Do Tunisian Young Children Have Equal Chances in Access to Basic Services? A Special Focus on Opportunities in Healthcare and Nutrition

Abstract

It is widely recognized in the public health literature that healthcare and nutrition during the early childhood period have important long-term and sometimes irreversible consequences on health and wellbeing later in life. In this paper, we endeavor to investigate deeply the patterns of inequality of opportunity in health and nutrition outcomes among under-five children in Tunisia. Several tools are used to do such analysis including comparison of distributions of the considered outcomes across circumstances groups; estimation of the Human Opportunity Index; and measurement of the relative contributions of different circumstances using the Shapley decomposition. The main findings reveal reasonable and low levels of inequality in access to all basic healthcare services and nutrition except access to improved water and sanitation. Besides, parents’ education, wealth status and location of residence are found to be key factors causing low level of inequalities. Thus, it is easy to conclude that without more inclusive and pro-poor policy interventions, there are few chances for children belonging in poor families and living in marginalized rural areas to spring out of the poverty lived by their parents.

1 Introduction

It is by now well recognized in the literature that risks to cognitive and physical development are heterogeneously distributed over the life-cycle and are typically higher in earlier stages of life (infancy and early childhood), with substantial long-term and sometimes irreversible effects on later stages of life (youth and adulthood). Risks are awfully high during the ages of 0–5 years, and may have a persistent and negative effects on education, health and labor market outcomes, and thus income-earning potential as adults (World Bank 2006). Several studies in public health reveal that health and nutrition are crucial at the early age to health and wellbeing later in life. Indeed, sickness and malnutrition during this early period of life may hinder a child’s subsequent cognitive and physical development, causing harmful health, productivity and wellbeing outcomes that can endure into adulthood. Any developmental shortages that happen in early childhood may be persistent and permanent (UNESCO 2006).

In developing countries, which Tunisia belongs to, access to basic housing, health and nutrition outcomes are unfairly distributed among children, depending on parental and household characteristics and public health inputs, such as the availability of clean water and sanitation facilities. Such inequality of opportunity can be considered as a main contributor to the observed inequality of opportunity later in life (de Barros et al. 2009). In Tunisia, according to the official statistics of the Tunisian National Statistics Institute (NSI), children aged less than 14 years represent, in 2014, more than 23.7% of the total population; 37.24% of them are aged between 0 and 4 years. It is well known that this part of the population is the most vulnerable in health care an in social terms. Notwithstanding the gap in access to basic services has been narrowed between rural and urban areas during the last decades, an obvious uneven distribution of basic childhood development outcomes still exist among various regions in Tunisia. Giving such regional disparities, more equitable access to main housing and health services in the early childhood is so needed to facilitate human capital accumulation, which in turn could lead to higher economic growth and poverty alleviation in the future (see Galor and Zeira 1993 for theoretical evidence; Birdsall and Londoño 1997 for empirical evidence).

Despite the fact that acceptable level of economic inequality in a society is controversial, policies that afford equal opportunity to all children, not considering of their socioeconomic background, are embraced across the political spectrum (Ersado and Aran 2014). Early childhood development policies are therefore commonly recognized as some of the few policy areas where the traditional equity-efficiency trade-off does not exist as stated by Heckman and Masterov (2007). It is thus imperative to comprehend how children’s opportunities develop and determine the appropriate policy interventions that contribute to alleviating the impact of predetermined factors. Nationally, further consideration should be accorded to inequality of opportunity in basic services for which child should not be held responsible. Few studies has focused on such inequality among children in Tunisia (Jemmali and Amara (2015a, 2015b) and Jemmali (2016)).

Giving the growing importance of equality of opportunity in public policy over the last decade in Tunisia, we attempt in the current study to focus on the patterns of inequality of opportunity in child health and nutrition in this country. The data used in the analysis come from the last wave of the Multiple Indicator Cluster Surveys¹ on children and women MICS4 conducted by the NSI and the UNICEF in 2011/2012. The aim of the paper is twofold: Firstly we attempt to analyze the extent of inequality of opportunity in health and nutrition among Tunisian children aged less than 5 years employing the Human opportunity index (HOI) methodology. Secondly, we endeavor, using the Shapley decomposition method, to estimate the contribution of different circumstances, such as gender, parental education, wealth status of parents and place of residence. Such circumstances, as explained above, are beyond the control of children and affecting their development outcomes. The main questions to answer are: How far Tunisia is from the objective of providing fair and universal access to a set of critical health and nutrition outcomes to all early children regardless of the aforementioned circumstances? And what circumstances play the main role in the distribution of the considered outcomes? To reply to these questions, we use concepts and ideas developed initially by Roemer (1998) and methodology developed in the recent and growing literature on assessment of inequality of opportunity (see for instance de Barros et al. 2009).

The remainder of the paper is structured as follows: The following section presents a brief literature review of the main studies focusing on the inequality of opportunity in health and nutrition outcomes. Section 3 presents the data and empirical methodology used to the measurement and the decomposition of inequality of opportunities among Tunisian children. Section 4 presents the main results and discussions, while Section 5 concludes the paper with some policy recommendations.

2 Literature Review

The literature on inequality of opportunity is drawn on to highlight the link between the differences in outcomes and differences due to predetermined circumstances over which an individual has no control. Any improvement in the coverage of a basic service may raises equality of opportunity and any improvement in coverage for the lagging groups of children may carries with it further decrease in inequality of opportunities (de Barros et al. 2009). Such equality of opportunity begin typically with the first few years of a child’s life. This critical period of life, in which development in social, cognitive, emotional and other developmental areas are in rapid change and progress, has been described as “an extended critical period, a window of opportunity for development, closed by age three” (UNESCO 2006, p. 109). Moreover, during this early period, children are particularly sensitive to the conditions in which they live without having a control over them. For instance, poverty, lack of healthcare providers, malnutrition are issues that harm its development is quite fragile in the face of issues like poor nutrition. For instance, being born to a poor family that couldn’t afford necessary nutrition and healthcare or living in an underserved geographic location, are entirely beyond the child’s control, but harms its development and determines its opportunities to accumulate crucial health assets.

Several studies have attempted to examine the patterns of inequality of opportunity on early childhood development in a number of countries. Drawing on the household income and expenditure survey data from South Africa, Zere and McIntyre (2003) analyzed the correlation between socioeconomic status and malnutrition among children aged less than 5 years. As expected, they found that stunting and wasting were most highly concentrated in the poorest and unprivileged regions of the country. The study found, then, that children living in poor family and poorest regions are more exposed to such diseases. One other finding of the study is the racial inequality: among the white population, no significant inequities were found, while nonwhite children from metropolitan regions showed the highest levels of stunting in the country. In the same line, Burgard (2002) assessed inequality of opportunities taking into account the racial inequalities in child stunting in Brazil and South Africa. The author found that racial differences and household socioeconomic status are strongly linked with stunting. Monteiro et al. (2010) focused, in a similar study undertaken in Brazil, on the relationship between the prevalence of child malnutrition in relation and income and basic services redistribution policies. The main finding of the study is that over the 33 years examined, the gap in terms of stunting between children from poor and rich households had shrinked considerably along with decline of income inequality, a rise in purchasing power, and an increase in access to healthcare and other basic services in unprivileged regions.

Using different inequality of opportunity indices, Singh (2011) measured inequalities in malnutrition and immunization for children in India, and finds significant regional disparities. In order to evaluate the trends for malnutrition among the early aged children, Pathak and Singh (2011) have used in another study, bivariate analysis, poor-rich ratio and concentration indices. Mohanty and Pathak (2009) used the same methodological approach to assess inequalities in access to maternal care services and child immunization. The common finding of these two studies, dealing with the Indian context, is the significant disparities in health and nutrition outcomes between the poor and rich. Limwattananon et al. (2010) in Thailand and Axelson et al. (2012) in Vietnam used the same concentration indices to investigate disparities in maternal health and early child health outcomes. Both studies found significant inequalities between poor and non-poor households. Employing the multivariate logistic regression, Anwar et al. (2008) examined inequities in the use of maternal healthcare services in Bangladesh. The authors found significant inequalities due to asset ownership, area of residence, and parental education.

There have been a few studies focusing on inequality of opportunity in health and nutrition in Arab countries. The majority of these studies are dealing with the extent of inequality in health in Egypt. Compared to 54 countries, Egypt was found as one of the least inequitable countries in terms of skilled birth attendant and measles immunization (Barros et al. 2012); it ranks 50th in the considered sample. Wagstaff et al. (2003), drawing on the 1995/6 Egypt DHS, stated that Egypt has high concentration indices (i.e., high inequality) in under-five mortality rate and infant mortality rate compared to other countries. Another study on Egypt by Boutayeb and Helmert (2011) examined inequities in maternal care between rich and poor women, as well as inequities between women living in urban and rural regions. Recently, Ersado and Aran (2014), in their study on inequality of opportunity in Egypt during the 2000s, found that Egypt has made considerable progress in the availability of and access to different basic services (water and sanitation, Education, Health, Nutrition) for children and mothers, in some cases with an overall pro-poor effect. The authors have found as well an appreciable improvements, translated by a decline in inequality of opportunity over the past decade in healthcare utilization before and during pregnancy and immunizations. The study reveals also that family background, especially parents’ education and wealth status, and geographic factors are the key factors affecting child development outcomes in Egypt.

In a recent study, Assaad et al. (2012) analyzed the extend of inequality of opportunity in child health outcomes in Egypt and some Arab countries and Turkey using a number of Demographic and Health Survey (DHS) data. They used indicators for stunting and wasting standardized by height and weight of children as outcome variables, instead of the standard Z-scores, computed by comparing the observed anthropometric measures to reference distributions of height and weight for healthy children of the same age and sex. They found mainly that total inequality in Egypt is increasing over time and geography is the main prevalent circumstance influencing height and weight of children, followed by demographic and educational parents’ characteristics.

Notwithstanding a number of studies have been undertaken to examine inequality of opportunities in Tunisia such as Jemmali and Amara (2015a, 2015b) and Jemmali (2016), no study has focused on health and nutrition issues. The common finding of the three previous studies is the large and significant disparities, particularly in access to sanitation facilities and secondary education between the East (Littoral) and Western (Inland) areas. Area of Location, parent’s socioeconomic and educational characteristics are found as the most important circumstances contributing to such regional disparities. To estimate the contribution of different circumstances in the inequality of opportunity, authors in the first two studies used the variance decomposition analysis employed previously by Son (2013), while in the third study, Jemmali (2016) used the Shapley decomposition method similarly to Ersado and Aran (2014) and Belhaj Hassine and Zeufack (2015).

The current study differs from the aforementioned studies in terms of outcomes variables. It shed more light on inequality of opportunity in health and nutrition using a number of outcomes and a set of circumstances that are presented in more details in the following section. To the best of our knowledge, this is the first paper providing a detailed analysis of inequality of opportunity in health and nutrition in the Tunisian context.

3 Data and Methodology

3.1 Data

We use data from the fourth round of the MICS4 conducted in Tunisia in 2011–2012 by the Ministry of development and international cooperation in collaboration with the NSI. Financial and technical support was provided by the UNICEF, the United Nations Population Fund (UNFPA) and the Swiss Cooperation Office in Tunisia. The Multiple Indicator Survey is a household survey program developed by the UNICEF in the mid 1990s in order to assist countries in filling data gaps for children and women. Using key indicators (such as children nutritional status, women fertility history, water and sanitation, HIV and AIDS, characteristics of household, and so on), the MICS survey enables countries, more specifically the middle income ones, to achieve the Millennium Development Goals (MDGs), the goals of a World Fit for Children (WFFC) and other nationally and internationally agreed commitments.²

As with MICS2 in 2000–2001 with 29,645 households and MICS3 in 2005–2006 with 9600 households, the current round of MICS (MICS4 Tunisia survey) highlights significant progress made over the past five years, particularly in terms of health, education and child protection, as well as in women’s health. The fourth round of Tunisian’s MICS were collected from a sample size of 9600 households, representative at the national level, for both urban and rural areas and at the level of nine regions of the country (Greater Tunis, North East, North West, Middle East, Governorate of Kairouan, Governorate of Sidi Bouzid, Governorate of Kasserine, South East and South West).

A stratified two-stage random sampling approach was used for the selection of the 9600 households. At the first stage of selection, 480 clusters (census enumeration areas) are randomly selected (with probability proportional to size (PPS)) among all clusters of the General Census of Population 2004. At the second stage of selection, 20 households are selected within each of the 480 clusters, to get a total sample size of 9600 households. Among the 9600 households selected for the sample, 9329 were identified at the time of the survey, and a total number of 9171 were successfully interviewed, resulting in a response rate of 98%. From the surveyed households, 10,514 women aged 15–49 years were identified and 10,215 of these were successfully interviewed, yielding a response rate of 97% within interviewed households (Table 1).

Table 1 Number of clusters, number of households and number of women aged 15–49 years by region

3.1.1 Outcome Variables

A total of eleven indicators have been selected from the MISC4s surveys and grouped in two outcome categories: (i) access to basic healthcare during pregnancy and birth and early postnatal periods and (ii) nutrition that includes indicators on malnutrition and micronutrient intake. Three composite indexes are developed afterward using the set of these indicators. Under the first set of outcomes, the subsequent variables, quite similar to variables used by Ersado and Aran (2014), are involved in the analysis: (i) lack of antenatal care (approximated by the occurrence of mother not having any blood tests during pregnancy³); (ii) birth not taking place at health facilities; (iii) birth not being attended by skilled health professional; (iv) child not having a postnatal check-up (within two months after birth); (v) health examination and regular immunizations within one year after birth; (vi) access to safe water and (vii) access to toilet (a health shield).⁴ The second category of outcomes is associated to levels and trends in malnutrition and micronutrient intake and access to clean water (a nutritional must) in the early years. To investigate the disparities in malnutrition levels across various circumstance groups, common anthropometric measures (Z scores⁵) such as (i) stunting⁶ (height-for-age), (ii) wasting⁷ (weight-for-height), and (iii) underweight⁸ (weight-for-age) are used to estimate the nutritional status of child (see Assaad et al. (2012) and O’Donnell and Wagstaff (2008) for more details about these measures). There is significant evidence that malnourishment, particularly in childhood, raises the risk of death, inhibits cognitive development, and can result in a higher disease risk later in life (O’Donnell et al. 2009). It’s also notable in this regard that children have the necessary level of micronutrient intake for healthy development. Access to iodized salt, iron tables during pregnancy and Vitamin A in early infancy are broadly used indicators for micronutrient intake. Giving that such indicators are not available in the MICS4 survey, we use in the current paper one variable that is important for determining the iron deficiency anemia during pregnancy: (v) whether the mother has had blood tests during her pregnancy.

The two Tables 2 and 3 illustrate respectively the basic descriptive statistics of health and nutrition outcomes across the country and the nine regions (Grand Tunis, North East, North West, Centre East, Kasserine, Kairouan, Sidi Bouzid, South East and South West). While the first columns of the first Table 2 show high and equitable access to health outcomes, a remarkable disparity in access to water and sanitation facilities is observed in the last two columns of the tables ranging respectively between 36.48% and 18.03% in Sidi Bouzid, the most lagging region in the country and 69.77% and 84.88% in the metropolitan region. In the same line, the Table 3 reveals that no significant disparity in nutritional status is shown across the different regions. Behind this pretended equitable distribution of both health and nutrition outcomes, access to these basic services may differ more significantly across various population sub-groups.

Table 2 Descriptive statistics of health outcome variables (by region) (2011/12)

Table 3 Descriptive statistics of nutrition outcome variables (by region) (2011/12)

3.1.2 Circumstantial Variables

On the other side, nine circumstance variables determining early child’s opportunity access to health and nutrition outcomes are involved in the analysis. Taking into account these circumstances, given below, population of children surveyed is divided in several subgroups. From these various subgroups, we focus later in the analysis on two specific and extreme groups, least and most advantaged group, in order to highlight the role of such circumstances in determining access of child to the considered outcomes. The nine circumstance variables used in the current application are:

• Region: Grand Tunis (metropolitan region), North East, North West, Centre East, Kasserine, Kairouan, Sidi Bouzid, South East and South West, (9 categories).

• Location: Urban and rural areas, (2 categories).

• Number of children aged under 5 years, (Discrete variable).

• Household composition: number of members in the household, (Discrete variable)

• Age of household head: in year, (Continuous variable).

• Gender of Household head: 0 if female and 1 if male, (2 categories).

• Father’s Education: None formal education, Primary/Lower secondary, Secondary, University (6 categories).

• Mother’s Education: None formal education, Primary/Lower secondary, Secondary, University, (6 categories).

• Economic wellbeing of household: Quintile of economic wellbeing (5 categories).

• Gender of the child: 0 if female and 1 if male, (2 categories).

These circumstances could be grouped into four dimensions: (i) child characteristics (e.g., gender of the child); (ii) demographic characteristics of the household (e.g., number of siblings aged under 5 years and household composition); (iii) geographical location of the household (e.g., Urban/rural status and region of residence); (iv) socio-economic characteristics of the father and mother (e.g., educational level of both father and mother, age and gender of the household head and household’s wealth quintile). Table 4 gives a summary of the main descriptive statistics of some circumstances variables across different regions for the year 2011/12.

Table 4 Descriptive statistics of circumstance variables (by region) (2011/12)

3.2 Empirical Methodology

As noted above, we aim in the current study to estimate and decompose the common and synthetic measure of the inequality of opportunity, the HOI. The index was firstly developed in 2006 and included to the report of the World Bank (2006). The first application of this novel index was developed in Latin America and the Caribbean (LAC) by de Barros et al. (2009). The main purpose of using such index is to assess the extent to which individual and household’s circumstances (such as birthplace, wealth, gender,...) influence an early child’s probability of accessing basic services required to have a good health and nutrition.

It’s noteworthy that we focus mainly here on inequality of opportunity among early children for two main reasons: Firstly, unlike adults and other ranges of age, children aged less than 5 years haven’t the capacity to access to basic health and nutrition outcomes by themselves. Accordingly, access to these services is considered during the early childhood as an opportunity which depends strongly on the family’s circumstances and other factors. Secondly, the selection of this range of age is explained as well by the fact that required interventions and policies aiming the alleviation of inequality between subgroups early in the life-cycle (children) of an individual are obviously more cost effective and relevant than interventions done later in life.

As noted above, the HOI combines measurements of both the absolute level of opportunities coverage and how fairly those opportunities are distributed among the different circumstances groups. The first component of the index is the average coverage rate of access to a specific opportunity, while the second component measure the fairness of distribution of this opportunity.⁹ Following de Barros et al. (2009), Son (2013), Jemmali and Amara (2015a, 2015b), and Jemmali (2016) and given the surveyed random samples of the population, a binary variable z _i is defined taking a value of 1 if the i ^th child has access to the health or nutrition opportunity and takes a value of 0 if he lacks access to this opportunity. It can be easily demonstrated that E(z _i ) = p _i  = P(z _i ), where p _i is the probability that the i ^th child has access to a specific opportunity. Such probability depends on a range of exogenous (explicative) variables linking to individual, household, and geographical characteristics outside the child’s control (circumstances), such as: gender, parental education, household wealth, geographic location and others. Before calculating the final index, circumstance groups are defined as a set of children sharing a common set of features. For instance, male children having 2 siblings and living in urban areas in the metropolitan region, with higher educated and wealthy parents, on the other side we find a group that contains female children having 4 siblings and living with no educated and poor parents in rural areas in Sidi Bouzid.

Giving a set of predefined k circumstance variables x _i1 , x _i2 ,  …  , x _ik , the probability p _i for each child can be estimated by means of a simple logit model as follows:

$$ {p}_i=\frac{e^{\left({\beta}_0+{\sum}_{j=1}^k{\beta}_j{x}_{ij}\right)}}{1+{e}^{\left({\beta}_0+{\sum}_{j=1}^k{\beta}_j{x}_{ij}\right)}} $$

(1)

The vector of parameters β _j of the logit model can be estimated using the maximum likelihood method. The obtained maximum likelihood estimate, ${\widehat{p}}_i$, provides an accurate estimate of the probability of access to the specific opportunity depending on circumstance variables defined above. Then, any gap in estimated probabilities between circumstances-groups is due to the existence of an inequality of opportunity among children belonging to such groups. After estimating the probability of access to opportunity for each group, a Dissimilarity index that gives a measure of the dissimilarity of access rates to a given service, is computed as follows (World Bank 2006):

$$ \hat{D}=\frac{1}{2\overset{-}{p}}{\sum}_{i=1}^n{w}_i\left|{\hat{p}}_i-\overset{-}{p}\right| $$

(2)

Where $\hat{D}$ is the estimated relative mean deviation, w _i is the population weight associated to the specific opportunity and $\overset{-}{p},$called level of coverage, is the average prevalence of access to the service in the surveyed sample, computed using the following formula:

$$ \overset{-}{p}={\sum}_{i=1}^n{w}_i{\hat{p}}_i $$

(3)

The weight w _i is equal to 1/n where n is the size of the selected sample.

The D-index is used to measure the level of inequality of opportunity explained by the different circumstances, while E which is equal to the difference (1 − D) is the measure of the equity of opportunity. As defined above, the value of D is ranging between 0 and 1 (0 to 100 in percentage terms). D = 0 means that each child in the population benefits from the same opportunities, while D = 1 means that merely one individual in the society benefits from all opportunities.

After estimating the two components: average access to opportunities ($\overset{-}{p}$) and D-index (D), the HOI is calculated as the product of $\overset{-}{p}$ and D-index (see Eq. 4):

$$ HOI=\overset{-}{p}\left(1-D\right) $$

(4)

Giving that D-index is ranging between 0 and 1, HOI is necessarily less than or equal to the level of coverage ($\overset{-}{p}$). This implies that the HOI, which can be identified as an equality-adjusted coverage rate, decrease when the associated D-index is more close to 1. Accordingly, each policy or intervention could increase the value of HOI by means of improving the total opportunity coverage ($\uparrow \overset{-}{p}$), boosting equity of opportunity (↓D) or both coverage and equity.

After estimating the level of equality of opportunity proxied by the HOI, it is promising to assess the contributions of different circumstantial variables in such equality of opportunity using the decomposition procedure proposed by Shorrocks (2013).¹⁰ This decomposition method, called Shapley decomposition, allow us to assess the marginal contribution of each circumstance (such as gender, education, location, family features) to inequality in access to basic health and nutrition services. It consists of estimating the marginal effect in the HOI of each inequality contributor (circumstance) in a specified sequence of elimination (Betti and Lemmi 2008; Shorrocks 2013). The same procedure is applied to all considered outcomes in order to estimate the relative contributions of different groups of circumstance to the observed variance of different outcomes.

Following de Barros et al. (2009), we assume that the dissimilarity index (and therefore the HOI) depends on a set of circumstances and adding other circumstances usually raises the value of the D-index. In other words, the impact of adding a circumstance A is given by the following formula:

$$ {D}_A={\sum}_{S\subseteq N\backslash \left\{A\right\}}\frac{\left|s\right|!\left(n-\left|s\right|-1\right)!}{n!}\left[D\left(S\cup \left\{A\right\}\right)-D(S)\right] $$

(5)

Where N is the set of all the n circumstances; and S is the subset of N circumstances obtained after omitting the circumstance A (i.e. S does not contain the circumstance A). D(S) is the dissimilarity index estimated with the set of circumstances S (after omitting the circumstance A) and D(S ∪ {A}) is the estimated dissimilarity index considering the set of circumstances S and the circumstance A. Then, using the shapely procedure, the contribution of the omitted circumstance A to the dissimilarity index can be estimated using the following formula:

$$ {M}_A=\frac{D_A}{D(N)} $$

Where ${\sum}_{i\in N}{M}_i=1$. This is a critical property satisfied by the Shapley decomposition, which means that the sum of contributions of all circumstances should add up to 1 (100%). To estimate the marginal effect of each contributor, among the nine circumstances, on inequality of access to health and nutrition opportunity, the above procedure is applied afterward.

4 Empirical Analysis

4.1 Circumstances and Access to Basic Healthcare and Nutrition

Before delving into analyzing the extent and the main drivers of inequality of health and nutritional opportunity among early children in Tunisia, a descriptive analysis of how some demographic, socioeconomic and geographical circumstances, which are naturally beyond the children’s control, affect these inequalities, is conducted afterward. Instead of considering each outcome alone, a new variable is created that take the value 1 if the child, respectively benefits from access to all healthcare services and is well protected from malnutrition risks and 0 if not. To emphasize the impact of some principal circumstances, including the location of residence, the wellbeing and parent’s education on the distribution of outcomes, we divide the population into two main and opposite groups depending on some circumstances variables: Most and least advantaged groups of early children. These two population groups make up both the two extremes of the set of groups constructed based on circumstances; they account for nearly 18.5% of the total number of early children aged between 0 and 5 years. The least advantaged group, as presented in the last columns contains children from rural areas, with poor parents having no formal education and having more than 2 children aged less than 5 years. On the other side, the most advantaged group contains children who are from urban area, living with wealthy parents having higher education level and having a number of siblings less than 2 children.

4.1.1 Access to Basic Healthcare Services

Figure 1 illustrates the gap between the two extreme groups (least and most advantaged) in access to basic healthcare services before and after birth (Health_1) and access to necessary immunizations (Health_2) as well as access to housing services: water and sanitation facilities (Health_3). It’s noteworthy that the aim of considering solely the two extreme groups is to highlight the disparity that could occur between children in access to health and nutrition outcomes. It is gleaned from the figure that children from the two groups benefit from considerable access to basic healthcare before and after birth including the necessary immunizations; more than 92.5% of the total number of children in each group have access to aforementioned services. The figure shows as well that the gap between the two extreme groups is narrow in access to these services.

Fig. 1

Access to basic healthcare and nutrition by groups (2011/12)

This preliminary result lead us to conclude that the government have played a great role, during the last decade, in providing these basic healthcare services to all the population independently of the geographical and socioeconomic circumstances that children haven’t no effect on. Nonetheless, only 51.47% of the total children living in better conditions and 0.52% living in difficult conditions have access to basic housing service (water and sanitation facilities), while these necessary services help improve hygiene and health standards. When comparing the two percentages of coverage of the two groups, it’s easy to find a large gap in access to water and sanitation between the least and most advantaged groups. This is due to the huge disparity in access to basic sanitation facilities between different groups particularly between rural and urban areas (see Table 4).

4.1.2 Access to Basic Nutrition Services

Similarly to healthcare services, the propensity of not having a malnutrition problem during the five years after birth differs a bit between children belonging to the two extreme groups. The Fig. 1 reveals that the gap is about 10.64 point for the first composite index (Nutrition_1¹¹), and 1.65 point for the second one (Nutrition_2¹²). It appeared from this comparison that living in urban areas with more wealthy and educated parents have played a considerable role in avoiding any malnutrition risk, while living in rural areas with poor and less educated parents may increase the propensity of having at least one aspect of malnutrition such as stunting, wasting or underweight. As noted above, no significant disparity in the propensity of having the required blood tests, which means that the government has succeeded in providing this service to the majority of mothers during their pregnancy independently of the household background variables. After this brief descriptive analysis of health and nutrition outcomes, we turn afterward to more deeply investigation of the inequality of opportunity in the two sectors with a special focus on the main drivers that shape the inequality pattern in health and nutrition facilities.

4.2 Inequality of Opportunity in Access to Healthcare Facilities and Basic Nutrition

4.2.1 Inequality of Opportunity in Access to Basic Healthcare

Figure 2 and Appendix Table 5 reveals the estimated coverage, Dissimilarity index and the different HOI levels of all basic healthcare services including some composite indicators such as the index of healthcare before and after birth and access (HEA1) and the housing component (HEA3). A child or mother (before birth) is assigned the value of 1 for the index HEA1 if all the antenatal and postnatal healthcare are undertaken during the pregnancy and the first two years of child’s life. Similarly, a child is assigned the value 1 if he benefits from access to both safe water and sewage facilities (HEA3). The Fig. 2 shows at the national scale greater level of both coverage and equality index (1-D-index), and accordingly higher level of HOI index. This high level of equality of opportunity, above 90%, indicates that the majority of children, in Tunisia, benefit from all healthcare services regardless of their circumstances living and region of residence.

Fig. 2

Inequality of opportunity in access to Healthcare services

Nonetheless, Regular access to improved drinkable water and sanitation services, which is a main condition to ensure basic health care and hygiene to population, shows a lower level of HOI, particularly in access to sanitation. In fact, only 34.22% of access to sanitation services is distributed equitably, and when considering the composite index HEA3 the level of the HOI decrease to the lower value, 22.52% (see Fig. 2). This leads us to mention that inequalities in access to improved drinking water and sanitation facilities still persist around the country. Giving that these basic housing services and hygiene¹³ are essential ingredients to ensure child health and survival, development and growth, government should improve access to clean drinking water and effective sanitation more specifically in unprivileged and marginalized rural areas. To highlight the main causes of such inequality in housing services, a Shapley decomposition is conducted.

The Shapley decomposition results presented in the Fig. 3 reveal that regional variables that are urban/rural status and region added to the household’s welfare explain all the largest part of the variations in access to immunizations and housing facilities. It appears from the figure that these geographical and economic variables explain more than 84% of the total variation of the composite housing index (HEA3). The decomposition results reveal, as well, that parent’s education and urban/rural status appear to be the main factors in determining whether the child has benefit from a professional assistance in birth and a postnatal checkup.

Fig. 3

Shapley decomposition of Healthcare opportunities

4.2.2 Inequality of Nutritional Opportunity

As noted above, three main malnutrition indicators, a composite index (NUT1) constructed from these indicators and access to blood tests during pregnancy are used in the current study to assess inequality of nutritional opportunity. The first three variables are linked directly to the nutritional status of the Child, while the second is associated indirectly with such nutritional status. It is gleaned from the Fig. 4 and Appendix Table 5 that malnutrition indicators such as stunting, wasting and underweight prevalence have been deteriorated significantly in Tunisia, inequality of opportunity for these indicators, measured by the D-index is very low, suggesting no significant disparity among circumstance groups. Unsurprisingly, the HOI for all the nutrition indicators are high and close to the coverage value.

Fig. 4

Inequality of opportunity in access to nutrition

The decomposition of the variability in anthropometric measures and access to blood test, illustrated in the Fig. 5, explained by circumstances, shows that geographical and economic circumstances added to parent’s education explain a large percentage of their variance. Therefore the malnutrition and non access to blood tests during pregnancy is a real problem for Tunisian children who belong to poor household living in unprivileged rural areas despite the low prevalence of malnutrition issues and the less inequality in nutritional opportunity in the country. The Shapley decomposition of inequality of opportunity in malnutrition indicators further shows that females are not more likely to be stunted, wasted or underweight than males like other developing countries such as Egypt (Ersado and Aran 2014).

Fig. 5

Shapley decomposition of Nutritional opportunities

5 Concluding Remarks and Policy Implications

Fairness, equity, and social justice in the distribution of different outcomes are no longer in the camp of philosophers and theorists; rather, they become recently in the realm of policy design and economic reform. Increased private participation in providing some basic health services does not exempt governments from their primary responsibility to ensure children’s access to basic healthcare and hygiene services such as assistance in birth, antenatal and postnatal care, immunizations, safe drinkable water and sanitation. Equal access to these services and others, regardless of geographical, socioeconomic, demographical circumstances that are beyond the control of children is a crucial and necessary step towards justice and fairness in the society. In this respect, the current study is focusing on the determination of the level in equity of the distribution of health and nutritional opportunities among Tunisian children. Drawing on the last wave of the MICS surveys, the MICS4, the analysis was carried out mainly using the Human Opportunity Index (HOI) methodology initially developed by the World Bank (2006). From the perspective of policy makers, such methodology may provide a practical diagnostic tool for policy analysis and an appropriate point of reference for gauging progress in the equality of opportunity thanks to its flexibility for application to various circumstances, opportunities, and population groups. Furthermore estimating the contribution of each circumstance, using the Shapley decomposition method, is a crucial step to analyze binding constraints and afford equitable opportunities to all Tunisian children across different subpopulation groups.

The main findings of the study reveal that Tunisia has experienced, during the recent decades, a significant progress with regards to the availability and access to basic healthcare services and ensuring a good nutrition for early children, in some cases with a pro-poor overall effect. In particular, significant improvements have been made in access to immunizations and antenatal and postnatal care. Accordingly, less inequality of opportunity in access to these basic services is observed despite the low coverage rate of these services in some rural inland areas. In these lagging regions, there are areas of persistent and emerging concerns that require urgent interventions, such as access to improved water and sanitation facilities. Geographical circumstances such as urban/rural and region of residence and educational and economic characteristics of households have played the main role in the inequality of health and nutritional opportunities.

Such main findings lead us to note that targeted interventions and more appropriate investments in favor of the less advantaged regions in rural areas and poor population may afford the significant potential for improving the overall equity in health conditions and nutrition status among children. In this respect a more inclusive approach and special efforts would be needed for those early children exposed naturally to various risk factors. From a policy perspective, evidence indicates that appropriate interventions and policies to maintain equality of opportunity between different children groups early in the lifecycle of an individual are noticeably more cost effective and successful than any interventions later in life.

Appendix

Table 5 Summary of human opportunity indices results in Tunisia