A Comparative Study of Household Willingness to Pay for Community-Based Health Insurance in Urban and Rural Local Government Areas in Lagos State

A COMPARATIVE STUDY OF HOUSEHOLD WILLINGNESS TO PAY FOR COMMUNITY-BASED HEALTH INSURANCE IN URBAN AND RURAL LOCAL GOVERNMENT AREAS IN LAGOS STATE

SUBMITTED BY

DR AGBO IJEOMA NKIRUKA EZIOMA

DEPARTMENT OF COMMUNITY HEALTH

LAGOS UNIVERSITY TEACHING HOSPITAL, IDI-ARABA, LAGOS

THE NATIONAL POSTGRADUATE MEDICAL COLLEGE OF NIGERIA IN PART

FULFILLMENT OF THE REQUIREMENTS FOR THE AWARD OF THE FINAL

FELLOWSHIP OF THE MEDICAL COLLEGE IN PUBLIC HEALTH (FMCPH)

NOVEMBER 2014

DECLARATION

I declare that this Part II Dissertation titled “A comparative study of household willingness to pay for community based health insurance in Urban and Rural LGAs in Lagos state” is my original and individual work. It was done under appropriate supervision.

It has not been submitted in part or in full for any other examination.

------

Dr Ijeoma N.E. Agbo Date

Department of Community Health

Lagos University Teaching Hospital

Idi-Araba, Lagos,

DEDICATION

This project is dedicated to God the Almighty

iii

CERTIFICATION

This is to certify that this dissertation “A comparative study of household willingness to pay for community based health insurance in Urban and Rural LGAs in Lagos state” was written by Dr. Ijeoma Nkiruka Ezioma Agbo of the Department of Community Health, Lagos

University Teaching Hospital, Idi-Araba, under my supervision.

------Supervisor Date

Prof. A.T. Onajole

Department of Community Health,

Lagos University Teaching Hospital,

Idi-Araba, Lagos

------Prof. A. T. Onajole Date

Head of Department

Department of Community Health,

Lagos University Teaching Hospital,

Idi-Araba, Lagos

ACKNOWLEDGEMENT

I would like to express my profound gratitude to the Almighty God for his grace in seeing me thus far and bringing this project to this end.

I am deeply grateful to my supervisor, Prof A. T. Onajole for his time, patience, guidance and the direction in supervising this work. Thank you Sir.

I thank Dr Ogunnowo for his attention to detail and his continued constructive critique at every stage of this work. I also appreciate Dr (Mrs) K.T. Odeyemi and the entire staff of

Community Health department for the training I have received from them.

To Drs Ariyibi, Okwor, Ikpeekha and my colleagues at the department, I say a big thank you for your contributions and advice on this project.

Thanks to the Medical officers of Health in Surulere and Ikorodu LGAs for putting up with my numerous demands. I am most grateful.

Thanks to Chika for helping with my data entry and your support in this work.

To my parents, Prof and Mrs G.C Onyedim, thanks for believing in me.

To my husband Chijioke and residency children, Nkemka, Kambinachi and Tobenna, despite my doubts you always see the best in me and for this I am grateful.

TABLE OF CONTENTS

DECLARATION...... ii

DEDICATION...... iii

CERTIFICATION ...... iv

ACKNOWLEDGEMENT ...... v

LIST OF TABLES ...... viii

LIST OF ABBREVIATIONS ...... xi

SUMMARY ...... xii

CHAPTER ONE ...... 1

INTRODUCTION ...... 1

CHAPTER TWO...... 10

LITERATURE REVIEW ...... 10

CHAPTER THREE ...... 32

METHODOLOGY ...... 32

CHAPTER FOUR ...... 44

RESULTS...... 44

CHAPTER FIVE ...... 93

DISCUSSION ...... 93

CHAPTER SIX ...... 106

CONCLUSION ...... 106

RECOMMENDATIONS...... 106

REFERENCES ...... 108

APPENDIX A: Questionnaire

APPENDIX B: Ethical Clearance

vii

LIST OF TABLES Pages

Table 1: Demographic characteristics of respondents ...... 44

Table 2: Socio-economic characteristics of respondents ...... 44

Table 3: Health seeking behaviour of most recent illness in the household ...... 48

Table 4: Amount spent on treatment by household head on most recent illness in the household in last 3 months ...... 50

Table 5: Household heads payment mechanisms for treatment of most recent illness in the household ...... 52

Table 6: Awareness of health insurance among the respondents ...... 54

Table 7: Perception of community based health insurance among the respondents ...... 56

Table 8: Level of perception of respondents to community based health insurance ...... 58

Table 9: Respondent’s acceptance of community based health insurance as a strategy for paying for their health ...... 59

Table10: Level of acceptance of community based health insurance among the respondents in the urban and rural LGA ...... 60

Table 11: Reasons for non-acceptance of community based health insurance among respondents ...... 61

Table 12: Respondents willingness to pay for a hypothetical CBHI package for their household ...... 62

Table 13: Respondents reasons for unwillingness to pay for the hypothetical community based health insurance package ...... 63

Table 14: Respondents willingness to pay bids of 500,450 and 350 naira as premium for

CBHI for individual enrolees per month ...... 64

viii

Table 15: The maximum amount respondents were willing to pay as premium for individual enrolees per month ...... 65

Table 16: The premium all respondents were willing to pay for the scheme for individual enrolee per month in case of inflation...... 67

Table 17: Respondents willingness to pay start, second and third bids (900, 850 and 750 naira) as premium for their household in the proposed scheme ...... 68

Table 18: Maximum amount respondents were willing to pay (those that refused the bids) for household ...... 70

Table 19: The premium all respondents were willing to pay for their household per month in case of inflation ...... 72

Table 20: Association between respondents’ gender and their willingness to pay for CBHI . 73

Table 21: Association between the age groups of respondents and their willingness to pay for

CBHI ...... 74

Table 22: Association between marital status of respondents and their willingness to pay for

CBHI ...... 75

Table 23: Association between the level of education of the respondents and their willingness to pay for CBHI ...... 76

Table 24: Association between the respondents occupation and willingness to pay for CBHI

...... 77

Table 25: Association between respondent’s household size and their willingness to pay for community based health insurance ...... 78

Table 26: Association between respondent’s income and their WTP for CBHI ...... 79

Table 27: Association between distance respondents’ travelled to health facility and their willingness to pay for CBHI ...... 80

Table 28: Association between respondent’s membership of a financial association/group and their willingness to pay for CBHI ...... 81

Table 29: Association between respondent’s previous use of health insurance and their willingness to pay for CBHI ...... 82

Table 30: Association between respondent’s level of confidence if funds are pooled and managed by community and their willingness to pay for CBHI ...... 83

Table 31: Relationship between Sociodemographic characteristics and mean amounts respondents are willing to pay as premium for single enrolees ...... 84

Table 32: Relationship between socioeconomic characteristics and the mean amounts respondents are willing to pay as premium for household per month ...... 86

Table 33: Logistic Regression of factors influencing willingness to pay for community based health insurance among the respondents...... 88

Table 34: Logistic Regression of factors influencing willingness to pay for community based health insurance among the respondents...... 91

LIST OF ABBREVIATIONS

CBHI: Community based health insurance

FMOH: Federal Ministry of Health

LGA: Local Government Area

MDG: Millennium Development Goal

NDHS: Nigerian Demographic Health Survey

NHIS: National Health Insurance Scheme

MOH: Medical Officer of Health

PHC: Primary Health Centre

SSA: Sub-Saharan Africa

THE: Total Health Expenditure

WHO: World Health Organisation

WTP: Willingness to pay

SUMMARY

INTRODUCTION: Community-based health insurance is increasingly being recognized as a potentially powerful instrument for granting households access to health services in a more equitable way.

AIM: The aim of the study is to assess and compare the perception and willingness to pay for community based health insurance among households in urban and rural LGAs in Lagos.

METHODS: A comparative cross-sectional design was used. The sample size was calculated using the formula for comparison of proportions of two independent groups and respondents were selected using a multistage sampling technique. A pretested semi structured interviewer administered questionnaire was used to collect information from 480 households in an urban and rural LGA in Lagos. This is a study of household WTP, rather than individual WTP, as the economic decision to purchase health care among these households is more likely to be a household decision and not an individual one. Contingent valuation method was used to elicit household willingness to pay for a proposed community based insurance scheme. Data analysis was done using SPSS 17.0. The Chi square and Fishers exact tests were used for comparison of proportions and the Student t test for comparison of means. Multiple logistic regressions were carried out to assess the relationship between predictors and willingness to pay. An association with p value of < 0.05 was considered statistically significant.

RESULTS: The results indicate that 86.3% of the households in the rural LGA and 78.6% of the households in the urban LGA were willing to pay for a proposed community based health insurance scheme. This difference was statistically significant (p=<0.001). It also elicited that the mean amount that the respondents were willing to pay for individual enrolees was

₦555.23/person/month in the urban and ₦542.19/person/month in the rural. This difference was statistically significant (p=0.001). The mean amount they were willingness to pay for

xii their household in the urban and rural area was ₦957.56 and ₦754.83/household/month respectively. The difference in the means was statistically significant (p<0.001).

Perceived reasons given by respondents for unwillingness’ to pay for the scheme includes the respondents’ lack of confidence in the managers of the scheme and that they felt they could not afford the premiums. Furthermore age, level of education, income, household size, and confidence in the managers of the scheme were determinants of willingness to pay.

CONCLUSION AND RECOMMENDATION: This study highlights the high willingness of the respondents to participate in the scheme and has the potential to increase access to quality health care in rural and urban areas in Lagos. The reported premiums gotten from this study can be applied to the populace. There is need for educating the population on the concept of insurance and risk management. A key task for policy makers is to win the trust of the population in relation to a health insurance system, particularly among those with relatively low education, low income groups, large households and the elderly.

xiii

CHAPTER ONE

INTRODUCTION

Background Information

Good health is necessary for well-being and essential in striving to achieve the Millennium

Development Goals with the ultimate goal of eradicating poverty.1

In Nigeria, over 70% of the overall population live below poverty line of about US1.25$/day2 and the percentage of households total expenditure on health is 0.73%.3 For households living close to the poverty line, even low levels of expenditure on health care is sufficient enough to tip them into poverty causing them to sacrifice consumption of other items that are necessary for their well being such as food or education.4 Access to healthcare has been reduced drastically for poor households due to their low purchasing power evidenced by earnings and expenditure patterns.2 As a result, many governments in developing countries have increased efforts to improve the provision of health care in both urban and rural areas. Despite these efforts, many developing countries are still far from achieving universal coverage.1

The situation is worse in rural areas due to the low standard of living and limited accessibility to quality health care services as a result of the absence of formal financial protection in the form of insurance schemes.5 Occurrence of illness usually requires payment at the time of occurrence and this restricts access and impoverishes households.6

In most developing countries like Nigeria, the major sources of health financing have been identified as through (i) the public sector that comprises Local, State and Federal

Governments (ii) the private sector (including the not-for-profit sector) financing which is done, directly or indirectly through health insurance of their employees (iii) households, through out-of-pocket expenditures, including user fees paid in public facilities; (iv) other

1 insurance-social and community-based; and (v) external financing (through grants and loans) from donor organization.7

Despite the health financing options so identified in Nigeria, there still exist disproportions in health system financing. For instance, it has been observed that severe budgetary constraints and uneven distribution of resources occur among the urban and rural areas with the rural areas mostly affected by inequitable budgetary health expenditure allocation.8 Payments for health care can be so high in relation to income that it results in “financial catastrophe” for individuals or households causing them to cut down on necessities.2 This catastrophic healthcare expenditure in Nigeria has further impoverished the poor.9 Nigeria’s health financial arrangement has shifted from health provisioning by government as a normal good towards a competitive market where greater proportion of health services are provided by ability to pay through out of pocket expenses (often referred to as user fees).7 Furthermore, excessive reliance on the ability to pay through Out-Of-Pocket payment reduces health care consumption, exacerbates the already inequitable access to quality care, and exposes households to the financial risk of expensive illness at a time when there are both affordable and effective health financing instruments to address such problems in low income setting.10

Health financing systems through general taxation or through the development of social health insurance are generally recognized to be powerful methods to achieve universal coverage with adequate financial protection for all against healthcare costs.11 This option responds to the goal of fairness in financing, in that beneficiaries are asked to pay according to their means while guaranteeing them the right to health services according to need.11 Risk- pooling is a core characteristic of these systems, enabling health services to be provided according to people’s need rather than to their individual capacity to pay for health services.11

Social health insurance traditionally started by insuring workers. Most functional health insurance schemes in Africa are associated with formal sector employment which requires

2 regular contributions compatible with formal sector earning.12 Such schemes do not cover individuals in the informal sector that predominantly include those that reside in rural areas.

A further nationally organized expansion of social health insurance to the self-employed and non-formal sector is especially demanding. Other financing methods which would circumvent these organisational difficulties are therefore explored; including the direct involvement of communities in health financing.13 In Nigeria, despite many poor people not having access to health care services and attributing the limited access to health care to an inability to pay out-of-pocket costs for health care services, a community-based health insurance scheme developed specifically to overcome this barrier and increase access to health care has received limited take-up among households in Nigeria.13

It could be argued that the system of health care financing in Nigeria is disproportionate, such that, it pushes the burden and risk of obtaining health services to the poor14 and community based health insurance has been advocated to protect them from this catastrophic nature of health financing.

Problem Statement

According to the World Health Organisation, 100 million people every year are driven into poverty due to catastrophic health expenditure15. Most of them reside in resource poor settings such as Sub Saharan Africa (SSA) with very weak modern health care systems and in most cases without any functioning health insurance schemes11,16. The result is high disease burden that has a risk of propagating a sickly, unproductive labour force. In Sub-Saharan

Africa, formal and well functioning health insurance schemes generally exist for the very few who are employed in the formal sector17. For the majority, health care is accessed through out-of-pocket expenditure, which in many instances may lead to suboptimal use of health care services. 17

Developing countries are unable to fulfil the healthcare needs of their population especially the poor due to the shrinking budgetary support of the health sector and the unacceptable low quality of public health services.18 The funding of health care in Nigeria is inadequate with budgetary provision to health hardly exceeding 3% of the nation’s total budgetary provisions19 ; this allocation falls below the World Bank recommendation of 15%.5 There is a growing concern about the economic impact of health care expenditure on households who face illness, particularly in areas where pre-payment mechanisms do not exist and households have to make out of pocket expenditures to use health services.20

In Nigeria, public expenditures accounts for about 25% of total health expenditure compared to private expenditure which accounts for almost 70% of total expenditure on health of which

90% is out-of-pocket.20,21 This high level of out-of-pocket expenditure implies that health care can place a significant financial burden on households.

Available health statistics show that the life expectancy of Nigerians increased from 47 years in 2003 to 52 in 201122. The disability adjusted life expectancy is estimated at 38.3 years and this places Nigeria in 187th position amongst 191 member countries of the World Health

Organisation.5

A constraining factor to access to health services is the negative effect of household health expenditures paid as user fees due to the limited funding from government.23 Nonetheless, it has been shown that the higher the level of health expenditures by the government, the higher the disease burden declines.23 Presently, public expenditure funded through general tax revenue in Nigeria accounts for only 20-30% of total health expenditure, while 70-80% comes from other sources, with the bulk of this taking the form of private payments.24,25

Payments for health services, in the form of user charges present a barrier to access26,27 and leads to increased disease burden because, both communicable and non-communicable diseases constitute a considerable burden to households, especially in the present health

4 system financing environment arising from prevailing user fees regimen and dominance of out-of-pocket payments.28 Previous studies show that the burden from these diseases is considerable and increasing.29, 30, 31

In most cases, user fees occur as a result of the scarcity of public financing; the prominence of the public system in the supply of essential healthcare; the government's inability to allocate adequate financing to its health system; the readiness of the poor and the better-off to pay fees as a way of reducing the travel and time costs of alternative sources of care; the low salaries of health workers; the limited public control over pricing practices by public providers; and the lack of key medical supplies such as drugs.32 Some other factors responsible for lack of access to healthcare include poverty, geographic inaccessibility and lack of knowledge.27

In efforts to improve health care financing, provision and utilization of health services in the country, some interventions have been started. The National Health Insurance Scheme

(NHIS) and community-based health insurance (CBHI) scheme are important means of improving access to the needed health services.33 Since the launch of the National Health

Insurance Scheme (NHIS) in 1999, there has been the major commitment to expand health insurance in Nigeria. However, as of mid-2012, NHIS still covered only about 3 percent of the population which is approximately 5 million individuals. 34

Community-Based Health Insurance is a form of health insurance whereby individuals, families, or community groups finance or co-finance costs of health services.35 The coverage is quite low with schemes only existing in a pocket of communities covering less than 1% of the population.36 It has been shown that even when charges are small, the very poor are unable to enrol.37 Thus, the existing inequalities may be worsened, since the less poor people are more likely to enrol and have improved access to care and financial protection. Low

5 enrolment rates is an important limiting factor in CBHI schemes and it has been reported in studies that evaluated willingness to pay for CBHI7,38 ; this greatly undermines the sustainability of the schemes. Enrolment is affected by factors such as attractiveness of the benefit package, affordability of the premium, quality of the health care and the communities trust in the managers of the scheme.38 Hence, the community based health insurance scheme has not started having significant effects on improving access to healthcare in the country.

Justification of the study

In Nigeria, private health expenditure is estimated to be over 70% of the total health expenditure with a major part of it coming from out of pocket expenses which places a significant financial burden on households.20 Most functional health insurance schemes in

Africa are associated with formal sector employment which requires regular contributions compatible with formal sector earning.7 Such schemes do not cover individuals in the informal sector that predominantly live in the rural areas.12 It is not surprising that compared to their urban counterparts; rural households tend to suffer disproportionately from higher levels of ill health, mortality, malnutrition and inadequate health care.39

Community-based health insurance will form an important source of health security, for the poor, unemployed, the informal sector and those living in the rural areas; most of whom are not under the NHIS or any private health insurance scheme that could provide financial protection against illness. In several countries, community-based prepayment schemes have proven to increase access to health care services, especially among children, pregnant women, rural households, and informal workers, a majority of whom are not covered by the existing formal health insurance.40 CBHI schemes have the potential to address the issue of inadequate funding of the health system.

There have been several studies on health financing and CBHI; however few have offered an insight on the willingness to pay for CBHI and little is known about the factors that influence the knowledge, perception and decisions to enrol in health insurance schemes. Also due to a lack of real world experience on community based health insurance among the population, willingness to pay (WTP) for health insurance by means of contingent valuation (CV) methods can be used to measure directly what individuals would be willing to pay for a hypothetical health insurance package.41,42

In general, willingness to pay data are rarely collected or used as part of designing health insurance schemes in developing countries12 and this can cause low enrolment rates in CBHI schemes and in situations where enrolment rates are high ; high drop-out rate43 because of a lack of evidence on willingness to pay before take-off of these schemes.44

Willingness to pay for community health insurance has been shown to be dependent on the place of residence of the enrolees as well as their socioeconomic status.45The characteristics of urban and rural areas differ and can influence perceptions and decisions to enrol in an insurance scheme. Households in urban areas have been found to be more willing to pay for

CBHI when compared to those residing in rural areas.46 This disparity has been attributed to the difference in geo-political area and cost of living in both locations.46 The households in rural areas do not readily accept the idea of paying for services that they might not use with regards to healthcare however this has been disputed in some other studies.12, 43

Willingness to pay studies provides valid and reliable information for a target population to facilitate the design and implementation of an insurance scheme. Baseline study on willingness to pay before commencement of community health insurance schemes is important for the planning and implementation of Community Health insurance.47

Examining the demand for community based health insurance as a mode of health care financing by households willingness-to-pay technique can provide important lessons and

7 recommendations to help researchers, policymakers and other stakeholders in improving the design, implementation and uptake of this health care financing scheme. This invariably would lead to increased access to quality health care1 in rural and urban areas in Lagos.

AIM AND OBJECTIVES

AIM:

The aim of the study is to assess and compare the perception, acceptability and

willingness to pay for community based health insurance among households in urban and

rural LGAs in Lagos.

OBJECTIVES:

1. To determine and compare the perception of households towards community based

health insurance in urban and rural LGAs in Lagos

2. To determine and compare the acceptability and willingness to pay of community

based health insurance among the households in urban and rural LGAs in Lagos.

3. To assess and compare monthly premiums that households are willing to pay for

community based health insurance in urban and rural LGAs in Lagos using a

contingent valuation method.

4. To identify factors that influence households’ willingness to pay for a community-

based health insurance scheme in urban and rural LGAs in Lagos.

CHAPTER TWO

LITERATURE REVIEW

The Healthcare System in Nigeria

Nigeria’s Ten Year Plan for Development and Welfare (1946-56) incorporated the first attempt at planning for Health Services in Nigeria. Since 1st October 1960 ( Nigeria’s

Independence), successive Nigerian Governments (Civilian and Military) have come up with the 2nd, 3rd and 4th National Development Plans all of which has substantial portions dedicated to addressing issues related to National Health care Systems.48 The organization of health services in Nigeria is complex. It includes a wide range of providers in both the public and private sectors. In the public sectors, Nigeria operates a decentralized health system run by the Federal Ministry of Health (FMOH), State Ministry of Health (SMOH), and Local

Government (LG). 48 The Federal Ministry of Health is the overall health policy formulating body thereby coordinating and supervising the activities of the other levels. In addition, it provides tertiary care through teaching hospitals and federal medical centres. The State

Ministry of Health provides secondary care through the state hospitals and comprehensive health centres while the Local Government provides health care services through the primary health centres with the primary level designed to take health care delivery literally to the doorstep of the populace and act as the gatekeeper to the health care system. Although the local government has the main responsibility of managing the primary health centres, all the three tiers of government and various agencies participate in the management of the PHC.48

Health seeking behaviour, determinants and utilisation of health facilities in Nigeria

There is growing literature on health seeking behaviours and the determinants of health services utilization especially in the context of developing countries. Few studies have carried out a review of the situation in Nigeria and related the similar factors responsible for shaping

10 up of health seeking behaviour and health service utilization in its health facilities.49 The factors determining the health behaviours may be seen in various contexts: physical, socioeconomic, cultural and political.49 Therefore, the utilization of a health care system, public or private, formal or non-formal, may depend on socio-demographic factors, social structures, level of education, cultural beliefs and practices, gender discrimination, status of women, economic and political systems environmental conditions, and the disease pattern and health care system itself.49

A study in a rural facility in Ekiti, Nigeria looked into the reasons for the seeking of medical services among the rural dwellers in the state. The major reasons identified were: the type of ailment suffered by the patients, availability of money at the time of sickness, age of the patients, religious beliefs, educational background, severity of sickness and position in the household.50 This finding is similar to that in another study assessing the determinants of health seeking behaviour of mothers for their children where a parent’s socioeconomic status, age of the child, cultural practices, accessibility to health care services, type and severity of disease determined their health-seeking attitude.51 On the factors responsible for utilisation of the medical establishment in Ekiti were its affordable cost, closeness to home, staff attitude, quality of service, knowledge of owners/staff, neatness of the environment, availability of services required and availability of drugs required.50 In a study carried out in a rural community in south west Nigeria52, education was positively associated with utilisation of

PHC services (P<0.05). Availability of maternal and child health services (45.4%), prompt attention at the health facility (23.0%), and appropriate outpatient services attracted 20.5% respondents to use PHC services. Poor education about when to seek care, poverty, perceived high cost of PHC services, lack of drugs and basic laboratory services, and a regular physician on site at the facility were identified as barriers to utilisation.52 The major factors that cause non-attendance of services in a health facility in an LGA in Ekiti included the high

11 costs of drugs (29%) and service charges (19%), easy access to traditional healers (39%) and difficulty in getting transport to a health facility (30%). The unfriendly attitude of the health workers (3.6%) and the wasting of patients' time at the facility (7.8%) did not constitute serious constraints at attendance of facilities for use of services.49 In a study in Ibadan, the findings showed younger women were less likely to use antenatal care service.53 Location also appeared to have effect on the utilization of antenatal care. It showed that women based in urban areas utilised antenatal care services than those in the rural areas.53

Healthcare Financing

A health care financing system involves the means in which funds are generated, allocated, and utilized for health care. It has three basic functions of collecting revenues, pooling resources, and purchasing services.32,54 The World health organisation defined health financing as the system of fund generation or credit, fund expenditures and flow of funds used to support the health services delivery system. Finances may come from foreign or domestic sources and may be private or public in origin.55 Increasingly, government-financed health services in most developing countries have come to depend on payments by patients.

This is particularly the case in low-income countries, which devote a smaller proportion of public spending to health than do high-income countries. In low-income countries (excluding

China and India), private health is nearly double public health expenditure55 ; meaning that households bear a substantial proportion of health costs while having little protection in the event of major illness or injury. The concept of fair financing has become one of three goals of most health systems ensuring that every member of society should pay the same share of their disposable income to cover their health costs.56

The way a country finances its health care system is a key determinant of the health of its citizens.54 Selection of an adequate and efficient method(s) of financing in addition to organizational delivery structure for health services is essential if a country is set to achieve its national health objective of providing health for all.54, 57 The commonly used mechanisms for implementing these functions include tax-based financing, out-of-pocket payments, donor funding, and health insurance (social and private).32 These methods are not mutually exclusive. In fact, most health systems adopt a mixture of various methods.58 The success of the different health financing methods can be measured by the overall effect on equity of access and health outcomes, revenue generation and efficiency, and the effects on user behaviour and provider.59

Health care in Nigeria is financed by a combination of tax revenue, out-of-pocket payments, donor funding, and health insurance (social and community).60 Nigeria's health expenditure is relatively low, even when compared with other African countries. The total health expenditure (THE) as percentage of the gross domestic product (GDP) from 1998 to 2000 was less than 5%, falling behind THE/GDP ratio in other developing countries such as Kenya

(5.3%), Zambia (6.2%), Tanzania (6.8%), Malawi (7.2%), and South Africa (7.5%).25

According to World Health Organization statistics, total health expenditure in Nigeria is around US$ 33 per capita, 63.4% of which comes directly out of pocket.56 Presently, public expenditure funded through general tax revenue in Nigeria accounts for only 20-30% of total health expenditure, while 70-80% comes from other sources, with the bulk of this taking the form of private payments.24, 30 Reliance on private payments transfers the financial burden on the poor20, 61 and this leads to increased disease burden which presents a barrier to access of health care services.26, 27 Health insurance is attracting more and more attention in low- and middle-income countries as a means for improving health care utilization and protecting households against impoverishment from out-of-pocket expenditures. The health financing

13 mechanism was developed to counteract the detrimental effects of out of pocket payments introduced in the 1980s, which now appear to inhibit heath care utilization, particularly for marginalized populations, and to sometimes lead to catastrophic health expenditures.62, 63, 64

The World Health Organization considers health insurance a promising means for achieving universal health-care coverage.65

Government Funded

Health financing systems where government revenues are the main source of health care expenditure are referred to as tax-based systems.66 Funds are usually generated through taxation or other government revenues. Revenues are raised at the federal, state, or local government levels. The states and local governments owing to their low internal revenue generation capacities, still largely depend on the allocation from the federal government66 thereby forming the majority of the funds for the other tiers of government. The total government health expenditure as a proportion of total health expenditure was estimated as

18.69% in 2003, 26.40% in 2004, and 26.02% in 2005.24 Remarkably, the federal budgetary component of health expenditure has increased over the years. It increased from 1.7% in 1991 to 7.2% in 2007.60 Nevertheless, the budgetary allocation for health is still below the 15% signed by the Nigerian government in the Abuja declaration.60

Donor Funding

Donor Funding refers to financial assistance given to developing countries to support socioeconomic and health development. It is external financing through grants and loans from donor organizations.24 The annual average official development assistance inflow from 1999 to 2007 was estimated at US$ 2.34 and US$4.67 per capita, respectively.67 These figures are way below the Sub-Saharan African average of US$28 per capita.24 The contribution of development aid to health care financing in Nigeria was estimated as N27.87 billion (4% of

THE) in 2003. This increased by 29% to N36.04 billion (4.6% of THE) in 2004 and by just

1% to N36.30 billion (4% of THE) in 2005.24 Although the international assistance to the

Nigerian health sector is increasing, it still accounts for a small proportion of public health expenditures. The subject of aid effectiveness and its macroeconomic impact has been debated and has raised concerns68; nonetheless it still remains an important health financing mechanism for developing countries such as Nigeria.

User fees/ Out of pocket payments

The payment for health care at the point of service is referred to as user fees. The scope of user fees is quite variable and can include any combination of registration fees, consultation costs, drug costs and medical material costs.69 Out-of-pocket payments account for the highest proportion of health expenditure in Nigeria and its expenditure as a proportion of

THE averaged 64.59% from 1998 to 2002.25 In 2003, it accounted for 74% of total health expenditure and decreased to 66% in 2004 and later increased to 68% in 2005.23 This implies that households bear the highest burden of health expenditure in Nigeria and has led to negative effects on health seeking behaviour and equity.69,70 In Nigeria, Kenya and Ghana; hospitals and clinics reported a dramatic 50% decline in use within two weeks of introducing charges.71 The ability to pay might require poor households sacrificing their longer term economic well-being and is referred to as catastrophic health expenditure.72 This has been shown to be high in Nigeria.4

Social Health Insurance

Social Health Insurance is a system of financing health care through contributions to an insurance fund that operates within a tight framework of government regulations. It is a form of mandatory insurance scheme, normally on a national scale.48 It provides a pool of funds to

15 cover the cost of health care and it also has a social equity function which eliminates barriers to obtaining health care services at the time of need especially for the vulnerable groups.36

The Social health insurance scheme includes the National Health Insurance Scheme (NHIS) and the Community Based Insurance scheme

a. National Health Insurance scheme

The National Health Insurance Scheme (NHIS) is a social health insurance programme designed by the Federal Government of Nigeria to complement sources of financing the health sector, and to improve access to health care for the majority of Nigerians.73 The

Nigerian government established the National Health Insurance Scheme (NHIS) under Act 35 of 1999 with the aim of improving access to health care and reducing the financial burden of out-of-pocket payment for health care services.74 The evolution of the National Health

Insurance Scheme dates back to 1962, when the need for health insurance in the provision of health care to Nigerian citizens was first recognized but several factors militated against its establishment.73 The Government had initially provided "free health care" for its citizens funded by its earnings from oil exports and general tax revenue; however, the global slump in oil prices in the 1980's greatly affected Nigeria's major source of income.73 Government could therefore no longer afford to provide free health, and subsequently introduced several cost recovery mechanism like user charges and drug revolving funds. Furthermore the introduction of the structural adjustment programme in 1986 adversely affected the health sector allocation.73 Other factors that led to the establishment of the National Health

Insurance Scheme include; the general poor state of the nation's health care services; the excessive dependence and pressure on government provided health facilities; the rising cost of health care services; the inadequate participation of private health services and inappropriate distribution of the health facilities in the country.73 The NHIS became fully

16 operational in 2005 and is organized into different social health insurance programmes however only the formal sector social insurance is currently fully operational.75 There has been a lag in the expansion of NHIS to achieve a considerable coverage since it became operational. A World Bank survey in 2008 reported that about 0.8% of the population was covered by NHIS.76 This has attracted a lot of censure since many people are left out and not benefiting from it.

b. Community-Based Health Insurance

Community-Based Health Insurance (CBHI) is any program managed and operated by a community based organisation that provides resource pooling to cover a part or the full costs of health services.68, 77 It is referred to as a mechanism whereby households in a community

(the population in a village, district or other geographical area, or a social-economic or ethnic population group) finance or co-finance the current and/or capital costs associated with a given set of health services, thereby also having some involvement in the management of the community financing scheme and organization of health services.16

The scheme is voluntary in nature and beneficiaries are associated with or involved in the management of community based schemes, or at least in the choice of healthcare services it covers.77 The term CBHI describes not for profit pre payment plans for healthcare with community control and voluntary membership that provides resource pooling to low income populations.78 CBHI is designed for people living in the rural area and people in the informal sector who cannot get adequate public, private, or employer-sponsored insurance.79 The function of the scheme ensures that sufficient financial resources are made available so that people can access effective healthcare.14 The effects of CBHI on equity, the quality, and efficiency health services are still ambiguous.56,58 Studies have shown that only a small

17 proportion of the nation’s population are aware of community based health insurance and the few schemes are underutilised as even fewer are willing to pay or contribute to the scheme.80

Despite the health financing options so identified in Nigeria, there still exist disproportions in health system financing. For instance, it has been observed that severe budgetary constraints and uneven distribution of resources exist among the urban and rural areas with the rural areas mostly affected by inequitable budgetary health expenditure allocation.11 Nigeria’s health financial arrangement has shifted from health provisioning by government as a normal good towards a competitive market where greater proportion of health services are provided by ability to pay through user fees.81 Furthermore, excessive reliance on user fees reduces health care consumption, exacerbates the already inequitable access to quality care, and exposes households to the financial risk at a time when there are both affordable and effective health financing instruments to address such problems13; it could be argued that the system of health care financing in Nigeria is disproportionate, such that, it pushes the burden and risk of obtaining health services to the poor.

Membership in most CBHI schemes is voluntary14 and makes these schemes vulnerable to adverse selection.30 Adverse selection results when high-risk or sick individuals are more likely to buy health insurance than the low-risk or healthy individuals. In the presence of adverse selection, the premiums which are fixed at the average risk in the population are not enough to cover all the claims.82 Hence, the financial sustainability of the scheme is jeopardized. To limit adverse selection some schemes have restricted enrolment to the group level. Group can be defined as a household, firm, school etc. And group enrolment ensures that all individuals in the group enrol which includes both high and low-risk individuals, thereby, reducing the risk of adverse selection.82

Two distinct CHBI models exist; the provider model in which the organizations under the model operate as both the health service provider and the insurer and the ‘mutual’ model, in

18 which organizations operate as the insurer but purchase healthcare services from a partner organisation.37

A report on estimation of WTP for CBHI in rural Nigeria reported that if varied forms of payment are allowed, households can choose to make contributions in whatever forms of payment they could afford and this can lead to an increase in enrolment rates.7

In a study in Anambra where the preference of benefit packages in rural and urban areas was compared, it showed that respondents in the rural areas and of a lower socioeconomic status preferred a comprehensive benefit package which included all inpatient, outpatient and emergencies services, while those in urban areas as well as those in the higher socioeconomic status group showed a preference for benefit packages which will cover only basic disease control interventions.41 When CBHI schemes are developed, it is important that they are designed and implemented in such a way as to ensure the relevance of the benefit packages to beneficiary communities.45 The scope of benefit package offered by a CBHI affects the community response to its introduction, acceptability, enrolment and overall sustainability.83

Impact and lessons learnt from Community based health insurance pilot schemes

In recent years, there has been a proliferation of community-based health insurance (CBHI) schemes designed to provide financial protection against the costs of health care and expand access to modern health-care services to the informal and rural sectors.84 A review systematically assessed the evidence of the extent to which community-based health insurance is a viable option for low-income countries in mobilizing resources and providing financial protection; results showed that there is evidence that CBHI provides financial protection by reducing OOP spending and by increasing access to healthcare, as seen by increased rates of utilization of care.84 There is was however weak or no evidence that schemes have an effect on the quality of care or the efficiency with which care is produced.84

A study in Nepal assessed the impact of CBHI schemes on access to health services and quality of health care85, they found out the CBHI schemes had achieved very limited coverage of the population. The overall utilisation rate for health services among members of the CBHI scheme was higher among the enrolees however the scheme was found to be non financially viable.85A study in Thiees region of Senegal that assessed the impact of community-based health insurance on the access to health care of their rural poor found out that where most people were deprived of access to health care of good quality, the introduction of the schemes made a substantial difference.37 This was found to have a potential positive effect on the ability of households’ to smooth their consumption, on labour supply and productivity and on the health status of the people insured.37 A simple probit model carried out in Rwanda17 suggested that the effect of membership to CBHI schemes led to high degree of utilization of health services and helped protect members from large and unforeseen catastrophic health related expenses. The results were extremely favourable to poor households than non-poor households.17 This result is similar to that obtained in the study undertaken in Senegal where an increase in utilization of hospitalization services was also observed however there was a failure of the program to address the needs of the poorest of the poor.37 In 2011, a pilot CBHI scheme was introduced in Ethiopia and evidence from this scheme showed increases in outpatient care services from modern providers and among different groups of the population, the benefit of the scheme in terms of creating access to care was more pronounced for the rich than for the poor.86

In Nigeria, CBHIs have been implemented on a relatively small scale, with the aim of extending health coverage to people in the informal sector. CBHI has been piloted in different states in the country. It was piloted and introduced in Anambra State in 2003. A study evaluated the impact of the Anambra state community health care financing scheme on

20 maternal health services and reported that the scheme was highly accepted and it provided adequate funds for maternal health services for a great proportion of the rural communities.82

However, since the change in government in 2005, the scheme has been dormant owing to the diminished government support and interest.79

The Lagos state Ikosi-Isheri Mutual Health Plan which is the pilot Community-Based Health

Insurance Scheme was launched in 2008. Presently, the State has three schemes running in 3

LGAs and 4 communities in the state.87 A 2010 programme assessment of the Lagos CBHI pilot scheme88 reported some positive outcomes as well as some challenges. The quality of services was good in terms of contributions to lower maternal and neonatal mortality rates, as well as drug availability and patient satisfaction. The community had an excellent relationship with the mutual health provider. On the other hand, high turnover was noted among the health providers, who were civil servants. Financial viability was a key issue; member attrition was high, and members who used the services infrequently were asking for a discounted premium. Furthermore, the LGA was paying the premiums for more than half of the 1,000 active member families.88

An assessment of the Hygeia Community Health Plan (CHP) in Kwara state, Nigeria89 showed that enrolment was below target, and many enrolees dropped out due to migration.

Some of the challenges the programme experienced were enrolees’ inability to pay premiums, inadequate number of providers, and uneven quality of services.89 Lessons learnt from this pilot scheme included the importance of community involvement in ensuring member retention; intensive marketing by community members or commissioned agents; pricing needed to reflect the ability, willingness of enrolees to pay and the disease prevalence in the community; continuous monitoring of service delivery was needed to ensure quality of care; and CBHIs that provided generous benefits couldn’t be funded solely by low-income communities.90Another key lesson was that private facilities could make important

21 contributions to CBHIs, even in rural areas, despite scepticism regarding the role of the private sector. Also, charging a token premium fee, even to the poorest members, ensures that enrolees value their membership; and it also discouraged inflation of enrolment rosters with non-existent members.89

Another evaluation of the Lagos and Kwara pilot schemes demonstrated that the use of health care had increased on average by over 20.5%.66 In addition, the program increased utilization of quality health care as measured by an increase in use of modern health care providers and private health facilities. There was also a 52% reduction in out of pocket expenditures.66

If CBHI pre-payment and risk-sharing can be encouraged, they are likely to have an immediate direct and indirect impact on poverty.42 The direct impact being prevention of impoverishment due to catastrophic health expenditures and the indirect impact ensures access to health and thereby improves health, thus allowing individuals to take advantage of economic and social opportunities.42 To enlarge poor and rural/ urban population access to health care, community-based health insurance schemes can be an important element and a valuable first step.22 The schemes are promising alternatives for a cost sharing health care system which hopefully also leads to better utilization of health care services, reduce illness related income shocks and eventually lead to a sustainable and fully functioning universal health care system.22

Perception of Community based health insurance

Perceptions’ relating to the scheme providers, insurance schemes and the community attributes plays a major role in household decisions to voluntarily enrol in the scheme.

Perceptions relating to insurance schemes are found to be most important.91 Price including premium and registration fees and the benefits of the scheme are factors that are significantly

22 associated with enrolment and retention in the scheme. Studies have shown that enrolment decreases if the price of the premiums is perceived to be high.91Also the convenience and administrative arrangements may be a possible barrier. The credibility of the health care system in relation to quality of care is a crucial factor in the way people perceive health insurance. Insured clients who perceive they are given poor quality of care or have longer waiting times in comparison to fee paying clients leads to a growing dissatisfaction with the scheme.91 Clients also perceive a constant supply of essential drugs is a prerequisite for the credibility of the scheme and for the quality of healthcare provided. In contrast to provider attitudes that were perceived negatively, the quality of care and service delivery in a study was perceived positively.91A community’s values and knowledge about health and risk sharing concepts of health insurance may influence household perceptions on need and participation in health insurance. Individual perceptions have a collective effect within a community and so positive and negative experiences of peers and opinion leaders affect people’s decision to enrol in a scheme.91

A study in Oyo state showed that 87% of the respondents were aware of the NHIS however majority of them wanted it discontinued. There was a lack of confidence in the programme as the scheme was not made compulsory.92 About 52% believed that sufficient drugs were available under the scheme and 87% did not believe there was a difference in health service delivery before and after NHIS was implemented.92 This is similar to what was seen in a study in Ghana91 that showed that perception about scheme factors have the strongest association with voluntary enrolment and retention decisions on NHIS and also in a study conducted in South Africa93 were half of the respondents thought it was important to provide healthcare coverage for all and indicated their preference for the NHIS.

Willingness to pay for Community based health insurance

The willingness to pay (WTP) for health insurance in low income countries by means of contingent valuation (CV) methods is used to gauge what individuals would be willing to pay for a hypothetical health insurance package due to lack of real life experience in that area.12,42

In general, willingness to pay data are rarely collected or used as part of designing health insurance schemes in developing countries.12 It has been shown by contingent valuation theory that studies could be undertaken in developing countries to obtain valid and reliable health-related willingness to pay data.12 Such valid and reliable willingness to pay information for a target population would facilitate scheme design and implementation.12 In

Nigeria enrolment in some community health insurance schemes have been low with small average premiums because of a lack of study on willingness to pay before such schemes took off.44 Studies in communities where WTP was not carried out before the scheme revealed a high drop-out rate.94 Studies have been carried out in Namibia and Ethiopia where WTP was done before instituting CHI and showed how important WTP study was to the planning and implementation of CBHI.95 Another study in a rural area in Southern Ghana revealed that

WTP were compatible with high membership in, and satisfactory performance of a proposed health insurance scheme.12

Some studies looked at individuals are willing to part with in monetary terms to restore their health state in event of deterioration since health is not directly tradable on the market like any other commodity. A study carried out in China examined WTP among informal sector workers and found that these workers are willing to pay the equivalent of 4 USD per person per month.95,96 Another study in India observed a median WTP for health insurance at the equivalent of 15 US Dollars per household per month.95 In rural Iran, the finding was that households are willing to pay three US Dollars per household per month on the average.95,97

A study in Namibia revealed fairly higher WTP of seven USD per person per month.42

Studies in rural parts in Nigeria found a WTP of 1.5 USD per household per month.97 On the other hand another Nigerian study compared WTP in an urban and rural community and found it to be higher in urban at 2.9 USD per person per month when compared to the rural of

1.7 USD per person per month.46 A study in Osun state studied the premiums that households preferred to pay for insurance. It showed that majority of the households (74.52%) were willing to pay less than N200 per month. Only 2.40% were willing to pay the highest presented premium of N250

In a study carried out in Nsukka, the mean and median WTP amounts for community based health insurance were N392 and N200 per quarter respectively.7 The mean amount of money that respondents were willing to pay was 522 Naira.41

Contingent valuation

Contingent valuation studies simulate a market. In health economics, contingent valuation is a method that elicits an individual's monetary valuations of health programmes or health states.1 Contingent valuation is a method for placing a monetary value on a good. It is a survey−based method, and a stated preference method, because it relies on people reporting how much they would be willing to pay for the good, with no actual transaction taking place.96 Willingness to pay (WTP), which contingent valuation aims at measuring, is the theoretically correct measure of the value that individuals place on the good in question. The objective is to elicit the maximum amount a non priced good is worth to the respondent.96

It is a monetary measure of welfare associated with a discrete change in the provision of an environmental good, by substituting one good for another or the marginal substitution of different attributes of an existing good.99 A variety of questioning formats are used to ask contingent-valuation questions which includes discrete choice experiment, bidding game,

25 open-ended questions, choice based conjoint analysis, contingent ranking, single or double- bounded dichotomous choice, paired comparisons, payment card e.t.c.100

The single or double dichotomous format however is commonly used as it has gained considerable acceptance due to its simplicity of use in data collection and it induces respondents to reveal their true preferences100. As is done for all contingent-valuation questioning formats, dichotomous-choice questions are implemented in a survey by presenting participants with a contingent market.99 The unique aspect of dichotomous-choice questions is that respondents are asked if they would pay a fixed sum of money for the item being evaluated. The single bound model comprises only one such question, while in the double bound model the first question is followed by another specifying a lower amount, if the answer to the first question was negative, and higher otherwise.100

Factors influencing household acceptance and willingness to pay for Community based health insurance

Designing, implementing, managing, and especially sustaining CBHI are complex. It requires a strong intuitional capacity, technical expertise, and management skills. Numerous challenges exist and could limit the success of the scheme in Nigeria97.

The results are mixed regarding the various factors associated with willingness to pay for community based health insurance among rural and urban households.

Age: The first of such factors is age. There is evidence that age of household heads is related to their mean WTP for health packages. The respondent’s age is found to have a positive effect on WTP in some studies97 while in others, it is the reverse.45 Findings in Namibia42 showed that the young respondents show more interest in joining and WTP for the scheme.

The premium level that individuals were willing to pay was also related to age of the respondents. The younger age group were willing to pay more.101

This is similar to that in a related study in Tanzania, where the age of household head appeared to affect WTP because seventy-four percent of respondents who were not willing to pay any amount had household heads who were aged fifty and above. Household heads that were 50 and above were willing to pay the least amount of premium with a mean amount of

373.2+264.8 Naira while household heads that were between the age of 30 to 39 had the highest with a mean amount of 622.3+209.7 Naira.40 In terms of age, a study in a rural community in Oshogbo, Nigeria found that a unit increase in age will increase WTP by 17% in the rural households in comparison to the urban areas where the younger age groups were more willing to pay and this was attributed to the younger age group being active and still productive whereas the elderly people usually retire at their villages after active services and are most times engaged in farming102; however a study in Kwara got contrary results.103

Gender: Another important factor that affects WTP is gender.42,46 It was noted that males were willing to pay higher amounts for insurance than females in Ghana as well as two different communities in Nigeria.46,101 Closely related is the finding in Namibia where thirty- one percent of individuals who live in male-headed households are insured compared with twenty-one percent of individuals living in female-headed households42 and a study in Benin

City where males show more willingness to participate in CBHI scheme than females.104 This is reverse to the finding in Ghana where seventy-eight percent of households who were not willing to pay anything for CHI had male household heads and twenty percent had female household heads.105 The mean amount of money that males were willing to pay was 504.9 +

260.0 naira and the mean amount that females were willing to pay was 624.4 + 283.4 naira.

Comparison of these mean amounts was statistically significant with a p-value of 0.005. Also

27 it has been reported that males were more willing to participate in rural health insurance scheme than the female headed households in rural areas of Nigeria.38

Marital status: Marital status has been shown to have a role in respondents’ willingness to pay. Recent research has shown a strong association between marital status and WTP with an increased likelihood among the married respondents.106 Contrary results were seen in a study in Jos, Nigeria where 93.7% of the single respondents were more willing to pay.18 However another study in India found no significant association between marital status and WTP.107

Educational level: The educational level of respondents also plays a significant role in their

WTP. There is positive correlation between educational attainment and WTP.35,46 People with more education had a higher WTP. Also the level of educational attainment of respondents had statistically significant influence on the premium that respondents were willing to pay.35,46 The mean amount of money that respondents were willing to pay increased with level of education. The more educated the respondents were, the higher the amount respondents were willing to pay and this was statistically significant7. Those with no formal education were willing to pay 396.4 + 238.7 Naira while the mean value that those with primary, secondary and tertiary education were willing to pay was 526.4 + 181.7, 620.4 +

252.3, and 730.9 + 268.8 Naira respectively. This is consistent with findings from previous studies where people with more education had a higher WTP. 7,35,38,46,103

Level of education was found to be important in the rural households, as a unit increase in the level of education increased WTP by 58% with use of cash i.e. the more educated in the rural communities the higher the WTP while for the use of cash this was found to be insignificant.102 In the urban households the higher the level of education the higher the WTP for the use of cash while for the use of commodities the reverse was the case i.e. the less educated were more willing to pay with commodities102 .

Income: The effect of income on their WTP was also statistically significant with a p-value of 0.000.45 Financial barrier was cited as one of the main reasons why households did not renew their membership.108 It was highlighted that the poor households found it difficult to pay for the scheme.108, 109 Studies show that the higher the income, the higher the willingness to pay. Less wealthy households or individuals were willing to pay lesser amounts.109 In most studies, income seemed to have a positive effect on WTP 45,108,109, however, there are some exceptions.104, 105 In a study in a rural community in Kwara, the mean value of each of the income group showed that the higher the income, the higher the WTP103, however in another study carried out in Oshogbo, income was also found to be significant with a negative signs which means a unit increase in income will decrease WTP by 53% i.e. the rich are less WTP.

The economic intuition behind this suggests that income is a very important variable in determining the demand for any good. This finding with respect to income has been the debate and argument about the WTP approach in health care valuation as the amount households or respondents are WTP is an increasing function of their ability to pay. In urban areas, the richer were less willing to pay showing that the rural areas are really in need of accessible and affordable health care. The urban household that were less willing still have access to health care although not adequate but they are not starved as rural households.102

Size of household: The mean WTP showed that those with household size of greater than 6 members were willing to pay the least amount of money.35,103 They were willing to pay 451.5

+ 263.7 Naira per annum per household member. However those that were willing to pay the highest amount of money were those with household size of 4 to 6 members willing to pay

566.0 + 235.1. This was statistically significant with a p value of 0.006.103 However contrary findings were seen in a Benin study104 where the respondents with larger families were found to be are more willing to participate in community based health insurance scheme than respondents with smaller households. This was attributed to the high financial burden faced

29 by large household when seeking health care services. This result agrees with the findings of other studies in rural areas where household size played a positive role in the willingness of people to participate in community based health insurance scheme.110, 111

Health status: The health status of people was found to influence their willingness to pay for

CBHI. In a study in 2003, there was a positive relationship between the health status of respondents and their willingness to participate in a CBHI scheme. Respondents with low or poor health status were more willing to participate in CBHIs than those with high or better health status.110 Contrary to this finding, another study in china found that rural dwellers with good health status were more willing to participate in CBHI than those in poor health.112

Distance to health facility: Distance to the nearest health facility was found to have a positive effect on WTP in some cases97 while in another study it had a negative effect.45

Distance to health centres was also found to be a contributing factor to WTP for CBHF among rural households with a unit increase in distance decreasing WTP by 18% respectively. A study from Ghana showed that distance to the nearest health facilities had the hypothesized sign and significance indicating that accessibility of health care will induce demand for community based health care financing.113 The willingness to pay increased with increase in distance to the health centre by 11% for the urban households which means they could only value the payment if the accessibility is not within the reach unlike the rural households.102

Mode of payment of premiums: The number in households preferring the use of cash as a payment choice was significant in urban households with the rural households preferring to pay with what they have.103 This finding is similar to that found in another study in Osun state where eight-one percent of rural households were WTP with commodities as compared to thirty-nine percent of the urban households while sixty–one percent of the urban households were WTP with cash as compared to eighteen percent of the rural households.114

Thus, the preferred mode of payment in the rural communities was the use of commodities while in the urban the preferred mode was cash. This was attributed to urbanization and increase economic activities in the urban areas where there may not be enough land for farming or other agricultural practices while in the rural areas the mainstay of their economy is agriculture.114

Membership of town unions/associations: This factor was also associated with WTP for

CBHI in a study in Benin, Nigeria were respondents who belonged to formal association were 3 times more likely to enrol in CBHIs than those who do not104 as those who are members of formal groups were familiar with the concept of weekly or monthly dues payment.

Level of trust in those organizing the scheme highly is very important in determining willingness to pay 44, 97, 104 as people are less likely to trust new initiatives especially if there has been mismanagement of previous initiatives and are therefore reluctant to invest their limited resources.

Other factors cited were attractiveness of the benefit package44, previous history of being involved in health insurance and history of large amount spent out of pocket (OOP) for health.46,101 Whilst previously paying out of pocket for health services was negatively related to willingness to pay, previously paying for health care using health insurance mechanism was positively related to willingness to pay.46

CHAPTER THREE

METHODOLOGY

Description of study area

The study was conducted in Lagos State which is an administrative division of Nigeria, located in the south western part of Nigeria and lies approximately on longitude 20 420E and

30 220E respectively and between latitude 600220N and 600 42N.115 Lagos state accounts for over five per cent of the national estimate115 with an estimated population total according to the official 2006 Nigerian Census of 9,013,534.87 Lagos State was created on May 27, 1967 and took off as an administrative entity in 1968 with Lagos Island serving the dual role of being the State and Federal Capital. However, with the creation of the Federal Capital

Territory of Abuja in 1976, Lagos Island ceased to be the capital of the State which was moved to Ikeja. Equally, with the formal relocation of the seat of the Federal Government to

Abuja on 12 December 1991, Lagos Island ceased to be Nigeria’s political capital.

Nevertheless, Lagos remains the centre of commerce for the country.116

Lagos State is divided into 20 Local Government Areas of which 16 comprise the urban

LGAs and the remaining four LGAs (Badagry, Ikorodu, Ibeju-Lekki and Epe) are classified as Rural LGAs.116 While the State is essentially a Yoruba-speaking environment, it is a socio- cultural melting pot attracting both Nigerians from different regions and foreigners alike.

There are two hundred and seventy six primary health care centres, twenty five state general hospitals, one state tertiary hospital distributed within the twenty local government areas in

Lagos. In addition are three federal tertiary centres, one federal medical centre and several privately owned health facilities in the state.87

Urban Local Government area

Surulere Local Government area

Surulere is a residential and commercial area, and an urban Local Government Area located on the Lagos mainland in Lagos State, Nigeria, with an area of 23 km². At the 2006 National census of 2006, there were 503,975 inhabitants, with a population density of 21,864 inhabitants per square kilometre.87 Surulere plays host to a large number of hospitality businesses which provides employment for some of its youth. Several other businesses exist in the general area fuelled by the proximity of the community to the Stadiums, Banks,

Markets and other small and medium scale enterprises in the area.87 Surulere is essentially inhabited by the Yoruba’s; prominent among them are the Ijebus, Egbas, Awories and Ilaje.

Other groups such as Oyo, Osun and Ekiti are represented in the Area. However, other Ethnic groups from the Eastern and Northern parts of the country are equally large in number in the community. There are several settlements in Surulere Local Government and these are delineated into 23 wards.87 There are eight primary health centres and 168 private hospitals in

Surulere Local Government area.87

Rural Local Government area

Ikorodu Local Government area

The history of Ikorodu LGA dates back to the period of British colonial administration in

Nigeria and has evolved through various reforms. It is the second largest of the Twenty LGAs in Lagos state. It is one of the four designated rural local government areas and is bordered by

Kosofe LGA on the west, Epe LGA on the east and the Lagos lagoon on the south. It covers a total area of 161.9 sq km with an estimated population of 527,917 according to the 2006

National census.117 The main occupations in this LGA is farming, hunting and fishing. Food

33 crops popular in this area are vegetables, melon, beans e.t.c. Fruits are also grown extensively. Trading and marketing has grown abundantly with modernisation and is a field dominated by the women folk. Majority of residents are of medium to low socio-economic status.117 Most residents are Ijebus, a Yoruba sub-group however a few non-indigenes reside in Ikorodu LGA namely the Fulani’s who are nomadic herdsmen; People from the middle belt and south-south Nigeria who are mostly wine-tappers and the Ghanaians who are also into fishing like the indigenes.117 Ikorodu Local Government area has a total of 30 wards which houses one General hospital, 19 primary health centres and 82 private health facilities.117

STUDY DESIGN

This study design is a comparative cross-sectional study to assess household willingness to pay for community based health insurance in urban and rural LGAs in Lagos

STUDY POPULATION

The respondents were heads of households in the urban and rural LGAs in Lagos state.

In the 2013 NDHS, a household was defined as a person or group of persons, related or unrelated, who usually live together in the same dwelling unit, have common cooking and eating arrangements, and acknowledge one adult member as the head of the household. A member of the household is any person who usually lives in the household.118

The head of a household is defined as any person who controls the maintenance of the household and exercises the authority to run the household.119

Inclusion criteria

 The person must aged eighteen or above.

 The resident must be a permanent resident of study area.

Exclusion Criteria

 Temporary residents or visitors of the study area were excluded.

 The three LGAs were CBHI has been piloted in Lagos state were excluded.

SAMPLE SIZE DETERMINATION

In determining the sample size, the formula for the comparison of proportions of two independent groups was used. n = (Zα + Zβ) 2 {P1 (1-P1) + P2 (1-P2)} 120

(P1-P2)2

Where: n = Minimum required sample size in each group.

Zα = Standard normal deviate (at 95% confidence interval) corresponding to the selected significance criteria = 1.96

Zβ = Standard normal deviate (at 80% power) corresponding to selected statistical power =

0.84

P1 = Proportion of people willing to pay for CBHI in urban community 54% or 0.54102

P2 = Proportion of people willing to pay for CBHI in rural community- 59.4% or 0.59121

P1 - P2= minimum expected difference between the 2 proportions n= (1.96+0.84)2{(0.59(1-0.59) + 0.54(1-0.54)}

(0.59- 0.54)2 n = 7.851 * 0.241+ 0.250

0.008 = 438

The calculated sample size was increased by 10%, 438* 0.1 = 43.8

Thus the desired sample size, n= 438+43.8=481.8, this was approximated to 480 each for the two groups.

Sampling Methodology

A multistage sampling technique was used to select the respondents.

Stage One

There are 16 urban and 4 rural LGAs in Lagos State. One Urban and one rural Local

Government each was selected from each by simple random sampling (balloting).

Surulere local Government was selected for the Urban LGA and Ikorodu LGA was selected for the Rural LGA

Stage Two

Two wards were selected by simple random sampling (balloting) from each selected local government.

The wards selected in Surulere local government area were Aralile and Gbaja/Obele Odan; while the wards selected in Ikorodu local government area were Agura/Gberigbe and

Ajaguro.

Proportional allocation of the sample sizes was done to the selected wards based on the population of each. Since it was not possible to get the population of the households in the respective wards, the population of the individuals in the wards was used as a proxy so as to proportionally allocate the sample sizes of these wards.

Urban: The population in Gbaja/Obele Odan is 24,140 while the population of Aralile is

21,103. The ratio of the two populations is 1: 1.1. Therefore 256 households were selected in

Gbaja/Obele Odan and 224 were selected in Aralile.

Rural: The population of Ajaguro is 66,964 while the population of Agura/Gberigbe is

20,219. The ratio of the two populations is 3.3: 1. Therefore 369 household heads were selected in Ajaguro and 111 were selected in Agura/Gberigbe.

Stage Three

The streets in each selected ward were enumerated and streets used were selected by simple random sampling from each selected ward.

Urban:

Gbaja / Obele Odan have a total of 59 streets. To select streets, 20 streets were selected by simple random sampling using the balloting method. If the selected street was non residential, another street was selected to replace it.

Aralile has a total of 26 streets. To select streets, 20 streets were selected by simple random sampling using the balloting method. If the selected street was non residential, another street was selected to replace it.

Rural:

Ajaguro has a total of 268 streets. To select streets sampled, 20 streets were selected by simple random sampling using a table of random numbers. If the selected street was non residential, another street was selected to replace it.

Agura/Gberigbe has 35 streets; however they were not well defined and formed 5 clusters of houses. Two clusters were randomly selected by balloting method.

Stage Four

To select houses, the number of houses required per street was gotten by splitting the sample size for each ward by the total number of selected streets per ward.

Urban

In Gbaja/Obele Odan: Each street had an average of 15 to 25 houses per street. Thirteen houses were randomly selected by balloting method on each of the selected 20 streets. Where there was more than one household living in a building, one of them was then selected by simple random sampling for questionnaire administration. Where the household head in the selected house was absent, another house was selected.

In Aralile: Each street had an average of 20 to 30 houses per street. Eleven houses were selected by simple random sampling on each of the selected 20 streets. However 4 extra houses were selected on Hogan Bassey Street (with 60 houses) as it was the longest street in the ward to get the required 224 number of respondents needed in that ward. Where there was more than one household living in a building, one of them was then selected by simple random sampling for questionnaire administration. Where the household head in the selected house was absent, another house was selected.

Rural

In Ajaguro: the streets had an average of 25 to 40 houses per street. Eighteen houses were randomly selected by balloting method on each of the selected 30 streets and 9 houses were randomly selected on the 31st street to get the required 369 number of respondents needed in that ward. Where there was more than one household living in a building, one of them was then selected by simple random sampling for questionnaire administration. Where the household head in the selected house was absent, another house was selected.

In Agura/Gberigbe, The required households of 110 were split equally between the two clusters. The 55 houses in each of the two clusters where visited successively until the required sample size was reached. Where there was more than one household living in a building, one of them was then selected by simple random sampling for questionnaire

38 administration. Where the household head in the selected house was absent, another house was selected.

Data collection tool and technique

Data collection tool

A pre-tested, semi-structured, interviewer administered questionnaire (Appendix 1) was used to collect data from houses. The questionnaire was adapted from the contingent valuation: a user’s guide and from literature46, 99 to suit the local context and for appropriateness of language. The questionnaire was translated Yoruba and back translated to English by a graduate of Yoruba language to suit the local language in the study area. It is in the dichotomous choice format with open ended questions to account for uncertainty in respondents responses and a bidding technique was used to elicit respondents’ willingness to pay for the proposed CBHI scheme.

The questionnaire included a short introduction explaining the reasons for conducting the study and consisted of four sections;

Section I: information about the sociodemographic characteristics of the respondents and their household was elicited. It included their gender, age, household size, marital status, highest level of education, household source of income and estimated monthly household income, Membership of a financial group or association.

Section II: It included questions focusing on information related to respondents and their household’s health status, health seeking behaviours, payment options, and level of health insurance coverage.

Section III: Assessed the respondents’ perception of CBHI and the opportunities it offers to improve payment for healthcare services and their acceptability of CBHI.

Section IV: Assessed respondent’s willingness to enrol in and pay for a proposed community based health insurance scheme for themselves and their household members and the factors that determine their willingness to pay as well as the amount they were willing to pay. The proposed scheme and benefits included Out-patient care including necessary consumables;

Essential drugs and diagnostic tests; Maternity care for up to four live births; Preventive care such as immunization, health education, family planning, antenatal care and postnatal care;

Consultation with specialists; Hospital care in a standard ward for a limited stay of cumulated

45 days per year; Eye examination and care which excluded provision of glasses or lens;

Preventive dental care and pain relief. They would be required to pay a premium in order to join the scheme and this could be paid monthly or once a year, however, no credit would be allowed. After payment, they receive services free of charge at the point of consumption.

Each registered person will be issued with an identity cared renewable yearly after payment of premium. Healthcare would be delivered by providers in the area which is registered with the CBHI management group and the scheme will be managed by a committee that the community will establish, supported by the state government. At the end of 12 months they could renew or drop out of the scheme.

The questionnaire also included the valuation scenario; which is the most important part of the CV survey. Guidelines for a valid contingent valuation analysis80 were followed as much as possible. Here, the scenario of a community health care prepayment scheme was presented to the respondent and was described in detail including the nature of the scheme, the organization, the membership criteria, and the expected benefits. Thereafter they were asked whether he or she would be willing to pay for the proposed bid. The bids were three different amounts presented to the respondent in decreasing order. The start bid amount was chosen based on the amount that was used in the pilot schemes in Lagos state. The second and third bids were chosen and modified from that used in literarure.1

Pretesting of the data collection tool

The questionnaire was pre-tested among 30 household heads in Oregun, a ward in Ikeja.

Leading and confusing questions were identified and revised in the questionnaire of the study. Also observed that some respondents did not understand some of the questions in

English and this prompted translation of the questionnaire into Yoruba. Pretesting also enabled observation of and rectifying gaps in the understanding of the questionnaire by the research assistants.

Training of Research Assistants

They received a 2-day training moderated by the researcher which lasted for 2 hours each day so as to ensure their mastery of the questionnaire and to achieve standardization in completing the questionnaires. The training method used was lecture/discussion in which the trainees were first quizzed on what they already knew about health insurance. They were then trained on healthcare financing options, health insurance and the concepts of community based health insurance. Also ensured at intervals that the trainees were following and understood the lecture.

Data collection

Data was collected by the researcher and 5 trained research assistants who are graduates of the school of health information technology and are fluent in reading and speaking Yoruba language. The complete questionnaire took approximately 25 minutes to fill per respondent.

Data was collected over a period of 3 months from February to April 2014.

DATA MANAGEMENT

Data entry and analysis was done using the Statistical Package for Social Sciences software

(SPSS) version 17. Discrete variables were presented as percentages while continuous variables were expressed as means + standard deviation. The proportions were compared using Pearson Chi square, while the different means were compared using Students t-test.

Level of significance is predetermined at P-value of 0.05. Variables attaining significance were further subjected to multivariate analysis using logistic regression to determine the predictors for willingness to pay. Results were expressed with 95% confidence intervals.

A grading system based on the responses of the respondents was adopted to assess the perception of the respondents about CBHI. A grading system based on the responses of the respondents was used to assess the perception of the respondents about CBHI. There were six statements to assess perception with each of them having 3 options of high, medium or low and each option was awarded a mark of 3, 2 and 1 respectively.122 The highest obtainable score was 18 and was converted to percentages. Scores were graded as poor, indifferent and good. Scores from 0-33% were graded poor, 34-66% as indifferent and 67-100% as good. To assess acceptability of CBHI, the highest level of acceptability was scored 10. Poorly acceptable was awarded to respondents with scores 1-3, acceptable to scores 4-7 and fully acceptable to scores 8-10.122

ETHICAL CONSIDERATION

Ethical approval for this study was obtained from the ethics and research committee of the

Lagos University Teaching Hospital. Permission was sought from the Chairman of the selected local government areas via their medical officers of health. Permission was also sought from the CDA chairmen of the selected wards. Informed consent was obtained from the respondents before the interview. The nature of study, participation status, benefits of the

42 study and confidentiality issues was made clear to the respondents before obtaining their consent. Those participating in this research were assured of the highest level of confidentiality on information given. Respondents were also informed of their right to opt out of the study if they choose to at any point of the study. Following the appropriate approvals, questionnaires were then administered.

CHAPTER FOUR

RESULTS

In this study, 480 household heads in the urban and 480 household heads in the rural LGA were found to be eligible and took part in the study.

Table 1: Demographic characteristics of respondents

Variables Urban n=480 Rural n=480 χ2 df p –value Freq (%) Freq (%) Age (years) 18 – 29 65 (13.5) 63 (13.1) 2.78 4 0.596 30 – 39 141 (29.4) 135 (28.1) 40 – 49 149 (31.0) 142 (29.6) 50 – 59 80 (16.7) 90 (18.8) > 60 45 (9.4) 50 (10.4) Mean age 43.17 ± 13.37 47.22 ± 11.31 t= 5.66 0.810 Sex Male 321 (66.9) 349(72.5) 3.87 1 0.049* Female 159 (33.1) 131(27.3) Marital status Single 30 (6.3) 22 (4) 12.34** 0.013* Married 393 (81.9) 391 (81.5) Separated 23 (4.8) 20 (4.2) Divorced 6 (1.3) 0 (0) Widowed 28 (5.8) 47 (9.8)

Household size 1-3 123(25.6) 110 (22.9) 4.24 2 0.120 4-6 255 (53.1) 289 (60.2) >7 102 (21.3) 81 (16.9) Mean household size 4.8 + 2.1 5.0+2.0 t=1.82 0.106 *p<0.05 = statistically significant **Fishers exact test

In the table above, the proportion of the respondents in the urban (31.0%) and the rural

(29.6%) were mostly within age range 40-49 years with a mean age in the urban and rural being 43.17+ 13.37 and 47.22 ± 11.31 years respectively. A large proportion of the household heads in the two groups were males with a proportion of 66.9% and 72.5% in the urban and rural LGAs respectively. A large proportion of the respondents in the urban (81.9%) and in the rural (81.5%) were married. This difference was statistically significant (p=0.013)

Majority of the household heads in the urban (66%) and the rural (70.8%) were male heads of household. The differences in terms of their demographic characteristics were all statistically significant except the difference in their age group and household size.

Table 2: Socio-economic characteristics of respondents

Variables Urban n=480 Rural n=480 χ2 df p-value Freq (%) Freq (%) Highest Level of education No formal schooling 22 (4.6) 36 (7.5) 12.42 4 0.027* Primary 71 (14.8) 144 (30.0) Junior secondary 45 (9.4) 72 (15.0) Senior secondary 227 (47.3) 161 (33.5) Tertiary 115 (24.0) 67 (14.0)

Occupation Professional 13 (2.7) 11 (2.3) 49.56 4 <0.001* Skilled 178 (37.1) 138 (28.8) Semi-skilled 109 (22.7) 59 (12.3) Unskilled 161 (33.5) 264 (55.0) Others 19 (4.0) 8 (1.7)

Estimated household income per month(Naira) <5000 23 (4.8) 308 (64.2) 44.42** <0.001* 5000-10000 91 (19.0) 87 (18.1) 10001-20000 116 (24.2) 65 (13.5) 20001-30000 83 (17.3) 15 (3.1) >30000 167 (34.8) 5 (1.0)

Estimated expenditure on food in the last month(Naira) < 5000 179 (37.3) 272 (56.7) 50.66** <0.001* 5000-10000 222 (46.3) 156 (32.5) 10001-15000 37 (7.7) 34 (7.1) 15001-20000 25 (5.2) 18 (3.8) 20001-25000 5 (1.0) 0 >25000 12 (2.5) 0 Mean 9485.42 ± 7442.29 ± 3333.25 t=5.95 0.002* 4989.58 Membership of a financial association/group

Yes 94 (19.6) 180 (37.5) 37.77 1 0.001* No 386 (80.4) 300 (62.5)

*p<0.05 = statistically significant **Fisher’s exact test

A higher proportion of the household heads in the urban (47.3%) and the rural area (33.5%) had secondary education (p=0.027).The respondents in the urban and rural LGA had an average of 4-6 people in their household at 53.1% and 60.2% respectively. The mean household size in the urban is 4.8+ 2.1 while that in the rural is 5.0 + 2.0. This difference was not statistically significant. A higher proportion of the respondents in the urban LGA were mostly skilled workers (37.1%) while those in the rural were mostly unskilled (55.0%). The urban household head (34.8%) made higher amount of money monthly (>30000 naira) in comparison to the rural heads (64.2%), who made a lesser amount monthly (<5000 naira)

(p=<0.001). The proportion of their income that the respondents spent on food was 20.2% and 63.8% in the urban and rural areas respectively.

A large proportion of the respondents in the rural (37.5%) as compared to the urban (19.6%) belonged to a financial group/association where money is managed for them. The differences in terms of their socio-economic characteristics were all statistically significant.

Table 3: Health seeking behaviour of the household

Variables Urban n=480 Rural n=480 χ2 df p value Freq (%) Freq (%) Reported illness in last 3 months Yes 221 (46.0) 240 (50.0) 10.42 4 0.029* No 259 (54.0) 240 (50.0)

Type of illness n=221 n=240 Malaria 134 (60.6) 148 (61.7) 5.08 4 0.279 Typhoid 24 (10.9) 34 (14.2) Diarrhoea 13 (5.9) 16 (6.7) RTI 24 (10.9) 27 (11.3) Others 26 (11.8) 15 (6.3) Where treatment was sought for illness Chemist 64 (29.0) 89 (37.1) 12.81** 6 0.046* Traditional healer 31 (14.0) 45 (18.8) PHC 65 (29.4) 57 (23.8) General hospital 14 (6.3) 19 (7.9) Private hospital 40 (18.1) 28 (11.7) Tertiary hospital 1 (0.5) 0 Others 6 (2.7) 2 (0.8) Form of transportation to facility Walked 61 (25.2) 121 (50.6) 22.54** <0.001* Okada/keke marwa 13 (5.9) 41 (17.2) Bus 100 (47.6) 72 (30.1) Taxi 13 (5.9) 5 (2.1) Personal vehicle 34 (15.4) 2 (1.2) Travel time to facility <15 mins 51 (23.1) 139 (58.4) 3.85** 0.258 15-30 mins 108 (48.9) 56 (23.5) 30-1hour 59 (26.7) 34 (14.3) >1hour 3 (1.4) 9 (3.8) Recovery after treatment Yes 196 (89.1) 214 (89.2) 0.00 1 0.993 No 25 (10.9) 26 (10.8) *p<0.05 = statistically significant **Fisher’s exact test

Table 3 shows the health seeking behaviour of the households. About half of the respondents reported an illness in their household with a proportion of 46% in the urban and 50% in the rural households in the last 3 months (p=0.029). Equal proportion of them in urban (60.6%) and rural (61.7%) reported that they had malaria. A higher proportion of the urban household members first sought treatment at the PHCs (29.4%) as compared to the rural who first sought treatment from chemists/patent medicine vendors (37.1%). Walking was the main form of transport in the urban (45.2%) and rural (50.6%). The time taken to get to the first point of seeking treatment was 15-30 mins in the urban (48.9%) compared to less than 15 minutes in the rural (58.4%). Equal proportion reported recovery for illness in their household after treatment among the urban (89.1%) and rural (89.1%) respondents. The differences between the respondents in terms of their health seeking behaviour were statistically significant except the difference in the type of illness, travel time to facility and recovery after first treatment.

Table 4: Amount spent on treatment by household head on most recent illness in the household in last 3 months

Variables Urban n=221 Rural n=240 χ2 df p-value Freq(%) Freq(%) Amount spent on treatment(Naira) >1000 98 (44.3) 143 (59.6) 13.69 3 0.003* 1001-5000 101 (45.7) 73 (30.4) 5001-10000 14 (6.3) 11 (4.6) 10001-20000 8 (1.7) 13 (5.4) Mean 4607.01+1500.75 3977.92+1110.18 t=2.04 0.035*

Amount spent on transport to health facility(Naira) 0-500 202 (91.4) 225 (93.8) 7.86** 0.016* 501-1000 19 (8.6) 10 (4.2) 1001-2500 0 (0) 1 (0.4) 2501-5000 0 (0) 4 (1.7) Mean 570.83+362.84 326.47+196.55 t=0.91 0.362

Total cost spent(treatment+ transport ) (Naira) >1000 75 (33.9) 137 (57.1) 33.43 <0.001* 1001-5000 114 (51.6) 73 (30.4) 5001-10000 22 (10.0) 17 (7.1) 10001-20000 6 (2.7) 13 (5.4) 20001-30000 4 (1.8) 0 (0) Mean 4832.35+1615.69 4234.17+1565.65 t=1.79 0.047*

Total amount spent on n= 25 n=26 further treatment (Naira) >5000 1 (4.0) 1 (3.8) 16.28 0.001* 5001-10000 7 (28.0) 20 (76.9) 10001-20000 7 (28.0) 4 (15.4) 20001-30000 8 (32.0) - 30000-40000 2 (8.0) 1 (3.8) Mean 12980+5159.61 6661.54+2782.31 t=3.72 <0.001* *p<0.05 = statistically significant **Fisher’s exact test

The table above shows the amount spent by the household heads on the household for their health in the last quarter. A higher proportion of the respondents in the urban (45.7%) had spent between 1000 and 5000 naira as compared to the rural where 59.6% had spent less than

1000 naira. The mean amount spent on treatment in the urban and rural was 4607.01+

1500.75 and 3977.92+1110.18 naira respectively. This difference was statistically significant

(p=0.035). An almost equal proportion of those in the urban (91.4%) and rural (93.8%) had spent less than 500 naira on transporting themselves to the health facility. The mean amount spent on transport was not statically significant with 570.83+362.84 naira spent by the urban and 326.47+196.55 naira by the rural respondents.

Similarly the total cost spent (treatment + transport) by the respondents showed that the urban

(51.6%) had spent between 1000 and 5000 naira as compared to the rural (57.1%) that had spent less than 1000 naira. This is reflected in the mean cost of treatment being more among the urban household members at 4832.35+1615.69 naira and 4234.17+1565.65 naira among the rural household members.(p=0.047)

Among those that did not recover following the first treatment, the rural (76.9%) spent between 5000 and 10000 naira on treatment till they recovered while the urban (32%) spent between 20000 and 30000 naira till recovery. A statistically significant difference was seen in the mean total cost of treatment till recovery with the cost in the urban at 12980+5159.61 naira being almost double the cost the cost among the rural at 6661.54+2782.31 naira(p=<0.001)

Table 5: Household heads payment source for treatment of most recent illness in the household

Variables Urban n=221 Rural n=240 χ2 df p-value Freq(%) Freq(%)

Mode of Payment Cash 181 (81.9) 219 (91.3) 10.75** 0.033* Employer 13 (5.9) 1 (0.4) Insurance 19 (8.6) 6 (2.5) In kind 5 (2.3) 4 (1.7) Instalment 3 (1.4) 10 (4.2)

Source of funds for payment Own money 176 (79.6) 196 (81.7) 17.81** 0.029* Borrowed/Loaned 4 (1.8) 20 (8.3) Sold household movable assets 2 (1.9) 0 (0) Sold Family Land/property 4 (1.8) 3 (1.3) Community solidarity/ someone else 16 (6.8) 7 (2.9) paid Health insurance 19 (4.0) 6 (2.5) Payment was subsidised/deferred 1 (0.5) 4 (1.7) Others 5 (1.8) 4 (1.7)

Reason for Health Insurance n=19 n=4 Financial protection 6(31.6) 0(0) 1.35** 0.898 Access to affordable healthcare 1 (5.3) 1 (16.7) Compulsory 9 (47.4) 5 (83.3) Work/Relative, Friend advised to 3 (15.8) 0 (0) enrol *p<0.05 = statistically significant **Fishers exact test

Table 5 shows the payment mechanism for health among the urban and rural respondents.

Majority of the urban (81.9%) and the rural respondents (91.3%) used cash as the form of payment for illness in their household. Among them, majority of the respondents in the urban

(79.6%) and the rural (81.7%) coped by using their own money to pay for their healthcare service. The main reason given for use of health insurance by respondents’ who used health insurance to pay for illness in their household in the urban (47.4%) and rural (83.3%) was because it was compulsory. The differences in payment and payment coping mechanisms were all statistically significant except for the reasons for using health insurance.

Table 6: Awareness of health insurance among the respondents

Variables Urban n=480 Rural n=480 χ2 df p–value Freq (%) Freq (%)

Ever heard of health insurance

Yes 197 (41.0) 69 (14.4) 85.20 1 <0.001* No 283 (59.0) 411 (85.6) n=197 n=69 Health insurance is defined as

Prepayment for healthcare 156 (79.2) 13 (18.8) 16.33** 4 0.002* Paying tax to Govt 16 (8.1) 40 (58.0) Free healthcare by Govt 20 (10.2) 9 (13.0) Others 4 (2.0) 3 (4.3) Don’t know 1 (0.5) 4 (5.8) Types of health insurance known

NHIS 67 (34.0) 28 (40.6) 1.838** 4 0.786 CBHI 3 (1.5) 0 (0) Private Health Insurance 103 (52.3) 35 (50.7) Others 21 (10.7) 5 (7.2) Don’t know 3 (1.5) 1 (1.4)

*p<0.05 = statistically significant **Fishers exact test

In the table above, more of the urban household heads in (41.0%) than the rural (14.4%) had heard of health insurance. The difference was statistically significant (p=<0.001). Majority of the urban 70.2% knew health insurance to be prepayment for healthcare compared to the rural where the proportion (18.8%) was much lower. This difference statistically significant. The urban (52.3%) and the rural (50.7%) knew private health insurance as a type of health insurance. Higher proportion of the rural (40.6 %) than the urban (34.0%) mentioned

National health insurance scheme (NHIS) as a type of health insurance. However this difference was not statistically significant.

Table 7: Perception of community based health insurance among the respondents

Variables Urban Rural χ2 df p–value n=480 n=480 Freq (%) Freq (%) Perceived ability of CBHI to make health more affordable Low 75 (15.7) 64 (13.4) 18.49 3 <0.001* Medium 210 (43.8) 182 (37.9) High 195 (40.6) 234 (48.8) Perceived potential of increasing access to affordable healthcare Low 87(18.1) 53 (11.1) 25.34 2 <0.001* Medium 227 (47.3) 261 (54.4) High 166 (34.6) 166 (34.6) Perceived potential to improve household health seeking behaviour Low 82 (17.0) 70 (14.6) 12.53 2 0.002* Medium 219 (45.6) 256 (53.3) High 179 (37.3) 154 (32.1) Potential to improve quality of services provided by Low 69 (13.4) 76 (15.8) 20.18 2 <0.001* Medium 245 (51.0) 238 (49.6) High 166 (34.6) 166 (34.6) Potential to ensure constant availability of drugs at facilities Low 89 (18.6) 116 (24.2) 21.13 2 <0.001* Medium 209 (43.5) 245 (51.0) High 181 (37.7) 119 (24.8) Perceived confidence in committee managing pooled funds in community Low 228 (47.5) 146 (30.4) 42.49 2 <0.001* Medium 146 (30.4) 283 (59.0) High 106 (22.1) 51 (10.6)

*p<0.05 = statistically significant **Fishers exact test

The respondents were divided in their perception of community based health insurance. The respondent’s perception of the potential ability of CBHI to make health affordable among the rural (48.8%) was high while that of the urban (43.8%) was medium. The respondent’s perception of the potential level of accessing affordable healthcare due to CBHI among the urban (47.3%) and the rural respondents (54.4%) was medium. The perception of CBHIs potential to improve household health consumption patterns was medium among the urban household head (45.6%) and the rural household heads (53.3%). The perception of CBHIs potential to improve quality of health services given in health care institutions was medium in

51.0% of the urban and 49.6% of the rural household heads. The perception of the potential of CBHI to ensure constant drug availability at facilities among the respondents in the urban

(43.5%) and majority of the rural (51%). was medium Majority of the respondents in the urban (47.5%) had a low perception of their funds being pooled and managed by community and a medium level of perception amongst the rural (59.0%). The differences in their perception among the household heads towards health insurance were all statistically significant.

Table 8: Level of perception of respondents to community based health insurance

Variables Urban n=480 Rural n=480 χ2 df P-value

Freq (%) Freq (%)

Poor 23 (6.0) 26 (5.4) 1.09 2 0.579

Indifferent 213 (44.4) 200 (41.7)

Good 238 (49.6) 254 (52.9)

*p<0.05 = statistically significant

The table above shows that a higher proportion of urban (49.7%) and the rural (52.9) had a good perception of community based health insurance. However there was no statistically significant difference in the respondents perception of community based health insurance in the urban and rural LGA (p=0.579).

Table 9: Acceptance of community based health insurance among the respondents in the urban and rural LGA

Variables Urban n=480 Rural n=480 χ2 df p value

Freq (%) Freq (%)

Yes 385 (80.2) 444 (92.5) 30.77 1 <0.001*

No 95 (19.8) 36 (7.5)

*p<0.05 = statistically significant

Table 9 shows a comparison of the respondent’s acceptance of CBHI as a strategy for paying for their health. A higher proportion of the rural respondents (86.5%) found CBHI acceptable as a strategy for paying for their health compared to their urban counterparts (73.8%). There was a statistically significant difference in the two groups (p=<0.001).

Table 10: Level of acceptance of community based health insurance among the respondents in the two groups

Variables Urban n=385 Rural n=444 χ2 df p-value Freq (%) Freq (%)

Poorly 67 (17.4) 51 (11.5) 26.64 2 <0.001* acceptable

Acceptable 267 (69.4) 272 (61.3)

Highly 51 (13.2) 121 (27.3) acceptable

*p<0.05 = statistically significant

The table above shows that majority of the urban respondents (69.4%) and the rural (61.3%) found CBHI acceptable as a strategy for paying for their health care. A higher proportion of rural (27.3%) found CBHI highly acceptable in comparison to the urban (13.2%). A higher proportion of urban (17.4%) found CBHI poorly acceptable as a strategy for paying for health care compared to the rural (11.5%). There was a statistically significant difference found between the two groups.

Table 11: Reasons for non-acceptance of community based health insurance among respondents

Variables Urban n=95 Rural n=36 χ2 df p- Freq (%) Freq (%) value

No confidence in scheme 25 (26.3) 9 (25.0) 8.52** 0.13 managers

Always in good health 19 (20.0) 14 (38.9)

Do not want the policy 17 (17.9) 5 (13.9)

Cannot afford the premium 23 (24.2) 7 (19.4)

Don’t know 10 (10.5) 0 (0)

Others 1 (1.1) 1 (2.8)

*p<0.05 = statistically significant **Fishers exact test

The table above shows a comparison of the respondent’s reasons for not accepting CBHI as a strategy for paying for their health. The commonest reason given by the respondents in the urban (26.3%) was lack of trust in the managers of the scheme while the reason given by the rural was they were always in good health (38.9%). The difference in their reasons for not accepting CBHI among respondents in the two groups was not statistically significant.

Table 12: Respondents willingness to pay for a hypothetical CBHI package for their household

Variables Urban n=480 Rural n=480 χ2 Df p-value

Freq (%) Freq (%)

Willing to pay for 354 (73.8) 415 (86.5) 24.32 1 <0.001* CBHI

Not willing to pay 126 (26.2) 65 (13.5) for CBHI

*p<0.05 = statistically significant

In the table above, a higher proportion of the rural respondents (86.5%) were willing to pay for the hypothetical community based health insurance scheme compared to the urban

(73.8%) respondents. This difference between the two groups was found to be statistically significant (p=<0.001)

Table 13: Respondents reasons for unwillingness to pay for the hypothetical community based health insurance scheme

Variable Urban=126 Rural n=65 χ2 df p value

Freq (%) Freq (%)

Don’t need health 19 (15.1) 19 (29.2) 26.62** 5 < 0.001* insurance

Healthy/Rarely ill 32 (25.4) 10 (15.4)

No confidence in the 52 (41.3) 12 (18.5) scheme

Don’t think I can afford 10 (7.9) 22 (33.8) the premium

I already have health 10 (7.9) 2 (3.1) insurance

*p<0.05 = statistically significant **Fishers exact test

The table above shows that the reasons for unwillingness to pay for CBHI. The commonest reason given by the respondents in the urban (41.3%) was no confidence in the scheme while the reason given by the rural was that they felt they could not afford the premium (33.8%).

The difference in their reasons for unwillingness to pay was statistically significant

(p=<0.001).

Table 14: Respondents willingness to bids as premium for CBHI for individual enrolees per month

Variable Urban n=354 Rural n=415 χ2 df p-value Freq (%) Freq (%) Willingness to pay start bid of 500 naira Yes 325 (74.4) 364 (89.4) 30.83 1 <0.001*

No 28 (25.6) 51 (10.6)

Willingness to pay n=28 n=51 second bid of 450 naira Yes 0 (0) 0 (0) 7.91** 1 0.005*

No 28 (100) 51 (100)

Willingness to pay 350 naira Yes 4 (12.9) 1 (2.0) 2.53** 1 0.272

No 24 (87.1) 50 (98.0)

*p<0.05 = statistically significant **Fishers exact test

The table above shows the proportion of respondents in the urban and rural willing to pay the starting bid premium of 500 naira per month. More of the rural (89.4%) compared to urban

(74.4%) were willing to pay the starting bid premium of 500 naira for individual enrolees

(p=<0.001). Among the respondents that were unwilling to pay the first bid of 500 naira, all in the rural (100%) and urban (100%) were not willing to pay the second bid premium of 450 naira for individual enrolees. The difference between the groups was statistically significant

(p=0.005). The table also shows the proportion of respondents who had refused the first and second bid in the urban and rural, willing to pay the third bid of 350 naira as premium per month. An equal proportion among the rural (2%) and the urban (12.2%) were willing to pay the third bid of 350 naira as premium for individual enrolees. This difference in the groups was not statistically significant.

Table 15: The maximum amount respondents are willing to pay as premium for individual enrolees per month

Variable Urban n=28 Rural n=51 χ2 df p –value Freq (%) Freq (%) Maximum amount willing to pay (Naira) 0 – 100 10 (32.3) 10 (19.6) 8.97** 0.018* 101 – 200 7 (22.6) 28 (54.9)

201 – 300 12 (38.7) 12 (23.5)

301 – 400 2 (6.5) 1 (2.0)

Mean + standard 204.84+103.57 194.44+78.38 t=0.37 0.003* deviation Final maximum amount willing to pay (Naira ) <100 3 (9.7) 11 (21.6) 2.17** 0.545 101 -200 8 (25.8) 17 (33.3) 201 – 300 14 (45.2) 20 (39.2) 301- 400 6 (19.4) 3 (5.9) Mean + standard 253.23+94.81 219.44+92.58 t=1.21 0.225 deviation *p<0.05 = statistically significant **Fishers exact test

The table above shows the maximum amounts that respondents who were unwilling to pay the starting bid of 500 naira were willing to pay as a premium per month. A higher proportion of the urban (38.7%) were willing to pay 201-300 naira while in the rural; 54.9% were willing to pay 101- 200 naira. This difference was statistically significant.

The table also shows the final maximum amount that those unwilling to pay the starting bid of 500 naira were willing to pay as a premium per month. Majority of the urban (45.2%) and

65 the rural (39.2%) were only willing to pay 201-300 naira. This difference was statistically significant between the two groups. The mean amounts that the respondents are willing to pay were 253.23+94.81 in the urban and 219.44+92.58 naira in the rural. However this difference was not statistically significant (p=0.225).

Table 16: The premium all respondents are willing to pay for the scheme for individual enrolee per month in case of inflation

Variable Urban n=354 Rural n=415 χ2 df p – Freq (%) Freq (%) value Amount willing to pay considering inflation

(Naira)

<250 116 (24.2) 47 (9.8) 50.60** 0.003*

251 – 500 155 (32.4) 152 (50.2)

501 – 750 155 (32.3) 241 (31.7)

751 – 1000 51 (10.6) 39 (8.1)

1001 – 1250 1 (0.2) 0 (0)

1251- 1500 2 (0.4) 1 (0.2)

Mean + standard deviation 555.23+221.01 542.19+317.67 t=3.38 0.001*

*p<0.05 = statistically significant **Fishers exact test

The table above shows the amount that all respondents were willing to pay for a single enrolee in case of inflation as a premium per month. A higher proportion of respondents in the rural (50.2 %) were willing to pay 251 – 500 naira while in the urban; an almost equal proportion (32.3%) and (32.4%) were willing to pay 251-500 and 501 –750 naira in the urban and rural LGA respectively. The mean amount that the respondents were willing to pay for individual enrolees was 555.23+221.01 in the urban and 542.19+317.67 naira in the rural.

This difference was statistically significant (p=0.001)

Table 17: Respondents willingness to bids (900, 850 and 750 naira) as premium for their household in the proposed scheme

Variable Urban n=354 Rural n=415 χ2 df p –value Freq (%) Freq (%)

WTP the starting bid of 900naira Yes 285 (84.6%) 373 (91.3%) 75.21 1 <0.001* No 68 (15.4%) 42 (8.8%)

WTP the second bid of n=68 n=42 850 naira Yes 0 (0) 0 (0) 0.407** 0.52 No 68 (100) 42 (100)

WTP third bid of 750 naira Yes 14 (19.4) 0 (0) 1.42** 1 0.23 No 54 (80.6) 42 (100) *p<0.05 = statistically significant **Fishers exact test The table above shows the proportion of respondents in the urban and rural willing to pay the starting bid premium of 900 naira per month for their households. Majority of the respondents in the rural (84.6%) and the urban (91.3%) were willing to pay the starting bid premium of 900 naira for their households’ monthly premium. This difference was statistically significant between the two groups (p=<0.001).

The table also shows the proportion of respondents who had refused the first bid in the urban and rural willing to pay the second bid premium of 850 naira per month. An equal proportion among the rural (100%) and the urban (100%) were not willing to pay the second bid premium of 850 naira for their households. This difference was not statistically significant. A

68 third bid of 750 naira was offered to the respondents that refused the second bid. Of these respondents, 19.4% in the urban and none of the respondents in the rural were willing to pay the third bid. This difference was not statistically significant between the two groups.

Table 18: Maximum amount respondents were willing to pay (wtp) (those that refused the bids) for household

Variable Urban n=68 Rural n=42 χ2 df p –value Freq (%) Freq (%) Maximum amount willing to pay for households (Naira) 27 (39.3) 26 (66.7) 8.15** 0.012* 0 – 250

251 – 500 31 (42.6) 12 (22.2)

501 – 750 11 (18.0) 4 (11.1)

Mean + standard 510.00+107.24 420.85+254.92 t=1.33 0.187 deviation

Final maximum amount wtp for households (Naira) 0 – 200 23 (30.7) 9 (38.9) 10.05** 0.024*

201 – 400 10 (13.3) 24 (44.4)

401- 600 24 (32.0) 8 (11.1)

601- 800 14 (18.7) 1 (5.6)

801-1000 4 (5.3) 0 (0)

Mean + standard 506.67+179.15 437.33+271.15 t=0.95 0.354 deviation

*p<0.05 = statistically significant **Fishers exact test

The table above shows the maximum amount that those unwilling to pay the starting bid of

900 naira were willing to pay as a premium per month for their household. Majority of the

70 rural (66.7%) were willing to pay less than 250 naira while majority of the urban (42.6%) were willing to pay between 251 and 500 naira. This difference was statistically significant.

The table also shows the final maximum amount that those unwilling to pay the starting bid of 900 naira were willing to pay as a premium per month. A higher proportion of the respondents in the urban (32.0%) were only willing to pay 401- 600 naira while a higher proportion of the respondents in the rural (44.4%) were willing to pay between 201 and 400 naira. This difference was statistically significant. The final mean amount reported by the respondents in the urban was 506.67+179.15 naira and in the rural were 437.33+271.15 naira.

However the difference in the two groups was not statistically significant

Table 19: The premium all respondents were willing to pay for their household per month in case of inflation

Variable Urban n=354 Rural n=415 χ2 df p –value (Naira ) Freq (%) Freq (%)

< 500 142 (29.6) 49(10.2) 64.80** 0.001*

501 – 1000 277 (57.7) 331 (69.0)

1001 -1500 54 (11.3) 82 (17.1)

1501 – 2000 4 (0.8) 15 (3.1)

2001 – 2500 1 (0.2) 2 (0.4)

2501 – 3000 2 (0.4) 1 (0.2)

Mean + standard 975.56+408.45 726.83+498.99 t=8.37 <0.001* deviation

*p<0.05 = statistically significant **Fischer’s exact test

The table above shows the amount that all respondents were willing to pay for their households in case of inflation as a premium per month. A higher proportion of the respondents in the urban (57.7 %) and the rural (69.0 %) were willing to pay 501 – 1000 naira. This difference was statistically significant. The mean amount reported by the respondents that they would pay while putting inflation into consideration was

975.56+408.45naira in the urban and 754.83+498.99 naira in the rural. The difference in the means was statistically significant (p=<0.001).

Table 20: Association between respondents’ gender and their willingness to pay for

CBHI

Variable Urban n=480 Rural n=480 Willingness to pay Willingness to pay Gender Yes No Total Yes No Total Female 114 (71.7) 45 (28.3) 321 (100) 101 (77.1) 30 (22.9) 131 (100)

Male 240 (74.8) 81 (25.2) 159 (100) 314 (90.0) 35 (10.0) 349 (100)

Total 354 (73.8) 126 (26.3) 480 (100) 444 (92.5) 36 (7.5) 480 (100)

χ2 = 0.52 p-value=0.472 χ2 = 13.40 , p-value=<0.001*

*p<0.05 = statistically significant

The table above shows the association between gender and willingness to pay for health

insurance among the respondents. A high proportion of male respondents in the urban

(74.8%) and in the rural (90.0%) were willing to pay for the proposed CBHI scheme. The

association between the household head’s gender and willingness to pay for CBHI was

statistically significant in the rural group with the male respondents more willing to pay for

the CBHI scheme than their female counterparts (p<0.001); however the association was not

statistically significant in the urban LGA (p=0.472).

Table 21: Association between the age groups of respondents and their willingness to pay for CBHI

Variable Urban n=480 Rural n=480

Willingness to pay Willingness to pay

Age Yes No Total Yes No Total

(years)

18 – 29 44 (67.7) 21 (32.3) 6 5(100) 58 (92.1) 5 (7.9) 63 (100)

30 – 39 95 (74.4) 31 (25.6) 126 (100) 120 (88.9) 15(11.1) 135 (100)

40-49 122 (75.4) 42 (24.6) 164 (100) 118 (86.0) 24(13.5) 142 (100)

50-59 65 (81.3) 15 (18.8) 80 (100) 76 (84.4) 14(15.6) 90 (100)

> 60 28 (62.2) 15 (33.3) 45 (100) 43 (83.1) 7(16.9) 50 (100)

Total 354(73.8) 126(26.3) 480 (100) 415 (86.5) 65(13.5) 480 (100)

χ2= 6.86 , p-value=0.044* χ2 = 18.55** , p-value= 0.039*

*p<0.05 = statistically significant **Fisher’s exact test

The table above shows the association between age group and willingness to pay for health insurance among the respondents. Willingness to pay for the scheme increased with age until the age group of 60 and above in the urban group while in the rural group; willingness to pay decreased with increasing age group until the age group 60 and above. A high proportion of the respondents who were in the age group 50-59 in the urban (81.3%) were willing to pay for the proposed CBHI scheme. Similar observation was seen in the rural (92.1%) amongst those aged 18-29 years who were willing to pay for the proposed CBHI scheme. There was statistically significant association between respondent’s age group and their willingness to pay for CBHI in the two groups.

Table 22: Association between marital status of respondents and their willingness to pay for CBHI

Variable Urban n=480 Rural n=480

Willingness to pay Willingness to pay

Marital Yes No Total Yes No Total Status

Single 22 (73.3) 68 (26.7) 30(100) 21 (95.5) 1 (4.5) 22 (100)

Married 295(75.1) 98 (24.9) 393 (100) 335 (85.7) 56 (14.3) 391 (100)

Separated 14 (60.9) 9 (39.1) 23 (100) 17 (85.0) 3 (15.0) 20 (100)

Divorced 4 (66.7) 2 (33.3) 6 (100) 0 (0) 0 (0) 0 (0)

Widowed 19 (67.9) 9 (32.1) 28 (100) 42 (89.4) 5 (10.6) 47 (100)

Total 354(80.2) 126(19.8) 480 (100) 444 (92.5) 36 (7.5) 480 (100)

χ2= 3.40** , p-value=0.482 χ2 ** = 1.76 , p-value= 0.615

*p<0.05 = statistically significant **Fisher’s exact test

Table 22 shows the association between respondent’s marital status and willingness to pay for health insurance. A high proportion of the respondents who were married in the urban

(75.1%) were willing to pay for the proposed CBHI scheme. Contrary to this finding was seen in the rural (95.5%) amongst the single were willing to pay for the proposed CBHI scheme. Marital status was not associated with respondent’s willingness to pay for CBHI in both the urban and rural areas.

Table 23: Association between the level of education of the respondents and their

willingness to pay for CBHI

Variable Urban n=480 Rural n=480

Willingness to pay Willingness to pay

Level of Yes No Total Yes No Total education

No formal 16 (72.7) 6 (27.3) 22 (100) 26 (71.4) 10 (28.6) 36 (100) schooling

Primary 60 (77.5) 11 (25.5) 71 (100) 127 (84.2) 17 (15.8) 144 (100)

Junior secondary 35 (77.8) 10 (22.2) 45 (100) 62 (86.1) 10 (13.9) 72 (100)

Senior secondary 162(82.8) 65 (17.2) 227 (100) 142 (87.2) 19 (12.8) 161 (100)

Tertiary 81 (87.7) 34 (12.3) 115 (100) 58 (88.6) 9 (12.4) 67 (100)

Total 354(73.8) 126 (26.3) 480 (100) 415 (86.5) 65 (13.5) 480 (100)

χ2 = 5.95, p-value=0.033* χ2 = 7.21 , p-value= 0.017*

*p<0.05 = statistically significant

The table above shows the association between level of education and willingness to pay for

the CBHI scheme among the respondents. Respondents who had tertiary level of education in

the urban (87.7%) and in the rural (88.6%) were more willing to pay for the proposed CBHI

scheme. Respondents highest completed educational level was significantly associated with

the respondent’s willingness to pay for the scheme as those with higher level of education

were more willing to pay for the scheme in the rural and urban areas.

Table 24: Association between the respondents occupation and willingness to pay for

CBHI

Variable Urban n=480 Rural n=480

Willingness to pay Willingness to pay

Occupation Yes No Total Yes No Total

Professional 11 (100.0) 0 (0) 11 (100) 4 (30.8) 9 (69.2) 13 (100)

Skilled 122 (88.4) 16 (11.6) 138 (100) 127 (71.3) 51 (28.7) 178 (100)

Semi skilled 53 (89.8) 6 (10.2) 59 (100) 84 (77.1) 25 (22.9) 109 (100)

Skilled 223 (84.5) 41 (15.5) 264 (100) 126 (78.3) 35 (21.7) 161 (100)

Others 6 (75.0) 2 (25.0) 8 (100) 13 (68.4) 6 (31.6) 19 (100)

Total 415 (86.5) 65 (13.5) 480 (100) 354 (73.8) 126 (26.3) 480 (100)

χ2 = 4.13**, p-value=0.345 χ2 = 13.78** , p-value= 0.007*

*p<0.05 = statistically significant ** Fischer’s exact test

The table above shows the association between respondent’s occupation and willingness to

pay for the CBHI scheme among the respondents. Respondents who were professionals in the

urban (100%) and those with skilled labour in the rural (78.3%) were more willing to pay for

the proposed CBHI scheme. Household source of income was significantly associated with

the respondent’s willingness to pay for the scheme in the rural area (p=0.007).

Table 25: Association between respondent’s household size and their willingness to pay

for community based health insurance

Variable Urban n=480 Rural n=480 Willingness to pay Willingness to pay Household Yes No Total Yes No Total size 1-3 90 (73.2) 33 (26.8) 123 (100) 98 (89.1) 12 (10.9) 110 (100)

4-6 199 (78.0) 56 (22.9) 255 (100) 244 (86.9) 25 (13.1) 286 (100)

>7 65 (63.7) 37 (36.3) 102 (100) 73 (85.3) 11 (14.7) 84 (100)

Total 354 (73.8) 126 (26.3) 480 (100) 415 (86.5) 65 (13.5) 480 (100)

χ2 = 7.74 , p-value = 0.021* χ2 = 0.985 p-value = 0.611

*p<0.05 = statistically significant

The table above shows the association between the respondent’s household size and their

willingness to pay for community based health insurance. A higher proportion of the

respondents in the urban (78.0%) who had a household size of 4-6 people were willing to pay

for the proposed CBHI scheme while respondents in the rural (95.5%) who had a household

size of 1–3 people (95.5%) were willing to pay for the proposed CBHI scheme. Household

size was associated with respondent’s willingness to pay for CBHI in the urban group

(p=0.021) as those with average household sizes were more willing to pay for the scheme;

however this was not so in the rural

Table 26: Association between respondent’s income and their WTP for CBHI

Variable Urban n=480 Rural n=480

Willingness to pay Willingness to pay

Income Yes No Total Yes No Total

(Naira)

<5000 20 (87.0) 3 (13.0) 23 (100) 271 (80.0) 37 (20.0) 308 (100)

5000-10000 59 (64.8) 32 (35.2) 91 (100) 73 (83.9) 14 (16.1) 87 (100)

10001-20000 79 (68.1) 37 (31.9) 116 (100) 52 (88.0) 13 (12.0) 65 (100)

20001-30000 59 (71.1) 24 (28.9) 83 (100) 14 (93.3) 1 (6.7) 15 (100)

>30000 137 (82.0) 30 (18.0) 167 (100) 5 (100.0) 0 (0) 5 (100)

Total 354 (73.8) 126 (26.3) 480 (100) 415 (86.5) 65 (13.5) 480 (100)

χ2 = 13.95 , p-value = 0.007* χ2 = 10.95**, p-value = 0.037*

*p<0.05 = statistically significant **Fisher’s exact test

The table above shows the association between the respondent’s income and willingness to

pay for community based health insurance. A high proportion of the respondents in the urban

(82.0%) and the rural (100%) who earned greater than 30000 naira monthly were willing to

pay for the proposed CBHI scheme. The respondent’s income was significantly associated

with their willingness to pay for community based health insurance as respondents with

higher income levels being more willing to pay for the scheme in the 2 groups (p=0.007,

0.037).

Table 27: Association between distance respondents’ travelled to health facility and their willingness to pay for CBHI

Variable Urban n=199 Rural n=211 Willingness to pay Willingness to pay for self Distance travelled Yes No Total Yes No Total to health facility < 15 mins 79 (73.1) 29 (26.9) 108 (100) 119 (85.6) 20 (14.4) 139 (100)

15 – 30 mins 43 (84.3) 8 (15.7) 51 (100) 47 (85.9) 9 (14.1) 56 (100)

30 – 60 mins 42 (71.2) 17 (28.8) 59 (100) 31 (91.2) 3 (8.8) 34 (100)

>1 hour 1 (33.3) 2 (66.7) 3 (100) 9 (100) 0 (0) 9 (100)

Total 165 (74.7) 56 (25.3) 221 (100) 206 (86.6) 32 (13.4) 238 (100)

χ2 = 5.68** , p-value=0.610 χ2 = 1.75**, p-value= 0.045 *

*p<0.05 = statistically significant **Fisher’s exact test The table above shows the association between distance travelled by the respondents to the health facility and willingness to pay for CBHI. A high proportion of the urban respondents who travelled 15-30 minutes to the health facility (84.3%) were willing to pay for the proposed CBHI scheme. Among the rural respondents, a high proportion of those that travelled for over an hour to the health facility (100%) were willing to pay for the proposed

CBHI scheme. Distance travelled to the health facility was associated with respondents’ willingness to pay for CBHI in the rural (p=0.045) as with increasing distance that the respondent travelled to access healthcare increased their willingness to pay for the scheme; however the association was not statistically significant in the urban area.

Table 28: Association between respondent’s membership of a financial

association/group and their willingness to pay for CBHI

Variable Urban n=480 Rural n=480 Willingness to pay Willingness to pay Membership Yes No Total Yes No Total of a financial group/ association Yes 66 (70.2) 28 (29.8) 94 (100) 156 (86.7) 24 (13.3) 180 (100)

No 288 (74.6) 98 (25.4) 386 (100) 259 (86.3) 23 (13.7) 300 (100)

Total 354 (73.8) 126(26.3) 480 (100) 415 (86.5) 65 (13.5) 480 (100)

χ2= 0.76, p-value=0.003* χ2 = 0.01 p-value=0.918

*p<0.05 = statistically significant

The table above shows the association between respondents membership status of a financial

association were money is being managed for them and their willingness to pay for

community based health insurance. Respondents who were members of an association/group

in the rural (86.7%) were willing to pay for the proposed scheme. In comparison, a higher

proportion of their urban counterparts who did not belong to an association (74.6%) were

willing to pay for the CBHI scheme. Non members of these financial groups were more

willing to pay for the CBHI scheme among the urban group. This association was statistically

significant (p=0.003); however this was not so among the rural group.

Table 29: Association between respondent’s previous use of health insurance and their

willingness to pay for CBHI

Variable Urban n=480 Rural n=480 Willingness to pay Willingness to pay Previous use Yes No Total Yes No Total of health insurance Yes 12 (63.2) 7 (36.8) 19 (100) 6 (100) 0 (0) 6 (100)

No 342 (84.7) 119 (15.3) 461(100) 309 (71.5) 65 (28.5) 374 (100)

Total 354 (73.8) 126 (26.3) 480(100) 415 (86.5) 65 (13.5) 480 (100)

χ2 = 1.89**, p-value = 0.789 χ2 = 1.83**, p-value = 0.397

*p<0.05 = statistically significant **Fisher’s exact test

The table above shows the association between the respondent’s use of health insurance and

willingness to pay for community based health insurance. A high proportion of the

respondents in the urban (84.7%) who had not used health insurance previously for paying

for their health care were willing to pay for the proposed CBHI scheme while those who had

used health insurance for paying for their health in the rural (100%) were willing to pay for

the proposed CBHI scheme. However respondents use of health insurance was not associated

with their willing to pay for CBHI in both the urban and rural areas.

Table 30: Association between respondent’s level of confidence if funds are pooled and

managed by community and their willingness to pay for CBHI

Variable Urban n=480 Rural n=480 Willingness to pay Willingness to pay Respondents Yes No Total Yes No Total level of confidence Low 107 (75.4) 35 (24.6) 142 (100) 115 (78.8) 31 (21.2) 146 (100)

Medium 154 (80.2) 38 (19.8) 192 (100) 252 (89.0) 31 (11.0) 283 (100)

High 93 (63.7) 53 (36.3) 146 (100) 48 (94.1) 3 (5.9) 51 (100)

Total 354 (73.8) 126 (26.3) 480 (100) 415 (86.5) 65 (13.5) 480 (100)

χ2= 11.94, p-value = 0.003* χ2 = 11.55 p-value = 0.003*

*p<0.05 = statistically significant **Fishers exact test

The table above shows the association between respondent’s level of confidence if funds are

pooled and managed by community and their willingness to pay for health insurance.

Respondent’s confidence in pooled funds being managed by community was significantly

associated with their willingness to pay for community based health insurance as those with

confidence were more willing to pay in the urban and rural areas (p=0.003, 0.003) with WTP

increasing with increasing confidence in the managers of the scheme.

Table 31: Relationship between sociodemographic characteristics and mean amounts respondents are willing to pay as premium for single enrolees

Variables Mean amounts willing to pay Urban n=385 Rural n=415

Age group (years)

18 - 29 585.12+204.49 381.11+286.69

30 – 39 575.65+175.76 438.46+309.71

40 – 49 550.47+250.42 497.16+307.57

50 – 59 497.73+191.16 503.75+312.38

> 60 466.67+291.64 508.72+337.66

f=3.79 p=0.005* f=1.89 p=0.007*

Sex

Male 557.88+208.28 486.76+328.21

Female 548.17+252.56 475.47+296.12

f=0.18 p=0.665 f=0.13 p=0.710 Highest level of education No formal schooling 472.86+285.78 415.91+284.26 Primary 520.18+231.45 466.33+288.59 Junior secondary 564.00+268.69 485.11+382.61 Senior secondary 565.88+189.40 514.08+245.70 Tertiary 611.19+239.60 515.89+326.46 F=2.63 p=0.024* F=1.53 p=<0.001* Household size 1-3 542.29+227.73 532.08+317.79 4-6 577.78+216.94 460.98+293.89 >7 572.73+204.83 371.20+319.53

F=1.27 p=0.277 F=9.48 p=<0.001* *p<0.05 = statistically significant

Table 31 shows the relationship between the respondents’ sociodemographic characteristics and the mean amounts that they were willing to pay for single enrolees per month for the scheme. The respondents within the age group 18-29 years in the urban LGA were willing to pay the highest premium for self per month of 508.72+337.66 while in the rural; the respondents aged > 60 years were willing to pay the highest premium of 585.12+204.49. The age group of respondents was significantly associated with the mean amounts they were willing to pay for single enrolees in the 2 groups.

The respondents’ gender was not associated with the mean amounts they were willing to pay for single enrolees in the urban and rural areas.

The educational level of respondents was significantly associated with the mean amounts they were willing to pay for single enrolees in the 2 groups with those with higher educational levels paying a higher mean amount as premium for single enrolees in the 2 groups (p=0.024, <0.001)

The household size of respondents was significantly associated with the mean amounts they were willing to pay for single enrolees in the rural group with those with smaller household sizes paying a higher mean amount as premium for single enrolees (p= <0.001); however this was not so among the urban respondents .

Table 32: Relationship between sociodemographic characteristics and the mean amounts respondents are willing to pay as premium for household per month

Mean amounts willing to pay Variables Urban n=378 Rural n=428

Age (years) 18 - 29 1040.91+408.04 686.67+640.17

30 – 39 1008.53+307.41 692.92+500.71

40 – 49 943.64+461.63 760.50+421.76

50 – 59 904.55+414.30 777.18+505.56

> 60 833.33+514.75 791.88+525.46

F=3.26 , p=0.005* F=0.65, p=0.027* Sex

Male 978.42+368.63 760.22+509.44

Female 967.94+500.85 743.96+478.58 F=0.06, p=0.795 F=0.11, p=0.741 Highest Level of education

No formal schooling 800.00+467.97 699.96+508.52 Primary 932.99+393.75 703.48+494.43 Junior secondary 971.81+367.07 722.73+576.47 Senior secondary 1024.53+373.00 727.56+511.34 Tertiary 1047.76+494.55 846.34+513.12

F=3.22 p=0.013* F=1.20 p=0.048* No of people in household 1-3 1031.36+426.58 857.85+487.02 4-6 943.67+403.78 699.11+457.51

>7 1013.58+393.23 532.61+506.85 F=2.27, p=0.989 F=16.55, p=<0.001*

*p<0.05 = statistically significant

Table 32 shows the relationship between the respondents’ socio-demographics and mean amounts that they were willing to pay for their household per month for the scheme. The age group of the respondents was significantly associated with the mean amounts they were willing to pay for single enrolees in the 2 groups as the amounts increased with increasing age in the 2 groups (p=0.005; 0.027)

The respondents’ gender was not associated with the mean amounts they were willing to pay for their households in the urban and rural areas.

The educational level of respondents was significantly associated with the mean amounts they were willing to pay in the 2 groups with those with higher educational levels paying a higher mean amount as premium in the urban and rural areas (p=0.013; 0.048 )

The household size of respondents was significantly associated with the mean amounts they were willing to pay for single enrolees in the rural group with those with larger household sizes paying a higher mean amount as premium (p= <0.001); however this was not so among the urban respondents .

Table 33: Logistic Regression of factors influencing willingness to pay for community based health insurance among the respondents.

Variables B OR 95% C.I P value

Lower Upper Sex Male 1.00 Female 0.825 2.28 1.33 3.92 0.003* Age group 18-29 1.00 30-39 -4.16 0.660 0.277 1.574 0.348 40-49 0.221 2.024 0.454 3.313 0.054* 50-59 0.861 1.424 1.152 1.181 0.001* >60 -0.22 3.360 1.458 6.412 0.003* Household size 1-3 1.00 4-6 0.326 1.386 0.705 2.723 0.344 >7 1.132 3.103 1.432 6.727 0.004* Income <5000 1.00 5001-10000 0.880 2.410 1.144 5.075 0.021* 10001-20000 0.637 3.891 2.35 8.55 0.012* 20001-30000 0.935 2.547 1.084 5.988 0.032* >30000 0.428 1.535 1.90 3.83 0.025* Occupation Professional 1.00 Skilled 0.644 0.525 0.217 1.269 0.153 Semi-skilled 0.824 0.439 0.174 1.107 0.081 Unskilled 0.816 0.442 0.184 1.062 0.068 Others 0.205 0.815 0.249 2.663 0.735

*p<0.05 = statistically significant OR = odd ratio CI = confidence interval

Table 33 shows the regression of the factors of willingness to pay for community based health insurance.

Respondents who were female were 2.28 times more willing to pay for CBHI than the males

(C.I 1.33 -3.92).

Respondents aged 30-39 were 0.66 times less willing to pay for CBHI (CI 0.28-1.57) compared to age group 18-29 years. The p value for this odd ratio was not statistically significant. Respondents in the age group 40-49 were 2.02 times as likely to be willing to pay for CBHI as age group 18-29 years (0.45-3.31). The p value for this odds ratio was statistically significant. Respondents in the age group 50-59 were 1.42 more willing to pay for CBHI compared to age group 18-29 years (1.15-1.18). The p value for this odds ratio was statistically significant. The respondents aged over 60 years were 3.36 times less willing to pay for CBHI than respondents in age group 18-29 years (1.46-6.62). The p value for this odds ratio was statistically significant.

Respondents with household size of 4-7 were 1.39 times more willing to pay for CBHI than respondents with household size 1-3 (CI 0.705-2.723). The p value for this odd was not statistically significant. Respondents with greater than 7 household members were 3.1 times as willing to pay for CBHI compared to respondents with household size 1-3 (CI 1.432-

6.727). The p value for this odds ratio is statistically significant (p=0.004).

Respondents that earned 5001-10000 naira monthly were 2.41 times as likely to be willing to pay for CBHI as those that earned < 5000 naira. (1.144-5.075). The p value for this odd ratio was statistically significant (p=0.021). The odds of willingness to pay among respondents that earned 10001-20000 naira monthly was 3.89 times as likely to be willing to pay for

CBHI as those that earned < 5000 naira. (2.35-8.55). The p value for this odd ratio was statistically significant. The odds of willingness to pay among respondents that earned

20001-30000 naira monthly was 2.55 times as likely to be willing to pay for CBHI as those

89 than earned < 5000 naira. (1.084-5.988). The p value for this odd ratio was statistically significant (p=0.032). Respondents who earned > 30000 naira were 1.54 times as likely to be willing to pay for CBHI as those that earned < 5000 naira. (CI 1.90-3.83). This association was statistically significant.

Table 34: Logistic Regression of factors influencing willingness to pay for community based health insurance among the respondents.

Variables B OR 95% C.I P value Lower Upper

Highest level of education No formal education 1.00 Primary 1.122 0.326 0.112 0.945 0.039* Junior secondary 1.070 1.343 1.096 2.221 0.045* Senior secondary 0.838 1.433 1.161 2.159 0.026* Tertiary 0.510 1.601 1.207 2.742 0.048* Distance travelled to health facility < 15 minutes 1.00 15 – 30 minutes 0.315 0.730 0.379 1.406 0.346 30– 60 minutes -0.163 0.850 0.438 1.647 0.630 >1 hour -0.132 1.141 0.221 5.882 0.875 Trust in managers of CBHI scheme Low 1.00 Medium -0.533 0.587 0.403 0.856 0.006* High 0.280 1.323 0.871 2.010 0.189 Membership of a financial group/Association Yes 1.00

No -0.255 0.775 0.443 1.356 0.372

*p<0.05 = statistically significant OR = odd ratio CI = confidence interval

Table 34 shows the logistic regression of the factors of willingness to pay for community based health insurance.

The respondents with primary school education were 0.33 times as willing to pay for CBHI as respondents with no formal education. The p value for this odd was statistically significant

(p=0.039). The respondents with secondary school education were 1.43 times as willing to pay for CBHI as respondents with no formal education (CI 1.16-2.16, p= 0.026). The respondents with tertiary educational level were 1.6 times as willing to pay for CBHI as respondents with no formal education (CI 1.20-2.74, p=0.048).

Respondents that travelled for 15-30 minutes to the health facility were 0.73 times as likely to be willing to pay for CBHI than those that travelled for less than 15 minutes (0.38-1.41).The p value for this odds ratio was not statistically significant. The respondents that travelled for

30-60 minutes to the health facility were 0.85% times less likely to be willing to pay for

CBHI than those that travelled for less than 15 minutes (0.45-1.65).The p value for this odds ratio was statistically significant. The respondents that travelled for over an hour to the health facility were 1.14 less likely to be willing to pay for CBHI than those that travelled for less than 15 minutes (0.22-5.88). The p value for this odds ratio was not statistically significant.

The respondents that had a medium level of confidence in those managing their pooled funds were 0.59 times less likely to be willing to pay for CBHI as respondents with low confidence

(CI 0.40-0.86). The p value for this odd ratio was statistically significant (p=0.006). The odd of willingness to pay among respondents with high confidence levels was 0.28 times as likely to be willing to pay for CBHI as respondents with low confidence (CI 0.87- 2.01). The p value for this odds ratio was not statistically significant.

Respondents who did not belong to any financial group or association were 0.78 times less likely to be willing to pay for CBHI than those that were members of financial associations

(CI 0.44-1.36). The p value for this odds ratio was not statistically significant.

CHAPTER FIVE

DISCUSSION

This study was carried out to compare the household willingness to pay for community based health insurance in Urban and Rural LGAs in Lagos state. The socio-demographic characteristics of respondents show that the mean age of heads of household for urban and rural households was 43.17 ± 13.4 years and 47.22 ± 11.3 years respectively. The difference in the mean age of the two groups was not statistically significant.

Majority of head of households in both settings were male which is common in most African household settings as the decision making in most settings is done by the men38 and is in line with the Nigeria demographic and health survey 2013.118

The average household size was 4.8 ±2.1 for urban households and 5.0 ± 2.0 for rural households which are similar to results from a study in Ilorin, Nigeria where the average household size was 5.0 ± 2.796 and also according to the Nigeria demographic and health survey 2013 where the mean household size for urban households is 4.2 and 4.9 for rural households.118 A higher proportion of the urban heads of households (31.5%) had tertiary education as compared to the rural households (14%) and this shows there is higher level of literacy in the urban communities which is characteristic of many urban communities in

Nigeria.

Majority of the household heads were earning income monthly with mean income for urban heads to be 32,000 ± 4,000 naira and 10,100 ± 3,300 naira for rural heads. This is similar to a study in Osun, Nigeria where majority of household heads average household income was

42,000 ± 5,200 naira and 10,000 ± 5,300 naira per month for urban and rural households102 as well as Enugu were the average household income was 10,141.20k naira per month for rural households.38

The factors determining the health behaviours may be seen in various contexts: physical, socio-economic, cultural and political.49 Therefore, the utilization of a health care system, public or private, formal or non-formal, may depend on socio-demographic factors, social structures, level of education, cultural beliefs and practices, economic conditions, and the disease pattern and health care system itself.49

A study in a rural facility in Ekiti, Nigeria looked into the reasons for the seeking of medical services among the rural dwellers in the state. The major reasons identified were: the type of ailment suffered by the patients, availability of money at the time of sickness, age of the patients, religious beliefs, educational background, severity of sickness and position in the household.50 Nearly half of the household heads in the two groups had an illness in the last three months of which malaria was the most reported illness with a prevalence of 60.6% in the urban and 61.7% in the rural. The first place that the urban household sought for treatment was the primary health centre (29.4%) which is the first point of contact that members of a community have with the health system. This may be based on the fairly reasonable availability of PHC facilities in the study area. This is in contrast to that in the rural where 37% of the respondents patronised the chemists/patent medicine vendors. A higher proportion of the rural (18.8%) compared to the urban (14%) patronised the traditional healers. This is similar to what is seen in other studies where higher proportions of the members of rural communities used traditional medicine. The major factors that cause non- attendance of services in a modern health facility in an LGA in Ekiti included easy access to traditional healers and patent medicine vendors among 39% of their respondents.

About half of the urban household heads used bus as the form of transport to the health facility in comparison to half of the rural household heads that walked to the health facility.

This is consistent with the findings in the rural area where a higher proportion of their

94 household heads patronised the chemists and traditional healers whose place of operation is within proximity in the community causing most respondents to walk to them.

The mean households’ expenditure on their health in the last quarter was 4832.35+1615.69 and 4234.17+1565.65 naira in the urban and rural areas which is 17.9% and 43.3% of their household income respectively. This signifies catastrophic spending among the rural households as a major proportion of the income (exceeding 40%) was spent on their health.

Majority of the households in the urban (81.9%) and rural (91.3%) areas used cash as form of payment for their health care and they likewise coped with this payment from out of pocket payments. This is consistent with literature where about 90% of health expenditure is from out of pocket payments20. Among the respondents, there was a low level of utilization of health insurance with only 4% in the urban and 2.5% in the rural utilising health insurance.

This value is similar to the health insurance coverage in Niger which is less than 5%.85 This further affirms the paucity of the health insurance mechanism and high level of out of pocket spending in Nigeria.

A higher proportion of the urban respondents (41%) compared to the rural (14.4%) had ever heard of health insurance. This could be due to their generally lower level of education in the rural areas. Similar low rates were recorded in Cameroon among rural dwellers, where 25% were aware of health insurance.1 Amongst those aware of health insurance; 79.2% of the urban respondents knew that health insurance was prepayment for health respondents.

However a higher proportion of rural respondents (58.0%) defined health insurance as paying tax to the government. This is in line with the high level of non-awareness of health insurance

(86%) seen among the respondents. Half of the respondents in the urban and rural knew private health insurance as a type of health insurance. This is possibly due to the sustained

95 marketing by the health management organisations on private health insurance in the study area.

Perceptions’ relating to insurance schemes, the scheme providers and the community attributes plays a major role in household decisions to voluntarily enrol and remain in the scheme.91 In this study the respondents in the two areas had a good perception of the CBHI scheme. Price including premium and registration fees and the benefits of the scheme are factors that are significantly associated with enrolment and retention in the scheme. In this study, 43.8% of the respondents in the urban and 48.8% in the rural perceived that CBHI had the ability to make healthcare more affordable for them. Studies have shown that enrolment decreases if the price of the premiums is perceived to be high.91 In addition, in Burkina Faso, found institutional rigidities of payment modalities to be a greater barrier than the premium per se and this indicates the need for policy makers to consider what methods of payment best suit households.108

Majority of the urban and rural households perceived that CBHI has the potential to improve the quality of services provided by healthcare givers but this was not statistically significant.

This finding is in keeping with a study in Oyo where a 60% of the households believed that

CBHI had the potential to improve quality of health services received in their establishments92. The credibility of the health care system in relation to quality of care is a crucial factor in the way people perceive health insurance. Insured clients who perceive they are given poor quality of care or have longer waiting times in comparison to fee paying clients leads to a growing dissatisfaction with the scheme. A study looked at provider attitudes as well as service delivery and quality of care pre and post scheme implementation.

Provider attitudes were perceived negatively but the quality of care and service delivery in the study was perceived positively91. Another study did not believe there was a difference in health service delivery before and after implementation of the scheme92.

Clients perceive that a constant supply of essential drugs was a prerequisite for the credibility of the scheme and for the quality of healthcare provided. In this study, the rural respondents

(51%) and their urban counterparts (43.5%) were indifferent about the potential of CBHI to ensure constant availability of drugs at health facilities in their community. This is similar to another study were about 52% of their respondents believed that sufficient drugs were available under the scheme92.

A high level of trust among CBHI scheme members is important for the functioning of effective risk pooling within a community. The urban households and rural respondents were asked regarding their perception about their level of confidence in the community managing their pooled funds. The urban respondents had a low perception (47.5%) of the level of confidence they had in the community managing their pooled funds compared to the rural

(59.0%) that were indifferent to it. This may be as a result of cultural difference in the two regions, whereby the rural people are used to groups were money is being managed for them popularly known as “Ajo” and “cooperative” e.t.c. Similar poor perception was seen in a study in Oyo state were majority of their respondents wanted the scheme discontinued due to lack of confidence in the programme92. CBHI schemes that are created in communities that have other risk sharing community groups are able to leverage the existing trust relationships to assist in the implementation and enrolment into the new CBHI scheme.108 Individuals living in areas where there are general practices of resource sharing in the community have a higher tendency of enrolling in CBHI schemes.108

Another study argues that the greater level of trust and solidarity within a community, the less likely it is that individuals will choose to pursue egoistic behaviours that may compromise the functioning of the CBHI scheme.1 Communities with higher degrees of trust are better able to handle common issues faced by insurance schemes, mainly moral hazard, adverse selection, and fraud.1

Acceptance of CBHI as a strategy for paying for their health care was scored among the urban and rural household heads. A higher proportion of the rural respondents (92.5%) accepted CBHI compared to the urban (80.2%). This difference was found to be statistically significant. The commonest reason given by the household heads who did not accept CBHI in the urban (26.3%) was lack of trust in the managers of the scheme while the reason given by the rural was they were always in good health (38.9%). The difference in their reasons for was not statistically significant. Despite the high proportions of households accepting CBHI as a healthcare payment option, there were disparities in the proportion that were willing to pay for the scheme.

About 73.8% of the urban households were willing to pay for community based health insurance scheme; while 86.5% of rural households were willing to pay for community based health insurance scheme. This disparity in acceptability and WTP for the scheme is probably due to the inability to pay for the scheme despite the strong desire to participate among the respondents. The high WTP rates is similar to what was found in north central Nigeria18 were the willingness to pay in a rural community was 93.6%. Much lower willingness to pay was reported in an urban community in south west Nigeria were the willingness to pay was

51.6%.102 This is contrary to the findings in Eastern part of Nigeria where < 7% of rural households was WTP and more urban households were willing to pay46. This may be as a result of difference in the geo-political zones and also cultural difference in the two regions, whereby the rural people in South Western part are used to having their money managed by financial groups and associations.

One would expect higher willingness to pay from urban households as evidenced by level of income, literacy level, e.t.c but reverse was the case and this may be as a result of lack of access to quality health care in the rural communities as compared to the urban centres thereby raising their interest in a programme that will improve their access to good health

98 care. Also because of strong earning power in the urban setting they may think they can afford the user fee anytime the need arises.

Reasons given by majority of the households in this study unwilling to pay for CBHI was that they had no confidence/trust in the scheme by 41.3% of the urban households and the main reason given by 33.8% of the urban households was that they felt they could not afford the premium. In another study, financial barrier was cited as one of the main reasons why households did not renew their membership.108 It was highlighted that the poor households found it difficult to pay for the scheme108, 109 while in urban areas, the richer were less willing to pay showing that the rural areas are really in need of accessible and affordable health care as was seen in this study. This differs from reasons in another study were the rural household were unwilling to pay on account of distance to the health facilities.18

In this study considering the premiums that households were willing to pay monthly, the mean WTP among the rural household heads was found out to be 542.19 + 317.67 naira (3.4

+ 1.98 USD) for individual enrolees per month and 754.83 + 498.99 naira (4.72 + 3.12 USD) per household per month while in the urban households it was found to be 555.23 + 221.01 naira (3.5 + 1.38 USD) per person per month and 957.56 + 408.45 (5.98 + 2.55 USD) naira per household per month. Similar WTP estimates were seen in Ilorin, Kwara were the researchers reported a mean WTP of 591.6 + 302.6 (3.48 + 1.78 USD) naira per person per annum for CBHI in a community with an average household size103 and in rural Iran, with a premium of 510 naira (3 USD) per household per month on the average.95

Lower values were seen in a study in Osun state where majority of the households (74.5%) were willing to pay less than 200 naira per month. Only 2.4% were willing to pay the highest presented premium of 300 naira98. In Eastern Nigeria, lower values were also found with a

WTP of 289 naira (1.7 USD) per person per month in a rural community while the WTP in

99 the urban was 493 naira (2.9 USD) per person per month in the urban area. Similar results were seen in another rural community were a mean WTP of 255 naira (1.5 USD) per household per month was observed.97 This disparity may not be because of the lack of desire to pay more but more because of the inability to pay due to their lower socioeconomic statuses.

A much higher mean WTP per person per month was gotten in a study in Namibia which revealed fairly higher WTP of 1190 naira (7 USD) per person per month.42 Similar high findings were found in Oshogbo to be 1,798.9 ± 134.7 naira (11.24 ± 0.84 USD) for urban households and 761 ± 255.5 naira (4.51± 1.57USD) in the rural households.102 A study carried out in Wuhan, China examined WTP among informal sector workers and found that these workers are willing to pay the equivalent of 680 naira (4 USD) per person per month.95,

96 This difference compared to our findings might be because of the differences in prevailing socio-economic situations and level of industrialization which obviously will affect the earning power.

Willingness to pay among respondents is affected by a dynamic interplay between individual, social, demographic, environment and economic factors. In this study, individual factors like age was significantly associated with WTP in both the urban and rural areas but this was not so for their gender.

In this study, willingness to pay declined with increasing age in the rural area which is in line with a study in Burkina Faso that found the elderly less willing to pay than the younger respondents with a confirmed decrease in WTP of 76% among the elderly.123 Willingness to pay increased with increasing age up to those aged 59 years and declined in the older age group of over 60 years of age in the urban area. The finding from the urban area in this study is in line with a study from Nsukka.7 Similar findings were seen in study in North central

Nigeria found an increase in willingness to pay with the age group 40-49 years and a decline

100 up to those who were over 60 years of age. Similar results were seen in Osogbo102 were a unit increase in age increased willingness to pay by 17% i.e. the elderly were more WTP.

Reasons for this might be that those in the older age group find it harder to accept new schemes or programmes easily. This can also be explained that the pay-off period for any human capital investment in health at older ages yield less and lesser compared to the younger respondents.123

The premium that individuals were willing to pay was also related to age of the respondents.

The premium that respondents in the urban area were willing to pay declined with increasing age while in the rural, it increased with increasing age. The finding among the urban respondents in this study was also observed in Namibia42 which showed that the young respondents were willing to pay more101 however a study in Kwara got contrary results.103

This study in North central Nigeria showed that the household heads that were 50 and above were willing to pay the least amount of premium while household heads that were between the age of 30 to 39 had the highest with a mean amount.103

Another important factor that affects WTP is gender42,46 as was observed in this study were a higher proportion of males in the urban (74.8%) and the rural (90%) were willing to pay in the proposed scheme. This was however significant in the rural area (p=<0.001). This is in line with findings in Oshogbo, for the rural households were gender was found to be significant with male gender increasing WTP by 12%102 and findings in Benin City where males show more willingness to participate in CBHI scheme than females.104 It has been reported that males were more willing to participate in rural health insurance scheme than the female headed households in rural areas of Nigeria.38 Closely related to the finding in the rural households is in Namibia where 31% of individuals who live in male-headed households are insured compared with 21% of individuals living in female-headed households.42 This may be as a result of men being the bread winners in most African

101 countries while most household headed by females are usually widows. This is however contrary from the finding in Tanzania40 where 78% of households who were not willing to pay anything for CBHI had male household heads and twenty percent had female household heads.

Marital status has been shown to have a role in respondents’ willingness to pay. Recent research has shown a strong association between marital status and WTP with an increased likelihood among the married respondents.106 In this study, majority of the respondents who were married in the urban were willing to pay into the scheme while majority of the single respondents in the rural area were willingness to pay. However no significant association between marital status and WTP as was seen in another study in India.107 A similar finding to the observation in the rural area was seen in a study in Jos, Nigeria where 93.7% of their single respondents were willing to pay into the scheme.18

Level of education was found to be another factor contributing to willingness to pay. In this study, the urban households (87.8%) and the rural (88.6%) with tertiary education were willing to pay for CBHI with similar findings seen in another study35,46 showing a strong correlation between education and WTP (p=0.033; 0.017). A study in Jos, found that the indication of their respondents WTP into a CBHI scheme was highest among those with tertiary education.18 Another study found that education plays a statistically significant role in determining the decision of respondents for WTP and went on to state that the marginal coefficient of education variable showed that a one grade increase in the highest grade completed will increase the probability of respondents to pay by 0.6%.42 Also the level of educational attainment of respondents had statistically significant influence on the premium that respondents were willing to pay.35,46 In this study the mean amount of money that respondents were willing to pay increased with level of education. The more educated the respondents were, the higher the amount respondents were willing to pay and this was

102 statistically significant.7 This is consistent with findings from previous studies where people with more education had a higher WTP.7, 35, 38, 46,103

Respondents’ occupation had an influence on their willingness to pay with a higher proportion of professionals in the urban area willing to pay into the scheme and a high proportion of the skilled in the rural area willing to pay into the scheme. However this was significant in the rural (p=0.007)

Income was found to be important factor in this study. The effect of income on their WTP was statistically significant as WTP increased with increasing income levels (p=0.007;

0.037). Similar findings were seen in a study among households in Osogbo, were a unit increase in income increased willingness to pay by 77%.102 Studies show that the higher the income, the higher the willingness to pay. In most studies, income had a positive effect on

WTP45, 103,108,109, however, there are some exceptions.104, 105 In another study carried out in

Oshogbo, income was also found to be significant with a unit increase in income decreasing

WTP by 53% i.e. the rich were less WTP.7 This finding with respect to income has been a source of debates about willingness to pay in giving value to health care as the amount households are willing to pay is an increasing function of their ability to pay.

The mean WTP showed that those with household size of 4 - 6 members in the urban (85.7%) and those with 1 – 3 members in the rural (95.5%) were willing to pay for the scheme

(p=0.021) and those with households greater than 6 members in number and similar results were seen with the mean amounts they were willing to pay. This is similar to another study where the mean WTP showed that those with household size of greater than 6 members were willing to pay the least amount of money35 ; however those that were willing to pay the highest amount of money were those with household size of 4 to 6 members.103 Contrary findings were seen in a Benin study104 where the respondents with larger families were found

103 to be are more willing to participate in community based health insurance scheme than respondents with smaller households. This was attributed to the high financial burden faced by large household when seeking health care services. This result agrees with the findings of other studies in rural areas where household size played a positive role in the willingness of people to participate in community based health insurance scheme.110.111

Distance to health centres was also found to be a contributing factor to WTP for CBHI in the rural area in this study. Willingness to pay increased with the farther the distance the respondent travelled to the health facility (p=0.045). Similar findings were seen in a

Ghanaian study were distance to the nearest health facilities had the hypothesized sign and significance indicating they could only value the payment if the accessibility is not within the reach.113 This is not in line with findings in another study were the willingness to pay decreased with a unit increase in distance to the health centre by 11% for their households124 which pointed out that people want to access health care close to them as much as possible.

This is also in line with principles of primary health care which says that the distance from home to a health centre should be ≤ 60km. Also in Oshogbo102 a unit increase in distance decreased WTP by 18% i.e. the farther the distance to the health centre the lesser the WTP because people wanted closer access to health care.

Membership of financial unions/associations popularly called “cooperative” and “Ajo” is a factor also associated with WTP for CBHI. In this study respondents who did not belong to a financial group/association in the urban area were willing to pay for the scheme (p=0.003).

Contrary to this finding was seen in the rural where 86.7% of those who belonged to these groups were willing to pay. In a study in Benin, Nigeria respondents who belonged to formal association were 3 times more likely to enrol in CBHIs than those who do not. A possible explanation for this is that those who are members of groups/associations are familiar with the concept of routine payments.104

104

Previously paying for health care using health insurance was positively related to willingness to pay46 but in this study there was no association between previous use of health insurance and WTP for the proposed CBHI scheme(p=0.789, 0.397).

Trust in the community managers of the scheme influences respondent’s willingness to pay.

In this study, there was association between respondent’s trust in community managing their pooled funds and their willingness to pay for health insurance (p= 0.003; 0.003). Other studies cited that clients’ level of trust in those organizing the scheme was positively related to willingness to pay. 44, 97, 104

Willingness to pay for community based health insurance was demonstrated to be high in both the urban and rural areas. However it was higher in the rural. The mean amount that the respondents were willing to pay as premium for individual and household enrolees was higher among the urban respondents.

The respondents’ age, household size, income, highest level of education and their level of trust in community managing the CBHI scheme were significantly associated with their willingness to pay for the CBHI scheme in this study.

105

CHAPTER SIX

CONCLUSION

The community-based health care insurance scheme is increasingly being recognized as a potentially powerful instrument for granting households access to health services in a more equitable way. Hence, the overall objective of this study was to assess the households’ perception and willingness-to-pay (WTP) and its determinants in urban and rural LGAs

Lagos for a proposed community based health insurance scheme.

The results of the study reveal that household heads in the urban (73.8%) and the rural

(86.5%) are willing to pay for the proposed scheme. This shows that they recognised the value of the scheme and its potential to increase access to good quality health services for their households, without having to pay at the point of service. Hence, this scheme could be well accepted in urban and rural areas of Lagos and has the potential to protect the households from any health risks. The urban households are willing to pay on average

₦540.73+235.50 for individual enrolees per month and ₦957.56+408.45 per household per month while the rural are willing to pay ₦479.38+319.91 for individual enrolees per month and ₦754.83+498.99 per household per month as premium for the proposed CBHI scheme.

Also respondents’ age, income, level of education, household size, and their level of confidence in the community managing the CBHI scheme are determinants of WTP.

This research showed that those with lower educational levels, lower incomes, smaller household sizes and the elderly were less willing to pay for the scheme than their counterparts.

106

RECOMMENDATIONS

Based on the findings in this study, the following recommendations are provided:

1. Based on the high willingness to participate in the scheme, the community-based

health care insurance scheme seems to be feasible in the study area. Lessons learnt

from the experience of the pilot schemes in place in the state can be applied to the

subject area.

2. Since there is limited data in setting premiums in community health care insurance

schemes, the current study can be used in setting premiums for the residents

concerned.

3. The premium to be charged for single enrollees and for households should be

anchored around reported premiums gotten from this study at ₦555.23 and ₦957.56

per month respectively.

4. Education of the population on the concept of health insurance and risk management

is required and this can be carried out via a social marketing strategy which would

involve mass media and public campaigns to inform and educate the populace on the

basic concept of community-based health insurance and its advantage. This may

encourage them to be involved in such scheme and hence improve their well-being.

5. Those with lower educational status, lower incomes and the elderly were less willing

to pay for this proposed scheme therefore in establishing community based health care

insurance scheme there should be selective targeting of these vulnerable groups

within the communities.

6. The Government can lend technical support to the communities managing and

running these schemes so as to strengthen their capacity. This can be in the form of

monitoring, reforming and where possible regulating their operations.

107

REFERENCES

1. Donfouet HP, Makaudze EM. The economic value of the willingness to pay for a

community based prepayment scheme in rural Cameroon. International Labor

Organization (ILO) under the Micro-insurance Innovation Facility 2011; Research

paper No 3.

2. World Health Organization. Health Systems: Improving Performance. The World

Health Report 2000 . Available at http//www.who.int/en/whr00_en.pdf. (Accessed

15th Feb 2013)

3. Consumption pattern in Nigeria. National bureau of statistics 2009/10. Preliminary

report March 2012.

4. Onoka C, Onwujekwe O, Hanson K, Uzochukwu B. Measuring catastrophic health

care expenditures in Nigeria: Implications for financial risk protection. The

Consortium for Research on Equitable Health Systems research brief 2010.

5. World Health organization. Report on the study of Effect of Abolition of cost sharing

in Uganda in collaboration with Ministry of Health Kampala 2002.

6. OECD/W.H.O. DAC Guidelines and Reference Series: Poverty and health,

Organization for economic Co-operation and Development/ World Health

Organization, Paris 2003.

7. Ataguba J, Ichoku H, Fonta W. An estimation of the willingness to pay for

community healthcare insurance scheme in rural Nigeria. A Research Report

submitted to PEP Research Network, Department of Economics, University of Laval.

2013.

8. Olaniyan O, Lawanson A. Health Expenditure and Health Status in Northern and

Southern Nigeria: A Comparative Analysis Using NHA Framework. Paper presented

108

at the 2010 CSAE conference held at St Catherine College, University of Oxford,

Oxford, UK. 2010.

9. Ichoku HE, Fonta WM. The Distributional Impact of Healthcare Financing in Nigeria:

A Case Study of Enugu State . PMMA Working paper 2006. Report No.17:3-22

10. Onwujekwe OE, Uzochukwu BS, Obikeze EN, Okoronkwo I, Ochonma OG, Onoka

CA. Investigating determinants of out-of-pocket spending and strategies for coping

with payments for healthcare in southeast Nigeria. BMC Health Serv Res 2010;10:67.

11. Carrin G, Waelkens MP, Criel B, Community-based health insurance in developing

countries: a study of its contribution to the performance of health financing systems.

Trop Med Int Health 2005;10(8):799-811.

12. Arkin-Tenkorang D. Health insurance for the informal sector in Africa: Design

features, risk protection and resource mobilization, World Bank's Human

Development Network Discussion Paper 2001;1-47

13. Carrin G, Criel B. Community-Based Health Insurance Schemes in Developing

Countries: Facts, Problems and Perspectives; Discussion Paper No.1. World Health

Organization, Department of Health System Financing, Expenditure and Resource

Allocation (FER), Geneva 2003.

14. Riman HB. Healthcare financing and health outcomes in Nigeria: a state level study

using multivariate analysis. International journal of humanities and social science

2002;2(15).

15. World Health Organization Geneva. Community based health insurance schemes in

developing countries: facts, problems and perspectives", Discussion Paper. 2003.

16. World Health Organization, World health survey, Geneva. 2005.

109

17. Shimeles A. Community Based Health Insurance Schemes in Africa: the Case of

Rwanda. Working Papers Series No 1 20, African Development Bank, Tunis, Tunisia

2005.

18. Banwat M, Agbo HA, Osagie JU, Ozoilo JU, Lassa S, Hassan S. Community based

health insurance knowledge and willingness to pay; a survey of a rural community in

Northern Central zone of Nigeria. Jos Journal of Medicine 2010;6(1).

19. Orubuloye IO, Oni JB, Health Transition Research in Nigeria in the Era of the

Structural Adjustment Programme. Health Transition Review (Supplement)

1996;1(6):301-24.

20. Velenyi E. In pursuit of more and ;better managed funds: policy options to purchase

better health for Nigeria: a feasibility study of the National Health Insurance Scheme

of Nigeria. Washington DC: World Bank 2005.

21. Soyibo A, Olaniyan O, Lawanson AA. Situational Analysis of the Nigerian Health

Sector. Report Submitted to Management Education Research Consortium,

Washington DC 2009.

22. The World Bank. The World Bank group 2013. Available at

http//data.worldbank.org/country/nigeria. (Accessed 17th june 2013)

23. Mays GP, Smith SA. Evidence links increases in public health spending to declines in

preventable deaths. Health Affairs 2011;4:55

24. Soyibo A, Lawanson O, Olaniyan L. National health accounts of Nigeria, 2003-2005:

University of Ibadan 2010.

25. Soyibo A, Lawanson O, Olaniyan L. National Health Accounts of Nigeria, 1998-

2002. Final report submitted to World Health Organization, Geneva: University of

Ibadan 2005.

110

26. Lagarde M, Palmer N.The impact of user fees on health service utilization in low- and

middle- income countries: how strong is the evidence? Bulletin of the World Health

Organization 2008;86(11):839-46.

27. Uguru NP, Onwujekwe O, Uzochukwu B, Igiliegbe G, Eze S. Inequities in incidence,

morbidity and expenditures on prevention and treatment of malaria in southeast

Nigeria. BMC International Health and Human Rights 2009;9(21).

28. Onwujekwe O, Onoka C, Uzochukwu B, Hanson K. Constraints to universal

coverage: inequities in health service use and expenditures for different health

conditions and providers. Int J Equity Health 2011;10(1):50.

29. Onwujekwe EO, Chima RI, Okonkwo PO. Economic burden of malaria illness versus

that of a combination of all other illnesses: A study in five malaria holo-endemic

communities. Health Policy 2000;54:143-59.

30. Chima RI, Goodman CA, Mills A. The Economic Impact of Malaria in Africa: a

Critical Review Evidence. Health Policy 2003;63:1-17.

31. Onwujekwe EO. Inequities in healthcare seeking in the treatment of communicable

endemic diseases in Southeast Nigeria. Social Science and Medicine 2005;61(2):455-

63.

32. Gottret P, Schneider G. Health financing revisited: a practitioner's guide overview.

Washington DC. The international Bank for reconstruction and Development/World

Bank 2006.

33. Uzochukwu B, Onwujekwe O, Eze S, Ezuma N, Obikeze E, Onoka C. Community

based health insurance scheme in Anambra State, Nigeria: an analysis of policy

development, implementation and equity effects: Consortium for Research on

Equitable Health Systems, London School of Hygiene and Tropical Medicine. 2009.

111

34. Dutta A, Hongoro. Scaling up National Health insurance in Nigeria: Learning from

Case Studies of India, Colombia, and Thailand. Washington, DC: Futures Group,

Health Policy Project 2013.

35. Adinma ED, Adinma BJ. Community based healthcare financing: An untapped option

to a more effective healthcare funding in Nigeria. Nigerian Medical journal

2010;51:95-100.

36. National Health Insurance Scheme (NHIS). Road map for the implementation of

Community-based health insurance scheme in Nigeria. In Consultation on Support to

NHIS CBSHIS roll-out in Programme States. Abuja. PPRINN-MNCH/NHIS 2011.

37. Jutting J. Do community-based health insurance schemes improve poor people's

access to health care? Evidence from rural Senegal. World development 2004;2:273-

88.

38. Ataguba JE. Community Health Insurance Scheme as a viable option for rural

population in Nigeria. A Paper submitted to the Centre for the Study of African

Economies (CSAE) Department of Economics, University of Oxford 2008.

39. Gwatkin DR, Wagstaff A, Yazbeck AS. Reaching the Poor with Health, Nutrition and

Population Services. What Works, What Doesn't, and Why. The World Bank,

Washington D.C. 2005.

40. Dror DM, Radermacher R, Koren R. Willingness to pay for health insurance among

rural and poor persons: field evidence from seven micro health insurance units in

India. Health Policy 2007;82(1):12-27.

41. Lewalen S, Geneau R, Mahande M, Msangi J, Nyumpumbwe S, Kitumba R.

Willingness to Pay for cataract surgery in two regions of Tanzania. Br J Ophthalmol

2006;90(1):11-3.

112

42. Asfaw A, Gustafsson-Wright E, VanderGaag J. Willingness to pay for health

insurance: An analysis of the potential market for new low-cost health insurance

products in Namibia, Amsterdam Institute for International Development; AIIDRS –

08 01/2: 1–22. AIIDRS 2008;08 01(2):1-22.

43. Brown W, Churchill C. Insurance provision in low-income communities: part 2-

initial lessons from micro-insurance experiments for the poor. Calmeadow,

Microenterprise Best Practices (MBP) Project, Development Alternatives Inc. 2000;

p. 1-4.

44. Onwujekwe O, Onoka C, Uzochukwu B, Okoli C, Obikeze E, Eze S. Is community-

based health insurance an equitable strategy for paying for healthcare? Experiences

from southeast Nigeria. Health Policy 2009;92(1):96-102.

45. Dong H, Kouyate B, Cairns J, Sauerborn R. Differential willingness of household

heads to pay community-based health insurance premia for themselves and other

household members. Health Policy Plan 2004 Mar;19(2):120-6.

46. Onwujekwe O, Okereke E, Onoka C, Uzochukwu B, Kirigia J, Petu A. Willingness to

pay for community-based health insurance in Nigeria: do economic status and place

of residence matter? Health Policy Plan 2010 Mar;25(2):155-61.

47. Mtei G, Mulligan J. Community health funds in Tanzania: A literature review, Ifakara

Health Research and Development 2007; 1-15

48. World Bank. Improving Primary Health Care Delivery in Nigeria: Evidence from

Four States. Working paper No 187. Washington DC: World Bank 2010.

49. Katung PY. Socio-economic factors responsible for poor utilization of PHC services

in rural community in Nigeria. Niger J Med 2001;10:28-9.

50. Omotosho O. Health Seeking Behaviour among the Rural Dwellers in Ekiti State,

Nigeria. African research review 2010;4(2):125-38.

113

51. Ajibade BL, Amoo PO, Adeleke MA, Oyadiran GO, Kolade OA. Determinants of

mothers health seeking behaviour for their children in a Nigerian Teaching Hospital.

IOSR Journal of Nursing and Health Science 2013;1(6):9-16.

52. Sule SS, Ijadunola KT, Onayade AA, Fatusi AO, Soetan RO, Connell FA. Utilization

of primary health care facilities: lessons from a rural community in Southwest

Nigeria. Niger J Med 2008;17(1):98-106.

53. Dairo MD, Owoyokun KE. Factors affecting the utilization of antenatal care services

in Ibadan, Nigeria. Niger J Med 2010;12(1):3-13.

54. Carrin G, Evans D, Xu K. Designing health financing policy towards universal

coverage. Bulletin of the World Health Organization 2007;85:652.

55. Spaan E, Mathijssen J, Tromp N, McBain F, Have A, Baltussen R.The impact of

health insurance in Africa and Asia: a systematic review. Bull World Health Organ

2012 Sep 1;90(9):685-92

56. World Health Organization. Bulletin of the World Health Organization

2010;88(5):321-40.

57. Drouin A. Methods of financing health care: A rational use of financing mechanisms

to achieve universal coverage. International Social Security Association Technical

report 05;2008.

58. Dreschler D, Jutting J. Private Health Insurance in Low and Middle-Income

Countries: Scope, Limitations, and Policy Responses; Issy-les Moulineaux: OECD

Development Centre 2005.

59. Palmer N, Mueller D, Gilson L, Mills A, Haines A. Health financing to promote

access in low income settings - How much do we know? Lancet 2004;364:1365-70.

60. World Health organization. WHO Country Cooperation Strategy: Nigeria 2008-2013.

Brazzaville: WHO Regional office for Africa 2009.

114

61. Preker AS. Feasibility of Mandatory Health Insurance - Challenges in Health Care

Financing; Abuja Flagship Courses, Abuja, the World Bank 2005.

62. Devadasan N, Criel B, Van Damme W, Ranson K, Van der Stuft P. Indian

community health insurance schemes provide partial protection against catastrophic

health expenditure. BMC Health Service Res 2007;7(43).

63. Hsiao W, Shaw RP. Social health insurance for developing nations. The World Bank

2007.

64. McIntyre D, Thiede M, Dahlgren G, Whitehead M. What are the economic

consequences for households of illness and of paying for health care in low- and

middle-income country contexts? Soc Sci Med 2006;62:858-65.

65. World Health Organization. Health systems financing: the path to universal coverage.

Geneva. World Health Report 2010.

66. Olakunde BO. Public health care financing in Nigeria: Which way forward? Ann

Nigerian Med 2004;6:4-10.

67. Cavagnero E, Lane C, Evansa DB, Carrina G. Development assistance for health:

Should policy-makers worry about its macroeconomic impact? Bull World Health

Organ 2008;86:864-70.

68. Tabor SR. Community based health insurance and social protection policy. The

World Bank 2005; Available from: http://www.worldbank.org/sp. (Accessed 20th

February 2013)

69. Lagarde M, Palmer N. Evidence from Systematic Reviews to Inform Decision-

Making Regarding Financing Mechanisms That Improve Access to Health Services

for Poor People: A Policy Brief Prepared for the International Dialogue on Evidence-

Informed Action to Achieve Health Goals in Developing Countries (IDEA Health) in

115

Khon Kaen; Thailand. 13-16 December 2006. Geneva: Alliance for Health Policy and

systems Research. 2006.

70. Uzochukwu B, Onwujekwe O, Eriksson B. Inequity in the Bamako Initiative

programme-Implications for the treatment of malaria in south-east Nigeria. Int J

Health Plann Manage 2004;19:107-16.

71. Russell S. Ability to pay for health care: Concepts and evidence. Health Policy Plan

1996;11:219-37.

72. Berki SE. A look at catastrophic medical expenses and the poor. Health Aff

(Millwood) 1986;5:138-45.

73. Office of the Head of Civil Service of the Federation 2013; Available from:

http://www.ohcsf.gov.ng/index-2.html. ( Accessed 22nd February 2013)

74. National Health Insurance scheme. Available from: http://www.nhis.gov.ng/.

(Accessed 22nd February 2013)

75. Kannegiesser L. National Health Insurance Scheme to boost generics market in

Nigeria. Available from: http://www.frost.com/prod/servlet/market-insight-

top.pag?Src=RSSanddocid=155485216. (Accessed 22nd February 2013)

76. World Bank. World bank-Administered Groba Launches Pre-paid Health Insurance

Scheme in Lagos, Nigeria. United Nations Development Programme. MDGs Needs

Assessment and Financing Strategy for Nigeria: Policy Brief 2008. Available from:

http://www.ng.undp.org/mdgs/policy_brief.pdf. (Accessed 22nd February 2013)

77. Catherine E, Vialle V, Dennis RD, Ntagamia J, Wagner A. Medicines coverage and

community based health insurance in Low income countries. Health research policy

and systems 2008;11.

78. Uzochukwu B, Onwujekwe O, Soludo E, Nkoli E, Uguru N. The District Health

System in Enugu State, Nigeria: An analysis of policy development and

116

implementation. Consortium for Research on equitable health systems 2009.Available

from:http:/www.crehs.lshtm.ac.uk/downloads/publications/District_health_system_in

Enugu_state.pdf. (Assessed 20th March 2013)

79. Onwujekwe O, Valenyi E. Feasibility of voluntary health insurance in Nigeria. The

World Bank 2006.

80. Ichoku EH, Fonta WM. 'Catastrophic Healthcare Financing and Poverty; Empirical

Evidence from Nigeria'. Journal of Social and Economic Development 2009;11(2):1-

16.

81. Cutler D, Zeckhauser R, Richard J. The anatomy of health insurance: Handbook of

Health Economics Volume 1. 1st edition. Edited by Culyer AJ, Newhouse JP.

Elsevier Inc, Amsterdam, the Netherlands; 2000:563-643.

82. Adinma ED, Nwakoby BA, Adinma BD. Integrating maternal health services into a

health insurance scheme: effect on healthcare delivery. Nig Q J Hosp Med 2010

Apr;20(2):86-93.

83. Poletti T, Balabanova D, Ghazaryan O, Kamal-Yann M, Kocharyan H, Arekelyan

K, Hakobyan M. Options for scaling up community-based health insurance for rural

communities of Armenia. Health System Development Programme, London. London

School of Hygiene and Tropical Medicine 2007.

84. Ekman B. Community-based health insurance in low-income countries: a systematic

review of the evidence. Health Policy Plan 2004 Sep;19(5):249-70.

85. Stoermer M, Fuerst F. Review of Community-based Health Insurance Initiatives in

Nepal 2012.

86. Mebratie A, Sparrow R, Yilma A, Abebaw D, Alemu G, et al., Impact of Ethiopian

pilot community-based health insurance scheme on health-care utilisation: a

household panel data analysis. the lancet 2013;381:S92.

117

87. Lagos state government, Lagos state. Available at http://www.lagosstate.gov.ng/c/portal/layout?p_1_id=PUB.1095.11 (Accessed 2nd April.

2013)

88. Zamba A. "Implementing Community Based Health Insurance Scheme: Experience of

Ikosi Isheri CBHI Scheme in Lagos." Conference presentation 2011. Available at:http://www.healthfinancenigeria.org/index.php?option=com_jdownloads&Itemid=195&vi ew=viewcategory&catid=12. (Accessed 22nd Feb 2013)

89. Improving financial access to health services for the poor in Nigeria 2011. http://www.healthpolicyproject.com/pubs/97_communitybasedhealthinsurance.pdf.

(Accessed 22nd Feb 2013)

90. Adeusi O. "Community Health Insurance Schemes-Kwara/Hygeia." Conference presentation 2011. Available at: http://www.healthfinancenigeria.org. (Accessed 22nd Feb

2013)

91. Jehu-Appiah C, Aryeetey G, Agyepong I, Spaan E, Baltussen R. Household perceptions and their implications for the enrolment in the NHIS in Ghana. Health policy and planning

2011;1-12.

92. Curt L, Nguyen T, Anders E, Lars L. Peoples willingness to pay for health insurance in rural vietnam. Cost Eff Resour Alloc 2008;6(16):6-16.

93. Shisana O, Rehle T, Dirwayi N, Dana P, Rispel L. Public perception on National health insurance: moving towards universal coverage in South Africa. S Afr Med J 2006;96:814-8.

94. VanderGaag J. Health care for the world's poorest: Is voluntary (private) health insurance an option? 2020 Focus Brief on World's Poor and Hungry People 2007.

95. Barnighausen T, Lin X, Zang X, Sauerborn R, Willingness to pay for social health insurance among informal workers in Wuhan, China: A contingent valuation study, London

Biomed Central Health Services Research 2007.

118

96. Asgary K, Taghvaei A, Rafeian M. Estimating rural households' willingness to pay for health insurance. European Journal of Health Economics 2004;5:581-7.

97. Ichoku E, Fonta W, Ataguba J. Estimating the willingness to pay for community health insurance schemes in Nigeria: A random valuation framework. The IUP Journal of Risk and

Insurance 2012;7(1&2):7-27.

98. Oyekale AO. Factors Influencing Households' Willingness to Pay for National Health

Insurance Scheme (NHIS) in Osun State. Nigeria Ethno Med 2012;6(3):167-72.

99. Carson RT. "Contingent Valuation: a User's Guide." Environmental Science &

Technology 2000;34:1413-1418.

100. Carson RT, Flores NE, Meade NF. 'Contingent Valuation: Controversies and Evidence' .

Environmental and Resource Economics 2001;19(2):173-210.

101. Lwambo NJ, Siza JE, Mwenda GC. Community's willingness to pay for a school-based chemotherapy program to control morbidity due to schistosomiasis and soil-transmitted helminthiasis in children in rural Tanzania. Tanzania Health Research Bulletin 2005;7:149-

53.

102. Usman AB. Willingness to Pay For Community Based Health Care Financing Scheme:

A Comparative Study among Rural and Urban Households in Osun State, Nigeria. IOSR

Journal of Dental and Medical Sciences (IOSR-JDMS) 2013;5(6):27-40.

103. Babatunde OA, Akande TM, Salaudeen AG, Elegbede OE, Ayodele LM, Aderigbigbe

SA. Willingness to Pay for Community Health Insurance and its Determinants among

Household Heads in Rural Communities in North-Central Nigeria. International Review of

Social Sciences and Humanities 2012;2(2):133-42.

104. Oriakhi HO, Onemolease EA. Determinants of Rural Household's Willingness to

Participate in Community Based Health Insurance Scheme in Edo State, Nigeria. Ethno Med

2012;6(2):95-102.

119

105. Asenso-Okyere WK, Nsowah-Nuamah N, Alberson P. "Characterizing the Poor in

Ghana: A Logit Approach", in Sustainable Food Security in West Africa 1997.ed by Asenso-

Okyere WK, Tims W. Kluwer Academic Publishers, Dordrecht.

106. Edward H. Demand for health insurance among women in Ghana: cross sectional evidence. International research journal of finance and economics 2009;33(1).

107. Suminder KB, Ruchita M. Awareness and willingness to pay for health insurance; An empirical study with reference to Punjab India. International journal of humanities and social science 2011;1(7).

108. De Allegri M, Sanon M, Sauerborn R. "To enrol or not to enrol?": A qualitative investigation of demand for health insurance in rural West Africa. Soc Sci Med

2006;62(6):1520-7.

109. Dong H, De Allegri M, Gnawali D, Souares A, Sauerborn R. Drop-out analysis of community-based health insurance membership at Nouna, Burkina Faso. Health Policy 2009

Oct;92(2-3):174-9.

110. Feeley F. Commercial Market Strategies in Sub-Saharan Africa: Lessons Learned in

Community Health Financing. Occasional Paper Series, USAID Commercial Market

Strategies Project, Washington DC. 2003.

111. Asfaw A, Von BJ. Innovations in health care financing: new evidence on the prospect of community health insurance schemes in the rural areas of Ethiopia. Int J Health Care Finance

Econ 2005 Sep;5(3):241-53.

112. Wang H, Yip W, Zhang L, Wang L, Hsiao W. Community-based health insurance in poor rural China: the distribution of net benefits. Health Policy Plan 2005 Nov;20(6):366-74.

113. Ajala OA, Lekan S, Adeyinka S.A. Accessibility to Healthcare facilities; a panacea for sustainable rural Development in Osun state Southwestern Nigeria. J Hum Ecol

2005;18(2):121-8.

120

114. Usman AB. Preferred Mode of Payments for Community Based Health Care Financing

Scheme among Households in Osun State, Southwestern Nigeria. Journal of Natural Sciences

Research 2013;3(1):93-102.

115. National population commission. National census figures, Abuja, Nigeria. NPC 2006.

116. Lagos state government, Welcome to Lagos state, centre of excellence: an authoritative guide book on Lagos state of Nigeria. 1 ed. 1998.

117. Ikorodu Local government, Brief History of Ikorodu local government; Information

Ikorodu local government 2008.

118. National Population Commission (NPC) [Nigeria] and ICF International. Nigeria demograhic and health survey 2013. Abuja, Nigeria, and Rockville, Maryland, USA: NPC and ICF International 2014.

119. Dilon A, Beaman L. Do Household Definitions Matter in Survey Design? Results from a Randomized Survey Experiment in Mali. International Food Policy Research Institute 2010; discussion paper 01009

120. Fleis JL. Statistical Methods for Rates and Proportions. New York, NY: John Wiley and

Sons, Inc. 1981.

121. Masud T, Farooq N. Equity shortfalls and Welfare State: Willingness to pay for health care at Government facilities in Jhelum. J Ayub Med Coll Abbottabad 2003;15(4):43-9.

122. Sanusi RA, Awe AT. Perception of National Health Insurance Scheme (NHIS) by

Health Care Consumers in Oyo State, Nigeria. Pakistan Journal of Social Sciences

2009;6(48):53.

123. Dong H, Kouyate B, Cairns J, Mugisha F, Sauerborn R, Willingness-to-pay for community-based insurance in Burkina Faso. Health Econ 2003 Oct;12(10):849-62.

124. National Population Commission and Orc Macro. Nigeria Demographic and Health

Survey (NDHS), Calverton, Maryland: 2008; 165

121

APPENDIX ONE

INFORMED CONSENT

Purpose of Research

You are invited to participate in a research titled “HOUSEHOLD WILLINGNESS TO PAY FOR COMMUNITY BASED HEALTH INSURANCE IN COMMUNITIES IN URBAN AND RURAL LGAs IN LAGOS: A COMPARATIVE STUDY.

As you probably know, high cost of health care prevents many people from accessing much needed health care every day. It is an issue that requires immediate strategic attention. Findings from this study will form recommendations to government and stakeholders on how to improve health care financing in Lagos state

Procedure of Research

If you choose to participate, the researcher and study staff will ask you some questions from a questionnaire on your socio-demographic characteristics, seek some information about your occupation, seek your healthcare seeking habits, as well as hear any suggestions you have for reducing the burden of healthcare cost for the people of your community. We would also like to understand your perceptions and willingness to participate in a community-based health insurance program.

Voluntariness

Your participation in this study is entirely voluntary. Your decision to participate or not will not prejudice your future relations with the investigator or his institution (LUTH, Idi Araba)

Confidentiality

All information collected in this study will be given code numbers and no name will be recorded. This cannot be linked to you in anyway and your name or any identifier will not be used in any publication or reports from this study. The results from this research study may be presented at scientific or medical meetings or published in scientific journals. However your identity will not be disclosed.

Participant Responsibility

As a participant, your responsibilities include;

 Complete your questionnaire as instructed.  Following the instructions of the Protocol Director and research staff.  Ask questions as you think of them.

122

 Tell the Protocol Director and research staff if you change your mind about staying in the study.

If you decide to participate, you are free to withdraw your consent and to discontinue participation at any time without penalty.

Benefits to researchers The results from this study will benefit researchers as it will add to the growing body of knowledge and in the long run better health care delivery.

Conflicting Interest No apparent or potential conflict of interest

Statement of the person obtaining informed consent

I have fully explained this research to______and have given sufficient information, including risks and benefits to make an informed decision.

DATE______SIGNATURE______

NAME______

Statement of person giving consent:

I have read the description of the research or have had it translated into language I understand. I have also talked it over with the doctor to my satisfaction. I understand that my participation is voluntary. I know enough about the purpose, methods, risks and benefits of the research study to judge that I want to take part in it. I understand that I may freely stop being part of this study at any time. I have received a copy of this consent form and additional information sheet to keep for myself.

DATE______SIGNATURE______

In case of any enquiry please contact:

Researcher:

Dr Ijeoma Agbo

Dept of Community Health and Primary care,

LUTH, Idi-Araba

08037250415

LUTH contact:

Health research & Ethical committee

Room 107, Administrative office, LUTH

123

INFORMED CONSENT

Purpose of Research

You are invited to participate in a research titled “HOUSEHOLD WILLINGNESS TO PAY FOR COMMUNITY BASED HEALTH INSURANCE IN COMMUNITIES IN URBAN AND RURAL LGAs IN LAGOS: A COMPARATIVE STUDY.

Procedure of Research

Voluntariness

Your participation in this study is entirely voluntary. Your decision to participate or not will not prejudice your future relations with the investigator or his institution (LUTH, Idi-Araba)

Confidentiality

Participant Responsibility

As a participant, your responsibilities include;

 Complete your questionnaire as instructed.  Following the instructions of the Protocol Director and research staff.  Ask questions as you think of them.  Tell the Protocol Director and research staff if you change your mind about staying in the study.

124

If you decide to participate, you are free to withdraw your consent and to discontinue participation at any time without penalty.

Benefits to researchers The results from this study will benefit researchers as it will add to the growing body of knowledge and in the long run better health care delivery.

Conflicting Interest No apparent or potential conflict of interest

Statement of the person obtaining informed consent

I have fully explained this research to______and have given sufficient information, including risks and benefits to make an informed decision.

DATE______SIGNATURE______

NAME______

Statement of person giving consent:

DATE______SIGNATURE______

In case of any enquiry please contact:

Researcher:

Dr Ijeoma Agbo

Dept of Community Health and Primary care,

LUTH, Idi-Araba

08037250415

LUTH contact:

Health research & Ethical committee

Room 107, Administrative office, LUTH

125

QUESTIONNAIRE ON COMPARTATIVE STUDY OF HOUSEHOLD WILLINGNESS TO PAY FOR COMMUNITY BASED HEALTH INSURANCE IN URBAN AND RURAL LGAs IN LAGOS STATE

I am a senior registrar from the Department of Community Health, LUTH, Idi-Araba undertaking a study on the above named subject, in partial fulfilment of the requirement for the award of the National Postgraduate Medical College of Nigeria in Public Health. Kindly answer the following questions. All information will be treated with utmost confidentiality.

Identification No......

Local Government Area......

Ward......

Street/House No......

Please, enter the appropriate number representing the answer given in the spaces provided. Please note that throughout the questionnaire, if YES write 1 in the box, if NO, write 0 in the box. All boxes MUST be filled.

SECTION 1: SOCIODEMOGRAPHIC INFORMATION 1. Do you consider yourself to be the main decision-maker in your household about what your household spends money on? Yes [ ] No [ ]

2. Who is the main income earner in this household? You [ ] spouse [ ] others [ ]

If others please specify______

3. What is your status in this household? a. female head of household[ ] b. male head of household[ ] c. Wife [ ] d. Grandmother [ ] e. representative of household[ ]

4. How old are you (as at last birthday)? ______

5. Respondent’s Sex? ______

6. How many people live in this household, including you? ______

7. How many adults (older than or equal to 18 years) live here? ______

126

8. How many younger people (less than 18 years) live here? ______

(Add the responses in questions 7-8 and see if they add up to question 6 response)

9. What was your highest completed education level? a. No formal education [ ] b. Primary [ ] c. Junior Secondary [ ] d. Senior Secondary [ ] e. Tertiary [ ] f. Others [ ] ______

10. What was the total number of years that you spent schooling? ______

11. What occupation is your major source of income? (Multiple answers allowed) a. Unemployed [ ] b. Farmer [ ] c. Petty trading [ ] d. Unskilled manual worker [ ] e. skilled manual worker [ ] f. Civil servant [ ] e. employed in private sector [ ] f. large scale business [ ] g. Self-employed professional [ ] h. others [ ] please specify:

______

12. What occupation is the household’s major source of getting money? ______

(Note that it may or may not be respondent’s occupation]

13. What is your estimated average monthly income in naira? a. < 5000 [ ] b. 5001-10,000 [

] c. 10,001-20,000 [ ] d. 20,001-30000 [ ] e. >30,000 [ ]

14. How much did your household spend to purchase food from the market in the past one week? ______Naira

15. Do you belong to any association or community group (Cooperative or Ajo) in whom money is being managed for you? Yes [ ] No [ ]

SECTION 2: HEALTH SEEKING, COST OF ILLNESS AND PAYMENT MECHANISMS

SECTION 2A: HEALTH SEEKING BEHAVIOUR

16. Has there been any illness in your household in the last three months? Yes [ ] No [ ] (If no illnesses go to section 2b)

127

17. What was the most recent type of sickness or poor health condition in your household within the last 3 months? a. Malaria (fever, body weakness) [ ] b. Typhoid (fever, abdominal pain, vomiting) [ ] c. Diarrhoea (stooling) [ ] d. Respiratory tract infections (catarrh and cough) [ ] e. Others [ ] please specify:______

18. Where was treatment for the illness first sought? [Do not read list. Mark first response only.] a. Chemist (patent medicine dealer) [ ] b) Traditional healers [ ] c. Primary health centre [ ] d. General hospital [ ] e. Tertiary centre [ ] f. Private hospital [ ] g. Others: [ ] Please specify______

19. How much did it cost to receive this treatment (including cost of registration/card, cost of drugs, laboratory tests, x-rays, etc)? ______Naira

20. What form of transportation was used to reach the location where treatment was obtained? a. Personal vehicle: [ ] b. Bus (public transport) [ ] c. Taxi [ ] d. Okada/keke-marwa [ ] e. Walked [ ] f. others [ ] Please specify ______

21. How long did it take to get to the location of treatment? a. Less than 15mins [ ] b. 15- 30mins [ ] c. 30-45mins [ ] d. 45- 1hour [ ] e. greater than 1 hour [ ] f. greater than 2 hours [ ]

22. How much did you spend on transportation to receive this treatment (to and fro)? ______Naira

23. Total cost spent? ______Naira [no 19 + 22]

24. Did you recover after receiving this treatment? ______[If YES, go to Section 2B, if NO, go to question 25]

25. What was the total cost you spent (transportation/drugs/others) until you recovered? ______

SECTION 2B: PAYMENT AND PAYMENT COPING MECHANISMS FOR RESPONDENT

128

26. How was the first treatment paid for? [Multiple responses are allowed] a. Out-of pocket/cash [ ] b. Employer [ ] c. Health Insurance [ ] Please specify type ______d. Instalment [ ] e. In-kind [ ] Please specify ______f. others (please specify)______

27. How did you cope with the payment for the treatment? (Multiple responses are allowed) a. Own money [ ] b. Borrowed money/took a loan [ ] Please specify source of the loan ______c. Sold household movable assets [ ] d. Sold family land [ ]e. Payment was subsidised [ ] f. Payment was deferred [ ] f. Community solidarity/someone else paid [ ] h. was exempted from payment [ ] i. others (specify) ______

For those that used any form of health insurance, ask question 28

28. Why did you or your household take out a health insurance policy? (Multiple responses are allowed) a. financial protection against the cost of illness [ ]b. Provide access for household members to affordable healthcare [ ]c. Provide good quality treatment for reduction of income loss due to ill health [ ]d. Relative/Friend asked me to join [ ] e. Others [ ] Please specify ______

SECTION 3a: KNOWLEDGE OF HEALTH INSURANCE 29. Have you heard of health insurance before? Yes [ ] No [ ] If Yes answer q30-31, if No move to section 3b 30. If yes, what type of health insurance do you know? (Tick all types mentioned) a. National Health Insurance Scheme [ ] b. Private health insurance [ ] c. Community Based Health Insurance [ ] d. Others [ ] please specify ______e. don’t know [ ] 31. What is health insurance about? A. Prepayment for healthcare [ ] b. Paying tax to government [ ] c. Free health delivery by government [ ] d. others [ ] please specify______e. Don’t know [ ]

SECTION 3b: PERCEPTIONS ABOUT THE USE OF COMMUNITY-BASED

HEALTH INSURANCE TO IMPROVE PAYMENT FOR HEALTHCARE

129

Health insurance is a program that puts together the risk of several people in an effort to decrease the amount that is paid by an individual at the time health care services are needed. Ill- health occurrence is largely unpredictable for individuals, as well as the need for health-care is often highly unpredictable and very costly for most individuals. It provides an opportunity to spread the financial burden of payment over several people, making health care more affordable for individuals. It can be either compulsory or voluntary. Health insurance can be offered at a community-level and is known as community-based health insurance. It is a non-profit type voluntary scheme where the people of the community pay a predetermined amount and depending on the benefits packages available, will have access to various health providers and services. CBHI provides some financial protection by reducing out-of-pocket spending which imposes heavy burden on households and it severely impedes access to health care utilization by particularly by those who stand in greatest need of care. CBHI offers opportunity for improvement in quality of services

32. From either your understanding, how would you rate the ability of community-based health insurance to make health care more affordable for you and the members of your household? [ ] 1 = low 2 = medium 3 = high

33. What do you think is the households’ potential level of accessing affordable healthcare due to community-based health insurance? [ ] 1 = low 2 = medium 3 = high

34. How would you rate the potential of community-based health insurance to improve household health consumption patterns? [ ] 1 = low 2 = medium 3 = high

35. How would you rate the potential of CBHI to improve the quality of services provided by healthcare givers? [ ] 1 = low 2 = medium 3 = high

36. How would rate the potential of CBHI to ensure constant availability of drugs at health facilities in your community? [ ] 1 = low 2 = medium 3 = high

37. How would you rate your level of confidence if funds are pooled together and managed by the community? [ ] 1 = low 2 = medium 3 = high

SECTION 3c: ACCEPTIBILITY OF COMMUNITY-BASED HEALTH INSURANCE

38. Is community-based health insurance acceptable to you as a strategy for paying for health care in this area? [ ] if yes (go to Q39) if no (go to Q40).

130

39. If Q50 is yes, please score your level of acceptability of CBHI from 1 to 10 [ ] where 1 is least preferred and 10 is most preferred.

40. If CBHI is not acceptable, why? ______(read responses and please circle just one)

1. I don’t trust the people that will manage the fund

2. I’m always in good health

3. I am afraid of the repercussion of the programme

4. I don’t want such policy

5. I cannot afford it

6. I don’t know

7. Other reasons (please specify)______

SECTION 4: WILLINGNESS TO PAY (WTP) FOR COMMUNITY-BASED HEALTH INSURANCE

I would briefly introduce a proposed CBHI scheme for yourself and other family members and thereafter ask a few questions.

The government has introduced a new way of helping Lagosians pay for medical care in your area. Your household could purchase a community based health insurance scheme for your family. Membership in the scheme is voluntary and you will pay a predetermined amount and depending on the benefits packages available, will have access to various health providers and services. You will be required to pay a premium in order to join the scheme. The premium can be paid monthly or once a year, however, no credit is allowed. After payment, you/your household will receive services free of charge at the point of consumption. Each registered person will be issued with an identity cared renewable yearly after payment of premium. The health scheme would consist of a basic health package, which includes Out- patient care including necessary consumables; Essential drugs and diagnostic tests; Maternity care for up to four live births; Preventive care such as immunization, health education, family planning, antenatal care and postnatal care; Consultation with specialists; Hospital care in a standard ward for a limited stay of cumulated 45 days per

131 year; Eye examination and care which excludes provision of glasses or lens; Preventive dental care and pain relief.

Healthcare would be delivered by providers in your area which is registered with the CBHI management group. The services will be accessed from primary healthcare centres, general hospitals, selected private hospitals and mission hospitals that the committee will register and consumers are restricted to the providers selected. The CBHI will be managed by a committee that your community will establish, supported by the state government. At the end of 12 months you could renew or drop out of the scheme.

41. Will you be willing to enrol in the CBHI scheme for you and your household if offered? Yes [ ] No [ ] b. If no to Q41, what is the major reason? a. Do not need insurance [ ] b. I am healthy/ rarely ill [ ] c. No confidence/trust in the scheme [ ] d. I don’t think I can afford the premium [ ] e. I have private insurance [ ] f. No close health facility [ ] g. Others [ ] Please specify ______

We would now like to know the maximum amount of money that you will be willing to pay for individual and for household.

42. WILLINGNESS TO PAY FOR INDIVIDUAL ENROLEES a. The price of a monthly insurance premium (contribution) is 500 Naira; are you willing to pay? Yes [ ] No [ ] Yes (Go to 42 g) No (go to 42b) Do not know (go to 42b) b. What is the maximum amount you are willing to pay? ______(Interviewer: if more or equal to 400 Naira go to 42c, but if less than 400 Naira, go to 42d) c. What if the premium is 450 Naira, will you be willing to pay? ______1 = yes 0 = No (Interviewer: no matter the answer, go to 42g). d. What if the premium is 350 Naira, will you be willing to pay? ____1 = yes (42g) 0 = No (42e) e. What really is the maximum amount you are willing to pay for the CBHI premium? ______(Interviewer: If more or equal to 400 Naira go to 42g, but if equal to or less than 400 Naira go to 42f)

132 f. The amount that you have quoted is too low, and cannot cover the cost of the premium, and so you will have to increase the amount if you really want to join the CBHI scheme. So what is the final maximum amount you are willing to pay per month to join the scheme? ______(Interviewer: No matter the answer, go to 42g) g. If due to inflation or other uncertainties, the premium for the CBHI scheme increases, what is the maximum amount you are very certain to pay bearing in mind your average monthly household income and money you spend on various items? ______

43. WTP FOR OTHER HOUSEHOLD MEMBERS a. The price of a monthly insurance premium is 900; are you willing to pay this amount of money for your household? [ ] 1 = Yes (43g) 0 = No (43b) 2=Do not know (43b) b. What is the maximum amount you are willing to pay? ______( if more or equal to 800 Naira go to 43c, but if less than 800 Naira, go to 43d) c. What if the premium is 850 Naira, will you be willing to pay? ______(Interviewer: no matter the answer, go to 43g). d. What if the premium is 750 Naira, will you be willing to pay? [ ] 1 = yes (43g) 0 = No (43e) e. What really is the maximum amount you are willing to pay for the CBHI premium? ______Interviewer: If more or equal to 800 Naira go to Q57g, but if less than 800 Naira go to Q43f) f. The amount that you have quoted is too low, and cannot cover the cost of the premium, and so you will have to increase the amount if you really want your other household members to enrol in the scheme. So what is the final maximum amount you are willing to pay per month per person? ______(Interviewer: No matter the answer, go to Q43g) g. If due to inflation or other uncertainties, the premium increases, what is the maximum amount you are very certain to pay per household member bearing in mind your average monthly household income and money you spend on various items? ______44. Are there comments that you wish to make about the CBHI scheme?

Thank you for your time

133

Ibeere ati idahun lori eko ati ife nipa eto ilera lese kuku ati labele pelu Igboro ni awon ijoba ibile ni Ipinle Eko

Mo je onisegun akekoo agba ni abala eka eko alabode ni Idi-Araba, mo n ko eko nipa ohun to wa loke yii, nipa ibamu lori awon amuye fun eto gbigba eko ijinle eleekeji ninu eto Eko

Isegun ti alaboode, e dahun awon ibeere wonyii. Gbogbo ohun ti e ba so ni a o se ni asiri.

Nonba idanimo ………………………………………………………………………………..

Agbegbe Ijoba Ibile ……………………………………………………………………………

Woodu ……………………………………………………………………………………..

Titi/ Nonba ile ………………………………………………………………………………….

E fi nonba to ba ye si idahun si awon alafo ti a ti pese.

Ninu gbogbo awon idahun si ibeere yii ti o ba se pe beeniko 1 sinu apoti, ti o ba se pe beeko ni, koo sile, Gbogbo apoti ni a gbodo ko nnkan si.

Ipin Akoko: Iroyin to je mo ikaniyan awujo

1. Nje o ka ara re kun eni to n se Ipinnu ninu ile nipa bi awon ara inu ile se n na owo?

Beeni [ ] Beeko [ ]

2. Bawo ni ipo re nipa igbeyawo? O ti gbeyawo [ ] Apon [ ] Eni to koko/aya [ ]

Eni ti o gbe papo mo/yapa [ ] Opo [ ]

3. Kinni ipo re ninu ile yii? (a) obinrin agba ile (b) okunrin agba ile (c) oko/aya

(d) Iya agba (e) asoju ile

4. Omo odun meloo ni o ? ………………………………………………………………

5. So boya obinrin/okunrin ni o? …………………………………………………………

6. Eniyan meloo lo n gbe ninu ile re pelu iwo? …………………………………..

7. Agba meloo (eni to dagba / to omo odun mejidinlogun) to n gbe ibi?

………………………..

8. Omode meloo (ti ko tii to odun mejidinlogun lo n gbe ibi?

134

Fi kun idahun ni ibeere 7 - 8 ki o wo boya ti a ba roo po mo ibeere ati idahun keefa

9. Ipele eko to ga ju wo lo ti pari? (a) ko lo si ile eko [ ] (b) alakobere [ ] (c) ile

eko girama kekere [ ] (d) Ile eko girama agba [ ] (e) Ile eko giga [ ] (f) Ti

omiran ba wa so [ ]

10. Iye odun meloo lo lo nile iwe? …………………………………………………………..

11. Awon ise wo ni owo ti n wole fun o? (o lee dahun ju eyo kan lo) (a) ko sise [ ]

(b) Agbe [ ] (c) Onisowo kekere [ ] (d) eni ti ko kose [ ] (e) eni to kose

mose [ ] (f) onise ijoba [ ] (g) eni to n sise adani [ ]

12. Iru ise woni ile fi n ri owo wole? (o lee ma je / o si le je ise eni to n dahun)

…………………………………

13. Kini iye owo osu to n wole? (a) ko to egberun marun-un naira [ ] (b) laarin

egberun marun si egberun mewaa naira [ ] (c) Egberun mewa si ogun egberun

naira [ ] (d) ogun egberun si ogbon egberun naira [ ] (e) o le ni ogbon egberun

naira [ ]

14. Eelo ni idile yin n na lati fi ra ounje ni oja ni ose kan seyin?

……………………………………………………….

15. N je o wa ninu egbe kan ni awujo yin bii (Ajo, cooperative) eleyii ti won n ba o toju

owo? Beeni [ ] Beeko [ ]

Ipin keji: Wiwa itoju, Iye owo aisan ati ona isanwo

Ipin akoko (a) Wiwa itoju

16. Nje idile yin se aisan laarin osu meta? Beeni [ ] Beeko [ ] (Ti o koba se aisan

lo si ibi keji)

17. Ewo lo je aisan tabi ilera ti koo eye ti idile yin ni laarin osu meta? (a) iba (ara riro) [

] (b) aisan iba jefun jefun (inu didun ati bibi [ ] (c) Igbe (gbuuru) [ ] (d)

aisan nipa mimi (ofinkin ati iko) [ ] (e) Awon miran [ ] si won ……………………..

135

18. Nibo lo ti koko gba itoju (ma ko pupo eyi ti o ba koko dahun nikan ni) (a) Itoju ile [

] (b) Odo atoogun oyinbo [ ] (c) Onisegun ibile [ ] (d) ile iwosan giga [ ]

(g) ile iwosan adani [ ] (h) awon miran [ ] so won ……………………………………..

19. Eelo lo na o lati gba itoju yii (pelu ati ti aworan yiya abi) ?

20. Iru ohun irinna wo lo lo lati fi de ibi ti o ti gba itoju? (a) Oko tire [ ] (b) O wo oko

boosi [ ] (c) takisi [ ] (d) okada [ ] (f) nnkan miran [ ] so won

………………………………………………

21. Wakati meloo lo gba o lati de ibi ti o ti gba itoju? (a) o kere ju iseju meedogun [ ]

(b) iseju meedogun si ogbo iseju [ ] (c) ogbon iseju si iseju marundinlaadota [ ] (d)

Iseju marundinlaadota si wakati kan [ ] (e) o ju wakati kan lo [ ] (f) o ju iseju

meji lo [ ]

22. Eelo lo na lori owo oko fun itoju fun (lilo ati bibo)?

………………………………………………………….

23. Iye owo ti o na? ………………………………..

24. Nje ara re ya leyin too gba itoju tan? (Ti o ba je beeni lo si ipin keji apa keji, ti o ba je

beeko, lo si ibeere kerinlelogun

25. Eelo ni gbogbo owo ti o na lori (oko wiwo, oogun ati ohun miran) titi ti ara re fi ya?

…………………..

Ipin keji abala keta: Sisan ati isanwowole fun oludahun

26. Bawo ni a se sanwo fun itoju akoko o lee dahun ju ikan lo (a) sisan owo lapo [ ] (b)

Olugbani sise [ ] (c) amojutofo ilera [ ] (so iru eyi to je ) Ni sisentenle [ ]

(e) loniruuru [ ] so ni pato

27. Bawo ni o se sanwo itoju naa (a) owo ara re [ ] (b) yawo tabi gba looni [ ] (c) ta

ohun elo ile [ ] (d) Ta ile ebi [ ] (e) Adinku ba owo to san [ ] (f) A sun owo si

136

iwaju [ ] (g) awon awujo da si /elomiran sanwo [ ] (h) A ni ko ma san owo [ ]

(i) Omiran ………… so won [ ]

Fun awon to n lo eto ilera adojutofo beere ibeere ketadinlogoji

28. Ki lo de ti awon are ile se mu kuro eto ilera adojutofo (o lee dahun ju eyo kan lo) (a)

Aabo eto isuna lodi si iye aisan [ ] (b) pipese ona abayo fun ara ile nipa ilera ti ko

gunpa [ ] (c) pipese ilera to dara lati lee din adanu ku nipa ailera [ ] (d) ore ati

ibatan so pe ki n darapo [ ] (e) Awon ohun miran [ ] so wo…………………………

Ipin keta abala kinni: Imo nipa eto ilera adojutofo

29. Nje o ti gbo nipa eto ilera adoju tofo ri? Beeni [ ] beeko [ ] Ti ibeere re ba je beeni

koja lo si ibeere kejilelogoji – ikerinlelogoji ti o ba je beeko koja lo si ipin keta abala

keji

30. Ti o ba je beeni, iru eto ilera adojutofo wo lo mo? (toka si gbogbo eyi ti a daruko (a)

ilera adojutofo gbogbogboo [ ] (b) ilera adojutofo aladani [ ] (c) eyi to je ilera

adojutofo ti abode [ ] (d) Omiran [ ] (so won) (e) o ko mo [ ]

31. Bawo ni o se mo nipa eto ilera adojutofo si? (a) sisanwo asansile fun ilera [ ] (b)

sisan owo ori fun ijoba [ ] (c) Eto ilera ofe lati odo ijoba [ ] (d) Awon miran [ ]

(e) A ko mo [ ] so ni pato

Ipin keta abala keji: Iwoye awon to n lo eto ilera adojutofo alabode lati jeki o rorun funs sisan owo itoju

Ailera maa n sele nigbati a ko roo fun enikookan, a si nilo ilera titoju fun enikookan eyi ti a ko rokan ti o si lowo lori fun enikookan.

Eto ilera Adojutofo yii n pese anfaani eyi ti o maa n din eru eto isuna ku lati sanwo fun opo eniyan, o maa n je ki eto itoju rorun fun enikookan. O lee je dandan tabi fifeese.

Eto ilera adojutofo a lee gbee kale ni alaboode, eyi ti a mo si eto ilera adojutofo eyi to wan i awujo. Kii se eyi to lowo lori tabi ti a n je ere le lori nibi ti awon eniyan yoo ti maa san owo

137 tabi eyi ti o da lori anfaani ti o wa nibe, eyi ti yoo fun ni ona si oniruuru eto ilera ati ojuse.

CBHI maa n pese aabo eto isuna nipa didikun ailowo lowo eyi ti o maa n wa eru wiwo lori awon eniyan nile ati nipataki ona lati ri itoju to peye nigbati awon to ba nilo re.

Eto itoju ilera, adojutofo yii maa n pese anfaani lati lee jeki atunse ba bi won se n seto

32. Nipa oye re, abwo ni waa se gbe eto adojutofo ilera alaboode yii wa fun iwo ati awon

ebi / aarale re? [ ] 1 = kekere 2 = aarin 3 = giga

33. Kinni o ro pe o je amuye ipele sise eto ilera nipa tie to ilera adojutofo abode? [ ]

1=kekere 2 = aarin 3 = giga

34. Bawo ni waa se gbe awon amuye eto ilera adojutofo alabode lati lee jeki ebi re tun dara

sii nipase eto ilera ati ilana? [ ] 1=kekere 2=aarin 3=giga

35. Bawo ni waa se gbe eto ilera yii ga lati lee jeki o darasi nipase pipese eto ilera fun awon

eniyan? [ ] 1=kekere 2=Aarin 3=Giga

36. Bawo ni waa se gbe eto ilera yii lati lee maa pese oogun loorekoore ati ilera ni agbegbe

re? [ ] 1=kekere 2=aarin 3=Giga

37. Bawo ni waa se gbe ipele igboya too ni ti a bag be owo kale lati fi sakoso eto yii ni

awujo? [ ] 1=kekere 2=aarin 3=giga

Ipin keta abala keta: Gbigba nipa eto ilera adojutofo alabode

38. Nje eto ilera adojutofo alabode je itewogba fun o gege bi ona lati fi san owo eto ilera ni

agbegbe yii? [ ] Ti o ba je beeni lo si ibeere kejilelaadota) ti o ba je beeko lo si ibeere

ketalelaadota)

39. Ti ibeere aadota ba je beeni, jowo ko iye ipele gbigba ti o gba eto ilera alaboode yii lati

(1-10) [ ] ibi ti o gba kere ati ibi ti o ti gba ju

40. Ti eto ilera adojutofo alabode ko ba je itewogba, kilo de? (ka idahun ki o si yi odo si eyi

kan) (1) Mi o jeri awon ti o maa nawo (2) Mo maa n wa ni ilera pipe (3) Eru lati

138

kopa ninu eto naa n ba mi (4) Mi o fe iru ise yii (5) Mi o mo (7) Awon idi miran

(so won)

Ipin kerin: Fife lati sanwofun eto ilera Adojutofo alabode

Ni kukuru maa so nipa eto yii fun iwo ati awon ebi re leyin eyi waa beere ibeere

Awon ijoba ti so ona titun miran ti won lee fi ran awon olugbe Ilu Eko lati san owo fun eto ilera ni adugbo re. Awon ara ile re lee ra nipa eto ilera adojutofo alabode fun ebi re Olukopa ninu eto yii gbodo je atinuwa waa si san owo ti ko gunpa eyi ti yoo da lori anfaani to ba wa, waa ni anfaani si oniruuru eto ilera ti won ba pese.

Waa reti lati san owo iwole lati san owo iwole lati lee dara po mo eto yii.

Owo iwole yii, o lee san losoosu tabi eekan lodun, ko si aaye fun awin.

Leyin owo sisan, iwo ati ara ileere yoo maa gbo itoju lofee nigba ti o ba fe. Gbogbo eni to ba foruko sile ni won yoo fun ni iwe idanimo eyi ti won yoo maa tunse ni odoodun leyin owo odun ti won ba san.

Eto ilera yii kun fun awon ilera ponbele eyi to se pe awon oogun to see lo fun ayewo lorisiirisii, eto: itoju leyin ibimo, awon abere ajesara eto feto somo bibi, eto itoju ninu oyun ati leyin ibimo, Riri dokita to muna doko, ayewo oju ati ipese awo oju idiwo nipa eyin ati oogun irora.

Ilera ni a o maa pese ni agbegbe ti a bat i foruko sile. Awon ilera yii ni a o maar i ni awon ile iwosan awon alabode awon ti ile igbebi ati ile iwosan adani.

Awon igbimo to da sile yii ni ijoba yoo maa ti won leyin.

Ni ipari odun o lee se atunse awon eso to jeyo ninu eto yii.

O lee mu eni titun niti eto ilera ti o ba fee lo iru nnkan bee fun osu mejila.

41. N je waa nife si ki elomiran foruko sile ninu ebi re fun eto yii ? Beeni [ ] beeko [ ]

42. Ti nomba ketalelaadota/ikerinlelaadota kinni idi Pataki too ni

139

(a) Se o nilo eto adojutofo [ ] (b) ara mi ya / mi kii sebe saisan [ ] (c) ko si igboya ninu eto yii [ ] (d) mi o rope mo lee see [ ] (e) mo ni eto adojutofo temi [ ] (f) ko si eto ilera to peye [ ] so ni pato

A o mo iye owo ti o nife lati san fun ara re ati awon miran ninu ebi re.

42. Ife lati san owo fun enikan

(a) iye owo osu eto adojutofo (sisan) je eedegbeta Naira: Nje o fe san-an) Beeni [ ] beeko [

] lo si ibeere karundinlogota ma lo si ipin keji

(b) Eelo lo fee san ti o nifee si (Ti o ba ju irinwo naira lo lo si ibeere karundinlogota bi o ba ti lo sile ju bee lo si ibeere karundinlogota

(42c)

(d) Eelo ni oodunrun le ni aadota? Se waa lee san?

(e) Eelo ni iye owo ti o nife si lati san fun eto ilera yii?

(f) iye owo ti o so yii ti lo sile ju ko si lee bori iye owo ti a so. O si ye ki o jeki owo yii po sit i o ba nife lati darapo mo eti yii

(g) Ti o ba sepe tori anidaniloju, owo odun eto yii lee lo siwaju Eelo ni iye owo to da o loju lati san, ni lokan idaji iye owo osu idile ati iye owo ti o n na lori awon nnkan miran?

43. Fun awon ara ile miran

(a) Iye owo osu eto adojutofo ni odun je eedegberun; N je o setan lati san iye owo yii fun ebi re? [ ] 1=Beeni 0=Rara, 2=Emi ko mo

(b) Eelo ni iye owo lapapo ti o nife lati san ……………………………….. (Ti o ba ju bee lo

/ to Egberin Naira lo si ibeere (g), ti o ba kere sii lo si Egberin Naira lo si ibeere abala’ d)

(d) Ti iye owo ba je Eedegberin Naira le ni aadota se waa nife lati san? [ ] 1=beeni

0=Rara

140

(e) Eelo ni iye owo ti o nife lati san fun eto yii lodun?

(f) Iye owo ti o so ti kere ju, ko si ni lee bori re lodun, o ye ki o tun fikun ti o ba fe ki awon ebi re foruko sile ninu eto yii. Ki wa ni iye owo ti o fee san ni osu pelu eni kookan ?

(g) Ti a ba n so nipa koba o le oro aje tabi aidaniloju eto yii ti owo yii ba lo sokelodun. Eelo ni iye owo to da o loju lati san lodun, ni lokan iye owo osu to n wole fun o losu ati iye owo ti o n san …………

44.Nje a ri awon ohun ti o fee so nipa eto ilera alabode adojutofo yii?......

A dupe fun akoko ti o lo.

141