April 2011

Integrated SMART Survey Nutrition, WASH, Food Security and Livelihoods Kitui District Kenya

Funded by

Table of Contents List of Tables ..................................................................................................................................................................3 List of Figures .................................................................................................................................................................3 Abbreviations.................................................................................................................................................................4 Acknowledgements .......................................................................................................................................................5 1. Executive Summary ...............................................................................................................................................6 2. Background information .......................................................................................................................................9 3. Survey Objectives ................................................................................................................................................10 4. Methodology.......................................................................................................................................................11 4.1. Sampling .....................................................................................................................................................11 4.2. Training and organization of survey teams ................................................................................................12 4.3. Data Quality Assurance Processes .............................................................................................................12 4.4. Data Collection ...........................................................................................................................................12 a) Anthropometric Indicators: ............................................................................................................................12 b) Mortality .........................................................................................................................................................13 c) Food Security and WASH ................................................................................................................................13 4.5. Data Entry and Analysis..............................................................................................................................13 a) Analysis of Acute Malnutrition .......................................................................................................................14 b) Analysis of Retrospective Mortality ................................................................................................................14 c) Additional health information ........................................................................................................................14 d) Food Security and Livelihoods ........................................................................................................................14 e) WASH ..............................................................................................................................................................15 5. Results & Discussion ...........................................................................................................................................15 5.1. Socio-Demographic Characteristics ............................................................................................................15 5.2. Nutritional Status .......................................................................................................................................16 5.3. Retrospective mortality ..............................................................................................................................19 5.4. Morbidity status, coverage of Vitamin A and Measles Immunization .......................................................20 5.5. Health seeking behavior and maternal & child care practices ...................................................................20 5.6. Food Security and Livelihoods ....................................................................................................................22 5.7. Water and sanitation .................................................................................................................................24 6. Conclusion and Recommendations.....................................................................................................................27 Annex 1: SMART Survey Anthropometric Form (April 2011) 6-59 months old children ............................................29 Annex II: Calendar of Events ........................................................................................................................................30 Annex III: Cluster Mortality Questionnaire ..................................................................................................................31 Annex IV: WASH and Food Security and Livelihood Questionnaire.............................................................................32 Annex V: Anthropometric data plausibility check (WHO) ...........................................................................................37

Action Against Hunger (USA), Integrated SMART Survey, April 2011 Kitui District, Kenya

2

List of Tables Table I: Summary of key Findings ................................................................................................................... 7 Table II: Demographic characteristics ........................................................................................................... 15 Table III: Distribution of age and sex of sample............................................................................................ 16 Table IV: Prevalence of acute malnutrition based on weight-for-height z-scores (and/or oedema) and by sex (WHO standards) ..................................................................................................................................................... 16 Table V: Prevalence of acute malnutrition based on weight-for-height z-scores (and/or oedema) and by NCHS ...................................................................................................................................................................... 16 Table VI: Prevalence of acute malnutrition by age based on weight-for-height z-scores and/or oedema WHO standards. ..................................................................................................................................................... 17 Table VII: Prevalence of acute malnutrition based on the percentage of the median and/or oedema ....... 17 Table VIII: Prevalence of malnutrition by age, based on weight-for-height percentage of the median and oedema ...................................................................................................................................................................... 18 Table IXI: Prevalence of GAM and SAM by MUAC ........................................................................................ 18 Table X: Prevalence of underweight based on weight-for-age z-scores by sex ............................................ 18 Table XI: Prevalence of underweight by age based on weight-for-height z-scores and oedema................. 19 Table XII: Prevalence of stunting based on height-for-age z-scores and by sex ........................................... 19 Table XIII: Prevalence of stunting by age based on height-for-age z-scores ................................................ 19 Table XIVI: Mean z-scores, Design Effects and excluded subjects ................................................................ 19 Table XV: Vitamin A supplementation, Measles Immunization Status, OTP/SFP and Morbidity ................. 20 Table XVI: Health Seeking Behaviour ............................................................................................................ 21 Table XVII: Maternal and children feeding practices. ................................................................................... 21 Table XVIII: Dietary Diversity Indicator for Children 6 – 23 Months ............................................................. 22 Table XIX: Source of food.............................................................................................................................. 22 Table XX: Coping Strategy ............................................................................................................................. 23 Table XXI: Livestock holding ......................................................................................................................... 23 Table XXII: Household Dietary Diversity Score ............................................................................................. 24 Table XXIII: Market Price .............................................................................................................................. 24 Table XXIV: Distance to Water Source .......................................................................................................... 25 Table XXV: Water treatment ........................................................................................................................ 25 Table XXVI: Hand Washing Practices ............................................................................................................ 26 Table XXVII: Household disposal of solid and human waste ........................................................................ 26

List of Figures Figure I: Areas Surveyed in the district. ........................................................................................................ 11 Figure II: Household head level of education ............................................................................................... 15 Figure III: GAM and SAM graph (WHO) ........................................................................................................ 17 Figure IV: Ranking for source of food ........................................................................................................... 23 Figure V: Source of Water............................................................................................................................. 25

Action Against Hunger (USA), Integrated SMART Survey, April 2011 Kitui District, Kenya

3

Abbreviations ALRMP ASAL BSFP CI CSB CTC DNOs ENA FFA GAM GFD HDDS IYCF KAP KFSSG MOH MUAC NGOs OTP SC SFP SMART VCT WFH WFP WHO

Arid Lands Resource Management Project Arid and Semi-Arid Lands Blanket Supplementary Feeding Programme Confidence Interval Corn-Soya Blend Community Therapeutic Care District Nutrition Officers Emergency Nutrition Assessment Food for Assets Global Acute Malnutrition General Food Distribution Household Dietary Diversity Score Infant and Young Child Feeding Knowledge, Attitudes and Practices Kenya Food Security Steering Group Ministry of Health Middle Upper Arm Circumference Non-governmental Organizations Outpatient therapeutic Programme Stabilization Centre Supplementary Feeding Programme Standardized Monitoring and Assessment of Relief and Transition Voluntary Counselling and Testing Weight for Height World Food Programme World Health Organization

Action Against Hunger (USA), Integrated SMART Survey, April 2011 Kitui District, Kenya

4

Acknowledgements ACF Kenya Mission would like to acknowledge the support of the following: • • •

• •

• • • •

Parents and caretakers for availing their children for assessment as well as for providing other relevant data for the study; UNICEF for providing financial support for the survey; Ministry of Health team in the Larger Kitui district, especially District Nutrition Officers from Mutomo and Kitui Districts. They provided relevant team support and vital information necessary for planning the survey; Team members from Kenya National Bureau of Statistics, The Ministry of Public Health and the Arid Lands Resource Management Project; District Commissioners, District Arid Lands Management Project office, District Health Planning Teams, Chiefs and Sub-chiefs from Kitui, Mutito and Mutomo districts. They all helped in mobilization, availed relevant population data, and introduced the survey team to important contacts. This both smoothed the planning and implementation of the survey; Village elders in Mutomo, Mutito, Ikutha and Mwitika for acting as guides to the teams in survey locations; ACF field staff on ground and in Nairobi for management of personnel, logistics and administrative issues; Data collectors and field survey supervisors for hard work and dedication; and, Survey drivers for timely and efficient transport and delivery.

Action Against Hunger (USA), Integrated SMART Survey, April 2011 Kitui District, Kenya

5

1. Executive Summary Research and anecdotal evidence show that Kitui is, for the most part, a food stressed district. Different studies have identified Kitui as a borderline food insecure district that easily regresses to food crisis after slight shocks. Food emergency operations have been conducted in the district since 2004. In 2009, when the last ACF study was carried out, emergency operations had to be scaled up due to a prolonged drought that devastated both crops and livestock. The situation stabilized somewhat in 2010 hence reducing the number of beneficiaries in 2011. This trend is exemplified in the number of beneficiaries under emergency food operations: 297, 000 in 2009 and 371 00 in 2011. Furthermore, the Kenya Food Security Steering Group (KFSSG) reports that Mutomo and Mutito Districts harvested 1 less than 15% of anticipated crop yield while areas like Athi and Ikutha experienced total crop failure . Despite reports that the prevailing nutritional status is stable cases of acute malnutrition have indicated in the marginal mixed farming livelihood zone. The nutrition survey was implemented using the Standardized Monitoring and Assessment of Relief and Transitions (SMART) methodology in Mutomo and Mutito districts the larger Kitui. Mutomo and Mutito districts fall under the Marginal farming livelihood zone. The specific sample areas studied were, Ikutha, Mutomo, Mwitika and Mutito. The survey was implemented in collaboration with the Ministry of Health (MoH), Arid Lands Resources Management Project (ALRMP) and the Kenya National Bureau of Statistics (KNBS). A 4 day SMART Methodology , training took place from April 13 to 16 2011, while data collection was carried out between April 18 and 29, 2011.

Methodology Two-stage cluster sampling with probability proportional to size (PPS) methodology was used. Population data was obtained from chiefs, sub-chiefs and village elders. Emergency Nutrition Assessment (ENA) for SMART software was used in determining the sample size using results of October 2009 Nutrition Survey. The survey results reported a GAM of 8.9% (7.0 – 10.9) and SAM of 1.5% (0.7 – 3). The upper limit of 10.9, precision of 3.5, a design effect of 2.0 and a 3% non-response rate were fed into ENA resulting in 609 children, (482 HH). For mortality, a prevalence of 0.15 per 10,000/day, precision of 0.18 and design effect of 2 resulted in a sample size of 477 households (population=2964). The maximum number of households (482) was used in the sample. This was translated into 37x13 cluster design with an overall sample size of 482 households with an estimate to cover 13 households each day. In the second stage, household selection was done using systematic random sampling. Lists of all households in respective clusters were provided by village elders. The total number of households in each village or cluster was divided by the number of households that could be visited by a team in a day (13) to determine the sampling interval. A random number was then chosen to select the first household and the sampling interval repeatedly added to determine the remaining sample households. Respondents were primarily heads of households and their spouses.

Survey Implementation Five survey teams, each comprising a team leader and four data collectors were constituted. The five team leaders were from the Ministry of Health (4) and KNBS (1).

1

KFSSG Long Rain Assessment, August 2010.

Action Against Hunger (USA), Integrated SMART Survey, April 2011 Kitui District, Kenya

6

th

th

A four day (13 to 16 April 2011) intensive training on SMART methodology was conducted. The field data th th collection took place on 18 to 29 April, 2011, covering 37 clusters and 13 households per cluster. The anthropometric and mortality data were entered and analyzed using ENA, October 2008 version. Food Security and Water & Sanitation data entry was done in SPSS version 12.

Survey Results A total of 678 children (351 male and 327 female) aged 6-59 months and for mortality 4,016 people from 561 households were surveyed. The mean household size and number of Under five children per household was 7.2 and 1.3 respectively. Global acute malnutrition (GAM) was 6.5% (95% CI: 4.5 – 9.3), severe acute malnutrition (SAM) was 0.9% (95% CI: 0.4-2.2) and there was one case (0.1%) of oedema. The Crude Death Rate finding was 0.08 (0.03-0.24) per 10,000 persons per day, below the emergency threshold of 1/10,000/day. Over half (58.7%) of the surveyed children had some form of illness in the two weeks prior to the survey an indication of a high incidence of disease. The most commonly reported illnesses included fevers, one associated with malaria (32%) and the other with cough (27.4%). Diarrheal incidents were reported at 12.4 %. Illness exacerbates a poor nutritional status which in turn reduces the body’s ability to utilize nutrients. Measles vaccination coverage was adequate at 70.1%, by card, as was coverage for vitamin A supplementation (55.5%). This coverage however is lower than the targeted 80%. 53.9% (16.2%, protected, plus 37.7%, unprotected) of households get water from shallow wells that mostly do not provide water all year round. 42.9% of the households spend more than one hour to reach a water source. The 55.3% of respondents that treat water several methods: Chemical treatment (34%), boiling (25.3%), decantation (8.4%) and filtration (1.8%). Water and sanitation practices indicated are below SPHERE standards, a potential health hazard and predisposing factor to malnutrition. The major source of food for most purchases was reported as purchases (87.9%), followed by cultivation at 10.9%. The average number of food groups consumed based on the 12 food groups and 24 hour recall period was 4.5. Despite an astonishing 99.5% respondents reporting farming as the major economic activity, only 21.9% reported having food stocks from the previous planting season. Table I: Summary of key Findings Index WHO (n=678)

Z- scores

Z-scores NCHS (n-678) % Median

2

2

Indicator Global Acute Malnutrition 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 W/H < 80% and/or oedema Severe Acute Malnutrition W/H < 70% and/or oedema

Results 6.5% [4.5-9.3] 0.9% [0.4- 2.2] 7.4% [5.2-10.3] 0.1% [0.0-1.1] 2.9% [1.8-4.8] 0.1 [0.0-1.1]

Results in brackets are at 95% confidence intervals

Action Against Hunger (USA), Integrated SMART Survey, April 2011 Kitui District, Kenya

7

Global Acute Malnutrition MUAC 65 cm Severe Acute Malnutrition MUAC > > > > >

Assessing the prevalence of acute malnutrition in children aged 6-59 months Estimating the Crude and under five mortality rates Determine the Infant and Young child feeding practices among children 0 – 23 months. Investigate household food security and food consumption patterns. Estimate Morbidity rates of children 6 – 59 months. Determine the proportion of households with access to safe water.

Action Against Hunger (USA), Integrated SMART Survey, April 2011 Kitui District, Kenya

10

Figure I: Areas Surveyed in the district.

Survey Locations

4. Methodology 4.1.

Sampling

A two-stage cluster sampling design with probability proportional to size (PPS) design was employed for this survey. The Emergency Nutrition Assessment (ENA) software for

Action Against Hunger (USA), Integrated SMART Survey, April 2011 Kitui District, Kenya

11

SMART was used to determine the sample size required. Village level population data were obtained from chiefs, sub-chiefs and elders in respective locations. The October 2009 Nutrition Survey was used to determine sample size which had a GAM rate of 8.9% (7.0 – 10.9 C.I) and SAM 1.5% (0.7 – 3.0 C.I). In this regard, the upper limit, 10.9 was used as highest estimated prevalence, precision of 3.5%, a design effect of 2 and a 3% non response rate giving 609 children (482 households). For mortality, an estimated prevalence of 0.15 per 10,000/day, precision of 0.18 and design effect 1.5. The sample size was determined at 477 households and a targeted population of 2964. The maximum number of households for anthropometry (482) was used. This was translated to 37x13 cluster design with an overall sample size 482 households, as 13 was the estimated maximum number of households a team could survey in one day. In the second stage, selection of household was done using systematic random sampling from a list of households availed by village elders. The total number of households in each village or cluster was divided by the required sample size per cluster (13) to determine the sampling interval. A random number was then chosen between 1 and the sampling interval to select the first household and the sampling interval repeatedly added to determine the remaining households. Respondents were primarily heads of households and spouses. Additional information was collected from the relevant household members.

4.2.

Training and organization of survey teams th

th

A four day intensive training held on 13 to 16 April, 2011 was done for 20 data collectors and five team leaders. The training focused on aspects of the survey implementation, objectives, household selection, MUAC, height and weight measurements and the Food Security and Livelihoods questionnaire. The five team leaders in Kitui were from the Ministry of Health (4) and KNBS (1). Five survey teams each comprising of a team leader, four data collectors were organized based on the number of clusters to be completed and households/children to be interviewed or measured per cluster. Each team did one cluster in a day (13 households).

4.3.

Data Quality Assurance Processes

To ensure data quality a number of steps were taken: (i) a standardization test was carried out on the second day of training but results were unsatisfactory because participants had not taken accurate and precise measurements. Consequently a second standardization was carried out leading to desirable and expected outputs from the participants; (ii) a field test was carried out in a village that was not in the sample, adjacent to Kitui town; (iii) a local events calendar developed by the survey data collectors was used in incidences where mothers or caretakers were unable to provide an immunization card with birth dates clearly indicated; and, (iv) at the end of each day, anthropometric data was entered into ENA, plausibility check performed and feedback relayed to the respective teams.

4.4.

Data Collection th

th

The field data collection was conducted from 13 to 29 of April, 2011, covering the 37 clusters/villages and 13 households from each cluster. The following categories of data were collected using three survey instruments. • Anthropometric data (Weight, Height, MUAC, Immunization and Disease prevalence) • Mortality questionnaires • Food security and livelihoods questionnaire, incorporating HINI indicators and IYCF practices.

a) Anthropometric Indicators: Children aged 6-59 months were measured using the standard survey form (see annexes) that captures the following key variables: •

Age in months-determined from child card or with the help of a local calendar of events

•

Sex- recorded as ‘m’ for male and ‘f’ for female

Action Against Hunger (USA), Integrated SMART Survey, April 2011 Kitui District, Kenya

12

•

Weight- Children were weighed to the nearest 100 g with a Salter Hanging Scale of 25 kg. All scales were calibrated daily by using a standard weight of 1 kg at the end of the survey exercise. In the field, it was calibrated with an empty weighing pant before each measurement.

•

Height- Children were measured on a measuring board (precision of 0.1cm).Children less than 85cm were measured lying down, while those greater than or equal to 85cm were measured standing up.

•

Mid-Upper Arm Circumference (MUAC) - measured in centimeters at mid-point of left upper arm to the nearest 0.1 cm with a MUAC tape.

•

Bilateral oedema - assessed by the application of moderate thumb pressure for at least three seconds to both feet (upper side) simultaneously. Only children with bilateral oedema were recorded as having nutritional oedema.

•

Measles vaccination- recorded for children aged 9-59 months from their vaccination cards. If no card was available at the time of the survey, the caretaker was asked if the child had been immunized against measles or not.

•

Vitamin A coverage- assessed by first describing what a Vitamin A capsule looked like, then asking the mother if the child received the content of that capsule in the past. The answer was then recorded depending on how many times the child had received it in the last one year.

•

Illness- assessed by asking each caretaker whether the child selected aged 6-59 months data was sick in the two weeks prior to the date of the survey. If the response was positive then the caretaker was further asked regarding the type of illnesses and the responses recorded.

b) Mortality The data required for estimating the death rate were collected using the SMART mortality survey form and 90 days th recall period. The recall period estimated from mid January (18 ) and the start of the survey. Each sample household regardless of having children 6-59 months of age was asked to enumerate current household members, indicate sex and age, members present at the time of the survey and at the beginning of the recall period, people joined or left during the recall period, and whether there was any birth or death in the recall period.

c) Food Security and WASH From the same households the mortality data were collected, the WASH and food security questionnaires were administered to the head of the household and/or the spouse regardless of whether the selected household had a child 6-59 months of age. The questionnaire used to gather data on health related variables from mothers with children under five, High Impact Nutrition Indicators data, water availability and accessibility, sanitation and hygiene practices, crop and livestock production, food sources, dietary diversity, income and expenditure and coping strategies.

4.5.

Data Entry and Analysis

The anthropometric and mortality data were entered and analyzed using the ENA Software, November 2008 version. The food security and WASH data entered and analyzed in SPSS. In assessing the nutritional status of children 6-59 months old, data on immediate and underlying causes of malnutrition such as disease, health seeking behavior, water and sanitation and food security and livelihood indicators were analyzed. Nutrition status is improved when individuals are healthy, have secure access to food and access to resources and livelihood options. This analytical approach provided the framework in identifying possible casual factors leading to the final outcome of malnutrition.

Action Against Hunger (USA), Integrated SMART Survey, April 2011 Kitui District, Kenya

13

a) Analysis of Acute Malnutrition Acute malnutrition rates are estimated from the weight for height (WFH) index values combined with the presence of oedema. The WFH indices are expressed in both Z-scores and percentage of the median, according to WHO 2005 and NCHS 1977 reference standards. Z-Score • Severe malnutrition is defined by WFH < -3 SD and/or existing bilateral oedema on the lower limbs. • Moderate malnutrition is defined by WFH < -2 SD and >-3 SD and no oedema. • Global acute malnutrition is defined by WFH < -2 SD and/or existing bilateral oedema. Percentage of Median • 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. • Global acute malnutrition is defined by WFH 0.05 >0.001 20 0 2 4 10 0 (5) Dig pref score - height Incl # 0-5 5-10 10-20 > 20 0 2 4 10 2 (6) Standard Dev WHZ Excl SD 0.05 you can consider the data normally distributed) Skewness WHZ -0.01 -0.01 -0.01 HAZ 0.53 0.53 0.17 WAZ 0.13 0.13 0.03 If the value is: -below minus 2 there is a relative excess of wasted/stunted/underweight subjects in the sample -between minus 2 and minus 1, there may be a relative excess of wasted/stunted/underweight subjects in the sample. -between minus 1 and plus 1, the distribution can be considered as symmetrical. -between 1 and 2, there may be an excess of obese/tall/overweight subjects in the sample. -above 2, there is an excess of obese/tall/overweight subjects in the sample Kurtosis WHZ 0.11 0.11 0.11 HAZ 1.87 1.87 0.07 WAZ 0.18 0.18 -0.09 (Kurtosis characterizes the relative peakedness or flatness compared with the normal distribution, positive kurtosis

Action Against Hunger (USA), Integrated SMART Survey, April 2011 Kitui District, Kenya

40

indicates a relatively peaked distribution, negative kurtosis indicates a relatively flat distribution) If the value is: -above 2 it indicates a problem. There might have been a problem with data collection or sampling. -between 1 and 2, the data may be affected with a problem. -less than an absolute value of 1 the distribution can be considered as normal. Test if cases are randomly distributed or aggregated over the clusters by calculation of the Index of Dispersion (ID) and comparison with the Poisson distribution for: WHZ < -2: ID=1.62 (p=0.007) WHZ < -3: ID=1.31 (p=0.083) Oedema: ID=1.00 (p=0.471) GAM: ID=1.70 (p=0.003) SAM: ID=1.22 (p=0.153) HAZ < -2: ID=0.99 (p=0.494) HAZ < -3: ID=0.97 (p=0.525) WAZ < -2: ID=1.55 (p=0.013) WAZ < -3: ID=1.41 (p=0.040) Subjects with SMART flags are excluded from this analysis. The Index of Dispersion (ID) indicates the degree to which the cases are aggregated into certain clusters (the degree to which there are "pockets"). If the ID is less than 1 and p < 0.05 it indicates that the cases are UNIFORMLY distributed among the clusters. If the p value is higher than 0.05 the cases appear to be randomly distributed among the clusters, if p is less than 0.05 the cases are aggregated into certain cluster (there appear to be pockets of cases). If this is the case for Oedema but not for WHZ then aggregation of GAM and SAM cases is due to inclusion of oedematous cases in GAM and SAM estimates. Are the data of the same quality at the beginning and the end of the clusters? Evaluation of the SD for WHZ depending upon the order the cases are measured within each cluster (if one cluster per day is measured then this will be related to the time of the day the measurement is made). Time SD for WHZ point 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 01: 0.84 (n=43, f=0) ## 02: 0.87 (n=43, f=0) ### 03: 0.77 (n=43, f=0) 04: 1.09 (n=42, f=0) ############ 05: 1.08 (n=43, f=0) ############ 06: 0.81 (n=43, f=0) 07: 1.00 (n=42, f=0) ######## 08: 0.84 (n=42, f=0) ## 09: 0.96 (n=42, f=0) ####### 10: 0.92 (n=40, f=0) ##### 11: 0.97 (n=40, f=0) ####### 12: 0.90 (n=37, f=0) #### 13: 0.92 (n=34, f=0) ##### 14: 0.84 (n=31, f=0) ## 15: 1.03 (n=28, f=0) ########## 16: 0.92 (n=25, f=0) ##### 17: 0.86 (n=22, f=0) OO 18: 0.77 (n=17, f=0)

Action Against Hunger (USA), Integrated SMART Survey, April 2011 Kitui District, Kenya

41

19: 0.89 (n=08, f=0) ~~~~ 20: 1.04 (n=06, f=0) ~~~~~~~~~~ 21: 0.69 (n=03, f=0) 22: 0.68 (n=02, f=0) (When n is much less than the average number of subjects per cluster different symbols are used: 0 for n < 80% and ~ for n < 40%; The numbers marked "f" are the numbers of SMART flags found in the different time points)

Analysis by Team Team 1 2 3 4 Percentage of values flagged with SMART flags: WHZ: 0.0 0.0 0.0 0.7 HAZ: 0.8 0.0 0.0 0.7 WAZ: 0.8 0.6 0.0 0.7 Age ratio of 6-29 months to 30-59 months: 0.59 0.80 0.58 0.68 Sex ratio (male/female): 1.02 1.07 1.26 0.96 Digit preference Weight (%): .0 : 12 7 11 10 .1 : 10 13 8 10 .2 : 13 10 5 9 .3 : 8 7 13 5 .4 : 12 15 17 12 .5 : 7 13 7 12 .6 : 12 10 11 8 .7 : 7 7 12 10 .8 : 9 6 8 12 .9 : 7 12 8 13 DPS: 8 9 10 7 10-20 poor and > 20 unacceptable) Digit preference Height (%): .0 : 11 12 11 17 .1 : 12 14 11 7 .2 : 14 12 9 15 .3 : 8 13 12 10 .4 : 10 12 14 6 .5 : 7 5 8 12 .6 : 12 8 13 12 .7 : 10 10 7 10 .8 : 6 9 5 4 .9 : 11 5 11 7 DPS: 7 10 8 13 10-20 poor and > 20 unacceptable) Standard deviation of WHZ: SD 0.90 0.94 0.79 0.92 Prevalence (< -2) observed: % Prevalence (< -2) calculated with current SD: %

5 0.0 0.6 0.0 0.79 1.14 10 13 9 5 8 8 13 6 14 12 10

Digit preference score (0-5 good, 5-10 acceptable,

10 11 8 11 11 14 12 9 6 7 7

Digit preference score (0-5 good, 5-10 acceptable,

0.99

Action Against Hunger (USA), Integrated SMART Survey, April 2011 Kitui District, Kenya

42

Prevalence (< -2) calculated with a SD of 1: % Standard deviation of HAZ: SD 1.02 0.92 1.05 observed: % 57.9 47.4 calculated with current SD: % 51.8 42.9 calculated with a SD of 1: % 51.9 42.6

1.17

1.06

38.6

37.0

38.8

36.2

37.0

35.4

Statistical evaluation of sex and age ratios (using Chi squared statistic) for: Team 1: Age cat. mo. boys girls total ratio boys/girls ------------------------------------------------------------------------------------6 to 17 12 11/14.2 (0.8) 9/13.9 (0.6) 20/28.1 (0.7) 1.22 18 to 29 12 13/13.8 (0.9) 12/13.6 (0.9) 25/27.4 (0.9) 1.08 30 to 41 12 20/13.4 (1.5) 15/13.2 (1.1) 35/26.5 (1.3) 1.33 42 to 53 12 12/13.2 (0.9) 17/12.9 (1.3) 29/26.1 (1.1) 0.71 54 to 59 6 5/6.5 (0.8) 7/6.4 (1.1) 12/12.9 (0.9) 0.71 ------------------------------------------------------------------------------------6 to 59 54 61/60.5 (1.0) 60/60.5 (1.0) 1.02 The data are expressed as observed number/expected number (ratio of obs/expect) Overall sex ratio: p = 0.928 (boys and girls equally represented) Overall age distribution: p = 0.230 (as expected) Overall age distribution for boys: p = 0.344 (as expected) Overall age distribution for girls: p = 0.477 (as expected) Overall sex/age distribution: p = 0.091 (as expected) Team 2: Age cat. mo. boys girls total ratio boys/girls ------------------------------------------------------------------------------------6 to 17 12 18/18.6 (1.0) 17/17.4 (1.0) 35/36.0 (1.0) 1.06 18 to 29 12 20/18.1 (1.1) 14/17.0 (0.8) 34/35.1 (1.0) 1.43 30 to 41 12 22/17.5 (1.3) 16/16.4 (1.0) 38/34.0 (1.1) 1.38 42 to 53 12 12/17.3 (0.7) 25/16.2 (1.5) 37/33.4 (1.1) 0.48 54 to 59 6 8/8.5 (0.9) 3/8.0 (0.4) 11/16.5 (0.7) 2.67 ------------------------------------------------------------------------------------6 to 59 54 80/77.5 (1.0) 75/77.5 (1.0) 1.07 The data are expressed as observed number/expected number (ratio of obs/expect) Overall sex ratio: p = 0.688 (boys and girls equally represented) Overall age distribution: p = 0.597 (as expected) Overall age distribution for boys: p = 0.560 (as expected) Overall age distribution for girls: p = 0.076 (as expected) Overall sex/age distribution: p = 0.022 (significant difference) Team 3: Age cat.

mo.

boys

girls

total

ratio boys/girls

Action Against Hunger (USA), Integrated SMART Survey, April 2011 Kitui District, Kenya

43

------------------------------------------------------------------------------------6 to 17 12 6/12.3 (0.5) 4/9.7 (0.4) 10/22.0 (0.5) 1.50 18 to 29 12 14/12.0 (1.2) 11/9.5 (1.2) 25/21.5 (1.2) 1.27 30 to 41 12 10/11.6 (0.9) 8/9.2 (0.9) 18/20.8 (0.9) 1.25 42 to 53 12 16/11.4 (1.4) 17/9.1 (1.9) 33/20.5 (1.6) 0.94 54 to 59 6 7/5.7 (1.2) 2/4.5 (0.4) 9/10.1 (0.9) 3.50 ------------------------------------------------------------------------------------6 to 59 54 53/47.5 (1.1) 42/47.5 (0.9) 1.26 The data are expressed as observed number/expected number (ratio of obs/expect) Overall sex ratio: p = 0.259 (boys and girls equally represented) Overall age distribution: p = 0.004 (significant difference) Overall age distribution for boys: p = 0.205 (as expected) Overall age distribution for girls: p = 0.017 (significant difference) Overall sex/age distribution: p = 0.001 (significant difference) Team 4: Age cat. mo. boys girls total ratio boys/girls ------------------------------------------------------------------------------------6 to 17 12 11/17.4 (0.6) 16/18.1 (0.9) 27/35.5 (0.8) 0.69 18 to 29 12 11/17.0 (0.6) 24/17.6 (1.4) 35/34.6 (1.0) 0.46 30 to 41 12 22/16.4 (1.3) 14/17.1 (0.8) 36/33.5 (1.1) 1.57 42 to 53 12 23/16.2 (1.4) 18/16.8 (1.1) 41/33.0 (1.2) 1.28 54 to 59 6 8/8.0 (1.0) 6/8.3 (0.7) 14/16.3 (0.9) 1.33 ------------------------------------------------------------------------------------6 to 59 54 75/76.5 (1.0) 78/76.5 (1.0) 0.96 The data are expressed as observed number/expected number (ratio of obs/expect) Overall sex ratio: p = 0.808 (boys and girls equally represented) Overall age distribution: p = 0.345 (as expected) Overall age distribution for boys: p = 0.056 (as expected) Overall age distribution for girls: p = 0.430 (as expected) Overall sex/age distribution: p = 0.011 (significant difference) Team 5: Age cat. mo. boys girls total ratio boys/girls ------------------------------------------------------------------------------------6 to 17 12 23/19.0 (1.2) 16/16.7 (1.0) 39/35.7 (1.1) 1.44 18 to 29 12 16/18.5 (0.9) 13/16.3 (0.8) 29/34.8 (0.8) 1.23 30 to 41 12 20/18.0 (1.1) 19/15.8 (1.2) 39/33.8 (1.2) 1.05 42 to 53 12 17/17.7 (1.0) 18/15.5 (1.2) 35/33.2 (1.1) 0.94 54 to 59 6 6/8.8 (0.7) 6/7.7 (0.8) 12/16.4 (0.7) 1.00 ------------------------------------------------------------------------------------6 to 59 54 82/77.0 (1.1) 72/77.0 (0.9) 1.14 The data are expressed as observed number/expected number (ratio of obs/expect) Overall sex ratio: p = 0.420 (boys and girls equally represented) Overall age distribution: p = 0.496 (as expected) Overall age distribution for boys: p = 0.681 (as expected) Overall age distribution for girls: p = 0.716 (as expected)

Action Against Hunger (USA), Integrated SMART Survey, April 2011 Kitui District, Kenya

44

Overall sex/age distribution: p = 0.280 (as expected)

Evaluation of the SD for WHZ depending upon the order the cases are measured within each cluster (if one cluster per day is measured then this will be related to the time of the day the measurement is made). Team: 1 Time SD for WHZ point 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 01: 0.88 (n=08, f=0) ### 02: 0.70 (n=08, f=0) 03: 0.71 (n=08, f=0) 04: 1.35 (n=08, f=0) ####################### 05: 1.18 (n=08, f=0) ################ 06: 0.94 (n=08, f=0) ###### 07: 0.55 (n=08, f=0) 08: 0.94 (n=08, f=0) ###### 09: 0.56 (n=08, f=0) 10: 1.05 (n=08, f=0) ########### 11: 0.76 (n=08, f=0) 12: 0.58 (n=06, f=0) 13: 0.73 (n=06, f=0) 14: 0.53 (n=05, f=0) 15: 1.19 (n=03, f=0) OOOOOOOOOOOOOOOO 16: 0.48 (n=03, f=0) 17: 0.63 (n=03, f=0) 18: 0.82 (n=03, f=0) O 19: 0.04 (n=02, f=0) (When n is much less than the average number of subjects per cluster different symbols are used: 0 for n < 80% and ~ for n < 40%; The numbers marked "f" are the numbers of SMART flags found in the different time points) Team: 2 Time SD for WHZ point 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 01: 1.24 (n=08, f=0) ################### 02: 0.68 (n=08, f=0) 03: 0.74 (n=08, f=0) 04: 0.98 (n=08, f=0) ####### 05: 0.66 (n=08, f=0) 06: 0.88 (n=08, f=0) ### 07: 1.27 (n=08, f=0) #################### 08: 0.85 (n=08, f=0) ## 09: 1.12 (n=08, f=0) ############# 10: 0.49 (n=08, f=0) 11: 1.38 (n=08, f=0) ######################## 12: 0.86 (n=08, f=0) ### 13: 0.78 (n=08, f=0) 14: 0.64 (n=08, f=0) 15: 1.04 (n=08, f=0) ########## 16: 0.93 (n=08, f=0) ##### 17: 1.00 (n=08, f=0) ######## 18: 0.94 (n=08, f=0) ######

Action Against Hunger (USA), Integrated SMART Survey, April 2011 Kitui District, Kenya

45

19: 0.67 (n=04, f=0) 20: 0.64 (n=04, f=0) (When n is much less than the average number of subjects per cluster different symbols are used: 0 for n < 80% and ~ for n < 40%; The numbers marked "f" are the numbers of SMART flags found in the different time points) Team: 3 Time SD for WHZ point 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 01: 0.51 (n=08, f=0) 02: 1.11 (n=08, f=0) ############# 03: 0.51 (n=08, f=0) 04: 0.79 (n=08, f=0) 05: 1.20 (n=08, f=0) ################# 06: 0.74 (n=08, f=0) 07: 0.65 (n=07, f=0) 08: 0.74 (n=07, f=0) 09: 1.01 (n=07, f=0) ######### 10: 0.68 (n=06, f=0) 11: 0.35 (n=06, f=0) 12: 0.53 (n=06, f=0) 13: 0.51 (n=03, f=0) 14: 0.49 (n=02, f=0) 15: 0.82 (n=02, f=0) ~ (When n is much less than the average number of subjects per cluster different symbols are used: 0 for n < 80% and ~ for n < 40%; The numbers marked "f" are the numbers of SMART flags found in the different time points) Team: 4 Time SD for WHZ point 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 01: 0.91 (n=10, f=0) ##### 02: 0.55 (n=10, f=0) 03: 1.03 (n=10, f=0) ########## 04: 0.91 (n=09, f=0) ##### 05: 1.39 (n=10, f=0) ######################### 06: 0.66 (n=10, f=0) 07: 0.79 (n=10, f=0) 08: 0.69 (n=10, f=0) 09: 0.74 (n=09, f=0) 10: 1.15 (n=09, f=0) ############### 11: 0.81 (n=09, f=0) 12: 0.92 (n=08, f=0) ##### 13: 1.06 (n=08, f=0) ########### 14: 0.94 (n=08, f=0) ###### 15: 1.21 (n=07, f=0) ################# 16: 0.99 (n=06, f=0) OOOOOOOO 17: 0.96 (n=05, f=0) OOOOOOO 18: 0.28 (n=03, f=0) (When n is much less than the average number of subjects per cluster different symbols are used: 0 for n < 80% and ~ for n < 40%; The numbers marked "f" are the numbers of SMART flags found in the different time points)

Action Against Hunger (USA), Integrated SMART Survey, April 2011 Kitui District, Kenya

46

Team: 5 Time SD for WHZ point 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 01: 0.67 (n=10, f=0) 02: 0.83 (n=09, f=0) # 03: 0.86 (n=09, f=0) ### 04: 1.27 (n=09, f=0) #################### 05: 0.81 (n=09, f=0) 06: 0.81 (n=09, f=0) 07: 1.34 (n=09, f=0) ####################### 08: 1.02 (n=09, f=0) ######### 09: 1.23 (n=09, f=0) ################## 10: 0.89 (n=09, f=0) #### 11: 1.23 (n=09, f=0) ################## 12: 1.09 (n=09, f=0) ############ 13: 0.97 (n=09, f=0) ####### 14: 1.17 (n=08, f=0) ################ 15: 0.69 (n=08, f=0) 16: 1.16 (n=07, f=0) ############### 17: 0.64 (n=06, f=0) 18: 0.74 (n=03, f=0) (When n is much less than the average number of subjects per cluster different symbols are used: 0 for n < 80% and ~ for n < 40%; The numbers marked "f" are the numbers of SMART flags found in the different time points)

Action Against Hunger (USA), Integrated SMART Survey, April 2011 Kitui District, Kenya

47