The nutritional and health status of women and children in households with and without crossbred cows in Holetta Wereda, Ethiopia¹

Jemal Haider¹, B.I. Shapiro², Tsegaye Demissie¹ and Alemu G/Wold³

1. Ethiopian Health and Nutrition Research Institute, Addis Ababa, Ethiopia

2. International Livestock Research Institute, Addis Ababa, Ethiopia

3. Ethiopian Agricultural Research Organization, Addis Ababa, Ethiopia

1. The authors would like to express sincere gratitude to the former staff of Ethiopian Nutrition Institute for their encouragement and support for this study, to ILRI and EARO for technical and financial support, to the people and the peasant association officials in the study site for their willingness to share information, and to the field enumerators for diligently collecting data.

Abstract

This study was conducted in Holetta Wereda, Ethiopia to evaluate the maternal and child health and nutritional status of owners and non-owners of crossbred cows so that policies and programmes may be identified to ensure positive impacts of the technology. The overall level of malnutrition as determined by stunting, wasting and underweight in pre-school children, and female adult malnutrition as measured by body mass index, were all found to be lower in households with crossbred cows than with local breed cows. Xerophthalmia and calorie and nutrient intake were also significantly higher in the crossbred cow group. No significant differences were found in maternal nutritional knowledge, attitude and practices in relation to childcare and feeding, frequency of illness and treatment of illness, and access to health amenities between the two household groups.

But access to piped water and immunisation of children against common diseases was higher in households with crossbred cows. Households with crossbred cows had higher levels of consumption of calorie, protein and other nutrients. They also consumed more dairy products compared to local cow owners. These differences do not necessarily imply impact of crossbred cows because the influence of other factors on such differences has not been controlled for. Therefore the results should be interpreted with some caution. However, increased food production and incomes brought about by the introduction of crossbred cows may potentially raise the nutrition and health status of the rural farm community, especially impact positively on the more vulnerable members of society, women and children. Current feeding practices also indicate a room for improvement of child health through nutrition education of mothers and carers.

Introduction

The Ethiopian Agricultural Research Organization (EARO), formerly the Institute of Agricultural Research (IAR), and the International Livestock Research Institute (ILRI) have been working on market-oriented dairy production (MODP) technologies which involve the use of crossbred cows (CBC) and complementary feed technologies to increase milk production and the incomes of smallholder farmers. The increased food production and incomes are meant to help improve the human nutrition and health status of individual household members in a sustainable manner, and especially the more vulnerable members of society, women and children. These impacts could be realised through higher milk production leading directly to higher consumption of dairy products or indirectly through the use of the cash income to grow more and different crops or to purchase more and better food.

In order to evaluate the impacts of the introduction of MODP on the nutritional and health status of women and children, a longitudinal study was undertaken in Holetta Wereda, Ethiopia. This site was chosen because of its good dairy potential, persistently high levels of both macro and micronutrient malnutrition, and the introduction of CBC technology through an on-farm research project. Structured questionnaires and clinical measurements were used to collect the relevant information to compare the households with CBC and those with local breed cows (LBC). Emphasis is given in this paper on the impacts of CBC on nutrition and health of women and children. An earlier study in the area reported malnutrition, particularly vitamin A deficiency as a major problem (ENI 1996a). In that study, officials of three peasant associations (PAs) in the Holetta area were asked to have families under their jurisdiction bring their pre-school children (PSC) for clinical eye examination. A total of 122 children were examined. Bitot's spots were found in the eyes of 12 children or about 10%. The prevalence of Bitot's spots was highest at three and four years of age and was somewhat higher among boys than girls, particularly at three years of age (Table 1). One child had a corneal scar in one eye and was blind in that eye. The mother reported that the scar appeared after the child had measles. Eighteen children were determined to have conjunctival xerosis (a pre-cursor of Bitot's spots). However, conjunctival xerosis can be due to vitamin A deficiency or environmental factors.

According to the World Health Organization, if prevalence rates of Bitot's spots rise above 0.5% in a population, vitamin A deficiency should be considered a public health problem (IVACG 1981; Eastman 1988; West 1994). Prevalence rates in Holetta was found to be very high, twenty times the suggested critical value. Thus, from a research viewpoint this site would appear to be suitable for evaluating the nutritional impacts of the MODP technologies, and in particular their effect in mitigating vitamin A deficiency.

Objectives

The overall objective of the study was to determine the effects of the CBC technologies on the nutrition and health status of mothers and children by comparing households with CBC and those with LBC. To achieve this, the following factors were assessed because differences in the two study groups were expected to be directly influenced by the MODP technologies:

• dietary intake and composition

• nutritional status of children under six years of age and adults–male and female

• the prevalence rate of micronutrient deficiency disorders in pre-school children by clinical examination

• the utilisation of the expanded programme of immunisations

• the incidence of disease and their treatment, including mortality, and

• environmental health factors like access to water.

• In addition, the following factors were also assessed because differences in these exogenously determined factors could also affect differences in nutritional status in the two study groups

• the knowledge, attitudes, practices (KAP) of mothers related to infant care

• access to health amenities and services.

Table 1. Results of eye examinations of pre-school children in Holetta, Ethiopia.

Age (years)	Males			Females
	Number of children	With  conjunctival Xerosis or Bitot’s spots  (%)	With  Bitot’s spots (%)	Number of children	With conjunctival Xerosis or Bitot’s spots (%)	With  Bitot’s spots  (%)
1	8	13	0	9	17	0
2	7	14	0	7	0	0
3	12	33	25	12	33	0
4	13	38	15	16	25	19
5	9	33	11	11	27	18
6	9	22	11	12	25	8
All ages	58	28	12	64	23	9

Source: ENI (1996a).

Subjects and methods

Study site

The study site, consisted of six PAs is situated within 5–15 km radius from the EARO research station in Holetta town which is located about 40 km west of Addis Ababa. The farming system in these areas is classified as a mixed crop–livestock system. The land area cultivated comprises between 1.0 and 2.5 ha and average livestock holdings are equivalent to about 8 tropical livestock units. The main crops grown are teff (Eragrostis tef), barely, wheat, and various types of legumes like faba beans, field peas, lentils and chick-peas. Livestock production is a key component of the farming system and is based primarily on cattle, small ruminants, and poultry. The Wereda has one health centre.

Study groups and sample sizes

From the 120 households participating in the on-farm trials with CBCs and the control with LBC, those with children aged between 6 and 59 months were enrolled in this study. A total of 80 households (39 CBC and 41 LBC) who fulfilled these criteria were monitored after the objective of the study was explained to each head of household. In the third and fourth quarters of the year of study, the sample size dropped to 79 households (38 CBC and 41 LBC) as one of the CBC group farmers lost his cows and was moved to the LBC group and one farmer from the LBC group dropped out of the study. Altogether, a total of 689 members were entered into the study including 152 PSC whose eyes were examined for ocular signs of vitamin A deficiency (xerophthalmia), anthropometry measurements were taken, and common type of child illnesses and dietary information were recorded.

Data collection

Data were collected for the year 1997. A clinical baseline was done in the middle of the first quarter. Food intake was done by 24-hour recall randomly once per month, except in the first quarter in which case it was done once in the quarter. Baseline and periodic anthropometric measurements, dietary intake, and morbidity information (2 week recall) was collected and ocular examination was performed. These aspects of the data collection were repeated once in the first three months and then continued at an interval of once a month starting April 1997 and continued to the end of the study period in December 1997 for all study subjects. The study was conducted with the help of two specially trained 12th grade enumerators, each handling 40 households. These enumerators were trained regarding the objectives of the study and data collection techniques. The principal investigators supervised the enumerators and visited the field sites regularly to ensure data quality.

Anthropometry

For children under two year of age, length was measured by having them lie down on a wooden length board with a flat surface as described by Gibson (1994). Older children and mothers were measured using a vertical height board. Length and height were recorded to the nearest 0.1 cm and weight of the children as well as the mother was measured using a digital bath scale to the nearest 0.1 kg.

Following the Waterloo classification method (Gibson 1994), children were then classified as normal (≤–2SD (standard deviation) height for age and weight for height), stunted (below –2SD height for age), wasted (below –2SD weight for height), or both wasted and stunted (below –2SD height for age and weight for height). The nutritional status of the mother was determined by body mass index (weight/height²). Wasting is considered as acutely malnourished, and stunting is considered as chronically malnourished. Children with acute or active malnutrition in both groups were compared and the difference in their nutritional status was determined.

Clinical examination

All children included in the study were clinically examined by a senior staff member of EHNRI following standard techniques for the presence of xerophthalmia and other common type of child illnesses (Sommer 1995). Data on morbidity, treatment of childhood illnesses, and mortality were recorded for all the study children.

Diet intake and composition

To calculate calories, protein, fat, iron, and vitamin intake, dietary data were collected for each household using 24-hours recall method, and the calorie and nutrient contents were computed using the food composition table developed by EHNRI for use in Ethiopia (ENI 1996b). For purposes of analysis, 74 types of ingredients reportedly consumed by both CBC and LBC groups were aggregated into seven groups: staples, cereals, pulses, meat/eggs, dairy, vegetables/oils, and sugar/spice.

Socio-demographics and ethical considerations

Primary caretaker or mothers of the children were interviewed using a pre-tested questionnaire, to document socio-economic, cultural, and health related factors potentially predictive of malnutrition. Household sizes were converted into adult equivalents using standard conversion factors as described in FAO/WHO (1985). Children who were observed to be ill during field survey were either treated on the spot or referred to a nearby clinic, depending on the severity of the illness.

Data handling and analysis

The data obtained were then entered into a computer using dBASE software and analysed by SPSS and ANTHROP software. Chi square and student t-tests were used to test for significance of difference between CBC and LBC groups with LBC as the control group. A P-value less than 5% are considered statistically significant. However, a significant difference between the two groups did not necessarily implied an effect of crossbred cow alone as the effects of other factors could not be controlled for in the simple partial analysis. Ideally, a comparison of situation before and after crossbred cow ownership and with and without crossbred cow ownership should be done to assess the net effect of crossbred cow ownership. Since the results presented in this paper are based only on a cross-sectional comparison between the type of cow ownership, the results should be interpreted with some caution.

Results and discussion

Some health related characteristics of the sample households

There was no significant difference between the CBC and LBC groups regarding KAP except that during the fourth quarter a significantly higher proportion of mothers of CBC households reported participating in village health programme and have taken vaccination when pregnant and they had slightly better knowledge about child feeding practices (Table 2). Mothers in CBC households also asked if the technology has improved the nutritional status of their index children or PSCs: 5.1%, 12.8% and 76.9% of the mothers responded that the situation was worse than before, same as before and better than before, respectively.

Table 2. Distribution of mothers (%) according to maternal knowledge, attitude and practices by technology group.                                                             

* Chi square significant at P<0.05.

A higher proportion of children of CBC households were vaccinated against diseases and a higher proportion of CBC households had access to better sources of water, e.g. protected well and piped water, which may contribute to better human health (Table 3). There was no significant difference between the two groups with respect to other health related factors.

There were significant differences between the two groups in respect of some child feeding practices notably breast feeding, type of food given at birth and the age at which supplementary food is given (Table 4). These differences may be attributed to belief and practices of the households or to the specific characteristics of the children. However, they also indicate possible room for improvement of child health through nutrition education.

Table 3. Distribution of households by technology group and access to health services and other environmental factors.                        

* Chi square significant at P<0.05.

Table 4. Distribution of households according to child feeding practices and other related characteristics by technology.                                                             

* Chi square significant at P<0.05.

Based on reports given by the households, there were some differences between the two groups in respect of incidence of sickness and the type of sickness among children, and the extent of treatment given while sick (Table 5). On clinical examination, fairly similar pattern was found except that a significantly higher proportion of children in LBC group had skin legion in the third quarter and otitis in the fourth quarter. The differences in the incidence of sickness in different quarters of the year could be partly due to seasonal effects, e.g. certain sickness may occur more frequently during rainy season (3rd quarter). Overall prevalence of Bitot's spots among the index children aged 6–71 months was 2.2%, and the most affected were the males. This level of prevalence is many times higher than the critical value of 0.05% beyond which high level of vitamin A deficiency is indicated (West 1994). By the 4th quarter the previously observed Bitot's spot in children of CBC group had disappeared and the incidence rate of conjunctivitis had decreased remarkably in both groups. But the problem remained at a higher level for LBC households.

* Chi square significant at P<0.05.

Nutritional status of the sample households

The overall nutritional status was better in the CBC group as a smaller proportion of children in this group suffered from malnutrition (Table 6). Furthermore, chronic malnutrition indicated by stunting, had decreased significantly over time and increased dairy output or cash income from dairy might have contributed to this improvement.

During the first quarter, about 12% of the male and female adult members in both groups of households were suffering from severe malnutrition (Table 7). By the fourth quarter, this ratio decreased to about 5.5%. Overall, there was no significant difference between the household groups in terms of pattern of prevalence of malnutrition among adult members.

Table 6. Distribution of households according to nutritional status of pre-school children by technology group in four quarters of the year.                                          

* Chi square significant at P<0.05.

The calorie and nutrient consumed by household members were calculated from a monthly 24-hour recall of foods consumed by the surveyed households. The different food components presented in Table 8 were calculated on the basis of the Ethiopian food composition tables (ENI 1994; ENI 1996b) and were converted to adult consumption equivalents.

Body mass index (BMI) = (Wt/Ht2 × 100).

Degree of malnutrition : Normal =BMI >18.5% , Mild = 17.4–18.4%, Moderate = 16.0–17.3%, Severe = <16.0%.

The USDA (United States Department of Agriculture) has fixed the minimum calorie requirement for Ethiopia at 2088 cal per day per adult equivalent. This is a very rough generalisation, but a useful reference, although it does not take into account individual activity levels and energy needs. Out of all surveyed households 30% did not meet this requirement, the majority belonging to the LBC household group.

As indicated in Table 8, there was significant difference in energy intake between the two groups. Households with CBC consumed almost 17% more calories than LBC households. This difference in energy intake corresponds to more fat and protein in the diet. Households with CBC consume 24% more fat and 13% more protein per adult equivalent. These results show significant differences in food consumption levels as well as diet composition between the two groups. Households with CBCs eat more calories and derive more nutrients from dairy products.

Table 8. Energy and nutrient consumption per day per adult equivalent by technology group in adult equivalents (per day/adult equivalent).                                                                        

The numbers in parenthesis are the coefficients of variation.

**, * Indicate significant at the 0.01 and 0.05 level, respectively.

Table 9 illustrates the diet composition of CBC and households, which is expressed by their respective share of calories obtained from different food groups. Calories consumed in the form of dairy products are 2.5 times higher in CBC households. Significant difference was also recorded for pulses, but not for staples or for meat. Vegetable consumption is not very common. Vegetables consumed are primarily potatoes and onions. The higher consumption of pulses might be due to sampling bias toward fasting days compared to non-fasting days. It may be noted that most of the people in the area are Orthodox Christians and they practice about 160 days of fasting in a year during which only pulses and other vegetables and no animal products are consumed. While interviewing households for a 24-hour recall every month, no distinction was made between fasting and non-fasting days.

Summary and conclusions

The overall prevalence of malnutrition among children and the mothers in the villages is high. A predominance of malnutrition in under five children was found. The finding that most of the children are abnormal (stunted) confirms the significance of the problem and the need for further preventive efforts. The clinical examination of the index children's eyes revealed an overall prevalence of Bitot's spots in the children aged 6 to 71 months old to be 2.2%, several times higher than the 0.5% level considered as the cutting point beyond which prevalence of vitamin A deficiency is indicated. Early sign of xerophthalmia was found to be common, but the disappearance of Bitot's spots in the CBC group by the fourth quarter indicated that access to more and better nutrition in the CBC households might have contributed to reduce the intensity of the problem. The environmental factors, which appear to play a role in human nutrition and health in the study area, are related to economic, cultural, and social shortcomings governing the farming practices, environmental sanitation, and feeding practices of the communities studied.

a. Mostly Kollo, a local snack which is a mixture of whole grains and where often pulses are added.

*, ** Indicate significant at the 0.05 and 0.01 level, respectively.

The nutritional status of the index children determined by stunting was significantly better in the CBC compared to the control group over the four quarters of the year under study. Common child illnesses and the presence of clinical signs for vitamin A deficiency were also lower in the CBC group. The overall level of malnutrition as determined by stunting, wasting and underweight in pre-school children, and adult malnutrition determined by body mass index were all found to be better in the CBC than the LBC group. Child morbidity, calorie and nutrient intake were also observed to be better in the CBC group.

Most of the energy consumed in both groups is obtained from staples followed by cereals and pulses. Dairy products as sources of protein and energy were more important for the CBC group indicating the contribution and potential of MODP technology in improving nutrition of poor households.

The implication of introducing MODP for improving resource productivity, income and nutrition need to be analysed more rigorously. In the meanwhile, on the basis of the preliminary results presented above, the following recommendations can be made

• diffusion of the crossbred cow technology to other Weredas appropriate for MODP

• greater efforts made in health and nutrition education, with more emphasis on early childhood feeding, consumption of safe water, and other factors affecting morbidity and thus nutritional status

• liaison between the nutritional and agriculture sectors to promote additional nutritionally supportive food production activities such as horticulture practices to complement the dairy activities, and improvement in storage and distribution of high quality food

• involvement of the community in recognising existing problems of childhood malnutrition and in identifying solutions.

References

Eastman S.J. 1988. Vitamin A deficiency and xerophthalmia: Recent findings and some programme implications. In: Mandl P.E. (ed), Assignment children 3. UNICEF, New York, USA. 84 pp.

ENI (Ethiopian Nutrition Institute). 1994. Nutritional tables for Ethiopia. ENI, Addis Ababa, Ethiopia.

ENI (Ethiopian Nutrition Institute). 1996a. Nutritional survey, Holetta, 1996. Technical Report, ENI, Addis Ababa, Ethiopia.

ENI (Ethiopian Nutrition Institute). 1996b. Food composition use for Ethiopia. ENI, Addis Ababa, Ethiopia.

FAO/WHO (Food and Agriculture Organization of the United Nations/World Health Organization). 1985. Coefficients for converting family size into standardised household size. FAO, Rome, Italy. 87 pp.

Gibson R.D. 1994. Principles of nutritional assessment. Oxford University Press, Oxford, UK. 691 pp.

IVACG (International Vitamin A Consultative Group). 1981. The symptoms and signs of vitamin A deficiency and their relation to applied nutrition. IVACG, Washington, D.C., USA. 9 pp.

Sommer A. 1995. Vitamin A deficiency and its consequences: A field guide to detection and control. Third edition. WHO (World Health Organization), Geneva, Switzerland. 69 pp.

West K. 1994. Vitamin A deficiency: Its epidemiology and relation to child mortality and morbidity. In: Blomhoff R. (ed), Vitamin A in health and disease. Marcel Dekker, Inc., New York, USA. pp. 585–614.