NUTRITIONAL ANTHROPOMETRIC SURVEYS ADJUMANI - UGANDA 20th - 25th

NUTRITIONAL ANTHROPOMETRIC SURVEYS ADJUMANI - UGANDA

20th - 25th April 1998

Carine SUBIT (Nutritionist)

Action Against Hunger – USA (ACF-USA)

Uganda

2

CONTENTS 1. TYPE OF SURVEY AND SAMPLE SIZE 9

1.1. SIMPLE SYSTEMATIC SAMPLING 9 1.2. CLUSTER SAMPLING 9

2. SAMPLING METHODOLOGY 10 2.1. SIMPLE SYSTEMATIC SAMPLING 10 2.2. CLUSTER SAMPLING 10

3. COLLECTED DATA 10 4. INDICATORS AND GUIDELINES USED 11

4.1. WEIGHT FOR HEIGHT INDEX 11 4.2. MID UPPER ARM CIRCUMFERENCE (MUAC) 11

5. FIELD WORK 11 6. ANALYSIS OF THE DATA 12 1. SETTLEMENTS RECEIVING 80 - 100% OF CEREALS 13

1.1. DISTRIBUTION BY AGE AND SEX 13 1.2. NUTRITIONAL ANTHROPOMETRIC ANALYSIS 14 1.3. MEASLES VACCINATION 18 1.4. ADDITIONAL INFORMATION 19 1.5. RELATIONSHIP BETWEEN CHILD'S NUTRITIONAL STATUS AND OTHER FACTORS 20

2. SETTLEMENTS RECEIVING 50% OF CEREALS 23 2.1. DISTRIBUTION BY AGE AND SEX 23 2.2. NUTRITIONAL ANTHROPOMETRIC ANALYSIS 24 2.3. MEASLES VACCINATION 28 2.4. ADDITIONAL INFORMATION 28 2.5. RELATION BETWEEN CHILD'S NUTRITIONAL STATUS AND OTHER FACTORS 29

3. SETTLEMENTS RECEIVING 0% OF CEREALS 32 3.1. DISTRIBUTION BY AGE AND SEX 32 3.2. NUTRITIONAL ANTHROPOMETRIC ANALYSIS 33 3.3. MEASLES VACCINATION 37 3.4. ADDITIONAL INFORMATION 38 3.5. RELATION BETWEEN CHILD'S NUTRITIONAL STATUS AND OTHER FACTORS 39

1. COMPARISON OF THE WEIGHT FOR HEIGHT DISTRIBUTION EXPRESSED IN Z-SCORE 41 2. COMPARISON OF THE WEIGHT FOR HEIGHT DISTRIBUTION EXPRESSED IN PERCENTAGE OF THE MEDIAN 42 3. ADDITIONAL INFORMATION 43 4. RELATION BETWEEN CHILD'S NUTRITIONAL STATUS AND OTHER FACTORS 44

3

SUMMARY METHODOLOGY Three surveys were carried out in Adjumani between the 20th and the 25th of April 1998: one survey in the settlements receiving 80 -100% of cereals, one in those receiving 50% of cereals and one in those receiving no cereals. The settlements receiving 80 - 100% of cereals are: Olua, Agojo East, Maaji 1 to 5, Mongola Esia, Dubaju, Umwia 2, Oliji, Magburu 2, Alere 1 and Robidire. The settlements receiving 50% of cereals are: Alere 2, Mongola Adidi, Mongola Aliwara, Mongola Opi, Kolididi, Nyeu, Keyo 1 to 3 and Arra. The settlements receiving no cereals are: Okusijoni, Obilonkogo, Agojo South, Nyimanzi 1 and 2, Umwia 1, Baratuku, Elema, Ibiaworo and Magburu 1. For the survey in settlements receiving 80 - 100% of cereals, the total number of children (aged from 6 to 59 months old) surveyed was 900, 30 clusters of 30 children. For the survey in settlements receiving 50% of cereals, the total number of children (aged from 6 to 59 months old) measured was 810, 30 clusters of 27 children. For the survey in settlements receiving no cereals, 390 children (aged from 6 to 59 months old) randomly chosen by simple systematic sampling were measured. RESULTS Global acute malnutrition of children aged from 6 to 59 months old : 80 - 100% of cereals: Z-Score: 5.0% [3.2% ↔ 7.6%] percentage of the median: 2.9% [1.6% ↔ 5.1%]. 50% of cereals: Z-Score: 7.5% [5.2% ↔ 10.7%] percentage of the median: 4.2% [2.5% ↔ 6.8%]. No cereals: Z-Score: 3.6% [2.2% ↔ 5.7%] percentage of the median: 1.8% [0.9% ↔ 3.5%]. Severe acute malnutrition of children aged from 6 to 59 months old: 80 - 100% of cereals: Z-Score: 0.8% [0.2% ↔ 2.3%] percentage of the median: 0.4% [0.0% ↔ 1.8%]. 50% of cereals: Z-Score: 0.9% [0.2% ↔ 2.5%] percentage of the median: 0.4% [0.0% ↔ 1.8%]. No cereals: Z-Score: 1.0% [0.4% ↔ 2.4%] percentage of the median: 0.8% [0.3% ↔ 2.2%]. Global acute malnutrition of children aged from 6 to 29 months old : 80 - 100% of cereals: Z-Score: 7.8% [4.8% ↔ 12.3%] percentage of the median: 4.3% [2.2% ↔ 8.1%]. 50% of cereals: Z-Score: 12.7% [8.6% ↔ 18.3%] percentage of the median: 7.2% [4.2% ↔ 11.9%]. No cereals: Z-Score: 5.6% [3.2% ↔ 9.4%] percentage of the median: 2.6% [1.1% ↔ 5.6%]. Severe acute malnutrition of children aged from 6 to 29 months old: 80 - 100% of cereals: Z-Score: 1.3% [0.3% ↔ 4.1%] percentage of the median: 0.6% [0.0% ↔ 3.1%]. 50% of cereals: Z-Score: 1.4% [0.3% ↔ 4.5%] percentage of the median: 0.5% [0.0% ↔ 3.1%]. No cereals: Z-Score: 1.5% [0.5% ↔ 4.1%] percentage of the median: 1.0% [0.2% ↔ 3.4%].

4

Measles coverage: 80 - 100% of cereals: proved by card, at or after 9 months: 40.4% according to the parents: 28.3% proved by card, but before 9 months: 15.3% 50% of cereals: proved by card, at or after 9 months: 42.2% according to the parents: 25.1% proved by card, but before 9 months: 16.6% No cereals: proved by card, at or after 9 months: 46.8% according to the parents: 27.1% proved by card, but before 9 months: 12.8% DISCUSSION The percentage of malnutrition of children aged from 6 to 59 months old is low. However this percentage is statistically higher for children aged from 6 to 29 months old than that of children aged from 30 to 59 months old, especially in the settlements receiving only 50% of cereals. The measles vaccination coverage is low : less than 50% are vaccinated at, or after, 9 months, proven by card. However, according to the card and to what the parents can recall, around 75% of the children have been vaccinated. SUGGESTED ACTIONS

- To make regular nutritional surveillance in this area, to study the evolution of nutritional status of children under five years old, especially in the settlements receiving no cerals. A second nutritional survey is planned (in October 1998) in order to monitor the nutritional situation in all the settlements.

- To establish a routine nutritional surveillance system within the health centres and refer malnourished children to feeding programs in order to curb acute malnutrition.

- To consider starting a supplementary feeding program, especially in the settlements receiving only 50% of cereals.

- To strengthen the health education programme, focusing mainly on childcare and nutrition, weaning habits and on reproductive health in the settlement.

- To improve the measles vaccination coverage through the planned vaccination campaign and improve the EPI system.

- To sensitise refugees about the advantage of giving birth in health centre or with the help of qualified personnel, and to encourage prenatal consultation.

- To realise some food security assessments in order to: . identify the more vulnerable groups in the settlements receiving no cereals. Even if

they seem to be able to develop income-generating activities and the malnutrition rate is low, it appears that many children are vulnerable and can easily become malnourished.

. define some actions to improve the nutritional status of children in the settlements receiving 50% of cereal, where less families cultivate and where children are at higher risk of becoming severe malnourished.

. define some actions to improve the autonomy and decrease the dependence, on food distribution, of the refugees in the settlements receiving 80-100% of cereal, where lots of them have no sources of income.

5

ACKNOWLEDGEMENT

The realisation of these surveys has only been possible through the help of all the implementing partners, the UN agencies and the individuals who have collaborated with us. Action Against Hunger - USA (ACF-USA) would like to thank UNHCR, AHA, LWF and WFP for their logistic support and for their help in sharing their knowledge of the field. I would also like to express thanks to the surveyors and the supervisors for their hard work during the survey.

6

INTRODUCTION Many refugees from South Sudan, Democratic Republic of Congo, Rwanda, and some from Somalia, Ethiopia and Kenya came into Uganda. The Sudanese refugees are staying mainly in the north of the country : Adjumani, Palorinya, Rhino Camp, Imvepi and Acholpii. Some are settled in the west, in Kiryandongo and Kyangwali. The other refugees are in the west and Southwest of the country, in Kyangwali (Sudanese, Congolese and Rwandese), Nakivale (Rwandese, Somalis, Ethiopians, Congolese and Kenyans) and Orunchinga (Rwandese). In order to assess the children's nutritional status in all the refugee settlements in Uganda, UNHCR requested ACF-USA to organise nutritional surveys in 1998 in all the settlements with the help of the different implementing partners. Two surveys were planned in 1998 in each settlement: one in April - June and one in October - December. In Adjumani, the total number of refugees was 62,638 in February 1998 scattered in 22 settlements. The food distribution depends on the date of arrival in the settlement and on the harvests. Since March 1998 the food distribution’s ration is as follows (in g/person/day):

Table 1 FOOD DISTRIBUTION IN ADJUMANI SETTLEMENTS

Items 100% 80% 50% cereals, 100% pulses

and oil

50% cereals pulses,

100% oil 50%

0% cereals, 50% pulses,

100% oil 0%

Maize Grains/ Sorghum 450 360 225 225 225 0 0

Or Maize meal 360 288 180 180 180 0 0 Pulses, legumes 60 60 60 30 30 30 0

Vegetable Oil 20 20 20 20 10 20 0 Calories (Kcal) 1838 Kcal 1538 Kcal 1089 Kcal 1009 919 260 Kcal 0 % of protein 11.2% 11.4% 11.9% 10.2% 11.2 10.2%

% of fat 18.6% 20.2% 24.3% 25.8% 18.6 70.7%

Settlements

Dubaju, Mongola Esia, Maaji 4

and 5, Magburu 2, Oliji, Olua,

Umwia 2

Alere 1,

Robidire, Agojo East, Maaji 1, 2

and 3

Alere 2

Kolididi, Nyeu,

Keyo 1, 2 and 3,

Mongola Adidi and Aliwara

Arra, Opi

Obilokongo, Okusijoni,

Agojo South

Umwia 1, Nyimanzi 1

and 2, Baratuku, Ibiaworo, Elema,

Magburu 1 The caloric value of the food given has been calculated for a ration composed of maize grain, the value of which is higher than for maize meal (flour). The actual quantities received by the refugees from November 1997 to March 1998 are given in the appendix (information from LWF). They are summarised below:

- in November 97, refugees received Cereals, Pulses and Sugar for 30 days and Oil for 15 days, - in December 97, they received Cereals for 30 days, Sugar and Oil for 15 days, - in January 98, all the refugees received Cereals for 30 days. The refugees from Magburu 2,

Umwia 2 and Maaji 5 received Pulses, Sugar and Oil for 30 days, in addition to the cereals. The others received only cereals.

- in February 98, refugees from all the settlements (except those from Magburu 2; Umwia 2, Maaji 5 and Olua) received Cereals, Pulses and Oil for 15 days. Those from Magburu 2, Umwia 2, Maaji 5 and Olua received Cereals, Pulses and Oil for 30 days. No more sugar and salt had been distributed since this month.

7

- in March 98, refugees from all the settlements (except those from Magburu 2; Umwia 2, Maaji 5, Mongola Esia, Mongola Oliji and Olua) received Cereals, Pulses and Oil for 15 days. Those from Magburu 2, Umwia 2, Maaji 5, Mongola Esia and Oliji received Cereals, Pulses and Oil for 30 days. Some refugees from Olua received Cereals, Pulses and Oil for 30 days, others received the same for 22 and 14 days respectively.

Most of the refugees have been given a residential plot and a farming plot according to family size. Many Non Governmental Organisations carry out various activities in refugee settlement in Adjumani, as indicated in the table below.

Table 2 AGENCIES WORKING IN ADJUMANI SETTLEMENTS

Activities

Agency Water / Sanitation

Medical Nutrition

Food security / Income

generation Education / Counselling

Camp management

Action Against Hunger - USA

ACF - USA

Borehole drilling, water resource development

Nutritional surveys

Agency for Cooperation in Research and Development -

ACORD

Agriculture, environment

protection and income generation

African Educational Fund - AEF

School hardware development in

Maaji and Mongola

African Humanitarian Action - AHA

Health centres

Therapeutic and

supplementary centres

Jesuit Refugee Service - JRS

School software development

and/or support

Lutheran World

Federation - LWF

Food distribution

Registration and transfers, road and other infrastructure developments

Finnish Refugee

Council - FRC

Community services, income

generation in Maaji and Mongola

Transcultural Psychosocial Organisation -

TPO

Counselling

Boreholes, schools, health centres have been built and are managed by the different implementing partners in all the settlements. The settlements with a health post or centre are: Mongola Esia, Mongola Aliwara, Maaji sites 1, 4 and 8, Okusijoni, Obilokongo, Ibiaworo, Nyumanzi 1 and 2, Elema, Olua, Kolididi, Alere, Keyo, Agojo, Nyeu, Arra, Magburu, Dubaju and Mireyi. The total number of consultations (for the refugees) was 13,407 in May 1998. 4,818 pathologies have been detected for the children under 5 years: 52% of these children seen in the health units were affected by malaria, 21% by respiratory tract infections, 8% by intestinal worms and 7% by diarrhoea.

8

There is one therapeutic feeding centre in Mongola (17 children, end of May), one in Alere (1 child) and one in Maaji 4 (9 children). 11 supplementary feeding centres were opened in April-May 98. Mongola Opi (43 children, end of May), Mongola Aliwara (10 children), Maaji site 4 (29 children), Okusijoni (24 children), Alere (23 children), Olua (42 children), Keyo (21 children), Magburu (12 children), Agojo (10 children), Elema (15 children), and one in the transit camp, Mireyi (12 children). AEF conducted nutritional surveys in January 1997 and in December 1997 in all the settlements except Mongola, Maaji and Olua (not yet opened) and in transit camps (closed in February 98). For both surveys, 600 children have been surveyed (30 clusters of 20 children each). The results of these two surveys were as follow.

Table 3 RESULTS OF THE SURVEYS DONE BY AEF

January 97 December 97 Settlements Transits Settlements Transits

Indicators Z-score % median Z-score %

median Z-score % median Z-score %

median Global acute malnutrition 11.2% 6.9% 8.0% 5.3% 4.2% 4.7% 4.7% 3.5%

Severe acute malnutrition 1.4% 1.5% 1.0% 0.5% 1.0% 0.6% 1.2% 0.8%

Presence of oedema 1.7% 1.7% 1.7% 0.7% 0.3% 0.3% 0.2% 0.2%

ACF-USA also conducted nutritional surveys in Mongola in April 96, December 96 and February 97, and also in Maaji in December 1997. These surveys revealed the following: in Mongola (in Z-score): 750 children (30 clusters of 25 children) were surveyed in December 96 and in February 97.

Table 4

RESULTS OF THE SURVEYS DONE BY ACF-USA IN MONGOLA

Indicators April 96 December 96 September 97 Global acute malnutrition

8.7% [5.3 - 13.9]

13.9% [10.6 - 18.0]

4.3% [2.5 - 7.0]

Severe acute malnutrition

2.1% [0.6 - 5.6]

3.5% [1.9 - 5.6]

0.1% [0.0 - 1.5]

in Maaji: 373 children were surveyed. The methodology chosen was the systematic sampling methodology.

0.8% [0.0 - 1.6%] severe malnutrition and 8.3% [5.8 - 10.8%] global malnutrition in Z-score, 0.5% [0.0 - 1.2%] severe malnutrition and 5.9% [3.7 - 8.1%] global malnutrition in percentage of the median.

In order to take into account the food distribution's disparity, three surveys were carried out in April 1998: one concerning the settlements receiving 80 - 100% of cereals, one concerning those receiving 50% of cereals and one concerning those receiving no cereals.

9

OBJECTIVES The general objective of the survey is :

• To evaluation the nutritional status of children aged from 6 to 59 months old. The specific objectives are :

• To evaluation of the measles coverage of children aged 9 to 59 months, • To relate information about the household and its sources of income with the child's nutritional

status, • To assess the use of the health facilities by pregnant mothers especially for delivery.

NOTE : Some relationships between malnutrition and other factors appeared during the analysis of the survey. However, this type of survey is not designed to show malnutrition "risks factors". To show causative relations, a case control study will be needed, involving a different way of calculating the sample size. Therefore, the relations appearing here are only suggestions, not established relations.

METHODOLOGY 1. Type of survey and sample size

1.1. Simple systematic sampling For the survey concerning the settlements receiving no cereals, systematic sampling was used. The minimum sample size which allows a reasonable precision (whatever the prevalence), is 450 children aged from 6 to 59 months old, when the population of children aged from 6 to 59 months old is more or equal to 5,000. However, in this case, the population was less than 5,000. Thus the sample size was revised. The aim was to take fewer children without losing precision. The revision formula is indicated below: new sample size = 450 / (1 + (450 / N)) where N is the number of children aged from 6 to 59 months old. The total population in these settlements was: 8,490 in February 1998 with an estimated number of children under 5 years of 1,700 (20% of the total population). Thus the revised sample size became 356.

1.2. Cluster sampling

For the surveys concerning the settlements receiving 50% of cereals and those receiving 80 - 100% of cereals, two stage cluster samplings were used. The minimum sample size which allows a reasonable precision (whatever the prevalence), is 900 children aged from 6 to 59 months old (30 clusters of 30 children), when the population of children aged from 6 to 59 months old is more or equal to 5,000. However, in case this population is less than 5,000 the sample size can be revised. The aim is to take fewer children without losing precision. The revision formula is as follows: new sample size = 2 * 450 / (1 + (450 / N)) where N is the number of children aged from 6 to 59 months old. The total population in the settlements receiving 50% of cereals was 17,696 in February 1998 with an estimated number of children under 5 years of 3,500. Thus the revised sample size became 810 (30 clusters of 27 children). The total population in the settlements receiving 80 - 100% of cereals was 36,452 in February 1998 with an estimated number of children under 5 years of 7,300. Thus the sample size was 900 (30 cluster of 30 children).

10

2. Sampling methodology The target population was children aged from 6 to 59 months old.

2.1. Simple systematic sampling

The total number of family heads in the settlements receiving no cereals was 2,993. Therefore, in order to obtain 356 children or more, we had to survey a child in each 8th family. A number between 1 and 8 was randomly chosen (number: 8), it represented the first family to survey in each settlement. Subsequently, the team arrived in the settlement, went in the centre and chose one direction, then counted 8 families and surveyed one child in this family. And thus, by counting 8 houses, the teams surveyed in total 390 children. If the 8th house had no child, the team went to the 7th house, 6th house, ... until it found one child.

2.2. Cluster sampling ]

Each settlement had been divided into clusters or blocks. The population figures by cluster or block were available. For each survey (50% and 80 - 100%) thirty clusters according to the population in each sub-division were selected. The team went to the centre of the chosen sub-division and threw a pencil to determine a direction. The first household visited was randomly selected among the houses counted between the centre and the limit of the cluster in the chosen direction. The other households of the cluster were then selected by choosing the nearest house to the first house and then the nearest to the second and so on. If there was more than one family in the house (a family was defined as people who shared the same meal), only one family was randomly selected. In one such selected family, one child aged from 6 to 59 months with a height between 65 and 110 cm, was randomly selected and measured. 3. Collected data

For each child chosen aged from 6 to 59 months old :

• Age was recorded with help of a local calendar of events. • Sex was recorded. • Weight (SALTER balance of 25 Kg, precision of 100 g). Children were weighed without clothes. • Height (Shorr measuring board, precision of 0.1 cm). Children less than 85 cm were measured

lying down. Children more or equal to 85 cm were measured standing up. • Mid upper arm circumference (MUAC) (precision of 0.1 cm). The MUAC was measured at mid

point of the left upper arm. • Applying thumb pressure for at least 3 seconds to both feet assessed existing bilateral oedema

on the lower limbs.

Other additional questions that had to be asked : • Measles vaccination, for children from 9 to 59 months : mothers were asked for vaccination

cards. If no vaccination card was available, mothers were asked if their children had been immunised against measles.

• Date of arrival: the family was asked for the date of arrival in Uganda (month and year). • Number of meals: this meant the number of meals the child had the day before. It did not

include breast-feeding and amount of fluids taken. Only solid food was considered as a meal (breakfast, family's meals, and porridge).

• Sources of income: families were asked their sources of income. The possible answers were: cultivating, live stock, trading food commodities, casual labour or formal job, shop or small scale commerce, others (specify), no income. Food distribution given by WFP was not counted as an income.

11

• Family size : this was considered as the number of people composing the family. • Number of children under 5 years : this was the number of children under 5 years (between 0

month and 5 years) that were in the family. • Head of household : the family was asked if both parents were staying together or if it was a

single parent family. The sex and the age of the head of household were also recorded. • Childbirth : the mother was asked :

where the child was born (tukul / home, hospital, health centre, health post, any other place or if she did not know). who attended to the mother during the delivery (a qualified health staff (nurse or doctor), a TBA (Traditional Birth Attendant) or CHW (Community Health Worker), an unqualified person, or if the mother gave birth alone, or if she did not remember).

4. Indicators and guidelines used

4.1. Weight for height index

For the children, acute malnutrition rates were estimated from the weight for height (W/H) index values combined with the presence of oedema. The W/H indexes are compared with NCHS1 references. W/H indexes were expressed both in Z-scores and in percentage of the median. The expression in Z score has true statistical meaning and allows inter-study comparison. The percentage of the median on the other hand is commonly used to identify eligible children for feeding programmes and both will be reported. Guidelines for the results expressed in Z Score :

• Severe malnutrition is defined by WFH < -3 SD and/or existing bilateral oedema on the lower limbs of the child • Moderate malnutrition is defined by WFH < -2 SD and >= -3 SD and no oedema.

Guidelines for the results expressed in percentage according to the median of reference : • Severe malnutrition is defined by WFH < 70 % and/or existing bilateral oedema on the lower limbs • Moderate malnutrition is defined by WFH < 80 % and >= 70 % and no oedema.

4.2. Mid upper arm circumference (MUAC)

The weight for height index is the most appropriate index to quantify wasting in the population in emergency situations where acute forms of malnutrition are the predominant pattern. However the mid-upper arm circumference (MUAC) is a useful tool for rapid screening of children at a higher risk of mortality. The MUAC is only taken for children taller than 75 cm. The guidelines used vary according to authors. The commonly accepted values are as follows :

• 125 mm <= MUAC < 135 mm : moderate malnutrition • MUAC < 125 mm : severe malnutrition

Or, for ACF-USA The cut-off values are as follow:

• 110 mm <= MUAC < 120 mm : moderate malnutrition • MUAC < 110 mm : severe malnutrition

5. Field work Nine teams of two people each were recruited from the AHA community health workers and LWF staff. One field officer or one health staff, from ACF-USA, AHA, LWF, UNHCR and WFP supervised each team. The supervisors received one day of specific training on nutritional surveys and their work as supervisors

1 NCHS: National Centre for Health Statistics (1977) NCHS growth curves for children birth-18 years. United States. Vital Health Statistics. 165, 11-74.

12

was clearly explained to them. Then, the supervisors and surveyors were trained on survey methodology and measurements for three days: - theoretical training on the different methodologies and the anthropometric measurements (one day), - practical exercise and normalisation test for the measurements (one day), preliminary survey (half a day), - general organisation of the survey (half a day). For the cluster methodology, one or two clusters per team per day were the set standard. For the systematic sampling methodology, 20 to 35 children per team per day were the set standard. A total of 6 days was necessary to complete these surveys. 6. Analysis of the data The data processing and analysis were carried out with EPI-INFO 5.0 programme. The calculation and the analysis of the anthropometrical indicators were done with the EPINUT computer software.

13

RESULT

1. Settlements receiving 80 - 100% of cereals These were : Olua, Agojo East, Maaji 1 to 5, Mongola Esia, Dubaju, Umwia 2, Oliji, Magburu 2, Alere 1 and Robidire. The final sample consists of 900 children aged from 6 to 59 months old. No data has been discarded due to aberrant values. The analysis of the results is therefore focused on 900 children. 1.1. Distribution by age and sex

Table 5 DISTRIBUTION BY AGE AND BY SEX

AGE

(In months) BOYS GIRLS Total Sex

ratio N % N % N %

06 – 17 109 51.2% 104 48.8% 213 23.7% 1.05 18 – 29 129 51.8% 120 48.2% 249 27.7% 1.07 30 – 41 125 53.6% 108 46.4% 233 25.6% 1.16 42 – 53 77 52.7% 69 47.3% 146 16.2% 1.12 54 – 59 37 62.7% 22 37.3% 59 6.6% 1.68

Total 477 53.0% 423 47.0% 900 100.0% 1.13

54-59

42-53

30-41

18-29

06-17

-50% -40 % -30 % -2 0% -10% 0 % 10% 2 0% 3 0% 40 % 50 % 6 0% 70%

54-59

42-53

30-41

18-29

06-17

Figure 1

Distribution by age and sexSettlements 80 - 100%, April 1998

Girls

Boys

The distribution by sex shows an imbalance : there are more boys than girls, especially for the age class 54 - 59 months. Two possibilities could explain this imbalance : either it is a real fact in the population, or the selection of samples was not correctly done [Table 5, Figure 1]. A nutritional survey was done by ACF in Maaji in December 1997 by different teams and the same imbalance was observed. As 33% of the children in our sample are from Maaji, it can be suggested that this imbalance is a real reflection of this population.

14

54-59

42-53

30-41

18-29

06-17

0% 5% 10% 15% 20% 25% 30%

54-59

42-53

30-41

18-29

06-17

Figure 2 Age distribution

Settlements 80 - 100%, April 1998

Age distribution shows an imbalance : the age groups 42-53 months and 54-59 months are under represented. This is a fact that has been observed in all the Sudanese refugee settlements in North Uganda during these surveys and also during the survey done in Maaji in December 97. It could be a demographic reality in this refugee population. [Table 5, Figure 2]. It also has to be noted that the age given by the parents is very often rather approximated. 1.2. Nutritional anthropometric analysis The results of the anthropometric analysis have been obtained by using the weight for height index expressed in Z Score and in the percentage of the median of the reference population.

1.2.1. Distribution of malnutrition in Z-Score

Table 6 WEIGHT FOR HEIGHT DISTRIBUTION BY AGE

In Z-SCORE

AGE (In months)

< -3 SD ≥ -3 SD & < - 2 SD ≥ -2 SD Oedema

N n % n % n % n % 06-17 213 4 1.9% 14 6.6% 193 90.6% 2 0.9% 18-29 249 0 0.0% 16 6.4% 233 93.6% 0 0.0% 30-41 233 0 0.0% 8 3.4% 224 96.1% 1 0.4% 42-53 146 0 0.0% 0 0.0% 146 100.0% 0 0.0% 54-59 59 0 0.0% 0 0.0% 59 100.0% 0 0.0%

TOTAL 900 4 0.4% 38 4.2% 855 95.0% 3 0.3%

15

Table 7 WEIGHT FOR HEIGHT vs. OEDEMA

< -2 SD ≥ -2 SD Marasmus/Kwashiorkor Kwashiorkor YES 1 0.1% 2 0.2% Oedema Marasmus Normal NO 42 4.7% 855 95.0%

F ig u r e 3

Z -sc o r e d i str i b u ti o n - W e i g h t fo r H e i g h t S e t t le m e n t 8 0 - 1 0 0 % , A p r il 1 9 9 8

02468

1 01 21 41 61 82 02 2

- 5 - 4 - 3 - 2 - 1 0 1 2 3 4 5

R e f e r e n c e

S e xC o m b in e d

There is no significant discrepancy on the left of the curve for the sample : the proportion of the children included between -3 SD and -1 SD is slightly more important in our sample than in the reference population. This doesn't show a precarious nutritional state.

Table 8 GLOBAL AND SEVERE ACUTE MALNUTRITION BY AGE GROUP In Z-SCORE

6-59 months (n = 900) 6-29 months (n = 462)

Global acute malnutrition 5.0% [3.2% ↔ 7.6%] 7.8% [4.8% ↔ 12.3%] Severe acute malnutrition 0.8% [0.2% ↔ 2.3%] 1.3% [0.3% ↔ 4.1%]

The rate of global malnutrition is slightly high, especially for children aged from 6 to 29 months old (Table 8). Statistical tests reveal that there is an extremely significant difference (p<0.05) in malnutrition rates between the two age groups. The uncorrected chi-square is equal to 15.58, the relative risk (RR) is 3.79 (95% confidence interval : 1.85 < RR < 7.78). This means that children less than 30 months have a risk multiplied by 3.79 of being malnourished, compared to children aged from 30 to 59 months old. Wasting affects children aged from 6 to 29 months old more than those aged from 30 to 59 months old.

16

Table 9 NUTRITIONAL STATUS BY SEX in Z-SCORE

Nutritional status Definition Boys Girls

n % n % Severe malnutrition Weight for Height < -3 or oedema 5 1.0% 2 0.5%

Moderate malnutrition -3 ≤ Weight for Height < -2 and no oedema

15 3.1% 23 5.4%

Normal -2 ≤ Weight for Height and no oedema 457 95.8% 398 94.1%

TOTAL 477 423

There is no significant difference as regards the nutritional status (in Z score) between boys and girls (p>0.05). The weight for height means are -0.477 (standard deviation : 0.869) for the boys and -0.485 (standard deviation : 0.959) for the girls and are not statistically different.

F igu r e 4Z-sc o r e d i str i b u ti o n - W e i g h t fo r H e i g h t

S e t t le m e n t 8 0 - 1 0 0 % , A p r il 1 9 9 8

02468

1 01 21 41 61 82 02 2

- 5 - 4 - 3 - 2 - 1 0 1 2 3 4 5

R e f e r e n c eB o y s

G ir ls

1.2.2. Distribution of malnutrition in percentage of the median

The expression in percentage of the median is commonly used as admission criteria in feeding centres.

Table 10 WEIGHT FOR HEIGHT DISTRIBUTION BY AGE in percentage of the median

AGE

(In months) < 70% ≥ 70% & < 80% ≥ 80% Oedema

N n % N % n % n % 06-17 213 1 0.5% 10 4.7% 200 93.9% 2 0.9% 18-29 249 0 0.0% 7 2.8% 242 97.2% 0 0.0% 30-41 233 0 0.0% 5 2.1% 227 97.4% 1 0.4% 42-53 146 0 0.0% 0 0.0% 146 100.0% 0 0.0% 54-59 59 0 0.0% 0 0.0% 59 100.0% 0 0.0%

TOTAL 900 1 0.1% 22 2.4% 874 97.1% 3 0.3%

17


< 80% ≥ 80% Marasmus/Kwashiorkor Kwashiorkor YES 1 0.1% 2 0.2% Oedema Marasmus Normal NO 23 2.6% 874 97.1%

Table 12

GLOBAL AND SEVERE ACUTE MALNUTRITION BY AGE GROUP In percentage of the median

6-59 months (n = 900) 6-29 months (n = 462)

Acute global malnutrition 2.9% [1.6% ↔ 5.1%] 4.3% [2.2% ↔ 8.1%] Acute severe malnutrition 0.4% [0.0% ↔ 1.8%] 0.6% [0.0% ↔ 3.1%]

The rate of global malnutrition is low, however it is higher for children from 6 to 29 months (Table 12). Statistical tests reveal that there is an extremely significant difference (p<0.05) in malnutrition rates between the two age groups. The uncorrected chi-square is equal to 7.02, the relative risk (RR) is 3.16 (95% confidence interval : 1.28 < RR < 7.80). This means that children less than 30 months have a risk multiply by 3.16 of being malnourished, compared to children aged from 30 to 59 months old. Wasting affects children aged from 6 to 29 months old more than those aged from 30 to 59 months old.

Table 13 NUTRITIONAL STATUS BY SEX

In percentage of the median

Nutritional status Definition Boys Girls n % n %

Severe malnutrition Weight for Height < 70% or oedema 3 0.6% 1 0.2% Moderate malnutrition 70 ≤ Weight for Height < 80 and no

oedema 6 1.3% 16 3.8%

Normal 80% ≤ Weight for Height and no oedema 468 98.1% 406 96.0%

TOTAL 477 423

There is a significant difference concerning the nutritional status (in percentage of the median) between the boys and the girls (p<0.05) : more girls are affected by malnutrition (global acute malnutrition in percentage of the median = 4.0%, compared to the one for the boys : 2.1%). The uncorrected chi-square is equal to 5.57, the relative risk (RR) is 2.73 (95% confidence interval : 1.14 < RR < 6.51). This means that girls have a risk multiply by 2.73 of being malnourished, compared to boys. The weight for height means are 96.2 (standard deviation : 8.7) for the boys and 95.8 (standard deviation : 9.3) for the girls and are not statistically different. The study of the relation between boys and girls nutritional status by the age group (6-29 and 30-59 months) shows a significant relationship for the children aged from 30 to 59 months.

18

Table 14 NUTRITIONAL STATUS BY SEX AND AGE GROUP In percentage of the median

6 - 29 months 30 - 59 months

Nutritional status Boys Girls Boys Girls n % n % n % n %

Severe malnutrition 2 0.8% 1 0.4% 1 0.4% 0 0.0% Moderate malnutrition 6 2.5% 11 4.9% 0 0.0% 5 2.5%

Normal 230 96.6% 212 94.6% 238 99.6% 194 97.5%

TOTAL 238 224 239 199

More girls are affected by malnutrition, especially for the age group 30-59 months, where the difference is statistically significant (p<0.05). For the children aged from 6 to 29 months old, the weight for height means are 92.8 (standard deviation : 8.3) for the boys and 93.9 (standard deviation : 9.6) for the girls and are not statistically different. For the children aged from 30 to 59 months old, the weight for height means are 99.6 (standard deviation : 7.8) for the boys and 97.9 (standard deviation : 8.5) for the girls and are statistically different (p<0.05).

1.2.3. Distribution of Mid Upper Arm Circumference by age

Table 15

MUAC DISTRIBUTION BY HEIGHT

MUAC HEIGHT N < 110 mm 110<=MUAC<120 >= 120 mm

N % N % N % < 75 cm 218 6 2.7% 22 10.1% 190 87.2%

75 <= H < 90 cm 417 1 0.2% 6 1.4% 410 98.3% >= 90 cm 265 0 0.0% 0 0.0% 265 100.0% TOTAL 900 7 0.8% 28 3.1% 865 96.1%

The mean for the whole sample is 144.6 mm with a standard deviation 13.9 mm. As research studies showed that MUAC is more relevant for children aged 12 to 47 months (corresponding to a height between 75 and 90 cm), it is interesting to also have the results for this specific height range : mean = 144.3 mm and standard deviation = 11.6 mm. 1.3. Measles vaccination

Table 16 MEASLES IMMUNISATION for children age 9 to 59 months

N % Measles vaccination proved by the card, and done at 9 months or after 345 40.4% Measles vaccination proved by the card and done before 9 months 131 15.3% Measles vaccination done according to the mother (card absent) 242 28.3% No vaccination on the card 59 6.9% No vaccination, according to the mother (card absent) 77 9.1%

The measles coverage is low according to the number of children holding a vaccination card and vaccinated at 9 months or after. According to the mother or to the card 68.7% of the children have been vaccinated, but according to the vaccination card, only 40.4% above or equal 9 months (age of vaccination against measles) are really protected against this disease. However, for those vaccinated before 9 months, this could be due to a mistake in recording of the birthdate or the vaccination date on the card.

19

The significant number of children without cards (37.4%) does not necessarily mean that these children have not been vaccinated. Therefore we cannot be entirely sure of the results. 1.4. Additional information

1.4.1. Date of arrival in Uganda

Table 17

DATE OF ARRIVAL IN UGANDA

Arrival before 88 88 - 92 in 93 in 94 in 95 in 96 in 97 in 98 n 16 509 115 209 8 21 13 3 % 1.8% 56.9% 12.9% 23.3% 0.9% 2.3% 1.5% 0.3%

43.5% of the refugees arrived in March 1989.

1.4.2. Number of meals

Table 18 NUMBER OF MEALS

Number of meals N % < 2 272 30.5% = 2 345 38.7% > 2 274 30.8% The mean is 2.01 meals taken the day before, with a standard deviation 0.82.

1.4.3. Sources of income

Table 19 SOURCES OF INCOME

SOURCES OF INCOME N %

Cultivating 385 43.0% Cultivating and casual labour or formal job 2 0.2% Cultivating and small commerce 5 0.6% Live stock 3 0.3% Trading food commodities 20 2.2% Trading food commodities and small commerce 1 0.1% Casual labour or formal job 185 20.7% Formal job and small commerce 2 0.2% Shop or small scale commerce 119 13.3% Borrow 1 0.1% No income 172 19.3% 43.8% of the families are cultivating their farm plot and selling part of their production, 20.9% are casual labourers or have formal jobs, 14.2% are engaged in small scale commerce and 19.3% have no source of income. Small-scale commerce mainly involves selling firewood, charcoal, local brew or even part of food ration.

20

1.4.4. Information on the family

Table 20: FAMILY SIZE Table 21: CHILDREN UNDER 5 FAMILY SIZE N % CHILDREN N %

<= 3 67 7.5% 1 452 50.4% 4 - 6 477 53.2% 2 356 39.7% 7 - 9 283 31.5% > 2 89 9.9%

>= 10 70 7.8% mean : 1.61, standard deviation : 0.71 Mean : 6.26, standard deviation : 2.21

Presence of the parents: 81.0% of the families have both parents and 19.0% are single parents. 39.6% of the heads of household are women and 60.4% are men. 0.6% of the heads of household have less than 20 years, 8.3% have between 20 and 24 years, 28.5% between 25 and 29 years, 43.4% between 30 and 39 years and 19.2% have more than 40 years. The mean is 33.0 with a standard deviation of 8.4.

1.4.5. Child birth

Table 22 BIRTH PLACE

Table 23 BIRTH ATTENDANT

PLACE N % ATTENDANT N % Tukul 501 56.2% Qualified health staff 257 28.8%

Hospital 47 5.3% TBA - CHW 511 57.4% Health centre 312 35.0% Unqualified person 110 12.3% Health post 8 0.9% Woman birthed alone 13 1.5%

Other* 24 2.7%

* Mainly “on the road” Only 41.2% of the women used the health facilities for delivery, 57.4% of the births were attended by a TBA or a CHW and only 28.8% by qualified health staff. This low percentage use of health facilities for delivering can be explained by the fact that mothers don't want to expose themselves to other people during labour. Moreover, they have no time to go to the health centre when the labour pains have started. It seems they use health centres only in case of complication and prefer to seek help from TBA or CHW. 1.5. Relationship between child's nutritional status and other factors NOTE : Some relationships between malnutrition and other factors appeared during the analysis of the survey. However, this type of survey is not designed to show malnutrition "risks factors". To show causative relations, a case control study will be needed, involving a different way of calculating the sample size. Therefore, these are only suggested relations, not established relations. For all the following tables, a child is malnourished if his weight for height index is less than -2 in Z score or if he has bilateral oedema. A child is not considered as malnourished if his W/H index is more or equal to -2 and if he has no bilateral oedema.

21

1.5.1. Relation between malnutrition, age and meals' number Table 24

NUMBER OF MEALS IN RELATION TO AGE GROUP

NUMBER OF MEALS AGE GROUP N < 2 = 2 > 2

N % N % N % 6 - 29 months 457 158 34.6% 166 36.3% 133 29.1% 30 - 59 months 434 114 26.3% 179 41.2% 141 32.5%

TOTAL 891 272 30.5% 345 38.7% 274 30.8%

The meals number's means are 1.95 (with a standard deviation of 0.85) for children aged from 6 to 29 months old and 2.07 (standard deviation : 0.78) for children aged from 30 to 59 months old, and are statistically different (p<0.05). Children aged from 6 to 29 months old took less meals the day before. This could be put in relation with the nutritional status of this age group, which is worse than that of the 30 - 59 months age group. Malnutrition rate is higher for children aged from 6 to 29 months old compared to that of children aged from 30 to 59 months old. Moreover meals' number differs between these two age groups. For these two reasons it is more interesting to study possible relations between malnutrition and meals number for each age group separately.

Table 25 NUTRITIONAL STATUS IN RELATION TO THE NUMBER OF MEALS

for CHILDREN AGED FROM 6 TO 29 MONTHS OLD

NUMBER OF MEALS < 2 = 2 > 2 NUTRITIONAL STATUS N

N % N % N % Malnourished 36 28 11.4% 14 8.4% 4 3.0%

No malnourished 421 140 88.6% 152 91.6% 129 97.0% TOTAL 457 158 166 133

This table shows a significant difference between the children having less than 2 meals, those having 2 meals and those having more than 2 meals (p<0.05) : the percentage of malnourished children decreases when the number of meals increases. If we compare the children having less than 2 meals and the other children, the uncorrected chi-square is equal to 13.73 (p<0.01), the relative risk (RR) is 2.77 (95% confidence interval : 1.58 < RR < 4.85). This means that children having less than 2 meals have a risk multiplied by 2.77 of being malnourished, compared to children having 2 or more meals. If we compare the children having more than 2 meals and the other children, the uncorrected chi-square is equal to 9.81 (p<0.01), the relative risk (RR) is 0.24 (95% confidence interval : 0.09 < RR < 0.65). This means that children having more than 2 meals have a risk divided by 4.17 (=1/0.24) of being malnourished, compared to children having 2 or less meals. Moreover, malnourished children took less meals the day before, than the well nourished children. The meals' number means are 1.58 (with a standard deviation of 0.73) for the malnourished children, and 1.98 (standard deviation : 0.86) for the no malnourished children. These two means are statistically different (p<0.01). But we have to be careful because several other factors interfere with meals number, like breast-feeding and meals quality. Unfortunately no data concerning breast-feeding and meals quality are available. Supplementary studies if they are to be carried out should include number of meals and their quality (quantitative and qualitative), breast-feeding, feeding time in the day, to be able to confirm or not the influence of the meals number on the nutritional status.

22



NUMBER OF MEALS < 2 = 2 > 2 NUTRITIONAL STATUS N



This table doesn't show a significant difference between the children having less than 2 meals, those having 2 meals and those having more than 2 meals (p>0.05). The meals' number means are 2.44 (with a standard deviation of 1.01) for the malnourished children, and 2.06 (standard deviation: 0.77) for the no malnourished children. These two means aren't statistically different (p>0.05). But, as explained before, we have to be careful : several other factors interfere with meals number, like meals quality thus supplementary studies are necessary.

1.5.2. Relationship between malnutrition and family composition

Table 27

MEANS OF DIFFERENT FACTORS FOR EACH NUTRITIONAL STATUS GROUP

NUTRITIONAL STATUS Family size Number of children under 5 years

Age of the head of household

N mean SD N mean SD N mean SD Malnourished 45 5.76 1.73 42 1.53 0.66 44 32.25 6.07

No malnourished 852 6.29 2.23 852 1.61 0.71 824 33.00 8.46 STATISTICALLY not different not different not different

SD = Standard deviation There is no statistical difference between the two nutritional status groups. However it seems that malnourished children belong to smaller families with less children under 5 years. One hypothesis could be that families with more people could generate more incomes or produce more food. It is also possible that when the family is bigger, there are more people to take care of the children under 5 years when the mother is absent.

Table 28 NUTRITIONAL STATUS IN RELATION TO SOCIAL STATUS

OF THE FAMILY HEAD

NUTRITIONAL PARENTS PRESENCE SEX OF FAMILY HEAD STATUS Both parents Single parent Woman Man

n % n % n % n % Malnutrition 38 5.2% 7 4.1% 19 5.4% 26 4.8%

Normal 689 94.8% 163 95.9% 336 94.6% 515 95.2% TOTAL 727 170 255 541

In this sample it seems there is no relationship between type of family head and malnutrition. But as explained before, we can't generalise.

23

2. Settlements receiving 50% of cereals These are : Alere 2, Mongola Adidi, Mongola Aliwara, Mongola Opi, Kolididi, Nyeu, Keyo 1 to 3 and Arra. The final sample consists of 810 children aged from 6 to 59 months old. No data has been discarded due to aberrant values. The analysis of the results is therefore focused on 810 children. 2.1. Distribution by age and sex


AGE

(In months) BOYS GIRLS Total Sex ratio

N % N % N % 06 - 17 114 54.3% 96 45.7% 210 25.9% 1.19 18 - 29 111 53.9% 95 46.1% 206 25.4% 1.17 30 - 41 97 48.5% 103 51.5% 200 24.7% 0.94 42 - 53 68 47.6% 75 52.4% 143 17.7% 0.91 54 - 59 24 47.1% 27 53.9% 51 6.3% 0.89 Total 414 51.1% 396 48.9% 810 100.0% 1.05

54-59

42-53

30-41

18-29

06-17

-60% -50% -40% -30% -20% -10% 0% 10% 20% 30% 40% 50% 60%

54-59

42-53

30-41

18-29

06-17

Figure 5 Distribution by age and sex

Settlements 50%, April 1998GirlsBoys

The distribution by sex does not show a significant imbalance. This fact allows us to validate the selection of samples. [Table 29, Figure 5]

24

54-59

42-53

30-41

18-29

06-17

0% 5% 10% 15% 20% 25% 30%

54-59

42-53

30-41

18-29

06-17


Settlements 50%, April 1998

Age distribution shows an imbalance : the age groups 42-53 months and 54-59 months are under represented. This is a fact that has been observed in all the Sudanese refugee settlements in North Uganda during these surveys and also during the survey done in Maaji in December 97. It could be a demographic reality in this refugee population. [Table 29, Figure 6]. It also has to be noted that the age given by the parents is very often rather approximated. 2.2. Nutritional anthropometric analysis The results of the anthropometric analysis have been obtained by using the indication weight for height expressed in Z-Score and in the percentage of the median of the reference population.

2.2.1. Distribution of malnutrition in Z-Score Table 30

WEIGHT FOR HEIGHT DISTRIBUTION BY AGE In Z-SCORE

AGE

(In months) < -3 SD ≥ -3 SD & < - 2 SD ≥ -2 SD Oedema

N n % n % n % n % 06-17 210 2 1.0% 24 11.4% 183 87.1% 1 0.5% 18-29 206 2 1.0% 23 11.2% 180 87.4% 1 0.5% 30-41 200 1 0.5% 4 2.0% 195 97.5% 0 0.0% 42-53 143 0 0.0% 1 0.7% 142 99.3% 0 0.0% 54-59 51 0 0.0% 2 3.9% 49 96.1% 0 0.0%

TOTAL 810 5 0.6% 54 6.7% 749 92.5% 2 0.2%



25

Figure 7Z-score distribution - Weight for Height

Settlement 50%, April 1998

02468

10121416182022

-5 -4 -3 -2 -1 0 1 2 3 4 5

Reference

SexCombined

There is a significant discrepancy on the left of the curve for the sample : the proportion of the children included between -3 SD and -1 SD is more important in our sample than in the reference population. This fact shows a precarious nutritional state.


6-59 months (n = 810) 6-29 months (n = 416)


The rate of global malnutrition is high, especially for children aged from 6 to 29 months (Table 32). Statistical tests reveal that there is an extremely significant difference (p<0.01) in malnutrition rates between the two age groups. The uncorrected chi-square is equal to 33.33, the relative risk (RR) is 6.27 (95% confidence interval : 3.02 < RR < 13.03). This means that children less than 30 months have a risk multiply by 6.27 of being malnourished, compared to children aged from 30 to 59 months old. Wasting affects the children aged from 6 to 29 months old more than those aged from 30 to 59 months old.

Table 33 NUTRITIONAL STATUS BY SEX In Z-SCORE

Boys Girls Nutritional status Definition n % n %

Severe malnutrition Weight for Height < -3 or oedema 2 0.5% 5 1.3% Moderate malnutrition -3 ≤ Weight for Height < -2 and no oedema 31 7.5% 23 5.8%

Normal -2 ≤ Weight for Height and no oedema 381 92.0% 368 92.9% TOTAL 414 396

There is no significant difference as regards the nutritional status (in Z score) between boys and girls (p>0.05). The weight for height means are -0.597 (standard deviation : 0.938) for the boys and -0.535 (standard deviation : 1.016) for the girls and are not statistically different.

26



02468

10121416182022

-5 -4 -3 -2 -1 0 1 2 3 4 5

Reference

Boys

Girls

2.2.2. Distribution of malnutrition in percentage of the median The expression in percentage of the median is commonly used as admission criteria in feeding centres.

Table 34 WEIGHT FOR HEIGHT DISTRIBUTION BY AGE in percentage of the median

AGE

(In months) < 70% ≥ 70% & < 80% ≥ 80% Oedema

N n % N % n % n % 06-17 210 0 0.0% 16 7.6% 193 91.9% 1 0.5% 18-29 206 0 0.0% 12 5.8% 193 93.7% 1 0.5% 30-41 200 1 0.5% 2 1.0% 197 98.5% 0 0.0% 42-53 143 0 0.0% 1 0.7% 142 99.3% 0 0.0% 54-59 51 0 0.0% 0 0.0% 51 100.0% 0 0.0%

TOTAL 810 1 0.1% 31 3.8% 776 95.8% 2 0.2%



27

Table 36 GLOBAL AND SEVERE ACUTE MALNUTRITION BY AGE GROUP In percentage of the median

6-59 months (n = 810) 6-29 months (n = 416)

Acute global malnutrition 4.2% [2.5% ↔ 6.8%] 7.2% [4.2% ↔ 11.9%] Acute severe malnutrition 0.4% [0.0% ↔ 1.8%] 0.5% [0.0% ↔ 3.1%]

The rate of global malnutrition is low, however it is higher for children aged from 6 to 29 months old (Table 36). Statistical tests reveal that there is an extremely significant difference (p<0.01) in malnutrition rates between the two age groups. The uncorrected chi-square is equal to 19.32, the relative risk (RR) is 7.10 (95% confidence interval : 2.53 < RR < 19.98). This means that children less than 30 months have a risk multiplied by 7.10 of being malnourished, compared to children aged from 30 to 59 months old. Wasting affects the children aged from 6 to 29 months old more than those aged from 30 to 59 months old.

Table 37 NUTRITIONAL STATUS BY SEX In percentage of the median

Nutritional status Definition Boys Girls

n % n % Severe malnutrition Weight for Height < 70% or oedema 0 0.0% 3 0.8%

Moderate malnutrition 70 ≤ Weight for Height < 80 and no oedema 16 3.9% 15 3.8% Normal 80% ≤ Weight for Height and no oedema 398 96.1% 378 95.5% TOTAL 414 396

There is no significant difference concerning the nutritional status (in percentage of the median) between the boys and the girls (p>0.05). The weight for height means are 95.1 (standard deviation: 8.5) for the boys and 95.4 (standard deviation: 9.2) for the girls and are not statistically different.


Table 38


MUAC < 110 mm 110<=MUAC<120 >= 120 mm HEIGHT N

N % N % N % < 75 cm 188 3 1.6% 24 12.8% 161 85.6%

75 <= H < 90 cm 391 1 0.3% 6 1.5% 384 98.2% >= 90 cm 231 0 0.0% 0 0.0% 231 100.0% TOTAL 810 4 0.5% 30 3.7% 776 95.8%

The mean for the whole sample is 144.2 mm with a standard deviation 13.7 mm. As research studies showed that MUAC is more relevant for children aged 12 to 47 months (corresponding to a height between 75 and 90 cm), it is interesting to also have the results for this specific height range : mean = 143.8 mm and standard deviation = 12.3 mm.

28

2.3. Measles vaccination Table 39

MEASLES IMMUNISATION for children age 9 to 59 months


The measles coverage is low according to the number of children holding a vaccination card and vaccinated at 9 months or after. But according to the mother and to the card 67.3% of children hav been vaccinated. This seems to be a better coverage, although still low, considering that the mother are sure that the children are effectively vaccinated against measles. However, the significant number without cards (30.2%) is worrying. 2.4. Additional information

2.4.1. Date of arrival in Uganda Table 40



41.2% of the refugees arrived in in March 1989.

2.4.2. Number of meals Table 41

NUMBER OF MEALS

Number of meals N % < 2 134 16.9% = 2 327 41.1% > 2 334 42.0%

The mean is 2.25 meals taken the day before, with a standard deviation 0.81.


Table 42 SOURCES OF INCOME

SOURCES OF INCOME N %

Cultivating 287 36.2% Cultivating and living stock 1 0.1% Cultivating and trading food commodities 7 0.9% Cultivating and casual labour or formal job 2 0.3% Cultivating, casual labour and small commerce 1 0.1% Cultivating and small commerce 24 3.0% Live stock 2 0.3% Trading food commodities 15 1.9% Casual labour or formal job 151 19.0% Formal job and small commerce 4 0.5% Shop or small scale commerce 191 24.1% No income 108 13.6%

29

40.6% of the families are cultivating their farm plot and selling part of their production, 19.9% are casual labourers or have formal jobs, 27.2% are engaged in small scale commerce and 13.6% have no source of income. Small-scale commerce mainly involves selling firewood, charcoal, local brew or even part of food ration.


Table 43

FAMILY SIZE Table 44

NUMBER OF CHILDREN UNDER 5 FAMILY SIZE N % CHILDREN N %

<= 3 63 7.9% 1 421 53.0% 4 - 6 412 51.9% 2 326 41.0% 7 - 9 245 30.6% > 2 48 6.0%

>= 10 74 9.3% Mean : 6.22, standard deviation : 2.24 mean : 1.52, standard deviation : 0.63

Presence of the parents : 83.8% of the families have both parents and 16.2% are single parent families. 39.6% of the heads of household are women and 60.4% are men. 0.4% of the heads of household have less than 20 years, 10.9% have between 20 and 24 years, 25.9% between 25 and 29 years, 42.2% between 30 and 39 years and 20.6% have more than 40 years. The mean is 33.3 with a standard deviation of 9.2.

2.4.5. Child birth





Other* 6 0.8%

* Mainly «on the road» Only 37.2% of the women used the health facilities for delivery. 49.1% of the births were attended by a TBA or a CHW and 27.5% by a qualified health staff. 2.5. Relation between child's nutritional status and other factors NOTE : Some relationships between malnutrition and other factors appeared during the analysis of the survey. However, this type of survey is not designed to show malnutrition "risks factors". To show causative relations, a case control study will be needed, involving a different way of calculating the sample size. Therefore, these are only suggested relations, not established relations. For all the following tables, a child is malnourished if his weight for height index is less than -2 in Z score or if he has bilateral oedema. A child is not considered as malnourished if his W/H index is more or equal to -2 and if he has no bilateral oedema.

30

2.5.1. Relation between malnutrition, age and meals' number

Table 47 NUMBER OF MEALS IN RELATION TO AGE GROUP

NUMBER OF MEALS

AGE GROUP N < 2 = 2 > 2 N % N % N %

6 - 29 months 408 82 20.1% 153 37.5% 173 42.4% 30 - 59 months 387 52 13.4% 174 45.0% 161 41.6%

TOTAL 795 134 16.9% 327 41.1% 334 42.0%

The meals number's means are 2.22 (with a standard deviation of 0.89) for children aged from 6 to 29 months old and 2.28 (standard deviation : 0.72) for children aged from 30 to 59 months old, and aren't statistically different (p>0.05). There is a significant difference between the two age groups (table 47 - p<0.05) : there is more children aged from 6 to 29 months old who took less than 2 meals the day before. This could be related to the nutritional status of this age group, which is worse than that of the 30 - 59 months age group.



NUMBER OF MEALS < 2 = 2 > 2 NUTRITIONAL

STATUS N N % N % N %

Malnourished 52 12 14.6% 21 13.7% 19 11.0% No malnourished 356 70 85.4% 132 86.3% 154 89.0%

TOTAL 408 82 153 173 This table doesn't show a significant difference between the children having less than 2 meals, those having 2 meals and those having more than 2 meals (p>0.05). The study of the relative risk doesn't show significant risk (they are all close to the value 1). However the percentage of malnourished children tends to decrease when the number of meals increases. The meals' number means are 2.08 (with a standard deviation of 1.03) for the malnourished children, and 2.24 (standard deviation: 0.87) for the no malnourished children. These two means aren't statistically different (p>0.05).



NUMBER OF MEALS < 2 = 2 > 2 NUTRITIONAL

STATUS N N % N % N %

Malnourished 7 1 1.9% 5 2.9% 1 0.6% No malnourished 380 51 98.1% 169 97.1% 160 99.4%

TOTAL 387 52 174 161 This table doesn't show a significant difference between the children having less than 2 meals, those having 2 meals and those having more than 2 meals (p>0.05). The meals' number means are 2.00 (with a standard deviation of 0.58) for the malnourished children, and

31

2.28 (standard deviation: 0.72) for the no malnourished children. These two means aren't statistically different (p>0.05).

2.5.2. Relationship between malnutrition and family composition Table 50


Family size Number of children under 5 years

Age of the head of household NUTRITIONAL STATUS


No malnourished 735 6.28 2.24 736 1.55 0.63 668 33.23 9.04 STATISTICALLY different (p<0.01) different (p<0.05) not different

SD = Standard deviation Significant difference appears : malnourished children are from smaller families. In fact the family size means are 5.44 for malnourished children and 6.28 for no malnourished children. Concerning the number of children under 5, it is lower for malnourished children. It is not what it is expected. But the differences are very low and it is not possible to conclude about a real relationship between these different factors and malnutrition.


OF THE FAMILY HEAD

PARENTS PRESENCE SEX OF THE FAMILY HEAD Both parents Single parent Woman Man NUTRITIONAL STATUS n % n % n % n %

Malnutrition 49 7.4% 10 7.8% 27 8.6% 32 6.7% Normal 617 92.6% 119 92.2% 288 91.4% 448 93.3% TOTAL 666 129 315 480

In this sample it appears there is no relationship between type of family head and malnutrition.

32

3. Settlements receiving 0% of cereals These are : Okusijoni, Obilonkogo, Agojo South, Nyimanzi 1 and 2, Umwia 1, Baratuku, Elema, Ibiaworo and Magburu 1. The final sample consists of 390 children aged from 6 to 59 months old. No data has been discarded due to aberrant values. The analysis of the results is therefore focused on 390 children. 3.1. Distribution by age and sex


AGE

(In months) BOYS GIRLS Total Sex ratio

N % N % N % 06 - 17 46 50.0% 46 50.0% 92 23.6% 1.00 18 - 29 60 57.7% 44 42.3% 104 26.7% 1.36 30 - 41 42 45.7% 50 54.3% 92 23.6% 0.84 42 - 53 33 45.8% 39 54.2% 72 18.5% 0.85 54 - 59 12 40.0% 18 60.0% 30 7.7% 0.67 Total 193 49.5% 197 50.5% 390 100.0% 0.98

54-59

42-53

30-41

18-29

06-17

-60% -50% -40% -30% -20% -10% 0% 10% 20% 30% 40% 50% 60%

54-59

42-53

30-41

18-29

06-17

Figure 9 Distribution by age and sex

Settlements 0%, April 1998GirlsBoys

The distribution by sex does not show a significant imbalance. This fact allows us to validate the selection of samples. [Table 52, Figure 9]

33

54-59

42-53

30-41

18-29

06-17

0% 5% 10% 15% 20% 25% 30%

54-59

42-53

30-41

18-29

06-17


Settlements 0%, April 1998

Age distribution shows an imbalance: the age groups 42-53 months and 54-59 months are under represented. This is a fact that has been observed in all the Sudanese refugee settlements in North Uganda during these surveys and also during the survey done in Maaji in December 97. It could be a demographic reality in this refugee population. [Table 52, Figure 10]. It also has to be noted that the age given by the parents is very often rather approximated. 3.2. Nutritional anthropometric analysis The results of the anthropometric analysis have been obtained by using the indication weight for height expressed in Z-Score and in the percentage of the median of the reference population.

3.2.1. Distribution of malnutrition in Z-Score Table 53

WEIGHT FOR HEIGHT DISTRIBUTION BY AGE In Z-SCORE

AGE

(In months) < -3 SD ≥ -3 SD & < - 2 SD ≥ -2 SD Oedema

N n % n % n % n % 06-17 92 1 1.1% 5 5.4% 85 92.4% 1 1.1% 18-29 104 1 1.0% 3 2.9% 100 96.2% 0 0.0% 30-41 92 1 1.1% 1 1.1% 90 97.8% 0 0.0% 42-53 72 0 0.0% 0 0.0% 72 100.0% 0 0.0% 54-59 30 0 0.0% 1 3.3% 29 96.7% 0 0.0%

TOTAL 390 3 0.8% 10 2.6% 376 96.4% 1 0.3%



34



02468

101214161820222426

-5 -4 -3 -2 -1 0 1 2 3 4 5

Reference

SexCombined

There is a significant discrepancy on the left of the curve for the sample : the proportion of the children included between -2 SD and -1 SD is more important in our sample than in the reference population. This shows a nutritional state, which can quickly deteriorate if the refugees meet some difficulties in getting food.


6-59 months (n = 390) 6-29 months (n = 196)


The rate of global malnutrition is low (Table 55). Statistical tests reveal that there is a significant difference (p<0.05) in malnutrition rates between the two age groups. The uncorrected chi-square is equal to 4.66, the relative risk (RR) is 3.63 (95% confidence interval : 1.03 < RR < 12.81). This means that children less than 30 months have a risk multiplied by 3.63 of being malnourished, compared to children aged from 30 to 59 months old. Wasting affected the children aged from 6 to 29 months old more than those aged from 30 to 59 months old.

Table 56 NUTRITIONAL STATUS BY SEX In Z-SCORE


Severe malnutrition Weight for Height < -3 or oedema 4 2.1% 0 0.0% Moderate malnutrition -3 ≤ Weight for Height < -2 and no oedema 8 4.1% 2 1.0%

Normal -2 ≤ Weight for Height and no oedema 181 93.8% 195 99.0% TOTAL 193 197

There is a significant difference as regards the nutritional status (in Z score) between boys and girls (p<0.05) : more boys are affected by the malnutrition (global acute malnutrition in Z-score = 6.2%, compared to the one of the girls : 1.0%). The uncorrected chi-square is equal to 7.62 (p<0.01), the relative risk (RR) is 6.12 (95% confidence interval : 1.39 < RR < 27.01). This means that boys have a risk multiplied by 6.12 of being malnourished,

35

compared to girls. One hypothesis could be that girls tend to spend more time than boys with their mothers. But, the proportion of girls around -1 Z-score is higher that that of the boys. The girls will be more vulnerable in case of difficulties in getting food. The weight for height means are -0.715 (standard deviation : 0.914) for the boys and -0.461 (standard deviation : 0.862) for the girls and are statistically different (p<0.05) : boys are thinner than girls.



02468

10121416182022242628

-5 -4 -3 -2 -1 0 1 2 3 4 5

Reference

Boys

Girls

The study of the relation between boys and girls nutritional status by the age group (6-29 and 30-59 months) shows a significant relation for the 6-29 months.

Table 57 NUTRITIONAL STATUS BY SEX AND AGE GROUP

In Z-score

6 - 29 months 30 - 59 months Boys Girls Boys Girls

Nutritional status

n % n % n % n % Severe malnutrition 3 2.8% 0 0.0% 1 1.1% 0 0.0%

Moderate malnutrition 7 6.6% 1 1.1% 1 1.1% 1 0.9% Normal 96 90.6% 89 98.9% 85 97.7% 106 99.1% TOTAL 106 90 87 107

More boys are affected by malnutrition, especially for the age group 6-29 months, where the difference is statistically significant (p<0.05). For this age group, the uncorrected chi-square is equal to 6.36 (p<0.05), the relative risk (RR) is 8.49 (95% confidence interval : 1.11 < RR < 65.06). This means that boys have a risk multiplied by 8.49 of being malnourished, compared to girls. For the children aged from 6 to 29 months old, the weight for height means are -0.96 (standard deviation : 0.87) for the boys and -0.63 (standard deviation : 0.76) for the girls and are statistically different (p<0.05). Boys are thinner than girls. For the children aged from 30 to 59 months old, the weight for height means are -0.42 (standard deviation : 0.88) for the boys and -0.32 (standard deviation : 0.92) for the girls and are not statistically different (p>0.05).

36

3.2.2. Distribution of malnutrition in percentage of the median The expression in percentage of the median is commonly used as admission criteria in the feeding centres.

Table 58 WEIGHT FOR HEIGHT DISTRIBUTION BY AGE In percentage of the median

< 70% ≥ 70% & < 80% ≥ 80% Oedema AGE

(In months) N n % n % n % n % 06-17 92 1 1.1% 1 1.1% 89 96.7% 1 1.1% 18-29 104 0 0.0% 2 1.9% 102 98.1% 0 0.0% 30-41 92 1 1.1% 0 0.0% 91 98.9% 0 0.0% 42-53 72 0 0.0% 0 0.0% 72 100.0% 0 0.0% 54-59 30 0 0.0% 1 3.3% 29 96.7% 0 0.0%

TOTAL 390 2 0.5% 4 1.0% 383 98.2% 1 0.3%



Table 60

GLOBAL AND SEVERE ACUTE MALNUTRITION BY AGE GROUP In percentage of the median

6-59 months (n = 390) 6-29 months (n = 196) Acute global malnutrition 1.8% [0.9% ↔ 3.5%] 2.6% [1.1% ↔ 5.6%] Acute severe malnutrition 0.8% [0.3% ↔ 2.2%] 1.0% [0.2% ↔ 3.4%]

The rate of global malnutrition is low (Table 60). Statistical test revealed that there was no significant difference in malnutrition rates between the age groups.

Table 61 NUTRITIONAL STATUS BY SEX In percentage of the median


Severe malnutrition Weight for Height < 70% or oedema 3 1.5% 0 0.0% Moderate malnutrition 70 ≤ Weight for Height < 80 and no oedema 4 2.1% 0 0.0%

Normal 80% ≤ Weight for Height and no oedema 186 96.4% 197 100.0% TOTAL 193 197

There is a significant difference as regards the nutritional status (in percentage of the median) between the boys and the girls (p<0.05) : more boys are malnourished (global acute malnutrition in percentage of the median = 3.6%, compared to the one for the girls : 0.0%). The weight for height means are 94.0 (standard deviation : 8.4) for the boys and 96.0 (standard deviation : 8.5) for the girls and are statistically different (p<0.05) : boys are thinner than girls. The study of the relation between boys and girls nutritional status by the age group (6-29 and 30-59 months) does not show a significant relation for each age group.

37

Table 62 NUTRITIONAL STATUS BY SEX AND AGE GROUP In percentage of the median

6 - 29 months 30 - 59 months

Boys Girls Boys Girls Nutritional status n % n % n % n %

Severe malnutrition 2 1.9% 0 0.0% 1 1.1% 0 0.0%Moderate malnutrition 3 2.8% 0 0.0% 1 1.1% 0 0.0%

Normal 101 95.3% 90 100.0% 85 97.7% 107 100.0%

TOTAL 106 90 87 107

There is no significant difference. However, some boys are affected by malnutrition whereas no girls are affected. Moreover, for the children aged from 6 to 29 months old, the weight for height means are 91.7 (standard deviation : 8.0) for the boys and 94.1 (standard deviation : 7.3) for the girls and are statistically different (p<0.05). For the children aged from 30 to 59 months old, the weight for height means are 96.6 (standard deviation : 8.2) for the boys and 97.7 (standard deviation : 9.2) for the girls and are not statistically different (p>0.05).


Table 63


MUAC < 110 mm 110<=MUAC<120 >= 120 mm HEIGHT N

N % N % N % < 75 cm 92 1 1.1% 4 4.3% 87 94.6%

75 <= H < 90 cm 181 0 0.0% 0 0.0% 181 100.0% >= 90 cm 117 1 0.9% 0 0.0% 116 99.1% TOTAL 390 2 0.5% 4 1.0% 384 98.5%

The mean for the whole sample is 144.2 mm with a standard deviation 12.5 mm. As research studies showed that MUAC is more relevant for children aged 12 to 47 months (corresponding to a height between 75 and 90 cm), it is interesting to also have the results for this specific height range : mean = 144.1 mm and standard deviation = 10.1 mm. 3.3. Measles vaccination

Table 64 MEASLES IMMUNISATION for children age 9 to 59 months


The measles coverage is low according to the number of children holding a vaccination card and vaccinated at 9 months or after. According to the mother or to the card 73.9% of surveyed children have been vaccinated, but according to the vaccination card, only 46.8% above or equal 9 months (age of vaccination against measles) are really protected against this disease. However, for those vaccinated before 9 months, this could be due to a mistake in recording of the birthdate or the vaccination date on the card.

38

The significant number of children without cards (32.6%) does not necessarily mean that these children have not been vaccinated. Therefore we cannot be entirely sure of the results. 3.4. Additional information

3.4.1. Date of arrival in Uganda

Table 65



3.4.2. Number of meals

Table 66 NUMBER OF MEALS

Number of meals N %

< 2 55 14.2% = 2 186 47.9% > 2 147 37.9%

The mean is 2.25 meals the day before, with a standard deviation 0.73.


Table 67

SOURCES OF INCOME

SOURCES OF INCOME N % Cultivating 132 33.8% Cultivating and live stock 3 0.8% Cultivating and trading food commodities 3 0.8% Cultivating and casual labour or formal job 20 5.1% Cultivating, casual labour and small commerce 1 0.3% Cultivating and small commerce 6 1.5% Live stock 3 0.8% Trading food commodities 20 5.1% Trading food commodities and formal job 4 1.0% Trading food commodities and small commerce 1 0.3% Casual labour or formal job 92 23.6% Formal job and small commerce 5 1.3% Shop or small scale commerce 76 19.5% No income 24 6.2% 42.3% of the families are cultivating their farm plot and selling part of their production, 31.3% are casual labourers or have formal jobs, 22.6% are engaged in small scale commerce and 6.2% have no source of income. Small-scale commerce mainly involves selling firewood, charcoal, local brew or even part of food ration.

39


Table 68 FAMILY SIZE

Table 69 NUMBER OF CHILDREN UNDER 5

FAMILY SIZE N % CHILDREN N % <= 3 29 7.4% 1 159 40.8% 4 - 6 206 52.8% 2 171 43.8% 7 - 9 120 30.8% > 2 60 15.4%

>= 10 35 9.0% Mean : 6.24, standard deviation : 2.18 mean : 1.77, standard deviation : 0.77

Presence of the parents : 82.8% of the families have both parents and 17.2% are single parents families. 33.8% of the heads of household are women and 66.2% are men. 0.3% of the heads of household have less than 20 years, 10.5% have between 20 and 24 years, 20.2% between 25 and 29 years, 42.1% between 30 and 39 years and 26.9% have more than 40 years. The mean is 34.5 with a standard deviation of 9.5.

3.4.5. Child birth 3.4.6.





Other* 2 0.5%

* Mainly “on the road” Only 52.7% of the women used the health facilities for delivery. 39.2% of the births were attended by a TBA or a CHW and 51.0% by a qualified health staff. Use of health facilities for delivering is higher than in the other settlements group (50%, 80-100%). It can be explained by the fact that refugees have been settled in these settlements before the others and therefore start to use heath facilities and and services of qualified health staff. 3.5. Relation between child's nutritional status and other factors NOTE : Some relationships between malnutrition and other factors appeared during the analysis of the survey. However, this type of survey is not designed to show malnutrition "risks factors". To show causative relations, a case control study will be needed, involving a different way of calculating the sample size. Therefore, these are only suggested relations, not established relations. For all the following tables, a child is malnourished if his weight for height index is less than -2 in Z score or if he has bilateral oedema. A child is not considered as malnourished if his W/H index is more or equal to -2 and if he has no bilateral oedema.

3.5.1. Relation between malnutrition, age and number of meals Table 72

NUMBER OF MEALS IN RELATION TO AGE GROUP

NUMBER OF MEALS < 2 = 2 > 2 AGE GROUP N

N % N % N % 6 - 29 months 194 31 16.0% 90 46.4% 73 37.6% 30 - 59 months 194 24 12.4% 96 49.5% 74 38.1%

TOTAL 388 55 14.2% 186 47.9% 147 37.9%

40

The meal number's means are 2.23 (with a standard deviation of 0.76) for children aged from 6 to 29 months old and 2.28 (standard deviation : 0.70) for children aged from 30 to 59 months old, and are not statistically different (p>0.05). Malnutrition rate is higher for children aged from 6 to 29 months old than for children aged from 30 to 59 months old. So it is more interesting to study possible relations between malnutrition and number of meals for each age group separately.



NUMBER OF MEALS < 2 = 2 > 2

NUTRITIONAL STATUS N



This table does not show a significant difference between the children having less than 2 meals, those having 2 meals and those having more than 2 meals (p>0.05). The meals' number means are 2.36 (with a standard deviation of 0.81) for the malnourished children, and 2.22 (standard deviation: 0.76) for the no malnourished children. These two means are not statistically different (p>0.05). But we have to be careful: several other factors interfere with number of meals, like breast-feeding and meals quality. Unfortunately, no data concerning breast-feeding and meals quality are available. Supplementary studies if they are to be carried out should include number of meals, their quality (quantitative and qualitative), breast-feeding, feeding time in the day, to be able to confirm or not the absence of influence of the number of meals on the nutritional status.



NUTRITIONAL NUMBER OF MEALS STATUS N < 2 = 2 > 2



This table does not show a significant difference between the children having less than 2 meals, those having 2 meals and those having more than 2 meals (p>0.05). However the percentage of malnourished children tends to decrease when the number of meals increases. The meals' number means are 1.67 (with a standard deviation of 0.58) for the malnourished children, and 2.29 (standard deviation : 0.70) for the no malnourished children. These two means are not statistically different (p>0.05).

41

3.5.2. Relationship between malnutrition and family composition Table 75


Family size Number of children under 5 years

Age of the head of household NUTRITIONAL STATUS


No malnourished 376 6.26 2.18 376 1.77 0.76 376 34.57 9.64 STATISTICALLY not different not different not different

SD = Standard deviation In this sample it seems there is no relation between family factors and malnutrition. But as explained before, we can not generalise. This absence of relation could be due to a real fact or to the sample.


OF THE FAMILY HEAD

PARENTS PRESENCE SEX OF THE FAMILY HEAD Both parents Single parent Woman Man NUTRITIONAL STATUS n % n % n % n %

Malnutrition 9 2.8% 5 7.5% 6 4.5% 8 3.1% Normal 314 97.2% 62 92.5% 126 95.5% 250 96.9% TOTAL 323 67 132 258

There is no significant relation between these two factors and the child's nutritional status (p>0.05). However it seems that children having only one parent are more vulnerable to malnutrition than children with both parents. It could be explained by the fact that in single parent families, the parent has attend to many things and has less time to take care of the child.

COMPARISON BETWEEN THE DIFFERENT SETTLEMENTS The three surveys were conducted by ACF - USA in April 1998. 1. Comparison of the weight for height distribution expressed in Z-score

Table 77

WEIGHT FOR HEIGHT DISTRIBUTION IN THE DIFFERENT SETTLEMENTS IN ADJUMANI IN APRIL 1998, In Z-score

< -3 ≥ -3 & < -2 ≥ -2 Oedema SETTLEMENTS

N n % n % n % n % 0% of cereals 390 3 0.8% 10 2.6% 376 96.4% 1 0.3%

50% of cereals 810 5 0.6% 54 6.7% 749 92.5% 2 0.2% 80 - 100% of cereals 900 4 0.4% 38 4.2% 855 95.0% 3 0.3%

42

Table 78 COMPARISON OF MALNUTRITION BETWEEN THE DIFFERENT SETTLEMENTS IN ADJUMANI IN

APRIL 1998, In Z-score

Children aged from 6 to 59 months old Children aged from 6 to 29 months old Settlements N Global acute malnutrition


Global acute malnutrition


0% of cereals 390 3.6% [2.2 - 5.7%] 1.0% [0.4 - 2.4%] 5.6% [3.2 - 9.4%] 1.5% [0.5 - 4.1%] 50% of cereals 810 7.5% [5.2 - 10.7%] 0.9% [0.2 - 2.5%] 12.7% [8.6 - 18.3%] 1.4% [0.3 - 4.5%]

80 - 100% of cereals 900 5.0% [3.2 - 7.6%] 0.8% [0.2 - 2.3%] 7.8% [4.8 - 12.3%] 1.3% [0.3 - 4.1%]


April 1998

02468

101214161820222426

-5 -4 -3 -2 -1 0 1 2 3 4 5

Reference80 - 100%50%0%

The percentage of malnutrition for children aged from 6 to 59 months old is low in all the settlements. But the percentage of malnutrition for children aged from 6 to 29 months is high, and extremely alarming, especially in the settlements receiving only 50% of cereals. In addition the number of children less than 5 years with a weight for height around -1 in Z-score is important in the settlements receiving no cereals and has to be monitored carefully. In fact, if the food situation becomes worse, these children will easily become malnourished. 2. Comparison of the weight for height distribution expressed in percentage of the median

Table 79 WEIGHT FOR HEIGHT DISTRIBUTION IN THE DIFFERENT SETTLEMENTS IN ADJUMANI IN APRIL

1998 In percentage of the median

< 70% ≥ 70% & < 80% ≥ 80% Oedema SETTLEMENTS

N n % n % n % n % 0% of cereals 390 2 0.5% 4 1.0% 383 98.2% 1 0.3%

50% of cereals 810 1 0.1% 31 3.8% 776 95.8% 2 0.2% 80 - 100% of cereals 900 1 0.1% 22 2.4% 874 97.1% 3 0.3%

43

Table 80 COMPARISON OF MALNUTRITION BETWEEN THE DIFFERENT SETTLEMENTS IN ADJUMANI IN

APRIL 1998 In percentage of the median

Children aged from 6 to 59 months old Children aged from 6 to 29 months old Settlements N





0% of cereals 390 1.8% [0.9 - 3.5%] 0.8% [0.3 - 2.2%] 2.6% [1.1 - 5.6%] 1.0% [0.2 - 3.4%] 50% of cereals 810 4.2% [2.5 - 6.8%] 0.4% [0.0 - 1.8%] 7.2% [4.2 - 11.9%] 0.5% [0.0 - 3.1%]

80 - 100% of cereals 900 2.9% [1.6 - 5.1%] 0.4% [0.0 - 1.8%] 4.3% [2.2 - 8.1%] 0.6% [0.0 - 3.1%] 3. Additional information

Table 81 DATE OF ARRIVAL IN UGANDA, FOR THE DIFFERENT SETTLEMENTS IN ADJUMANI IN APRIL

1998

Settlements N Arrival in 89 Arrival in 94 Arrival before January 92

Arrival before January 95

0% of cereals 390 35.6% 3.9% 75.0% 97.4% 50% of cereals 810 43.2% 29.3% 55.0% 95.8%

80 - 100% of cereals 900 45.9% 23.4% 53.5% 95.0%

Table 82 SOURCES OF INCOME IN THE DIFFERENT SETTLEMENTS IN ADJUMANI IN APRIL 1998

Settlements N Cultivating Casual labour or

formal job Shop or small scale

commerce No income

0% of cereals 390 42.3% 31.3% 22.6% 6.2% 50% of cereals 810 36.2% 19.9% 27.2% 13.6%

80 - 100% of cereals 900 43.8% 20.9% 14.2% 19.3%

44

4. Relation between child's nutritional status and other factors Table 83

SUMMARY OF THE POSSIBLE INFLUENCE OF DIFFERENT FACTORS ON THE NUTRITIONAL STATUS IN THE DIFFERENT SETTLEMENTS IN ADJUMANI

SETTLEMENT

child's age

child's sex

number of meals family

composition social status of the family

head

0% of cereals

YES. Children aged from 6 to 29 months old are

at a higher risk of malnutrition. RR = 3.63

YES. More boys

are affected by

malnutrition (RR = 6.12).

NO, for both age groups NO. NO.

50% of cereals



NO.

NO. But, for children aged from 6 to

59 months old, the percentage of

malnourished children tends to decreases slightly when the number of meals

increases.

YES. Malnourished

children belong to smaller families, with less children

under 5 years

NO.

80 - 100% of cereals



NO.

YES for children aged from 6 to 29 months old.

The percentage of malnourished children decreases when the

number of meals increases.

LITTLE. Malnourished

children belong to smaller families, with less children

under 5 years

NO.

A child is malnourished if his weight for height index is less than -2 in Z score or if he has bilateral oedema. A child is not considered as malnourished if his W/H index is more or equal to -2 and if he has no bilateral oedema. NOTE : Some relationships between malnutrition and other factors appeared during the analysis of the survey. However, this type of survey is not designed to show malnutrition "risks factors". To show causative relations, a case control study will be needed, involving a different way of calculating the sample size. Therefore, these are only suggested relation not established relations.

45

DISCUSSION - SUGGESTIONS The percentage of malnutrition of children aged from 6 to 59 months old is low. However this percentage is statistically higher for children aged from 6 to 29 months old than that of children aged from 30 to 59 months old, and is extremely alarming especially in the settlements receiving 50% of cereals. In the settlements receiving no cereals, many children have a weight for height around -1 Z-score. It means that they can easily become malnourished if there are any difficulties with the food supplies. According to food security assessments carried out by ACF-USA (report to follow) in the settlements receiving no cereals (Elema, Baratuku, Ibiaworo, Nyumanzi 1 and 2 in March 98 and Magburu 1 in April 98), the situation appears to be heading to stability in terms of food security. In Magburu 1 the population has many opportunities for sources of income. This is as a result of good integration with the national population. As such, there should not be a real problem in the coming months for this population to get food. The only fear is lack of rain : many of the refugees had not yet planted during this period because of the lack of rain. In conclusion, if the weather is good, food supplies should not be scarce, in the settlements receiving no cereals. Some relationships between malnutrition and other factors appeared during the analysis of the survey. However, this type of survey is not designed to show malnutrition "risks factors". To show causative relations, a case control study will be needed, involving a different way of calculating the sample size. Statistically, children less than 30 months are at a higher risk of becoming malnourished compared to children aged from 30 to 59 months old. This means that this age group (children 6 to 29 months) is more vulnerable to malnutrition and must be carefully monitored. This difference in terms of age group could be explained by the fact that in general mothers tend to abruptly stop breast feeding and start to give a normal diet to their children as soon as they conceive another child. The sudden weaning could induce malnutrition and occurs most of the time before the age of 30 months. In the settlements receiving 50% of cereal, there is a real need to treat the 6-29 months in these settlements. There could be a problem of availability of food adapted to this age group (porridge, weaning food) on top of the education problem. For the settlements receiving 50% of cereals and those receiving 80-100% of cereals, number of meal difference between the two age groups has been found. The children aged from 6 to 29 months old took in general, less meal the day before the survey than the older children. It could be explained by the fact that when the mothers go to cultivate they take the small children with them for the whole day and as a result these children do not have a real meal during the day. The bigger children stay at home and are generally fed by the caretakers / neighbours. For the settlements receiving no cereal, this difference is not observed between the two age groups. Some hypotheses can explain this difference between the settlements groups. It seems that refugees from settlements receiving no cereal tend to give more meals to their children and the same number of meals regardless the age. But this result has to be carefully interpreted and can not be generalised because breast-feeding influences the meal number. In fact many children less than 24 months are still breast-fed and therefore get less meal than the others. It can be possible that mothers in these settlements stop breast-feeding earlier and therefore give more meals to the younger children than in the other settlements. In fact, few refugees in settlements receiving no cereal have no sources of incomes (6.2% compared to 13.6% in settlements receiving 50% of cereal and 19.3% in those receiving 80-100% of cereal). Many mothers in these settlements go for casual labour (31.3%). It could be interesting to know how often they go for casual labour and if they take the younger children with them. It could be possible that when they go for casual labour, children stay at home and, as they could not breast-fed, they take more meals, even those aged from 6 to 29 months old. In order to conclude on the relation between number of meals and age group it would be necessary to take into account breast-feeding. For the children aged from 6 to 29 months old, the percentage of malnourished children decreases when

46

the number of meals increases. But we have to be careful because several other factors interfere with number of meals, like breast-feeding and meals quality. Unfortunately no data concerning breast-feeding and meals quality are available. Supplementary studies if they are to be carried out should include number of meals and their quality (quantitative and qualitative), breast-feeding, feeding time in the day, to be able to confirm or not the influence of the number of meals on the nutritional status. Concerning the measles vaccination, the coverage is low according to the vaccination card (less than 50%, in all the settlements are vaccinated at or after 9 months). However, for those vaccinated before 9 months, it could be due to a mistake in the recording of the birthdate or the vaccination date on the card. A measles campaign was planned the week after the nutritional surveys in all district around Adjumani. During this survey it appears that women did not use the health facilities for delivering: more than 45% delivered in the tukul (home). One of the reasons given for this rather low use of health facilities is that they did not want to expose themselves to people during labour. The high percentage of women who delivered in the tukul, alone or with the help of an unqualified person can be due to the fact that women may have continued to work until the on set of labour. Such behaviour is dangerous for pregnant women. Most accidents occur when women deliver at home. That is why it is important to sensitise women about the advantage of delivering in health centre or with the help of qualified personnel. SUGGESTED ACTIONS

- To make regular nutritional surveillance in this area, to study the evolution of nutritional status of children under five years old, especially in the settlements receiving no cereals. A second nutritional survey is planned (in October 1998) in order to monitor the nutritional situation in all the settlements.

- To establish a routine nutritional surveillance system within the health centres and refer malnourished children to feeding programs in order to curb acute malnutrition.

- To consider starting a supplementary feeding program, especially in the settlements receiving only 50% of cereals.

- To strengthen the health education programme, focusing mainly on childcare and nutrition, weaning habits and on reproductive health in the settlement.

- To improve the measles vaccination coverage through the planned vaccination campaign and improve the EPI system.

- To sensitise refugees about the advantage of giving birth in health centre or with the help of qualified personnel, and to encourage prenatal consultation.

- To realise some food security assessments in order to: . identify the more vulnerable groups in the settlements receiving no cereals. Even if

they seem to be able to develop income-generating activities and the malnutrition rate is low, it appears that many children are vulnerable and can easily become malnourished.

. define some actions to improve the nutritional status of children in the settlements receiving 50% of cereal, where less families cultivate and where children are at higher risk of becoming severe malnourished.

. define some actions to improve the autonomy and decrease the dependence, on food distribution, of the refugees in the settlements receiving 80-100% of cereal, where lots of them have no sources of income.

NUTRITIONAL ANTHROPOMETRIC SURVEYS ADJUMANI - UGANDA 20th - 25th

Documents