NHIS membership non-renewal decision in rural Ghana : exploring reasons not to renew the National Health Insurance membership in the Bongo District, Ghana

(1)

NHIS Membership non-Renewal Decision in Rural Ghana

E

XPLORING REASONS NOT TO RENEW THE

N

ATIONAL

H

EALTH

I

NSURANCE MEMBERSHIP IN THE

B

ONGO

D

ISTRICT

,

G

HANA

Vera Lütke Holz

UvA Student Number: 11126485

Local Supervisor: Dr. Kennedy Alatinga

M.Sc. International Development Studies

Second Reader: Dr. Winny Koster

Graduate School of Social Sciences

Amsterdam, 22 June 2016

(2)

Acknowledgements

I would like to take the opportunity to thank the people who have been supporting me during the time in Ghana and the thesis writing process. First, I am very grateful to my thesis supervisor Dr. Nicky Pouw. I would like to thank you for the constant advice and support you gave me. Especially during the thesis writing process, I have really benefitted from your critical input and your ideas a lot. Thank you.

Upon arrival in Ghana, I was welcomed by my local supervisor, Dr. Kennedy Alatinga. Before I arrived, you already organized interviews with officials for me and I was lucky to be able to make use of your local contacts. You also constantly provided me with valuable input and advice and I was inspired by your local knowledge and experience. It was an honour to work with you.

I would also like to thank my research assistants, Daniel and Wisdom. Your support was essential to achieve my research goals. Daniel, you did not only accompany me every day during the data collection, you were also my first contact person in Bongo. Whatever I needed, you organized it for me and made my time in Bongo much more comfortable. I really appreciated your reliability and your commitment to our work. Wisdom, I am grateful for your input and help during the data collection. You did most of the transcriptions for me and I know it was hard work. I especially appreciate your critical thinking and professional inputs concerning research methods.

The study would also not have been possible without all the lovely people in the Bongo District, who were willing to take part in the research and to share their thoughts with me. I was surprised about your openness, honesty and willingness to support me. I learned that the people in the Bongo District are incredibly kind and helpful and in every single household people were willing to take part in the research.

Finally, I would like to thank my family, especially my parents Herbert and Renate. Your support and positive thinking brought me where I am today. You are always there when I need help. By supporting all my projects, field studies and travels abroad, even though being constantly worried about me; you help me to achieve my goals. Thank you for everything!

(3)

ii

Abstract

In 2003 Ghana introduced a National Health Insurance Scheme which was intended to enable the poor and marginalized access to healthcare. However, 13 years later many poor are still not registered for the insurance and many have decided not to renew their membership. This is a threat to the

sustainability of the scheme which is largely dependent on premium payments of members. In order to improve the health insurance, prevent people from dropping out and ensure inclusiveness of the scheme it is critical to analyse reasons for non-renewal of the membership in a rural poverty context. This research thus analyses factors that influence people’s non-renewal decision and strategies that help to increase re-enrolment in the Bongo District, Ghana. The study derives from a

multi-dimensional framework which combines poverty research and health decision-making theories. It is framed in the context of inclusive development and takes a dynamic and relational view on poverty. A mixed methodology is applied including interviews, focus group discussions, observations and surveys among poor individuals who did not renew their health insurance membership. The findings reveal that high premiums, mismanagement of the insurance and co-payments in health facilities are the main reasons for dropout. Main strategies introduced by the poor, are the reduction of costs for poor

members, improved quality of care and better management of the insurance. The findings support the theoretical assumptions of the dynamic and relational approach to poverty by stressing the locality as a main factor which creates poverty, unveiling the influence of power relations and showing the

inability of the poor and vulnerable to benefit from social protection mechanisms. Finally, the findings indicate that Ghana’s National Health Insurance, which is a strategy for inclusive development, has not yet reached complete inclusiveness.

Keywords: NHIS, Ghana, membership non-renewal, decision-making, strategies, poverty, inclusive development

(4)

iii

List of Figures and Tables

Figure 1. Map of the Bongo District, Ghana 18

Figure 2. Conceptual scheme – NHIS membership non-renewal 20

Figure 3. FGD with women 24

Figure 4. Filling surveys with participants 25

Figure 5. Data collection in the Bongo District 26

Figure 6. Typical housing compound in the Bongo District 30

Figure 7. Education household head 31

Figure 8. Women weaving in the dry season 33

Figure 9. Monthly income per livelihood activity 35

Figure 10. Reasons for NHIS non-renewal 38

Figure 11. Ward Bongo District hospital 45

Figure 12. Negative cycle of deficient refunding 48

Figure 13. Strategies and actions to improve the NHIS 55

Figure 14. Students discuss about NHIS card-renewal in schools 60 Figure 15. Adjusted conceptual scheme – NHIS membership non-renewal 67

Table 1. Livelihood activities household members 32

Table 2. Healthcare access in groups 36

Table 3. Ranking of most important benefits 42

Table 4. Importance of healthcare quality 44

Table 5. Importance of waiting time 49

(8)

vii

Acronyms

CHIP = Community-based health and Planning Services DMHIS = District Mutual Health Insurance Schemes FGD = Focus Group Discussion

Ghc = Ghana Cedi

GoG = Government of Ghana

GPRS = Ghana Poverty Reduction Strategy GSS = Ghana Statistical Service

HDI = Human Development Index ILO = International Labour Organisation

LEAP = Livelihood Empowerment against Poverty

MoGCSP = Ministry of Gender, Children and Social Protection MoH = Ministry of Health

NHIS = National Health Insurance Scheme NSPS = National Social Protection Strategy SP = Social Protection

SSNIT = Social Security and National Insurance Trust UNDP = United Nations Development Programme UNRISD = United Nations Research Institute VAT = Value Added Tax

(9)

1

Chapter 1: Introduction

Globally, people are suffering from exclusion from basic healthcare. Around 400 million people in the world are still excluded from access to basic healthcare (WHO, 2015). Most of them are living in sub-Saharan Africa. In reaction to the challenge of access to healthcare, National Health Insurance Programmes are increasingly gaining attention as development tools. Ghana, as one of the first sub-Saharan African countries, has introduced a National Health Insurance Scheme (NHIS) in 2003 which aimed at the affordability of healthcare for the poor and marginalized.

Even though the NHIS has increased the access to healthcare, especially for the poor, it also faces many obstacles and still has not achieved universal health coverage. A major problem of the NHIS is the non-renewal of membership. Members need to renew their insurance annually and re-enrolment is critical to ensure the sustainability of the insurance which is partly financed through premium payments. Even though the topic has high relevance, there is not much literature about factors that influence the NHIS non-renewal decision yet. This study tries to fill the empirical gap by analysing people’s dropout factors. As non-renewal is higher in rural areas, characterized by

widespread poverty and exclusion, this research is conducted in the Bongo District in the rural North of Ghana.

Analysing the factors that influence the NHIS dropout is relevant in the context of poverty reduction and inclusive reforms. The study contributes to the inclusiveness debate as it combines poverty approaches, the relational approach and the dynamic approach with health decision-making theories, which was not done before in this particular setting. Moreover, it incorporates literature on Ghana’s NHIS which adds a local perspective on the subject matter and makes the framework more holistic. Combining these theories is an interesting and new attempt to shine light on the inclusiveness of social protection (SP) strategies.

The practical relevance of my study for the NHIS and similar schemes is rooted in the disclosing of shortcomings since it offers entry points for potential improvements, to increase re-enrolment and consequently support the sustainability of the scheme. Moreover, my study is

practically relevant for the inclusion of the poor as it serves the goal to make the NHIS more inclusive and help the poor to access healthcare, profit from human capital and have equal opportunities in society.

To achieve these empirical and social contributions I try to answer the following research question: What are the factors that influence the decision of poor people not to renew the membership of the National Health Insurance Scheme (NHIS) in rural areas of Ghana and how can the reasons for non-renewal be overcome?

The nature of this research is explorative since there is not much information about the topic yet and I try to capture the perceptions of the poor without specific predefined assumptions. I apply a

(10)

2

mixed methodology approach with a main focus on qualitative methods to put the voice of the researched central but at the same time achieve complementarity and validity. First, I present the theoretical framework of the study, namely inclusive development, the dynamic and the relational approach to poverty, SP and health decision-making theories in chapter 2. After an introduction to the research context in chapter 3, I present the research methodology in chapter 4. The findings are divided in household characteristics (chapter 5) reasons for non-renewal (chapter 6) and strategies to improve the NHIS (chapter 7). The thesis closes with an answer to the main research question and recommendations for policy and future research (chapter 8).

(11)

3

Chapter 2: Theoretical Framework

In this chapter, I present different theories which form the theoretical background of the study. First, I define and explain the concept of inclusive development. Second, I introduce two approaches to poverty, the dynamic approach and the relational approach. Afterwards, I describe the concept of SP followed by the presentation of different health decision-making theories which help to understand the NHIS non-renewal decision of individuals. The theoretical framework closes with the description of different potential factors that might influence the dropout decision and strategies, informed by empirical research, which might help to overcome insurance dropout.

2.1 Inclusive Development

Inclusive development is the broad framework on which this study is based. To define inclusive development it has to be distinguished from inclusive growth first. For decades, the main focus of development policies was on the economic growth aspect. Inclusive growth is the increase of per capita income whereby the most marginalized are also included in profiting from the growth (Rauniyar & Kabur, 2010). Cook (2006) argues that even though many Asian countries made huge development improvements focussing mainly on growth, it is not sufficient to only take growth into account when aiming to achieve poverty reduction. Besides, fast growth alone entails the risk of undermining social aspects and creating inequalities (Cook, 2006).

For this reason, it is critical to not only focus on inclusive growth but rather on inclusive development (Cock, 2006). Gupta, Pouw and Ros-Tonen (2015) define inclusive development as “[…] development that includes marginalized people, sectors and countries in social, political and economic processes for increased human wellbeing, social and environmental sustainability, and empowerment” (p. 546). I use this definition of inclusive development in this paper as well. It is distinguishable from inclusive growth as it adds social dimensions to the former discourse of development (Rauniyar & Kanbur, 2009). Conceptualizing inclusive development usually incorporates the three dimensions, economics, education and health (Rauniyar & Kanbur, 2009). These dimensions are also included in the Human Development Index (HDI), the most common measurement of inclusive development (Rauniyar & Kanbur, 2009).

Also the inclusiveness aspect of inclusive development is relevant in different regards. First, from a normative approach it is critical to include everyone, also the poorest and most marginalized, in the process of development (Gupta et al., 2015) and enable them to lift themselves out of poverty. As Sen (2000) explained, “To be excluded from common facilities or benefits that others have can certainly be a significant handicap that impoverishes the lives that individuals can enjoy” (p. 44). Second, from an economical angle, inclusiveness is a tool to improve human capital (Gupta et al., 2015)asinequality and exclusion can harm growth and development (Easterly, 2007) and compromise economic agency (Pouw & McGregor, 2014). Furthermore, from a human rights and democratic perspective, everyone has the right to be included in society and take part in decision-making

(12)

4

processes and the sharing of resources (Fraser, 2001). Moreover, inclusive development should enhance voice and empowerment of the poor through interactive governance and participation (Gupta et al., 2015). Finally, the relational approach1 is central to inclusive development (Gupta et al., 2015) since it argues that poverty derives from the societal context and the actions of others, and can therefore only be eliminated through inclusiveness (Harriss-White, 2005). The relational approach provides reasons why inclusive development is often not implemented (Gupta et al., 2005). As people make trade-offs between individual and collective wellbeing, different wellbeing domains and present and future wellbeing (Pouw & McGregor, 2014) institutions and people in power in a capitalist society might enhance their individual wellbeing at the expense of collective wellbeing (Gupta et al., 2015, p.548). The relational approach argues that existing power relations and capitalist institutions reinforce inequality and exclusion and explains that the wellbeing of the poor can be increased through

empowerment and inclusion (Harriss-White, 2006).

Since the NHIS targets the inclusion of the poor regarding healthcare access, inclusive development forms an appropriate framework for the present study. Moreover, the health insurance is a tool which should foster development and include the most marginalised in the process. Therefore it is critical to analyse reasons for exclusion from it in front of the background of inclusive development.

2.2 Conceptualisation of Poverty

The debate around inclusive development is essentially built on the goal to reduce poverty and inequality and ensure stable livelihoods and SP. Since there is no one-size-fits-all recipe to poverty reduction, I first discuss what poverty entails. Moreover, in this study I target people in a rural poverty context, which might influence their health decision-making. As lack of healthcare access and poverty are closely related, the conceptualisation of poverty has to be discussed. I use two related poverty approaches. First, the approach of poverty as a dynamic process as argued by Narayan, Pritchett and Kapoor (2009). Second, the relational approach to poverty, introduced by Harriss-White (2006) and Harriss (2007).

2.2.1 Poverty as a Dynamic Process

Different definitions of and approaches to poverty have been developed over time. For decades, poverty was defined in economic terms as the lack of income or the inadequacy of consumption levels. Introduced by Rowntree (1891) this definition of poverty was made on the measurement of earnings which are at a status of poverty insufficient to maintain physical efficiency. According to Carter and Barrett (2006) the conventional money-metric approach regards poverty as a static state of people who live under the national poverty line. Many scholars challenge this traditional definition of poverty and argue for a more holistic and multi-dimensional approach not only including economic characteristics but also factors like personal assets, vulnerability, freedom and equality (Sen, 1981; Carter and

1

(13)

5

Barrett, 2006). One such approach is the human development approach based on the capabilities approach by Sen (1999), which understands poverty as deprivation of capabilities and basic freedoms. The human development approach draws a more holistic picture of poverty, including social, political and economic aspects and argues that the deprivation of those limits one’s freedom (Laderchi, Saith & Stewart, 2003). Similarly, the multi-dimensional approach advocates to incorporate various indicators in the measurement of poverty, as lack of education, health and living standards (Tsui, 2002).

A more dynamic approach to poverty was introduced by Narayan et al. (2009) which is distinguishable from traditional poverty approaches since it does not regard poverty as a static characteristic but as a dynamic temporary condition. Poverty is a dynamic process as people fall into and lift themselves out of poverty (p.95). Four processes can be distinguished; moving out of poverty, falling into poverty, staying poor and staying rich (p. 95). Accordingly, not everyone who can be classified as ‘poor’ at a certain point of time is caught in permanent poverty (p.88). Narayan et al. (2009) explain that only some people might be poor because of personal reasons, as chronical illnesses (p.98). These are often characterized by “insecurity, limited citizenship, spatial disadvantages, social discrimination and poor economic opportunities” (p.98). Moreover, even people caught in chronical or structural poverty are usually not poor because of personal characteristics but rather lacking

opportunities and social barriers (p.99). Rich people have better possibilities to access markets and establish businesses (p.219). They have capital to invest; better connections to local authorities and better bargaining power due to a greater production scale (p.194).

Narayan et al. (2009) argue that poverty should not be measured on a fixed poverty line but rather be defined individually since it is an individual, social-contextual phenomenon (p.105). Thereby, an emphasis on locality is important since it is a crucial factor that influences poverty, for instance in the form of connectedness, economic opportunities and ecological circumstances (p.219). Finally, the findings of Narayan et al. (2009) reveal that local governance plays a major role in improving conditions like healthcare infrastructure which make it easier for people to escape poverty (p.228).

The dynamic approach to poverty is also used in the study ‘Escaping poverty and becoming poor in 20 Kenyan villages’ by Krishna, Kristjanson, Radeny and Nindo (2004), since they also regard poverty as being context related and dynamic. Moreover, they argue that the factors which help people to escape poverty differ from the ones that prevent people from falling into poverty. According to the findings of Krishna et al. (2004) poor health is the main reason for falling into poverty. Health is a crucial factor not only because of the heavy healthcare expenses but also the knock on effect where other households are involved in taking care of bereaved members where healthcare is lacking (p.17). “Until health constraints are overcome, decline into poverty cannot be feasibly arrested” (p.17). Other factors which potentially precipitate people into poverty are high expenses for funerals, large family sizes and small land holdings (p.13). Opposed to earlier assumptions of more conventional poverty

(14)

6

approaches, laziness and drunkenness are not considered as being important factors which drag people into poverty. This supports the findings of Narayan et al. (2009) who argue that poverty is usually not the fault of the poor themselves but rather caused by conditions and circumstances.

Unlike the reasons that drive people into poverty, escape from poverty is usually related to a diversification of income (Krishna et al., 2004) which occurs when a household member gets involved in a new livelihood activity. A similar conclusion is drawn by the Sustainable Livelihood Approach which argues that livelihood diversification helps to spread risks, recover from shocks and maintain or enhance income and capabilities (Hussein & Nelson, 1998). However, not everyone has equal chances to get involved in a new livelihood activity. The opportunity of getting a job offer is usually dependent on skills, contacts and capital access which are not equally available for everyone (Krishna, et al., 2004). This is again in line with the assumption of Narayan et al. (2009) that poor face greater barriers to access markets and establish businesses.

2.2.2 A Relational Approach to Poverty

As the dynamic approach to poverty, the relational approach also breaks away from conventional poverty research (Harriss, 2007). As elaborated in chapter 2.1 the relational approach fits well into the framework of inclusive development (Gupta et al., 2015).

Poverty and inequality are also regarded as social-political processes within the relational approach (Harriss, 2007). Poverty is created by relations and the structure of society and should not be seen independently but rather contextually (Harriss, 2007). It is a result of how modern society and the market economy function rather than the outcome of the behaviour of the poor themselves (Harriss, 2007). This is in line with Narayan et al.’s (2009) argument that poverty is a condition rather than a characteristic and occurs from the lack of opportunities in society. Mosse (2010) states, that poverty should be regarded as an effect of prejudicial treatment by others rather than individual incapacity.

In addition, the relational approach emphasises the influence of power and unequal power relations. As Mosse (2010) argues, categorical exclusion and exploitation of groups in society are the effects of this. Thereby, the wealth of some is directly linked to the poverty of others within society (Harriss, 2007) and poor benefit less from and are less able to invest in social capital (Harriss, 2007). Relational approach scholars see the main origin of poverty in todays’ capitalist society (Harriss, 2007; Mosse, 2010; Harriss-White, 2006). Poverty is embedded in institutions and processes of capitalism and the capitalist state of production (Harriss-White, 2006). The relationships which arise from the development of capitalism create and recreate power relations and consequently poverty and inequality (Harriss, 2007). The ways in which capitalism creates poverty are for instance small scale household forms of production, unemployment, commodification, crisis and environmental destruction (Harriss-White, 2006).

Harriss-White (2006) claims that it is necessary to regulate capitalism in order to eliminate poverty and that only the state is able to fulfil this task (Harriss-White, 2006). Relational approach

(15)

7

scholars call for a social welfare system of the state. They support the idea that through SP, healthcare and education people have better and more equal chances to contribute to and profit from human capital. With regard to this claim, the relational approach authors are also in favour of a health

insurance system which provides healthcare for marginalized groups (Harriss-White, 2006). A similar claim was also made by authors of the dynamic approach with their argument that poverty cannot be eliminated without the access to sufficient healthcare (Krishna et al., 2004; Narayan et al., 2009).

In this study, I use a combination of the dynamic and the relational approach to conceptualize poverty. Poverty is regarded as a dynamic process which people can be driven into and escape from and as a contextual phenomenon whereby locality is a crucial aspect. Furthermore, the societal context plays an important role in my conceptualisation of poverty as a factor which makes it difficult for people to escape poverty due to power relations, lacking opportunities and other barriers poor individuals face. Even though my study does not engage in poverty measurement, the

conceptualisation of poverty forms a critical background as my study is situated in a rural poverty context and the conceptualisation of poverty is therefore needed to understand people’s decisions and obstacles. The poverty of the rural population might influence their household characteristics and their reasons for the NHIS dropout.

2.3 Social Protection

Against this background of inclusive development and the dynamic and relational approaches to poverty, social safety mechanisms are increasingly seen as an appropriate tool to improve the conditions of the poor in developing countries. Especially with the strong focus on inclusiveness, safety mechanisms that include the poorest and most marginalized, help to foster development in an appropriate manner by including everyone in the development process. They can be fruitful in protecting living standards and providing basic levels of consumption to those living in poverty (United Nations Research Institute (UNRISD), 2010). Besides, “Universal social protection can contribute to human security, reduce poverty and inequality, and build social solidarity” (UNRISD, 2010, p. 135). SP incorporates both, social insurance, like pension schemes and health insurances and social assistance in form of cash transfers (Pouw, 2014). The International Labour Organisation (ILO) views SP as human right and defines it as “entitlement to benefits that society provides to individuals and households – through public and collective measures – to protect against low or declining living standards arising out of a number of basic risks and needs” (van Ginneken, 2006, p.11). Considering SP as a human right implicates that governments and policy makers are obligated to include SP into their governance framework (Barrientos & Hulme, 2008).

Opposed to these today dominating conceptualisations of SP, in the 1980s and early 1990s, in which neoliberal thinking was dominant in western countries, it was widely assumed that supporting economic growth and private sector activities would automatically reduce poverty and increase living

(16)

8

standards (Adésinà, 2009; DeHaan, 2015). This approach was also imposed on developing countries by the provision of credits from multilaterals (e.g. World Bank (WB) and IMF) with the condition to push liberalisation and privatisation in the country (DeHaan, 2015, p.607). Rising inequality, poverty, illiteracy and ill-health in many sub-Saharan African countries were the consequences of market failures to compensate for missing public policies in protecting vulnerable citizens and providing safety mechanisms (Adésinà, 2009). The neoliberal reaction to this failure was the provision of social safety nets to absorb temporary income shortfalls (Adésinà, 2009). The earlier WB (2001) definition fits into this thinking as they define SP as: “[…] all public interventions that help individuals, households, and communities to manage risk or that provide support to the critically poor” (p.9). However, Adésinà (2009) argues that many were excluded, whereby “people did not merely fall through the net of social provisioning – they died” (p.44). Also Sabates-Wheeler and Devereux (2008) critique the WB approach by arguing that SP should not only include economic protection in terms of public responses to livelihood shocks but also focus on social wellbeing, empowerment, equity and social rights. A shift was made from simple safety nets to more emphasis onfinancial redistribution, social inclusion and equality (DeHaan 2015, p.608/609). This view on SP fits into the framework of inclusive development and is supported by the relational and the dynamic approach to poverty which are against a purely income related definition of poverty and promote SP as a development tool (Harriss-White, 2006; Krishna et al., 2004).

One example of a SP mechanism is the health insurance. According to Alatinga and Fielmua (2011) “Health and social security are human rights and indispensable prerequisites for poverty reduction, economic growth and development” (p.125). Therefore, it is critical to provide an appropriate health financing system to offer everyone the possibility to access healthcare. Xu et al. (2007) found that 100 million people, of whom more than 90% live in low-income countries, are forced into poverty because of heavy healthcare expenses every year. That is in line with findings of Krishna et al. (2004) who state that the main factor which drives people into poverty, are healthcare payments. Xu et al. (2007) advocate the introduction of a national health financing system as a critical SP tool that prevents people from falling into poverty.

2.4 Health Decision-Making

As the present study analyses health decision-making factors, in this paragraph I present theoretical background information about making theories with a particular focus on health decision-making.

One main decision-making theory is the rational decision-making, based on the rational economic man who is completely informed, sensitive and rational and choses the alternative that leads to greatest utility (Doyle, 1999). However, the rational theory is widely criticised since people’s decisions are not only based on utility maximisation but are for instance also influenced by emotions,

(17)

9

risks (

Sanfey, Rilling, Aronson, Nystrom, & Cohen, 2003), expectations, experiences and

social preference (Camerer, Loewenstein & Rabin, 2011).

Pouw and McGregor (2014) explain that neoclassical thinking about the rational man neglects essential factors like lack of awareness and information, and different subjectivities like cultural values. “Neoclassical theories have failed to explain individual level decision-making processes as influenced by social and cultural institutions, which in themselves have a history” (Pouw & McGregor, 2014, p.12). Additional, economic agents face different economic problems which influence their making (p.8). In economic decision-making people try to improve, maintain and protect their wellbeing by choosing the best combination of three alternative elements of wellbeing: material, social-relational, cognitive/subjective (p.18). Thereby economic maximisation is only part of the story while people make trade-offs between individual and collective wellbeing and present and future wellbeing (p.18).

Especially in the health sector different factors might occur that influence people’s decision-making (Schneider, 2004) and people have to make trade-offs between different wellbeing dimensions (Pouw & McGregor, 2014). The expected utility theory for instance argues that health decision-making has to be analysed in the context of uncertainty since people make choices between the uncertainty of being uninsured and the financial loss when insured (Schneider, 2004). Additionally, the fuzzy-trace theory argues that past experiences influence people’s health decisions and someone who did not experience the benefits of the health insurance in the past tends to drop out (Reyna, 2008). Moreover, according to the state-dependent utility theory, the insurance decision is also influenced by the individual conditions, like socio-economic and health conditions (Schneider, 2004). Finally, Propper (1993) found, that demographic characteristics, income, quality of care and level of premium payment influence the insurance enrolment decision. Concluding, both cost-effectiveness (Eichler, Kong, Gerth, Mavros & Jönsson, 2004) and other factors based on individual circumstances,

experiences (Schneider, 2004; Reyna, 2008) and general conditions (Propper, 1993; Schneider, 2004) might influence people’s insurance decision while they make trade-offs between different wellbeing dimensions (Pouw & McGregor, 2014), for example between present consumption and future security.

What I take out from the description of decision-making theories for this study is the understanding of how people make health decisions. It helps to understand why people consider certain factors as important and to realize the underlying mechanisms that form part of the decision-making process as for instance the different trade-offs an individual has to make. Moreover, it indicates that decision-making is not straightforward and usually influenced by different factors on various levels. This is critical to understand the heterogeneity of individual decisions.

2.5 Potential Factors and Strategies

With this background of the before mentioned literature, this study aims to discover the factors which influence the decision of poor people in rural Ghana not to renew their NHIS membership. Since there

(18)

10

is not much literature in existence about this topic, my choice of factors is mainly based on literature about the NHIS enrolment decision and general health insurance decision-making theories. Moreover, since the nature of this research is explorative, further factors may yet emerge during field research for future consideration. Finally I introduce how strategies can be developed and provide an example of potential strategies.

2.5.1 Household Characteristics

Households are defined according to the definition of the Ghana Statistical Service (GSS) (2014) as a person or group of persons who live together and are either relatives or non-relatives. Household characteristics as the socio-economic situation and income levels of the household might influence the insurance decision (Schneider, 2004). Moreover, characteristics of the household head like age, education and gender influence the insurance decision (Adamba, 2010; Alatinga & Williams, 2015). Furthermore, the household education status and size might influence the dropout decision (Adamba, 2010). Finally also the health status of a household might play a role (Adamba, 2010; Schneider, 2004) as households that experienced less illness in the past are less likely to renew their health insurance.

2.5.2 Renewal Costs

Costs that occur with the renewal of the NHIS membership might be of relevance for the non-renewal decision. Renewal costs include all expenses an individual or household has to cover when renewing the NHIS membership, including premium payments and renewal fees. According to Schneider (2004) premium payments play a critical role in health insurance decision-making. Especially for poor households, premiums might influence the decision not to insure (Alatinga & Fielmua, 2011; Jehu-Appiah et al., 2011; Adamba, 2010).

2.5.3 Renewal Benefits

I define renewal benefits as the personal benefits someone receives from the renewal of the NHIS membership. According to existing literature, risk-reduction is a main benefit people obtain from an insurance membership (Schneider, 2004; Adamba, 2010; Blanchet, Fink & Osei-Akoto, 2012). People are usually risk averse and especially people with low socio-economic status try to avoid financial risks (Schneider, 2004). Moreover, the insurance renewal enables poor individuals to access treatment and medical care they would otherwise not be able to access (Adamba, 2010) which is also seen as benefit.

2.5.4 Healthcare Factors

I use the term healthcare factors to refer to factors that are connected to the healthcare system in Ghana. One main healthcare factor is the health infrastructure (Alatinga & Fielmua, 2011; Durairaj, D’Almeida & Kirigia, 2010; Nguyen, Rajkotia & Wang 2011). Different studies found that people decide not to enrol for the NHIS because of the lack of nearby facilities (Alatinga & Fielmua, 2011;

(19)

11 Durairaj et al., 2010).

Moreover, Schneider (2004) argues that the quality of care has significant influence on the insurance decision. Also Jehu-Appiah et al. (2011) disclosed that both the technical care quality and the quality of health stuff influence the NHIS enrolment and renewal decision. This might especially influence the renewal decisions since those individuals might have experienced the quality of care before and can make their decision based on this knowledge.

Additional cash payments at the point of treatment might also influence people’s dropout decision. According to findings of Nguyen et al. (2011), members of the NHIS often have to make co-payments in health facilities in order to receive treatment and medicine. These additional co-payments are illegal and might make the NHIS membership renewal unappealing (Nguyen et al., 2011; Adamba, 2010).

2.5.5 Personal and Additional Reasons

The personal attitude of members and other personal reasons could influence the non-renewal decision. The personal attitude can be formed by political, religious and ethical beliefs or the social and cultural context. According to Jütting (2005) societal norms and values influence health decision-making in rural areas. Moreover Tabor (2005) found that religious and spiritual beliefs hinder people in rural Benin from insuring. As argued by the state-dependent utility theory (Schneider, 2004) the non-renewal decision could also be influenced by the health state of the individual. Healthy

individuals might less likely renew their insurance. This research is open to discover other reasons for the NHIS dropout, which have as yet not been identified in previous research.

2.5.6 Strategies and Actions to Overcome Non-renewal

Based on the non-renewal factors, I design strategies and actions that help to overcome membership dropout. Potential strategies could be the increase of membership renewal benefits or the decrease of renewal costs. In their research, Alatinga and Williams (2015) presented actions the management of the NHIS should undertake to increase enrolment. They recommended the approval of a more flexible payment method for premiums and subsidies for poor groups who are not able to afford the premium payment but are not classified as indigent either (Alatinga & Williams, 2015). Similar strategies could be identified through my research. They could be related to knowledge- and awareness-raising about the NHIS and the improvement of the healthcare system.

2.6 Integrated perspective on the subject matter

All approaches mentioned, have in common that they neglect the exclusive positioning of economics by emphasizing the need to consider respectively, social aspects of development, the social-relational and dynamic nature of poverty, the need for social services as protection mechanism and the

consideration of different wellbeing dimensions in decision-making. Thereby they stress the

(20)

12

poverty context is regarded as multi-dimensional and taking place on different levels. Inclusive development enables putting the findings in a broader context and not only regarding them as simple outcomes of individual decision-making but rather as consequences of lacking inclusiveness and failure of SP. Additional, a combination of poverty research and decision-making theories is needed to achieve a holistic picture of people’s decision-making and understand the reasons for their choices in the poverty context. Even if the research does not try to measure poverty or estimate the value of decision-making theories they form the background influencing people’s health insurance decision.

2.7 Conclusion

The conceptual framework of the study is holistic, including different theoretical angles. First the concept of inclusive development and inherent SP strategies is used which offer a comprehensive view on development and stress the need for inclusiveness. The conceptualisation of poverty as being dynamic and relational enables to see poverty as non-static, dependent on external circumstances and influenced by the structure of society. The poverty conceptualisation is necessary to understand its influence on people’s NHIS dropout decision. Finally, health decision-making is seen as a

comprehensive process including different rationales, trade-offs and factors described in various theories. A combination of these decision-making principles might influence people’s non-renewal decision together with the influence of people’s poverty.

(21)

13

Chapter 3: Research Context

Before addressing the research methodology and research findings, I outline the contextual

background in which this study is embedded. Large poverty rates in rural areas, especially in the North of the country, due to dependency on agriculture in a dry climate play an important role in the

understanding of my research. Furthermore, the context of SP in Ghana is a significant backbone to my study. After a general description of the political, economic and social situation of Ghana I present the issue of SP with a special regard to the NHIS in Ghana. Moreover, I introduce the research area, the Bongo District, and SP in this area.

3.1 Geographical, Political, Economic and Social Characteristics of Ghana

Ghana is a sub-Saharan African country which is located in the western part of Africa and shares boarders with Togo, the Ivory Coast and Burkina Faso. Ghana has a tropical climate, whereby the South of the country is hot and humid and the North is hot and dry which makes the North vulnerable to droughts. The population of Ghana is 25.9 million (United Nations Development Programme (UNDP), 2015) and is split up into about 80 different ethnical groups. The largest group are the Akan (47.3%) followed by the Mole Dagbani (16.6%), Ewe (13.9%) and Ga-Dangme (7.4%) (GSS, 2013). While English is the official language, most ethnical groups have their own language and many Ghanaians are not able to speak English. The dominant religion in Ghana is Christianity with 70% whilst the second largest religion is Islam which covers 17.6% of the population followed by Traditionalists with 5.2% (GSS, 2013).

Ghana gained independence from the colonial powers in 1957 as the first British colony in Africa (UNDP, 2015). In 1960, Ghana turned into a republic with D.Kwame Nkrumak as its first president. After some periods of struggle, Ghana became a democracy in 1992 with a political multi-party system (UNDP,2015). The Ghanaian democracy gained increasingly more stability during the past years in four terms of successful and peaceful elections (UNDP, 2015). Ghana is often regarded as a role model for other African countries concerning political and economic reforms because of political stability fast economic growth and successful poverty reduction (UNDP, 2015).

In 2014 the country had a GDP of $1.426 US per capita and it is classified as a lower middle income country (GSS, 2015). The largest industry in Ghana is agriculture, employing 42.0% of the workforce (GSS, 2013). Other major industries are wholesale and retail at 18.7%, manufacturing at 10.7% and food and service activities employing 5.4% of Ghana’s workforce. The dominant rural employment activity is agriculture which accounts for 69.4% of the rural workforce (GSS, 2013).

According to the UNDP (2015), the life expectancy in Ghana is estimated at 61.13 years. The adult literacy rate reaches 71.5% of the population above 15 years (UNDP, 2015). Even though Ghana is often regarded as one of the more developed sub-Saharan African countries, still many Ghanaians

(22)

14

suffer from extreme poverty, vulnerability and exclusion (Agbaam & Dinbabo, 2014). In 2012, 24.2% of the population was living below the national poverty line (WB, 2012). Moreover, poverty is especially spread out in rural areas with 80% of the poor living in rural regions (Agbaam & Dinbabo, 2014) and huge rural-urban disparities (UNDP, 2015). Rural areas are lacking sufficient infrastructure (UNDP, 2015) which is critical for economic growth and social progress. Furthermore, gender differences are still a major problem in Ghana. Women’s access to land and financial means is very poor and men are usually regarded as household head (Social Institutions and Gender Index, 2015). 3.1.1 Social Protection and the NHIS in Ghana

Ghana has a long history in SP strategies. In 1965 the country introduced a nationwide Social Security Scheme which provided flat payments for elderly, invalids and survivors (Abebrese, 2011). In 1991 it developed into a pension scheme, the Social Security and National Insurance Trust (SSNIT), due to the implementation of a social security law. 17.5% of a workers monthly salary was paid to the scheme and was used to pay for old age pension, invalidity pension and death-survivors payment (Abebrese, 2011). In 2002, the Ghana Poverty Reduction Strategy (GPRS) was implemented. It aims at the reduction of poverty and consists of different programmes for the excluded and vulnerable (Abebrese, 2011). One major component is the implementation of the NHIS. In 2007/2008 the second broad SP strategy was implemented. The National Social Protection Strategy (NSPS) had the main goal of reducing poverty and achieving the Millennium Development Goals (Abebrese, 2011). The strategy consisted of three main components, the establishment of a new social grant scheme, better poverty targeting of existing programmes and a package of complementary inputs (Abebrese, 2011). The Livelihood Empowerment against Poverty (LEAP) social grant programme is a main component of the strategy. SP in Ghana is mainly financed by the Government of Ghana (GoG), with support of development partners like UNICEF and WB loans. Different programmes have different financial sources as the NHIS is mainly financed by the GoG and member contributions and LEAP is mostly financed by development partners (ILO, 2015). According to Pouw (2014) SP can be divided in social insurance, aimed at health, unemployment and old age; and social assistance, including cash transfers to poor and vulnerable. Ghana’s SP includes both social insurance and social assistance and consists of a variety of programmes.

Ghana’s in 2003 introduced insurance is regarded as a pro-poor policy (Adamba, 2010, Blanchet et al., 2012). The goal of the NHIS was to replace the existing system in which citizens had to pay for healthcare in cash at the point of service. It aims at enabling the poor access to healthcare (Adamba, 2010) and to increase the affordability and utilisation of health services and drugs (Blanchet et al., 2012). The insurance system consists of District Mutual Health Insurance Schemes (DMHIS) responsible for the insurance within different regions of Ghana (Blanchet et al., 2012). The NHIS is a pre-defined benefit package covering around 95% of the diseases common in Ghana (Adamba, 2010; Blanchet et al., 2012; Brugiavini & Pace, 2011). For formally employed workers, whose contribution

(23)

15

is deducted from their salaries, the health insurance is mandatory (Adamba, 2010). Since most of Ghana’s citizens work in the informal sector, most people can decide for themselves whether to insure or not (Brugiavini & Pace, 2011). Officially, people who subscribe pay an income-adjusted premium (Adamba, 2010; Brugiavini & Pace, 2011). Nevertheless since it is hard to examine the income of households working in the informal sector, in practice people usually pay a flat rate of Ghc20 (4.50€)2 annually (Adamba, 2010). The membership expires after a period of one year and has to be renewed for a fee of Ghc5 (1.10€). Some groups, identified as especially vulnerable, are excluded from paying insurance premiums. These are people older than 70 years, children below 18, pregnant women, and indigents, who are classified as having no income, no residence and not living with anyone employed or with a permanent residence (Blanchet et al., 2012). These groups, with exception of the indigents, are still obliged to pay the annual renewal fee (Blanchet et al., 2012). The NHIS is mainly financed by a 2.5% Value Added Tax (VAT) on most goods and services, 2.5% of the formal sector workers salary as a contribution to the SSNIT, premium payments of informal sector workers and other funds from the parliament and donors (Schieber, Cashin, Saleh & Lavado, 2012; Agyepong et al., 2008). The financing of the NHIS has implications on the sustainability of the scheme since it is, due to high levels of informal employment, largely dependent on premium paying members and the revenue is not sufficient to sustain the scheme long term (Schieber et al., 2012).

The NHIS brings significant advantages for the citizens of Ghana. It increases the access to healthcare as insured use health facilities on average about three times more than uninsured (Alatinga & Fielmua, 2011). It decreases the risk of being in need of an urgent treatment which is not affordable, especially for the poor (Adamba, 2010; Alatinga & Fielmua, 2011; Blanchet et al., 2012). The free insurance for vulnerable people has significantly positive effects, as insured pregnant women are for instance considerably more likely to give birth in a hospital and receive prenatal care (Brugiavini and Pace, 2011; Blanchet et al., 2012).

However, the NHIS also entails disadvantages especially regarding the exclusion of poorest groups in society. Nguyen et al. (2011) explain that much needs to be done still to make the NHIS affordable and attractive for everyone. First, insured members are still required to make out-of-pocket payments for many services (Nguyen et al., 2011). Second, the health infrastructure is insufficient, particularly in rural areas (Durairaj et al., 2010; Nguyen et al., 2011), which makes it unappealing to pay for an insurance one cannot profit from. Third, insured Ghanaians are generally wealthier than uninsured and the poorest are still in risk of being excluded because of the lacking ability to pay premiums (Alatinga and Fielmua, 2011; Jehu-Appiah et al., 2011). Jehu-Appiah et al. (2011) identified that huge parts of the population (28.5%) live below the poverty line while not classified as indigents. Those cannot afford premium payments but are also not exempt from it (Jehu-Appiah et al., 2011). Durairaj et al. (2010) obtained similar findings and argue that the strict definition of indigents causes

2

(24)

16

the exclusion of many marginalized who should actually be the beneficiaries of the programme (Durairaj et al., 2010). Moreover, huge dropout rates indicate problems in the system and compromise the sustainability of the scheme (Adamba, 2010).

3.1.1.1 Current policy debates and targets of SP and the NHIS in Ghana

Ghana’s SP strategy is constantly debated. To improve collaboration between the different programmes, the GoG has introduced the Ministry of Gender, Children and Social Protection (MoGCSP) in 2013 to coordinate the various programmes and, as Minister Lithur (MoGCSP) states, “to provide a coherent framework for delivering social protection, effectively and efficiently” (GBN, 2015). At the conference on inclusive development and SP in April 2016 a shift from growth to inclusion and equality was emphasized (NDCP, 2016), since inequality increased in Ghana (NDCP, 2016). It is stressed that Ghana’s development has to become fair and include everyone equally (NDCP, 2016). The ‘Coordinated Programme for Economic and Social Development’ (2014-2020) emphasizes several targets regarding SP: develop a comprehensive social development policy, improve the targeting, enhance funding and cost-effectiveness, enhance institutional arrangements for sectoral collaboration and poverty reduction, develop economic and social intervention for vulnerable and marginalized and reduce income disparities (NDCP, 2014). On a conference on 5 January 2016 Minister Lithur, announced that the Ministry has established a Ghana National Household Registry Database to identify and select the actual beneficiaries for various SP programmes (GoG, 20161). Moreover, she mentioned that they established an electronic disbursement to beneficiaries and service providers of SP interventions (GoG, 20161). On 13 June 2016 the GoG in cooperation with the MoGCSP, development partners and civil society launched a new National Social Protection Policy including all different SP programmes and focussing on income support, livelihood empowerment and improved access to SP schemes (GoG 20162). Between 2004 and 2012, poverty reduction expenditure in Ghana increased from Ghc516 million to Ghc3.423 million (ILO, 2015).

Also the NHIS as component of Ghana’s SP strategy is constantly discussed and currently aiming towards an upscaling of the process. Main actors involved are the National Health Insurance Authority (NHIA) which regulates and governs the DMHIS, different DMHIS providers who are responsible for the implementation on district level, the Ministry of Health (MoH), involved in organisation and evaluation and the MoGCSP, responsible for SP in Ghana in general (Blanchet et al., 2012; NHIS 2013). Other important stakeholders are the public healthcare provider, providing the actual service and national and international development agencies supporting the NHIS (Blanchet et al., 2012; NHIS 2013). The Government expenditure for the NHIS increased from Ghc0 in 2004 to Ghc587million in 2012 (ILO, 2015).

In 2014 the NHIS introduced a biometric registration system to digitalize and professionalize the process, including the use of Mobile Kits for outside-office registration (GoG, 2014). The

(25)

17

However it also brought challenges like long waiting times, network problems and insufficient Mobile Kits for field registration (GoG, 2014). Moreover, it builds a barrier for old and vulnerable people who are not able to go for card-renewal in person, which is required with the biometric registration. Besides the biometric registration, the NHIS is trying to improve the refunding of health facilities by speeding up the process and improving the communication with health providers (GoG, 2016). The NHIS also wants to improve the indigent targeting. In November 2015 a pilot project was launched financed by the Africa Health Market for Equity to find a new method of identifying poor and vulnerable and registering them free of charge (GoG, 20163).

3.2 Geographical, Political, Economic and Social Characteristics of the Bongo

District

This research is conducted in the Bongo District, a remote area in the Upper East region of Ghana (figure 1). The Bongo District shares boundaries with Burkina Faso to the North, the Kassena-Nankana East District to the West, Bolgatanga Municipal to the South West and Nabdam District to the South East. The vegetation in the district is a Guinea Savanna landscape which stems from the dry climate with temperatures up to 40°C and only one annual rain season (GSS, 2014). The landscape is characterized by big granite rocks which are spread over most parts of the district. Bongo has a population of 84.545 which is relatively young and predominantly living in rural areas (94%) (GSS, 2014). The district shows a high cultural diversity with two main ethnical groups in the region, Bossis and Gurunsis, accompanied by several other ethnicities. The dominant languages are Bonni and Guruni (GSS, 2014). The main religions in the Bongo District are Christians (45.1%), Traditionalists (44%), and Muslims (7.2%) (GSS, 2014).

In 1988 the Bongo District Assembly was created with the mission to improve the quality of live in the district. The district has one constituency, seven Area Councils and fifty-one Unit

Committees and the District Assembly is headed by the District Chief Executive who is nominated by the president and approved by the Assembly members.

Most of the employed population in the Bongo District works in the informal sector (95.5%) whereby more females (97.6%) are employed informally than males (92.8%) (GSS, 2014). 3.2% of the workforce works in the public sector and 1% in the private formal sector. The main economic activity in the region is farming with 72.2% of the employed population being engaged in the farming sector (GSS, 2014). The industrial sector employs 15.5% whilst 12.3% are employed in the service sector. Besides agricultural activities, the majority of women are engaged in small production activities like shea butter processing, pito brewing and handicraft production whereas most men are engaged in sale of poultry, cattle and small ruminants (GSS, 2014).

(26)

18

Figure 1. Location of the Bongo District, Ghana. Fißmer (2013)

The Bongo District has a high illiteracy rate with only 47.7% of the population above 15 years being literate. The District Health Management Team (DHMT) is in charge of the management of health services. The health infrastructure of the Bongo District consists of one hospital, five health centres, one clinic, 36 functional Community-based health and planning service (CHIP) zones, 59 outreach points, 10 feeding centres and one rehabilitation centre(GSS, 2014). The child mortality rate is relatively high with only two-thirds of the children born surviving (GSS, 2014). Moreover the general death rate is higher than the average in the Upper East region with 12.9 out of 1.000 people dying annually (GSS, 2014). Furthermore, 67.4% of the population live under the national poverty line of Ghc3.6 (0.82€) per person per day (GSS, 2015).

3.2.1 Social Protection and the NHIS in the Bongo District

As the numbers reveal, poverty rates in the Bongo District are high and the population is highly dependent on SP strategies. In 2014 Ghc357.666.0 were distributed to 1,803 LEAP beneficiaries in the district (GNA, 2014). Additionally the Bongo District was one of ten districts which were selected to be part of the 1000 programme in 2014, a modified part of the LEAP programme specialized on pregnant women and infants (GNA, 2014).

With the Bongo District being one of the poorest areas of Ghana, the NHIS is supposed to play a central role in the district in order to enable people to access healthcare. 23.000 people in the district were registered for the NHIS free of charge under the classification of indigent in the end of 2015 and

(27)

19

beginning of 20163. These high numbers reveal the lack of financial capabilities of people in the district and the need for SP. In total 58.337 people were registered in the NHIS in the Bongo District in 2015. Further information about the NHIS in the Bongo District will be presented in the following chapters.

3.3 Conclusion

In this chapter I presented the national and district context for this research. I defined the economic situation with agriculture being the dominant industry both on the national and on the district level. Poverty levels are relatively high in the country especially in rural areas like the Bongo District. In order to reduce poverty and enforce social security, Ghana has introduced different SP strategies as the NHIS which is seen as a pro-poor policy.

3

(28)

20

Chapter 4: Methodology

In this chapter I explain the research methodology of this study. I present the research questions and units of analysis, the epistemological positioning of the research, the research methodology and a description of the data collection methods and the data analysis. Moreover, I elaborate on the sampling strategy and finally on ethical considerations and limitations of the research.

4.1 Conceptual Scheme

Figure 2. Conceptual scheme – NHIS membership non-renewal. Created by author, 2016 Figure 2 displays the conceptual scheme underlying this study. The grey boxes on top show the theoretical background in which the study is framed. The rural marginalized context potentially creates poverty in the district which then influences the factors for the NHIS membership non-renewal. These factors are at the same time influenced by health decision-making strategies. The potential factors are, as justified in the theoretical framework, costs and benefits, healthcare system, household

characteristics and personal factors. Moreover due to the explorative nature of the study, it is open to incorporate further factors. These factors might influence the non-renewal decision displayed in green. This forms the base for developing strategies and actions to improve the NHIS and overcome non-renewal. Those strategies might also influence the poverty of the people as they could enhance healthcare access for the poor and diminish social exclusion.

(29)

21

4.2 Research Questions and Unit of Analysis

4.2.1 Research Questions

Against the background of the theoretical framework used in this research, in which I describe SP as a tool of inclusive development that helps to eliminate relational and dynamic poverty, the main goal is to understand why poor households in rural Ghana do not renew their NHIS membership, which can be classified as a SP strategy. Therefore the main research question this study aims to answer is as follows:

What are the factors that influence the decision of poor people not to renew the membership of the National Health Insurance Scheme (NHIS) in rural areas of Ghana and how can the reasons for non-renewal be overcome?

SQ1: What are the household characteristics of people who do not renew their NHIS membership? SQ2: How do costs and benefits of the NHIS influence the decision not to renew the membership? SQ3: How do healthcare factors like the healthcare infrastructure, quality of care and co-payments influence the decision not to renew the NHIS membership?

SQ4: What personal reasons do people provide that influence their decision not to renew their NHIS membership and how do they influence their non-renewal decision?

SQ5: What (other) reasons do people provide for not renewing their NHIS membership?

SQ6: What potential strategies and actions could be adopted to overcome the reasons not to renew the NHIS membership?

4.2.2 Units of Analysis

This research aims at analysing one main unit of analysis, individuals who dropped out of the NHIS in the Bongo District, Ghana. During the data collection process I discovered that the membership renewal is not a household decision as I assumed beforehand. Consequently, the data collection not only included household heads but different individuals and results refer to the individual level. The second unit of analysis are poor and vulnerable people in the Bongo District who dropped out of the NHIS in general. Since individual reasons for dropout of the NHIS were similar throughout the district, I conducted quantitative data to replenish the qualitative findings and to be able to generalize the results to a district level.

4.3 Epistemic Assumptions and Implications

This research derives from a pragmatic epistemic approach which mainly stems from its

(30)

22

questions. The pragmatic approach is suitable since my definitions of development and poverty show a clear distinction from neoliberal, positivist definitions of development economics (Van Staveren, 2011). Moreover, both complete objectivity and complete subjectivity cannot answer my research question, which on the one hand calls for individual perspectives and on the other hand aims to find broader patterns. The pragmatic approach supports the use of different research methods, both

qualitative and quantitative, by putting the research purpose and question central on the basis of which appropriate methods are chosen (Morgan, 2007; Cresswell & Plano Clark, 2011; Johnson &

Onwuegbuzie, 2014). In this research I also used the research question as guideline to choose appropriate methods; a combination of qualitative and quantitative methods. Furthermore, from a pragmatic viewpoint the results obtained from one data collection method can be used for the

development of another instrument (Morgan, 2007) which suits my research in which I developed the quantitative data collection instrument based on qualitative findings.

As the pragmatic epistemic assumption advocates a use of different methods or any suitable method, and thereby brings together researcher from different paradigms (Johnson & Onwuegbuzie, 2014) it allows for incorporating other epistemologies. This research incorporates elements of the participatory approach and the subjectivist approach with minor elements of objectivism.

The participatory approach is influenced by political issues and addresses topics like empowerment and marginalization (Cresswell & Plano Clark, 2011). This study can be classified as partially participatory due to its emphasize on political inequalities in a context of marginalization. The participatory approach advocates including different stakeholders, also the beneficiaries and their knowledge, values and experiences in the process of policy making and governance (Cornwall, 2002; Pouw et al., 2016). In this research, policy recommendations are also based on the ideas of

beneficiaries. According to Pouw et al. (2016) the participatory approach has the potential to link the theoretical inclusive development strategies to the practical situation by including the excluded in the process of defining development strategies.

Subjectivist elements are also integrated in my research since individual reasons for non-renewal are the focus of the study and form the starting point of the data collection which makes it a bottom-up approach, typical for the subjective epistemology (Cresswell & Plano Clark, 2011). Moreover, the subjectivist approach derives from the perspective that individual perceptions are the main source of knowledge and that research should focus on individual opinions rather than a general truth (Holden and Lynch, 2004). This can be identified in my research as well.

Finally, some minor objectivist elements are included. Because the pragmatic approach is often criticized for lacking validity, the objectivist elements of the study increases the validity and reliability of the research (Holden & Lnych, 2004) and thereby compensate for the rather pragmatic, context specific choice of research instruments. Even if objectivism does not necessarily call for

NHIS membership non-renewal decision in rural Ghana : exploring reasons not to renew the National Health Insurance membership in the Bongo District, Ghana