MALNUTRITION AND SOCIOECONOMIC GAPS IN 1 MALNUTRITION AND SOCIOECONOMIC GAPS IN MALNUTRITION IN GHANA Van de Poel E 1 Department of Applied Economics, Erasmus University Rotterdam,

1

MALNUTRITION AND SOCIOECONOMIC GAPS IN MALNUTRITION IN GHANA

Van de Poel E 1 Department of Applied Economics, Erasmus University Rotterdam, The Netherlands

Department of Economics, University of Antwerp, Belgium

Hosseinpoor A Equity, Poverty and Social Determinants of Health, Evidence and Information for

Policy, World Health Organization, Switzerland

Jehu-Appiah C Policy Planning Monitoring and Evaluation Division, Ghana Health Service, Ghana

Speybroeck N

Equity, Poverty and Social Determinants of Health, Evidence and Information for

Policy, World Health Organization, Switzerland

Summary. – This study identifies factors that are associated with malnutrition and socioeconomic gaps in

malnutrition in Ghana using data from the Ghana 2003 Demographic Health Survey. Socioeconomic

inequality is measured in terms of a concentration index of malnutrition, which is decomposed into

inequality into the determinants of malnutrition. The results show that malnutrition in Ghana is related to

poverty, education, health care and family planning services and regional characteristics. To reduce poor-

rich disparities in malnutrition, health policies in Ghana should further be directed at

strategies/interventions to reduce poverty and to improve the use of health care services and maternal

education among the poorer population groups. Furthermore, regional disparities should be tackled.

1Correspondence to:

Ellen Van de Poel

Department of Applied Economics

Erasmus University Rotterdam

Burg. Oudlaan 50

3000 DR Rotterdam,The Netherlands

Tel: +31 10 408 1502

Fax: +31 10 408 91 41

E-mail: [email protected]

2

INTRODUCTION

Background

In the developing world, an estimated 230 million (39%) children under the age of five

are chronically malnourished and about 54% of deaths among children younger than 5

are associated with malnutrition (UNICEF, 2000). Malnutrition is a major public health

and development concern, certainly in Sub-Saharan Africa, and has foregoing health and

socioeconomic impacts. In Sub-Saharan Africa, the prevalence of malnutrition among the

group of under-fives is estimated at 41% (UNICEF, 2000). It is the only region in the

world where the number of child deaths is increasing and in which food insecurity and

absolute poverty are expected to increase (United Nations, 2000; Smith, Obeid & Jensen,

2000; Smith & Haddad, 2000). Malnutrition in early childhood is also associated with

significant functional impairment in adult life and reduced work capacity, hereby

affecting economic productivity (Pelletier, Frongillo & Habicht, 1993; Pelletier &

Frongillo, 1995; Vella et al, 1992; Delpeuch et al, 2000; Mendez & Adair, 1999;

Schroeder & Brown, 1994). Children who are malnourished not only tend to have

increased morbidity and mortality but are also more prone to suffer from delayed mental

development, poor school performance and reduced intellectual achievement (Pelletier,

Frongillo & Habicht, 1993, Pelletier & Frongillo, 1995, Schroeder & Brown, 1994).

Chronic malnutrition is usually measured in terms of growth retardation. It is widely

accepted that children over the world have much the same growth potential, at least to

seven years of age. Environmental factors, diseases, inadequate diet, and the handicaps of

poverty appear to be far more important than genetic predisposition in producing

deviations from the reference. These conditions, in turn, are closely linked to overall

standards of living and the ability of populations to meet their basic needs. Therefore, the

assessment of growth not only serves as one of the best global indicators of children’s

nutritional status, but also provides an indirect measurement of the quality of life of an

entire population (de Onis, Frongillo & Blossner, 2000; Lavy et al, 1996; Martorell et al,

1992).

3

Large scale development programs such as the Millennium Development Goals (MDGs)

have also picked up on the importance of the under-fives' nutritional status as indicators

for evaluating progress (UN Millennium Project, 2006). Progress towards the MDGs and

in meeting the needs of the worlds poorest in general, should benefit all people,

irrespective of their socioeconomic status. When aiming at reducing childhood

malnutrition, it is important not only to consider averages, which can obscure large

inequalities across population groups. Failure to tackle these inequalities is acting as a

brake on making progress towards achieving the MDGs and is a cause of social injustice

(UNDP, 2005; Bambas et al, 2005).

GHANA

Against this background, Ghana provides an interesting case study. The country

experienced remarkable gains in health from the immediate post independence era. Life

expectancy improved over the years and the prevention of a range of communicable

diseases improved child survival and development. However in the last decade despite

increasing investments in health, Ghana has not achieved envisaged health outcomes.

There has been no significant change in Ghana’s under-five and infant mortality rates

between 1993 and 2003. In the last couple of years, under-five mortality is actually

slightly increasing. Life expectancy has also stagnated between 57.42 years in 2000 and

56 years in 2005 (Ghana Statistical Service, 2004). Ghana’s human development index

(HDI) is worsening too; after improving from 0.444 in 1975 to 0.563 in 2001, the HDI

dropped to 0.520 in 2005 (UNDP, 2005). Since 1988, there has been no definite trend in

malnutrition (in terms of height-for-age). Apparent gains between 1988 and 1998 were

reversed in 2003 (ORC Macro, 2005). Although the Ghana DHS 2003 final report

(Ghana Statistical Service, 2004) recommends caution when using data from the various

DHS surveys to assess the trend in the nutritional status, it is noted that there is a trend

over the past five years of increased stunting compared to a decrease of wasting and

underweight. Further, there is a trend of continued high values of stunting in the North

compared to the South (Ghana Statistical Service, 2004; Shepherd et al, 2004). However,

a further note of care is necessary when assessing trends in malnutrition as comparisons

between the 2003 and 1997 CWIQ (Core Welfare Indicators Questionnaire) data show

4

increases in the percentage of children stunted, wasted and underweight in every region

of the country, except in the three Northern regions (ISSER, 2005). 1

A paradigm shift in Ghanaian health policy is taking place in 2006. The theme for the

new Health policy in Ghana is ‘Creating Wealth through Health”. One of the fundamental

hypotheses of this policy is that improving health and nutritional status of the population

leads to improved productivity, economic development and wealth creation (Ministry of

Health, 2006). Since this policy adopts an approach that addresses the broader

determinants of health, it has thus generated interest in socio-economic inequalities in

health and malnutrition. It is further recognised that not paying attention to malnutrition

inequalities during the early years of life is likely to perpetuate inequality and ill health in

future generations and thus defeat the aims of the new health policy.

Malnutrition in Ghana is most prevalent under the form of Protein Energy Malnutrition

(PEM), which results in growth retardation and underweight. About 54% of all deaths

beyond early infancy are associated with PEM, making this the single greatest cause of

child mortality in Ghana (Ghana Health Service, 2005 a).

The contribution of this paper is twofold. First it delivers evidence on the determinants of

malnutrition in Ghana. Secondly, to our knowledge this is the first study to provide

insight into the factors behind socioeconomic inequality in malnutrition in Ghana.

METHODS

Measuring malnutrition

Nutritional status was expressed in terms of height-for-age z-scores. An overview of

other nutritional indices and why height-for-age is the most suited for this kind of

analysis is provided in Pradhan et al (2003). The z-scores are calculated as:

height of child - median height of reference population

standard deviation of reference population.

Generally, children whose height-for-age z-score is below minus 2 standard deviations of

the median of the (US) reference population are considered chronically malnourished or

stunted. In the regression models, the negative of the z-score is used as dependent

5

variable ( y ). This facilitates interpretation since it has a positive mean and is increasing

in malnutrition (Wagstaff et al, 2003). There are two advantages of using the z-score

instead of a binary or ordinal variable indicating whether the child is

(moderately/severely) stunted. First, the z-score contains more information on the depth

and duration of malnourishment instead of simply indicating whether or not the child is

malnourished. Second, it allows for linear regression analysis, which facilitates the

interpretation of coefficients and the decomposition of socioeconomic inequality.

The concentration index as a measure of socioeconomic inequality

Assume y is the negative of the height-for-age z-score that is linearly increasing in

malnutrition. A concentration index (C) of malnutrition results from a concentration

curve. This curve plots the cumulative proportion of children, ranked by socioeconomic

status, against the cumulative proportion of y , as illustrated in Figure 1.

If all children, irrespective of their socioeconomic status, have the same y , the

concentration curve would coincide with the diagonal. The concentration curve lies above

the diagonal if y is larger among the poorer children and vice versa. The further the

curve lies from the diagonal, the higher the socioeconomic inequality in ill health. A

concentration index is a measure of this inequality and is defined as twice the area

between the concentration curve and the diagonal. If children with low socioeconomic

status suffer more malnutrition than their better off peers, the concentration curve lies

above the diagonal and the concentration index will be negative (Wagstaff & Van

Doorslaer, 2004). 2

Decomposition of socioeconomic inequality

More formally, a concentration index can be written as:

1

2

1

1 −=

∑

∑

=

=n

i

i

n

i

ii

y

Ry

C

where iy refers to the i -th individual and iR is its respective fractional rank in the

income distribution. If iy is linearly modelled

6

∑=

++=K

k

iikki xy1

εβα

, Wagstaf et al (2003) showed that the concentration index of y can be decomposed into

inequality in the determinants of y as follows:

µµβ εGC

Cx

C k

K

k

kk +

=∑

=1

where µ is the mean of y , kx is the mean of kx , kC is the concentration index of kx and

εGC is the generalized concentration index of the residuals. The latter term reflects the

socioeconomic inequality in y that cannot be explained by the model and is calculated as

∑=

=n

i

iiRn

GC1

2εε .

Because of the survey nature of the data, all estimates take into account clustering and

sample weights.

DATA

Data is used from the 2003 Ghana Demographic Health Survey (DHS) and restricted to

children under the age of 5. Anthropometric measures are missing for 12.3% of children

in this age group. The final sample contains information on 3000 children.3 The

nutritional status of a child is specified to be a linear function of a vector of child-level

characteristics such as age, gender, duration of breastfeeding, size at birth; a vector of

maternal characteristics such as education, mother's age at birth, birth interval, parity,

marital status, use of prenatal health services, occupation and finally a vector of

household-level characteristics such as wealth, type of toilet facility, access to safe water,

region and urbanization. The explanatory variables are described in Table 1. All have

well documented relevance in the literature (Smith & Haddad, 2000; Vella et al, 1992; de

Onis et al, 2000; Lavy et al, 1996; WHO, 1997; Wagstaff et al, 2003; Ukuwuani &

Suchindran, 2003; Alderman, 1990; Tharakan & Suchindran, 1999; Larrea & Kawachi,

2005; Brakohiapa et al, 1988; Ruel et al, 1999). 4

7

RESULTS

Summary statistics

In 2003 DHS data for Ghana, 29.35% of children under the age of 5 are stunted. The

concentration index for stunting in children under the age of 5 equalled -0.167

(SD=0.018). This negative value implies that poor children had a higher probability of

being stunted than their better off peers. Figure 2 illustrates the strong socioeconomic

inequality in childhood stunting. The malnutrition rate among children in the poorest

quintiles was more than twice the rate of children in the richest quintile. Figure 3 shows a

comparative picture of stunting and socioeconomic inequality in stunting across the Sub-

Saharan African region. 5

Summary statistics and bivariate relationships between explanatory variables and stunting

are shown in Table 2. Stunting is defined as height-for-age being below minus 2 SD from

the median of the reference population.

Determinants of malnutrition

The regression coefficients and their significance are shown in the first two columns of

Table 3. Note that the dependent variable is increasing in malnutrition, such that a

negative coefficient should be interpreted as lowering malnutrition.

Malnutrition increased with the child's age in a non-linear way. Children who were very

small at birth had a higher probability to be stunted than children with normal size. Male

children were more prone to malnutrition than their female peers. Long duration of

breastfeeding had a borderline significant negative impact on children's growth.

With respect to maternal characteristics, the existence of a short birth interval and higher

parity were significantly increasing malnutrition. Children of women that (highly)

accessed health services and education were less prone to being malnourished. Maternal

occupation showed no clear effect.

Household wealth and regional variables showed a strong significant association with

childhood malnutrition. Sanitation variables had no significant effect on malnutrition. As

compared to the Northern region all regions were associated with lower malnutrition.

This effect was the largest for the Accra and Volta region. The high regional disparities in

malnutrition are further illustrated in Figure 4. The four most deprived regions in Ghana

8

(Northern, Central, Upper East and Western regions) exhibited the greatest burden of

malnutrition.

Decomposition of socioeconomic inequality in malnutrition

Table 3 also shows the concentration index and the absolute and relative contributions of

each determinant to socioeconomic inequality in childhood malnutrition. For the ease of

interpretation, the last column shows the grouped contribution of the categorical variables.

A negative contribution to socioeconomic inequality implies that the respective variable

is lowering socioeconomic inequality and vice versa. A variable can contribute to

socioeconomic inequality in malnutrition both through its effect on malnutrition and

through its unequal distribution across wealth groups. The extent to which each of the

explanatory variables is unequally distributed across wealth is reflected by its C value. A

negative C means that the determinant is more prevalent among poorer households.

Wealth explained the major part (45%) of socioeconomic inequality. Other important

contributors were regional variables (15%), the use of health care services (15%) and

maternal education (9%). The age of the child was contributing negatively to

socioeconomic inequality (-9%). This means that the combined effect of its coefficient

and its distribution by wealth was lowering socioeconomic inequality in malnutrition.

Older children are more likely to be stunted and are more prevalent in higher wealth

quintiles. The latter is reflected by the positive and significant C of the variable age3

(SD=0.0056). The contribution of the error term only amounted to about 4%, meaning

that the decomposition model functioned well in explaining socioeconomic inequality in

malnutrition.

DISCUSSION

Relative to other Sub-Saharan countries, Ghana appeared to have a rather low level of

average stunting, combined with relatively high socioeconomic inequality in stunting.

Determinants of malnutrition

Malnutrition in Ghanaian children accumulated over time. The same age pattern was

found in Vella et al (1992), Wagstaff et al (2003) and Tharakan & Suchindran (2005).

The higher prevalence of malnutrition among boys as compared to girls, and the negative

9

effect of long breastfeeding are also commonly found in the literature (Vella et al, 1992;

Wagstaff et al, 2003; Brakohiapa et al, 1988; Larrea & Kawachi, 2005). Long duration of

breastfeeding may be associated with higher malnutrition because it reflects lack of

resources to provide children with adequate nutrition. It is also possible that children who

are breastfed for a long time are more reluctant to eat other foods, as was found by

Brakohiapa et al (1988) in their study on a cohort of Ghanaian children.

Short birth intervals and high parity affected childhood growth negatively by placing a

heavy burden on the mother’s reproductive and nutritional resources, and by increasing

competition for the scarce resources within the household (Brakohiapa et al, 1988).

Children of younger mothers could be more prone to malnutrition because of

physiological immaturity and social and psychological stress that come with child

bearing at young age (Heaton, 2005).

Maternal education was only significantly lowering childhood malnutrition from the level

of secondary education on. This may reflect low quality primary education for women

currently in childbearing years. It can also point to education only generating the

necessary income to purchase food from the level of secondary education on. However,

although education is often suggested to be a measure of social status, the effect stayed

significant after controlling for household wealth and living conditions. A high level of

maternal education could also lower childhood malnutrition through other pathways such

as increased awareness of healthy behaviour, sanitation practices and a more equitable

sharing of household resources in favour of the children (Smith & Haddad, 1999; Vella et

al, 1992).

Sanitation in terms of having a toilet and access to safe water was not significantly

affecting malnutrition. Ukwuani et al (2003) also reported this result, but they did find a

significant effect of sanitation on wasting (which reflects current nutritional status).

Therefore they concluded that good sanitation can avoid episodes of diarrhoea and hereby

affecting current nutritional status, while it may not be sufficient for long term child

growth.

The higher levels of malnutrition of the population living in the northern regions of

Ghana have already been observed many decades ago (see e.g. Alderman, 1990). This

regional pattern reflects ecological constraints, worse general living conditions and

10

access to public facilities in the Northern regions. In addition, the persistence of this

regional inequality can point to an intergenerational effect of malnutrition. Since women

who were malnourished as children are more likely to give birth to low-birth-weight

children, past prevalence of child malnutrition is likely to have an effect on current

prevalence.

Decomposition of socioeconomic inequality in malnutrition

The high socioeconomic inequality in childhood malnutrition is mainly associated with

wealth, regional characteristics, use of health care services and maternal education.

Wealth was responsible for almost half of the socioeconomic inequality in malnutrition.

This means that poorer children were more likely to be malnourished, mainly because of

their poverty. The regional contribution results from the fact that poorer children are

more likely to live in regions with disadvantageous characteristics. The regional

inequality in Ghana originates from both geographical and historical reasons. Much of

the North is characterized by lower rainfall, savannah vegetation, periods of severe

drought and remote and inaccessible location. Further, the colonial dispensation ensured

that northern Ghana was a labor reserve for the southern mines and forest economy and

the post-colonial failed to break the established pattern (Shepherd et al, 2004).

Health services use was also responsible for a substantial proportion of socioeconomic

inequality in malnutrition. This comes from the combined effect of the positive effects of

health services use on childhood growth and the unequal use across socioeconomic

groups. The reason for the lower health care use amongst the poor may be due to several

barriers including the cost of care, cost of transportation and lower awareness on health

promoting behavior (Lindstrom & Munoz-Franco, 2006). User fees were introduced in

Ghana in 1985 as a cost-sharing mechanism at all public health facilities. To ensure

access to health care services for the poor and vulnerable the government introduced fee

exemptions. Then again in 2003, a new policy for exempting deliveries from user fees in

the four most deprived regions of the country, namely Central, Northern, Upper East and

Upper West regions were introduced. To further bridge the inequality a key

recommendation of the Ghana Poverty Reduction Strategy (GPRS I) was to allocate 40%

of the non-wage recurrent budget to the deprived regions. However, experience to date

indicates that Ghana has not been able to implement an efficient exemption mechanism

11

or commit to the 40% budgetary allocation to achieve the principal purpose. In addition

to these financial hurdles, poorer people are often also located further from health centers.

The ratios of population to nurses and doctors are the highest the poorest regions of

Ghana. For example the ratio of population to doctors in the northern region is 1:81338

compared to the national average of 1:17733. Trends also show that since 1995 the

Northern region has had the lowest average number of outpatient visits per capita in the

country (Ghana Health Service, 2005 b).

The negative contribution of age comes from the combined facts that older children are

more likely to be malnourished and at the same time more prevalent in the richer wealth

quintiles. The latter could be related to higher child and infant mortality rates amongst

poorer households that cause the proportion of older children to be lower among poor

households as compared to richer households.

Considerations and limitations

There exist some limitations of this study. First, for 12.3% of the children below 5 years

of age, anthropometric scores were missing. To the extent that these missing values are

not random this could introduce a problem. A logit model explaining the selection in the

sample and a Heckman sample selection model (using different exclusion restrictions)

were used to check for this (Wooldridge, 2002). Both tests did not reveal large sample

selection problems, and coefficients in the Heckman model were very similar to those in

the model presented here. Second, DHS only collects information on the recent food

consumption of the youngest child under three years of age living with the mother.

Restricting the sample to these children would substantially reduce the number of

observations. However, the analysis was also conducted on this sub sample, using food

consumption as one of the determinants of malnutrition (indices were created similar to

Tharakan & Suchindran, 1999; Larrea & Kawachi, 2005). Since the regression and

decomposition results did not alter much, these are not presented in this paper (but are

available with the authors upon request). Third, it is important to note that this paper is

not modelling causal pathways, but merely showing the factors that are associated with

malnutrition and socioeconomic inequality in malnutrition and the magnitude of these

associations. It must be admitted that the model is likely to suffer from endogeneity in the

sense that e.g. mothers who use health care services are likely to differ in other –

12

unobservable – characteristics that influence children’s nutritional status.6 This problem

however should not be exaggerated as the aim of the paper is to model correlations, not

causality.7

CONCLUSIONS AND POLICY IMPLICATIONS

The regression results show that malnutrition in Ghana is a multisectoral problem.

However in Ghana it often falls through the cracks since it has no institutional home.

Tackling malnutrition therefore calls for a shared vision and should be viewed and

addressed in a broader context (World Bank, 2004). Therefore special attention needs to

be given to policies aimed at reducing malnutrition based on the magnitude and nature of

determinants of malnutrition, such as poverty, education, health care and family planning

services and regional characteristics. Currently in Ghana, various interventions are being

implemented to reduce both PEM and micro nutrient deficiencies. These include the

Infant and Young Child Feeding Strategy (IYCF) and Community Based Nutrition and

Food Security project among others. However these initiatives address only the

symptoms of malnutrition and cannot have a sustained impact in the long term as they do

not deal with the root causes.

If equity goals are to be achieved, health policies in Ghana should further be directed at

strategies/interventions to reduce poverty and to improve the use of health care services

and maternal education among the poorer population groups. Furthermore, regional

disparities should further be tackled to narrow the gap in malnutrition between the poor

and the rich. A starting point could be for policy makers to include under-five

malnutrition differentials to set criteria to guide resource allocation to regions. Moreover,

the strong regional contributions to socioeconomic inequality, even after controlling for

other factors such as household wealth and education, bring forward the issue of

geographical targeting. Further targeting public programs towards the central and

northern regions would substantially reduce socioeconomic inequality in malnutrition and

is administratively easier than targeting the poor. The latter argument is relevant for

Ghana, where pro-poor policies (redistribution schemes and exemption policies) are not

having the aimed effect because of problems in identifying the poor (Bosu et al, 2000;

Bosu et al, 2004). Geographic targeting reduces leakage of program benefits to the non-

13

needy compared to untargeted programs, although under coverage of the truly needy can

increase. “Fine-tuning" the targeting by basing it on smaller geographic units increases

efficiency, but in some circumstances may be costly and politically unacceptable (Baker

& Grosch, 1994).

With respect to Ghana, regional averages should be interpreted with caution as there is

large heterogeneity between districts in each region and indeed among socio-economic

groups within districts. In this case, polices aimed at reducing child malnutrition based on

regional averages may lead to under coverage of those in need.8 Nonetheless, there is a

need for additional research to further decompose regional malnutrition inequalities to

generate valuable information for policy making decisions. The Ghana Growth and

Poverty Reduction Strategy (GPRSII) for 2006 – 2009 states that one of the strategies to

be implemented is developing and implementing high impact yielding strategies for

malnutrition (GPRS II, p122.) This would mean targeting areas at the greatest risks of

malnutrition, replicate best practices and expand coverage. This then should result in

decreasing malnutrition rates among children particularly in rural areas and northern

Ghana.

14

0

20

40

60

80

100

0 20 40 60 80 100

cumulative % of children ranked by socioeconomic

status

cumulative % of ill health (y)

Figure 1: Concentration curve of ill health (example).

Inequality in stunting, by wealth quintile (GDHS 2003)

0

5

10

15

20

25

30

35

40

45

Q1 Q2 Q3 Q4 Q5

wealth quintiles

% of children

0.25

.5.75

1

Cumul % hfa_2sd

0 .25 .5 .75 1Cumul % ranked by pcares

Figure 2: Distribution of stunting across wealth quintiles and concentration curve of stunting in

function of wealth.

NOTE: Stunting is defined as height-for age z-score being below minus 2 standard deviations from the

median of the reference population.

Zambia

Uganda

Tanzania

Rwanda

Nigeria

Namibia

Mozambique

Mauritania

Mali

Malawi

Madagascar

Kenya

Ghana

Gabon

EthiopiaChad

Cameroon

Burkina Faso Benin

0

0.05

0.1

0.15

0.2

0.25

0.3

0 10 20 30 40 50 60

Average stunting in children<5 years (%)

(-) Concentration Index of stunting

Figure 3: Average stunting versus socioeconomic inequality in stunting in under-five children, Sub-

Saharan African countries with recent DHS surveys.

NOTE: Stunting is defined as height-for age z-score being below minus 2 standard deviations from the

median of the reference population.

15

Inequality in stunting in children<5 years, by region

(GDHS 2003)

0

10

20

30

40

50

60

western

central

greater accra

volta

eastern

ashanti

brong ahafo

northern

upper west

upper east

% of children

Inequality in stunting in children<5 years by

grouped regions (GDHS 2003)

0

5

10

15

20

25

30

35

40

45

northern middle southern accra

regions (grouped)

% of children

Figure 4: Inequality in stunting by regions and grouped regions (as in Larrea & Kawachi, 2005).

16

Variable Description

breastfeeding duration of breastfeeding (in months)

age age of child split into 3 categories:age1≤6 months; 6 months<age2≤12 months; age3>12

size size of child at birth in 5 categories: very large, large, normal, small, very small

Sex sex of child: male(1), female (0)

region region of residence: Western, Central, Accra, Volta, Eastern, Ashanti, Brong Ahafo,

Upper West, Upper East, Northern

urban urban location (1), rural location (0)

wealth

Wealth quintiles (quintile 1) based upon principal component analysis. The wealth

indicator is estimated on household level and combines the following assets: electricity,

radio, TV, fridge, bike, motor, car, phone and the type of the flooring material (Filmer &

Pritchett, 2001).

toilet having a toilet (flush toilet, traditional pit toilet, ventilated improved pit latrine) (1-0)

water

Whether the household has access to safe water available (1-0). The following sources of

water supply were regarded as safe water: piped water (piped into dwelling, piped into

yard, plot, or public tap); water from open well (manually pumped water, protected

well); covered well or borehole (public well or private well); tanker truck or vendor; and

bottled water.

parity number of children the mother has given birth to

birth interval whether there were less than 24 months between the child's birth and the birth of the

previous child (1-0)

married whether the child's mother is married or living together (1-0)

education education level of the mother split into 3 categories: no education, primary, secondary

and higher

health service Use of health services (low, moderate, high) estimated by principal component analysis.

The indicator combines skilled birth attendance, antenatal care and proportion of

recommended vaccinations. (see Larrea & Kawachi, 2005)

maternal occupation professional, technical, managerial; clerical; sales; agriculture; services; manual; not

working

maternal age maternal age at birth in years

Table 1: Description of independent variables

NOTE: Reference categories for categorical variables used in the regression model are underscored.

17

variables Stunting

% ni/nk Cramer's V/Phi p

duration of breastfeeding

not breastfed 12.86 3.5/27 0.12 0.00

< 12 months 15.42 111/723

12-24 months 32.18 420/1306

> 24 months 36.88 348/944

Age of child

< 6 months 9.13 32/352 0.13 0.00

6-12 months 17.56 63/357

>24 months 34.21 784/2291

size at birth

very large 18.93 62/326 0.05 0.00

large 28.65 247/862

normal 28.82 367/1273

small 38.69 138/356

very small 38.81 71/183

sex of child

male 32.39 490/1514 0.06 0.00

female 26.35 391/1486

region

Western 28.19 81/286 0.07 0.00

Central 32.25 65/201

Accra 11.73 30/259

Volta 24.35 52/212

Eastern 26.01 69/267

Ashanti 28.42 130/458

Brong Ahafo 28.89 102/353

Upper West 33.69 96/284

Upper East 30.62 61/200

Northern 48.2 231/480

urban/rural location

urban 20.21 166/821 0.11 0.00

rural 33.93 739/2179

wealth

quintile 1 38.81 433/1117 0.08 0.00

quintile 2 31.15 178/573

quintile 3 28.95 165/570

quintile 4 22.82 72/316

quintile 5 14.71 62/424

toilet

yes 25.19 457/1814 0.11 0.00

no 39.08 463/1186

safe water

yes 26.61 484/1820 0.06 0.00

no 33.93 400/1180

parity

1-2 25.25 266/1054 0.05 0.00

3-5 30.11 360/1197

>5 34.5 258/749

risky birth interval

yes 36.33 101/279 0.04 0.02

no 28.63 779/2721

married

yes 29.16 805/2759 0.01 0.53

18

no 31.5 76/241

education

no education 37.9 532/1403 0.10 0.00

primary 24.78 368/1487

secondary & higher 10.21 11/110

health services

low 35.60 423/1189 0.08 0.00

moderate 31.00 293/945

high 20.69 179/866

maternal age

<20 years 33.2 108/325 0.02 0.31

20-40 years 28.83 696/2416

<40 years 29.75 77/259

occupation

not working 29.7 84/283 0.13 0.00

prof, tech, man 14.68 8/52

sales 23.01 147/638

agriculture 34.88 516/1481

services 18.49 23/126

manual 28.62 120/420

Table 2: Summary statistics

Note: Stunting is defined as height-for-age z-score below minus 2 SD from the median of the reference

population.

19

Table 3: Regression and decomposition results (dependent variable: negative height-for-age z-score).

Number of observations= 3000, R2= 0.2395, C of dependent variable=-0.1387.

variables coefficient p-value C contribution % con % con

breastfeeding 0.0079 0.0920 -0.0180 -0.0019 1.37 1.37

age

age2 0.6918 0.0000 -0.0161 -0.0010 0.73 -8.25

age3 1.4407 0.0000 0.0143 0.0125 -8.98

size at birth

size very large -0.2188 0.0100 0.1378 -0.0028 2.01 3.08

size large -0.0246 0.7020 -0.0159 0.0001 -0.07

size small 0.2718 0.0010 -0.0790 -0.0020 1.42

size very small 0.3872 0.0000 0.0285 0.0005 -0.39

sex of child 0.2321 0.0000 -0.0017 -0.0002 0.11

region

Western -0.4564 0.0000 0.0345 -0.0013 0.94 14.85

Central -0.3392 0.0230 -0.0923 0.0022 -1.59

Accra -0.6948 0.0000 0.5043 -0.0296 21.30

Volta -0.5689 0.0000 -0.1582 0.0057 -4.12

Eastern -0.4683 0.0010 0.0792 -0.0031 2.27

Ashanti -0.2616 0.0360 0.1822 -0.0070 5.04

Brong Ahafo -0.3167 0.0120 0.0297 -0.0008 0.57

Upper West -0.5419 0.0000 -0.4102 0.0056 -4.04

Upper East -0.6726 0.0000 -0.2837 0.0076 -5.51

urban -0.0843 0.2990 0.4231 -0.0093 6.72 6.72

wealth

quintile 1 0.4248 0.0000 -0.6782 -0.0731 52.71 45.19

quintile 2 0.3633 0.0030 -0.1705 -0.0091 6.55

quintile 3 0.2454 0.0180 0.2187 0.0086 -6.20

quintile 4 0.2158 0.0690 0.5406 0.0109 -7.86

toilet -0.1149 0.2340 0.1318 -0.0083 6.01 6.01

water -0.0135 0.8190 0.1288 -0.0009 0.62 0.62

parity 0.0341 0.0780 -0.0625 -0.0064 4.63 4.63

birth interval 0.2195 0.0140 0.0364 0.0006 -0.44 -0.44

married -0.0430 0.6690 0.0082 -0.0003 0.18 0.18

education

primary -0.0773 0.2650 0.1350 -0.0046 3.33 9.33

secondary & higher -0.3876 0.0010 0.6541 -0.0083 6.00

health services

moderate -0.0777 0.2600 -0.0383 0.0008 -0.54 14.72

high -0.2706 0.0020 0.3002 -0.0212 15.26

maternal age -0.0129 0.0510 -0.0049 0.0014 -1.02 -1.02

occupation

prof, tech, man -0.0308 0.8470 0.6792 -0.0003 0.24 -0.56

sales -0.2110 0.0370 0.2607 -0.0107 7.70

agriculture -0.1480 0.1530 -0.2607 0.0136 -9.81

services 0.0027 0.9850 0.2381 0.0000 -0.02

manual -0.1453 0.2350 0.1165 -0.0018 1.32

constant 0.5940 0.0130

error -0.0050 3.57 3.57

total -0.1387 100.00 100.00

20

NOTES

1 This could be due to feeding programs which could have allowed the percentage of children underweight

and stunted to either decline or remain stable in those regions.

2 It should be noted that the concentration index is not bounded within the range of (-1,1) if the health

variable of interest takes negative, as well as positive values. This clearly is the case with the height-for-age

z-scores. Rescaling the z-scores to positive values would be one option. This would leave the

decomposition results unchanged; however the value of the concentration index would then be sensitive to

the transformation chosen (Erreygers, 2005). Since the purpose of this paper is to identify the determinants

of malnutrition and the factors behind socioeconomic inequality in malnutrition and not to interpret or

compare concentration indices, the problems with having negative health outcomes are limited, and

therefore ignored. However one should bear this comment in mind when interpreting results.

3 Because of the nature of the data, only children of living mothers could be included. Children of mothers

who have died may have worse health status and their mothers may have had systematically different

characteristics (see e.g. Lindblade et al, 2003). This could lead to a selection problem, but since neither

these children nor their mothers could be observed, this problem could not be corrected for.

4 No information on mother’s nutritional status was included in the set of explanatory variables. Since

about 10% of women in the dataset were pregnant at the time of interview, their BMI did not provide an

accurate measure of their nutritional status. Furthermore, BMI reflects current nutritional status and may

not be relevant for children born 5 years prior to the interview. Inclusion of mother’s height-for-age had no

significant effect on results.

5 Stunting and socioeconomic inequality in stunting are calculated for each country on DHS data in

exactly the same way as is described for the Ghana DHS. For each country, the same set of assets was used

to construct the wealth index.

6 The same problem holds for wealth. Although focusing on child health avoids the direct feedback of

income and health that is usually present in microeconomic studies, there remains the problem that both

nutritional status and wealth may be jointly influenced by unobserved factors, hereby inducing simultaneity.

7 To model causal pathways, one should ideally use panel data that allow to condition upon individual

effects. Further, it would be better to have health care variables regarding proximity or prices of health care

21

services than regarding usage. The latter is both demand and supply driven and it may be possible that

women experiencing difficult pregnancies are more likely to seek antenatal care. This would result in an

underestimation of the positive effects of health care use on children’s health (see e.g. Conway & Deb,

2005). However, no such data were available in the 2003 Ghana DHS. Another option would be to predict

health care use, but we were not able to find strong predictors for health care.

8 Morris et al (1999) expose some important limitations of geographic targeting if used to place poverty-

alleviation or nutrition interventions within cities. Using data from Abidjan (Cote d'Ivoire) and Accra

(Ghana), they found significant clustering in housing conditions; however they did not find any sign of

geographic clustering of nutritional status in either city. This implies that geographic targeting of nutrition

interventions in these and similar cities has important limitations. Geographic targeting would probably

lead to a significant under coverage of the truly needy and, unless accompanied by additional targeting

mechanisms, would also result in significant leakage to non-needy populations.

REFERENCES

Alderman H. (1990). Nutritional status in Ghana and its determinants. Social Dimensions

of Adjustments in sub-Saharan Africa. World Bank Working Paper nr 3. World

Bank, Washington.

Available at: www.worldbank.org/afr/poverty/pdf/docnav/02956.pdf

Baker J, Grosch M. (1994). Poverty reduction through geographic targeting: How well

does it work? World development, 22, 7, 983-995.

Bambas L, Braveman P, Dachs JNW, Delgado I, Gakidou E, Moser K, Rolfe L, Vega J,

Zarowsky C. (2005). Strengthening health information systems to address health

equity challenges. Bulletin of the World Health Organization, 83, 8, 561-640.

Bosu WK, Nsowah-Nuamah N, Ward PM. (2000). A Profile of Health Inequalities in

Ghana. Ministry of Health, Ghana.

Bosu WK, Larya-Adjei GL, McIntyre D. (2004). A Review of the Ghana Health sector's

Pro-poor Agenda. Ghana.

Brakohiapa LA, Bille A, Quansah E, Kishi K, Yartey J, Harrison E, Armar MA,

Yamamoto S. (1988). Does prolonged breastfeeding adversely affect a child's

nutritional status? Lancet, 332, 8608, 20, 416-418.

22

Conway KS, Deb P. (2005). Is prenatal care really ineffective? Or, is the ‘devil’ in the

distribution? Journal of Health Economics, 24, 3, 489-513.

Delpeuch F, Traissac P, Martin-Pre Y, Massamba JP, Maire B. (2000). Economic crisis

and malnutrition: socioeconomic determinants of anthropometric status of

preschool children and their mothers in an African urban area. Public Health

Nutrition, 3, 1, 39-47.

de Onis M, Frongillo EA, Blossner M. (2000). Is malnutrition declining? An analysis of

changes in levels of child malnutrition since 1980. Bulletin of the World Health

Organization, 78, 10, 1222-1233.

Erreygers, G. (2005). Beyond the health concentration index: an Atkinson

alternative to the measurement of socioeconomic inequality of health. University

of Antwerp. Antwerp

Filmer D, Pritchett L. (2001). Estimating wealth effects without expenditure data - or

tears: An application to educational enrolments in states of India. Demography,

38, 1, 115-132.

Ghana Growth and Poverty Reduction Strategy Paper (GPRSI). (2003). Ghana Poverty

Reduction Strategy 2003-2005: An Agenda for Growth and Prosperity.

Available at :

http://siteresources.worldbank.org/GHANAEXTN/Resources/Ghana_PRSP.pdf

Ghana Growth and Poverty Reduction Strategy Paper (GPRSII). (2005). National

Development Planning Commission, Ghana. Available at :

http://siteresources.worldbank.org/INTPRS1/Resources/Ghana_PRSP(Nov-

2005).pdf

Ghana Health Service. (2005).

a) Imagine Ghana free of malnutrition. Ghana.

b) Facts and Figures 2005. Ghana.

Available at:

http://www.ghanahealthservice.org/includes/upload/publications/FACTS%20

AND%20FIGURES%202005.pdf

Ghana Ministry of Health. (2006). National Health Policy Draft. Ghana.

23

Ghana Statistical Service (GSS), Noguchi Memorial Institute for Medical Research

(NMIMR), and ORC Macro. (2004). Ghana Demographic and Health Survey

2003. Calverton, Maryland.

Heaton T, Forste R, Hoffmann J, Flake D. (2005). Cross-national variation in family

influences on child health, Social Science and Medicine, 60,97-108.

ISSER (Institute of Statistical, Social and Economic Research). (2005). University of

Ghana, Ghana. Available at: http://www.isser.org/69%20Cavalcanti.pdf

Larrea C, Kawachi I. (2005). Does economic inequality affect child malnutrition? The

case of Ecuador. Social Science and Medicine, 60, 165-178.

Lavy V, Strauss J, Thomas D, de Vreyer P. (1996). Quality of health care, survival and

health outcomes in Ghana. Journal of Health Economics, 15, 333-357.

Lindblade KA, Odhiambo F, Rosen DH, De Cock KM. (2003). Health and nutritional

status of orphans < 6 years old cared for by relatives in western Kenya. Tropical

Medicine and International Health, 8, 67-72.

Lindstrom DP, Munoz-Franco E. (2006). Migration and maternal health services

utilization in rural Guatemala. Social Science and Medicine, 63, 706-721.

Martorell R , Rivera J, Kaplowitz H, Pollitt E. (1992). Long-term consequences of

growth retardation during early childhood. In: Hernandez M, Argente J, eds.

Human growth: basic and clinical aspects. Amsterdam, Elsevier Science

Publishers, p. 143–149.

Mendez MA, Adair LS. (1999). Severity and timing of stunting in the first two years of

life affect performance on cognitive tests in late childhood. Journal of Nutrition,

129, 1555–1562.

Morris S, Levin CE, Armar-Klemesu M, Maxwell D, Ruel MT. (1999). Does geographic

tareting of nutrition interventions make sense in cities? Evidence from Abidjan

and Accra. World Development, 27, 11, 2011-2019.

ORC Macro. (2005). Trends in Demographic, Family Planning, and Health

Indicators in Ghana, 1960-2003: Trend Analysis of Demographic and Health

Surveys Data. Calverton, Maryland, USA.

Pelletier DL , Frongillo EA. (1995). The effects of malnutrition on child mortality in

24

developing countries. Bulletin of the World Health Organization, 73, 4, 443–448.

Pelletier DL, Frongillo EA, Habicht JP. (1993). Epidemiological evidence for a

potentiating effect of malnutrition on child mortality. American Journal of Public

Health, 83, 8, 1130-1133.

Pradhan M, Sahn DE, Younger SD. (2003). Decomposing world health inequality.

Journal of Health Economics, 22, 271-293.

Ruel MT, Levin CE, Armar-Klemesu D, Maxwell D, Morris S. (1999). Good care

practices can mitigate the negative effects of poverty and low maternal schooling

on children’s nutritional status: Evidence from Accra. World Development,

27, 11, 1993-2009.

Schroeder DG, Brown KH. (1994). Nutritional status as a predictor of child survival:

summarizing the association and quantifying its global impact. Bulletin of the

World Health Organization, 72, 569–579.

Shepherd A, Gyimah-Boadi E, Gariba S, Plagerson S, Wahab Musa A. (2004). Bridging

the North South divide in Ghana. World Development Report 2006, Background

Papers. Available at:

http://siteresources.worldbank.org/INTWDR2006/Resources/477383-

1118673432908/Bridging_the_North_South_Divide_in_Ghana.pdf

Smith LC, Haddad L. (2000). Explaining child malnutrition in developing countries: a

cross country analysis. International Food Policy Research Institute, Food

Consumption and Nutrition Division discussion paper nr. 60.

Available at: http://www.ifpri.org/pubs/abstract/111/rr111.pdf

Smith LC, Obeid AEE, Jensen HH. (2000). The geography and causes of food insecurity

in developing countries. Agricultural Economics, 22, 199-215.

Smith LC, Ruel MT, Ndiaye A. (2005). Why is child malnutrition lower in urban than in

rural areas? Evidence from 36 developing countries. World Development, 33, 8,

1285-1305.

Tharakan CT, Suchindran CM. (1999). Determinants of child malnutrition - An

intervention model for Botswana. Nutritional Research, 19, 6, 843-860.

Ukuwuani FA, Suchindran CM. (2003). Implications of women's work for child

25

nutritional status in sub-Saharan Africa: a case study of Nigeria. Social Science

and Medicine, 56, 2109-2121.

Vella V, Tomkins A, Borghesi A, MiglioriGB, Adriko BC, Crevatin E. (1992).

Determinants of child nutrition in north-west Uganda. Bulletin of the World

Health Organization, 70, 5, 637-647.

UNICEF (United Nations Children’s Fund). (2000). The state of the world's children.

Available at: http://www.unicef.org/sowc00/

UN (United Nations). (2000) Sub-Saharan Africa – the human costs of the 2015

‘business-as-usual’ scenario. Available at:

http://hdr.undp.org/docs/events/Berlin/Background_paper.pdf

UNDP (United Nations Development Program). (2005). Human Development Report.

Available at: http://hdr.undp.org/reports/global/2005/

United Nations Millennium Project. (2006). About the Goals: What they are.

Available at: http://www.unmillenniumproject.org/goals/

Wagstaff A, van Doorslaer E, Watanabe N. (2003). On decomposing the causes of health

sector inequalities with an application to malnutrition inequalities in Vietnam.

Journal of Econometrics; 112, 207-223.

Wagstaff, A., & van Doorslaer, E. (2004). Overall versus socioeconomic health

inequality: a measurement framework and two empirical illustrations. Health

Economics, 13, 297-301.

WHO (World Health Organization). (1997). Global database on malnutrition. De Onis M,

Blossner M. World health Organization, Geneva.

Wooldridge JM. (2002). Econometric analysis of Cross Section and Panel Data, MIT

Press, Cambridge, 752 p.

World Bank (2004). Improving, Health, nutrition and population outcomes in Sub-

Saharan Africa. World Bank, Washington, 245 p.