**1. Introduction**

#### **1.1 Background**

Poverty is a pervasive problem in Africa and especially in Nigeria (World Bank, 2008). About 50.3% of the population of Sub-saharan Africa is reported to be living below the International Poverty Line of US\$1.25 (UN, 2008). In Nigeria, about 55% of the population is living below the poverty line (World Bank, 2008). There is a geographical and sectoral dimension to the poverty situation in Nigeria. Poverty in Nigeria is more intense in the rural areas than the urban areas (Aigbokhan, 2000; Aigbokhan, 2008). Majority of Nigerians living in the rural areas are engaged either directly or indirectly in agriculture (NBS, 2006) and these are the people who are mostly trapped in poverty.

To develop appropriate policies to address poverty, there is a need for proper measurement of poverty. The use of money metric measures in indicating the level of poverty is gradually yielding place to other indicators of welfare which include deprivations in health, educational attainment, enjoyment of citizenship rights, social participation, life expectancy at birth and; maternal and child mortalities, among others (Okunmadewa, 1999; Srinivasan, 2001; Anderson, 2010). Among these indicators, health status and access to health facilities are keys to lifting people out of poverty or preventing them from falling into it (Republic of Sierra Leone, 2008). This is probably the reason while these health-related indicators are weighted heavily in the computation of the Human Development Index which is used for ranking countries in respect of welfare status (Herero *et al.* 2010).

Inadequate access to health services is one of the components of rural poverty which is prevalent in Nigeria (NBS, 2006). Inadequate access to health services determines, to a large extent, the decision of rural households to either patronize orthodox medicine (OM) or traditional medicine (TM) (Mafimisebi & Oguntade, 2010).

#### **1.2 Justification for and focus of the study**

Inadequate access to health services is a major issue confronting the poor in Nigeria. The Nigeria Core welfare Indicator study (NBS, 2006) revealed that 55.1% of Nigerians have access to OM health facilities while 7.5% consulted traditional healers in the four weeks preceding the survey. Obviously, Nigerians use both OM and TM for the maintenance of

Health Infrastructure Inequality and Rural-Urban Utilization of

urban dichotomy.

perfect equality

N is the number of LGAs.

Where

Where,

2010).

indicators.

<sup>2</sup> *x* = Household size

Orthodox and Traditional Medicines in Farming Households: A Case Study of Ekiti State, Nigeria 199

of persons and the land area per OM infrastructure, were analyzed focusing on the rural-

Gini Coefficient measures the degree of concentration (inequality) of a variable in a distribution of its elements. It compares the Lorenz Curve of a ranked empirical distribution with the line of perfect equality. The Gini Coefficient ranges from 0, where there is no concentration (perfect equality), to 1 where there is total concentration (perfect inequality). The ID is the summation of vertical deviations between the Lorenz Curve and the line of perfect equality. The closer the ID is to 1, the more dissimilar the distribution is to the line of

The extent of inequality in the distribution of the health infrastructure was explored with

*G Y YX X*

1 1

*i ii i*

*i i*

1 ( )( )

1 0.5 *N*

*i ID X Y* 

N is the number of LGAs (Castillo-Salgado et.al., 2001; Dixon et.al., 1987; Rodrigue et.al.,

For the primary data, qualitative description was used in presenting the result of the FGD. Descriptive statistics, which include frequencies and percentage, were used to describe the primary data on socio-economic and demographic characteristics of the respondents. The logistic regression was adopted in analyzing the influence of postulated independent variables on the probability of use of TM separately in the urban and in the rural locations. In using the logistic regression, we developed a dichotomous variable indicating whether the household uses TM more often than OM. This dichotomous variable is in this study called household's use of TM (HUTM). HUTM is 1 if a household uses TM more often and zero otherwise. The predictor variables are a set of socio-economic and demographic status

> 0 11 22 77 ( ) ln( ) <sup>1</sup> *<sup>p</sup> Logit p xx x*

the Gini Coefficient and the ID. The Gini Coefficient is calculated as:

0

*i*

σX is cumulative proportions of the populations or land areas of the LGAs;

X is the cumulative proportion of the populations or land areas of the LGAs,

The estimating equation of the binary logit model is specified as follows:

*p*

Y is the cumulative proportion of the number of OM infrastructure in the LGAs; and,

σY is the number of OM infrastructure in the LGAs; and

The Index of Dissimilarity is calculated as:

*p* = probability that the household uses TM <sup>1</sup> *x* = Age of household head (in years)

*N*

their health. In deciding which of these to use, access, in terms of availability and affordability, plays a significant role (Mafimisebi & Oguntade, 2010). Public policy affects both availability and affordability of OM services whereas for TM, availability and affordability are affected by the location of the prospective users (Mafimisebi & Oguntade, 2010). To this extent, the distribution of OM facilities requires public policy attention to ensure equitable access in terms of availability and affordability such that the decision to use either OM or TM will depend on users' preference. Given that affordability is a more critical factor in the rural and agriculture dependent areas because of higher level of poverty, public policy attention needs to be focused on access to OM services in the rural areas (Mafimisebi & Oguntade, 2010).

This study assesses the distribution of OM health infrastructure in Ekiti State, Nigeria, focusing on the rural-urban dichotomy that is prevalent in the establishment of OM health infrastructure in most states of Nigeria (NBS, 2007). It further looks at the use of OM and TM among farming households with special emphasis on the rural-urban dichotomy.

#### **1.3 Approach to the study**

This study was carried out in Ekiti State, Nigeria. It is one of the six states in South-west Nigeria and it has 16 Local Government Areas (LGAs). It is located between longitude 4o 45' to 5o 45' East of the Greenwich Meridian and latitudes 7o 15' – 8o 5' North of the Equator. Based on 2006 census, the state has a total population of 2,384,212 (National Bureau of Statistics (NBS), 2010). Ekiti State is largely agrarian (NBS, 2006) and hence it is typical of most states in Nigeria. The state was selected for this study because it is one of the states in the catchment area of the Federal University of Technology, Akure, the institutional base of the authors of this paper.

In this study, secondary data were used to assess the distribution of OM infrastructure. These data, which comprise the names and addresses, Local Government Area (LGA), ownership status and legal status of all orthodox health institutions in Ekiti State, were collected from the State Ministry of Health. The data were compared with similar data that were accessed from the NBS (NBS, 2007). In addition, the population figures by LGAs were also accessed from NBS (NBS, 2010) while the land areas of the LGAs were collected from the State Surveyor-General's office. For the assessment of rural-urban utilization of OM and TM, primary data were collected from farming households in two LGAs of Ekiti State, one of which is urban and the other rural. Two sets of primary data were collected; first, through the use of structured and pre-tested questionnaire administered on household heads and second, through focus group discussions (FGD) guided with a checklist of desired information. For the administration of the structured questionnaire, the multi-stage sampling method was used in selecting the respondents. In the first stage, Ado, an urban LGA , and Irepodun/Ifelodun, a rural LGA, were purposively selected. In the second stage, three communities in each LGA were randomly selected from the list of farming communities while in the third stage; twenty (20) households were systematically selected from the list of farming households in each community. This yielded a total of sixty (60) households each in the urban and rural LGAs. For the FGD, 206 other farmers participated. These FGD participants were not privileged to provide responses to the questionnaire and were not necessarily household heads.

The secondary data were analyzed through the use of Gini Coefficient and Index of Dissimilarity (ID) with a view to assessing the level of inequality in the distribution of health infrastructure in Ekiti State. To further assess the source of the inequality, both the number of persons and the land area per OM infrastructure, were analyzed focusing on the ruralurban dichotomy.

Gini Coefficient measures the degree of concentration (inequality) of a variable in a distribution of its elements. It compares the Lorenz Curve of a ranked empirical distribution with the line of perfect equality. The Gini Coefficient ranges from 0, where there is no concentration (perfect equality), to 1 where there is total concentration (perfect inequality). The ID is the summation of vertical deviations between the Lorenz Curve and the line of perfect equality. The closer the ID is to 1, the more dissimilar the distribution is to the line of perfect equality

The extent of inequality in the distribution of the health infrastructure was explored with the Gini Coefficient and the ID. The Gini Coefficient is calculated as:

$$G = 1 - \sum\_{i=0}^{N} (\sigma Y\_{i-1} + \sigma Y\_i)(\sigma X\_{i-1} - \sigma X\_i)^2$$

Where

198 Health Management – Different Approaches and Solutions

their health. In deciding which of these to use, access, in terms of availability and affordability, plays a significant role (Mafimisebi & Oguntade, 2010). Public policy affects both availability and affordability of OM services whereas for TM, availability and affordability are affected by the location of the prospective users (Mafimisebi & Oguntade, 2010). To this extent, the distribution of OM facilities requires public policy attention to ensure equitable access in terms of availability and affordability such that the decision to use either OM or TM will depend on users' preference. Given that affordability is a more critical factor in the rural and agriculture dependent areas because of higher level of poverty, public policy attention needs to be focused on access to OM services in the rural areas (Mafimisebi

This study assesses the distribution of OM health infrastructure in Ekiti State, Nigeria, focusing on the rural-urban dichotomy that is prevalent in the establishment of OM health infrastructure in most states of Nigeria (NBS, 2007). It further looks at the use of OM and TM among farming households with special emphasis on the rural-urban dichotomy.

This study was carried out in Ekiti State, Nigeria. It is one of the six states in South-west Nigeria and it has 16 Local Government Areas (LGAs). It is located between longitude 4o 45' to 5o 45' East of the Greenwich Meridian and latitudes 7o 15' – 8o 5' North of the Equator. Based on 2006 census, the state has a total population of 2,384,212 (National Bureau of Statistics (NBS), 2010). Ekiti State is largely agrarian (NBS, 2006) and hence it is typical of most states in Nigeria. The state was selected for this study because it is one of the states in the catchment area of the Federal University of Technology, Akure, the institutional base of

In this study, secondary data were used to assess the distribution of OM infrastructure. These data, which comprise the names and addresses, Local Government Area (LGA), ownership status and legal status of all orthodox health institutions in Ekiti State, were collected from the State Ministry of Health. The data were compared with similar data that were accessed from the NBS (NBS, 2007). In addition, the population figures by LGAs were also accessed from NBS (NBS, 2010) while the land areas of the LGAs were collected from the State Surveyor-General's office. For the assessment of rural-urban utilization of OM and TM, primary data were collected from farming households in two LGAs of Ekiti State, one of which is urban and the other rural. Two sets of primary data were collected; first, through the use of structured and pre-tested questionnaire administered on household heads and second, through focus group discussions (FGD) guided with a checklist of desired information. For the administration of the structured questionnaire, the multi-stage sampling method was used in selecting the respondents. In the first stage, Ado, an urban LGA , and Irepodun/Ifelodun, a rural LGA, were purposively selected. In the second stage, three communities in each LGA were randomly selected from the list of farming communities while in the third stage; twenty (20) households were systematically selected from the list of farming households in each community. This yielded a total of sixty (60) households each in the urban and rural LGAs. For the FGD, 206 other farmers participated. These FGD participants were not privileged to provide responses to the questionnaire and

The secondary data were analyzed through the use of Gini Coefficient and Index of Dissimilarity (ID) with a view to assessing the level of inequality in the distribution of health infrastructure in Ekiti State. To further assess the source of the inequality, both the number

& Oguntade, 2010).

**1.3 Approach to the study** 

the authors of this paper.

were not necessarily household heads.

σX is cumulative proportions of the populations or land areas of the LGAs; σY is the number of OM infrastructure in the LGAs; and

N is the number of LGAs.

The Index of Dissimilarity is calculated as:

$$ID = 0.5 \sum\_{i=1}^{N} |X\_i - Y\_i|$$

Where,

X is the cumulative proportion of the populations or land areas of the LGAs,

Y is the cumulative proportion of the number of OM infrastructure in the LGAs; and,

N is the number of LGAs (Castillo-Salgado et.al., 2001; Dixon et.al., 1987; Rodrigue et.al., 2010).

For the primary data, qualitative description was used in presenting the result of the FGD. Descriptive statistics, which include frequencies and percentage, were used to describe the primary data on socio-economic and demographic characteristics of the respondents. The logistic regression was adopted in analyzing the influence of postulated independent variables on the probability of use of TM separately in the urban and in the rural locations. In using the logistic regression, we developed a dichotomous variable indicating whether the household uses TM more often than OM. This dichotomous variable is in this study called household's use of TM (HUTM). HUTM is 1 if a household uses TM more often and zero otherwise. The predictor variables are a set of socio-economic and demographic status indicators.

The estimating equation of the binary logit model is specified as follows:

$$Logit(p) = \ln(\frac{p}{1-p}) = \pounds\_0 + \pounds\_1\mathbf{x}\_1 + \pounds\_2\mathbf{x}\_2 + \dots + \pounds\_T\mathbf{x}\_T$$

*p* = probability that the household uses TM

<sup>1</sup> *x* = Age of household head (in years)

<sup>2</sup> *x* = Household size

Health Infrastructure Inequality and Rural-Urban Utilization of

distributions), equity is *fairness* of distributions

health and life expectancy.

*et.al*. 2010).

Orthodox and Traditional Medicines in Farming Households: A Case Study of Ekiti State, Nigeria 201

can be defined as differences in health status or in the distribution of health determinants between different population groups (WHO, 2009). They are the result of 'a complex system operating at global, national and local levels which shapes the way society, at national and local level, organizes its affairs and embodies different forms of social position and hierarchy. The place people occupy on the social hierarchy affects their level of exposure to healthdamaging factors, their vulnerability to ill health and the consequences of ill health (Marmot, 2009: 14). Health inequality refers to differences or variations in health-related quality of life

The causes of urban health inequalities are associated primarily with socio-economic status, income, poverty, deprivation levels, unemployment, incapacity, worklessness, skills and educational level, housing conditions and social mobility as well as life chances (O'Brien

Inequality in health is not the same as inequity in health. Inequalities in health status between individuals and populations are inevitable consequences of genetic differences, of different social and economic conditions, or a result of personal lifestyles. Inequities occur as a consequence of differences in opportunity which result, for example in unequal access to health services, nutritious food, adequate housing and so on. In such cases, inequalities in health status arise as a consequence of inequities in opportunities in life (WHO, 1998). It should however be noted that public policy-induced inequality in HI and other socioeconomic conditions will contribute to inequities in opportunities. According to Whitehead (1992), health inequities are 'differences in health which are not only unnecessary and avoidable but, in addition, are considered unfair and unjust'. This means that not all inequalities can be described as inequities. Whereas equality means sameness (equality of

Health status affects economic growth and sustainable development. There is evidence that investing in health brings substantial benefits to the economy (Anyanwu & Erhijakpor, 2007). According to WHO (2001), increasing life expectancy at birth by 10% will increase the economic growth rate by 0.35% a year. On the other hand, ill health is a heavy financial burden. About 50% of the growth differential between rich and poor countries is due to ill-

Harttgen & Misselhorn (2006) found that access to health infrastructure is important for child mortality which is one of the health outcomes covered by the MDGs. On the other hand, socio-economic factors, especially poverty, are often found to be strong determinants of health outcomes (Nolte & Mckee, 2004; Young, 2001; Leger, 2001). In most developing countries, health attainment indicators for the poor tend to be worse than the national average (Tandon, 2007). Also, the extent to which such health inequalities exist varies significantly across countries. Empirical evidence suggests that health inequalities have been persistent over time and, in many cases, have been growing (ADB, 2006). The rich can ignore government finance and health facilities; and access private sector health facilities on their own while the poor are more dependent on the public sector OM infrastructure and governments often do not have enough resources to expend on pro-poor health programmes and interventions (Tandon, 2007). Sachs (2004) has hence been calling for a scaling up of government health programmes in order to attain health-related MDGs.

The MDGs had three out of eight goals directed at promoting health. These are reduction in child mortality, improvement in maternal health and combating HIV/AIDs, malaria and other diseases (UNDP, 2003). The first goal, which is the eradication of extreme poverty and

**2.2 Health Infrastructure Inequality and Health Policy in Nigeria** 

and length of life profiles of different population groups in a nation (WHO, 2009).


The equation is estimated by the maximum likelihood method because the procedure does not require the assumptions of normality or homoscedasticity of errors in predictor variable. The model was fitted separately for rural and urban households.
