NUTRITIONAL ANTHROPOMETRIC AND MORTALITY SURVEY FINAL REPORT

Transcription

1 NUTRITIONAL ANTHROPOMETRIC AND MORTALITY SURVEY FINAL REPORT HABASWEIN AND WAJIR SOUTH DISTRICT NORTH EASTERN PROVINCE, KENYA 16 th - 26 th January

2 Acknowledgements Save the Children takes this opportunity to express special thanks to;. European Commission Humanitarian Aid (ECHO) for their financial support. The Provincial administration, United Nations Children s Fund (UNICEF), Arid Lands Resource management Project (ALRMP), Ministry of Agriculture (MoA), Ministry of Health (MOH) and District Development Office (DDO) through their respective district focal persons for the support provided during the entire survey period. Survey team (coordinator, supervisors, team leaders, enumerators, and drivers) for their tireless efforts to ensure that the survey was conducted professionally and on time. The Monitoring and Evaluation team who tirelessly ensured the survey data entry and analysis was done. Community members who willingly participated in the survey and provided the information needed. Rahab Kimani, Nutrition coordinator Save the Children 2

3 Table of Contents Acknowledgements... 2 Table of Contents... 3 List of Tables... 4 Table of figures... 4 Abbreviations Executive summary Introduction Survey Methodology Sampling procedure and sample size for Anthropometric and mortality data Sample size calculation for anthropometry Sample size calculation for mortality Sampling procedure: selecting households and children Case definitions and inclusion criteria Questionnaire, training and supervision Data analysis Nutritional indices Results Survey sample description Anthropometric results (based on WHO standards 2006): Mortality results (retrospective over 3 months/days prior to interview) Children s morbidity Water and sanitation Food Security and Utilization Infant and Young Child Feeding Practices Food diversity at household levels Main livelihoods activities Discussion Nutritional status Mortality Targeted Feeding Programme coverage Blanket feeding programme Causes of malnutrition Conclusions Recommendations References Appendices

4 List of Tables Table 1 : Results summary for acute malnutrition... 7 Table 2: Results summary for Mortality... 8 Table 3: Results summary for children s morbidity, immunization and supplementation... 8 Table 4: Results summary for water, hygiene and sanitation... 8 Table 5: Results Summary for food and income sources... 9 Table 6: Amount of rainfall received in the district during the last short rains Table 7: Food aid basket Table 8: Other relief programmes in the area Table 9: WFH indices Table 10: MUAC indices Table 11: Height for Age indices Table 12: Mortality Thresholds Table 13: Survey Sample Selection Table 14: Acute Malnutrition definitions: Table 15: Distribution of age and sex of sample Table 16:prevalence of malnutrition based on weight for height z-scores (and/oedema) and by sex Table 17: Prevalence of acute malnutrition by age, based on weight for height z-scores and/or oedema Table 18: Distribution of acute malnutrition and oedema based on weight-for-height z-scores Table 19: Prevalence of underweight based on weight-for-age z-scores by sex Table 20: Prevalence of underweight by age, based on weight-for-age z-scores Table 21: Prevalence of stunting based on height-for-age z-scores and by sex Table 22: Prevalence of stunting by age based on height-for-age z-scores Table 23: Mean z-scores, Design Effects and excluded subjects Table 24: Plausibility Checks for Anthropometric Data Table 25: Mortality rates Table 26: Prevalence of reported illnesses in children in the two weeks prior to interview (n=267) Table 27: Vaccination, deworming and Vitamin A coverage Table 28:Handwashing time Table 29: Dietary diversity amongst children 6-23 months Table 30: Minimum meal times months Table 31:Comparison of acute malnutrition expressed by MUAC 2011 and Table of figures Figure 1: Map of Wajir South Figure 2: Population age and sex pyramid Figure 3: Weight for height z-scores Figure 4: Symptoms breakdown in the children in the last two weeks prior to interview (n=267) Figure 5: Caretaker s health seeking behaviour Figure 6: Vitamin supplementation coverage Figure 7: source of water Figure 8: Time taken on water collection Figure 9: Methods of water treatment Figure 10: Types of toilet facilities Figure 11: Initiation of breastfeeding Figure 12: Exclusive breastfeeding in the last 24 hours Figure 13: Food consumption 7 days and 24-hour recall Figure 14: Main Livelihood activities Figure 15:Comparison of malnutrition rates based on WFH Z-scores in Wajir South ( ) Figure 16: Comparison mortality rates

5 Abbreviations ALRMP II - Arid Lands Resource Management Project II APHIA Plus Aids Population Health Integrated Assistance Project ASAL - Arid and Semi-Arid Lands CMR - Crude Mortality Rate CI - Confidence Interval CMR - Crude Mortality Rate CSB - Corn Soya Blend ECFF European Commission Food facility ECHO - European Commission Humanitarian Aid ENA - Emergency Nutrition Assessment EPI - Extended Programme of Immunization GAM - Global Acute Malnutrition GFD - General Food Distribution HAZ - Height-for-Age Z-score HNSP - Hunger Safety Net Project IYCF - Infant and Young Child Feeding Practices KFSSG - Kenya Food Security Steering Group L/HAZ - Length/ Height for Age Z-score MOH - Ministry of Health MUAC - Mid-Upper Arm Circumference OPV - Oral Polio Vaccine OTP - Out-patient Therapeutic Program RUSF - Ready to Use Supplementary Food RUTF - Ready to Use Therapeutic Food SAM - Severe Acute Malnutrition SC - Stabilization Centre SCiK - Save the Children in Kenya SD - Standard Deviation SFP - Supplementary Feeding Programme SMART - Standardized Monitoring and Assessment of Relief and Transitions U5MR - Under Five-Mortality Rate UNICEF - United Nations Children s Fund URTI - Upper Respiratory Tract Infection WASDA - Wajir South Development Association WAZ - Weight-for-Age Z-score WFP - World Food Programme WHM - Weight for Height Median WHO - World Health Organization WHZ - Weight-for-Height/length Z-scores 5

6 1.0 Executive summary Wajir South is one of the districts that form Wajir County in North Eastern Province (NEP) and is gazetted as part of the Arid and Semi-Arid Lands of Kenya (ASAL). The district is located in the North West horn of Kenya bordered by Somalia republic to the east, Wajir West district to the West, Lagdera to the south and Wajir East district to the North. The district was in 2010 subdivided into Habaswein and Wajir South districts. The larger Wajir South district administratively consists of 5 divisions including Habaswein, Sabuli, Banane, Kulaaley and Diif. Within the five divisions there are a total of 14 government health facilities including Habaswein district hospital. Rainfall in the district is unpredictable, erratic and inadequate amounting to mm annually on average and the district experiences an annual evapo-transpiration of 2500mm apart from the last short rains which was adequate it is also characterized by long dry spells and short rainy seasons which are erratic, unreliable and poorly distributed. Temperatures are normally high ranging between C. Soils are mainly sandy and sandy loams. The districts are characterized by chronic food insecurity and high rates of malnutrition. The community is largely pastoralist and pre-dominantly Somali. Due to the drought experienced in 2011 in the Horn of Africa (HOA) which was declared by the Kenyan president as a national Disaster, many livestock died, dams dried up and there was an escalation in the rates of malnutrition. Thereafter, in the months of October to November, the district as in many parts of Kenya experienced torrential rains that resulted in massive flooding rendering many parts of the district inaccessible. Though there water and pasture availability improved, there was an increase in prevalence of diseases affecting both humans and livestock. Insecurity further affected the district since September 2011 with the main affected sites being Dagahaley, Gerilley, Dadajabulla, Sarif and Diff. The district population is currently estimated at 137,991 1 persons with a growth rate of 3.7%. About 60% -70% of the people depend largely on livestock for their livelihood. The main form of land use is nomadic pastoralism which is seen as the most efficient method of exploiting the range lands hence pastoral activities are practiced all over the district. The prominent ethnic group is Somali-Muslim. Survey objectives The overall goal of this survey was to assess the current nutrition situation given the on-going nutrition emergency interventions. This assessment constituted a nutrition surveillance system as well as providing information for program future planning. The specific objectives of the survey were to estimate: 1. The prevalence of acute and chronic malnutrition in children aged 6-59 months; 2. The nutrition status pregnant women and mothers with children <5 years ; 3. The crude and under five mortality rate and causes of death; 4. The proportion of households with access to improved water and sanitation; 5. Infant and young child feeding practices 6. The coverage and content of the general food distribution; 7. The food access and dietary diversity at household level; 8. The Coverage of measles and BCG vaccination among target children; 9. The Coverage rate of Vitamin A. supplementation and de worming; 10. The Morbidity rates of children 6-59 months 2 weeks prior to the survey; 11. To recommend appropriate interventions based on the survey findings; 1 According to the projected population based on 2009 National population census 6

7 Area Covered The survey was conducted from 16 th to 26 th January 2012 in the five divisions of Wajir South namely Habaswein, Sabuli, Banane, Diif and Kulaaley. Methodology A two stage cluster sampling using SMART (Standardised Monitoring of Assessment in Relief and Transition) methodology was employed with identification of clusters being proportional to the population size. The target population for the anthropometric survey was children aged 6-59 months and mothers with children aged between 0-59 months. A total number of 38 clusters were selected and 645 households were visited. Data was collected on anthropometry, morbidity, vaccination and de-worming status, Vitamin A, zinc and iron folate supplementation, Infant and young child feeding practices, hygiene and sanitation practices and food security. Retrospective information on mortality was collected using the current household census method, with a recall period of 90 days (from 20 th October 2011-Kenyatta Day), from all households visited including those without children under the age of five. A total of 38 clusters and 645households were visited and 863 children from 6 to 59 months were assessed for anthropometry and other indicators. The final analysis was on 847 children after exclusion of 16 records. Anthropometric and mortality data were analyzed using the ENA software version November Qualitative and quantitative data was analyzed using the Epi Info/ ENA software. Key findings Table 1 : Results summary for acute malnutrition INDEX INDICATOR RESULTS WHO(2006) Z- score Global Acute Malnutrition (196) 23.1 % N=847 NCHS(1977) N=847 MUAC ANALYSIS [N=847] Z-score W/H< -2 z and/or oedema Severe Acute Malnutrition W/H < -3 z and/or oedema Global Acute Malnutrition W/H< -2 z and/or oedema Severe Acute Malnutrition W/H < -3 z and/or oedema Global Acute Malnutrition MUAC (<125 mm) ( CI) (39) 4.6 % ( CI ) (172) 20.3 % ( C.I.) ( 17) 2.0% ( C.I.) (80) 9.4% ( CI) Severe Acute Malnutrition MUAC (<115 mm) At risk of malnutrition MUAC ( 125 and <135mm) (13) 1.5% (270) 31.9% Oedema present 0 7

8 Table 2: Results summary for Mortality Child Mortality CMR (total deaths/10,000 people / day) 0.30 ( ) U5MR (deaths in children under five/10,000 children under five/ day) 0.54 ( ) Table 3: Results summary for children s morbidity, immunization and supplementation Child Morbidity BCG SCAR MEASLES 9-59 MONTHS OPV 1 OPV 3 De-worming Vitamin A supplementation (in the last 6 months)-6-11 Months Vitamin A Months Total children sick Diarrhoea Fever, cough, difficult breathing Fever with chills like malaria Vomiting Present Absent By card According to caretaker By card According to caretaker By card According to caretaker Given Once Given twice Once Twice Once Twice 25.6% (267) 60% (160) 63%(168) 40% (108) 34% (90) 96.3% (817) 3.7% (31) 51.20% (434) 44.14% (374) 54.8% (468) 42.3% (359) 54.2% (460) 42.1% (357) 50.4% (427) 29.0% (246) 70% (61) 20% (17) 61% (466) 34% (257) Table 4: Results summary for water, hygiene and sanitation Source of water Borehole 46 (224) Public water pan 27%(135) Unprotected well 9% (41) Water tankers 11% (51) Protected well 5% (23) Water purification Did nothing to water 88.7 %(424) Access to toilet facility Yes No Place of relief (defecation) for people with no access to toilet Bush defecation Open field defecation Hand Washing Before eating food After visiting toilet After cleaning children s bottoms Before preparing food Before feeding the child 26% (123) 74% (345) 64% (207) 6% (31) 36.4% (174) 15.7 % (75) 6.1% (29) 6.8%% (32) 6.3%(30) 8

9 Table 5: Results Summary for food and income sources Livelihood activity Livestock herding 38.4% Wage/casual 9.6% Petty trade 10.2% Firewood/charcoal 1.2% Source of food Purchase 81.3% Food consumption (7 day recall period) Cereal Sweet sugar honey Milk and Milk products Oil/fats Dark green leafy vegetables Vitamin A rich foods Source of income Sale of livestock Petty trade Casual labour 99.1% 91.2% 90.2% 89.8% 5.4% 4.5% 43.9% (215) 12.4% (61) 15.5% (76) Conclusions The prevalence of acute malnutrition in wajir south district is considered critical with global acute malnutrition (GAM) of 23.1 ( % C.I.) and Severe Acute Malnutrition (SAM) rate of 4.6 % ( % C.I.). Compared with the survey undertaken in 2011 which indicated GAM of 28.5% ( % CI) and SAM of 4.5% ( ), the acute malnutrition rates have reduced though not statistically significant due to overlapping confidence intervals. Other indicators like Vitamn A supplementation, deworming,have also improved compared to the previous survey. RecommendationsImmediate Continue supporting to the MOH with OJT,supportive supervision and logistical support. Scale up community mobilization activities through the empowerment of the community i.e. Traditional Births Attendants (TBAs), sheikhs and traditional healers in the detection and referral of acutely malnourished children <5 years, pregnant and lactating mothers. Need for provision of water trucking and water storage equipment to areas affected by water stress as well as fuel subsidies and repairs. Scale up Behaviour change communication (BCC) and formation of Community Mother to Mother support groups to improve Infant and Young Child nutrition status. Medium term MOH to develop a health workers retention strategy to reduce the high staffs turn over. Continued health outreach in locations inaccessible to health facilities to offer basic primary health care package Strengthen continuous nutrition surveillance through regular nutrition assessments and ongoing MUAC screening Strengthening of hygiene practices to reduce the incidence of diarrhoeal disease including health and nutrition promotion to educate the community on basic WASH i.e. domestic treatment of drinking water and proper disposal of faecal waste. 9

10 Long term Provision of toilet facilities through community participatory approaches coupled with awareness campaign on the importance of using such facilities i.e. Community Led Total Sanitation (CLTS) and Participatory Hygiene and Sanitation Transformation (PHAST) approaches Disaster risk reduction strategy in programming.this includes but not limited to strategic re stocking, educating the community on management of disaster risk reduction, pasture growing and provision of fuel for irrigation, and encouraging the communities to establish pasture range reserve /reseeding to avoid mass losses of animals (livelihood) during drought. Need for defined linkage of nutrition sector cluster with other sectors such as Water Sanitation and Hygiene (WASH) in the longer term. Advocacy for recruitment and retention of health workers i.e. nurses, Clinical Officers (Cos) and nutritionists in North Eastern province Government of Kenya (GOK) to strengthen community health strategy in the ASALS to foster empowerment of CHWs to participate in health and nutrition promotion and management of minor childhood ailments. 10

11 2.0 Introduction Wajir South is one of the districts that form Wajir County in North Eastern Province (NEP) and is gazetted as part of the Arid and Semi-Arid Lands of Kenya (ASAL). The district is located in the North West horn of Kenya bordered by Somalia republic to the east, Wajir West district to the West, Lagdera to the south and Wajir East district to the North. The district was in 2010 subdivided into Habaswein and Wajir South districts. The larger Wajir South district administratively consists of 5 divisions including Habaswein, Sabuli, Banane, Kulaaley and Diif. Save the Children operates in all the Divisions of Wajir South districts. Within the five divisions there are a total of 14 government health facilities including Habaswein district hospital. The districts are characterized by chronic food insecurity and high rates of malnutrition. The community is largely pastoralist and pre-dominantly Somali. Due to the drought experienced in 2011 in the horn of Africa (HOA) which was declared by the Kenyan president as a national Disaster, many livestock died, dams dried up and there was an escalation in the rates of malnutrition. Thereafter, in the months of October to November, the district as in many parts of Kenya experienced torrential rains that resulted in massive flooding rendering many parts impassable. Though there water and pasture availability improved there was an increase in prevalence of diseases affecting both humans and livestock. Table 6: Amount of rainfall received in the district during the last short rains DIVISION Amount of rain receive in mm(monthly) October 2011 November 2011 December 2011 Total Diff Habaswein Insecurity has further affected the district over the year with the main affected sites being Dagahaley, Gerilley, Dadajabulla, Sarif and Diff. Rainfall is unpredictable, erratic and inadequate amounting to mm annually on average and the district experiences an annual evapo-transpiration of 2500mm. It is also characterized by long dry spells and short rainy seasons which are erratic, unreliable and poorly distributed. Temperatures are normally high ranging between C. Soils are mainly sandy and sandy loams. The district population is currently estimated at 137,991 2 persons with a growth rate of 3.7%. About 60% -70% of the people depend largely on livestock for their livelihood. The main form of land use is nomadic pastoralism which is seen as the most efficient method of exploiting the range lands hence pastoral activities are practiced all over the district. The prominent ethnic group is Somali-Muslim. The larger Wajir South district has a total population of 137,991. The population is predominantly Muslim and of Somali ethnicity, and is divided into clans, with community elders being in charge of daily affairs. Ogaden is the predominant clan and other clans include Masare, Garre, Degodia, Murule and Ajuran. Ogaden clan is further divided into four sub-clans namely-garre (GK), Bah Gere, (BG) Muhamed Zuera (MZ) and Makabul (MK). 2 According to the projected population based on 2009 National population census 11

12 Figure 1: Map of Wajir South The District consists largely of a featureless plain. There are three swamps namely Boji, Lagbogol and Lorian all of which are found in Habaswein division. The area receives bi-modal rains with the onset of the long rains in April-May. The months succeeding the long rains, June to September, are very dry but vegetation continues to thrive because the lower temperatures reduce the rate of evaporation. The short rains fall from October to December. However in the past short rains have been more reliable than the long rains. The annual precipitation is about 280mm which varies in amount and distribution from year to year. The district s climatic condition is characterized by recurrent droughts and unreliable rainfall that hinders crop production and growth of pasture for livestock keeping. These cyclic shocks have retarded development in the area since gains of a particular season are wiped out by drought and famine 3. Save the Children has been implementing programmes in Wajir South district since July Save the Children s strategy for Wajir South aims at improvement of and access to health facilities, the protection and improvement of the nutritional status of beneficiaries and improved food security and livelihoods of beneficiaries through community management structures and social protection. Save the children integrated approach including health, Nutrition, Food security and Livelihoods Support program, Water Sanitation and Hygiene (WASH), child protection and Education programmes aims to address the immediate and underlying causes of malnutrition through enhancement of house hold food security and livelihoods in the medium term while at the same time linking this to long term livelihood support. The World food programme (WFP) through its implementing partner Wajir South Development Association (WASDA), has been undertaking General food Distribution (GFD) in the area since In the Month of December 2011 WFP through the Kenya Food Security Steering Group (KFFSG) increased the GFD target beneficiaries to 308,700 persons against the total population of 606,000 persons in the entire Wajir County. All GFD are distributed are at a 75% ration scale of the daily per capita energy requirement 4 as follows; Table 7: Food aid basket Commodity Cereals Pulses CSB Vegetable Oil Ration sizes 10.35kgs 1.80kgs 1.20kgs 0.60kgs The ministry of special programmes through the office of the president (OOP) has also been supplying relief food commodities to various communities and institutions as agreed by the District Steering Group (DSG). 3 Wajir South District Development plan (Final Draft) 4 Based on the UNHCR/UNICEF/WFP guidelines for food and nutrition in emergencies 12

13 Table 8: Other relief programmes in the area Organization Activities Horn Relief Cash relief and cash for work APHIA plus Health (HIV), child protection World Vision Wajir South ADP Sponsorship, Water and Sanitation, food aid, education and Health VSF Suisse Improved Community Response to Drought (ICDR Project) Save The Children in Kenya(SCUK) WFP/ WASDA WARDA(Waaso Resource Development Agency) Integrated management of acute malnutrition, health, food security and livelihoods, WASH for health facilities, education and child protection Food Aid, Disaster Risk Reduction, school feeding programme Health, Education and food aid The overall goal of this survey was to assess the current nutrition situation given the on-going nutrition emergency interventions. This assessment constituted a nutrition surveillance system as well as providing information for program future planning. The specific objectives of the survey were to estimate: 1. The prevalence of acute and chronic malnutrition in children aged 6-59 months; 2. The nutrition status pregnant women and mothers with children <5 years ; 3. The crude and under five mortality rate and causes of death; 4. The proportion of households with access to improved water and sanitation; 5. Infant and young child feeding practices 6. The coverage and content of the general food distribution; 7. The food access and dietary diversity at household level; 8. The Coverage of measles and BCG vaccination among target children; 9. The Coverage rate of Vitamin A. supplementation and de worming; 10. The Morbidity rates of children 6-59 months 2 weeks prior to the survey; 11. To recommend appropriate interventions based on the survey findings; 3.0 Survey Methodology The survey used the standardised monitoring and Assessment of relief and transitions (SMART 2006) sampling methodology. Emergency Nutrition Assessment (ENA) for SMART software was used to determine sample size for anthropometry and retrospective mortality. Children from 6 to 59 months of age were included in the sampling. Two stage cluster sampling was applied to assign clusters and to select the survey population. A local events calendar was used to aid mothers in accurate determination of the ages of their children. The principle of Probability Proportional to Population Size (PPS) was used in the identification of clusters using ENA for SMART software version A total population of 137,991 was estimated for the survey area covering the five divisions. Information on population figures was obtained from the District Development Officer (DDO) in Wajir south based on projections from the 2009 National Census. The survey training took place from 16 th to 19 th January 2012 while data collection was conducted from 20 th to 26 th January

14 3.1 Sampling procedure and sample size for Anthropometric and mortality data Sample size calculation for anthropometry The sample size was calculated using the ENA for SMART software package. The number of children under 5 (U5) in the survey area was estimated at 20% of the total population (26,014 children). In addition to the under-five population data, the following basic assumptions were made: 1) The estimated prevalence of malnutrition is 28.5 % 5 ) 2) The design effect is 2 and the standard margin of error is 5% (95% CI). 3) The number of children less than 5 years per household is estimated at ) The average number of persons per household is 6 and 1 mother per household. To calculate number of clusters to visit, the total sample for anthropometry and IYCF was used. Number of households (645) was divided by number of HH to be reached per day (17) resulting to 38 clusters Sample size calculation for mortality The sample size for mortality was determined by using the following factors: Total population of 137,991 The estimated Crude Mortality Rate (CMR) was 0.38 deaths/10,000 persons /day (Wajir south May 2011 nutrition survey). Level of the desired precision was 0.3% Design effect was 2 Recall period of 90 days Average family size of 6 ENA/SMART software automatically calculated the sample size for mortality as 3924 persons using the above-mentioned data. Considering average household size of 6 persons, and expected non response rate of 3%, ENA further generated 578 households as the minimum sample size. As the sample size for anthropometry (645 HHs) was larger than the sample size for mortality 578 HHs), it was decided to use the higher sample size higher precision for mortality results. Given the operational circumstances and based on the fact that one cluster had to be completed in a day, 17 households were estimated to be visited in one day which yielded 38 clusters (645/17). A recall period of 90 days was selected to enable the respondents recall the number of deaths without bias. The retrospective data was obtained within a period of 3 months specific dates being 20 th October 2011(Kenyatta Day) to 20 th January first day of the survey data collection. 3.2 Sampling procedure: selecting households and children EPI 7 method was utilized to select the households. A household was defined as a group of people who sleep under the same roof and eat from one cooking pot. Members of a household may not necessarily 5 GAM rates May 2011 Wajir South nutrition survey 6 From May 2011 Wajir South nutrition survey 7 SMART Methodology,

15 be related to one another. If there were several structures within the same compound but each had their own cooking pot, then they were considered as separate households. Each survey team with the help of a village guide per cluster moved to the centre of the village where they spun a pen and walked to the direction the point of the pen faced to the edge of the village. While at the edge of the village, the team spun a pen again and walked along this second line counting each house on the way. Using random numbers list, the team selected the first house to be visited by randomly selecting a number between 1 and the total number of households counted. The randomly selected number became the first household to be visited and thereafter they moved in the right hand direction after every household. The respondent was the primary caregiver of the child/children. For the empty households the team confirmed from the neighbours if the owners were coming back the same day, they noted on the absentees form and went back later. In case the household members had migrated the teams noted and walked to main entrance where they selected the next household by moving to the right hand direction from that household. In the event that the team reached the end of the cluster and had not attained the target households, a different direction was determined by randomly spinning a pen and the process repeated until the expected households per cluster was achieved. All children aged between 6 and 59 months of the same household were included in the survey for anthropometric measurements, while all children age between 0-23 months from the same household were included in the Infant and young child feeding practices(iycf) survey. If the children were absent their names were noted in the absentees form and the teams returned back later to the same household to measure the children. A retrospective mortality questionnaire using the current household census method was conducted for all the households sampled in each cluster, regardless of the presence of children under five years. 3.3 Case definitions and inclusion criteria Age: All children aged 0-23 months in the households visited were included in the IYCF survey while all children 6-59 months in the households visited were included in the anthropometric survey. Caregivers were requested by the survey teams to produce the Child Welfare Clinic (CWC) cards or birth certificates at the beginning of the interviews to verify age and other information required i.e. Immunisation and Vitamin A supplementation status. In case the care givers did not have the CWCs cards or birth certificates, age was determined using a local events calendar which was developed during the survey teams training based on remarkable events that had taken place and could be recalled by most respondents in the community(see annex 4). Age for children 0-59 months was recorded in months. Weight: Children were measured using the electronic scale (UNISCALE 8 ). All children were weighed with minimal clothing and no shoes. In case of children who could not stand on the scale, the caregivers were requested to stand on the weighing scale then after their weight was noted the scale was switched to read the weight of the child. Weight was read to the nearest 0.1kg Height: was measured using the height board to the nearest 0.1cm. Recumbent length was taken for children less than 87 cm or less than 2 years of age while those greater or equal to 87 cm or more than 2 years of age were measured standing up. All children were measured with no shoes, hijabs and caps. Height was taken by two measurers while the third staff recorded the readings. Bilateral oedema: Oedema was assessed by the application of normal thumb pressure for at least 3 seconds to both feet at the same time. The presence of a pit or depression(with supervisor s verification) on both feet was recorded as oedema present and no pit or depression as oedema absent. MUAC: Mid Upper Circumference (MUAC) was measured at the midpoint of the left upper arm to the nearest 0.1cm using the standard MUAC tapes. Morbidity data: Information on two-week morbidity prevalence was collected by asking the mothers or caregivers if the child had been ill in the two weeks preceding the survey and including the day of the 8 Recommended as a valid weighing instruments in the guidelines for nutrition and Mortality assessments in Kenya 15

16 survey. Illness was determined based on respondent s recall and was not verified by a health worker. Immunization status: For all children 6-59 months, information on Pentavalent 1 and OPV 1 and Pentavalent 3 and OPV 3 and measles vaccination was collected using Child Welfare Clinic cards and recall from caregivers. The vaccination coverage was calculated as the proportion of children immunized based on CWC records and recall. Further explanations were given to the caregivers to aid her in recalling i.e. the dosage whether injection or oral and the age of immunization. BCG: For all children 6-59 months, the information was collected by checking whether the characteristic BCG scar was present or not. Vitamin A supplementation status: For all children 6-59 months of age, information on Vitamin A supplementation was collected using the child welfare cards and recall from caregivers. Information on how many times the child had received supplementation in the last 6 months was collected. Vitamin A capsules were also shown to the mothers to aid in recall. De-worming status: Information was solicited from the care takers as to whether their child/children 6-59 months had been de-wormed in the last 3 months. A local calendar of events was used to refer to 3 months recall period Sedax bilood aan sodhafney. Samples of Deworming tablets were also shown to aid the caregivers in recall. Water and sanitation: information on access to clean drinking water and toilet facility and disposal of children faecal waste was collected in all households visited. Food security, livelihood and dietary status: Information on livelihoods activities, source of food, registration for general food distribution and coping strategies was collected in all the households visited. Mortality data: Retrospective mortality data was collected using the current household census method in all the visited households, including those with no children aged less than five years old. The recall period was 90 days starting from 20 th October 2011, Mashujaa day. Information was collected on the age and sex of the household members, the number of household members present within the recall period, the number of persons who joined or left within the recall period, and the number of births and deaths over the recall period. The presumed causes of death were recorded based on the following case definitions: 1= Diarrhoea (minimum of 3 watery stools/24hrs). 2= Bloody Diarrhoea; 3= Measles (fever with rash); 4= Fever; 5= Lower respiratory tract infection (fever, productive cough, chest pain, difficulty breathing) 6= Malnutrition; 7= Injury; 8= Other (Specify); 9= Unknown 3.4 Questionnaire, training and supervision The overall survey was coordinated and supervised by Save the Children nutrition coordinator, 1 Ministry of Health DHMT member, I Save the Children nutrition officer and 1 UNICEF Nutrition Support Officer. For data collection, a total of 6 teams were recruited for the survey. The survey teams comprised 1 team leader and 3 enumerators/measurers. The 12 enumerators were selected through an interview process while team leaders were staffs from line ministries including 3 MOH staffs, 1 Ministry of Agriculture staff, 1 staffs from District Development Office and 1 staff from Arid Lands Resource Management Project II (ARLMP). The survey was preceded by a 4 days thorough training by 3 Save the Children and 1 MOH staffs and covered the following topics;- A brief introduction on Signs and symptoms of malnutrition, conceptual framework of malnutrition; The objectives and purpose of the survey; 16

17 Roles and responsibilities of the survey staffs The survey design and methodology; Sampling procedure; Household selection, data collection procedures, accurate completion of questionnaires and interview skills; How to take anthropometric measurements using standardized procedures. Standardization test for the measurements were conducted; Development of a calendar of events ; and Pre-testing of the questionnaires, translation into local language and the data collection procedure were undertaken before the actual survey. This was followed by the review of the questionnaires based on the feedback from pre-testing where no changes were made. During training, a standardization test was also done whereby all survey team members (18) and the supervisors measured ten (10) same children each two times. Based on the results of the first test the precision and accuracy were not good and hence, the whole test was re-done after which the survey teams were formed based on individual team member s strength. Based on the standardization report, team members with difficulty in taking measurements were retrained again until it was ascertained they could take accurate measurements. Teams were closely supervised by 3 supervisors (3 Save the Children staff and 1 Deputy District Public Health Nurse (D.DPHN) throughout the survey period. The supervisors were also trained on how to control the quality of work and collect qualitative data. Teams were trained to check anthropometric measurements twice before recording and ensured that all questions had been filled before leaving each household. At the end of every day during the de-briefing session, all the questionnaires were checked and any errors or omissions in data recording were verified and if information was missing the particular households were revisited. 3.5 Data analysis Data entry was done by 4 data entry clerks who were supervised by Save the Children Monitoring and Evaluation Officer. The data entry clerks also participated in the four days initial training to be able to understand all the parameters in the questionnaire. Anthropometry data was entered on ENA for SMART 2011 version on a daily basis and the plausibility checks done. Any irregularity in the data was discussed with the team the following day in the morning before they set off to the field. Due to vastness of the district teams spent in the field but questionnaires were sent to the base on a daily basis. Anthropometric, mortality and quantitative data entry, cleaning, processing and analysis was conducted using EPI info version 3.5.1, ENA for SMART 2011 and SPSS version 16 software. The software flagged off any extreme, potentially incorrect or out of range values. Additional analysis for frequencies was conducted using windows SPSS and MS excel. 3.6 Nutritional indices Acute malnutrition indices: Weight-for-height (WFH) index Acute malnutrition rates were estimated from the weight for height (WFH) index values and oedema. The values were derived from comparison of children in the survey to WHO 2006 references and are reflective of current nutritional conditions. WHI indices were expressed in Z-scores. Table 9: WFH indices Weight for Height z-score Global Acute Malnutrition <-2 SD and/or oedema Moderate Acute Malnutrition <-2 SD and -3 SD Severe Acute Malnutrition <-3 SD and/or oedema Global acute malnutrition was therefore defined as the proportion of children presenting with a weight for height index less than -2 Z scores. 17

18 MUAC The guidelines used are as follows: Table 10: MUAC indices MUAC <11.5 cm MUAC 11.5 cm and <12.5cm MUAC 12.5cm and <13.5cm MUAC 13.5cm severe acute malnutrition and high risk of mortality moderate acute malnutrition at risk of moderate malnutrition adequate nutritional status Chronic Malnutrition Index: Height-for-Age (HFA) -Stunting Chronic malnutrition rates were estimated from the height-for-age (HFA) index values. The HFA indices were compared with WHO standards and are reflective of long-term/ chronic malnutrition. HFA indices are expressed in Z-scores. Table 11: Height for Age indices Global Chronic Malnutrition Moderate Chronic Malnutrition Severe Chronic Malnutrition Height for Age z-score <-2 SD <-2 SD and -3 SD <-3 SD Global chronic malnutrition was therefore defined as the proportion of children presenting with a weight for age index less than -2 Z scores. Mortality Indices The crude mortality rate (CMR) was determined for the entire population surveyed from the time of recall period to the time of survey. His was calculated using the SMART software. The proportion of deaths among children under-five years of age (U5MR) is also calculated the same way using the under-five population data. The thresholds are defined as follows: Table 12: Mortality Thresholds Total population CMR Under-five population U5MR Alert level: 1/10,000 people/day 2/10,000 children/day Emergency level: 2/10,000 people/day 4/10,000 children/day Survey Data Validation Process Close supervision of data entry was done on daily basis and data cleaned and validated at the end before actual analysis. Validation of the data was based on the following parameters: Out of usual range values flags Age and sex distribution Digit preference scores The standard deviation Skewness (This is a measure of the degree of asymmetry of the data around the mean) Kurtosis (This shows the relative flatness of the data compared to a normal distribution). During data analysis, the ENA for SMART software flagged any missing data, extremes or potentially 18

19 incorrect z-scores values. All flagged z-scores were excluded from the analysis. A triangulation of information was done to verify the findings of the quantitative data; information was collected through key informants, secondary data and observation. The survey results were also presented to the Nutrition Information Technical Working Group (NITWG) under the Nutrition technical Forum (NTF) for validation during which the survey was validated. 19

20 4.0 Results 4.1 Survey sample description Table 13: Survey Sample Selection Number of children 6-59 months surveyed 863 Number of children 6-59 months analyzed by WHO 847 Number of children 6-59 months analyzed by NCHS 847 Number of total population surveyed for mortality 3724 Number of children under five surveyed for mortality 1024 Number of HH covered in the mortality survey 663 Number of persons who joined the household during the recall period 93 Number of persons who left the household during the recall period 33 Number of under five children who joined the household during the recall period 17 Number of under five children who left the household during the recall period 17 Number of births during the recall 10 Average Number of persons per HH 7 Average Number of children per HH 2 % of children under five in the population 27.5% Of all the households visited with children between 6-59 months, 10 children were absent. The children had either been taken to the grazing lands baadia following the livestock movements in order to access milk or to their grandmothers who took care of them when their mother gave birth to younger siblings. 4.2 Anthropometric results (based on WHO standards 2006): Table 14: Acute Malnutrition definitions: WFH z-score MUAC Global Acute Malnutrition < -2 SD and/or oedema <12.5 CM and/or Oedema Moderate Acute Malnutrition < -2 SD and -3 SD 11.5cm and <12.5cm Severe Acute Malnutrition < -3 SD and/or oedema <11.5cm and /or oedema Table 15: Distribution of age and sex of sample Boys Girls Total Ratio AGE (mo) no. % no. % no. % Boy: girl Total Of the children measured 51.9% were boys while 48.1% were girls. The overall sex ratio was 1.1 which is within the recommended range of

21 Figure 2: Population age and sex pyramid There was a slight overrepresentation of boys aged 6-17, and months which may be attributed to age recall bias as age determination in the absence of birth certificate and child welfare cards was based on local events calendar. Compared to the previous surveys, this does not represent a typical pattern for those age groups. Table 16: prevalence of malnutrition based on weight for height z-scores (and/oedema) and by sex All n = 847 Boys n = 440 Girls n = 407 Prevalence of global malnutrition (<-2 z-score and/or oedema) (196) 23.1 % ( (113) 25.7 % ( (83) 20.4 % ( Prevalence of moderate malnutrition (<-2 z-score and >=-3 z-score, no oedema) Prevalence of severe malnutrition (<-3 z-score and/or oedema) The prevalence of oedema is 0.0 % 95% C.I.) (156) 18.4 % ( % C.I.) (40) 4.7 % ( % C.I.) 95% C.I.) (86) 19.5 % ( % C.I.) (27) 6.1 % ( % C.I.) 95% C.I.) (70) 17.2 % ( % C.I.) (13) 3.2 % ( % C.I.) Table 17: Prevalence of acute malnutrition by age, based on weight for height z-scores and/or oedema Severe wasting (<-3 z-score) Moderate wasting (>= -3 and <-2 z- score ) Normal (> = -2 z score) Oedema Age Total No. % No. % No. % No. % (mo.) no Total

22 Table 18: Distribution of acute malnutrition and oedema based on weight-for-height z-scores Oedema present Oedema absent <-3 z-score >=-3 z-score Marasmic kwashiorkor Kwashiorkor No. 0 No. 0 (0.0 %) (0.0 %) Marasmic No. 40 (4.7 %) Not severely malnourished No. 807 (95.3 %) Figure 3: Weight for height z-scores Table 19: Prevalence of underweight based on weight-for-age z-scores by sex Prevalence of underweight (<-2 z-score) Prevalence of moderate underweight (<-2 z-score and >=-3 z-score) Prevalence of severe underweight (<-3 z-score) All n = 847 (168) 19.8 % ( % C.I.) (135) 15.9 % ( % C.I.) (33) 3.9 % ( % C.I.) Boys n = 440 (96) 21.8 % ( % C.I.) (79) 18.0 % ( % C.I.) (17) 3.9 % ( % C.I.) Girls n = 407 (72) 17.7 % ( % C.I.) (56) 13.8 % ( % C.I.) (16) 3.9 % ( % C.I.) The weight for Age (WFA) indices which measures underweight is a combination measure of both acute and chronic malnutrition (wasting and stunting) therefore it indicates a reflection of both current and past nutritional experience of the community. A Percent of underweight was seen in boys than girls from the community surveyed. 22

23 Table 20: Prevalence of underweight by age, based on weight-for-age z-scores Severe underweight (<-3 z-score) Moderate underweight (>= -3 and <-2 z- score ) Normal (> = -2 z score) Oedema Age Total No. % No. % No. % No. % (mo.) no Total Table 21: Prevalence of stunting based on height-for-age z-scores and by sex Prevalence of stunting (<-2 z-score) Prevalence of moderate stunting (<-2 z-score and >=-3 z-score) Prevalence of severe stunting (<-3 z-score) All n = 847 (164) 19.4 % ( % C.I.) (126) 14.9 % ( % C.I.) (38) 4.5 % ( % C.I.) Boys n = 440 (87) 19.8 % ( % C.I.) (66) 15.0 % ( % C.I.) (21) 4.8 % ( % C.I.) Girls n = 407 (77) 18.9 % ( % C.I.) (60) 14.7 % ( % C.I.) (17) 4.2 % ( % C.I.) The height-for-age (HFA) indices assesses linear growth and hence it reflects the cumulative effects of chronic nutritional inadequacy and/or recurrent chronic illness which may result in a child having a low HFA (referred to as stunting) indicated by shortness when compared to his/her age cohorts. It is not affected by seasonality but is rather related to the effects of socio-economic development and longstanding food security situation in a community. The results (Table 21) indicate a global stunting rate of 19.4% ( % CI) and severe stunting rate of 4.5 %( %CI ). However this is higher than that of the last survey where the Global stunting was at 12.2% ( CI) and Severe stunting at 2.3% ( CI) 95% which may be indicative of a worsening situation. According to the results boys are more affected than girls though not significantly different Table 22: Prevalence of stunting by age based on height-for-age z-scores Severe stunting (<-3 z-score) Moderate stunting (>= -3 and <-2 z- score ) Normal (> = -2 z score) Age Total No. % No. % No. % (mo.) no Total

24 Mean Weight for height Z scores WHO Standards 2006 The mean weight-for-height Z scores was -1.20±01.14 with a design effect of 1.77 whereas the weightfor-age mean Z scores -1.25±0.94 with a design effect of 1.74 and the height-for-age was 0.79±1.30 with a design effect of There were no Z scores that were not available nor out of range as shown in the table below. Table 23: Mean z-scores, Design Effects and excluded subjects Indicator n Mean z- Design Effect z-scores not z-scores out scores ± SD (z-score < -2) available* of range Weight-for-Height ± Weight-for-Age ± Height-for-Age ± * contains for WHZ and WAZ the children with oedema. Table 24: Plausibility Checks for Anthropometric Data Indicator Survey Digit preference weight 0 (5) Digit preference height 2 (7) WHZ (Standard deviation) 0 (1.07) WHZ (Skewness) 0 (0.20) WHZ (Kurtosis) 0 (0.10) Percentage of flags 1.1% Age ration: ages 6-29: Overall sex ratio: p-value = (boys and girls equally represented) Overall age distribution: p-value = (significant difference) Overall age distribution for boys: p-value = (as expected) Overall age distribution for girls: p-value = (significant difference) Overall sex/age distribution: p-value = (significant difference) 4.3 Mortality results (retrospective over 3 months/days prior to interview) Table 25: Mortality rates CMR (total deaths/10,000 people / day): 0.30 ( ) (95% CI) U5MR (deaths in children under five/10,000 children under five / day): 0.54 ( ) (95% CI) Information on mortality was collected. A total number of 3274 people were included in the mortality survey. Out of these, 1024 of them were children below 5years of age. Of all deaths reported during the survey 2 (20%) deaths occurred in children under five years while the other 8 (80%) deaths occurred in persons over five years. Both mortality rates were within the acceptable levels for emergency situations. The causes of deaths for the children under five years were due to diarrhoea. Moreover, causes of death for adults were lower respiratory tract infections (2), injury (1), others (ageing -1, suspected dengue fever -1) and unknown causes (3). 24

25 4.4 Children s morbidity The prevalence of reported illness was determined based on a two-week recall period (this was inclusive of the day of the survey). As shown in the table below, more than a third (31.5%) of the children (267) under five years were cough (63%), diarrhoea (60%), fever 94%) and vomiting (34%). Upsurge of pneumonia and diarrhoea cases was also observed in outreach mobile clinics, health facilities and stabilization centre. Table 26: Prevalence of reported illnesses in children in the two weeks prior to interview (n=267) Illness Number 6-59 months Diarrhoea % Cough % Fever % Measles 0 0% Other(vomiting) 90 34% Eye infection 10 4% Stomach ache 7 3% Malnutrition 6 2% Skin infections 5 2% Figure 4: Symptoms breakdown in the children in the last two weeks prior to interview (n=267) 25

26 Figure 5: Caretaker s health seeking behaviour Health care seeking behaviours determines the preference and quality of health services obtained whenever a child falls ill. Quality health of health services and the duration before a sick child receives medical attention contributes to the severity of the illness. Of the 267 children who were reported to be ill within the last two weeks, only 1% who did not seek ant medical assistance. However 4% reported to have sought assistance from traditional healers. The Percent of those who did not seek assistance had greatly improved from 16% in the previous survey to 1% reported. Table 27: Vaccination, deworming and Vitamin A coverage BCG SCAR Present Absent MEASLES 9-59 MONTHS By card According to caretaker OPV 1 By card According to caretaker OPV 3 By card According to caretaker De-worming Given Once Given twice Vitamin A supplementation (in Once the last 6 months)-6-11 Months Twice Vitamin A Months Once Twice 96.3% (817) 3.7% (31) 51.20% (434) 44.14% (374) 54.8% (468) 42.3% (359) 54.2% (460) 42.1% (357) 50.4% (427) 29.0% (246) 70% (61) 20% (17) 61% (466) 34% (257) By using both Child Welfare Clinic (CWC) card and caretakers recall from the caregivers, there is good coverage for all the vaccinations, which are above the recommended Kenya Expanded Programme on Immunization (KEPI) target of 80%. Only 42.7% respondents reported to have immunization cards. This is an improvement compared to the same survey conducted last year which was at 25%.Deworming coverage had greatly improved from last survey that is from 34.6% in May 2011 to 79.4%.This may have been attributed by the scale up of interventions during the emergency drought. However approximately 42% of the coverage reported here was based on recall and not evidenced by and EPI card. 26

27 Figure 6: Vitamin supplementation coverage Of the children surveyed, over half (56%) reported to have received Vitamin A supplementation once while slightly less than a third (30%) reported to have received supplementation twice in the last one year. Compared to the previous survey there was an improvement which may have been attributed to the two Malezi bora sessions in 2011 and the emergency Blanket Supplementary Feeding programme which was covering all children under five years, pregnant and lactating mothers. 4.5 Water and sanitation Almost half (46%) of the households surveyed said boreholes were their main current sources of water, with 27% getting water from water pans which had been filled by the short rains experienced between October and early December 2011.eleven Percent (11%) reported to be receiving water from water trucking as shown in the graph below. Figure 7: source of water 27

28 Figure 8: Time taken on water collection Water treatment Majority (88.7%) of the households interviewed did nothing to their water before drinking. This may have been the major contributor to high incidences of diarrhoea and vomiting reported both in the health facilities and outreach sites mobile clinics. It s worth noting that some locations reported to be using PUR to purify water from the water pans However there is an improvement compared to last survey report of 98% who had been reported not doing anything to water before drinking. Figure 9: Methods of water treatment 3. 28

29 Access to toilet facilities Slightly less than a third (26%) reported having access to a toilet facility which was either their own, neighbour or a public toilet.71% of those who reported not to have access 74.8% reported to be defecating in the bush. Figure 10: Types of toilet facilities Handwashing Only 22% of the households surveyed reported to be washing their hands with water and soap while and 0.2 % washed with water and ash. The rest wash with water only. The community reported to their hand washing times to be as shown in the table below; Table 28: Hand washing time Hand washing time Count % After visiting the toilet % Before feeding the child % Before eating % Before preparing food % When dirty % When water available % After cleaning children s bottoms % Other (Before prayers) % 29

30 5.0 Food Security and Utilization 5.1 Infant and Young Child Feeding Practices Information on Infant and young child feeding practices (IYCF) based on a 24-hour recall, in line with WHO guidelines to minimize recall bias and thus obtain a more valid data. Information on breastfeeding practices was obtained for children aged 0-23 months. Figure 11: Initiation of breastfeeding All mothers reported to have breastfed their children. As shown above on 28% of the children were reported to have been put on breast in the first hour of birth. Figure 12: Exclusive breastfeeding in the last 24 hours Half the proportion of children 0-5 months was reported to have been exclusively breastfed in the last 24 hours as indicated in the graph above. 30

31 Dietary Diversity For this indicator only a third (33%) mothers who reported to have practiced a dietary diversity of three or more food groups while feeding the children with complementary foods. Most of them reported that they fed their children with what was commonly available and accessible in the market. Table 29: Dietary diversity amongst children 6-23 months Dietary diversity-6-23 months No. Percent 3+ Food groups breastfed 78 33% 4+ Food groups non-breastfed 1 0% Minimum meal times for children 6-23 months For the average healthy breastfed infant, complimentary foods 9 time should be 2-3 times per day at 6-8 months, 3-4 times per day at 9-11 months of age, with additional snacks offered 1-2 times per day as desired (WHO IYCF participants manual pp ) Table 30: Minimum meal times months Minimum meal times No. Percent At least 2 times a day-6-8 months 32 13% At least 3 times a day-6-23 months % 4 times for non-breastfed 1 0% 5.2 Food diversity at household levels According the t survey the main source of food was purchase at 81.3% with 43.9% indicating livestock sale as the main source of income. Food diversity analysis done on the survey findings was based on the foods consumed in the last 7 days and 24-hour recall period. The graphs below show that most common foods consumed by households were;-cereal based foods (mainly pasta, maize, maize meal and Corn Soya Blend (CSB), milk and milk products, sweets, oils/fats and pulses/legumes. Availability of milk and milk products was as a result of heavy rainfall experienced in the country between October and December 2011.However the milk supply was reported to be low as many livestock s did not calf in the previous year due to the drought experienced. Despite the fact that Wajir South community was a pastoralist community, only a small proportion of households consumed meat. The low consumption of meat was explained by lack of butcheries in most of the locations visited. Moreover most of the animals had died due to the prolonged drought experienced previously and the community valued the only few left for bringing back the life of their livestock s herds hence were not keen in slaughtering them. Households also cited high food prices as a result of inflation and closure of border points had led to consumption of less preferred foods. This in turn led to reduction in sizes and number of meals consumption by the households. Of the households interviewed, 74% had received food aid from WFP in the last three months. However food aid delivery was not done for two months (October and 9 Complimentary food any foods given to children from 6 months of age while they are still breastfeeding 31

32 November 2011) due to pipeline breakdown as a result of heavy rains that rendered roads impassable hence hampering the delivery of food aid commodities both to and from the warehouse. Figure 12 shows that almost all (99.1%) the respondents reported to have used cereals for the last seven days prior to the survey, followed by sweets (91.2%), milk and milk products (90.2%) and oils/fats (89.8%).These results indicates that this pastoralist community was able to access milk due to pastures regeneration. On the other hand, the proportion of respondents that consumed meat was quite low at 18%, those that consumed green leafy and vitamin A rich vegetables are negligible at 5.4% and 4.5% respectively. Figure 13: Food consumption 7 days and 24-hour recall. 5.3 Main livelihoods activities The survey showed that 38.4% of the households main livelihood activities were livestock herding. This was followed by petty trade at 10.2% which included sale of firewood and small kiosks and sale of milk and milk products near trading centres. Waged /casual labour at 9.6% which mainly included use of donkey carts services 32

33 Figure 14: Main Livelihood activities 6.0 Discussion 6.1 Nutritional status A total number of 847 children aged between 6-59 months were measured and included in the analysis of this survey. The overall age sex ratio was 1.04 which is within the acceptable range of The GAM rates at 23.1% ( ) 95 CI scores are above the emergency threshold of >15% 10. Compared with the survey undertaken in May 2011 which indicated GAM of 28.5% ( % CI) and SAM of 4.6% ( ), the acute malnutrition rates have reduced though not statistically significant due to overlapping confidence intervals. Figure 15: Comparison of malnutrition rates based on WFH Z-scores in Wajir South ( ) 10 Interpretation of level; Global Acute malnutrition(gam): prevalence of GAM <5 termed as acceptable, 5-9% poor, 10-14% serious and > 15 critical 33