Integrated Nutrition Survey in Habiganj district of Bangladesh, October 2015.

Photo: Training of Survey enumerators

Conducted by: Supported by:

1 | P a g e

Executive Summary Habiganj is one of the under developed (110,000) and underweight (96,000). Food district, emerged as a zila in 1984, with security and livelihoods: Around 98% of around 2 million population and among the the surveyed households were working districts of highest poverty rate. Main every day. Almost half (44.9%) of the sources of income are Agriculture 42.26%, family’s main income sources were Agriculture labourer 20.55 non-agricultural dependent on unskilled/skilled daily labourer 4.64%, industry 1.7%, commerce labour, rickshaw pulling and fishing. 8.2%, service 4.69% and others 13.42%. Estimated 53.5% of the households Habiganj has also disaster vulnerabilities surveyed were purchasing staple foods for like frequent occurrences of flooding, pre- daily consumption with more than 90% monsoon flooding, river erosion, cyclones who were purchasing fish, legumes & etc. Every year flooding affects the rice pulse, meats & eggs and fruits. Overall, production, limits the working opportunities food consumption was very high with among farmers leaving households with 86.7% of the households with acceptable food insecurity across 8 Upazilas of the food consumption scores in past week. district. Floods also affect embankments, Although the diet was mostly dominant by roads and sanitation facilities every year. fish, vegetables and staples with very low Objective: To determine current consumption of meat & eggs, milk and nutritional status of children aged 6-59 dairy products, legumes & pulses. Dietary months, food security, water and consumption among reproductive aged sanitation situation of Habiganj District. women showed that around 55.5% women Methodology: Two-staged cluster had acceptable level (minimum 5 or above sampling recommended by SMART food groups) of diversity in last 24 hours. survey methodology was applied to Further analysis revealed that 98.9% of conduct the survey. At first stage, list of the women had vitamin A rich foods villages (smallest administrative unit) with (mostly from plant sources) and 92.6% population was entered into ENA to select had iron rich food in last 24 hours. Food estimated number of clusters applying insecurity situation was better reflected in PPS method. In second stage, all the finding of coping strategy index scores households from selected clusters were which explored that more than half of the updated to generate sampling frame. The households (57.8%) were adapting with list of households was then used to select medium or high level of coping strategy estimated number of households per scores. Among the coping strategies cluster using simple random sampling adapted by families, 49.3% households technique. Results: Nutrition: Global were depending on loan from relatives for acute malnutrition in Habiganj district was food, 49.4% eat less than necessary, 13.9% (10.3 – 18.5 95% C.I.) with a 42.5% households’ elder ate less to allow severe acute malnutrition rate of 0.7% younger to eat more and strikingly 29% (0.2-2.0 95%C.I.). The district hosts a total households reduced their meals to cope of 34,500 acutely malnourished children up with household food shortage. Water with around 1500 severely malnourished and Sanitation: Around 99.4% children who are in needs of immediate households had access to improved life-saving treatment for their recovery and drinking water sources (98.3% of them growth. Stunting and underweight rate accessing through tube wells: shared - were 43.7% (37.2-50.4, 95% C.I.) and 59.9% & household-37.4%). About 74,3% 39.0 % (32.6-45.8 95% C.I.) respectively of the surveyed households had access to in the district. The district has significantly improved sanitation facilities. Discussion exceeded WHO thresholds for nutritional and Conclusion: Nutrition situation in emergency with high burden of stunting Habiganj district was found to be at 2 | P a g e serious level (13.9%) with aggravating 3) Design and implement micronutrient factors (such as food insecurity at supplementation for children less than 5 household level, disaster prone areas, years and PLWs pocket areas with very high rate of morbidity). The population high rate of chronic undernutrition were adapting to poverty and food 4) Reinforce Growth Monitoring & insecurity by reducing consumption of food Promotion activities in health facilities both in terms of quality and quantity. There focusing on identifying growth failure and is high risk of deterioration in nutritional promotion of age appropriate IYCF status of this population which is predicted practices by the high proportion of households 5) Design and implement community coping with restricting adult’s food based interventions for promoting Infant (43.6%), reducing meals (30%), eating and young child feeding practices less than necessary foods (52%) and targeting 1000days families (considering borrowing or taking loan of food (49.6%). high rates of stunting & underweight rates) The district with high burden of chronic including essential nutrition actions undernutrition needs to put in place at 6) Design and implement integrated food scale promotion of preventive security-livelihoods and nutrition interventions focussing on essential interventions targeting vulnerable families nutrition actions. To complement to these (who are adapting with negative coping community based nutrition specific mechanism) for improved quality and interventions, livestock interventions to quantity of food consumption in the district. promote animal food production and increased consumption is important in the 7) Support vulnerable families to ensure context. Reviewing all nutrition actors in low cost sanitary latrine facilities and its the district and its interventions to identify utilization at household level gaps and appropriate strategies to 8) Sensitizing and mobilizing government address high burden of undernutrition and non-government stakeholders at could be helpful for improved coverage district level for targeting families at risk of and outcomes. undernutrition for district level Recommendations: 1) Implement programming (nutrition sensitive) through therapeutic feeding program for severely the district multi-sectoral coordination malnourished children in the Upazila platform. Health Complex and District Hospital. 9) Facilitate Capacity Gap Analysis of the 2) Identify Upazilas with high rates of existing health system for identifying undernutrition and implement targeted specific gaps related to nutrition service supplementary feeding program for delivery (in line with NNS) and define moderately malnourished children in the appropriate strategies for reinforcement of district the health system’s capacity.

3 | P a g e

Acknowledgement

Action Contre la Faim would like to acknowledge and express gratitude to the following organization for their support, collaboration and contribution: - Institute of Public Health and Nutrition for their support and cooperation in planning and coordination with District Health, Family Planning and Administrative authorities in Habiganj District. - Health and Family Planning staff at district, Upazila and community level for their active support in generating the sampling frame for selected clusters during the survey. - District and Upazila administration office, local representatives’ office - Food and Agriculture Organization, IPC unit, for their support and collaboration in conducting the training for survey supervisors, reviewing the report and inputs in additional indicators’ analysis - Concern Worldwide for their extended support in recruitment of local enumerators, support for effective coordination with district authorities. - UNICEF for funding the nutrition survey and reviewing reports for Habiganj District

ACF would like to acknowledge the community representatives and community people who have actively participated in the survey process for successful completion of the survey.

Finally, ACF is thankful to all of the survey enumerators, team leaders and supervisors for their tremendous efforts to successfully complete the survey in the district.

4 | P a g e

Acronyms

ACF Action Contre la Faim ARI Acute Respiratory Infection CI Confidence Interval CMAM Community Management of Acute Malnutrition

FCS Food Consumption Score FSL Food Security and Livelihoods GAM Global Acute Malnutrition HAZ Height-for-Age z-score HH Household MoH Ministry of Health MAM Moderate acute malnutrition MUAC Mid-Upper-Arm-Circumference NGO Non-Governmental Organization

SAM Severe Acute Malnutrition SD Standard Deviation SFP Supplementary Feeding Programme SMART Standardized Monitoring and Assessment of Relief and Transition WaSH Water, Sanitation and Hygiene WAZ Weight-for-Age z-score WHO World Health Organization WHZ Weight-for-Height z-score

5 | P a g e

Table of Contents Executive Summary ...... 2 Acknowledgement ...... 4 Acronyms ...... 5 Table of Contents ...... 6 Introduction ...... 8 1.1 Survey Objectives ...... 9 2. Methodology ...... 10 2.1 Survey Area ...... 10 2.2 Type of survey ...... 10 2.3 Sample size ...... 10 2.4 Sampling procedure: selecting clusters ...... 11 2.5 Sampling procedure: selecting households and children ...... 12 2.6 Case definitions and inclusion criteria ...... 12 2.8 Data analysis ...... 14 3. Results ...... 18 3.1 Household and family composition ...... 18 3.2 Anthropometric results (based on WHO standards 2006): ...... 18 3.2.1 Acute Malnutrition ...... 19 3.2.2 Underweight ...... 20 3.2.3 Stunting ...... 21 3.3 Childhood Morbidity ...... 22 3.4 Food Security and Livelihoods ...... 22 3.4.1 Household Source of Income ...... 22 3.4.2 Source of Food ...... 23 3.4.3 Food Stock/Preservation ...... 24 3.4.4 Food Consumption Score ...... 24 3.4.5 Coping Strategy Index ...... 25 3.4.6 Women dietary diversity score ...... 25 3.5 Water and Sanitation ...... 26 3.5.1 Main sources of water by percentage of HH ...... 26 3.5.2 Type of Sanitary latrines ...... 26 4. Discussion and Conclusion ...... 27 4.1 Nutritional status ...... 27 4.2 Food Security and Livelihoods ...... 28 4.3 Water and Sanitation ...... 28 4.4 Causes of malnutrition ...... 28 5. Recommendations and priorities ...... 29 7. Appendices ...... 30

List of tables

Table 2.1: Details of Administrative areas with population ...... 10 Table 2.2: Sampling parameters ...... 11 Table 2.3: Case definitions used for analysis ...... 13 Table 2.4: Categorization of food into proposed 9 food Groups for estimation of women dietary diversity score ...... 15 Table 2.5: Micronutrient rich food groups with category of foods ...... 16 Table 2.6: Acute under nutrition thresholds for children 6-59months ...... 17 Table 2.6: Classification of prevalence ranges of low height-for-age and low weight-for- age for children under 5 years of age ...... 17 Table 2.7: Categorization of households based on Food Consumption Scores for Bangladesh ...... 17 Table 3.1: Household information ...... 18

6 | P a g e

Table 3.2: Distribution of age and sex of sample ...... 18 Table 3.3: Prevalence of acute malnutrition based on weight-for-height z-scores (and/or oedema) and by sex ...... 19 Table 3.4: Prevalence of acute malnutrition by age, based on weight-for-height z-scores and/or oedema ...... 19 Table 3.5: Prevalence of acute malnutrition based on MUAC cut off's (and/or oedema) and by sex ...... 20 Table 3.6: Prevalence of acute malnutrition by age, based on MUAC cut off's and/or oedema ...... 20 Table 3.7: Prevalence of underweight based on weight-for-age z-scores by sex ...... 20 Table 3.8: Prevalence of underweight by age, based on weight-for-age z-scores ...... 21 Table 3.9: Prevalence of stunting based on height-for-age z-scores and by sex ...... 21 Table 3.10: Prevalence of stunting by age based on height-for-age z-scores ...... 22 Table 3.11: Mean z-scores, Design Effects and excluded subjects ...... 22 Table 3.12: Percentage of children aged 6-59 months by type of illness (last 14 days) ... 22 Table 3.13: Distribution of main occupation at household level ...... 23 Graph 1: Distribution of households by sources of food for different category of foods ... 23 Graph2: distribution of households by the duration of their preservation category ...... 24 Table 3.14: Food Consumption Score in Habiganj District ...... 24 Table 3.15: type of coping strategies by % of HH (at least once in last 7 days)...... 25 Table 3.16 Severity level of coping ...... 25 Graph 3: Vitamin A and Iron rich food consumption pattern among reproductive aged women (15-49 y) ...... 26 Table 3.17: Main source of water used by % of HH ...... 26 Table 3.18: category of sanitary latrines by % of households ...... 26 Table 3.19: Distribution of undernutrition rates across different Upazilas ...... 27 Graph-4: Distribution of GAM (WHZ<-2SD) cases in clusters ...... 49 Graph-5: Distribution of Stunted cases (HAZ <-2SD) in clusters ...... 49 Graph-6: Distribution of underweight cases (WAZ<-2SD) in clusters Appendix-7: Questionnaire ...... 50

List of Appendix

Appendix 1: Plausibility Report ______30 Appendix 2: Assignment of Clusters ______41 Appendix3: Evaluation of enumerators ______43 Appendix4: Maps of the areas ______46 Appendix 5: Result tables for NCHS reference ______47 Appendix6: Distribution of wasted, stunted and underweight cases in clusters ______49 Appendix7: Questionnaire ______50

7 | P a g e

Introduction

Bangladesh is the most climate vulnerable country in the World. Not only that, Bangladesh is also one of the densely populated and poorest countries of the World and has some of the highest rates of under nutrition, with millions of children under the age of five suffering from severe malnutrition. The country’s health system is further undermined due to inadequate sanitation and hygiene facilities or practices, recurring natural disasters such as rampant seasonal flooding during monsoons, cyclones, earthquakes etc. Natural disasters, political violence and instability, poor sanitation and hygiene are common features in Bangladesh and have a severe impact on the nutrition status. Malnutrition is identified as an underlying cause for nearly half of all child deaths (WHO Factsheet 2016). The Haor basin in north-eastern Bangladesh is one of the poorest regions of the country. It suffers from extensive annual flooding and devastating flash- floods, which limit livelihood opportunities for the poor, including agricultural production and enterprise growth. Haor dwellers are extremely vulnerable and their suffering is heightened by a lack of proper communication and transportation systems, hindering economic growth, access to markets (ie off-farm employment opportunities), and existing social services (i.e. health and education)..

Habiganj is one of the under developed district, emerged as a zila in 1984, Habiganj district is bounded by Sunamganj district on the north, Tripura State of India on the south, Maulvi bazar and Sylhet districts on the east, Brahmanbaria and Kishoreganj districts on the west. Annual average temperature is maximum 33.2C, minimum 13.6C. Total rainfall 3,334 mm. Noted rivers are Khowai, Sutang, Korangi, Kalni, Kushiyara, Gopala, Ratna, Barak.

Administration: Habiganj subdivision was established in 1874 under sylhet district and was turned into a district in 1984. It consists of 4 municipality, 36 wards, 124 mahallas, 8 upazilas, 77 union parishads, 1,241 mouzas and 2,143 villages. The upazilas are Ajmiriganj, Baniachang, Bahubal, Chunarughat, Habiganj sadar, Lakhai, Madhabpur and Nabiganj. The total population of Habiganj district is 20,89,001; male 49.94%, female 50.06%1. Muslim 80.23%, Hindu 19.12%, Buddhist 0.05%, Christian 0.13% and others 0.47%; ethnic nationals include Khasia and Manipuri. The district has higher literacy rate is 59.9% compared to national rate 51.8%.

Main occupations are Agriculture 42.26%, agricultural laborer 20.55%, wage laborer 6.45%, commerce 8.2%, service 4.69%, industry 1.7%, fishing 2.73% and others 13.42%. Main crops Paddy, tea, wheat, potato, jute, ground nut, betel leaf and oil seed.

The district host 23% landless population and is identified as one of the districts with highest poverty rate in the poverty map (BBS, WFP and WB).

Habiganj district is one of the most disaster prone districts in Bangladesh with frequent occurrences of flooding, pre-monsoon flooding, river erosion, cyclones etc. Every year flooding affects the rice production, limits the working opportunities among farmers leaving households with food insecurity2 across 8 Upazilas of the district. Floods also affect embankments, roads and sanitation facilities every year. The overall flood situation of the north eastern haor basin deteriorated on August 2014. Some people, forced to take refuge to safer places have started to return to their homes to begin the demanding clean-up operation removing huge deposits of mud and damaged fences. Work opportunities in this haor for the poorest, particularly day labours was disrupted leaving people with no income; while others have lost cattle and poultry due to post-flood

1 Bangladesh Population and Housing Census 2011 2 District Disaster Management Plan, Hobiganj. Accessed on December 2015 at - http://www.hobigonj.gov.bd/sites/default/files 8 | P a g e infections3. Children and women were reported suffering from skin diseases. Heavy rainfall triggered flash flood in the district including the district headquarters during 2015.

The district has a regular health system including Health centres Zila sadar hospital 1, Upazila health complex 7, Union sub-centers 18, Family Welfare Centers 51, satellite clinic 1; maternity 31 and veterinary hospital. On top of these government clinics there are 25 private clinics providing health care services. The district was identified with very high level (>40%) of underweight and stunted children in the undernutrition map jointly done by WFP, IFAD and BBS (2012).

1.1 Survey Objectives

Overall Survey Objective

To determine current nutritional status of children aged 6-59 months, food security and water & sanitation situation of Habiganj District.

Specific Survey Objectives

 To determine the current global acute malnutrition rate among children aged 6-59 months  To assess current rate of stunting and underweight among children aged 6-59 months  To understand the current level of household food insecurity by measuring food consumption score  To explore the existing coping mechanism through measuring household Coping Strategy Index (CSI)  To assess the current household women dietary diversity score among women of reproductive age (17-49 year)  To explore current rate of use for improved water and sanitation facilities in Habiganj District

3 JNA report 2014 and CARE Bangladesh 9 | P a g e

2. Methodology 2.1 Survey Area

The integrated SMART nutrition survey took place in Habiganj District during 31st October to 11th November 2015 which is considered as a lean period in Bangladesh. All 2,297 villages & wards (wards are the smallest unit in municipalities) from 8 Upazilas (including municipalities) of the district were included in the survey.

Table 2.1: Details of Administrative areas with population

Name of Upazila Number of Number of villages/wards Total unions Population Ajmirigonj 5 133 114265 Bahubal 7 342 197997 Baniachong 15 359 332530 Chunarughat 10 373 302110 Habiganj Sadar 10 248 329093 Lakhai 6 65 148811 Madhabpur 11 267 319016 Nabigonj 13 356 345179 Total 86 2143 20,89,001

2.2 Type of survey

This survey followed Standardized Monitoring and Assessment of Relief and Transition (SMART) survey methodology with a modified questionnaire defined in coordination with Bangladesh Nutrition Custer and Food Security clusters. Two stage cluster sampling recommended by SMART methodology was used for sampling and data collection of the survey. The key objective of the survey was to assess the nutrition situation of the district through anthropometry while additional FSL and WaSH indicators were incorporated for Integrated Food Security Phase Classification analysis. Household was considered as the basic sampling unit in the second stage and village was selected as primary sampling unit (Cluster) at the first stage.

Basic Sampling Unit

At the second and final stage, households were selected as the basic sampling unit. For Anthropometry, all eligible children aged 6-59 months were included from the selected HH and information for other additional indicators were collected from every household selected. Caregivers of household alone or with husband were considered as primary respondent for the survey. All eligible women aged 15-49 year old present in the selected household were included in the survey to measure women dietary diversity score. Mothers/caregivers of children were asked questions about food security; water, sanitation and hygiene conditions of the household. 2.3 Sample size

In the absence of recent data at district level for global acute malnutrition rate, GAM prevalence for the region (Sylhet) from BDHS-2014 was used to estimate the sample size. A precision of 3.5% and design effect defined as 1.5 was used for estimation of sample size. An estimated number of 545 children were estimated to be statistically

10 | P a g e representative for anthropometric survey in the District. And factoring with family size, non-response rate (4% considering possible absentees due to movement from Haor and Baor region), and U5 population parameters (given in below table), the minimum number of HHs estimated was 894 to reach the sampled number of children.

Table 2.2: Sampling parameters

Parameters for Value Assumptions based on context Anthropometry Estimated Prevalence of BDHS 2014 – Regional data (Sylhet) 12.1 % GAM (%) ± Desired precision Since the regional prevalence is between 10-15%, 3.5% of 3.5 % precision was considered as a rule of thumb of SMART sample size estimation. Design Effect No previous data exists to refer. As such 1.5 as rule of 1.5 thumb is used. Children to be included 545 Average HH Size 5.30 Census Bangladesh Bureau of Statistics 2011 % Children under-5 13.30% Census Bangladesh Bureau of Statistics 2011 % Non-response Considering possible absentees due to movement from 4% Households the Haor and Baor region. Households to be included 894

Selection of number of household per cluster / per day

Based on the following points, a calculation was done for each team to estimate the number of household to be surveyed per day. o Time spent on the field is (arrival time – departure time) – (travel time, round trip) = 660-120min=540minutes. o Work time spent on the field = time spent on the field – break times=540- 50=490 minutes o Time allocated for households visits = work time spent in the field – time for introduction and household selection = 490-20=470 minutes o No. of households to survey per day = total time allocated for households visits/ (average time spent in each Household + travel time between households) = 470/25=18.8≈19 HHs per day.

19 Households were feasible to visit and complete the questionnaire by each team in every day. Therefore, considering 19 HHs needed from each cluster, a total of 47 clusters were estimated to be selected for first stage sampling from all the villages and wards available in the Habiganj District.

2.4 Sampling procedure: selecting clusters

Government Household and Population Census data 2011 was used for estimating the survey sample size. Villages (wards in municipality areas) were the smallest unit with administrative boundary in the locality and thus villages were taken to be selected as cluster. A total of 47 clusters were selected from the targeted areas of Habiganj District using Emergency Nutrition Assessment (ENA version 9th July 2015) software. List of smallest unit (village/ward) with total population was entered into the ENA software (version July 2015) and 47 clusters were randomly selected based on Probability Proportional to Size technique. PPS method ensured that every household in the district had an equal chance to be selected irrespective the size of the village/ward.

11 | P a g e

Reserved clusters were planned to be included only when equal or more than 10% clusters cannot be surveyed and if only less than 80% of the sampled households could be reached from 47 clusters. Therefore, no reserved clusters were included in the survey since we could access all the 47 clusters for conducting the survey in Habiganj and also reached minimum number of HHs expected (equal or more than 80% of the 894 sampled HHs).

2.5 Sampling procedure: selecting households and children

Sampling frame, list of household, for all selected clusters were updated in coordination with local health & family planning teams and local community representatives before the training and data collection of the survey. And then at the second stage, simple random sampling technique was used to select households within each selected cluster. All the eligible children aged 6-59 months in the household were included for anthropometry; reproductive aged (15-49y) women were included for women dietary diversity score questionnaire.

Since updating the list of households with support from health and family planning staffs and volunteers were feasible and completed pre-hand to the survey, simple random sampling technique was applied for selection of household instead of systematic sampling. ENA software was used to generate random table with 19 household selected randomly for each cluster. Therefore, a total of 893 households were targeted for collecting information during the survey.

All empty/absent households were revisited during the survey day or in the reserve day and their status was reported through cluster control form. 2.6 Case definitions and inclusion criteria

Household definition: In this survey, household is defined as a group of people who live together and share a common cooking pot. ▬ Polygamous Families was counted as one household as long as they are living together and sharing a common cooking pot.

▬ Polygamous families or any other families living in the same house but not sharing a common cooking pot were counted as separate house hold. In such cases if the house is selected for the survey, both households were included in the survey with a different household number.

Inclusion criteria of children: All Children aged from 6-59 months were included for anthropometry. If age could not be defined by any means i.e. birth certificate, vaccination card or event calendar then height between 65 and 110cm was used as secondary inclusion criteria for anthropometry.

Height of children less than 2 years old was measured in laying down position while children more than 2 years were measured standing position. In the absence of age, height 87cm was used to define whether to measure standing (≥87cm) or lying (<87cm) position.

12 | P a g e

Table 2.3: Case definitions used for analysis

Indicator Definitive criteria WHZ <-2SD and/or Global Acute Malnutrition Acute Presence of Bilateral Odema Malnutrition4 Moderate Acute Malnutrition WHZ <-2 and ≥-3 Severe Acute Malnutrition WHZ <-3 and/or oedema Overall stunting HAZ <-2 Stunting Moderate Stunting HAZ <-2 and ≥-3 Severe stunting HAZ <-3 Overall Underweight WAZ <-2 Underweight Moderate Underweight WAZ <-2 and ≥-3 Severe Underweight WAZ <-3

WHO growth reference 2006 was used to estimate the prevalence of under nutrition. And rate of acute malnutrition by MUAC criteria was analysed and reported.

2.7 Questionnaire, training and supervision

Questionnaire

A modified version of questionnaire including additional indicators of FSL, WaSH sectors with anthropometry was developed by ACF Bangladesh team in close collaboration with nutrition cluster and IPC team of FAO. The purpose of adding additional indicators was to analyse additional data and feed into the integrated phase classification of food security situation in the district. Since the questionnaire is part of National Nutrition Survey guideline in country and has been in use for more than 3 years in the mission, survey team directly translated it into Bangla and used it in training, field testing and data collection5. Copy of the questionnaire is attached in the (Annex-7).

Survey teams and supervision

ACF team recruited 20 enumerators from Habiganj district and Upazilas for conducting the survey. Considering the importance of local translators for appropriate data collection, ACF ensured that all enumerators recruited are from the district/Upazila. Thus, 20 enumerators and 5 team leaders were divided into 5 teams of 5 persons. All enumerators had graduate level education or intermediate level with extensive experience in SMART survey conducted by ACF previously. All team leaders had master’s degree in Nutrition or social science. Survey Supervisor was experienced with conducted SMART survey with ACF previously with academic background in Applied Nutrition. Team leaders were assigned with each team for monitoring and supervision of the survey data collectors to ensure that appropriate sampling technique defined in the methodology was followed during selection of household. Team leaders were also responsible to supervise the enumeration process, anthropometric measurement process in the field to ensure that quality data is collected. Survey supervisors monitored the survey teams in the field and responsible for overall management of the survey in the district. Day to day and periodic guidance on data quality was provided from ACF technical department in Dhaka through plausibility checks. Management decisions for any unforeseen administrative or accessibility issues were taken in consultation with partners (UNICEF/Concern Worldwide/IPHN) for smooth conduction of the surveys.

4Acute malnutrition was also estimated using MUAC criteria as GAM represented MUAC <125mm; SAM represented children with MUAC <115mm and MAM represented children with MUAC ≥115mm and <125mm 5 This similar questionnaire was in use in the mission for integrated nutrition survey and that is why back translation was not required 13 | P a g e

Training

ACF trained the assessment team (5 team leaders and 1 survey supervisor) on SMART methodology nutrition survey for 5 days during August 2015. The team had previous experience as well as recent experience of conducting SMART survey in Cox’s Bazar district. Survey Supervisor with support from the team leaders conducted the 5 days field level training for enumerators. The SMART enumerator’s training tools and presentation was adapted in line with the survey objective and used during the enumerator’s training. The training for enumerators covered survey objectives, household selection techniques, demonstration on anthropometric measurements and standardization test (evaluation result is given in appendix-3), data collection and interview skills with group works & field testing. During the field testing, all the survey teams went to communities and conducted at least 4 HH surveys including anthropometry by each team.

2.8 Data analysis

During the data collection survey team leaders entered the anthropometric data into ENA for plausibility check. Survey supervisor with support from Dhaka technical team guided survey teams’ through-out the survey data collection for revisiting households and re- measurements of children if necessary.

Once the data collection was completed, all data was entered by a data entry officer based in Dhaka. Pre-defined excel database was used for data entry. During the data entry, random checks on sampled questionnaires were conducted to identify errors and minimizing data entry errors.

Anthropometric data was analysed using ENA software (version July 2015) and additional indicators were analysed using SPSS software. All flagged data using SMART flags6 (observed mean) was excluded from the analysis. For analysing household level indicators, all identical household’s information was extracted from the main database and analysed using SPSS software (version 20).

Reduced Coping Strategy Index7

Considering the district’s highest undernutrition rates and severity in the poverty map including disaster vulnerabilities, the coping strategy index was included to evaluate the food security situation. Coping Strategy Index (CSI) is often used as a proxy indicator of household food insecurity. CSI is based on a list of behaviors (coping strategies). CSI combines: (i) the frequency of each strategy (how many times each strategy was adopted?); and (ii) their (severity) (how serious is each strategy?) for households reporting food consumption problems. Higher CSI indicates a worse food security situation and vice versa. CSI is a particularly powerful tool for monitoring the same households or population over time. In this survey, reduced coping strategy index was used to evaluate the food security situation. Households were asked about how often they used a set of five short-term food based coping strategies in situations in which they did not have enough food, or money to buy food, during the one-week period prior to interview. The information is combined into the CSI which is a score assigned to a household that represents the frequency and severity of coping strategies employed. First, each of the five strategies is assigned a standard weight based on its severity. These weights are:

6 SMART flagging criteria (from observed mean) WHZ -3 to 3; HAZ -3 to 3; WAZ -3 to 3 7 WFP, CARE, USAID: The coping strategy index: Field Methods Manual, second edition, 2008 14 | P a g e

▬ Relying on less preferred and less expensive foods (=1.0); ▬ Limiting portion size at meal times (=1.0); ▬ Reducing the number of meals eaten in a day (=1.0); ▬ Borrow food or rely on help from relatives or friends (=2.0); ▬ Restricting consumption by adults for small children to eat (=3.0).

Household CSI scores are then determined by multiplying the number of days in the past week each strategy was employed by its corresponding severity weight, and then summing together the totals.

Food Consumption Scores8

Food Consumption is a standardized and transparent methodology to understand food consumption of families over a reference period of time recommended by World Food Programme. Food Consumption Score is a composite score based on the dietary diversity, food frequency and relative nutritional importance of the different food groups. FCS is similar to dietary diversity but is stronger because it captures the both quality and frequency and also weighs the nutritional values of food groups consumed. The cut-off thresholds for FCS recommended for Bangladesh is given in table 2.7 and used for the categorization of food consumption groups in Bangladesh due to the high consumption of oil and sugar.

Women Dietary Diversity Scores The WDDS reflects the probability of micronutrient adequacy of the diet and therefore the food groups included in the score are tailored towards this purpose. Vegetables and fruits were further segregated into specific categories based on their content of micronutrients’ availability. Consumption of foods was grouped into 9 categories proposed by FAO Guidelines for measuring household and individual dietary diversity to estimate the WDSS scores. A dietary diversity score below 5 was considered low in the analysis considering the diversified micronutrient needs from both plant as well as animal source of food for reproductive aged women. Additional variables for Vitamin A rich food consumption and iron rich food consumption were created to understand the dietary adequacy of these two micronutrients in their diet.

Table 2.4: Categorization of food into proposed 9 food Groups for estimation of women dietary diversity score

Food category Examples of foods Starchy Staples Rice, wheat, muri, potatoes, sweet potatoes, maize, kichuri

Legumes and Nuts Dal, cooked dry beans, peas, peanuts, other seeds/beans/ khichuri Dark green leafy All kinds of leafy vegetables vegetables Ripe mangoes, papaya, jackfruits other red/yellow or orange fruit Other vitamin A orange sweet potato, pumpkin, carrot or other yellow or orange rich fruits and Foodsvegetable containing oil, fat, butter vegetables Other fruits & Guava, Strawberry, Lemon, Orange, Lychees, Pineapple, vegetables Gourd,Mango,Grapes Broccoli, Cauli flower, tomatoes, Green Cabbage Cabbage, turnips, , bananas, apples Eggs Hen/duck, other birds, or fish eggs Organ meat Liver, kidney, gizzards

8 WFP 2008: Food Consumption Analysis-Calculation and use of the food consumption score in food security analysis 15 | P a g e

Small Fish Eaten Whole with Bones (i.e.Kachki, Mola, dhela, chapila, batashi, small prawn) Meat and Fish Beef, Pork, Veal, Lamb, Goat, Chicken, Duck, Large Whole Fish and Shell Fish Dairy Milk, cheese, yogurt or other milk products

Micronutrients of interest and corresponding food groups used to generate variables and estimation of Vitamin A and Iron rich food consumption

Table 2.5: Micronutrient rich food groups with category of foods

Micronutrient Food groups Vitamin A Plant-based food groups Vitamin A rich vegetables And Dark green leafy vegetables Vitamin A rich fruits (e.g. mangos, apricots) food group with red palm oil or products made from red palm oil Vitamin A Animal-based food groups Organ Meat Eggs Milk and dairy products Iron rich food groups Organ Meat (i.e. liver) Flesh Meat (chicken, beef etc.) Fish and sea food

Water and Sanitation

Improved drinking water sources: Access to safe drinking water was estimated by the percentages of the population using improved drinking water sources as defined below: - Household connection - Public standpipe - Borehole - Protected dug well - Protected spring - Rainwater collection

Improved Sanitation Facilities (UNICEF): Access to sanitary means of excreta disposal is estimated by the percentage of the population using improved sanitation facilities. Facilities that ensure hygienic separation of human excreta from human contact. They include: ▬ Improved sanitation facilities: ▬ Flush or pour-flush toilet/latrine to: o piped sewer system o septic tank o pit latrine ▬ Ventilated improved pit (VIP) latrine ▬ Pit latrine with slab ▬ Composting toilet

Classifying food security and Nutrition situation in regards to public health significance

16 | P a g e

WHO emergency thresholds of undernutrition rates for management of nutrition in major emergency was used for classification of undernutrition severity in the district.

Table 2.6: Acute under nutrition thresholds for children 6-59months

Severity of under nutrition Prevalence of GAM (% <- Mean weight for height Z- 2SD) score Acceptable < 5% > -0.40 Poor 5-9% -0.40 to 0.69 Serious 10-14% -0.70 to -0.99 Critical ≥ 15% ≤ -1.00

Table 2.7: Classification of prevalence ranges of low height-for-age and low weight-for-age for children under 5 years of age9

Prevalence ranges (% <-2SD) Level of severity Low height-for-age Low weight-for-age (stunting) (underweight) Low < 20% < 10% Medium 20-29% 10-19% High 30-39% 20-29% Very High ≥ 40% ≥ 30%

Household food insecurity status can be further classified as follows:

Table 2.8: Categorization of households based on Food Consumption Scores for Bangladesh

Food Consumption Groups FCS Thresholds Poor ≤28 Borderline 28.5-42 Acceptable >42

Based on the country’s context, the total CSI score is the basis to determine and classify the level of coping: into three categories: ▬ No or low coping (CSI= 0-3), ▬ Medium (CSI = 4-9), ▬ High coping (CSI ≥10).

9 WHO 2000: The management of nutrition in major emergencies 17 | P a g e

3. Results

3.1 Household and family composition

A total of 878 households were surveyed during the data collection. Household data revealed that only 4.9% households were led by women in Habiganj district against 95.1% of household by Men. Average family size in Habiganj district was 5.52. However, proportion of children (11.3%) aged less than 5 years has been decreased compared to census data from BBS 2011 (13.3%) in the district.

Table 3.1: Household information

Category/Indicator Number of Value Proportion/Mean observations % of Women Headed Household 859 43 4.9% % of Men Headed Household 859 816 95.1% Average age of HH Head 878 - 41.97 Mean Family Size 878 - 5.52 % of Male members 878 2451 50.8% % of Female members 878 2372 49.2% % of Infants in the Households 878 116 2.4% (<6 months) % of Children less than 5 years 878 431 8.9% % Children aged 6-17 years 878 1516 31.4% % Adult members (18-49y) 878 2192 45.4% % Elderly (50 years and above) 878 576 11.9%

3.2 Anthropometric results (based on WHO standards 2006):

Anthropometric data from a representative sample of 455 children were measured and analysed excluding z-scores from observed mean (SMART flags): WHZ-3 to3; HAZ -3 to 3; WAZ -3 to 3. Therefore, a total of 2 children for WHZ, 1 child for WAZ and 4 children HAZ indicator were excluded from the analysis with SMART flags.

Distribution of Age and Gender in the sample

In the sample, representation of boys and girls was equal (p value=0.815). Overall age distribution was as expected (p=0.923) with an equal representation from 6-29m and 30- 59m age groups (p value=0.394). Distribution of sex per age categories was as expected with equal representation in all categories (p value=0.703).

Table 3.2: Distribution of age and sex of sample Boys Girls Total Ratio AGE (mo) no. % no. % no. % Boy:girl 6-17 49 48.5 52 51.5 101 22.2 0.9 18-29 53 53.5 46 46.5 99 21.8 1.2 30-41 54 50.5 53 49.5 107 23.5 1.0 42-53 47 47.0 53 53.0 100 22.0 0.9 54-59 22 45.8 26 54.2 48 10.5 0.8 Total 225 49.5 230 50.5 455 100.0 1.0

18 | P a g e

3.2.1 Acute Malnutrition

Weight for height indicator reflects to acute crisis i.e. food shortage, inadequate food consumption due to various reasons in any context as weight changes rapidly compared to height. Global acute malnutrition rate in Habiganj district was identified as 13.9% which is considered as serious level based on WHO emergency thresholds. The district hosts a total of approx.1500 severely malnourished children with a rate of 0.7% In October 2015 who is in need of immediate treatment for their recovery and growth to the fullest potential.

Table 3.3: Prevalence of acute malnutrition based on weight-for-height z-scores (and/or oedema) and by sex All Boys Girls n = 453 n = 225 n = 228 Prevalence of global (63) 13.9 % (37) 16.4 % (26) 11.4 % malnutrition (10.3 - 18.5 (11.5 - 23.0 (7.3 - 17.4 (<-2 z-score and/or oedema) 95% C.I.) 95% C.I.) 95% C.I.) Prevalence of moderate (60) 13.2 % (35) 15.6 % (25) 11.0 % malnutrition (9.7 - 17.9 (10.7 - 22.1 (6.9 - 17.0 (<-2 z-score and >=-3 z-score, no 95% C.I.) 95% C.I.) 95% C.I.) oedema) Prevalence of severe (3) 0.7 % (2) 0.9 % (1) 0.4 % malnutrition (0.2 - 2.0 (0.2 - 3.5 (0.1 - 3.3 (<-3 z-score and/or oedema) 95% C.I.) 95% C.I.) 95% C.I.)

The prevalence of oedema is 0.0 %

Distribution of acute malnutrtion across different age category in below table shows that younger children had higher rate of severe acute undernutrition.

Table 3.4: Prevalence of acute malnutrition by age, based on weight-for-height z-scores and/or oedema

Severe wasting Moderate Normal Oedema (<-3 z-score) wasting (> = -2 z score) (>= -3 and <-2 z-score ) Age Total No. % No. % No. % No. % (mo) no. 6-17 101 2 2.0 15 14.9 84 83.2 0 0.0 18-29 99 1 1.0 19 19.2 79 79.8 0 0.0 30-41 106 0 0.0 12 11.3 94 88.7 0 0.0 42-53 99 0 0.0 10 10.1 89 89.9 0 0.0 54-59 48 0 0.0 4 8.3 44 91.7 0 0.0 Total 453 3 0.7 60 13.2 390 86.1 0 0.0

Prevalence of Acute Malnutrition by MUAC Global acute malnutrition rate according to MUAC cut-offs was significantly lower than Weight for height cut-offs. Prevalence of global acute malnutrition by MUAC was 4.4%. Surprisingly, no severely malnourished children were identified by MUAC cut-offs (<115mm). Comparatively high number of younger children was identified as acutely malnourished by MUAC criteria (table 3.6).

19 | P a g e

Table 3.5: Prevalence of acute malnutrition based on MUAC cut off's (and/or oedema) and by sex

All Boys Girls n = 455 n = 225 n = 230 Prevalence of global (20) 4.4 % (10) 4.4 % (10) 4.3 % malnutrition (2.6 - 7.5 (2.0 - 9.5 (2.1 - 8.7 (< 125 mm and/or oedema) 95% C.I.) 95% C.I.) 95% C.I.) Prevalence of moderate (20) 4.4 % (10) 4.4 % (10) 4.3 % malnutrition (2.6 - 7.5 (2.0 - 9.5 (2.1 - 8.7 (< 125 mm and >= 115 mm, no 95% C.I.) 95% C.I.) 95% C.I.) oedema) Prevalence of severe (0) 0.0 % (0) 0.0 % (0) 0.0 % malnutrition (0.0 - 0.0 (0.0 - 0.0 (0.0 - 0.0 (< 115 mm and/or oedema) 95% C.I.) 95% C.I.) 95% C.I.)

Table 3.6: Prevalence of acute malnutrition by age, based on MUAC cut off's and/or oedema

Severe wasting Moderate Normal Oedema (< 115 mm) wasting (> = 125 mm ) (>= 115 mm and < 125 mm) Age Total No. % No. % No. % No. % (mo) no. 6-17 101 0 0.0 11 10.9 90 89.1 0 0.0 18-29 99 0 0.0 4 4.0 95 96.0 0 0.0 30-41 107 0 0.0 2 1.9 105 98.1 0 0.0 42-53 100 0 0.0 2 2.0 98 98.0 0 0.0 54-59 48 0 0.0 1 2.1 47 97.9 0 0.0 Total 455 0 0.0 20 4.4 435 95.6 0 0.0

3.2.2 Underweight Underweight is a composite form of undernutrition that reflects to both acute changes as well as chronic deficits of nutrients & energy over time in nutritional status.

In Habiganj district, 39% (32.6-45.8%) of the children aged 6-59 months were identified as underweight which leads to an estimated total of around 95,500 underweight children. The underweight situation is considered as very high based on the WHO classification of the severity in emergencies.

Table 3.7: Prevalence of underweight based on weight-for-age z-scores by sex

All Boys Girls n = 454 n = 225 n = 229 Prevalence of underweight (177) 39.0 % (90) 40.0 % (87) 38.0 % (<-2 z-score) (32.6 - 45.8 (32.1 - 48.5 (30.4 - 46.2 95% C.I.) 95% C.I.) 95% C.I.) Prevalence of moderate (142) 31.3 % (71) 31.6 % (71) 31.0 % underweight (25.6 - 37.6 (25.2 - 38.7 (23.7 - 39.3 (<-2 z-score and >=-3 z-score) 95% C.I.) 95% C.I.) 95% C.I.) Prevalence of severe (35) 7.7 % (19) 8.4 % (16) 7.0 % underweight (5.2 - 11.4 (5.3 - 13.1 (4.2 - 11.4 (<-3 z-score) 95% C.I.) 95% C.I.) 95% C.I.)

20 | P a g e

Table 3.8: Prevalence of underweight by age, based on weight-for-age z-scores

Severe Moderate Normal Oedema underweight underweight (> = -2 z score) (<-3 z-score) (>= -3 and <-2 z-score ) Age Total No. % No. % No. % No. % (mo) no. 6-17 101 7 6.9 23 22.8 71 70.3 0 0.0 18-29 99 8 8.1 44 44.4 47 47.5 0 0.0 30-41 107 8 7.5 34 31.8 65 60.7 0 0.0 42-53 99 10 10.1 22 22.2 67 67.7 0 0.0 54-59 48 2 4.2 19 39.6 27 56.3 0 0.0 Total 454 35 7.7 142 31.3 277 61.0 0 0.0

3.2.3 Stunting

Height for age, stunted growth, reflects a process of failure to reach linear growth potential as a result of suboptimal health and/or nutritional conditions. A population with high levels of stunting is said to be associated with poor socioeconomic conditions and increased risk of frequent and early exposure to adverse conditions such as illness and/or inappropriate feeding practices (WHO).

Habiganj district, nearly half of the children is stunted with a rate of 43.7% (95% CI; 37.2- 50.4). The district is categorized as having very high level of stunting children with severe public health concern. The district hosts an estimated total of approx. 1,08,200 stunted children out of which 28,000 are severely stunted.

Table 3.9: Prevalence of stunting based on height-for-age z-scores and by sex

All Boys Girls n = 451 n = 225 n = 226 Prevalence of stunting (197) 43.7 % (93) 41.3 % (104) 46.0 % (<-2 z-score) (37.2 - 50.4 (34.5 - 48.5 (37.6 - 54.6 95% C.I.) 95% C.I.) 95% C.I.) Prevalence of moderate stunting (144) 31.9 % (70) 31.1 % (74) 32.7 % (<-2 z-score and >=-3 z-score) (26.4 - 38.1 (24.3 - 38.9 (25.9 - 40.4 95% C.I.) 95% C.I.) 95% C.I.) Prevalence of severe stunting (53) 11.8 % (23) 10.2 % (30) 13.3 % (<-3 z-score) (8.7 - 15.7 (6.5 - 15.8 (9.0 - 19.2 95% C.I.) 95% C.I.) 95% C.I.)

21 | P a g e

Table 3.10: Prevalence of stunting by age based on height-for-age z-scores

Severe Moderate Normal stunting stunting (> = -2 z score) (<-3 z-score) (>= -3 and <-2 z-score ) Age Total No. % No. % No. % (mo) no. 6-17 100 6 6.0 27 27.0 67 67.0 18-29 97 15 15.5 32 33.0 50 51.5 30-41 106 15 14.2 31 29.2 60 56.6 42-53 100 11 11.0 38 38.0 51 51.0 54-59 48 6 12.5 16 33.3 26 54.2 Total 451 53 11.8 144 31.9 254 56.3

Table 3.11: Mean z-scores, Design Effects and excluded subjects

Indicator n Mean z- Design z-scores z-scores scores ± Effect (z- not out of SD score < -2) available* range Weight-for- 453 -1.04±0.91 1.57 0 2 Height Weight-for-Age 454 -1.72±0.93 2.09 0 1 Height-for-Age 451 -1.81±1.03 2.00 0 4

* contains for WHZ and WAZ the children with edema.

3.3 Childhood Morbidity

Overall, 53.2% children aged 6-59m had reported illness in last 2 weeks preceding the survey. Diarrheal prevalence among children was identified as 8.1% with 83.8% (31 out of 37) of children who had diarrhoea were given ORS. Every 1 in out of 3 children reported to have fever in last 2 weeks while rate of acute respiratory infections among children was identified as 25.7%.

Table 3.12: Percentage of children aged 6-59 months by type of illness (last 14 days)

Type of Illness Proportion (95% CI) Diarrhoea 8.1% (5.5-10.8) Fever 39.1 % (34.7-43.8) Acute Respiratory Infections 25.7% (21.8-29.6) Other illnesses 0.4% (0.0-1.1)

No significant correlation was found between presence of illness and acute malnutrition (p=0.342). Acute malnutrition was not correlated with presence of acute respiratory infections (p=0.238) and diarrhoea (p=0.351) in the surveyed population. 3.4 Food Security and Livelihoods

3.4.1 Household Source of Income Overall, 98.6% of the households were found to have worked every day in past two months. However, household with no work represents 1.4% and they did not have work due to illness or disability or they did not get work. Major occupation identified in Habiganj

22 | P a g e district were Agricultural rice (20.4%), Agricultural day labour (15.5%), skilled day labour (10.5%), employment (6.5%) and rickshaw puller/driver-taxi, van (4.4%). Detail source of income are given in below table. The data shows that a total of 44.9% of the family’s main earning sources were dependent on daily labour (skilled or unskilled)/fishing/rickshaw pulling which does not ensure that every day they will have a consistent earning.

Table 3.13: Distribution of main occupation at household level

Occupation % HH 95% CI Agriculture 20.4 17.7-23.3 Agricultural day labour 15.5 13.1-17.8 Skilled day labour 10.5 8.5-12.5 Small Business or Petty business (<10000BD per 10.1 8.2-12.2 month) Unskilled day labour 9.6 7.6-11.5 Employment 6.5 4.9-8.2 Business 5.4 3.9-6.8 Rickshaw/Van/Baby Taxi/Driver 4.4 3.1-5.7 Fishermen 4.9 3.4-6.4 Agriculture other than rice 1.5 0.8-2.4 Do not earn 1.5 0.7-2.3 Others (doctors, engineer, housemaid, aquaculture, 8.5 n/a livestock, handicraft etc.)

3.4.2 Source of Food

In the graph below, bars present type of foods and stacks represents percentage of households who has consumed the food at least once in last week. Households who did not consume specific type of food in last 7 days were excluded to estimate their source of food for that particular food type.

More than half (53.7%) of the households purchased staple foods for consumption in last week, more than 90% households purchased Legumes & pulses, meats & eggs, fish and fruits for their consumption. Only 46.3% of households had staple foods from their own production with 25.3% household who had own production of milk and milk based foods for their consumption. Data shows that even vegetables, fruits, legumes & pulses and eggs were highly dependent on purchase which makes it difficult for poor families with uncertain income sources to have more diversified & nutritious foods.

Graph 1: Distribution of households by sources of food for different category of foods

23 | P a g e

3.4.3 Food Stock/Preservation In general, 51.8% of the surveyed households did not have stock of staple foods in their households during the survey. Therefore, only 48.2% of households (468 out of 970 HHs) who had storage of staple foods in their households were further categorized based on the duration of the food stock. The graph below shows that 70.9% households had food preserved for more than 60 days, 14.5% had food preserved for 30-60 days, 4.9% for 30 days, 3.6% for 15-30 days and 5.8% of households had food preserved for less than 15 days.

Graph2: distribution of households by the duration of their preservation category

3.4.4 Food Consumption Score Overall, the food consumption score in Habiganj district was very good with an estimated 85.5% of households who had acceptable food consumption scores in past week. Further analysis revealed that more than 90% households had staple foods, fats & oils in past 7 days with 48.8% households who consumed fish in all 7 days. Around 62.7% and 32% households had consumed vegetables and pulse & legumes for 3 or more days in past week. However, around 60%, 35.4% and 35.2% of households did not consume milk & dairy food, fruits and meats & eggs in past week which indicates that these households were heavily dependent on fish, vegetables and staple foods.

This should be noted that high consumption of fish in past week might have significantly influenced the weighting of meat & fish food group scoring. Among the surveyed households only 11.5% of HH consumed meat for 3 or more days while 91.8% had consumed fish for 3 more days (48.8% had fish for 7 days) in past week.

Table 3.14: Food Consumption Score in Habiganj District

Food Consumption Groups Prevalence (95% CI value) Poor (≤28) 0.6% (0.1-1.1) Borderline (28.5-42) 13.9% (11.6-16.1) Acceptable (>42) 85.5% (83.3-87.8)

24 | P a g e

3.4.5 Coping Strategy Index

Interestingly with very high proportion of households with acceptable food consumption, one in every three households is adapting with high coping strategy score with shortage of food. Moreover, 22.1% of surveyed households were identified with medium level coping strategy scores in Habiganj district. Among the coping strategies, eating less preferred foods was strikingly high (69.1%). In addition, half (49.6%) of the surveyed households were adjusting with taking foods from relatives on loan, eating less than necessary. This should be noted that 43.6 % households were identified with adults restricting foods to allow children more which is considered as severe stage of coping with food shortage at household level.

Table 3.15: type of coping strategies by % of HH (at least once in last 7 days)

Type of coping strategies % of HH Depended on foods which was less preferred or lower price 69.1 Depended on loan or relatives for food 49.6 Eaten less than necessary 51.9 Elder member eaten less to allow children more 43.6 Reduced number of meals due to lack of food 30.2

Table 3.16 Severity level of coping

Level of coping % of HH 95% CI No or low coping 41.1 37.6-44.3 Medium level of coping 22.1 19.1-25.1 High level of coping 36.8 33.7-40.1

3.4.6 Women dietary diversity score

In Habiganj district, mean women dietary diversity was 4.85. Overall, almost half (46.2, 95% CI 43.2-49.5) of the women surveyed were considered as consuming low diversified foods in their diet lacking adequate nutrition for this age group. The food consumption in last 24 hours was dominant by staple foods (99.5%), meat and fishes (92.4%), Vitamin A rich fruits and vegetables (98.2%), dark green leafy vegetables (66.4%). The consumption was very low for milk and dairy foods (14.1%), legumes, nuts and seeds (48%), egg consumption (19%) and organ meat consumption (3.4%). This needs to be pointed that meat and fish category was high due to higher proportion of small fish consumption (83%) while meat consumption was only 13.7%.

Data was further analysed to explore vitamin A and Iron rich food consumption among women. It was identified that overall 98.9% of women consumed vitamin A foods from either plant or animal sources. Only 27.9% women consumed vitamin A rich foods from animal food sources10. In addition, a total of 92.6% women of reproductive age consumed iron rich foods11 in last 24 hours.

10 Vitamin A rich fruits (yellow/red/orange colored fruits such as ripped mango, papaya, jack fruits etc) and vegetables (yellow/red/orange colored vegetables such as carrot, sweet potatoes, pumpkin etc) 11 Iron rich foods estimated by combining the consumption of organ meat, flesh meat and fish & seafoods 25 | P a g e

Graph 3: Vitamin A and Iron rich food consumption pattern among reproductive aged women (15-49 y) 3.5 Water and Sanitation

3.5.1 Main sources of water by percentage of HH Major sources of water were shared tube well or bore whole 59.9% and household tube well or bore whole 37.4%. Overall, 99.4% (95% CI; 98.9-99.8%) of surveyed households had access to improved drinking water sources in Habiganj. Type of unimproved drinking water sources (1.3%) were unprotected dug well and open ground water (river, pond, canal).

Table 3.17: Main source of water used by % of HH

Source of water % of HH Shared Tube well/Bore whole 59.9 Household tube well/ Bore Whole 37.4 Piped to dwelling 1.4 Others (Piped to yard, public tap, surface 1.3 water) Total 100

3.5.2 Type of Sanitary latrines Overall, 73.3% (95% CI; 70.5-76.2) of the population had access to improved sanitation facilities in Habiganj district. However, still one-fourth of the population were found to be using unimproved sanitary facilities (pit latrine without a slab or platform, hanging latrine). About 27% households were using shared latrine with 72.3% household using own latrine facilities in the district.

Table 3.18: category of sanitary latrines by % of households

Type of sanitary facility % of HH Latrine with septic tank 19 Water sealed rink slab toilet 12.9 Ring slab toilet without water seal 10.5 Closed pit latrine 43.1 Open pit latrine 12.1 Hanging latrine 1.7 Latrine linked with sewerage line 0.4 No latrine (defecate here and there) 0.3

26 | P a g e

4. Discussion and Conclusion

4.1 Nutritional status

In the survey, boys and girls were equally represented (p=0.815). Age distribution was also equally distributed across different categories (p=0.923). Overall, age and sex ratios were as expected with no differences (p=0.703).

Statistical test (Sapiro-wilk test for normality) result for normal distribution for WHZ, HAZ and WAZ suggest that data was not normality distributed in the surveyed population with p-value <0.5 (See plausibility report in appendix-1). Skewness for WHZ, HAZ and WAZ suggest that there are excess of obese, tall and overweight children in the sample. Further, kurtosis value for HAZ and WAZ suggest normal distribution without any differences in body size and tails. However, kurtosis value for WHZ indicated that there might have been more children with relatively large tails and small body shape in the surveyed population. Index of dispersion results shows that all forms of underweight (wasting, stunting and underweight) cases are not randomly distributed across clusters rather aggregated into certain clusters (Appendix-1 and 6). And detail segregated analysis of undernutrition rates by Upazilas suggest relatively higher rates in specific Upazilas.

Table 3.19: Distribution of undernutrition rates across different Upazilas

(N)

(n)

rate

unting

ht(n)

t

ht rate ht

Upazila

SAM(n)

GAM(n)

Stunting S

Name Name of

SAMrate

GAM rate GAM

Sample Underweig Underweig Amirganj 28 5 1 18 15 17.9% 3.6% 64.3% 53.6% Bahubal 55 5 0 13 12 9.1% 0.0% 23.6% 21.8% Baniaghang 84 8 0 44 36 9.4% 0.0% 51.8% 42.4% Chunarughat 52 9 0 27 26 17.3% 0.0% 51.9% 50.0% Habiganj 71 10 1 28 20 13.9% 1.4% 38.9% 27.8% Sadar Lakhai 29 8 0 16 14 27.6% 0.0% 55.2% 48.3% Madhobpur 89 12 1 40 41 13.5% 1.1% 44.9% 46.1% Nabiganj 45 6 0 11 13 13.3% 0.0% 24.4% 28.9% Total 453 63 3 197 177 13.9% 0.7% 43.7% 39.0%

Nutrition situation in the district is at SERIOUS level based on WHO threshold of emergency for acute malnutrition. Additional vulnerabilities i.e. food insecurity as high level of negative coping strategies; poor sanitation stress the current situation. Both underweight and stunting rates exceeded WHO emergency thresholds of emergency in the district. The district is currently hosting an estimated total of 34,000 wasted children out of which around 1500 are severely wasted; 95,500 underweight children and 108,208 stunted children. Although SAM rate can be considered as low, high rate of moderate acute malnutrition (13.2%) along with all above nutritional vulnerabilities increase the risk of deterioration and increased rate of SAM. Chronic undernutrition in the district is lower than the region, Sylhet while underweight and acute malnutrition is similar to the regional finding reported in BDHS report 201412. Although the sampling methodology and size, timing and geographical coverage of BDHS survey and this SMART survey is different, the comparison is made for an overall indication which is statistically weak.

12 BDHS 2014 finding for Sylhet: wasting 12.1%; Underweight- 39.8% and stunting 49.6% 27 | P a g e

District’s high burden of undernutrition urges appropriate and quality promotion of infant and young child feeding practices including regular growth monitoring of children below five years. This will also trigger the increased detection and referral of SAM children for appropriate nutritional treatment and management. The nutrition specific preventive services needs to be delivered through community based interventions to include most of the vulnerable children under the services. Additional complementary food security and livelihood interventions seems inevitable to put in place with the community based nutrition preventive activities since almost nearly half of the population were adapting with negative coping mechanisms due to food insecurity.

Reviewing the current nutrition actors, specific and sensitive interventions including the government health and family planning service delivery would be effective for re- emphasize the focus and efforts to tackle chronic undernutrition.

4.2 Food Security and Livelihoods

Overall, 51.8% households had uncertain income sources (daily work either skilled or unskilled, agricultural labour, fishing or rickshaw pulling). Half of the population were depending on staple food purchase while more than 90% of households had purchased other foods including legumes & pulses, meat and eggs, fish and fruits & vegetables for their consumption. The vulnerability and access to food was further heightened with the fact that around 51.8% of the households did not have food stock and were consuming foods on day to day purchase. This is a serious concern for those vulnerable households who have uncertain income sources to ensure enough nutritious foods for household members. The severity of the food insecurity was further explained by the coping strategy scores as one-third of the households had high coping strategy scores meaning that they were adapting with restricting food consumption, reducing meals and eating less than necessary. Additional 21.6% households had medium level of coping strategy scores to adapt with the food shortage. However, it is interesting to note that high level of undernutrition exists in the district despite in contrast to the very high proportion of food consumption identified at household level. Complementary food security livelihood interventions in the areas should be delivered in an integrated approach with nutrition activities to increase access to food, dietary diversity and the consumption of animal food sources. 4.3 Water and Sanitation Almost every household had access to improved drinking water sources. However, this definition of improved source of water has its limitation evaluate those sources with excess level of minerals or metal (Fe, Arsenic) which has its health impact in the longer run. Overall access to sanitation facilities has been identified as high in the district.

4.4 Causes of malnutrition

From the analysis of the relevant data within the scope of this survey, it is assumed that food insecurity, lower income with uncertain income sources for nearly half of the population in the district had severe negative impact on the dietary intake for children below five years. Also very low consumption of protein rich food which is considered as building blocks for growth could be linked with undernutrition. In addition, poor complementary feeding in the region with only 24.7% children (6-23 months) who had food from 4 food groups and 17.2% children who had minimum acceptable diet (BDHS 2014) might have been greatly influencing undernutrition. Inadequate growth monitoring and promotion of age specific essential nutrition actions for vulnerable children through

28 | P a g e the existing health system are partly contributing to the overall burden of undernutrition in the district.

5. Recommendations and priorities13

5.1 Implement therapeutic feeding program for severely malnourished children in the Upazila Health Complex and District Hospital. 5.2 Identify Upazilas with high rates of undernutrition and implement targeted supplementary feeding program for moderately malnourished children in the district 5.3 Design and implement micronutrient supplementation for children less than 5 years and PLWs pocket areas with very high rate of chronic undernutrition 5.4 Reinforce Growth Monitoring & Promotion activities in health facilities focusing on identifying growth failure and promotion of age appropriate IYCF practices 5.5 Design and implement community based interventions for promoting Infant and young child feeding practices targeting 1000days families (considering high rates of stunting & underweight rates) including essential nutrition actions 5.6 Design and implement integrated food security-livelihoods and nutrition interventions targeting vulnerable families (who are adapting with negative coping mechanism) for improved quality and quantity of food consumption in the district. 5.7 Support vulnerable families to ensure low cost sanitary latrine facilities and its utilization at household level 5.8 Sensitizing and mobilizing government and non-government stakeholders at district level for targeting families at risk of undernutrition for district level programming (nutrition sensitive) through the district multi-sectoral coordination platform. 5.9 Facilitate Capacity Gap Analysis of the existing health system for identifying specific gaps related to nutrition service delivery (in line with NNS) and define appropriate strategies for reinforcement of the health system’s capacity.

13 For generating recommendation guidance from these tools/policies were considered 1) Management of nutrition in emergencies and 2) MAM decision tool 2014 29 | P a g e

6. Appendices

Appendix 1: Plausibility Report

Plausibility check for: BD_HABIGANJ_OCTOBER_2015.as

Standard/Reference used for z-score calculation: WHO standards 2006 (If it is not mentioned, flagged data is included in the evaluation. Some parts of this plausibility report are more for advanced users and can be skipped for a standard evaluation)

Overall data quality

Criteria Flags* Unit Excel. Good Accept Problematic Score

Flagged data Incl % 0-2.5 >2.5-5.0 >5.0-7.5 >7.5 (% of out of range subjects) 0 5 10 20 0 (0,4 %)

Overall Sex ratio Incl p >0.1 >0.05 >0.001 <=0.001 (Significant chi square) 0 2 4 10 0 (p=0,815)

Age ratio(6-29 vs 30-59) Incl p >0.1 >0.05 >0.001 <=0.001 (Significant chi square) 0 2 4 10 0 (p=0,394)

Dig pref score - weight Incl # 0-7 8-12 13-20 > 20 0 2 4 10 0 (3)

Dig pref score - height Incl # 0-7 8-12 13-20 > 20 0 2 4 10 2 (8)

Dig pref score - MUAC Incl # 0-7 8-12 13-20 > 20 0 2 4 10 2 (8)

Standard Dev WHZ Excl SD <1.1 <1.15 <1.20 >=1.20 . and and and or . Excl SD >0.9 >0.85 >0.80 <=0.80 0 5 10 20 0 (0,91)

Skewness WHZ Excl # <±0.2 <±0.4 <±0.6 >=±0.6 0 1 3 5 1 (0,27)

Kurtosis WHZ Excl # <±0.2 <±0.4 <±0.6 >=±0.6 0 1 3 5 1 (0,22)

Poisson dist WHZ-2 Excl p >0.05 >0.01 >0.001 <=0.001 0 1 3 5 3 (p=0,001)

OVERALL SCORE WHZ = 0-9 10-14 15-24 >25 9 %

The overall score of this survey is 9 %, this is excellent.

There were no duplicate entries detected.

Percentage of children with no exact birthday: 0 %

Anthropometric Indices likely to be in error (-3 to 3 for WHZ, -3 to 3 for HAZ, -3 to 3 for WAZ, from observed mean - chosen in Options panel - these values will be

30 | P a g e flagged and should be excluded from analysis for a nutrition survey in emergencies. For other surveys this might not be the best procedure e.g. when the percentage of overweight children has to be calculated):

Line=31/ID=50: WHZ (5,384), WAZ (3,677), Weight may be incorrect Line=178/ID=1: HAZ (1,439), Age may be incorrect Line=237/ID=92: HAZ (2,012), Age may be incorrect Line=259/ID=127: HAZ (1,943), Height may be incorrect Line=306/ID=57: HAZ (2,194), Age may be incorrect Line=396/ID=118: WHZ (2,147), Height may be incorrect

Percentage of values flagged with SMART flags:WHZ: 0,4 %, HAZ: 0,9 %, WAZ: 0,2 %

Age distribution:

Month 6 : Month 7 : ###### Month 8 : ################ Month 9 : ########### Month 10 : ######### Month 11 : ############ Month 12 : ######### Month 13 : ############# Month 14 : ### Month 15 : ######## Month 16 : ####### Month 17 : ##### Month 18 : ####### Month 19 : ####### Month 20 : #### Month 21 : ######### Month 22 : ####### Month 23 : ########## Month 24 : ############ Month 25 : ###### Month 26 : ######### Month 27 : ########### Month 28 : ####### Month 29 : ######### Month 30 : ##### Month 31 : ########## Month 32 : ###### Month 33 : ####### Month 34 : ########### Month 35 : ########### Month 36 : ################## Month 37 : ######### Month 38 : ############

31 | P a g e

Month 39 : ######## Month 40 : ####### Month 41 : #### Month 42 : ### Month 43 : ############ Month 44 : ######### Month 45 : #### Month 46 : ############# Month 47 : ############# Month 48 : #### Month 49 : ########### Month 50 : ######### Month 51 : ###### Month 52 : ######## Month 53 : ###### Month 54 : ##### Month 55 : ######## Month 56 : ######## Month 57 : ########### Month 58 : ########### Month 59 : ####### Month 60 : ##

Age ratio of 6-29 months to 30-59 months: 0,78 (The value should be around 0.85).: p-value = 0,394 (as expected)

Statistical evaluation of sex and age ratios (using Chi squared statistic):

Age cat. mo. boys girls total ratio boys/girls ------6 to 17 12 49/52,2 (0,9) 52/53,4 (1,0) 101/105,6 (1,0) 0,94 18 to 29 12 53/50,9 (1,0) 46/52,0 (0,9) 99/102,9 (1,0) 1,15 30 to 41 12 54/49,3 (1,1) 53/50,4 (1,1) 107/99,8 (1,1) 1,02 42 to 53 12 47/48,5 (1,0) 53/49,6 (1,1) 100/98,2 (1,0) 0,89 54 to 59 6 22/24,0 (0,9) 26/24,5 (1,1) 48/48,6 (1,0) 0,85 ------6 to 59 54 225/227,5 (1,0) 230/227,5 (1,0) 0,98

The data are expressed as observed number/expected number (ratio of obs/expect)

Overall sex ratio: p-value = 0,815 (boys and girls equally represented) Overall age distribution: p-value = 0,923 (as expected) Overall age distribution for boys: p-value = 0,918 (as expected) Overall age distribution for girls: p-value = 0,881 (as expected) Overall sex/age distribution: p-value = 0,703 (as expected)

Digit preference Weight:

Digit .0 : ################################################ Digit .1 : ####################################### Digit .2 : ################################################### Digit .3 : ################################################## Digit .4 : ######################################## 32 | P a g e

Digit .5 : ############################################### Digit .6 : ########################################### Digit .7 : ####################################### Digit .8 : ################################################## Digit .9 : ################################################

Digit preference score: 3 (0-7 excellent, 8-12 good, 13-20 acceptable and > 20 problematic) p-value for chi2: 0,873

Digit preference Height:

Digit .0 : ############################################################ Digit .1 : ######################################################## Digit .2 : ########################################################## Digit .3 : #################################################### Digit .4 : #################################### Digit .5 : #################################################### Digit .6 : ##################################### Digit .7 : ###################################### Digit .8 : ####################### Digit .9 : ###########################################

Digit preference score: 8 (0-7 excellent, 8-12 good, 13-20 acceptable and > 20 problematic) p-value for chi2: 0,001 (significant difference)

Digit preference MUAC:

Digit .0 : ########################### Digit .1 : ############################################### Digit .2 : ################################### Digit .3 : ############################################### Digit .4 : ########################################## Digit .5 : ################################### Digit .6 : ########################################################## Digit .7 : ######################################################### Digit .8 : ############################################# Digit .9 : ##############################################################

Digit preference score: 8 (0-7 excellent, 8-12 good, 13-20 acceptable and > 20 problematic) p-value for chi2: 0,003 (significant difference)

Evaluation of Standard deviation, Normal distribution, Skewness and Kurtosis using the 3 exclusion (Flag) procedures

. no exclusion exclusion from exclusion from 33 | P a g e

. reference mean observed mean . (WHO flags) (SMART flags) WHZ Standard Deviation SD: 0,97 0,92 0,91 (The SD should be between 0.8 and 1.2) Prevalence (< -2) observed: calculated with current SD: calculated with a SD of 1:

HAZ Standard Deviation SD: 1,08 1,08 1,03 (The SD should be between 0.8 and 1.2) Prevalence (< -2) observed: 43,3% 43,3% 43,7% calculated with current SD: 41,9% 41,9% 42,8% calculated with a SD of 1: 41,3% 41,3% 42,5%

WAZ Standard Deviation SD: 0,96 0,96 0,93 (The SD should be between 0.8 and 1.2) Prevalence (< -2) observed: calculated with current SD: calculated with a SD of 1:

Results for Shapiro-Wilk test for normally (Gaussian) distributed data: WHZ p= 0,000 p= 0,011 p= 0,034 HAZ p= 0,000 p= 0,000 p= 0,011 WAZ p= 0,000 p= 0,000 p= 0,001 (If p < 0.05 then the data are not normally distributed. If p > 0.05 you can consider the data normally distributed)

Skewness WHZ 0,88 0,33 0,27 HAZ 0,48 0,48 0,25 WAZ 0,69 0,69 0,37 If the value is: -below minus 0.4 there is a relative excess of wasted/stunted/underweight subjects in the sample -between minus 0.4 and minus 0.2, there may be a relative excess of wasted/stunted/underweight subjects in the sample. -between minus 0.2 and plus 0.2, the distribution can be considered as symmetrical. -between 0.2 and 0.4, there may be an excess of obese/tall/overweight subjects in the sample. -above 0.4, there is an excess of obese/tall/overweight subjects in the sample

Kurtosis WHZ 3,96 0,37 0,22 HAZ 0,41 0,41 -0,28 WAZ 1,80 1,80 0,03 Kurtosis characterizes the relative size of the body versus the tails of the distribution. Positive kurtosis indicates relatively large tails and small body. Negative kurtosis indicates relatively large body and small tails. If the absolute value is: -above 0.4 it indicates a problem. There might have been a problem with data collection or sampling. -between 0.2 and 0.4, the data may be affected with a problem. -less than an absolute value of 0.2 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=1,79 (p=0,001) WHZ < -3: ID=0,96 (p=0,556) GAM: ID=1,79 (p=0,001) SAM: ID=0,96 (p=0,556) HAZ < -2: ID=1,80 (p=0,001) HAZ < -3: ID=1,30 (p=0,086) WAZ < -2: ID=2,08 (p=0,000) WAZ < -3: ID=1,49 (p=0,018) 34 | P a g e

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 > 0.95 it indicates that the cases are UNIFORMLY distributed among the clusters. If the p value is between 0.05 and 0.95 the cases appear to be randomly distributed among the clusters, if ID is higher than 1 and 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 likely 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: 0,82 (n=47, f=0) # 02: 1,12 (n=47, f=0) ############# 03: 1,23 (n=47, f=1) ################## 04: 0,94 (n=47, f=0) ###### 05: 1,01 (n=44, f=0) ######### 06: 0,89 (n=43, f=0) #### 07: 0,86 (n=38, f=0) ### 08: 0,88 (n=34, f=0) ### 09: 1,11 (n=33, f=1) ############# 10: 0,73 (n=22, f=0) 11: 0,64 (n=14, f=0) 12: 0,99 (n=13, f=0) OOOOOOOO 13: 0,84 (n=07, f=0) ~~ 14: 0,51 (n=07, f=0) 15: 1,67 (n=06, f=0) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 16: 0,51 (n=04, f=0) 17: 1,06 (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 5 n = 97 97 73 81 107 Percentage of values flagged with SMART flags: WHZ: 0,0 0,0 1,4 0,0 0,9 HAZ: 3,1 1,0 0,0 0,0 0,0 WAZ: 0,0 0,0 1,4 0,0 0,0 Age ratio of 6-29 months to 30-59 months: 0,87 0,67 0,87 0,65 0,88 Sex ratio (male/female): 1,31 0,80 1,03 0,62 1,23 Digit preference Weight (%): .0 : 11 6 14 10 12 .1 : 10 11 4 5 10 35 | P a g e

.2 : 8 13 7 19 9 .3 : 14 9 11 10 10 .4 : 9 8 10 5 11 .5 : 6 9 12 14 11 .6 : 10 8 12 7 9 .7 : 9 5 11 9 9 .8 : 11 13 10 12 8 .9 : 9 15 10 10 8 DPS: 7 11 9 13 4 Digit preference score (0-7 excellent, 8-12 good, 13-20 acceptable and > 20 problematic) Digit preference Height (%): .0 : 12 21 11 12 9 .1 : 9 12 10 16 14 .2 : 23 11 14 7 8 .3 : 11 10 5 12 16 .4 : 4 10 12 7 7 .5 : 11 12 8 7 16 .6 : 7 6 8 9 10 .7 : 9 7 8 11 7 .8 : 7 4 4 6 4 .9 : 5 5 19 11 9 DPS: 16 15 14 10 13 Digit preference score (0-7 excellent, 8-12 good, 13-20 acceptable and > 20 problematic) Digit preference MUAC (%): .0 : 2 9 15 2 3 .1 : 14 11 10 7 8 .2 : 11 5 11 9 4 .3 : 8 9 0 14 18 .4 : 12 9 1 9 12 .5 : 5 9 5 15 5 .6 : 15 10 11 11 15 .7 : 11 10 15 15 12 .8 : 7 11 11 6 13 .9 : 12 14 21 12 10 DPS: 13 7 20 13 16 Digit preference score (0-7 excellent, 8-12 good, 13-20 acceptable and > 20 problematic) Standard deviation of WHZ: SD 1,01 0,88 1,03 0,95 0,97 Prevalence (< -2) observed: % 13,4 8,2 Prevalence (< -2) calculated with current SD: % 12,9 15,9 Prevalence (< -2) calculated with a SD of 1: % 12,7 15,2 Standard deviation of HAZ: SD 1,13 1,08 1,09 1,02 1,06 observed: % 43,3 38,1 43,8 43,2 47,7 calculated with current SD: % 37,2 37,6 43,2 44,1 48,1

36 | P a g e calculated with a SD of 1: % 35,6 36,6 42,6 44,0 48,0

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 12/12,8 (0,9) 10/9,7 (1,0) 22/22,5 (1,0) 1,20 18 to 29 12 11/12,4 (0,9) 12/9,5 (1,3) 23/21,9 (1,0) 0,92 30 to 41 12 12/12,1 (1,0) 7/9,2 (0,8) 19/21,3 (0,9) 1,71 42 to 53 12 15/11,9 (1,3) 10/9,1 (1,1) 25/20,9 (1,2) 1,50 54 to 59 6 5/5,9 (0,9) 3/4,5 (0,7) 8/10,4 (0,8) 1,67 ------6 to 59 54 55/48,5 (1,1) 42/48,5 (0,9) 1,31

The data are expressed as observed number/expected number (ratio of obs/expect)

Overall sex ratio: p-value = 0,187 (boys and girls equally represented) Overall age distribution: p-value = 0,803 (as expected) Overall age distribution for boys: p-value = 0,883 (as expected) Overall age distribution for girls: p-value = 0,776 (as expected) Overall sex/age distribution: p-value = 0,330 (as expected)

Team 2:

Age cat. mo. boys girls total ratio boys/girls ------6 to 17 12 8/10,0 (0,8) 14/12,5 (1,1) 22/22,5 (1,0) 0,57 18 to 29 12 8/9,7 (0,8) 9/12,2 (0,7) 17/21,9 (0,8) 0,89 30 to 41 12 12/9,4 (1,3) 11/11,8 (0,9) 23/21,3 (1,1) 1,09 42 to 53 12 11/9,3 (1,2) 16/11,7 (1,4) 27/20,9 (1,3) 0,69 54 to 59 6 4/4,6 (0,9) 4/5,8 (0,7) 8/10,4 (0,8) 1,00 ------6 to 59 54 43/48,5 (0,9) 54/48,5 (1,1) 0,80

The data are expressed as observed number/expected number (ratio of obs/expect)

Overall sex ratio: p-value = 0,264 (boys and girls equally represented) Overall age distribution: p-value = 0,469 (as expected) Overall age distribution for boys: p-value = 0,773 (as expected) Overall age distribution for girls: p-value = 0,518 (as expected) Overall sex/age distribution: p-value = 0,168 (as expected)

Team 3:

Age cat. mo. boys girls total ratio boys/girls ------6 to 17 12 8/8,6 (0,9) 8/8,4 (1,0) 16/16,9 (0,9) 1,00 18 to 29 12 12/8,4 (1,4) 6/8,1 (0,7) 18/16,5 (1,1) 2,00 30 to 41 12 7/8,1 (0,9) 9/7,9 (1,1) 16/16,0 (1,0) 0,78 42 to 53 12 6/8,0 (0,8) 7/7,8 (0,9) 13/15,8 (0,8) 0,86 54 to 59 6 4/3,9 (1,0) 6/3,8 (1,6) 10/7,8 (1,3) 0,67 ------6 to 59 54 37/36,5 (1,0) 36/36,5 (1,0) 1,03

The data are expressed as observed number/expected number (ratio of obs/expect)

37 | P a g e

Overall sex ratio: p-value = 0,907 (boys and girls equally represented) Overall age distribution: p-value = 0,863 (as expected) Overall age distribution for boys: p-value = 0,688 (as expected) Overall age distribution for girls: p-value = 0,732 (as expected) Overall sex/age distribution: p-value = 0,367 (as expected)

Team 4:

Age cat. mo. boys girls total ratio boys/girls ------6 to 17 12 8/7,2 (1,1) 12/11,6 (1,0) 20/18,8 (1,1) 0,67 18 to 29 12 4/7,0 (0,6) 8/11,3 (0,7) 12/18,3 (0,7) 0,50 30 to 41 12 8/6,8 (1,2) 14/11,0 (1,3) 22/17,8 (1,2) 0,57 42 to 53 12 7/6,7 (1,0) 7/10,8 (0,6) 14/17,5 (0,8) 1,00 54 to 59 6 4/3,3 (1,2) 9/5,3 (1,7) 13/8,6 (1,5) 0,44 ------6 to 59 54 31/40,5 (0,8) 50/40,5 (1,2) 0,62

The data are expressed as observed number/expected number (ratio of obs/expect)

Overall sex ratio: p-value = 0,035 (significant excess of girls) Overall age distribution: p-value = 0,188 (as expected) Overall age distribution for boys: p-value = 0,780 (as expected) Overall age distribution for girls: p-value = 0,225 (as expected) Overall sex/age distribution: p-value = 0,012 (significant difference)

Team 5:

Age cat. mo. boys girls total ratio boys/girls ------6 to 17 12 13/13,7 (0,9) 8/11,1 (0,7) 21/24,8 (0,8) 1,63 18 to 29 12 18/13,3 (1,3) 11/10,9 (1,0) 29/24,2 (1,2) 1,64 30 to 41 12 15/12,9 (1,2) 12/10,5 (1,1) 27/23,5 (1,2) 1,25 42 to 53 12 8/12,7 (0,6) 13/10,4 (1,3) 21/23,1 (0,9) 0,62 54 to 59 6 5/6,3 (0,8) 4/5,1 (0,8) 9/11,4 (0,8) 1,25 ------6 to 59 54 59/53,5 (1,1) 48/53,5 (0,9) 1,23

The data are expressed as observed number/expected number (ratio of obs/expect)

Overall sex ratio: p-value = 0,288 (boys and girls equally represented) Overall age distribution: p-value = 0,596 (as expected) Overall age distribution for boys: p-value = 0,404 (as expected) Overall age distribution for girls: p-value = 0,733 (as expected) Overall sex/age distribution: p-value = 0,118 (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: 1,19 (n=10, f=0) ################ 02: 1,65 (n=10, f=0) #################################### 03: 0,68 (n=10, f=0) 04: 0,94 (n=10, f=0) ###### 38 | P a g e

05: 0,93 (n=10, f=0) ###### 06: 1,03 (n=10, f=0) ########## 07: 0,83 (n=10, f=0) # 08: 0,90 (n=09, f=0) #### 09: 0,90 (n=08, f=0) #### 10: 1,12 (n=04, f=0) OOOOOOOOOOOOO 11: 0,15 (n=03, f=0) 12: 1,32 (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)

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: 0,68 (n=09, f=0) 02: 0,92 (n=09, f=0) ##### 03: 1,34 (n=09, f=0) ####################### 04: 0,78 (n=09, f=0) 05: 1,04 (n=09, f=0) ########## 06: 0,66 (n=09, f=0) 07: 0,79 (n=08, f=0) 08: 0,83 (n=07, f=0) # 09: 1,11 (n=07, f=0) ############# 10: 0,55 (n=06, f=0) 11: 0,45 (n=04, f=0) 12: 0,28 (n=04, f=0) 13: 0,09 (n=02, f=0) 14: 1,16 (n=02, f=0) ~~~~~~~~~~~~~~~ 15: 1,82 (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: 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: 0,72 (n=10, f=0) 02: 0,78 (n=10, f=0) 03: 2,16 (n=10, f=1) ######################################################### 04: 0,80 (n=10, f=0) 05: 0,65 (n=08, f=0) 06: 0,82 (n=08, f=0) # 07: 0,53 (n=06, f=0) 08: 0,64 (n=04, f=0) 09: 0,66 (n=04, f=0) 10: 0,90 (n=03, f=0) OOOO

(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: 0,71 (n=09, f=0) 02: 1,06 (n=09, f=0) ########### 03: 0,23 (n=09, f=0) 04: 1,29 (n=09, f=0) #################### 05: 1,15 (n=08, f=0) ############### 06: 0,85 (n=07, f=0) ## 07: 1,30 (n=05, f=0) ##################### 08: 1,14 (n=05, f=0) ############## 09: 1,16 (n=05, f=0) ############### 10: 0,95 (n=03, f=0) OOOOOO 11: 1,20 (n=02, f=0) OOOOOOOOOOOOOOOOO 12: 0,20 (n=02, f=0) 13: 1,12 (n=02, f=0) OOOOOOOOOOOOOO 14: 0,03 (n=02, f=0) 39 | P a g e

15: 0,01 (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: 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: 0,63 (n=09, f=0) 02: 1,12 (n=09, f=0) ############# 03: 0,83 (n=09, f=0) # 04: 0,91 (n=09, f=0) ##### 05: 1,23 (n=09, f=0) ################## 06: 1,08 (n=09, f=0) ############ 07: 0,88 (n=09, f=0) ### 08: 0,91 (n=09, f=0) ##### 09: 1,47 (n=09, f=1) ############################ 10: 0,39 (n=06, f=0) 11: 0,80 (n=05, f=0) 12: 1,59 (n=04, f=0) OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 13: 0,62 (n=03, f=0) 14: 0,24 (n=03, f=0) 15: 0,89 (n=02, f=0) ~~~~ 16: 0,81 (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)

(for better comparison it can be helpful to copy/paste part of this report into Excel)

40 | P a g e

Appendix 2: Assignment of Clusters Cluster list Habiganj Cluster Geographical unit HHs Population size Union Upazila 1 Mamudpur 115 618 Badalpur Ajmirganj 2 Kamalpur 1007 5490 Kakailseo Ajmirganj 3 Bahubal 480 2463 Bahubal Bahubal 4 Jasmangal 44 225 Bhadeshwar Bahubal 5 Tarapasha 171 791 Lamatashi Bahubal Paschim Bijuli (Dubai 6 Bazar) 129 618 Putijuri Bahubal 7 Bakatapur 261 1582 Satkapan Bahubal Uttar Purba 8 Majlish Mahalla 424 2593 Baniachang Baniachang Dakshin Purba 9 Chowdhury Para 181 1016 Baiachang Baniachang Dakshin Paschim 10 Buruj Para 107 593 Baiachang Baniachang 11 Noagaon 266 1456 Baraiuri Baniachang 12 Chandpur 168 1020 Kagapasha Baniachang 13 Lama Ujirpur 27 175 Khagaura Baniachang Murda Kaharam 14 (Uttar Sangar) 1235 6595 Mandari Baniachang 15 Daraya 119 735 Pukhra Baniachang 16 Athukura 288 1885 Subidpur Baniachang 17 Amo T.E. 1289 6034 Ahmadabad Chunarughat 18 Ramsree 239 1145 Chunaraughat Chunarughat 19 Dudh Patil 342 1708 Gazipur Chunarughat 20 Satchhari T.G. 320 1458 Paik Para Chunarughat 21 Deutali 191 996 Samkholi Chunarughat 22 Sreebari T.E. 441 1975 Shatiajuri Chunarughat Hobiganj 23 Jasher Abda 581 2880 Paurashuba,W-02 Habiganj Sadar Hobiganj 24 Mohanpur 517 2510 Paurashuba,W-08 Habiganj Sadar 25 Bangaon 180 980 Laskarpur Habiganj Sadar 26 Sukri Para 175 889 Nizampur Habiganj Sadar 27 Putia 169 835 Nurpur Habiganj Sadar 28 Raziura 358 1719 Raziura Habiganj Sadar 29 Kadamtali 229 1364 Saistaganj Habiganj Sadar 30 Bamai 2326 12123 Bamai Lakhai 31 Swajangaon 1322 7299 Lakhai Lakhai 32 Muriauk 953 5150 Muriak Lakhai 33 Alalpur 116 620 Adair Madhabpur 34 Puraikala 377 2089 Bagasura Madhabpur 35 Bulla 772 4115 Bulla Madhabpur 41 | P a g e

36 Chhatiain 1031 5497 Chhatain Madhabpur 37 Manoharpur 74 427 Chowmohani Madhabpur 38 Bejura North 734 4192 Jagadishpur Madhabpur 39 Mirnagar 386 1974 Amdiurauk Madhabpur 40 Ratanpur 132 688 Shahjahan Pur Madhabpur Paurashuba ,W- 41 Kali Kanaipur 288 1533 04 Nabiganj 42 Maijgaon 170 1054 Bausa Nabiganj 43 Uttar Debpara 212 1214 Bebpara Nabiganj 44 Uttar Kamargaon 173 975 Dighalbak Nabiganj 45 Raiapur 370 2195 Kurshi Nabiganj 46 Paniumda 1255 6575 Paniumda Nabiganj Paschim Bara 47 Sonapur 538 2828 Bhakhair Nabiganj RC Kakaus 187 866 Mirahi Union Chunarughat RC Anandapur 313 1734 Gopaya Union Habiganj Sadar RC Harinakona 141 758 Karab Union Lakhai RC Holimpur 232 1513 Inathganj Union Nabiganj RC Gumgumia 512 2930 Kargaon Union Nabiganj

42 | P a g e

Appendix3: Evaluation of enumerators Standardization test results Precision Accuracy OUTCOME

subject Technical TEM/me Coef of Bias from Bias from Weight mean SD max result s error an reliability superv median # kg kg kg TEM (kg) TEM (%) R (%) Bias (kg) Bias (kg) Supervisor 10 10.8 2.8 0.1 0.07 0.6 99.9 - 0.62 TEM acceptable Enumerator 1 10 10.8 2.7 0.2 0.06 0.5 100 0.03 0.64 TEM acceptable Enumerator 2 10 10.8 2.8 0.3 0.09 0.8 99.9 -0.02 0.6 TEM acceptable Enumerator 3 10 15.5 20.8 92.7 20.73 133.7 0.5 4.68 5.3 TEM reject Enumerator 4 10 10.9 2.8 0.2 0.07 0.7 99.9 0.04 0.66 TEM acceptable Enumerator 5 10 10.9 2.8 0.2 0.07 0.7 99.9 0.07 0.69 TEM acceptable Enumerator 6 10 10.9 2.8 0.2 0.06 0.6 99.9 0.08 0.7 TEM acceptable Enumerator 7 10 10.8 2.8 1.9 0.44 4 97.5 -0.03 0.58 TEM reject Enumerator 8 10 10.9 2.7 0.1 0.05 0.5 100 0.05 0.66 TEM acceptable Enumerator 9 10 10.9 2.8 0.6 0.14 1.3 99.8 0.07 0.69 TEM poor Enumerator 10 10 10.8 2.8 0 0 0 100 0 0.62 TEM good Enumerator 11 10 10.9 2.8 0.1 0.04 0.4 100 0.06 0.68 TEM acceptable Enumerator 12 10 10.9 2.8 0.1 0.04 0.4 100 0.05 0.66 TEM good Enumerator 13 10 10.9 2.8 0.1 0.06 0.6 99.9 0.11 0.73 TEM acceptable Enumerator 14 10 10.9 2.8 0.1 0.06 0.6 99.9 0.12 0.73 TEM acceptable Enumerator 15 10 10.9 2.7 0.1 0.06 0.6 99.9 0.04 0.66 TEM acceptable Enumerator 16 10 10.9 2.8 0.1 0.06 0.5 100 0.08 0.69 TEM acceptable Enumerator 17 10 10.9 2.8 0.1 0.06 0.6 99.9 0.1 0.71 TEM acceptable Enumerator 18 10 10.9 2.8 0.1 0.07 0.6 99.9 0.07 0.69 TEM acceptable Enumerator 19 10 10.9 2.8 0.1 0.05 0.5 100 0.09 0.71 TEM acceptable Enumerator 20 10 10.9 2.8 0.2 0.07 0.6 99.9 0.1 0.71 TEM acceptable enum inter 1st 20x10 11.3 7.1 - 6.56 57.9 13.7 - - TEM reject enum inter 2nd 20x10 10.9 2.7 - 0.09 0.8 99.9 - - TEM good inter enum + sup 21x10 11.1 5.2 - 3.24 28.7 58.8 - - TEM reject TOTAL intra+inter 20x10 - - - 6.56 59 -50.5 0.29 0.89 TEM reject TOTAL+ sup 21x10 - - - 6.4 57.7 -48.6 - - TEM reject 43 | P a g e

STANDARDISATION TEST RESULT FOR HEIGHT subject Technical TEM/me Coef of Bias from Bias from Height mean SD max result s error an reliability superv median # cm cm cm TEM(cm) TEM (%) R (%) Bias (cm) Bias (cm)

Supervisor 10 85.3 11.9 0.1 0.06 0.1 100 - 0.86 TEM good Enumerator 1 10 84.8 11.9 0.6 0.16 0.2 100 -0.46 0.41 TEM good Enumerator 2 10 84.8 11.9 0.5 0.15 0.2 100 -0.46 0.4 TEM good Enumerator 3 10 84.9 12 0.3 0.1 0.1 100 -0.38 0.49 TEM good Enumerator 4 10 84.9 12 0.3 0.1 0.1 100 -0.4 0.47 TEM good Enumerator 5 10 84.3 12 1 0.27 0.3 99.9 -0.96 -0.1 TEM good Enumerator 6 10 85 12.1 0.3 0.11 0.1 100 -0.24 0.62 TEM good Enumerator 7 10 84.8 11.9 0.8 0.19 0.2 100 -0.43 0.43 TEM good Enumerator 8 10 84.5 12.2 1.8 0.41 0.5 99.9 -0.81 0.06 TEM acceptable Enumerator 9 10 85 11.9 0.2 0.07 0.1 100 -0.3 0.56 TEM good Enumerator 10 10 85.2 11.7 0.1 0.04 0.1 100 -0.04 0.82 TEM good Enumerator 11 10 85.2 11.7 0.3 0.1 0.1 100 -0.1 0.76 TEM good Enumerator 12 10 85.1 11.7 0.2 0.07 0.1 100 -0.15 0.71 TEM good Enumerator 13 10 84.8 11.9 1 0.23 0.3 100 -0.47 0.39 TEM good Enumerator 14 10 84.8 11.9 0.1 0.07 0.1 100 -0.43 0.43 TEM good Enumerator 15 10 84.8 11.9 1 0.23 0.3 100 -0.47 0.39 TEM good Enumerator 16 10 85.1 11.7 0.2 0.07 0.1 100 -0.14 0.73 TEM good Enumerator 17 10 84.8 12 0.2 0.08 0.1 100 -0.43 0.43 TEM good Enumerator 18 10 84.8 11.9 0.1 0.07 0.1 100 -0.43 0.43 TEM good Enumerator 19 10 84.8 11.9 0.2 0.07 0.1 100 -0.43 0.43 TEM good Enumerator 20 10 84.9 11.9 0.2 0.08 0.1 100 -0.41 0.45 TEM good enum inter 1st 20x10 84.9 11.7 - 0.73 0.9 99.6 - - TEM acceptable enum inter 2nd 20x10 84.9 11.6 - 0.74 0.9 99.6 - - TEM acceptable inter enum + sup 21x10 84.9 11.6 - 0.74 0.9 99.6 - - TEM acceptable TOTAL intra+inter 20x10 - - - 0.75 0.9 99.6 -0.4 0.48 TEM acceptable

TOTAL+ sup 21x10 - - - 0.75 0.9 99.6 - - TEM acceptable

44 | P a g e

STANDARDIZATION TEST RESULT FOR MID UPPER ARM CIRCUMFERENCE subject Technical TEM/me Coef of Bias from Bias from MUAC mean SD max result s error an reliability superv median TEM # mm mm mm TEM (%) R (%) Bias (mm) Bias (mm) (mm) Supervisor 10 14.1 1.1 0.2 0.07 0.5 99.6 - -0.38 TEM good Enumerator 1 10 14.1 1.2 0.6 0.14 1 98.5 0.02 -0.37 TEM good Enumerator 2 10 14 1.2 0.7 0.2 1.4 97 -0.1 -0.48 TEM good Enumerator 3 10 14.1 1.2 0.7 0.2 1.4 97.2 0 -0.38 TEM good Enumerator 4 10 14.1 1.1 0.8 0.21 1.5 96.2 -0.03 -0.41 TEM good Enumerator 5 10 14 1.1 0.5 0.13 0.9 98.6 -0.11 -0.5 TEM good Enumerator 6 10 14.2 1.2 0.6 0.15 1.1 98.3 0.05 -0.34 TEM good Enumerator 7 10 14.2 1.2 0.1 0.04 0.3 99.9 0.06 -0.33 TEM good Enumerator 8 10 14 1.1 0.5 0.13 1 98.5 -0.09 -0.47 TEM good Enumerator 9 10 14.1 1.2 0.4 0.1 0.7 99.3 -0.01 -0.39 TEM good Enumerator 10 10 14.1 1.2 0.1 0.04 0.3 99.9 -0.07 -0.45 TEM good Enumerator 11 10 14.1 1.1 0.1 0.07 0.5 99.6 -0.01 -0.39 TEM good Enumerator 12 10 14.1 1.2 0.2 0.07 0.5 99.7 -0.02 -0.41 TEM good Enumerator 13 10 14.1 1.1 0.3 0.09 0.6 99.4 -0.02 -0.4 TEM good Enumerator 14 10 14.1 1.1 0.1 0.07 0.5 99.6 -0.06 -0.44 TEM good Enumerator 15 10 14.2 1.1 0.1 0.03 0.2 99.9 0.04 -0.34 TEM good Enumerator 16 10 14.2 1.1 0.2 0.08 0.5 99.5 0.06 -0.32 TEM good Enumerator 17 10 14.1 1.1 0.2 0.07 0.5 99.6 0.01 -0.37 TEM good Enumerator 18 10 14.1 1.1 0.1 0.06 0.4 99.7 0.02 -0.36 TEM good Enumerator 19 10 14.1 1.1 0.1 0.07 0.5 99.6 0.01 -0.38 TEM good Enumerator 20 10 14.1 1.1 0.1 0.06 0.4 99.7 0.01 -0.37 TEM good enum inter 1st 20x10 14.1 1.1 - 0.17 1.2 97.5 - - TEM good enum inter 2nd 20x10 14.1 1.1 - 0.14 1 98.4 - - TEM good inter enum + sup 21x10 14.1 1.1 - 0.15 1.1 98 - - TEM good TOTAL intra+inter 20x10 - - - 0.2 1.4 96.9 -0.01 -0.39 TEM good TOTAL+ sup 21x10 - - - 0.19 1.4 97 - - TEM good 45 | P a g e

Appendix4: Maps of the areas

46 | P a g e

Appendix 5: Result tables for NCHS reference

Table 3.19: Prevalence of acute malnutrition based on weight-for-height z-scores (and/or oedema) and by sex

All Boys Girls n = 500 n = 254 n = 246 Prevalence of global (66) 13.2 % (35) 13.8 % (31) 12.6 % malnutrition (10.0 - 17.3 (9.4 - 19.8 (8.5 - 18.3 (<-2 z-score and/or oedema) 95% C.I.) 95% C.I.) 95% C.I.) Prevalence of moderate (63) 12.6 % (33) 13.0 % (30) 12.2 % malnutrition (9.4 - 16.7 (8.7 - 18.9 (8.1 - 17.9 (<-2 z-score and >=-3 z-score, no 95% C.I.) 95% C.I.) 95% C.I.) oedema) Prevalence of severe (3) 0.6 % (2) 0.8 % (1) 0.4 % malnutrition (0.2 - 1.8 (0.2 - 3.1 (0.1 - 3.1 (<-3 z-score and/or oedema) 95% C.I.) 95% C.I.) 95% C.I.) The prevalence of oedema is 0.0 %

Table 3.20: Prevalence of acute malnutrition by age, based on weight-for-height z-scores and/or oedema Severe wasting Moderate Normal (<-3 z-score) wasting (> = -2 z score) (>= -3 and <-2 z- score ) Age Total No. % No. % No. % (mo) no. 6-17 105 2 1.9 10 9.5 93 88.6 18-29 112 1 0.9 26 23.2 85 75.9 30-41 119 0 0.0 12 10.1 107 89.9 42-53 111 0 0.0 10 9.0 101 91.0 54-59 53 0 0.0 5 9.4 48 90.6 Total 500 3 0.6 63 12.6 434 86.8

Table 3.21: Prevalence of acute malnutrition based on the percentage of the median and/or oedema

n = 500 Prevalence of global acute (36) 7.2 % malnutrition (4.9 - 10.4 95% (<80% and/or oedema) C.I.) Prevalence of moderate acute (36) 7.2 % malnutrition (4.9 - 10.4 95% (<80% and >= 70%, no oedema) C.I.) Prevalence of severe acute (0) 0.0 % malnutrition (0.0 - 0.0 95% (<70% and/or oedema) C.I.)

Table 3.22: Prevalence of underweight based on weight-for-age z-scores by sex

47 | P a g e

All Boys Girls n = 500 n = 254 n = 246 Prevalence of underweight (231) 46.2 % (119) 46.9 % (112) 45.5 % (<-2 z-score) (40.1 - 52.4 (39.5 - 54.3 (38.5 - 52.7 95% C.I.) 95% C.I.) 95% C.I.) Prevalence of moderate (194) 38.8 % (100) 39.4 % (94) 38.2 % underweight (33.1 - 44.8 (33.0 - 46.1 (31.4 - 45.5 (<-2 z-score and >=-3 z-score) 95% C.I.) 95% C.I.) 95% C.I.) Prevalence of severe (37) 7.4 % (19) 7.5 % (18) 7.3 % underweight (4.9 - 11.1 (4.7 - 11.6 (4.4 - 11.8 (<-3 z-score) 95% C.I.) 95% C.I.) 95% C.I.)

Table 3.23: Prevalence of stunting based on height-for-age z-scores and by sex

All Boys Girls n = 498 n = 254 n = 244 Prevalence of stunting (176) 35.3 % (86) 33.9 % (90) 36.9 % (<-2 z-score) (29.5 - 41.6 (27.6 - 40.7 (29.7 - 44.8 95% C.I.) 95% C.I.) 95% C.I.) Prevalence of moderate stunting (137) 27.5 % (75) 29.5 % (62) 25.4 % (<-2 z-score and >=-3 z-score) (22.8 - 32.8 (23.6 - 36.2 (20.1 - 31.5 95% C.I.) 95% C.I.) 95% C.I.) Prevalence of severe stunting (39) 7.8 % (11) 4.3 % (28) 11.5 % (<-3 z-score) (5.6 - 10.9 (2.2 - 8.3 (7.6 - 16.9 95% C.I.) 95% C.I.) 95% C.I.)

48 | P a g e

Appendix6: Distribution of wasted, stunted and underweight cases in clusters

Graph-4: Distribution of GAM (WHZ<-2SD) cases in clusters

Graph-5: Distribution of Stunted cases (HAZ <-2SD) in clusters

49 | P a g e

Graph-6: Distribution of underweight cases (WAZ<-2SD) in clusters

50 | P a g e

Appendix-7: Questionnaire

INTEGRATED NUTRITION SURVEY HABIGANJ – 2015

“Hello my name is ______. I am working with ACF (explain what ACF does?). Please let me introduce you to the other team members: ______and ______. We are here today to gather household information related to nutrition, child feeding practices and food security. We have randomly identified required number of villages from the district and required number of households from the selected villages for conducting the survey. Your village and household is one of those selected village and household. If there are any children under 5 in the household, we would like to take some measurements (weight, height, MUAC/ explain) to find out the percentage of under 5 children who are malnourished in this Village/District. All the information collected will be used only for analytical purposes and will remain confidential. Please note that it is not currently known what actions/ if any will be taken after the results of the survey are finalized. Is it okay if we ask you some questions? (If yes) Do you have any questions before we start?”

Must write answer using a Pen but not Pencil

Upazila: ------Union: ------Ward: ------Village: ------Consent: ------Date: ------Team N0: ------Cluster N0: ------HH N0: ------HH Serial N0: ------1. Household Demographyt

Household Information Family Size Total male Total female Gender of Age of the 1. Male 2. Female HH Head HH Head

2. Age composition of HH members

Age Categories Total number of family members Infants (0-12m): Young Children (<5 Yrs) Young Children (6 – 17ys): Adults (18-50ys): Elderly (> 50ys): …

3. Anthropometry (6-59m)

Ask if soap is available in the household for hand washing before touching the

child for measurement? (Yes/No)

Child SEX Date of AGE Weight Height/ Oedema MUAC N0 (Male/Female) Birth (Months) (Kg) Length14 (Y/N) (cm) (mm/dd/yy) ±0.1kg (cm) 0.1cm

1 2

14 Height measurement standing when child is ≥24 months (height proxy ≥87 cm) and lying down when child is < 24 months (< 87 cm) 51 | P a g e

3

4. Morbidity (6-59 MONTHS) In the past 2 WEEKS has your child had the following illnesses? Put tick sign () if yes, if not then do not ask the following questions and go to next questions. Child Diarrhoea Is the If the child is not given Only Acute Respiratory others 0 N child ORS then why? (Did not fever Infections (Cough, (specify) given go to Clinic=1, No ORS in breathing difficulties, ORS? Clinic=2, Clinic is far=3, chest in-drawing and Others (specify)=4 high respiratory rate) 1 2 3

5. From what type of labour did this household member primarily earn income in the past two months?

5.1 Main Occupation 11 = Petty businessman 5.2 Reason for Not Working 01 = Selling of paddy 12 = Household help 0 = Worked everyday 02 = Selling of cash crops or 13 = Selling of Jhum crop 1 = There was no work/did other crops 14 = No income not get work 03 = Agriculture Day labourer 15= Commercial sale of Poultry/ 2 = was sick/disable/unable to 04 = Unskilled day labourer Livestock/Dairy work 05 = Skilled day labourer 16 = Small scale sale of Poultry/ 3 = There was no need to 06 = Rickshaw/van/cart puller Livestock/Dairy raised or work/ unwillingness /baby, taxi driver /boatman produced at home 4 = went to visit relatives 07 = Fisherman 16= Selling of Handicrafts 5 = Students/Housewives 08 = Salaried worker products 6 = Others 09 = Professional 17= Vegetable Cultivation 7 = Not applicable 10 = Large to medium 18= Fish cultivation & selling of businessman fish 19 = Day labour in fish farms 55= Others (Specify…………………) 88= Don’t know

6. Food consumption score

I would like to ask you about all the different food items that your household members have eaten in the last 7 days. Could you please tell me how many days in the last 7 days your household member (s) have eaten the following foods? (We are asking about food groups eaten by the entire household members and if the quantity of food consumed is less than one table spoon do not consider that food item. If two or more food of the same food group is eaten in one day the number of day eaten will be one. No. of Days the food What was the main source of group is eaten in last 7 this food in the last 7 days? days (from 0 to 7) 1= Own production; 2= Purchase Food Group on cash; 3= Purchase on credit; 4= Bartering; 5= Gifts/Charity; 6= Collecting wild foods; 7= Food aid (Gov, NGO/WFP) Starchy foods (rice, wheat, muri, potatoes, sweet potatoes, maize, khichuri)

Vegetables (any type)

52 | P a g e

Dal or peanuts (any type) (khichuri) Edible Oil Meat, off all type, fishes and eggs

Milk or milk product (yogurt, milk sweets) Fruits

Sugar, gur

7. If your household depends on own cereal production how long the cereal stock lasts following the harvest?

1. less than 15 days  2. 15-30 days  3. 30 days  4. 30-60 days  5. 60+ days

8. Selected woman (YESTERDAY) (one per household) DIETARY DIVERSITY

Food category Examples of foods Yesterday during the day or night, did you consume the following food items? No Yes Starchy Staples Rice, wheat, muri, potatoes, sweet potatoes, 0 1 maize, kichuri Legumes and Nuts Dal, cooked dry beans, peas, peanuts, other 0 1 seeds/beans/ khichuri Dark green leafy All kinds of leafy vegetables 0 1 vegetables Red/orange/yellow Ripe mangoes, papaya, jackfruits other 0 1 fruits red/yellow or orange fruit Red/orange/yellow orange sweet potato,pumkin, carrot or other 0 1 vegetables yellow or orange vegetable Vitamin C rich fruits Guava, Strawberry, Lemon, Orange, Lychees, 0 1 Pineapple, Mango,Grapes Vitamin C rich Gourd, Broccoli, Cauli flower, tomatoes, Green 0 1 vegetables Cabbage Other vegetables or Cabbage, turnips, , bananas, apples 0 1 fruits Eggs Hen/duck, other birds, or fish eggs 0 1 Organ meat Liver, kidney, gizzards 0 1 Small Fish Small Fish Eaten Whole with Bones (i.e.Kachki, 0 1 Mola, dhela, chapila, batashi, small prawn) Flesh Foods and Small Beef, Pork, Veal, Lamb, Goat, Chicken, Duck, 0 1 animal Protein Large Whole Fish and Shell Fish Dairy Milk, cheese, yogurt or other milk products 0 1 Edible Oil Foods containing oil, fat, butter 0 1 Sugar, honey, 0 1 molasses

53 | P a g e

WASH (Water, Sanitation and Hygiene)

9. What type of latrine does Piped sewer system……………….. 01 your household use? Septic tank…………………………….. 02 Ring slab with water seal………… 03 Ring slab without water seal….. 04 Pit latrine with slab………………… 05 Pit latrine without slab………….. 06 Hanging latrine……………………. 07 No facility (Bush/open field/river pond side)……………… 08 Others, specify……………………… 09 10. Do you share this latrine No………………………………………… 0 with other households? Yes……………………………………… 1 No facility………………………………. 9 11. What is the main source of PIPED WATER drinking water of your Piped to dwelling……… 01 household? Piped to yard/plot……… 02 Public tap…………………… 03 TUBE WELL OR BOREHOLE Shared………………………. 04 Household…………………. 05

DUG WELL 06 Protected…………………… 07 Unprotected……………… 08 RAINWATER

SURFACE WATER (Pond/ River/ 09 Canal/Hawar/ irrigation 10 channels)……………………………….. WATERTANKER………………………

12. Reduced Coping Strategy Index (related to food consumption)

In the past 7 days, if there have been times when you did not have Frequency: enough food or money to buy food, how often has your household had No. of Days the strategy repeated in to: last 7 days (from 0 to 7) a Rely on less preferred and less expensive foods |___| . b . Borrow food, or rely on help from a friend or relative |___|

c . Limit portion size at mealtimes |___|

d Restrict consumption by adults in order for small children to eat . |___|

e Reduce number of meals eaten in a day . |___|

54 | P a g e