Ethiopia

Save the Children (UK) Nutritional Survey 22-26 November 2000 in Fik Zone, Somali National Regional State

Attachments


1. EXECUTIVE SUMMARY
The Somali Region has been severely affected by drought since early 2000. SC (UK) has been active in four woredas in Fik zone since May 2000, providing emergency assistance to those most affected by the drought. An anthropometric and food security assessment was carried out in the four woredas of Fik Zone between 21 and 26 November 2000 to assess the nutritional status of the population and determine programme activities. A total of 1829 children were seen in two thirty by thirty cluster surveys. The results showed a decline in acute malnutrition as compared with previous surveys conducted in April and July in Fik Zone.

Fig 1: Prevalence of Acute malnutrition in Fik Zone

Anthropometrical Measurement
Fik and Hamero
Segeg and Dihun
Global Acute Malnutrition (GAM) <80%
12.2%
15.5%
Severe Wasting < 70% )
0.1%
0.8 %
Mean Weight for Height %
88.43%
87.0%
Z Scores global acute malnutrition <-2.0
17.2 %
24.2 %
Z Scores severe acute malnutrition <-3.0
1.6%
2.3%
Median weigh-for-height <80%
12.2%
15.5%
Median weight-for-height <70%
0.2%
0.8%

No oedema was recorded.

The prevalence of acute malnutrition is above the internationally accepted norms, although rates have decreased compared to surveys conducted earlier in the year.

Reported mortality rates have fallen from 20-25 per week in May-June'00 to 3-5 per week in November.

The food security situation appears to be improving. Pastoralists that migrated out of Fik Zone during the worst of the drought have been returning over the last couple of months with herds of camels and cattle. The condition of the livestock has improved with the increased availability of water and pasture.

Access to clean drinking water and health services are now thought to be the key factors affecting nutritional status rather than lack of access to food. Current leading health problems include malaria, TB, measles and diarrhea. The lack of functioning health facilities in the Zone is having a negative impact on the health status of the population.

Despite the overall improvements in food security in 2000, the nutritional situation continues to be fragile, particularly for the under-five population who is more vulnerable to the poor health environment. Close monitoring of the food security and nutritional status of the population are required as it remains to be seen what effect the livestock ban and closure of supplementary feeding programmes will have in the Region.

2. BACKGROUND

SC (UK) experience in Somali Region (Region 5) dates back to the early 1970's, when nutritional surveillance in the area began. Since 1994, activities have focused on rehabilitation and development, mainly in education, water, agriculture, restocking and veterinary projects but also maintaining food security surveillance and emergency response capacity. In 1998, for example, SC (UK) distributed 55,000 mt of food aid in Region 5.

In response to the acute food crisis caused by three successive years of drought, SC (UK) started emergency relief interventions in Somali Region in May 2000. These included:

  • The provision of 7,770 mt of wheat grain as a general ration for Degahabour, Gode and Fik Zones.
  • Blanket supplementary distributions to the entire under five populations and their mothers in four woredas in Fik Zone.
  • Targeted supplementary feeding programmes in Fik and Hamero Woredas for moderately malnourished children.
  • Three therapeutic feeding centres for severely malnourished children in Fik, Hamero and Gasangas Woredas.
  • Distribution of blankets and plastic sheeting to the most vulnerable populations in Fik Zone.

3. OBJECTIVES

2.1 To assess rates of global and severe acute malnutrition among children aged between 6 and 60 months in four Woredas in Fik Zone.

2.2 To collect socio-economic information to assist in the analysis of the nutritional data and to assess the food security situation in the Zone.

2.3 To review programme activities based upon the results of the survey.

4. METHODOLOGY

4.1 Selection of sample

Two two-stage cluster sample surveys were conducted. The first survey covered Fik and Hamero woredas and the second covered Segeg and Dihun woredas. A total of 915 children were seen in Fik and Hamero and 914 in Segeg and Dihun, giving a total of 1829 children.

Fik and Hamero woredas and Segeg and Dihun woredas were grouped together as factors such as ecological conditions, accessibility to pasture, relief distributions and clan relations were very similar within each grouping.

Clusters were selected in each survey using population proportion to size techniques. Population estimates were based on distribution lists used and verified over the previous months during SC(UK) food distributions. The selection criteria for the villages chosen included a minimum of fifty households per settlement, accessibility and security considerations. Note that this method did not result in an entirely random selection of children being measured. However, key informants have indicated that the results presented below are representative of the population as a whole and samples based on population sizes were taken from urban, peri-urban and rural communities across the four woredas.

From the central point of each selected cluster, a random direction was picked by spinning a pen to determine the direction and selection of individual households1.

4.2 Anthropometric Measurements

Weight was recorded with the minimum clothing to the nearest 0.1kg, using 25kg Salter Scale. The scale was checked after every 7-10 children were weighed with a stone of a known weight.

All children were measured lying down. Length was measured to the nearest 0.5cm. One centimetre was subtracted from the length of children who were greater than 85cm in length during analysis of data to compensate for increased recumbent length measurement2. This allowed the data to be compared to the reference standards that use height data. The range of children selected was 70cm -110cm.

Data collected included age, sex, weight, height, presence of oedema, measles and polio vaccination and inclusion in SC (UK) targeted and blanket feeding programmes.

Nutritional oedema was defined by the presence of bilateral oedema. This was detected by applying normal thumb pressure to the feet for three seconds. If the shallow pit remained when the thumb was lifted on both feet, nutritional oedema was confirmed.

Results for weight for height were compared against Reference Population Table of the NCHS/CDC/WHO 1982.

Data was collected by four teams of four people, each including a team leader, data recorder and two measuring assistants.

EPI INFO 6 and Excel Software were used to analyse the results of the survey.

4.3 Collection of Socio-Economic Data

Information, covering issues of food-security, rainfall, condition of livestock, mortality and morbidity, was gathered through discussions with key informants, typically three to four community leaders in the form of group discussion based upon a questionnaire (Annex 4). All team leaders have had training and experience in participatory rural appraisal techniques. Where possible, women from the community were also involved. Information was crossed checked with secondary sources from other agencies and by observations made by SC (UK) staff.

5. RESULTS

5.1 The sample population

Ratio of Females to Males

Fik and Hamero: 432 Female: 483 Male - 47.2%: 52.8%
Segeg and Dihun: 466 Female: 448 Male- 51%: 49%

5.2 The prevalence of acute malnutrition

Figures 1 and 2 show the prevalence of global, moderate and severe acute malnutrition in the two surveys. The prevalence of severe acute malnutrition in both surveys is low. There were no cases of oedema in either sample. The estimated prevalence of moderate acute malnutrition was higher in Segeg and Dihun (14.8% <80%WFH) than in Fik and Hamero (12.0% <80%WFH).

The graphs showing the distribution of weight-for-height z scores in both surveys (annex 1 and 2) indicate that the population's weight-for-height has shifted to the left in comparison to the reference population. This implies that the population has lower weight for height than the reference population.

Fig 2: Summary of prevalence of acute malnutrition (Weight for Height percent of the median)

Indicator
Fik and Hamero
Segeg and Dihun
Total no. of children
915
914
Prevalence of moderate acute malnutrition 70- 80%
12.0%
14.8%
Confidence interval
9.0-15.0%
12.0-17.6%
Prevalence of severe acute malnutrition <70%
0.2%
0.8%
Confidence interval 95%
0.0-0.5%
0.3-1.3%
Global Acute Malnutrition (GAM) <80%
12.2%
15.5%
Confidence interval 95%
9.2-15.2%
12.7-18.4%
Mean Weight for Height %
88.43%
87.0%

Fig 3: Summary of prevalence of acute malnutrition (Weight for Height Z score)

Indicator
Fik and Hamero
Segeg and Dihun
Total no. of children
915
914
Prevalence of moderate acute malnutrition -3z - <-2z
15.5%
21.9%
Confidence interval
12.1-18.9%
18.6-25.2%
Prevalence of severe acute malnutrition <-3z
1.7%
2.3%
Confidence interval 95%
0.7-2.8
1.4-3.2%
Global Acute Malnutrition (GAM) <-2z
17.2%
24.2%
Confidence interval 95%
13.4-21.1%
20.7-27.7%
Mean Weight for Height %
88.43%
87.0%

As is usual in anthropometric surveys, the results presented as z scores show higher levels of acute malnutrition than percentage of the median values.

5.3 The prevalence of acute malnutrition by age and sex

Figure 4 shows that rates of global and severe acute malnutrition show greater prevalence of acute malnutrition in children under 24 months old in both surveys.

There is no significant difference in the prevalence of acute malnutrition between boys and girls.

Fig 4: Prevalence of wasting by age group (Mean Weight for Height)

Age Group
Severely Malnourished
<70% W/H
Moderately Malnourished
70-80% W/H
Global Acute Malnutrition (GAM)
<80% W/H
Fik and Hamero
6-24 Months
0.10%
6.66%
6.76%
25-36 Months
-
2.19%
2.19%
37-60 Months
0.1%
3.17%
3.27%
Total 915 Children
12.2%
Segeg and Dihun
6-24 Months
0.33%
6.8 %
7.13%
25-36 Months
0.11%
2.6 %
2.71%
37- 60 Months
0.33%
5.36 %
5.69%
Total 914 Children
15.5 %

Fig 5: Prevalence of Wasting by Gender (Mean Weight for Height)

Severely Malnourished
< 70 % W/H
Moderately Malnourished
70-79.9% W/H
Global Acute Malnutrition (GAM)
<80% W/H
Fik and Hamero
Boys(n=483)
0.2%
12.6%
12.8%
Girls (n=432)
0.0%
11.3%
11.3%
Total 915 Children
0.1%
12.1%
12.2%
Segeg and Dihun
Boys (n=448)
0.7%
13.1%
13.8%
Girls (n=466)
0.9%
16.3%
17.2%
Total 914 Children
0.8%
14.7%
15.5 %

5.4 Changes in Nutritional Status and the food security situation

Indications that the nutritional status of the population has improved include:

i. The prevalence of acute malnutrition estimated in these surveys, in terms of both z scores and medians, is considerably lower than those estimated in surveys conducted earlier in the year. The survey results are not are not directly comparable due to methodological differences, but certainly imply an improvement in the situation.

(The survey conducted in April3 estimated a GAM of 57% for Fik and GAM of 65% for Hamero and in July4 a GAM of 28.1% for Fik and Hamero and 37.2% for Segeg and Dihun).

ii. Admissions into the therapeutic feeding centres (TFCs) have fallen over the last two months, the TFC in Fik closed at the end of October and from total admissions of 585 children in all three sites, only 54 children in need of therapeutic feeding remain in Gasangas and Hamero as at 21 Nov' 00.

iii. Reported mortality rates have fallen from 20-25 per week in May-June'00 to 3-5 per week in November.5

Other factors indicting that the food security situation of the population of Fik Zone has improved over the last few months and has reached acceptable levels include the following:

i. Pastoralists that migrated out of Fik Zone during the worst of the drought have been returning over the last couple of months with herds of camels and cattle. The condition of the livestock can be seen to have improved with the increased availability of water and pasture.

ii. Those populations that moved into urban areas to benefit from relief distributions have now started to move back to areas of origin in outlying areas.

iii. The amount of relief wheat being sold to buy other essentials, such as sugar, oil, tea and soap, has increased.

5.5 Health Factors

Based upon information gained from the questionnaire, the current leading health problems in the survey area include malaria, TB, measles and diarrhea. Discussion with key informants suggested that the lack of functioning health facilities in the Zone is having a very negative impact on the health status of the population.

The EPI programme coverage is low as can be seen below.

Fig 6: Coverage of Measles and Polio Vaccination Fik Zone Nov’00

Woreda
No of Children Seen
% of Children Vaccinated Against Measles
% of Children
Vaccinated against Polio
Fik
577
44
49
Hamero
338
46
21
Segeg
425
27
24
Dihun
489
36
46
Total
1829
38%
37%

6. DISCUSSION

6.1 Reasons for the improvement in nutritional status

The main factors contributing to these improvements include:

(i) The onset of the Dehr rains in September has led to an improvement in pasture. Although the amount of rainfall has been less than previous years, it has been sufficient to regenerate pasture, which has allowed herds to return. This is particularly important for the under five population, for whom camel milk is the stable diet. Camel milk is now available across the Zone, with the price falling to 0.50 cents per litre from 2 birr per litre in April. Originally it was reported that there was very heavy loss of livestock over the drought period, but it is difficult to assess the extent to which losses have actually occurred.

(ii) Due to the imposition of the livestock ban on imports by the Gulf States in September '00, animal prices are somewhat depressed but can be seen to have improved since the worst of the drought in February-May' 00. Even though there has been less demand for livestock, particularly shoats, the terms of trade for livestock has improved against relief wheat, the main source of available cereals.

(iii) The availability of water for drinking and cooking has improved, but quality still remains an issue, particularly in Bermil and Garasley.

(iv) Food distributions by SC(UK), DPPC, WFP, ICRC and IRO since April 2000 have included:

- General ration distribution of wheat grain

- Blanket distribution of CSB, sorghum and oil

- Blanket supplementary distributions of CSB and oil to the entire under five population and lactating mothers

- Supplementary feeding programmes for children under 80% W/H.

(v) SC (UK) therapeutic feeding centres in Fik, Hamero and Gasangas gave support to 737 severely malnourished children under 70%W/H over the period from June to December 2000.

6.2 Differences in nutritional status between Fik and Hamero and Segeg and Dihun

The difference between the prevalence of global acute malnutrition, as measured by z scores, in the two survey areas are significant (risk ratio=0.71, confidence interval 0.55-0.93), but are not significant when measured by medians. This result implies that there are marginal differences in the prevalence of acute malnutrition between the two areas. These differences may be explained by:

(i) Ecological factors such as lower rainfall in deeper Ogaden and better access to water for drinking and livestock in Fik and Hamero and closer proximity to better pasture in the Fafen Valley around Babille and Besidimo

(ii) Segeg and Dihun received less food aid, as they are less accessible, particularly over the rainy season.

(iii) A bias towards Fik and Hamero in the compilation of the DPPC population figures for the distribution of food.

(iv) Better access to markets for populations in Fik and Hamero, in terms of both greater availability of food-stuffs and lower prices.

6.3 Reasons behind the differences by age and sex

Younger children in both surveys had higher prevalences of acute malnutrition. In general, younger children are more vulnerable to infection than older children and a higher prevalence of acute malnutrition in the younger age groups often implies that the malnutrition seen is more closely linked to health factors than access to food. This supposition is supported by the health and food security data reported above.

There are no significant differences in the prevalence of acute malnutrition by sex in either area.

6.4 Implications for SC(UK)'s emergency programmes

The prevalence of acute malnutrition is above the internationally accepted norm6 in both areas and the nutritional status of both populations is classified as "poor" according to the DPPC guidelines (see fig. 7).

Fig 7: Disaster Prevention and Planning Commission Intervention Guidelines

The Transitional Government of Ethiopia Relief and Rehabilitation Commission Guidelines on Nutritional Status and Food Relief. Early Warning Department 1995

Nutritional Status
Mean Weight for Height
Median weight-for-height<80%
Good
>95%
<5%
Satisfactory
90-95%
6-10%
Poor
85-89%
11-20%
Serious
80-84%
21-30%
Emergency
<80%
>30%

Despite the fact that the prevalence of acute malnutrition remains above optimal levels, observations of the population's nutritional status, morbidity, mortality and movements, indicate that the nutritional status and general food security situation of the population has improved over the last six months.

It is thought that access to clean drinking water and health services are now the key factors affecting nutritional status rather than lack of access to food. TB remains one of the major health problems in Somali region, particularly prevalent in the under five population7.

For these reasons, it has been decided to discontinue both the supplementary distributions and the provision of therapeutic feeding in 2001. A further food distribution was conducted in Dihun district in response to the survey findings. In addition, those children classified as acutely malnourished at the end of the programme in Fik district will be given a further one month's ration and follow-up medical treatment where necessary in the community in January 2001.

However, the results of these surveys suggest that overall nutritional situation remains fragile, particularly in Segeg and Dihun. Households that have lost livestock and the poorest sections of the communities are still vulnerable and it remains to be seen how they cope once food aid is suspended. Similarly, the extent to which the imposition of the livestock ban will impact upon livelihoods in Somali region is not clear. It will be necessary to monitor the situation closely over the coming months.

7. RECOMMENDATIONS

SC(UK)'s supplementary food distributions should cease at the end of the year, but general ration distributions should continue for a three to four month transitionary period, by when it is hoped that local agricultural produce will be available.

SC(UK)'s therapeutic activities should close at the end of December. The remaining children should be given follow-up care in the community for a further month, as no health facilities are functioning within the Zone.

The food security situation and the nutritional status of the population should be closely monitored for a further four months until the onset of the Gu rains in April. This may warrant a further nutritional and food security assessment in three months time.

Rates of EPI coverage are low in Fik Zone, this issue should be addressed by the relevant authorities in the coming months.

Consideration as to how best to improve access to clean drinking water and health services is required.

Footnotes :

1 'Selection of Children in the Clusters' p58 MSF Nutrition Guidelines 1995

2 As recommended by WHO (Physical Status: the use and interpretation of anthropometry, WHO Technical report series, 1995

3 Fik Zone Assessment 26-03 May 2000 SC(UK)

4 Anthropometric and Food Security Assessment, Fik, Hamero, Segeg and Dihun Woredas of Fik Zone 14-17 July 2000 SC(UK)

5 Information collected from Community leaders during the three nutritional surveys carried out by SC(UK) in Fik Zone, April-Nov'00

6 Humanitarian Charter and Minimum Standards in Disaster Response The Sphere Project 1998

7 No figures are available, but Action Contre La Faim have diagnosed 23.3% of children in the therapeutic feeding centre in Kebri Dehr as suffering from TB, 'Emergency Intervention Korahai Zone Somali National Regional State'. Activity Progress Report 26 Sept to 26 Oct 2000 and it can be assumed that levels of TB infection are similar across the region for pasturalist communities.

to download Annexes 1-4 in zip format