace all the information you provided back to you.

185 privatization from the perspective of users, something that has been neglected in Sri Lanka. Finally this study will be a way for us to come together as a community to discuss the problems we face in accessing health services.

Will you receive compensation? If the interview is held during your regular working hours, I will compensate you for your work time. In addition, you will receive a pack of food items as a token of appreciation for participating in an interview. If you travel by 3-wheeler to the focus group discussion, I will reimburse your travel expenses.

Consent to be interviewed

By writing my name, I express my willingness to be interviewed. I also agree for this interview to be audio recorded.

Name: Date:

Consent to be contacted for clarifications

By writing my name, I express my willingness to be contacted for clarifications in the future.

Name: Date:

186

APPENDIX L

Script to invite participants to the final meeting

As you know, during the past few months, I’ve been talking to women in different communities and neighbourhoods in the xxxxx Division to learn about how their experiences with healthcare might be affected by the expansion of private health care.

I am nearing the end of my data collection, and am in the process of organizing a final group discussion. I am inviting you, along with other women who have participated in the project, to this last gathering. The meeting has several purposes. One is to give you and other participants the opportunity to meet each other and learn about each other’s experiences of accessing and receiving healthcare. Another is for me to present preliminary study findings and hear your thoughts on them. I would also like to discuss ways to use the study findings to achieve positive changes in healthcare.

This final discussion will last about 1.5 to 2 hours. Tea will be served at the end of the meeting.

Would you like to participate in this discussion? If so, please write your name below to indicate that I can contact you about it.

Name:

Date:

187

APPENDIX M

Information letter and consent form for the final meeting

What is this meeting about? As you know, I’ve been studying women’s experiences with healthcare to understand how the expansion of private health care shapes healthcare access in the xxxxx Division. I am nearing the end of my data collection, and now invite you, along with all other participants, to a final meeting. The purpose of this gathering will be: 1) for participants to meet each other and learn about each other’s experiences of accessing and receiving healthcare through a presentation of preliminary study findings from the group discussions and individual interviews; 2) for us to talk about healthcare access issues faced by women in the xxxxx Division; 3) to identify the most important healthcare access issues affecting residents of this Division; and 4) to discuss possible ways to use my findings to achieve positive changes in healthcare.

What will this meeting involve? The meeting will include a short presentation of preliminary study findings (about 15 minutes), followed by a discussion among participants about them (about 45 minutes). I will invite everyone to tea after the meeting. You do not need to say anything during the discussion unless you want to. If you want to leave the gathering at any time, you may do so.

Who is participating in this meeting? Women who have participated thus far in this study, either in a group discussion or an individual interview. Your attendance at this final meeting is completely voluntary. This means you do not need to take part unless you want to. Before you agree to participate, however, it is very important that you understand the information on this page. If there is anything that is not clear to you, please ask me.

Will I mention your name? No. I will never mention your name. I will keep your personal information confidential. You do not have to provide any information that identifies you during the meeting, and please do not say anything that you do not wish others to know. We must all agree to keep everything that is shared in this discussion confidential. This means not sharing anything you learn about others at the gathering. I will ask all participants to sign a confidentiality agreement, but in the end, I cannot guarantee that each person will respect the agreement. The discussion will not be recorded; my research assistant and I will make notes during the meeting. Eventually, the written notes will be destroyed. I may use some of what you say when I write reports of my research, but will remove any information that identifies you.

Can you withdraw from this meeting? If you wish to withdraw from this meeting, you may do so by informing me. It will not be possible for me to remove everything you said at the meeting 4 weeks after it is held because the data will be broken down and mixed with others’ ideas by that time. Thus I may not be able to trace the information you provide back to you.

How will this focus group be useful? There may be no direct benefit to you from participating in this meeting. But the meeting will provide an opportunity for me to share some general, preliminary findings with you, as well as receive your feedback on these findings. The meeting will also provide an occasion for us to come together as a community to discuss the problems we face in accessing health services.

188

Who is doing the research? I, Ramya Kumar, am a PhD student in Public Health at the University of Toronto in Canada. This research is funded by the International Development Research Centre and the University of Toronto. For further information, you may contact the Office of Research Ethics at the University of Toronto (Email: [email protected]; Phone: +1-416-946-327). You may also contact the faculty member supervising this work: Dr. Peggy McDonough (Email: [email protected]; Phone: +1-416-946- 7936).

Consent for participation in the final focus group discussion

By writing my name, I express my willingness to participate in this discussion.

Name: Date:

Confidentiality agreement

By writing my name, I agree that I will not share any information I learn about others with those not present at this discussion.

Name: Date:

189

APPENDIX N

Focus group discussion guide

Thank you for agreeing to participate in this discussion today. As you may be aware, we have a system of public as well as private healthcare and nowadays we seem to rely more and more on private healthcare. I am interested in your experiences using public and private healthcare. The discussion today will not be about your specific health conditions. Instead, I would like you to talk about what you do and where you go when you have a health problem.

I will start by asking you all a few general questions about your recent healthcare experiences to get our discussion going. I may also use an example of a health situation and ask for your thoughts on how the situation could be addressed. As you share your opinions and experiences, I will listen attentively. I may ask questions to clarify something that’s been said. I may also ask questions to refocus the conversation back to the purpose of this research.

Because this is a group interview, I want to stress once again the importance of maintaining confidentiality. By participating in this study, I ask that you all respect that what gets talked about here should not be discussed outside of this group. I invite you to determine what you would like to share with the group, knowing that confidentiality cannot be guaranteed. Nonetheless, we count on everyone to cooperate in keeping confidentiality. Please do not use your own name or refer to others in the group by their name during this discussion. Please also do not state the names of specific healthcare providers in our discussion.

As I mentioned earlier, I will audio-record this interview and write down your words at a later time because I will not be able to write as fast as you share your ideas and thoughts. I also want to be able to focus on what you are saying and capture what you share accurately. If at any time you do not want me to record something, let me know and I will turn off the recorder. The information you share with me will be kept confidential, which means that in no way will your words be identified to you.

Questions: 1. What is your overall experience of using health services? 2. How do you feel about the health services available to you? 3. Think back to the last time you had a health problem. What did you do? 4. How did you go about making your decision? What influenced you most in making your decision? 5. How satisfied have you been you with your decision? Please tell me more about why you feel this way. 6. I am interested in your thoughts about our current health care system – particularly about public and private health services. What do you think are the pros public healthcare? What about the cons of public healthcare? What about the pros and cons of private healthcare? 7. If you could change one thing about the health services available to you, what would it be?

Scenarios: 1. Rifana is 55 years of age. She was recently diagnosed with diabetes and heart disease and directed to the closest government medical clinic for treatment and follow up. She has to attend this clinic once a month to get her medicines. If you were Rifana, what would you do? Probes: Do you think who Rifana is and what she does may matter for her to obtain the care she needs? Follow up: Rifana experiences mild chest pain while working in the garden. After doing several tests, she is told that she needs by-pass surgery. If you were Rifana, what would you do?

190

2. Anula is 40 years of age and has had irregular periods for about 3 months. She now wants to see a doctor. She is a working woman and has three children below 12 years of age. If you were Anula, what would you do?

Probes: Would it matter that it’s a “women’s problem” when Anula decides where to go for healthcare? What if Anula needs an operation? Follow up: Anula finds out that she is pregnant. If you were Anula, what would you do?

Thank you for participating in this discussion. I appreciate the time you spent with us today.

If you have any questions or concerns about the discussion, please contact me.

191

APPENDIX O

Individual interview guide

(i) For those who did not participate in a focus group discussion:

I will start by asking you a few general questions about your recent healthcare experiences to get our discussion going. I may also use an example of a health situation and ask for your thoughts on how the situation could be addressed. As you share your experiences, I will listen attentively. I may ask questions to clarify something that you’ve said. I may also ask questions to refocus the conversation back to the purpose of this research.

As I mentioned earlier, I will audio-record this interview and write down your words at a later time because I will not be able to write as fast as you share your ideas and thoughts. I also want to be able to focus on what you are saying but at the same time capture what you share accurately. If at any time you do not want me to record something, let me know and I will turn off the recorder. The information you share with me will be kept confidential, which means that in no way will your words be identified to you. Along these lines, please do not state the names of specific healthcare providers in our discussion.

Questions: 1. What is your overall experience of using health services? 2. How do you feel about the health services available to you? 3. Think back to the last time you had a health problem. What did you do? 4. How did you go about making your decision? What influenced you most in making your decision? 5. How satisfied have you been you with your decision? Please tell me more about why you feel this way. 6. What was the health care system like when you were growing up? 7. I am interested in hearing your thoughts about our current health care system – particularly about public and private health services. What do you think are the pros and cons of public healthcare? What about private healthcare? 8. If you could change one thing about the health services available to you, what would it be?

192

Thank you for participating in this interview. I appreciate the time you spent with me today. If you have any questions about the interview, please contact me. If you think of anything else that you would like to add or if you have any concerns about something that we have discussed today, please let me know.

(ii) For those who did participate in a focus group discussion:

Thank you for agreeing to participate in this interview today. As we discussed at the group meeting, we have a system of public and private healthcare in place and nowadays we seem to rely more and more on private healthcare. I am interested in your experiences using public and private healthcare. The discussion today will not be about specific health conditions. Instead, I would like you to talk about what you do and where you go when you have a health problem. I would also like to know more about some of the ideas you shared at the recent meeting.

Like the other day, I will start by asking you a few general questions about your recent healthcare experiences to get our discussion going. As you share your experiences, I will listen attentively. I may ask questions to clarify something that has been said. I may also ask questions to refocus the conversation back to the purpose of this research. I may ask you some of the questions we discussed the other day to understand your perspective better.

As I mentioned earlier, I will audio-record this interview and write down your words at a later time because I will not be able to write as fast as you share your ideas and thoughts. I also want to be able to focus on what you are saying but at the same time capture what you share accurately. If at any time you do not want me to record something, let me know and I will turn off the recorder. The information you share with me will be kept confidential, which means that in no way will your words be identified to you. Along these lines, please do not state the names of specific healthcare providers in our discussion.

1. What is your overall experience of using health services? 2. How do you feel about the health services available to you? 3. Think back to the last time you had a health problem. What did you do? 4. How did you go about making your decision? What influenced you most in making your decision? 5. How satisfied have you been you with your decision? Please tell me more about why you feel this way. 6. What was the health care system like when you were growing up? 7. I am interested in your thought about our current health care system – particularly about public and private health services. What do you think are the pros and cons of public healthcare? What about private healthcare? 8. If you could change one thing about the health services available to you, what would it be? 9. Did you find any of the topics that came up at our recent discussion relevant to you? 10. Were there any topics that you thought were not addressed? 11. Did anyone say anything that surprised you? 12. You mentioned that day that …………………….. Would you mind if we discussed this further today?

Thank you for participating in this interview. I appreciate the time you spent with me today.

193

If you have any questions about the interview, please contact me. If you think of anything else that you would like to add or if you have any concerns about something that we have discussed today, please let me know.

194

APPENDIX P

Guide for the final meeting

Introduction: Thank you for agreeing to come to this meeting today. As I mentioned earlier, I will start with a brief presentation of study findings, and then invite you to share your thoughts about them. I would also like to hear your opinion about the access issues we discuss today, and your ideas for ways to use the study findings to achieve positive changes in healthcare. After the discussion, tea will be served.

Because this is a group discussion, I want to stress the importance of maintaining confidentiality. I ask you to determine what you would like to share with the group, knowing that confidentiality cannot be guaranteed. Please make sure that you do not talk about what is discussed today outside this group. Please also do not state the names of healthcare providers during the course of the discussion. This discussion will not be audio-recorded like previous discussions. My research assistant and I will take notes during the meeting. We will not include any personal information (e.g. your name or any other person’s name) in our notes.

Guide: 1. Welcome [~3-5 mins]

2. Presentation [~15 mins] Preliminary findings from the focus groups and individual interviews will be presented with the use of visual aids/handouts based on the following themes: . What women do to access healthcare for themselves and their families; . Perceptions on the availability and affordability of public and private healthcare; . Perceptions on public and private healthcare institutions and providers; and . Suggestions for change expressed at focus group discussions and interviews.

3. Discussion [~45 mins]

Questions: . Based on these findings, what do you think are the most important healthcare access issues facing the residents of xxxxx? [A list will be developed on a board] . Do you think the listed changes would address these problems? [Refer back to visual aid/handout on suggested changes and go through them one by one] . Do you have any new suggestions? [Additions will be listed on a board] . How do you think the study findings could be used to achieve these changes? . Would you like to be involved in an initiative to work toward these changes?

4. Closing: Thank participants for attendance and invite to tea [5 mins]

195

APPENDIX Q

Handout for priority-setting exercise

Relevant to the public sector Reduce healthcare expenses

 All drugs and tests should be made available at public hospitals;  Quality of medicines in government hospitals should be as good as in the private sector;  Patients should not have to bring a list of items when admitted for a delivery or any other surgical procedure;  Patients should not need to pay for stenting procedures or buy medical devices in the private sector.

Reduce wait-times and delays

 Reduce waiting times at clinics, pharmacies and laboratories; address waiting lists for surgical procedures;  Relatives/friends of health workers should not be allowed to jump the queues;  Reorganize clinics with appointments or at least times on certain dates for specific groups of people; patients attending clinics should not need to stand in line from 5 am;  All counters should be kept open in the pharmacy for quick delivery; they should not have to stand 2-3 hours to get their medicines;  Hours must be extended for laboratory testing;

Address health worker issues

 Address gaps in field midwifery services, and appoint an additional midwife to the area;  Resolve issues for health sector employees to prevent strike action;  Increase salaries of public sector healthcare providers so that they can devote themselves to work in public hospitals without working part-time in the private sector;  Eliminate channel practice so that specialists spend more time in the public sector.

System-wide change

 Introduce a village-based system of family practice to bring health services closer to people, like for maternal and child health services.

Relevant to the private sector

 Introduce a proper referral system so that patients are directed to the correct specialist who will be able to address the problem;  Introduce an appointment system for channeling to reduce waiting times; reduce inconveniences like the requirement to pay upfront at time of booking;  Introduce a minimum time for each consultation (e.g. 30 minutes);  User-charges for consultations, treatment, investigations and in-patient care should be regulated.

Relevant to both sectors

BOTH SECTORS  Patients should be spoken to politely;  Toilets should be improved and kept clean;  Implement language policy – all health workers should know both languages;  All patients should be treated alike irrespective of language or ethnicity;  A complaint mechanism should be introduced;

196

 Elderly and disabled patients need special facilities; all hospitals should be accessible for people with disabilities;  Establish a system of record keeping that supports continuity of care.  Address the “real” issues experienced by people that affect their health. It is not only the lack of health services that needs to be addressed but also the reasons why people are getting sicker over time. People need to have better education and steady employment so that they can lead a stress-free life.

197

APPENDIX R

Coding framework

Patterns Concepts

Place ABROAD Affordability Amaya Availability Bogambara Clinic Changes over time Children's Hospital Class/SES COLOMBO COMMUNITY Kandy Channeling Center CONFIDENCE Kandy Hospital CONNECTIONS MCH Clinic CONVENIENCE Peradeniya Hospital Decisions Suduhumpola dispensary DISABILITY Suwasevana Hospital DISAGREE/DISSENT EMPOWERMENT Type of problem Children's health Ethnicity and language DENTAL Gender Emergency/complex Hygiene/toilets EYE Inequality/injustice/negligence FEVER/INFECTION MISTRUST/MOTIVES Minor ailments Organization NCDs OVERSTRETCHED PAIN OR RELATED Privacy Preventive/public health Quality Women's reproductive health Referral/follow up "SMALL" ILLNESS SDOH "LONG-TERM" ILLNESS STRIKE "BIG" ILLNESS TECHNOLOGY TIME WOMEN DOING Sector/ Institution Private ayurveda Researcher role Private channeling Private family practice Private in-patient Private labs Private pharmacies

Public ayurveda Public clinics Public in-patient Public labs Public OPD Public pharmacy

Healthcare providers

OOP spending INSURANCE PRESIDENT'S FUND Private sector OOP Public sector OOP

Proposals for change

Lower case: pre-defined codes; upper case: in vivo codes