Maternal Mortality among Pastoralists in Kenya

1 Maternal Mortality among Pastoralists in Kenya Submitted to Institute of Development Studies University of Sussex Brighton In partial fulfilment of the requirement for the degree of Masters of Arts in Development Studies Candidate Number: 161638 2 Summary My dissertation focusses on the relevance of applying an inclusive social protection framework for the reduction of maternal mortality among pastoralists in Kenya. I will discuss this using the case study of the Free Maternity Services program introduced in Kenya in 2013 which aims at providing access to free maternal health services to the poor and economically disadvantaged sections of its population. Using evidence from reports on the implementation of the program, I will provide a detailed analysis of applying an inclusive approach to this intervention in order to make it more equitable and accessible to the pastoralist communities in the region. 3 Acknowledgement After an intensive period of three months, I am writing this note of gratitude for my supervisor, Linda Waldman for her wonderful guidance and feedback throughout the course of my dissertation. I would like to thank her for always willing to help me, and for all the opportunities I was given to present my ideas before her and further my research. In addition, I would like to extend my gratitude to my tutors, Hayley Macgregor and Jalia Kangave, for providing me with the tools that I needed to choose the right direction and successfully complete my dissertation. I would also like to thank my friend, Devanik Saha for his kind words of encouragement and a sympathetic ear. 4 List of Tables Table 1: Regions of Kenya with highest MMR Table 2: Categorization of counties based on percentage of ASAL coverage List of Abbreviations FMS: Free Maternity Services GHA: Greater Horn of Africa WHO: World Health Organization MDG: Millennium Development Goal SDG: Sustainable Development Goal RMNCH: Reproductive, Maternal, Newborn and Child Health MMR: Maternal Mortality Ratio UNFPA: United Nations Population Fund ASAL: Arid and Semi-Arid Land UNDP: United Nations Development Program TBA: Traditional Birth Attendant SBA: Skilled Birth Attendant ODI: Overseas Development Institute ILO: International Labor Organization PAHO: Pan American Health Organization UNICEF: United Nations Children’s Fund KHSSP: Kenya’s Health Sector Strategic Plan ANC: Antenatal Care ICU: Intensive Care Unit 5 Table of Contents Summary .................................................................................................................................... 2 Acknowledgement ..................................................................................................................... 3 List of Tables ............................................................................................................................. 4 List of Abbreviations ................................................................................................................. 4 Introduction ................................................................................................................................ 6 Chapter 1 Background ............................................................................................................... 9 1.1 Maternal Mortality: A Global Challenge ......................................................................... 9 1.2 Maternal Mortality in Kenya: Policies and Demographic Trends ................................. 10 1.3 The Pastoral Context ...................................................................................................... 14 1.3.1 Who are Pastoralists? .............................................................................................. 14 1.3.2 Poverty .................................................................................................................... 14 1.3.3 Geographical Dispersion and Availability of Data ................................................. 15 1.3.4 Discrimination in Service Provision ....................................................................... 15 1.3.4 Mobility .................................................................................................................. 16 1.3.5 Gender Disparity ..................................................................................................... 17 Chapter 2 Inclusive Social Protection ...................................................................................... 18 2.1 Pastoralists’ Vulnerability: A Political Consequence? .................................................. 18 2.2 The Social Protection Dividend in Maternal Health ...................................................... 19 2.3 Adding the Element of Inclusion ................................................................................... 21 2.4 The Influence of State- Citizen Dynamics ..................................................................... 22 Chapter 3 Discussion ............................................................................................................... 24 3.1 Case Study: Kenya’s Free Maternity Services Program ................................................ 24 3.2 Challenges and Recommendations ................................................................................ 25 3.2.1 Quality of Care........................................................................................................ 25 3.2.2 Accessibility............................................................................................................ 28 3.2.3 Lack of Understanding of FMS among Users ........................................................ 30 3.2.4 Out of Pocket Payments and Disaggregated Data .................................................. 31 3.2.5 Lack of Motivation to implement Policy ................................................................ 31 Conclusion ............................................................................................................................... 33 Bibliography ............................................................................................................................ 36 6 Introduction My dissertation explores the significance of applying an inclusive social health protection approach in national policy frameworks in order to reduce maternal mortality among pastoralists in Kenya. The research question that will drive my dissertation is ‘How can Kenya’s Free Maternity Services Program be made more inclusive towards pastoralist communities?’ Pastoralists refer to mobile livestock herders in the dimension of production or livelihood. Nomadic and seasonal rearing of domesticated animals apart from farming or enclosed ranching are the essential forms of pastoralism. A clan is generally the basis of their organization, which is responsible for the control of territory and management of the livestock (Dong, 2016). Through my dissertation, I aim to address the gap that emerges in the framing of policies for pastoralists when it comes to providing them with maternal health services. This is important as pastoralists are usually excluded in the dominant narrative which leads to policies being insufficiently tailored to their needs (Ali, Hobson, 2009; Ogachi, 2011). Furthermore, I will highlight the importance of framing inclusive policies to make maternal health interventions more accessible to pastoralists. For this, I will analyze the application of the inclusive social protection framework provided by Roelen and Devereux, (2013) and UNICEF (2012) to maternal health policies, particularly the Free Maternity Services (FMS) program which was introduced in Kenya in 2013 (Ministry of health, 2015). I have chosen this program for my analysis because it is the largest and the most prominent maternal health scheme formulated in Kenya in the recent years. In pastoralist communities’ lives, reproductive health comes up as a crucial issue in policy outcomes, which lack a general sense of commitment to the pastoralist way of life. This has led to the continued exclusion of pastoral communities in national development interventions and a consistent lack of access to social services on the part of the pastoralists, especially in the Greater Horn of Africa (GHA) region (Maro et. al, 2012). Their lifestyle of moving from one place to another for subsistence seems to deprive them from basic reproductive health services. This trend is further complicated by geographical isolation, physical distance to formal health services, gender disparities, traditional birth practices and more importantly, inadequate service provision tailored to the pastoralists’ needs at health facilities (Ibid.). 7 Although governments appear willing to develop comprehensive policies, these are more often than not, inappropriate and lack follow-through on their implementation, especially at lower levels of the bureaucracy (Humanitarian Policy Group, 2010). For my research, I will look at a range of secondary, web based resources which talk about the implementation and uptake of the FMS program. These resources will include government reports (Ministry of Health, 2015), studies comprising interviews of health facilities’ staff (Wamalwa, 2015; Njuguna, Kamau, Muruka, 2017) and studies on uptake of scheme in pastoral counties (Caulfield et. al, 2016). I will also draw upon theoretical perspectives on maternal mortality (Keats et. al, 2017) and ethnographic evidence (Schelling, Weibel, Bonfoh, 2008) to highlight the trends in recent maternal health interventions in Kenya. One of the limitations of my dissertation is that it will use secondary resources and literature rather than primary data. I acknowledge that this might constrain the scope of my arguments within the realm of literature available to me on the internet, however, I will try to look at different kinds of material based on authorship to present a well-balanced and diverse perspective on the issue I am trying to tackle. Another limitation that I perceive in the process of my dissertation is the lack of disaggregated official data on the community that I am trying to engage with (Humanitarian Policy Group, 2010). Accurate data on pastoralists is scarce not only because they are a mobile community but also due to the failure in part of the government to define them in the national narrative (Ibid.). I will try to highlight this issue strongly in my dissertation and try to look at resources which specifically talk about maternal mortality in the context of pastoralism. Furthermore, my positionality as an economically privileged Indian student, studying in the United Kingdom may affect my analysis of the pastoral context. The lens through which I will analyze the implication of the Free Maternity services program on pastoralists may obliterate some hidden complexities which I will not be aware of due to a lack of knowledge about the diverse ethno racial groups in Kenya (Kenya National Bureau of Statistics, 2009) and a lack of primary level research on pastoralist communities within this region. I will make all attempts to address the unintended mistakes that might arise from any pre conceived notions about pastoralists or the Kenyan society from my positionality as a researcher. I would also like to point it out, that my dissertation does not aim to disregard any of the well intentioned political interventions undertaken by Kenya in its quest to reduce the rate of maternal mortality. I am only trying to examine the relevance of using an inclusive framework to add a fresh orientation 8 to policies that aim at addressing the above issue, in order to make them more equitable towards the marginalized communities. 9 Chapter 1 Background 1.1 Maternal Mortality: A Global Challenge According to WHO (World Health Organization), Maternal death or maternal mortality is defined as ‘the death of a woman while pregnant or within 42 days of termination of pregnancy, irrespective of the duration and site of the pregnancy, from any cause related to or aggravated by the pregnancy or its management but not from accidental or incidental causes’ (WHO, 2017)1. In circumstances wherein the cause of death has not been identified, a new categorical definition has been introduced. Pregnancy-related death has been defined as ‘the death of a woman during pregnancy or within 42 days of termination of pregnancy, irrespective of the cause of death’ (WHO, 2017). Since the past three decades, efforts have been made to address the issue of maternal death on a global level (Kassebaum et. al, 2014) which have led to a 44% decrease in the rate of maternal mortality from 1990, especially during the years 2010 to 2014 (Bohren et.al, 2015; Chou, Daelmans, Jolivet, Kinney, 2015; Say et. al, 2014). Callister and Edwards (2017) have observed that the causes for maternal deaths vary worldwide, among countries and regions within those countries. Globally these causes can be broken down as follows: Cases of hemorrhage (27%), indirect causes (27%), hypertension (14%), sepsis (10%), other direct causes (10%), abortion (9%), and embolism (3%) have been reported apart from 70% of indirect causes being related to pre-existing medical conditions (Say et. al, 2014). Since these factors vary contextually, it is difficult to address the underlying causes and factors that contribute to maternal mortality. Although, some people and countries might argue that a lot of progress has been made to address this issue globally, the enormity of this problem becomes more evident when it is confronted on a daily basis with 800 women dying everyday due to pregnancy related complications (Kassebaum et. al, 2014). It has been observed that nearly 99% of maternal deaths occur in developing regions, with the largest, i.e. 83.8% occurring in the Sub- Saharan African and Southeast Asian region (Kassebaum et. al, 2014; Say et.al, 2014). According to a report by the WHO (factsheet 348, 2015), structural 1 This definition has been taken from the website of WHO and therefore, there is no page number mentioned in the reference. 10 factors such as poverty, lack of education, traditional belief systems and limited access to healthcare are also significantly related to higher rates of maternal mortality in these regions. The Millennium Development Goals (MDGs) that were introduced in 2000 with the objectives of poverty reduction, halting the spread of HIV/AIDS, and the provision of universal primary education by the target of 2015, gave way to the implementation of the Sustainable Development Goals (SDGs) officially known as ‘Transforming Our World: The 2030 Agenda for Sustainable Development’, in order to build on the progress achieved through MDGs (United Nations, 2015). The new SDG goals, particularly target 3.1 which aims at the reduction of the global maternal mortality ratio to less than 70 per 100,000 live births by 2030 is in lieu of the agenda that was set by MDG 5 which had aimed to reduce the maternal mortality ratio by three quarters and to achieve universal access to reproductive health care by 2015 (United Nations, 2015). The components of SDG 3 are located under a broader theme of ensuring “that no one is left behind” in the process (United Nations, 2015, p. 11). This highlights that there is a need to engage with the issue of maternal mortality in a more contextualized manner with the objective of achieving positive outcomes which are inclusive (Devereux, Roelen, 2015) of all groups irrespective of their economic or social status/ identity. 1.2 Maternal Mortality in Kenya: Policies and Demographic Trends Maternal mortality ratio is defined as the number of maternal deaths within a given reference period (usually 1 year) per 100,000 live births during the same reference period (same year). Live birth is defined by the WHO as the complete removal or expulsion of the fetus which shows any evidence of life after it’s extraction from the mother, regardless of the duration of the pregnancy (WHO, 2004). Despite the global efforts to improve maternal health in an equitable manner and increase accountability of governments and stakeholders, progress in Kenya’s reproductive, maternal, newborn and child health (RMNCH) has been very inconsistent (Keats et. al, 2017). Kenya failed to achieve its MDG target of reducing maternal deaths by three-quarters i.e. bringing it down to 147 per 100,000 livebirths (United Nations, 2015). Between the years 1990 and 2015, maternal mortality declined at half the rate of under5 mortalities. Increased rates of maternal mortality were also observed in the early period from 1990 to 1999 whereas the rate of neonatal mortality remained more or less the same during that time frame (Keats et. al, 2017). This unsystematic and slow progress was characterized by several factors including low investment in healthcare, political instability, devolution of funds 11 and resources from central to county governments and structural disparities in terms of inequity in the coverage of health interventions by region, rural/ urban classification, wealth and level of education (Ibid.). Kenya’s national policy framework around reproductive, maternal, newborn and child health has seen a number of good policies and strategic interventions over the years, however none of these have been able to take into consideration mechanisms of accountability towards the marginalized populations2 (Abuya, 2014). Some of those policy initiatives were the Kenya health Sector Policy Framework 1994- 2010, the Reproductive Health Policy (2012- 2030), the National Health Sector Strategic Plan, Vision 2030 (a long term national development agenda which included a component on Health) and the National Roadmap 2010 (Ibid.). The National Road Map which was borne out of a status review report on the MDG goals set by Kenya by WHO in 2007, was a strategic framework that aimed at accelerating the reduction of maternal and newborn mortality and morbidity (Republic of Kenya, 2010). Despite of setting clear objectives, the framework could not provide information on the health financing system that the government planned to incorporate in order to support the implementation of its interventions (Abuya, 2014). The Reproductive Health Policy (Republic of Kenya, 2007) which was connected with the National Health Sector Strategic Plan, was structured around the need to provide equitable access to reproductive health backed by an efficient system of service delivery and better response to the needs of the users. However, the actions implemented failed to address the broad strategies outlined by the policy (Abuya, 2014). In order to ensure that the policies and commitments translate into improving the health and well-being of mothers, the Constitution of Kenya in 2013 introduced a decentralized system of governance which aimed at enhancing the access to health services for all Kenyans, especially those living on rural areas. With this new parameter, the commitment shifted towards implementing health policies at the county level with full involvement of policy makers and health functionaries (Republic of Kenya, 2012). 2 The draft Constitution of Kenya 2004 refers to the minority or indigenous communities in Article 306 as a ‘marginalized group or community’. 12 Although, national level trends in the above government interventions have seen a slight improvement, some geographical bias remains in counties located in the northern, eastern and Rift Valley regions (Keats et. al, 2017). A report by the United Nations Population Fund in Kenya listed out the counties in Kenya with the highest Maternal Mortality Ratio. This study was conducted in 2014 (UNFPA, 2014). Table 1: Regions of Kenya with highest MMR Maternal Mortality Ratio Region (deaths per 100,000 live births) MANDERA 3795 WAJIR 1683 TURKANA 1594 MARSABIT 1127 ISIOLO 790 SIAYA 691 LAMU 676 MIGORI 673 GARISSA 646 TAITATAVETA 603 KISUMU 597 HOMABAY 583 VIHIGA 531 SAMBURU 472 WESTPOKOT 434 Source: UNFPA It is possible to identify from the above table, that several counties with the highest MMR (except Siaya, Kisumu and Vihiga) have high pastoralist populations. These counties are also categorized as ASAL3 counties in Kenya (Ministry of Devolution and Planning, 2015). From Table 1, Mandera, Wajir, Turkana, Marsabit, Isiolo and Garissa, which are shown to have the 3 Pastoral communities in Kenya are found in areas identified as ASAL (Arid and Semi-Arid Lands). 13 highest ratio of Maternal Mortality are classified as 100% ASAL as demonstrated in the table below. Table 2: Categorization of counties based on percentage of ASAL coverage Category Districts 100% ASAL Turkana, Moyale, Marsabit, Isiolo, Wajir, Mandera, Garissa, Ijara 85-100% ASAL Kitui, Makueni, Tana River, Taitataveta, Kajiado, Samburu 50-85% ASAL Machakos, Mwingi, Mbeere, Tharaka, Laikipia, West Pokot, Kwale, Kilifi, Baringo, Meru North 30-50% ASAL Lamu, Narok, Transmara, Malindi, Keiyo, Marakwet 10-25% ASAL Nyeri (Kieni), Rachuonyo, Suba, Kuria, Thika, Koibatek Source: Republic of Kenya (1992) Arid and Semi- Arid Lands (commonly known as ‘ASALs’ in Kenya) have never enjoyed the kind of political attention which takes into account their unique potential and the challenges faced by the people living in these regions. The dominant system of production in much of the region is pastoralism which has been widely misunderstood by governments as these areas differ from the rest of the country in several aspects (Njoka, 2016). The communities living in these regions are usually mobile pastoralists, especially in arid areas. However, both arid and semi- arid regions experience chronic food insecurity, degraded ecosystems and climate change. Their demography is characterized by low population density and high population growth, in contrast to having the lowest development indicators and high levels of poverty. These factors along with being a repository of invaluable indigenous knowledge makes them different from the other parts of the country and yet, they are hardly accommodated in the national narrative. (Elmi, Birch, 2013). In this context, it is important to understand the barriers that limit the access to health care facilities for pastoralists in order to inculcate them into the broader policy discourse for achieving positive outcomes on maternal health. 14 1.3 The Pastoral Context 1.3.1 Who are Pastoralists? Constructing a definition for marginalized groups is critical in shaping policies around their demographics and livelihood patterns. Aspects of ethnicity and socio- cultural arrangements, production forms, degree of mobility or sedentarization, dependence on livestock for food or income, geographical location, degree of engagement with the markets are all factors that add to the definition of a particular indigenous community. However, it has been observed that governments do not have an unvaried approach to defining pastoralists (Humanitarian Policy Group, 2010). It is important to note that in Kenya, ASAL districts are clearly demarcated and yet are not officially labelled as pastoralist regions. This unclear and non- uniform approach can affect policy outcomes in negative ways (Ibid.). Kenya’s political drive towards decentralization of governance has led to the implementation of health care approaches that are moving beyond resource allocation and health services to determinants such as income and infrastructure to ensure better access to the governmental health systems. However, this move alone, without political improvement of the situation of a marginalized population, will not have the positive health impact that is expected (Schelling, Weibel, Bonfoh, 2008). Even with efficient health service facilities in pastoral zones, significant barriers to service delivery will continue to exist. States such as Kenya, Tanzania and other governments in the horn of Africa, in their attempt to promote settlement and privatization have been responsible for the breakdown of their traditional systems and for neglecting them in economic development programs and establishment of public services (Pratt et.al, 1997; World Initiative for Sustainable Pastoralism, 2007). In order to understand the determinants of increased maternal mortality among pastoral communities, one needs to first delve into the background of pastoralists and the challenges they face. Some key factors are discussed below: 1.3.2 Poverty Pastoralists make up a significant proportion of the livestock-reliant population in Kenya. A census conducted in 2009 indicated that ASAL citizens made up 36% of the population - 12% in arid counties and 24% in semi-arid counties in a country home to 4 million pastoralists (Oxfam Briefing Paper, 2008). A report on one of the pastoral counties referred to as Turkana 15 stated that 64% of its population was reliant on pastoralism and another 16% on agropastoralism. However, these regions are in fact reflecting the country’s poorest welfare indicators characterized by weak infrastructure, widespread insecurity, frequent droughts and limited livelihood options. The Human Development Index of the former North Eastern Province stands at 0.417 compared to the national average of 0.562 and Central Province’s 0.624 (UNDP, 2010). These HDI disparities clearly highlight the development deficit faced by pastoral regions, which in turn comprise a significant portion of the country’s demographics. 1.3.3 Geographical Dispersion and Availability of Data The geographical dispersion of mobile pastoralists and their spatial relationship with urban settlements and service delivery facilities are significant factors in limiting their access to social services (Schelling, Weibel, Bonfoh, 2008). In Kenya’s Marsabit district, the average distance to the nearest health facility is 60 km. The other pastoralist divisons like Loiyangalani and Laisami are reported to have the longest distances to local health facilities. This problem is further exacerbated by factors such as lack of transportation and time pressure. In order to transfer the critically ill, pastoralists have to resort to walking to reach the facilities. It is not unusual for sufferers to experience death on their way to the service points (Duba, MurVeeman, Raak, 2001). Furthermore, official data on pastoralist populations is poor and inconsistent in terms of broader national and regional trends. Accurate statistics are often difficult to obtain not only due to cultural reasons but more importantly, due to how pastoralists are defined by the governments. This leads to pastoral areas being omitted from national censuses at various official and bureaucratic levels. Until 2003, the Demography and Housing Survey in Kenya did not cover Northern Kenya, which is a hub of the most number of pastoralist districts (Humanitarian Policy Group, 2010). This shows that the failure to collect statistics only amplifies the larger problem of exclusion of pastoralist communities in government policies (discussed further in Chapter Three). 1.3.4 Discrimination in Service Provision Despite the lack of demographic and other statistics for pastoral areas, there are studies that show that the pastoralist areas have the highest incidences of poverty and the least access to any basic services in the country, especially in the Northern pastoralist districts (Oxfam Briefinf paper, 2008). The staff of service facilities usually lack proper training and exposure 16 to different cultural settings and therefore, when there is a shortage of drugs and vaccines at a health facility, mobile pastoralists are likely to be excluded from treatment or vaccination. (Schelling, Weibel, Bonfoh, 2008). This is one of the key factors that deter the pastoralists from accessing services from the fear of not receiving proper care and treatment at the facilities, eventually leading to a sense of mistrust between them and the service providers. Traditional birth practices are therefore preferred over formal healthcare delivery services, which are available easily and are relied upon heavily (Ibid.). In Kenya, traditional birth attendants are untrained midwives who help women deliver at home, usually in resource limited settings, when the women do not have access to skilled midwives in formal health facilities. Their activities have been discouraged by governments for several years owing to widely circulated reports of TBAs using outdated and potentially harmful practices in terms of obstetric complications and newborn care (Bucher et. al, 2016). TBAs are known to practice several risky methods including the use of poorly chosen instruments at the time of delivery, not referring anemic patients or those suffering from antepartum hemorrhage to antenatal care, poor handling of the chord during birth, lack of sterility and asepsis, delay or lack of referrals in a case of prolonged labour, and use of external cephalic version without the knowledge of contraindications (Solomon, Rogo, 1989). In many cases, service provision is also used as a political tool to influence pastoralists. Provision of basic services are typically provided by emergency and relief aid agencies during seasonal crisis instead of adequately addressing the underlying causes that perpetuate their perils. While aid was meant to be temporary and short-term, it is transformed into a permanent solution for reaching out to the pastoralists due to the lack of clear strategies, and an incomplete understanding of pastoral livelihoods. This phenomenon is referred to as ‘aid dependency’, where pastoralist households depend on the provision aid for their survival (Humanitarian Policy Group, 2010). 1.3.4 Mobility Pastoralist societies move across rangelands using a combination of herd management and camp mobilization strategies. Their households either establish a base at a particular rangeland or move seasonally or they are on a move continuously. The use of their rangelands is organized according to the access and availability of resources, seasonal variations, and climatic disruptions caused by both natural and man-made events (Reid, Fernandez-Gimenez, Galvin, 17 2014). This pattern of livelihood needs to be inculcated in any development strategy pertaining to pastoralists. However, in the majority of the development reports and evaluations, this aspect is only mentioned fleetingly in relation to service provision (see Chapter Two for more on policy and pastoralism). Mobility as a barrier in accessing health and other social services is hardly pondered on in government strategies and interventions (Humanitarian Policy Group, 2010). 1.3.5 Gender Disparity Gender disparity, which is often deeply rooted in pastoral societies, is an important determinant of access to services, primarily health services. This is because access to external resources, credit, health care services, land and livestock ownership, treatments and knowledge of urban practices are mostly controlled by men. It is also evidenced that households headed by females are more likely to fall into poverty than households headed by men (Eneyew, Mengistu, 2013). Apart from these challenges and burdens on pastoral women, their access to services lying outside of their community structure, often depend on factors such as the social support system in their household and the networks within their communities. In terms of health, this social support system refers particularly to the male members of the family, whom they have to rely on for their treatment and care. This is due to fact that in most patriarchal pastoral societies, women are not allowed to visit health centers or other public facilities without the company or permission of their male counterparts such as fathers, husbands or brothers. This kind of gender disparity is an important aspect that needs to be factored in by governments while designing relative interventions (Schelling, Weibel, Bonfoh, 2008). Pastoral societies exhibit many of the above features of poverty and deprivation in Kenya, living in the most remote and impoverished geographical areas. There is also little reference to pastoralism in relation to mainstream policies in Kenya. The next chapter looks more specifically at this relationship between policies and pastoralism in Kenya and the implications for maternal health of pastoralist women. It talks about the vulnerability that pastoralists face in terms of misrepresentation in political contexts and how an inclusive social protection framework can help make policies more effective in reaching out to the pastoral populations. 18 Chapter 2 Inclusive Social Protection 2.1 Pastoralists’ Vulnerability: A Political Consequence? As discussed in the previous chapter, pastoral communities have generally been excluded from state policies and strategies with policies being insufficiently tailored to their needs (Ogachi, 2011). Ogachi argues that governments often use the difficulty of providing services to nomadic pastoralists as an excuse to encourage sedentary settlements. Their marginalization by national governments over the years, stems from the fact that the mainstream society views them as having a fundamentally flawed way of life (Ali, Hobson, 2009). This form of misrepresentation of pastoral communities plays a pervasive role in the generation of unintended policy outcomes. Although many formal institutions in Kenya recognize the importance of including pastoralist communities in development ventures, this recognition remains more in theory than in practice (Ibid.). Policy makers have long been disapproving of the pastoralist way of life. Critical aspects of pastoralism- particularly, their mobility and reliance on indigenous knowledge systems are seen as archaic and backward in comparison to the modern settings (Technical Centre for Agricultural and Rural Cooperation, 2012). This is a result of several interrelated factors. Apart from the larger problem of the lack of desire among policy makers to listen to the pastoralists, their (pastoralists) absence from regional and national policy processes is, also, due to their inability to articulate the rationale of their livelihood and to organize themselves to influence the power dynamics which can be explained by the fact that these communities tend to be in minority, often thriving in geographically remote areas, away from urban settlements usually being the centers of economic and political activity (Overseas Development Institute, 2009). These factors contribute to them being excluded from the dominant narrative surrounding policy interventions in African countries. Furthermore, formal delivery systems which do not pay attention to how specific pastoralist contexts may shape the policies, in a sense, undermine pastoralists and their communities by replacing their traditional coping strategies and weaken their community’s social fabric only to increase their perils (Ali, Hobson, 2009). 19 Social protection in health has been regarded as a strategy that ensures all people have access to essential care to alleviate the burden posed by ill health including its social and economic impact such as death, loss of income, lower productivity, short term or long term disability and hampered social development and education of children. It had been defined by the international Labor organization (ILO) as a series of measures against social distress and economic loss by the reduction of productivity, income or costs of treatment due to ill health. A combination of various financing and organizational mechanism can help to provide benefit packages to protect people from ill health and its unintended consequences (International Labor Organization, 2007). However, the majority of formal social health protection policies focus on the provision of services which are essentially designed for sedentary populations, while the concept of providing state assistance and equity in pastoral communities is virtually nonexistent (Ali, Hobson, 2009). In this context, I argue for a different approach to social health protection for pastoralist communities by applying a framework of inclusion (Roelen, Devereux, 2013) to formal maternal health interventions. However, it is first important to understand the importance of social protection in maternal health interventions. 2.2 The Social Protection Dividend in Maternal Health The Overseas Development Institute (ODI) defines Social Protection as a set of ‘public actions that should be taken in response to levels of vulnerability, risk and deprivation which are deemed socially unacceptable within a given polity or society’ (Norton, Conway, Foster, 2001, p. 7). This indicates that social protection can be introduced across a diverse range of contexts, to achieve positive policy impacts on a range of well-being outcomes, be it increased income and consumption to increase in access to education or health care (Roelen, Devereux, 2013). Social protection in providing maternal health services has been regarded all over the world as an essential prerequisite for achieving gender equality and championing women’s rights (Gacheri, 2016). Ensuring effective access to quality maternal healthcare is a key element of national strategies to promote development and socio- political stability. Premature deaths of women, regardless of the age at which they occur, have huge socio economic effects on their families, especially if they are female headed. Apart from the loss of a caregiver, families are often driven to destitution, poverty, income loss and huge health expenditures. An integrated approach to maternal health using existing infrastructure, financing, and cost effective interventions need to be implemented in order to address the burden of maternal mortality globally (Knaul et. al, 2016). Article 43(1)(a) in the Constitution of Kenya gives every person 20 the right to the highest attainable standard of health, including the right to health care services, and reproductive health care. It also states that a person has the right to treatment in case of an emergency (Article 43(2)). In addition to this, it provides for a decentralized system of governance which has ratified county governments with the responsibility of providing healthcare services, hence providing a window to address historical inequities in access to health services (Gacheri, 2016). Despite the existing political support and an enabling policy environment for maternal health, access to quality maternal health services continues to be a challenge for pastoral communities. Moreover, public expenditure on health in Kenya accounts for only 6% of the total government expenditure (Gacheri, 2016). Affordability of health services in terms of alleviating out of pocket payments4 among vulnerable groups also remains one of the most significant obstacles. Ensuring access to maternal healthcare thus becomes a critical component of social protection. Public interventions directed at allowing mothers to meet their health needs and demands through access to health care goods and services in adequate condition should therefore, be based on a few important parameters ((Pan American Health Organization, 2008). These include: • The ability to increase equitable access to care or the utilization of health services; • Offset social determinants that affect the health status or the demand for health care such as in the case of pastoralists (long distances from service delivery points, traditional birth practices, gender disparities, knowledge gap) and finally; • Eliminate exclusion of vulnerable groups from the designed health care policies (PAHO, 2008). In order to address the health financing gaps and to improve service coverage, especially among vulnerable populations, several countries have increasingly started considering the implementation of a variety of social health protection mechanisms. Alternative health financing mechanisms like the free provision of tax-funded national health services, cash transfer schemes, Social Health Insurance and regulated private- or community-based not-forprofit health insurance schemes are beginning to see the light of the day (United Nations Children’s Fund, 2009). However, the assumption that a generic approach will work for 4 According to the OECD glossary of statistical terms, out of pocket payments refer to the expenses of medical care that are not reimbursed by insurance due to cost sharing, self-medication or other expenditures paid directly by the user regardless of whether the services are covered or not by the health care system. 21 everyone results in a failure to adapt to the needs of vulnerable communities like the pastoralists who’s needs differ from the rest of the population (Tanner, 2005). 2.3 Adding the Element of Inclusion Social Protection has begun to be conceptualized in a more political and transformative sense rather than just confining its scope to targeted income and cash transfers. (Devereux, Wheeler, 2004). In the early 1900s and 2000s, the objectives of implementing social protection programs were limited to ‘safety nets’ during crisis and the reduction of poverty. More recently, this concept has been borrowed to achieve multiple development initiatives such as those articulated in the Sustainable Development Goals (Cluver et. al, 2016). However, social protection as a stand-alone program cannot achieve the objective of an integrated development system unless it ensures that it reaches everyone who needs it. In this context, the introduction of an inclusive approach to social protection can ensure that the policies do not target a fortunate few for a fixed period of time as well as recognize the fact that drivers of vulnerability are more structural and institutional rather than individual (Roelen, Devereux, 2013). Inclusive social protection is a response to different dimensions of exclusion and vulnerability such as gender, geographic dispersion, discrimination in service delivery, traditional social norms preventing usage of services, limited economic assets, knowledge gap, etc., to explicitly promote social inclusion and equity while designing and implementing programs (UNICEF, 2012, p. 79). In order to understand the element of inclusion, we need to first analyze these dimensions of exclusion (discussed in detail in Chapter One) that operate within a pastoralist setting (Kipuri, Ridgewell, 2008). Instruments that specifically address these dimensions in accessing services and securing adequate standards of care are imperative in mainstreaming inclusion in the design and implementation of maternal healthcare policies. UNICEF’s framework of integrated social protection highlights two ways in which social protection policies can further social inclusion to remove barriers of discrimination and vulnerability. The first tool is to use policy instruments that directly aim to eliminate discrimination and inequities. Lessons can be drawn from the Red de Oportunidades conditional cash transfer program which was introduced in 2006 in Panama to reach the 10 percent of the population that identified as indigenous. Majority of the people who resided in Panama’s indigenous reserves lived in utmost poverty (UNICEF, 2012, p. 84). Thereby, all those families were automatically directed into the ambit of the cash transfers program without 22 having to undergo the rigorous proxy means test to determine their eligibility. To add to this, the indigenous families were consulted regarding the program to figure out the most effective and culturally sensitized way to disburse the cash transfers and provide health information to co inhabitants of the area. Local community liaison officers were delegated to provide assistance and support as well as information on nutrition and health practices to the families (Ibid.). Even when social protection interventions do not directly address inequities, they can be tailored to specific vulnerabilities or impacts of exclusion on communities. An inclusive design should be able to identify the appropriate building blocks towards integrating different contexts as well as the most appropriate sequence of interventions needed to support different groups (UNICEF, 2012, p. 85). Sustainable mechanisms of financing and service delivery should be applied to address the dimensions of vulnerability in social protection programs (WHO, 2011). Increasing the availability of disaggregated data in terms of livelihoods to inform and monitor programs is also an important method to create more inclusive social protection interventions (Medicus Mundi International Network, 2013). The discourse on inclusive social protection and expansion of public service delivery systems has been on a global level, centered around policy and technical aspects. However, it’s implementation on a country or regional level is also majorly, if not completely, influenced by its local political context (Stuckler et. al, 2010). It is critical to understand and evaluate the political forces at play including the role of donor agencies and international organizations in advocating for sustainable commitments to these programs. For example, countries with long histories of implementing welfare mechanisms and protective policies may be more inclined towards advocating universal access to healthcare, whereas countries which have a strong donor presence may be more favorable towards targeted interventions vis-a-vis universal coverage (UNICEF, 2012, p.74). 2.4 The Influence of State- Citizen Dynamics The dynamics shared between the state and the communities in question, particularly vulnerable groups like the pastoralists, are often reflected in policies as well as their implementation and evaluation. It matters whether social protection is seen as a right or as a benefit for pastoralists; as a short-term intervention to sedentarize them or a long-term strategy to restore their livelihoods. Decision-makers and their interests also determine social protection 23 agendas in terms of public expenditures and financing of programs. This often leads to welfare programs being implemented for political survival rather than to uplift he needy sections of the population (Botlhale et. al, 2015). Political instability owing to pressure created from crossparty politics, or elections can influence social protection programs in both positive and negative ways. Moreover, program delivery and implementation can also be weakened when decentralized at lower levels of decision making, irrespective of a strong political will at the center (UNICEF, 2012, p. 74). It is therefore, imperative to prevent political misuse of the intervention in terms fund mismanagement for political interests and increase its sustainability for the recipients. In the above context, the next chapter is a discussion on Kenya’s recently introduced Free Maternity Services (FMS) program which aims at the reduction of infant and maternal mortality and strengthening of delivery systems by providing free access to maternity services in all public health facilities (Mwaura, 2013). I analyze this policy through the lens of Inclusive Social Protection to determine which of its aspects need to be revised in order to make this policy more inclusive towards pastoralists. 24 Chapter 3 Discussion 3.1 Case Study: Kenya’s Free Maternity Services Program The Government of Kenya in 2013 launched a key intervention in Maternal and Child Health to increase access to skilled delivery services and reduce infant and maternal mortality. The President of the Republic of Kenya in his declaration on 1st June 2013, promulgated the Free Maternity Services Program (Ministry of Health, 2015), which declared maternity services free in all public health facilities in effect from that date to help Kenya in moving towards achieving the Millennium Development Goal (MDG) 4 & 5. Although a move too late to address the MDG targets, the Free Maternity Services (FMS) was hailed as a major stride on the road to Universal Health Coverage which is a part of Kenya’s Health Sector Strategic Plan (KHSSP) 2014-2018 (Ibid.). This section aims to look at the impact of this intervention to document areas that require an inclusive approach in terms of policy and delivery mechanisms to include the needs of pastoral communities. The efficiency and effectiveness of an inclusive social protection program depends on a few basic factors. Devereux and Roelen (2015) formulated three key principles for the post 2015 agenda that would ensure that future social protection programs are grounded in terms of rights, equity and sustainability. According to them, a program can be designed in a truly inclusive manner if, firstly, it reaches everyone that needs to be reached; secondly, it does not place unrealistic expectations and responsibilities on those individuals seeking care to overcome barriers or constraints over which they have little or no control; and thirdly, it ensures that the inclusive policies are well integrated with other schemes that work on vulnerability at the structural level (Ibid.). The need to involve all stakeholders into the program: the mothers or the end users, the facility-based care workers, the health management teams, governance teams, the national government and the community at large, is also paramount (Ministry of Health, 2015). In this chapter, I evaluate the findings of four studies to present a set of challenges that need to be tackled in order to make the Free Maternity Services program more effective and inclusive towards pastoralists. Study One was conducted at formal health facilities in five group ranches 25 in Laikipia County and three group ranches in Samburu County using in-depth interviews and focused group discussions with Skilled birth attendants, Community health workers and key informants such as relatives and clients (Caulfield et. al, 2016). Study Two was a report on the monitoring of the Free Maternity Service Program conducted by the Ministry of Health and Consultants from the World Health Organization and evidence was gathered from a large team of sampled staff, facility administrators, and key informants- mothers (Ministry of Health, 2015). Study Three was a cross-sectional descriptive study carried at the Rift Valley (pastoral region) Provincial General Hospital and Bondeni maternity to assess the implementation of the program using questionnaires and interview guides to collect data from the medical staff as well as the superintendent of the facility (Wamalwa, 2015). Finally, Study Four was conducted in several county referral hospitals as well as low-cost private hospitals to compare their deliveries and antenatal attendance. The data was extracted from Kenya’s Health Information System (Njuguna, Kamau, Muruka, 2017). I have chosen these four pieces of work because they were conducted after the implementation of the Free Maternity Services Program. One out of them (Study Two) is a government report and the others are independent studies. I chose to inculcate three independent and one government report in order to understand the perceptions of FMS among different stakeholders. It is also important to note that Study One, which is not directly analyzing the impact of the FMS program, is still relevant in this context as it was conducted within specific pastoral counties and gives a very nuanced idea of the challenges that pastoralists face with regard to health care. These studies were used to identify some key challenges, namely Quality of Care, Accessibility, Lack of understanding of FMS among users, Out of pocket expenditures and Lack of Motivation to implement Policy. Each of these is now discussed in turn below, highlighting what the key issue is and providing recommendations. 3.2 Challenges and Recommendations 3.2.1 Quality of Care Assessment of quality of care by the end users although subjective, can provide useful inputs to help the government and the providers to understand and formulate acceptable standards of services (Ministry of health, 2015). Study Two which was conducted in the health facilities 26 concluded that the management of service delivery may not have been satisfactory. It was observed that a majority of the staff members were tasked to work in multiple departments whenever they were on duty, especially in the lower level facilities. Despite the increased workload, there was minimal provision of mentorship for staff as was reported by key informants in the study (Ibid.). The study indicated that the informants, who were among a total of 603 mothers interviewed about the perceptions of FMS, in 24 counties from Rift Valley, Central, Coast, Eastern and Nyanza regions, also complained about the long waiting time at the health facilities. This was attributed to the increase in the number of mothers attending the Antenatal Care Services with inadequate numbers of staff to serve the mothers. One key informant observed that with introduction of Free Maternity Service, there had been an increase in utilization of services by up to 26% increase in deliveries, 22% increase in caesarean section and 50% increase in Antenal care (ANC) visits5, which indicated that there had been a drastic increase of workload in that respective health facility. However, the functionality of various amenities and equipment for maternal and neonatal care were found inadequate especially at the lower level health facilities (Njuguna, Kamau, Muruka, 2017; Ministry of health, 2015). For example, about 50% of the available ambulances were not functioning at the time when the survey was conducted, basic amenities such as toilets and bathrooms were not in use and most of the facilities did not have the provision of running water in the maternities. Non-functioning equipment was more commonly found at the lower level health facilities. The Twenty-four Hours Maternity in-patient services were offered by 82% of the facilities, but most of the facilities that did not offer twenty-four-hour maternity in patient services were level 2 (lower level) facilities (Ministry of Health, 2015) which are usually located in rural or areas relatively closer to pastoral zones (Schelling, Weibel, Bonfoh, 2008). Study One which was conducted among the pastoralists living in the Laikipia and Samburu counties documented the inadequate interpersonal skills of the staff. Many community respondents highlighted the negative attitudes and low quality services provided by the SBAs (skilled birth attendants) that deterred them from going to the health facilities (Caulfield et. al, 2016). Most women complained about hearing about or directly being subjected to verbal or physical abuse in health facilities. These experiences evidently played a huge role in 5 Despite the increase in the utilization of the services, the study does not mention if there was an increase in utilization on behalf of the pastoralist communities living in those regions. 27 reinforcing adherence to traditional birth practices instead of formal health services leading to pastoral women viewing TBAs (Traditional Birth Attendants) in a more positive light (Schelling, Weibel, Bonfoh, 2008). Increasing workload and inadequate staffing at the lower level, could, quite apart from a general disconnect with the pastoral population, be a significant reason behind the dismissive attitude and unfriendly treatment being meted out these women (Njuguna, Kamau, Muruka, 2017). Study One suggest that one of the many reasons why pastoral women are deterred from attending formal health facilities is because they are frequently left alone during the early stages of facility-based deliveries, a practice which is perceived as a very uncaring attitude on behalf of the SBAs (Caulfield et. al, 2016). This makes the health facility appear as an unpleasant place in which to deliver. In pastoral birthing practices, it is customary for women to be held by other women throughout the delivery in order to provide support and comfort to the mother. This is based on community members’ belief that a warm environment ensures a speedy delivery and protects the child and the mother from any kind of illness. TBAs provide vital support to the pastoralist mothers throughout pregnancy, childbirth and the postpartum period and are considered the backbone of the communities in terms of health and reproductive care. (Maro et. al, 2012). As is evidenced by these findings, there is a need to ensure that the services at the health facilities meet the recommended standards and quality. The inadequacies at the service facilities affect the quality of care (Ministry of Health, 2015). The Ministry of Health in Kenya and other stakeholders need to ensure that they put in place the required inputs so as to address the high service demand and balance this by strengthening the standards of health systems to maintain quality of care (Caulfield et. al, 2016). The engagement of various stakeholders needs to be increased in order to oversee the implementation and monitoring of the program. This needs a specific instruction to health system workers to pay attention to the needs of the pastoralists and to treat them properly. Integrating culturally responsive strategies tailored to the needs of women in pastoral areas, such as allowing relatives to support women during deliveries, allowing women to choose their own preferred delivery positions, and ensuring warm and friendly health facility environments are important measures that can be adopted to make services more acceptable and responsive to these women (Bedford et. al, 2013). Incorporating TBAs into the delivery system may also help in increasing the utilization of formal health services by pastoral women (Maro et. al, 2012). It has been established that 28 pastoralist women highly value TBAs over skilled birth attendants due to their ready availability and a mutual sense of trust. Integrating them in the system by imparting them with proper training, and skilled birth practices would not only ensure a culturally sensitized environment in the health facilities, but will also establish a long term relationship between the health facility and the pastoral mothers (Kitui et. al, 2017). The TBAs can also be recruited by healthcare facilities to educate pregnant women about the importance of delivering in health facilities as in the case of the Yatta district of Kenya, where TBAs were offered a small stipend for each pregnant woman they brought to a facility for a Skilled Birth Attendant delivery. The outcome of this initiative was that the rate of skilled birth deliveries increased in health facilities as the TBAs established a link between the facilities and the women living in marginalized communities who earlier were inhibited to seek formal health care (Tomedi, Tucker, Mwanthi, 2013). 3.2.2 Accessibility Despite the government making maternity services freely accessible, it was estimated that 39% of pregnant women were still not delivering in a health facility (Kenya National Bureau of Statistics, ICF International, 2015). Facility-based delivery is influenced by not only cost (Gabrysch, Campbell, 2009) but also various other factors. Physical accessibility, maternal education, parity, rural-urban residence, gender disparities, household income, distance from the residence to the health facility are also strong factors that influence the uptake of formal maternal health services (Moyer, Mustafa, 2013). Study Four pointed out that only 11% of women in Kenya cited cost as a reason for not delivering in a formal health facility. Study Four indicated that there was an increase in the number of deliveries by 22% in the financial year 2013-2014 (Njuguna, Kamau, Muruka, 2017), however in comparison antenatal attendance declined in private facilities. However, the increase in the number of deliveries can be attributed to the population size of the counties, as stated in Study Four, counties with higher population had higher number of women of reproductive age and hence, the increase in the number of deliveries (Ibid.). It was also observed that higher level facilities registered more deliveries than lower level facilities indicating that pastoralists, who would ideally have access only to lower level facilities may not have delivered in the health facilities (Schelling, Weibel, Bonfoh, 2008). 29 Distance, cost, lack of transportation and poor quality of roads often influence the decision to seek formal services among pastoralists (Bedforrd et. al, 2013; Ng’anjo Phiri et. al, 2014). These are not just barriers but a disincentive to seeking healthcare (Thaddeus, Maine, 1994). Unable to walk the long distance to a health clinic in a pregnant state, majority of women deliver at home with the assistance of a Traditional Birth Attendant (TBA) who are highly relied on by these communities as they are almost always available to assist them (Caulfield et. al, 2016). In addition to the problem of long distances from health facilities, is the lack of availability of transport, cited by 42% of the respondents, in areas where these communities live. In order to travel to formal health facilities, the families have to hire vehicles which in most cases are unavailable when needed as indicated by the Kenyan Demographic and Health Survey (Kenya National Bureau of Statistics, ICF Macro, 2010). There is a strong need for better targeting of geographically isolated populations in the context of designing Health Service Delivery programs (UNICEF, 2012, p. 85). The geographic location of vulnerable indigenous communities needs to be taken into account to optimize the use of facilities among these groups. Many social transfer programs, such as FMS which is discussed in more detail below, are established in areas with ample supply of services while these same facilities are scarce or non-existent in remote areas with low demographic density. Improving ambulance services in such areas can help in increasing the number of women delivering in facilities (Sipsma et. al, 2013). Sufficient fuel provision for the ambulance; access to smaller health posts on its route; sufficient number of ambulances to serve the needs of the community; and passable roads are some of the measures that need to be taken to ensure success of this strategy (Caulfield et. al, 2016). In order to establish a link between the health facilities and the pastoral areas, a combined static and mobile outreach intervention can be successful in creating an integrated system of health service delivery. According to Aliou (Aliou, 1992), it is possible to organize static health services which can cater to pastoralists by having mobile clinics6 which match the mobility of the community they serve. In accordance with the pastoralists’ patterns of movement, seasonal circuits can be established corresponding to the geographical locations of these communities (UNICEF, 2012, p. 85). One or more of these areas should be served by an intermediate fixed/ 6 Mobile Clinics are vehicles that are customized to travel to the heart of rural areas and communities where indigenous populations thrive to provide prevention and healthcare services, in this case, maternal health services to overcome the barriers of accessibility. These clinics are specifically tailored to suit the needs of vulnerable populations. 30 static health facility. This system of integrated fixed- and mobile-facilities can be carried out in each operational area on the basis of a flexible schedule that can be adjusted according to the seasonal movements of the pastoral communities. This method will not only produce greater coverage but will also be more sustainable in terms of future referrals and easy identification Aliou, 1992). 3.2.3 Lack of Understanding of FMS among Users The scope of the Free Maternity Service program was found to be inadequate in terms of its coverage among the users (Ministry of Health, 2015). The FMS program which was initiated before the two ministries of Medical Services and Public health and Sanitation was a strategy to enhance the access of maternal services to the poor. According to the circular generated by the Ministry of health in Kenya, it covered several services including, Antenatal Care, Deliveries, Postnatal Care and other complications of delivery including Intensive Care Unit (ICU) care, renal dialysis and pregnancy related medical diseases. This circular also communicated the reimbursement for health facilities based on their performances. Despite this communication, key informant interviews from Study Two reported that the mothers had different understandings of what the FMS program offered (Ibid.). This indicates that the scope of FMS was not defined properly to increase clarity about its interventions among the people who accessed its services. Study One respondents in pastoral communities suggested that a lack of awareness among women about the facility-based services contributed to its low uptake. Even in areas where the Community Health Workers (CHWs) conducted awareness raising campaigns regarding the Free Maternity program, it was reported that only better educated or younger women utilized the services of Skilled Birth Attendants (Caulfield et. al, 2016). Limited educational opportunities in pastoralist areas, is identified as one of key obstacles in women receiving formal healthcare. Consultation and information sessions with the leaders of pastoral communities as well as short term health information programs specifically targeted to these communities can play a huge role in increasing the understanding and knowledge of women with little or no education, regarding the benefits of this program (Mekonnen, Mekonnen 2003; Wilunda et. al, 2014). Educating men and women about the benefits of delivering at health facilities will equip them in making informed decisions about pregnancy and delivery and can have a significant impact on maternal mortality. Not just this, but the government also has to take efforts to overcome the lack of trust that pastoralists have of the formal health system, by treating them with respect 31 (and pastoralists’ notion of respect might differ from that of the health system staff). Involvement of indigenous community leaders in the planning, implementation and evaluation processes of the program will not only increase accountability but will also help in improving maternal health outcomes among pastoralists (UNICEF, 2012, p. 85). 3.2.4 Out of Pocket Payments and Disaggregated Data As highlighted earlier, cost is an important factor in accessing healthcare in Kenya’s pastoral regions. Although FMS services are free, a third (28%) of the respondents in Study Two reported to have paid some fee for various health care services that ranged from laboratory tests, drugs, registration, x-ray services and delivery charges (Ministry of Health, 2015). The cost of transport was also prohibitive in accessing the health facility for majority of the respondents as the cost of hiring a local vehicle ranged from KSH1500 up to KSH4000 (USD 16–42), depending on the distance between the pastoral areas (ranch) and the facilities (Caulfield et. al, 2016). It is observed that in Kenya, women who belong to wealthier households and women who are covered with some form of health insurance are more likely to deliver in a facility based service compared to women belonging to poorer households and those without insurance (Kitui, Lewis, Davey, 2008) It was also noted that patient records in the health facilities are not adequately documented with post-natal records being the most neglected (Ministry of Health, 2015). In one of the surveys conducted in Study Two, it was recorded that out of a total 301 maternal deaths that occurred in a county health facility, only 132 were audited. In some cases, the facilities did not have any evidence of maternal death audits despite having recorded some maternal deaths (Ibid.). Thereby, defining a more comprehensive, institutionalized system for financing the Free Maternity Services by exploring the scope of health insurance mechanisms (Gadeberg, 2016) among pastoral communities will be a crucial step towards creating equity in financing the program. The FMS policies should integrate efforts to reduce the pastoralists’ financial barriers to accessing health services. The integration of disaggregated data in terms of livelihoods will not only help in keeping a record of the users and but will also enhance the evaluation process of the program (UNICEF, 2012, Chapter VI). 3.2.5 Lack of Motivation to implement Policy Apart from the issue of inadequate staffing at the health facilities, Study Three reported that majority of the respondents (62%) were not motivated to implement the free maternity services 32 policy in their health facilities (Wamalwa, 2015). Failure to boost workforce during implementation of such a policy apart from the fact that the staff was overworked could be one of the reasons that lead to reduced morale and motivation. This could also be a result of the irregular reimbursements that were being channeled through the central county revenue accounts to the health facilities (Ministry of Health, 2015). This had made the health service facilities to have no control of the Free Maternity Service Funds. Many of the facilities reported irregular or no reimbursements from the government since the introduction of the policy whereas some were not informed at all on their reimbursements (Ibid.). This challenge has greatly affected the quality of care at these facilities. The inadequate flow of funding is said to create friction not only between the communities who are seeking care and the health staff but also between the managers of these facilities and higher levels of the maternity program systems which can eventually lead to failure of the program failure or can force health staff to demand out- of- pocket payments from users for the ostensibly-free health services that they want to avail (Wamalwa, 2015). It is therefore, imperative to boost the motivation of health workers through several implementation strategies like reduction in number of working hours, improving the overall work environment and providing the health workers with regular incentives. Training opportunities in maternal health services for staff to enhance their skills and recruitment of more staff can also help improve the service quality and outcomes (Wamalwa, 2015). It is also very important on behalf of the government to ensure regular release of reimbursement funds, provision of more logistics and supplies, and regular supervision of the health facility (Witter, Anthony, Aikins, 2007) to improve the overall management of the program. In this regard, it is advised that the Ministry of Health in tandem with the County Governments should come up with initiatives to regularly evaluate the disbursement mechanisms for successful implementation of the policy. 33 Conclusion Improving the rate of maternal mortality has been a global imperative with the emergence of several international interventions including the Millennium Development Goals which have been significant in shaping up maternal health strategies across the world. Despite efforts, the progress in reducing maternal mortality in Kenya has been inconsistent (Keats et. al, 2017). As I have discussed in the first section, this slow progress has been attributed to the fact that the national policy framework surrounding maternal health in Kenya has not been able to take into consideration, mechanisms to include marginalized populations within the ambit of its policy interventions (Abuya, 2014). Pastoralist communities, as evidenced by my findings, face a number of challenges in accessing social services, as promised by the governments. As in the case of maternal health programs, most of them are inappropriately designed and fail to offset the determinants that limit pastoralists’ accessibility to maternal healthcare (Schelling, Weibel, Bonfoh, 2008). Some of those factors include: poverty, the geographical dispersion of pastoralist communities and their spatial relationship with service delivery centers, mobility, lack of availability of disaggregated data on pastoralists, gender disparities within their communities’ social fabric and a general sense of disconnect between formal health systems and these communities, evident in the form of inadequate and low quality of services and disrespect meted out to the pastoralists. Throughout my dissertation, I have emphasized strongly on the aspect of pastoralists’ vulnerability which is a result of their misrepresentation in the dominant political narrative, owing to their cultural beliefs and practices being identified at odds with the modern setting (Humanitarian Policy Group, 2010). This as a consequence has translated into a general neglect in policy outcomes with interventions being designed for sedentary populations, especially in the context of maternal healthcare (Ali, Hobson, 2009). My argument stems from this lack of contextualization of policies to weigh in the application of an all-inclusive social protection framework for maternal health interventions in order to increase equitable access to services among pastoralist populations in Kenya (Roelen, Devereux, 2013). Maternal deaths have huge socio economic effects on families, who are driven to poverty, loss of income and destitution, especially if they are headed by a female (Knaul et. al, 2016). Social protection programs with an integrated approach to maternal health using existing financing 34 mechanisms, infrastructure, and cost effective interventions in the form of free, tax- funded services, health insurances, etc., are being formulated in order to address the burden of maternal mortality. (Gacheri, 2016). However, social protection as a stand- alone mechanism cannot achieve the objective of integrating marginalized populations within its framework (Ibid.). In this context, the application of Roelen and Devereux’s inclusive social protection framework takes into account the vulnerability of pastoralists and ensures that the responsibility of offsetting the above mentioned barriers or determinants are not placed upon them (Roelen, Devereux, 2013). My dissertation has identified two ways of applying an inclusive framework to social health protection policies (UNICEF, 2012). The first method is to directly target policy interventions towards vulnerable pastoralists by designing specific schemes or programs for this section of the population. The second method, and the one which I focus more upon, is integrating inclusive strategies within already existing interventions to reach out to them. The latter is a more sustainable and cost effective strategy which ensures equitable access to all sections of the population (Ibid.). However, I have also taken into account the importance of political will and international support in designing these strategies which is extremely essential in ensuring the effective implementation of any such initiatives. The Free Maternity Services program implemented in Kenya in 2013 is one such intervention that has the potential to be transformed into a more inclusive program for the pastoralists. The program which aims at targeting the poor and people who lack access to maternal services (Ministry of Health, 2015), is briefly analyzed in the final section of my dissertation. Irrespective of the fact that FMS was a well-intentioned policy intervention, I have observed that it failed to create a strong link between the health facilities and pastoralists which lead to its low uptake among this section of the society. Implementation challenges which limit the access or uptake of its services among pastoralists are discussed broadly within this section. In order to identify these bottlenecks, I have analyzed data from four main studies conducted after the implementation of FMS to compare the perception of the program among various stakeholder including the Ministry of Health (2015), pastoralists (Caulfield et. al, 2016) and other independent researchers (Wamalwa, 2015; Njuguna, Kamau, Muruka, 2017). Strengthening the standards of health care facilities by integrating culturally responsive practices among the staff (Bedford et. al, 2013) and increasing the quality of service provision in lower level health facilities (Caulfield et. al, 2016) has been identified as one of the important 35 steps in making FMS more inclusive for pastoralists. Training of the health facility workers to make them more receptive towards pastoralists, and recruiting the traditional birth attendants within the formal systems (in case of shortage of staff) by training them in safe birthing practices can be a solution to improving the quality of services offered to them (Maro et. al, 2012). I have also laid stress on the need to effectively target geographically isolated populations, in order to eliminate the barriers that they face in accessing formal health services (UNICEF, 2012). This can be done by increasing the supply of workforce, ambulances (Sipsma et. al, 2013) and other services in lower level facilities and incorporating new mixed methods service delivery like the combined provision of mobile and static clinics (Aliou, 1992) to reach out to the pastoralists. The scope of FMS needs to be defined properly in order to increase clarity about its interventions among these communities (Ministry of health, 2015). Increasing involvement of representatives from pastoralist communities to propagate the benefits of FMS and educating both men and women about the importance of safe maternal health practices is imperative for increasing an understanding of the intervention among all users (Mekonnen, 2003; Wilunda et. al, 2014). A more comprehensive financing mechanism (Gadeberg, 2016) to reduce the financial constraints in the access to services and an increase in the availability of disaggregated data (UNICEF, 2012) on indigenous populations are steps that need to be taken to augment the uptake of FMS among pastoralists. Lastly, strategies to boost the motivation of health facility workers in the form of better reimbursements to health facilities, reduction in the number of working hours, better work environment, better training and regular incentives will improve the overall management of the program, especially in lower level facilities located in rural areas (Wamalwa, 2015). The above recommendations are just some of the key elements of good practice that I was able to incorporate within the scope of my dissertation. There are a lot of other factors that may have been obscured in my research owing to my positionality. However, I have made all attempts to present the above information in a clear and concise manner. 36 Bibliography Abuya, T. (2014). Reproductive, Maternal, Newborn, and Child Health in a Devolved State: The Kenya Context. [ebook] Mobilising Advocates from Civil Society. Available at: http://www.familycareintl.org/UserFiles/File/MACS_Kenya_brief2014.pdf [Accessed 2 Jul. 2017]. Ailou, S. (1992) ‘What health system for nomadic populations?’, World Health Forum,13, pp. 311–314. Ali, A. and Hobson, M. (2009). Social protection in pastoral areas. [online] Humanitarian Policy Group, Overseas Development Institute. Available at: http://www.fao.org/fileadmin/user_upload/drought/docs/Social%20Protection.pdf [Accessed 15 Jun. 2017]. Bedford, J., Gandhi, M., Admassu, M., Girma, A. (2013) ‘A normal delivery takes place at home: a qualitative study of the location of childbirth in rural Ethiopia’. Matern Child Health J, pp. 230–239. Bohren, M., Vogel, J., Hunter, E., Lutsiv, O., Makh, S., Souza, J., Aguiar, C., Saraiva Coneglian, F., Diniz, A., Tunçalp, Ö., Javadi, D., Oladapo, O., Khosla, R., Hindin, M. and Gülmezoglu, A. (2015). The Mistreatment of Women during Childbirth in Health Facilities Globally: A Mixed-Methods Systematic Review. PLOS Medicine, 12(6). Botlhale, E., Mogopodi, L., Mothusi, B. and Motshegwa, B. (2015) ‘A Political Economy Analysis of Social Protection Programmes in Botswana’. Partnership for African Social and Governance Research Working Paper No. 001, Nairobi, Kenya. Bucher, S., Konana, O., Liechty, E., Garces, A., Gisore, P., Marete, I., Tenge, C., Shipala, E., Wright, L. and Esamai, F. (2016). Self-reported practices among traditional birth attendants surveyed in western Kenya: a descriptive study. BMC Pregnancy and Childbirth, 16(1). Callister, L. and Edwards, J. (2017). Sustainable Development Goals and the Ongoing Process of Reducing Maternal Mortality. Journal of Obstetric, Gynecologic & Neonatal Nursing, 46(3), pp.56-64. Caulfield, T., Onyo, P., Byrne, A., Nduba, J., Nyagero, J., Morgan, A. and Kermode, M. (2016). Factors influencing place of delivery for pastoralist women in Kenya: a qualitative study. BMC Women's Health, 16(1). Chou, D., Daelmans, B., Jolivet, R., Kinney, M. and Say, L. (2015). Ending preventable maternal and newborn mortality and stillbirths. BMJ, pp. 4255. 37 Cluver, L.D., Orkin, F.M., Meinck, F., Boyes, M.E., Yakubovich, A.R., Sherr, L. (2016) Can Social Protection Improve Sustainable Development Goals for Adolescent Health? PLoS ONE 11(10). Dong, S. (2016) ‘Overview: Pastoralism in the World’, Building Resilience of HumanNatural Systems of Pastoralism in the Developing World, pp 1-37. Duba, H., Mur-Veeman, I. and Van Raak, A. (2001). Pastoralist health care in Kenya. International Journal of Integrated Care, 1(1). Elmi, M. and Birch, I. (2017). Working Paper: Creating Policy Space for Pastoralism in Kenya. 68th ed. [ebook] Future Agricultures. Available at: http://www.fao.org/fileadmin/user_upload/drought/docs/Elmi%20Creating%20Policy%20Sp ace%20for%20Pastoralism%20in%20Kenya%20Agric%20Futures%20201309.pdf [Accessed 20 Aug. 2017]. Eneyew, A and Mengistu, S (2013) ‘Double Marginalized Livelihoods: Invisible Gender Inequality in Pastoral Societies’, Societies, 3, pp. 104–116 Gabrysch, S., Campbell, O. (2009) ‘Still too far to walk: Literature review of the determinants of delivery service use’, BMC Pregnancy Childbirth, 9:34. Gacheri, A. (2016) ‘Tackling High Maternal Deaths in Kenya’ Policy brief, Parliament of Kenya. Available at: https://www.afidep.org/download/research-briefs/policybriefs/Anneceta-Gacheri.pdf [Accessed 5 Jun. 2017]. Gadeberg, M. (2016) ‘World’s first insurance for African pastoralists provides buffer against climate shocks’ Drylandsystems, [online] Available at: http://drylandsystems.cgiar.org/outcome-stories/world%E2%80%99s-first-insurance-africanpastoralists-provides-buffer-against-climate-shocks [Accessed 5 Aug. 2017]. Humanitarian Policy Group (2009). Demographic trends, settlement patterns and service provision in pastoralism. Synthesis Paper. [online] Overseas Development Institute. Available at: https://www.files.ethz.ch/isn/116336/2009-04_Demographic.pdf [Accessed 19 Aug. 2017]. International Labor organization. (2007) ‘Social Health Protection: An ILO strategy towards universal access to healthcare’, Issues in Social Protection, Discussion paper 19, ILO. Kassebaum, N., Bertozzi-Villa, A., Coggeshall, M., Shackelford, K., Steiner, C., Heuton, K. and Gonzalez-Medina, D. (2014). Global, Regional, and National Levels and Causes of Maternal Mortality During 1990-2013. Obstetric Anesthesia Digest, 35(4), pp.196-197. 38 Keats, E., Ngugi, A., Macharia, W., Akseer, N., Khaemba, E., Bhatti, Z., Rizvi, A., Tole, J. and Bhutta, Z. (2017). Progress and priorities for reproductive, maternal, newborn, and child health in Kenya: a Countdown to 2015 country case study. The Lancet Global Health, 5(8), pp. 782-795. Kenya.unfpa.org. (2014). UNFPA Kenya | Counties with the Highest Burden of Maternal Mortality. [online] Available at: http://kenya.unfpa.org/news/counties-highest-burdenmaternal-mortality [Accessed 29 Aug. 2017]. Kenya National Bureau of Statistics, (2009) ‘The 2009 Kenya Population and Housing Census Volume II – Population and Household Distribution by Social Economic Characteristics’ pp. 397-398. Kenya National Bureau of Statistics and ICF International. (2015)’ Kenya Demographic and Health Survey 2014: Key Indicators.’ Calverton: KNBS and ICF International. Available at: http://reliefweb.int/sites/reliefweb.int/files/resources/Kenya%20Demographic%20and%20He alth%20Survey.pdf [Accessed 18 Jun. 2017]. Kenya National Bureau of Statistics and ICF Macro. (2010) ‘Kenya Demographic and Health Survey 2008–09’,Calverton Maryland: KNBS and ICF Macro. Kipuri, N., Ridgewell, A. (2008) ‘A Double Bind: The Exclusion of Pastoralist Women in the East and Horn of Africa’, Minority Rights Group International. Kitui, J., Dutton, V., Bester, D., Ndirangu, R., Wangai, S. and Ngugi, S. (2017). Traditional Birth Attendant reorientation and Motherpacks incentive’s effect on health facility delivery uptake in Narok County, Kenya: An impact analysis. BMC Pregnancy and Childbirth, 17(1). Kitui, S., Lewis, S., Davey, G. (2013) ‘Factors influencing place of delivery for women in Kenya: an analysis of the Kenya demographic and health survey, 2008/2009’, BMC Pregnancy Childbirth, pp. 13 (40). Knaul, F., Langer, A., Atun, R., Rodin, D., Frenk, J. and Bonita, R. (2016). Rethinking maternal health. The Lancet Global Health, 4(4), pp. 227-228. Maro, G. Z. et al. (2012). Understanding Nomadic Realities. Case Studies on Sexual and Reproductive Health and Rights in East Africa. AMREF/KIT. Medicus Mundi International Network (2013). Realising the Right to Health with Disaggregated Data — MMI. [online] Available at: http://www.medicusmundi.org/contributions/reports/2014/realising-the-right-to-health-withdisaggregated-data [Accessed 19 Aug. 2017]. 39 Mekonnen, Y., Mekonnen, A. (2003) ‘Factors influencing the use of maternal healthcare services in Ethiopia’. J Health Popul Nutr, 21, pp. 374–382. Ministry of Devolution and Planning (2015). National Policy for the Sustainable Development of Northern Kenya and other Arid Lands : ‘Unlocking Our Full Potential for Realization of the Kenya Vision 2030 ’. The Presidency, Republic of Kenya. Ministry of Health (2015). Status of Implementation of Free Maternity Services Program in the Devolved health system in Kenya: A Comprehensive assessment report. Ministry of Health. Moyer, A., Mustafa, A. (2013) ‘Drivers and deterrents of facility delivery in sub-Saharan Africa: a systematic review’, Reprod Health, 10 (40). Mwaura, N. (2013). Free Maternal Care in Kenya. GirlsGlobe. [online] Available at: https://www.girlsglobe.org/2013/08/28/free-maternal-care-in-kenya/ [Accessed 6 Aug. 2017]. Ng’anjo, P.S., Kiserud, T., Kvale, G., Byskov, J., Evjen-Olsen, B., Michelo, C., Echoka, E., Fylkesnes, K. (2014) ‘Factors associated with health facility childbirth in districts of Kenya, Tanzania and Zambia: a population based survey’, BMC Pregnancy Childbirth. Njoka, J.T. (2016). Kenya: Country situation assessment Working paper, Pathways to Resilience in Semi-arid Economies, 01. Njuguna, J., Kamau, N. and Muruka, C. (2017). Impact of free delivery policy on utilization of maternal health services in county referral hospitals in Kenya. BMC Health Services Research, 17(1). Norton, A., Conway, T., Foster, M. (2001), Social Protection Concepts and Approaches: Implications for policy and practice in International Development, Overseas Development Institute, p. 7. Ogachi, I.O. (2011). ‘Transforming Education and Development Policies for Pastoralist Communities in Kenya: Through the Integration of Indigenous Knowledge Systems’, African Books Collective. Overseas Development Institute (2009). Pastoralism, policies and practice in the Horn and East Africa: A review of current trends. Synthesis Paper. [online] Overseas Development Institute. Available at: https://www.odi.org/sites/odi.org.uk/files/odi-assets/publicationsopinion-files/4315.pdf [Accessed 21 Jun. 2017]. Oxfam Briefing paper (2008). Survival of the fittest: Pastoralism and climate change in East Africa. [online] Oxfam International. Available at: 40 https://www.oxfam.org.nz/sites/default/files/reports/Pastoralism%20and%20Climate%20Cha nge%20in%20East%20Africa.pdf [Accessed 5 Jul. 2017]. Pan American Health organization, (2008) Social Protection in Health Schemes: Lessons Learned from the Latin American Region for mother, newborn and child populations. Available at: http://www1.paho.org/english/AD/THS/OS/SPHS-eng.pdf?ua=1 [Accessed 7 Aug. 2017]. Pratt, D., Le Gall, F. and Haan, C. (1997). Investing in pastoralism. Washington, D.C.: World Bank Technical Paper, 365. Reid, R., Fernández-Giménez, M. and Galvin, K. (2014). Dynamics and Resilience of Rangelands and Pastoral Peoples Around the Globe. Annual Review of Environment and Resources, 39(1), pp. 217-242. Republic of Kenya (1992) ‘Development Policy for Arid and Semi-Arid Lands’. Nairobi: Republic of Kenya. Republic of Kenya (2007). National Reproductive Health Policy: Enhancing Reproductive Health Status for all Kenyans. Ministry of Health. Republic of Kenya (2010). National Road Map: For Accelerating the Attainment of the MDGs related to Maternal and Newborn Health in Kenya. Ministry of Medical Services, Ministry of Public Health and Sanitation. Republic of Kenya (2012). Kenya Health Policy 2012 - 2030. Ministry of Medical Services, Ministry of Public Health and Sanitation. Roelen, K. and Devereux, S. (2013) ‘Promoting Inclusive Social Protection in the Post-2015 Framework’ IDS Policy Briefing 39, Institute of Development Studies. Sabates-Wheeler, R. and Devereux, S., (2008). ‘Transformative social protection: the currency of social justice’. Social protection for the poor and poorest: Concepts, policies and politics, pp. 64-84. Say, L., Chou, D., Gemmill, A., Tunçalp, Ö., Moller, A., Daniels, J., Gülmezoglu, A., Temmerman, M. and Alkema, L. (2014). Global causes of maternal death: a WHO systematic analysis. The Lancet Global Health, 2(6), pp. 323-333. Schelling, E., Weibel, D. and Bonfoh, B. (2008). Learning from the delivery of social services to pastoralists: Elements of good practice. [ebook] World Initiative for Sustainable Pastoralism. Available at: https://cmsdata.iucn.org/downloads/social_services_to_pastoralists__english__2.pdf [Accessed 11 Aug. 2017]. 41 Sipsma, H., Thompson, J., Maurer, L., Bradley, E., Curry, L. (2013) ‘Preferences for home delivery in Ethiopia: provider perspectives’, Global Health Public, 8, pp. 1014–1026. Solomon, M. and Rogo, K. (1989). A needs assessment study of traditional birth attendants in rural Kenya. International Journal of Gynecology & Obstetrics, 30(4), pp. 329-334. Stuckler, D., B Feigl, A., Basu, S. and McKee, M. (2010). The political economy of universal health coverage: Background paper for the global symposium on health systems research. [ebook] First Global Symposium on Health Systems Research. Available at: http://www.pacifichealthsummit.org/downloads/UHC/the%20political%20economy%20of% 20uhc.PDF [Accessed 2 Aug. 2017]. Tanner, M. (2005) ‘Strengthening district health systems’, Bulletin of the World Health Organization 83, pp. 403–404. Technical Centre for Agricultural and Rural Cooperation, (2012) ‘Protecting and promoting pastoralism in Africa’, CTA Policy Brief 06, Available at: https://brusselsbriefings.files.wordpress.com/2012/07/no-6-protecting-and-promotingpastoralism-in-africa.pdf [Accessed 24 aug. 2017]. Thaddeus, S., Maine, D. (1994) ‘Too far to walk: maternal mortality in context’,Soc Sci Med. 38, pp. 1091–1110. Tomedi, A., Tucker, K. and Mwanthi, M. (2013). ‘A strategy to increase the number of deliveries with skilled birth attendants in Kenya’. International Journal of Gynecology & Obstetrics, 120(2), pp.152-155. UNICEF. (2009) Maternal and Child Health: The Social Protection Dividend, UNICEF Regional Office for West and Central Africa. UNICEF. (2012) Chapter V , p. 74, Available at: https://www.unicef.org/socialprotection/framework/files/UNICEF_SPSFramework_Chapter_ 5.pdf [Accessed 6 jun. 2017]. UNICEF. (2012) Chapter VI, p. 79, Available at: https://www.unicef.org/socialprotection/framework/files/UNICEF_SPSFramework_Chapter_ 6.pdf [Accessed 6 jun. 2017]. UNICEF. (2012) Chapter VI, p. 84, Available at: https://www.unicef.org/socialprotection/framework/files/UNICEF_SPSFramework_Chapter_ 6.pdf [Accessed 7 jun. 2017]. 42 UNICEF. (2012) Chapter VI, p. 85, Available at: https://www.unicef.org/socialprotection/framework/files/UNICEF_SPSFramework_Chapter_ 6.pdf [Accessed 7 jun. 2017]. United Nations. (2015) The Millennium Development Goals Report 2015, United Nations, Avaiable at: http://www.undp.org/content/dam/undp/library/MDG/english/UNDP_MDG_Report_2015.pd f [Accesssed 25 aug. 2017]. United Nations. (2017). [online] Available at: https://www.un.org/pga/wpcontent/uploads/sites/3/2015/08/120815_outcome-document-of-Summit-for-adoption-of-thepost-2015-development-agenda.pdf [Accessed 29 Aug. 2017]. United Nations Development Program (2010). The Real Wealth of Nations: Pathways to Human Development. Human Development Report 2010, 20th Anniversary Edition. [online] UNDP. Available at: http://hdr.undp.org/sites/default/files/reports/270/hdr_2010_en_complete_reprint.pdf [Accessed 10 Jun. 2017]. Wamalwa, E. (2015). Implementation challenges of free maternity services policy in Kenya: the health workers’ perspective. Pan African Medical Journal, 22. Who.int. (2017). WHO | Maternal mortality ratio (per 100 000 live births). [online] Available at: http://www.who.int/healthinfo/statistics/indmaternalmortality/en/ [Accessed 29 Aug. 2017]. Witter, S., Anthony, K. AND Aikins, M. (2007) ‘Working practices and incomes of health workers: evidence from an evaluation of a delivery fee exemption scheme in Ghana’, Human Resources for Health, 5(2). World Health Organization, (2011) ‘Sustainable health financing structures and universal coverage’, Sixty-Fourth World Health Assembly, Agenda item 13.4, WHA64.9. World Health Organization. (2017). Maternal mortality. [online] Available at: http://www.who.int/mediacentre/factsheets/fs348/en/ [Accessed 29 Aug. 2017]. World Health Organization (2004) Maternal mortality ratio (per 100 000 live births). [online] Available at: http://www.who.int/healthinfo/statistics/indmaternalmortality/en/ [Accessed 17 Aug. 2017]. World Initiative for Sustainable Pastoralism. (2007). ‘Pastoral Institutions for managing natural resources and landscapes’, WISP policy brief , 6. 43 44