Nutrition Survey Using SMART Method · PPS Probability Proportion to Size PSU Primary Sampling Unit...

i

Nutrition Survey Using SMART Method In Typhoon Haiyan-affected areas of Regions VI, VII and VIII

The Philippines: 03 February – 14 March 2014

FINAL REPORT

March 2014 smartmethodology.org

ii Nutrition Survey using SMART Methodology for Typhoon Haiyan-affected areas of Regions VI, VII and VIII The Philippines: 03 February-14 March 2014

For additional information on the Survey,

Contact Details:

Jocelyn A. Juguan, Ph.D.

SMART Survey Consultant Manager

E: [email protected]

National Nutrition Council

http://www.nnc.gov.ph/about-nnc/contact-us


National Nutrition Council - Surveillance Officer


Nutrition Cluster - Information Management Officer


http://www.nnc.gov.ph/about-nnc/contact-us

mailto:[email protected]

mailto:[email protected]

iii Nutrition Survey using SMART Methodology for Typhoon Haiyan-affected areas of Regions VI, VII and VIII The Philippines: 03 February-14 March 2014

ACKNOWLEDGEMENTS

Grateful and sincere acknowledgements are extended to the following organizations and

individuals who contributed to the successful implementation of the Survey and the

completion of this report:

The National Nutrition Cluster led by Assistant Secretary of Health and National Nutrition

Council (NNC) Executive Director Maria-Bernardita T. Flores for her unceasing guidance and

support to the conduct of the SMART Survey, and for allowing NNC staff to participate as

team leaders in the SMART survey;

The Assessment and Monitoring Working Group of the National Nutrition Cluster for the

technical inputs and for formulating the initial questionnaires to be used in the Survey;

UNICEF (United Nations Children’s Fund) for technical support and funding; Ms. Victoria

Sauveplane, SMART (Standardized Monitoring and Assessment of Relief and Transitions)

Program Manager, ACF Canada for her technical assistance in designing the survey and

training the team in Tacloban City; Ms. Christine Klotz and Fanny Cassard, Chair of Assessment

and Monitoring Working Group, and Monitoring and Evaluation Officer, respectively of the

World Food Programme for their technical assistance in analyzing the data of the

Anthropometric Phase of the Survey and writing the preliminary and final report, and for

monitoring the teams during data collection in Region 6;

Mr. Alessandro Iellamo, UNICEF IYCF specialist, for his commitment and assistance in doing

the data analysis and write-up of the Maternal, Infant and Young Child Nutrition; and his

guidance in the report writing and power point presentation;

The Regional Nutrition Clusters: Region 6 led by NNC-Nutrition Program Coordinator (NPC)

Nona B. Tad-y; Region 7 by NNC-NPC Parolita A. Mission; and Region 8 by NNC-OIC Segundina

Devota A. Dilao; for their unselfish assistance in coordinating and endorsing the SMART

Survey and teams to the sample municipalities/cities and clusters (barangay);

The Local Nutrition Committees of the Local Government Units (LGUs), from provincial to

municipal level for their warm reception to the teams and assistance in the conduct of the

SMART survey in their respective clusters (barangays);

Barangay Chairpersons of sample barangays for their cooperation and unwavering support in

providing the teams with updated lists of households and other important information

relevant to data collection;

iv Nutrition Survey using SMART Methodology for Typhoon Haiyan-affected areas of Regions VI, VII and VIII The Philippines: 03 February-14 March 2014

ACF Tacloban and Roxas Base for providing administrative and logistical assistance to the

Survey Teams as they move from one base to the other; the support team from CDO for

facilitating the survey; Dr. Martin Parreño, Medico-Nutrition Coordinator, ACF Manila and Ms.

Elena Rivero, Nutrition Advisor, ACF Spain, for their technical support. Also, special thanks to

Mr. Oscar Fudalan and the nutrition team from Cotabato for their strong commitment and

unrelenting support;

Head of households that were sampled for their wholehearted cooperation to join the Survey,

and allowing their children to be weighed and measured;

Community assistants for leading the teams to the sample households and joining them

enthusiastically in whatever landscape it maybe, and to the drivers for taking good care of the

teams while on the road;

And last but not the least, to the Team Leaders and surveyors for braving the burning heat,

for hiking for hours, crossing hanging bridges, traversing schistosomiasis-endemic waters to

reach sampled households, who without their teamwork, commitment and dedication to

locate and reach sampled households, collected data will not be excellent.

v Nutrition Survey using SMART Methodology for Typhoon Haiyan-affected areas of Regions VI, VII and VIII The Philippines: 03 February-14 March 2014

Persons Responsible for the Survey

Survey

Consultant

Victoria Sauveplane ACF – International Canada

Survey Coordination Committee

Training Victoria Sauveplane SMART Program Manager, ACF –

International Canada

Dr. Jocelyn A. Juguan SMART Survey National Coordinator,

ACF – International Philippines

Assessment and

Monitoring

Working Group

(SMART Survey)

Lindsey Horton

Christine Klotz

Fanny Cassard

Chair, Assessment and Monitoring

Working Group – National Nutrition

Cluster, World Food Programme

Dr. Martin Parreño Medico-Nutrition Coordinator, ACF –

International Philippines

Henry Mdebwe Health and Nutrition Specialist –

UNICEF

Hygeia Ceres Catalina B. Gawe Chief, Nutrition Surveillance Division –

National Nutrition Council

Richard Wecker Information Management Officer,

Nutrition Cluster – UNICEF

Alessandro Iellamo IYCF Specialist – UNICEF

Frederich Christian S. Tan Nutrition Officer – National Nutrition

Council

vi Nutrition Survey using SMART Methodology for Typhoon Haiyan-affected areas of Regions VI, VII and VIII The Philippines: 03 February-14 March 2014

The Survey Team

Survey Team 1 Survey Team 3

Team Leader: Frederich Christian S. Tan Team Leader: Margarita D.C. Enriquez

Ayessa Eona A. Arante Mark Angelo C. Juanico

Cyril D. Amor Irene Mae C. Cuayzon

Survey Team 2 Survey Team 4

Team Leader: Ma. Corazon A. Oñate Team Leader: Jeline Marie M. Corpuz

Jonathan Renier B. Verzosa Jamella Reina S. Picorro

Sarrah Din A. Astang Vengiebelle V. Tiong

vii Nutrition Survey using SMART Methodology for Typhoon Haiyan-affected areas of Regions VI, VII and VIII The Philippines: 03 February-14 March 2014

PREFACE

On 25 March 2013, Secretary of Health Enrique T. Ona signed Department Personnel Order

2007-2492-A which designates the National Nutrition Council as the new chair of the Nutrition

Cluster. The members of the Nutrition Cluster include local and international non-government

organizations, and national government agencies. The Nutrition Cluster aims to ensure that

the nutritional status of affected populations especially the most vulnerable groups -- infants,

children, pregnant women and breastfeeding mothers, older persons, people with disabilities,

and people living with debilitating conditions -- will not worsen or deteriorate due to the

impact of emergency and disaster through linking with other cluster/sector groups and

establishing capacities at all levels.

After Typhoon Haiyan hit the country on November 8, 2013 and devastated Regions VI, VII

and VIII, the National Nutrition Cluster developed a Strategic Response Plan (SRP) to guide

the implementation of nutrition services in 81 municipalities that were identified as priority

areas in the three regions. To date, the Nutrition in Emergencies (NiE) activities have focused

on the promotion and protection of Infant and Young Child Feeding (IYCF), the prevention of

acute malnutrition and micronutrient deficiencies, and treatment of acute malnutrition.

The objectives of the SRP are to:

a. Promote optimal IYCF practices in emergencies, to protect and support 80% of the

breastfed and non- breastfed girls and boys aged between 0-23 and pregnant/lactating

women (PLW) in eight provinces of three affected regions during one year following the onset

of the emergency;

b. Ensure access to programmes that treat and prevent acute malnutrition to at least 50%

of vulnerable populations (boys and girls between 0-59 months, pregnant and lactating

women (PLW) and older people in seven provinces across three regions affected by Haiyan;

c. Ensure access to programmes that prevent and control micronutrient deficiencies

(anemia, vitamin A and other micronutrient deficiencies) in at least 50% of vulnerable

populations (children aged between 6-59 months and PLW) in 7 provinces;

d. Determine and evaluate the IYCF practices and nutritional status of vulnerable groups in

three regions affected by Haiyan between 3 months and 9 months following the onset of the

emergency; and

e. Ensure a timely, coordinated and effective nutrition response to all typhoon-affected

populations.

viii Nutrition Survey using SMART Methodology for Typhoon Haiyan-affected areas of Regions VI, VII and VIII The Philippines: 03 February-14 March 2014

Table of Contents

Acknowledgement ........................................................................................................ iii

Preface ......................................................................................................................... vii

List of Acronyms ........................................................................................................... x

Executive Summary ...................................................................................................... xii

1. Introduction ............................................................................................................. 1

1.1 Rationale for the Survey .............................................................................................. 2

1.2 Objectives of the Survey .............................................................................................. 2

2. Methodology ............................................................................................................ 3

2.1 Survey Design .............................................................................................................. 3

2.2 Target Population, Sampling Frame and Sample Size Calculation .............................. 3

2.3 Sampling Procedure ..................................................................................................... 4

2.4 Reserved Clusters ........................................................................................................ 6

2.5 Household Selection and Definition ............................................................................ 6

2.6 Questionnaire and Data Collected .............................................................................. 7

2.7 Training ........................................................................................................................ 8

2.8 Pilot Survey .................................................................................................................. 9

2.9 Anthropometric Measurements .................................................................................. 9

2.10 Standardization of the Anthropometric Tools .......................................................... 12

2.11 Coordination and Communication ............................................................................ 12

2.12 Survey Teams and Supervision .................................................................................. 13

2.13 Data Collection .......................................................................................................... 15

2.14 Data Management (Data Encoding and Data Validation) ......................................... 18

2.15 Data Analysis ............................................................................................................. 19

2.16 Limitations of the Survey and Recommendations to the Next Survey ..................... 21

3. Results ...................................................................................................................... 22

3.1 Description of Sample ................................................................................................. 22

Households ............................................................................................................... 22

Displacement ............................................................................................................ 22

4Ps ............................................................................................................................ 22

Children 6 – 59 months old ...................................................................................... 23

Women of Reproductive Age ................................................................................... 23

3.2 Review of Data Quality ................................................................................................ 24

3.3 Child Nutritional Status ............................................................................................... 25

Prevalence of Global Acute Malnutrition ................................................................. 26

Prevalence of Acute Malnutrition According to MUAC ........................................... 29

Prevalence of Chronic Malnutrition ......................................................................... 30

Prevalence of Underweight ...................................................................................... 32

ix Nutrition Survey using SMART Methodology for Typhoon Haiyan-affected areas of Regions VI, VII and VIII The Philippines: 03 February-14 March 2014

3.4 Morbidity/Health Status ............................................................................................... 34

3.5 Children Access to Programs/Services ........................................................................ 35

Measles Vaccination ................................................................................................. 35

Vitamin A Supplementation ..................................................................................... 36

Deworming ............................................................................................................... 36

MUAC Screening Coverage ....................................................................................... 38

Maternal, Infant and Young Children ....................................................................... 38

3.6 Women Nutritional Status ......................................................................................... 39

Mid-Uppder Arm Circumference (MUAC) Measurement of Women ....................... 39

Prenatal Care ............................................................................................................ 40

Iron and/or folic Acid Supplementation ................................................................... 40

3.7 Infant and Young Child Feeding .......................................................................... 41

4. Conclusion and Recommendations ............................................................................ 45

Annexes ....................................................................................................................... 49

Annex 1: List of Surveyed Barangays ............................................................................... 49

Annex 2: SMART Survey Questionnaires ......................................................................... 52

Annex 3: Data Dictionary ................................................................................................. 57

Annex 4: Lessons Learned and Recommendations ......................................................... 64

Annex 5: Data Quality ...................................................................................................... 75

x Nutrition Survey using SMART Methodology for Typhoon Haiyan-affected areas of Regions VI, VII and VIII The Philippines: 03 February-14 March 2014

List of Acronyms

ACF Action Contre La Faim

ARI Acute Respiratory Infection

CI Confidence Interval

CMAM Community-based Management of Acute Malnutrition

DALYS Disability Adjusted Life Years

DHS Demographic and Health Survey

DOH Department of Health

ENA Emergency Nutrition Assessment

FNRI Food and Nutrition Research Institute

GAM Global Acute Malnutrition

GP Garantisadong Pambata

HAZ Height-for-Age Z-Scores

HFA Height-for-Age

IEC Information, Education and Communication

IFA Iron-Folic Acid

IYCF Infant and Young Child Feeding

IYCF-E Infant and Young Child Feeding in Emergencies

LGU Local Government Unit

MAM Moderate Acute Malnutrition

MDG Millennium Development Goal

MHO Municipal Health Officer

MNAO Municipal Nutrition Action Officer

MNP Micronutrient Powder

MUAC Mid-Upper Arm Circumference

NIE Nutrition in Emergencies

NSCB National Statistical Coordination Board

NDRRM National Disaster Risk Reduction and Management

NDRRMC National Disaster Risk Reduction and Management Council

NGO Non-Government Organization

NNC National Nutrition Council

NNS National Nutrition Survey

NPC Nutrition Program Coordinator

PHO Public Health Office

PNAO Provincial Nutrition Action Officer

PPS Probability Proportion to Size

PSU Primary Sampling Unit

RC Reserve Cluster

RN Registered Nurse

xi Nutrition Survey using SMART Methodology for Typhoon Haiyan-affected areas of Regions VI, VII and VIII The Philippines: 03 February-14 March 2014

RND Registered Nutritionist-Dietitian

RN HEALS Registered Nurses for Health Enhancement and Local Service

SAM Severe Acute Malnutrition

SRP Strategic Response Plan

SD Standard Deviation

SMART Standardized Monitoring and Assessment of Relief and Transitions

UNICEF United Nations Children’s Fund

VAS Vitamin A Supplementation

WAZ Weight-for-Age Z-Scores

WFA Weight-for-Age

WFH Weight-for-Height

WFP World Food Programme

WHO World Health Organization

WHZ Weight-for-Height Z-Scores

WRA Women of Reproductive Age

4Ps Pantawid Pamilyang Pilipino Program

xii Nutrition Survey using SMART Methodology for Typhoon Haiyan-affected areas of Regions VI, VII and VIII The Philippines: 03 February-14 March 2014

Executive Summary

The SMART survey was conducted from 03 February –14 March 2014 in Typhoon Haiyan-

(locally known as Yolanda) affected areas in Region VI, VII, and VIII, Philippines. Data were

collected from a total of 1,386 households, 645 children 6-59 months old, 265 children 0-23

months old and 1,424 women of reproductive age (15-49 years) in 60 randomly selected

clusters (barangays). The survey aimed to: 1) determine the nutritional status of children 6-

59 months old and women 15-49 years old in typhoon affected areas; 2) assess coverage of

Vitamin A supplementation, deworming, and measles vaccination; 3) determine the

prevalence of child illness (diarrhea and acute respiratory infection); and 4) provide

information on infant and young child feeding practices.

A cross-sectional cluster sample survey with three stage sampling design was employed to

undertake the survey. Sample size was calculated to be representative of typhoon affected

areas for the three regions. The nutritional status of children was analyzed using WHO Child

Growth Standards and SMART flags were excluded (-3/+3 SD) from the observed survey mean.

SMART flags are children whose measurement exceeds ±3SD from the observed mean.

The cut-off of <-2 SD was used to determine GAM, stunting and underweight using weight-

for-age, height-for-age and weight-for age, respectively. MUAC was also used to assess acute

malnutrition among children 6-59 months and women of reproductive age, at cut-off of

125 mm and 210 mm, respectively.

Shown below are the key findings of the survey:

Child Nutritional Status

GAM1 rate based on Weight-for-Height Z-scores 4.1% (95% CI: 2.9-5.9)

SAM2 based on Weight-for-Height Z-scores 0.3% (95% CI: 0.1- 1.3)

GAM rate based on Measurement of Upper-Arm Circumference (MUAC)

0.8% (95% CI: 0.3-1.9)

Prevalence of edema 0.0%

Stunting among children 6-59 months old 30.6% (95% CI: 25.6-36.0)

Underweight children 6-59 months old 20.7% (95% CI: 17.3-24.6)

Morbidity/Health Status

Children aged 6-59 months that reportedly suffered from diarrhea during the past 24 hours

4.3% (95% CI: 2.3-6.4)

Children aged 6-59 months that reportedly suffered from ARI during the past two weeks

37.2% (95% CI: 29.2- 45.2)

Children Access to Programs/Services

1Global Acute Malnutrition (GAM) – the sum of Moderate Acute Malnutrition (MAM) and Severe Acute Malnutrition (SAM) 2 Severe Acute Malnutrition (SAM)

xiii Nutrition Survey using SMART Methodology for Typhoon Haiyan-affected areas of Regions VI, VII and VIII The Philippines: 03 February-14 March 2014

Measles vaccination for children aged 9-59 months based on those with vaccination card records

47.7%

Measles vaccination for children aged 9-59 months based on recall and those with cards.

91.2% (95% CI: 88.4-94.1)

Children Access to Programs/Services

Vitamin A received in the last 6 months (based solely on recall)

78.3% (95% CI: 73.3-83.3)

Deworming of children aged 12-59 months received in the last 6 months

54.7% (95% CI: 48.1- 61.3)

Women Nutritional Status

Undernutrition among women of reproductive age based on MUAC (<210 mm)

3.6% (95% CI: 2.5-4.8)

Infant and Young child Feeding Practices SMART survey (2014) Other data sources

Ever breastfed children 0-23 months old 86% (95% CI: 82-90) 96% (NNS-2011 regional average)

Never breastfed children 0-23 months old 14% 4% (NNS-2011 regional average)

Initiated breastfeeding of children 0-23 months old within 1 hour

58% (95% CI: 52-64)

Exclusive breastfeeding of infants less than 6 months of age

Less than 50% 3 57% (NNS-2011 regional average)

Infant formula given to infants less than 6 months of age the day before the interview

41% (95% CI: 31-53) 36% (DHS 2008)

Bottle fed children 0-23 months 46%(95% CI: 41-51) 39% (NNS-2011 regional average)

Solid/Semisolids introduced to children 6 to 8 months of age

90%

Iron-rich food consumed by children 6 to 8 months of age

66%

Unfortunately, in this survey no data are available for specific complementary feeding

practices, namely: introduction of complementary food, diversity and frequency. The GAM

rate of 6.0% recorded in 200 children 6-23 months of age being higher than the 4.1% overall

GAM rate, suggests that breastfeeding and complementary feeding practices after the age of

six months are far from optimal and appropriate.

3 Not possible to provide conclusive information on the infant feeding practices below six months of

age because of the small sample size for this age group.

xiv Nutrition Survey using SMART Methodology for Typhoon Haiyan-affected areas of Regions VI, VII and VIII The Philippines: 03 February-14 March 2014

Recommendations

The large discrepancy of results between WHZ and MUAC merits further study

WHZ and MUAC should both be used as independent criteria for enrollment of children in feeding programs

Results have programmatic implications as follows: o Recalibration of targets for acute malnutrition o Community-based Management of Acute Malnutrition (CMAM) activities should

target areas with high acute malnutrition, 10% GAM and above since this can be considered as a pre-existing ‘emergency’ situation

o Urgent need to identify and address bottlenecks in the implementation of Infant and Young Child Feeding interventions

Stunting of children 6-59 months old was observed in 1 out of 3 cases – a major

aggravating factor of ill health, malnutrition and underdevelopment among children.

Refocus activities toward the prevention of stunting of children during the ‘window of

opportunity’ (the time of pregnancy until the end of the first two years of life) through:

o Increase investments on the establishment of community, health and nutrition

facilities and spaces for promoting, supporting and protecting exclusive

breastfeeding for the first six months of life and continued breastfeeding up to two

years of age and beyond

o Support community-based programs to provide information and counseling on

optimal and appropriate complementary feeding practices

o Link with livelihood, food security and social welfare clusters and programs to

ensure increased access by vulnerable families to appropriate and safe diets

o Reduce infections by educating households on proper care and hygiene practices

and improving health seeking behavior for management of children’s infections

o Educate pregnant women about the importance of prenatal care and protect

maternal nutrition and health to prevent low birth weight babies;

o Promote regular growth monitoring and include measurement of length/height (not

just weight) in nutrition programs

o Invest in mass communication campaign for development based on preventive

activities: nutrition of pregnant women, promotion of exclusive breastfeeding,

complementary feeding and continued breastfeeding, good hygienic practices, the

production and consumption of complementary foods

o Continue distribution of micronutrient powder (MNP) for children 6-23 months old

after the end of the emergency operation

Strengthen efforts to increase coverage of vitamin A supplementation and deworming

(80% target) to reach and address key concerns, including:

o Raising awareness of mothers on micronutrient supplementation and deworming

campaigns

xv Nutrition Survey using SMART Methodology for Typhoon Haiyan-affected areas of Regions VI, VII and VIII The Philippines: 03 February-14 March 2014

o Strengthening distribution channels of vitamin A and deworming supplies and

monitoring and evaluation of campaigns

o Planning the achievement of mass activities around supplementation and deworming

at least twice a year, through Garantisadong Pambata days (April and October)

Eliminate suboptimal and inappropriate infant and young child feeding practices, that may

affect the health and development of children. Preliminary data show the need to ensure

that supportive environment for optimal and appropriate infant and young child feeding

practices are established or maintained through:

o Organization of IYCF community support systems through deployment of trained

local health and nutrition volunteers, NGOs, other civil society organizations

o Ensuring that high quality IYCF counseling and support services are integrated in the

health care delivery system

o Ensuring that all hospitals and health facilities assisting deliveries should comply with

AO 2007-0026 on the Revitalization of the Mother Baby Friendly Hospital Initiative in

Health Facilities with Maternity and Newborn Care Services

o Integrating IYCF counseling and monitoring in routine public health programs

(prenatal, EPI, post-partum) both health facility-based and out-reach efforts

o Strengthening the enforcement and accountability mechanisms for key legislations

like the Milk Code (EO51), the Enhanced Breastfeeding Promotion Act of 2009 (RA

10028) and the Rooming-in Act, (RA 7600)

o Developing and investing in massive communication/behavior and social change

campaign based on a sound formative research

o Monitoring and tracking the progress being made in reaching pregnant women and

mothers of children 0-23 with IYCF counseling services

1 Nutrition Survey using SMART Methodology for Typhoon Haiyan-affected areas of Regions VI, VII and VIII The Philippines: 03 February-14 March 2014

1. Introduction

Background

The Philippines is located along the typhoon belt in the Pacific, and

is visited by an average of 20 typhoons every year. It is also

situated in the “Pacific Ring of Fire” which makes it vulnerable to

frequent earthquakes and volcanic eruptions. Furthermore, its

geographical location and physical environment also contributes

to its high-susceptibility to tsunami, sea level rise, storm surges,

landslides, (flash) flooding, and drought.4

On 8 November 2013, the Philippines was severely hit by Typhoon Haiyan (Yolanda). Three of

the country’s 17 regions, namely Regions VI, VII and VIII, were devastated to unprecedented

levels in history. In Typhoon Haiyan’s wake, 14.1 million people were affected and 4.1 million

displaced. Infrastructure, water and sanitation, food security and medical services were

severely affected. Prior to the typhoon, GAM was estimated at 7.3% (NNS 2011), and

worldwide, the Philippines is one of the ten countries with high number of wasted under-5

children (769,000). In the severely affected regions, the prevalence of GAM ranged from 5.8%

in Region VI to 7.8% in Region VIII. Ten percent of lactating mothers were undernourished

and 16-33% of pregnant women were nutritionally at risk (NNS 2011).

After the declaration of a level 3 emergency category following Typhoon Haiyan, the

Philippines National Nutrition Cluster, led by the National Nutrition Council (NNC) of the

Department of Health (DOH), and the Cluster Lead Agency – UNICEF – convened with the

Global Nutrition Cluster. Four technical working groups were established to provide effective,

timely, and coordinated nutrition response: assessment and monitoring, advocacy and

communications, CMAM, and IYCF-E. The Nutrition Cluster developed a Strategic Response

Plan (SRP) for Nutrition interventions that was agreed and endorsed by the Strategic Advisory

Group. The 12-month Nutrition Cluster SRP focuses on the implementation of priority

nutrition interventions in emergency, namely support and promotion of IYCF, prevention and

management of acute malnutrition, and prevention of micronutrient deficiencies in 81

municipalities that have been prioritized across severely affected regions: VI, VII, and VII.

4 http://www.adrc.asia/nationinformation.php?NationCode=608&Lang=en&Mode=country


1.1 Rationale for the Survey

The SMART survey was carried out by the National

Nutrition Cluster in Regions VI, VII, and VIII to provide

empirical evidence and baseline information on the

nutrition situation of nutritionally-vulnerable groups

(children 6-59 months old and women 15-49 years

old) in severely affected regions. The survey, done in

partnership with UNICEF, ACF, WFP, and other

NGOs, was also intended to increase national

capacity in nutritional assessment using the SMART

methodology, and data analysis of relevant

indicators.

The results of the survey will support the National and Regional Nutrition Cluster in prioritizing

resources and planning programs to ensure that the needs of the affected population are

being met for improved nutritional status of the most vulnerable groups. More importantly,

the results will be the basis for the realignment of targets of the Nutrition Cluster SRP.

1.2 Objectives of the Survey

The main objectives of the survey are:

1. To provide updated information on the nutrition situation of children 6-59 months old in typhoon-affected villages the results of which will be used to improve programming of Nutrition Cluster partners

2. To assess coverage of Vitamin A supplementation and measles vaccination 3. To provide information on infant and young child feeding practices 4. To recommend immediate and medium/long-term interventions to save lives and

support livelihoods. Specifically, the survey aims to:

1. measure the prevalence of acute malnutrition in children aged 6-59 months old 2. measure the prevalence of stunting in children aged 6-59 months old

3. determine the coverage of vitamin A supplementation in the last six months among

children aged 6-59 months old

4. determine the coverage of measles vaccination among children aged 9-59 months old

5. determine the prevalence of child illness (acute respiratory infection and diarrhea) among children 6 to 59 months of age

6. estimate the prevalence of maternal malnutrition using MUAC among pregnant and still breastfeeding women

7. investigate infant and young child feeding practices among children 0-23 months old

8. recommend interventions to improve the nutrition situation


2. Methodology

2.1 Survey Design

The survey is a cross-sectional cluster survey that adopted the Standardized Monitoring and

Assessment of Relief and Transitions (SMART) methodology in determining the nutritional

status of the target population. The survey is of two parts: Part I is an Anthropometric Survey

that aims to determine the nutritional status of target population, and Part II is a survey on

the IYCF practices of children with ages 0-23 months old.

2.2 Target Population, Sampling Frame and Sample Size Calculation

The target population of the survey is households in areas that were severely hit by Typhoon

Haiyan in Regions VI, VII and VIII. The list of severely-affected areas from the National Disaster

Risk Reduction and Management (NDRRM) Situation Report as of 30 November 2013 was

used as the sampling frame for the selection of survey areas. Sample size was calculated using

a cross sectional anthropometric survey of the Emergency Nutrition Assessment (ENA)

software, November 2013 version. In calculating sample size, GAM prevalence of 10% and the

desired precision ±3% with design effect of 1.7 was used; Seven hundred eleven (711) was

computed to be the survey’s required sample size for children 6-59 months old. The

assumptions considered for each parameter are shown in Table 1.

Table 1. Sample Size Calculation for Children 6-59 Months Old

Parameter Value Assumption

Estimated Prevalence of

Global Acute Malnutrition

(GAM) in percentage (%)

10%

In the 2011 National Nutrition Survey, the

prevalence of GAM across Regions VI, VII and

VIII, ranged from 5.3% to 7.8%. To be on the

safe side, 10% was set, the upper limit of the

range was used as the prevalence.

± Desired precision ±3% Since the GAM prevalence is higher and it is a

baseline survey, a precision of ±3% was chosen.

Design Effect 1.7

This was set to allow the difference between

rural and urban, and between barangay and

evacuation centers. Certain regions have been

affected by flooding. Others have had more

NGO interventions.

Children to be included 711 Based on the formula above computed using

ENA

The SMART Methodology recommends converting the required number of children into a

number of households (fixed household method) for the following reasons: 1) it is easier to


create lists of households than lists of children in the area; 2) sample size calculated based on

number of children can encourage teams to skip households without any children, thus

introducing bias for household-level indicators; and 3) households can provide a common

metric for comparing sample size of many indicators. In doing this, the average household

size, percentage of children under-5 to total population and a non-response rate (%) of 10

was likewise used, and the result was 1,287. Table 2 shows the parameters and assumptions

used in this conversion.

Table 2. Sample Size Calculation for Household

Parameters Value Assumptions

Average household

size 5.05 Based on 2011 National Nutrition Survey

% Children under-5 13.5

Population data for the 81 municipalities as of

November 2013*

Total population - 397,071 = 13.5

Total number of children U5 - 2,941,268

% Non-response

households 10

The percentage of non-response households chosen

was relatively high to take into account that evacuation

camps were included within the sampling frame, and

there were a lot of people who were returning back to

their homes

Households to be

included (accord ing

to ENA)

1,287 households

*NDRRM, 30 November 2013

2.3 Sampling Procedure

The Three-stage Random Sampling

Due to the big size of the population of interest, a three-stage random sampling design was

employed in selecting sample households. Following the administrative division in the

country, the first stage was the selection of municipalities in the affected provinces. This

serves as the primary sampling unit (PSU). Then the second stage involved the selection of

barangay(s) from the PSUs based on probability proportional to size (PPS) method. Thus, in

areas where more population was affected, such as Tacloban City and Ormoc City, four

barangays from each city was randomly selected, while the rest have only one barangay per

PSU. The final stage of selection is the selection of households, as the ultimate sampling unit.


From these households, all children 0-59 months old and WRA became the subjects of the

survey.

Figure 1. Map showing the primary sampling units

Selection of Cluster (Barangay)

In determining the number of barangays to be included in the survey, importance was given

to ensuring high quality of data collection. In this regard, the number of households that can

be completed in a day by a survey team was first determined, and the time spent by a survey

team for the following activities were considered: 1) travel from home-base to survey area

and back, and preliminary activities like paying courtesy call to governors/mayors and

nutrition/health officers; and 2) time spent during actual data collection. This included

briefing of sample barangay and households on the survey objectives, methodology and how

households in the barangay were selected, getting household consent, interviewing and

measuring target population, lunch break and other procedures done such as sampling,

mapping and locating households, and encoding anthropometric measurements of children

before leaving the survey area.

After accounting for all inputs, it was decided that each Survey Team will have a workload of

12 households in a day, with the goal of covering one barangay in two days. In this set-up,

household coverage for each barangay was set at 24. The number of barangay samples was

then determined by dividing the required households (calculated by ENA) by the number of

household coverage in a barangay. As shown below, a total of barangays required as survey

areas was calculated at 54.

Number of Sample

Barangays =

1,287 households based on ENA

calculation = 54

24 households in each barangay


Barangays were selected at random using ENA, except for barangays in the municipalities of

Can-avid, Oras, and Guiuan, in Eastern Samar due to unavailability of list of households. These

areas were the most directly affected by Typhoon Haiyan. Data/record on the extent of

damage in terms of households affected was not available during the survey, making

barangay selection not feasible prior to data collection. Sampling was done only when the

Survey Team arrived in the area for data collection and had interviewed local officials or

agencies concerned.

Segmentation Technique

In this technique, the Team Leader asked municipal mayors/officials or agencies concerned

about the existing household population of the municipality and of each barangay. The Team

Leader listed all barangays and their corresponding household population in the

Segmentation Sheet as estimated. On the 3rd column of the form, the Team Leader computed

the cumulative population (in range) of each barangay and for all the barangays. Once done,

the Team Leader requested any of the attending officers to choose a number from the Table

of Random Numbers. The barangay whose cumulative population range included the random

number chosen, became the survey area.

2.4 Reserve Clusters

In the SMART methodology, the ENA software automatically pre-selected Reserve Clusters

(RC) when 10% of the survey areas could not be surveyed because of a peace and order

situation, inaccessibility of the area, or refusal by the sample household. In the survey, the

number of children with ages 6-59 months old in the households surveyed was low, thus the

six RCs were included to increase coverage of children bringing to 60 the clusters or barangays

surveyed. Annex 1 shows the complete list of surveyed barangays.

2.5 Household Selection and Definition

Households were selected by random, either by

simple or systematic methods. For each barangay

selected, the latest household list of the barangay

was used to make household selection feasible

without introducing selection bias. In the survey,

household is defined as a group of people, related

or not related, living together and sharing the

same cooking pot. Thus, a family living with

relatives or non-relatives (ie. parents, siblings and

neighbors) is considered a household. For


evacuation centers, based on the camp managers’ definition, a household will refer to all

individuals who respond to the same household leader. In evacuation centers, two or three

families sharing the same cooking pot, were considered one household. A household was

considered absent when its members slept there the night before the survey, but was out

during the two days of scheduled data collection in the barangay.

During the survey, with the exception of a few, families living in evacuation centers/camps

had already returned back to their respective homes. No households staying in evacuation

centers/camps were sampled. The survey was done three months after Typhoon Haiyan.

Before the team arrived in the assigned barangay, the local chief executive or contact person

was informed in advance by phone, of the Survey objectives and methods as well as the

procedure of selecting the area and households. The Survey team then requested the contact

person to ensure that an updated list of households in the barangay would be available during

data collection. When the team arrived in the barangay the Team Leader checked the

household list and inquired if any of the households listed moved out or were temporarily out

or absent. When household lists were confirmed, household populations were determined. If

a barangay had less than 250 households, simple random sampling was employed, and when

it had more than 250 households, segmentation was done.

2.6 Questionnaire and Data Collected

The questionnaire used during the survey was developed through a series of meetings of the

Assessment and Monitoring Working Group of the National Nutrition Cluster. The

questionnaire has two sections: a) The Anthropometric Survey for children 6-59 months old

and women of reproductive age (WRA) 15-49 years old; and b) the IYCF practices of children

0-23 months old. The Anthropometric questionnaire also included the introduction and

consent statement, questions on selected household characteristics like displacement, and

participation in 4P’s, and access to nutrition programs and services such as vitamin A

supplementation, deworming, measles vaccination, and pre-natal care. A copy of the

questionnaire and the data dictionary appear in Annexes 2 and 3, respectively.

The IYCF questionnaire contained questions on breastfeeding and complementary feeding,

and consumption of certain food items. These information were asked if the sample

household has a child aged 0-23 months old (1 questionnaire per child aged 0-23 months old).

Furthermore, the questionnaire was administered to the mother of the child or the main

caregiver who is responsible for feeding the child. Terms used are described in the data

dictionary.


The questionnaire was translated in Tagalog language, and pretested among mothers of

children who were the subjects in standardization test. Revisions were then made after the

pre-test. The Tagalog questionnaire was then used during the pilot survey. Again, all

comments and suggestions encountered were reviewed before the finalization and

reproduction of the questionnaires.

2.7 Training

To ensure data quality of the survey, a 5-day training was conducted prior to the data

collection. A total of 12 surveyors including the team leaders completed the training. The

training focused on the different aspects of the data collection emphasizing on building the

skills of the measurers on getting precise and accurate anthropometric measurements. A pre-

test was given to the surveyors to assess their level of knowledge prior to the training. The

training discussions covered the overview of the SMART survey, nutritional anthropometric

indicators and steps in measuring weight, length/height and MUAC and identification of

bilateral edema.

The consultant/trainer on SMART methodology from ACF-Canada, was the main trainor. Steps

and practical techniques in measuring weight, length/height and MUAC were demonstrated.

After the actual demonstration, each surveyor/trainee took turns in measuring the children.

In addition, criteria and steps of household random selection were also discussed in the

training. Emphasis was given when to do the segmentation in a cluster as well as when to opt

for systematic and simple random sampling.

Aside from the technical aspects in SMART training, standardization in writing numbers in the

questionnaires was also emphasized to avoid wrong data entry of the encoder due to

misreading of filled-up forms.

A data encoder also attended the training, and a special discussion was allotted for the

orientation on how to use ENA and ENA/EPI Info software. The data encoder participated in

the installation of the ENA and ENA/EPI Info software in the laptops assigned for each team.

Standardization Test

A standardization test was done during the training to gauge the accuracy and precision of

each surveyor/trainee in taking measurements. In this test, a total of 12 children with their

mothers/caregivers from one barangay were recruited. Ten of these children served as the

subjects while the remaining two served as substitutes in case the other children refused to

be measured. Each child was measured for weight (in kg), height (in cm) and MUAC (in mm).

Two series of measurements were done by the surveyor/trainee, each taking turn as main

measurer and as assistant measurer. These measurements were then compared with the

“gold standard” taken by the SMART ACF-Canada consultant/trainor. Enumerators’

measurements were then evaluated for precision and accuracy.


Evaluation of training

Pre- and post-tests were given to each surveyor/trainee at the start and end of the training.

Surveyor/trainee’s scores were compared (pre-test and post) to gauge the level of learning

from the training. Results of the pre and post-tests and the standardization test were then

carefully reviewed by the trainors. And based on this and with the consensus of trainors, the

surveyor/trainee with the highest score was given the slot of a Team Leader to complete the

required number of four. And the next four with highest scores were assigned as main

measurers while the rest as the assistant measurers.

2.8 Pilot Survey

After each surveyor had been assigned with their tasks and duties as main and assistant

measurers and the four Survey Teams had been identified, a pilot survey was done in two

barangays in the municipality of Sta Fe, Leyte, not a survey area. The Pilot Survey aimed to

pre-test the methods and the questionnaires in actual scenario in preparation for the data

collection proper. Two teams were deployed in each of the two barangays.

Each team undertook each step in data

collection, from courtesy call to the barangay

officials, random selection of households,

actual visit to households, interviews, and

measurements of weight, length/height, and

MUAC, identification of bilateral edema, and

data entry in ENA software.

2.9 Anthropometric Measurements

Weight

In a household with children 6-59 months old, the

children’s weight was measured using Seca

Model 876. The scale is digital with 0.1 kg

graduation. Before measuring weight, the

weighing scale was carefully placed in a flat and

steady surface in the household. If there was no

flat surface in the household, the scale was

placed on a piece of plywood that each Survey

Team carried. Tared or double-weighing mother

and child was done for children who were not yet

able to stand and who were crying and/or afraid of the equipment. Older children who are

able to stand alone were measured independently.


The assistant measurer ensured that the child being measured was standing in the center of

the weighing scale. When the child was still, the measurer read aloud the measurement to

the team leader. The team leader would repeat the measurement reading to the measurer

prior to recording the measurement to ensure

correct data recording on the questionnaire.

Children were measured without clothing.

Children who refused to remove their clothes,

were weighed with clothes-on, and then the

team asked the mother if she can change the

clothes of the child so the team can measure

the clothes worn by child when the

measurement was taken. The weight of the

clothes was then subtracted from the weight of the child with clothes-on.

Length/Height

In taking the length or height of a child, both

main and assistant measurers made sure that

the child properly positioned before taking

height/length. Height boards (from UNICEF)

with 0.1 cm gradation was used in measuring

both recumbent length and standing height.

Shoes/slippers and hair accessories were

removed from the child to avoid interference in

getting the correct length and height

measurements.

Children 6-23 months were measured lying down to get recumbent length while standing

height was measured in children 24 months and above. In getting the length, the assistant

measurer with the help from the mother/caregiver carefully placed the head of the child

against the headboard. The main measurer, on the other hand, made sure that the child’s

body and legs were straight against the board while the heels were flat against the footboard

when reading the measurement.

Standing height, on the other hand, was measured in children 24 months and above. With the

help of the mother/caregiver, the child was made to stand straight against the height board.

The assistant measurer checked that the child’s buttocks, calves and heels were touching

against the board. When these were certain, the assistant measurer expressed this to the

main measurer who in turn checked that the child’s head and shoulder blades were also flat

against the height board. When these were done, the measurer read the height measurement

aloud to the team leader who repeated the reading loudly to the measurer, then record

measurement on the questionnaire.


In cases where height measurement was not possible, for example a child having cerebral

palsy, this was noted in the questionnaire. Other possible measurements such as weight and

MUAC were taken, but appropriately noted in the questionnaire.

Mid-Upper Arm Circumference (MUAC)

MUAC measurement was taken from children

6-59 months and women 15-49 years old

using MUAC tapes for children as no MUAC

tape for adults was available at the time of the

Survey. MUAC was measured on the left arm

of the children and women. In getting the

MUAC, the midpoint of the left arm was first

identified by bending the arm at 90 degrees

and then locating the tip of the shoulder blade

and tip of the elbow.

The midpoint was then lightly marked with a pen then the arm was put to relaxed position.

The MUAC tape was then carefully positioned around the arm making sure that it was not

too tight or too loose.

MUAC measurements of children were recorded in millimetres. For women, “1” was recorded

if MUAC is greater the 210 mm, and actual measurement was recorded if MUAC was less than

210 mm.

Bilateral Edema

The presence of bilateral edema was determined among children 6-59 months old by the

main measurer. Using the thumb finger of both hands, a light pressure was applied to both

feet for three seconds. Edema was present if an imprint on both feet were left for a few

seconds on the part of the foot where pressure was applied. If edema was detected, the team

leader double checked for verification and then recorded in the questionnaire with a “yes”

for presence of edema or a “no” for none.


2.10 Standardization of the Anthropometric Tools Prior to visiting each household, the teams

calibrated the weighing scales. Calibration

is done to make sure that the scale is

working properly and will get the same

measurement when measuring standard

test weights. Different reading of the test

weights signifies that the scale is not

working properly. During the survey,

standard test weights were not available,

so improvised test weights made of sealed bottled water weighing 0.4 kg were used.

However, after the monitoring visit of the head of the AWG, it was suggested to use the car

“brick” or “jack” to calibrate the weighing scales. The car “brick” or “jack” weighed more

compared to the improvised test weights and therefore more reliable to use as test weights.

Readings of the calibration were recorded daily in the anthropometric calibration form. On

the other hand, calibration of the UNICEF height boards was not possible due to unavailability

of standard instrument for height board such as a piece of wood with known length.

In addition to the daily calibration of the weighing scales, conditions of the weighing scales

and UNICEF height boards were checked before leaving the base. Batteries were replaced if

necessary. Each team also brought spare Seca digital weighing scale, height boards and MUAC

tapes in the field in case any of the equipment is not working properly.

2.11 Coordination and Communication

During the preparatory stage, a meeting was held with Nutrition Program Coordinators (NPCs)

of Regions VI, VII, and VIII of NNC to orient them on the rationale, objectives, methodology,

and coverage of the SMART survey. Using existing coordination structure of the NNC at the

national to subnational levels, letters addressed to mayors and barangay captains were

coursed through the regional offices of NNC for dissemination. The information was then

conveyed to the Provincial Health Office/Provincial Nutrition Office using email and or

facsimile to the sampled municipalities and barangays. Figure 2 shows the communication

process employed by the survey team during field coordination.

Figure 2. Communication Process on Field Coordination

SMART Survey

•Survey Manager

NNC Regional Office

•NPC

Provincial Office

•PHO/PNAO

Mayor's Office

•MHO/MNAO

Barangay Captain's Office

•Midwives/Volunteer Workers


Prior to data collection, the Team Leaders communicated and reconfirmed with the

respective MHO/MNAOs and barangay captains on the scheduled data collection in the

barangay to ensure that documents on updated household list and barangay map and its

boundaries were available.

2.12 Survey Teams and Supervision

The survey team was composed of 17-18 members: 1 Survey Manager, 4 team leaders, 4

main measurers, 4 assistant measurers, 2 data encoders, 1-2 logisticians, and 1 administrator.

The organogram is shown in Figure 3.

Figure 3. SMART Survey Organogram

Technical personnel were divided into 4 teams. Each team is composed of a Team Leader,

Main Measurer and Assistant Measurer. Three of the Team Leaders came from NNC and one

was identified in the training based on the results of the evaluation done by the trainors.

Nine of the surveyors were Registered Nurses (RN) of which seven came from Tacloban City

and one each from Negros Occidental and Davao del Sur. The three Team Leaders are

Registered Nutritionist-Dietitian (RND) of NNC, two of which are from the Nutrition

Surveillance Division and one from Nutrition Policy and Planning Division.

Survey Manager

Team 1

Team Leader

Measurer

Assistant Measurer

Team 2

Team Leader

Measurer

Assistant Measurer

Team 3

Team Leader

Measurer

Assistant Measurer

Team 4

Team Leader

Measurer

Assistant Measurer

1-2 Logisticians

Administrator

2 Data Encoders


Most of the surveyors had worked in other INGO projects, in public health under the RN HEALS

program of DOH and as hospital nurses.

The main roles and responsibilities of each member of the team and their core functions were

as follows:

1. The team Leader was responsible for introducing the team and the survey objective to

local officials, filling up the anthropometric and IYCF questionnaires, as well as data entry and

review of data quality;

2. The measurer was responsible for reading the anthropometric measurements (weight,

height/length and MUAC), checking bilateral edema in 6 -59 months old children and MUAC

of women of reproductive age;

3. The assistant measurer was responsible for assisting in getting weight and height and also

checking of edema of 6-59 months old children.

Since there were no field supervisor for this survey, the consultant from ACF-Canada, who

also trained the team, and the Survey Manager joined the team during the first few days of

data collection to guide them during their field work. A monitoring visit to the four teams was

also conducted by one of the partners, halfway during the implementation of the survey, to

observe how teams carry out the actual data collection in the field.

To compensate for the absence of field operation supervisors, each team diligently met after

data collection to discuss what had transpired during the day and to address immediate issues

and concerns. Team leaders also communicated with each other as often as possible to guide

and help each other during data collection. A staff meeting was also held at least once a week

to discuss field operations (technical, administrative and logistics matters). One team leader

performed as field supervisor and team leader at the same time, to make up for the absence

of field supervisor and to ensure that implementation of data collection is efficiently and

effectively done.

A two-day processing activity was done before covering the six (6) reserve clusters. During

this activity, the rationale of including the reserve clusters, and the team assignment were

discussed. A workshop was done to discuss survey experiences, bottlenecks, actions taken,

and recommendations. Each team, including the data encoder’s team reported their output

including a summary of lessons learned and recommendations for the conduct of the survey.

The last activity undertaken by the group was a psychological debriefing, wherein the team

shared freely their experiences during the implementation of the survey.


2.13 Data Collection

Courtesy Call to Local Officials

Before data collection, the Survey Teams

proceeded to the governor/mayor’s office with

the P/MHO/P/MNAO for courtesy call to

explain the rationale, objectives and

methodology of the Survey. The three stage

cluster sampling was also discussed for the

local officials to understand how barangay(s)

was/were selected. The mayor’s

approval/endorsement will then signify that

the team can proceed to the sampled barangays.

In the barangay, the Survey Team first met with the barangay officials and volunteer workers

to also explain once more the rationale, objective, and methodology of the survey. Using the

updated list of households, household selection was done.

Due to the devastation brought by the typhoon, there were cases that household list and map

were not available, and the Survey team together with the barangay officials drew the

barangay map.

In instances when preparing the list and map was not feasible, the Survey team did complete

enumeration of the households, and once completed, applied simple random or systematic

random sampling.

Once the twenty four (24) households were identified, the

Survey Team asked the barangay if someone from the

officials/volunteer workers, who knows the area, could serve as

their guide in locating the households. These assistants also

served as translators/interpreters when

the need arose.

Survey team with Mayor Edward C. Codilla (center) and MNAO Rheisa Lydia S. Nastor

(center left)during the courtesy call in Ormoc City


Data Collection in Households

Once a household was located, the Survey team met with the head of the household and

briefly explained the purpose of the visit. Consent was also solicited prior to proceeding in

anthropometric measurement and interview.

During the interview, sample vitamin A supplements, Plumpy Doz, micronutrient powder and

high energy biscuits were shown to the mothers/caregivers to facilitate easy recall of the

respondents on the children’s health services received and infant and young child feeding

practices. All interviews were conducted by the team leaders. In cases where the respondents

could not understand Tagalog, enumerators or community assistants translated the question

into local dialect and the respondent’s answers were translated to Tagalog.

WRA 15-49 years of age in the household were interviewed with basic questions on marital

and pregnancy status, and whether the pregnant women paid visits to the local health clinic

for prenatal check-up and whether they took iron-folic acid supplement. Left MUAC was also

measured in women present in the household at the time of visit. Women who were at work

or at school at the time of visit were revisited when possible and information was asked from

the respondents from the household, usually the husband, grandparents, relatives and

mothers of the women.

All anthropometric measurements and other information were written in a standard

questionnaire for children 6-59 months old, WRA and IYCF practices, respectively. Before

proceeding to the next household, the team checked the questionnaires to verify if datalines

were not filled-out correctly; and at the end of the survey in a given area, the team verified

all data prior to encoding. The team leader encoded all the anthropometric measurements

taken from the children in the ENA software with assistance from the two measurers.


When the 24 households in a cluster had been visited, the team updated the data encoder on

the coverage of the survey via text message, which included the total number of households

visited, households absent, eligible children, children measured, eligible women, women

measured, children with normal nutritional status, MAM, SAM and acute malnutrition cases,

as well as pregnant and breastfeeding women. This information was regularly collated for

updates on the coverage of the survey.

Identification of MAM and SAM Cases

After measurements of the eligible child were taken, nutritional

status was identified and the caregiver/mother was informed. A

referral form was issued if the child was identified as either MAM or

SAM. The team referred the identified MAM or SAM cases to a health

worker or to the rural health unit. A unisex table was initially used as

reference. However, it was observed that there were missed cases

using unisex table as compared to those identified by ENA software

based from the Z-score generated for weight for height.

After a thorough discussion among survey teams, it was agreed to use the WHO Child Growth

Standard Table instead (weight-for-length/height for boys and girls 0-23 and 24-59 months

old) in the latter part of the survey. Missed cases and misclassifications were minimized.

A master list was prepared for SAM and MAM cases identified during this survey using the

WHO 2006 Child Growth Standard Table. The list was shared to the Nutrition Cluster partners

for the management of acute malnutrition.


2.14 Data Management (Data Encoding and Data Validation)

After data collection, the team leaders submitted survey questionnaires by barangay to data

encoder. Data encoders then checked the submitted questionnaires and made a control list.

Data was encoded using the EPI info software.

The data were encoded in four sections, namely: (1) general information, (2) child

anthropometric data, child morbidity/health status and access to programs and services, (3)

women nutritional status, and (4) IYCF practices.The process of data entry is shown in Figure

4 below.

Double data entry was not done in the survey because of time constraints. To ensure that

data were correctly encoded, validation was done and the process is shown in Figure 5.

Figure 4. Data Entry Process

Figure 5. Data Validation Process

YES

NO

DATA on

General Info

Child Anthro

WRA

IYCF

Each team to validate

their own data Correction

Team to accomplish

data validation form

Encoder collects data

validation forms

Encoder updates the

record END

YES

END

Check content

and count

survey forms

Return folder

to Team Leader

Assign # in

survey form

Encode the ff:

General Info

Child Anthro

WRA

IYCF

Complete

NO


2.15 Data Analysis

Nutritional Anthropometric Indicators

The anthropometric data of children 6-59 months old were analysed using the latest version

of ENA 2011 (16 November 2013). The ENA software calculates the Z-scores for weight-for-

height (WH), height-for-age (HA) and weight-for-age (WA). Using Z-scores in reference to the

WHO Child Growth Standards, the following cut-offs were used to determine the prevalence

of wasting, stunting and underweight:

Table 3. Cut-off points for definition of Global, Moderate, and Severe Acute Malnutrition using

WHZ (WHO 2006)

Classification of Acute

Malnutrition Weight-for-Height Z-Scores

Global < -2 and/or bilateral edema

Moderate < -2 SD and > -3 SD, no edema

Severe < -3 SD and/or bilateral edema

Table 4. Cut-off points for definition of Stunting and Underweight using HAZ and WAZ (WHO

2006)

Height-for-Age Weight-for-Age

Classification Z-Scores Classification Z-Scores

Stunted < -2 SD Underweight < -2 SD

Moderately Stunted < -2 SD and > -3 SD Moderately

Underweight < -2 SD and > -3 SD

Severely Stunted < -3 SD Severely

Underweight < -3 SD

Acute Malnutrition/Wasting

Wasting was estimated according to the weight-for-height of each child and/or presence of

bilateral edema. Weight-for-height is an age independent indicator that assesses body mass

or weight in relation to length/height;it detects whether children are wasted or severely

wasted. Wasting is usually a result of current lack of food or illnesses that leads to acute and

severe weight loss and possible onset of malnutrition. It is particularly useful to consider this


variable in an emergency situation. Below are the cut-off points in determining acute

malnutrition:

Underweight

Underweight was estimated according to weight-for-age index. Weight-for-age is an indicator

that reflects body weight relative to the child’s age. It is used to assess whether a child is

underweight or severely underweight. It takes into account both acute and chronic

malnutrition.

Stunting

On the other hand, stunting was determined using height-for-age index. It is an indicator of

linear growth of a child, and reflects prolonged or chronic lack of food and repeated

infections. The effects of low height-for-age or stunting are largely irreversible by the end of

2 years of age which includes delayed motor development, impaired cognitive function and

poor school performance. Stunting represents long-term effects of malnutrition and is not

sensitive in assessing acute malnutrition.

For the measures of MUAC, the standards in Table 5 are taken from the WHO Child Growth

Standards and the identification of severe acute malnutrition in infants and children, 2009.

Moreover, MUAC is used in rapid screening of acute malnutrition for children 6-59 months at

a high risk of mortality associated with malnutrition. Below are the cut-off points for MUAC:

Table 5. Cut-offs for Definition of Acute Malnutrition Defined by MUAC, Children

Classification of Acute Malnutrition MUAC

Global <12.5 cm and/or edema

Acute <12.5 cm and >11.5 cm

Severe <11.5 cm and/or edema

For WRA 15-49 years old, the survey team used the cut-off point of 210 mm for identification

of nutritional risk among WRA. WRA with MUAC value of > 210 mm were recorded as “1”,

while the exact MUAC measurement was written if MUAC was < 210 mm.

Other data were processed using SPSS version 17 and STATA version 12. Frequency counts and cross-tabulations were mainly used for statistical analysis. For the nutritional status of women (pregnant and non-pregnant) data processed was transformed into the following: 1) results of the MUAC measurement, 2) access to antenatal


care services by pregnant women at the time of the interview, and 3) consumption of IFA by pregnant women at the time of the interview.

For the analysis of IYCF practices, standard indicators were generated: 1) early initiation of breastfeeding for children 0-23 months, 2) exclusive breastfeeding rate (0-<6 months old), 3) consumption of complementary foods of children 6-8 months old, and 4) bottle-feeding for children 0-23 months old.

Other indicators were generated, like the consumption of infant formula by infants less than six (6) months. One of the major limitations of the data analysis was the small sample size available to review and process data related to IYCF practices of children 0-23 months.

2.16 Limitations of the Survey and Recommendations for the Next Surveys

The sample size was calculated to be representative of typhoon affected areas for the 3

regions (VI, VII, VIII), but not representative for each region. The malnutrition rate for wasting,

stunting and underweight best reflects the nutrition situation of the three regions but does

not surface malnutrition rate for each region. Also, the sample size was based on attaining

the number of 6-59 months old children to represent the three regions.

During the implementation phase, no regular monitoring of the teams was done due to the

absence of a field supervisor. However, this was compensated through close coordination and

regular meetings between the field teams.

In taking the MUAC measurements of WRA, MUAC tapes for children were used in the

absence of WRA MUAC tapes in the Philippines at the time of the survey. The MUAC tapes for

children were used to estimate the midpoint of the left arm of WRA.

For the next survey, it is recommended that adequate time be allocated to undertake

implementation-related issues and concerns. For the smooth implementation of the survey,

there should be a field supervisor to closely monitor the survey team and its overall operation

in the field. Annex 4 shows the summary of lessons learned and recommendations of the

SMART Survey Teams developed during a processing activity conducted before the survey

ended. The lessons will be helpful in planning for the next SMART Surveys.


3. Results

3.1 Description of Sample

Households

Among the 54 clusters that were planned to be reached, all were surveyed and 6 reserve

clusters were added in order to increase the sample size of children 6-59 months of age. Data

were collected from a total of 1,386 households, 645 children 6-59 months, 265 children 0-

23 months and 1,424 women of reproductive age (15-49 years) in 60 clusters (barangays).

The number of households planned and the number of households with completed interviews

are shown in Table 6. The table also shows the number of children 6-59 months old planned

and surveyed.

Table 6. Number and percentage of households and children 6-59 months old planned and

interviewed.

Number of

HH planned

Number of

HH surveyed

% surveyed/

planned

Number of

children 6-

59 months

planned

Number of

children 6-

59 months

surveyed

% surveyed/

planned

1440 1386 96.3 711 645 90.7

Displacement

Among the 1,385 households surveyed, 874 (63.1%) have been displaced since Haiyan.

Of those displaced, more than 70% were living in their original homes and less than 30%

were living either in spontaneous settlements, transitional shelters or host family at the

time of interview (Table 7).

Table7. Type of residence of households at the time of interview

N

Current Residence at the Time of Interview

Original Home

% (n)

Evacuation Centers

% (n)

Transitional Shelters

% (n)

Spontaneous Settlements

% (n)

Host Family % (n)

874 73.3

(641) 0.1 (1)

6.0 (52)

12.7 (111)

7.9 (69)


4Ps

Pantawid Pamilyang Pilipino Program or 4Ps is a conditional cash transfer program of the

Philippine government administered by the Department of Social Welfare and Development

(DSWD). It aims to eradicate extreme poverty in the Philippines by investing in health and

education particularly among children 0–14 years old.

The poorest among poor families as identified by the 2003 Small Area Estimate (SAE) survey

of the National Statistical Coordination Board (NSCB) are eligible for the program. 4Ps

households are identified through a proxy-means test using economic indicators such as

ownership of assets, type of housing, education of the household head, livelihood of the

family and access to water and sanitation facilities. Additional qualification is the presence in

a household of children 0–14 years old and/or pregnant women during the assessment. 4Ps

household have to comply with all the conditionalities set by the government to remain in

the program.

The survey found that 20.7% of households were recipients of 4Ps (Table 8).

Table 8. Percentage of households which are recipients of 4Ps

N Recipient of 4Ps Not Recipient

n % n %

1386 287 20.7 1099 79.3

Children 6-59 months old

The ratio of boys to girls among children 6-59 months of age was 1.07 boys to 1 girl, which is

within the acceptable threshold of 0.8 to 1.2.

Table 9. Distribution of children 6-59 months old by gender

N Boys Girls Ratio Boys/Girl

645 333 312 1.07

Women of Reproductive Age

In this survey, 1,424 women of reproductive age were interviewed. Forty-four percent were

married, 32.3% single, and about 20% were living together but were not married.

Furthermore, 78 or 5.5% were pregnant.


Table 10. Marital status of women 15-49 years of age

N

Single/

Never

married

%

(n)

Currently

married

%

(n)

Living

together

%

(n)

Seperated/

Divorced/

Annulled

%

(n)

Widowed

%

(n)

Don’t

know/

Missing

data

1424 32.3

(460)

44.0

(627)

19.7

(280)

1.0

(14)

1.0

(15)

2.0

(28)

Of the pregnant women (n=78), the mean age is 29.7 years and Figure 6 shows that 11.5%

are of ages 15-19 and more than half are 20-29.

Figure 6.Percent of pregnant women by age groups (N=78)

3.2 Review of Data Quality

Data were collected from 96.3% of households selected for the survey. Overall, 645 children

6 to 59 months of age and 1,424 women in of reproductive age (15-49 years) were

interviewed.

Based on the result of the Plausibility Check of ENA, 99.9% of children interviewed were found

to have age calculated from exact day, month and year. One team used the calendar of local

events only for one child.

Age distribution of children under five shows that slightly more children aged 30-59 months

were surveyed than those aged 6-29 months but this difference was not significant. The age

11.5

55.1

29.6

3.8

15-19 years 20-29 years

30-39 years 40-49 years


ratio of 6-29 months to 30-59 months was 0.76, while the value should be around 0.855

according to the Plausibility Check of ENA. Age distribution of women shows a slight under-

representation of older women compared to younger women but this difference was not

significant and no peak in age was found.

Details of data quality is are shown in Annex 5 of the report. The data quality review was done

after excluding anomalies using SMART flags. The Plausibility Check highlighted the excellent

quality of the anthropometric data, both in terms of sample representativeness and quality

of anthropometric measurements.

Standard deviation for the distribution of Weight-for-Height (0.95), Height-for-Age (1.05) and

Weight-for-Age (1.02) in Z-score fell within the acceptable range (0.8–1.2).

The sex ratio was found good (1.07).

There were no significant digit preferences for weight, height and MUAC measures.

For 35 children (5.5%), the measure of height did not follow the protocol (standing height for

children of 24 months or more and recumbent length for children less than 24 months). ENA

software applied a correction to these 35 children.

3.3 Child Nutritional Status

The nutritional status of children was analyzed using the WHO Child Growth Standards.

SMART flags (-3/+3 Z-scores) from the observed survey mean were used to exclude extreme

values. Table 11 shows the Z-scores, design effect, and the number of children with flag signs

and were excluded in the analysis.

Table 11. Mean Z-scores, design effects and excluded subjects using SMART flags (WHO

2006) for children 6-59 months

Indicator N Mean Z-scores

± SD

Design Effect

(Z-score < -2)

Z-scores

not

available

Z-scores

out of

range

Weight-for-Height 628 -0.44 ± 0.94 1.00 11 6

Height-for-Age 628 -1.48 ± 1.04 2.02 11 6

Weight-for-Age 632 -1.17 ± 1.02 1.27 4 9

5The standard demographic distribution used for the Plausibility Check for all surveys; based on the analysis of several

demographic studies and it takes into account child mortality as well.


Prevalence of Global Acute Malnutrition

Wasting or acute malnutrition is the condition represented by measures of thinness or

bilateral edema and represents current nutritional status. It represents child’s failure to

receive adequate nutrition in the period before the measurements and may be the result of

inadequate food intake or a recent episode of illness causing loss of weight and the onset of

malnutrition.

Figure 7. Distribution of Weight-for-Height in Z-score compared to WHO Standards (2006)

The above figure shows that the survey distribution of Weight-for-Height (in red) follows close

to Gaussian distribution (in green). The mean of Weight-for-Height in Z-score was - 0.44 with

a Standard Deviation (SD) of 0.95. A SD which is between 0.8 and 1.2 reflects that the data of

weight and height is of good quality.

Furthermore, the curve of the Survey population is slightly shifted to the left of the curve of

the reference population, indicating that the surveyed population has more malnourished

children than the reference population. Table 12 shows the prevalence of acute malnutrition

in children 6-59 months by sex and by age group.


Table 12. Prevalence of acute malnutrition (global, moderate and severe) in children 6-59

months of age, by sex and by age group (WHZ, WHO 2006)

N

GAM (WHZ <-2 Z-score and/or edema)

MAM (WHZ <-2 z-score &>-3

Z-score, no edema)

SAM (WHZ<-3 Z-score and/or edema)

% [95% CI]

(n)

% [95% CI]

(n)

% [95% CI]

(n)

All 628 4.1

[2.9-5.9] (26)

3.8 [2.6-5.6]

(24)

0.3 [0.1-1.3]

(2)

Boys 323 4.0

[2.3-7.0] (13)

3.7 [2.0-6.7]

(12)

0.3 [0.0-2.3]

(1)

Girls 305 4.3

[2.6-6.9] (13)

3.9 [2.3-6.5]

(12)

0.3 [0.0-2.4]

(1)

6-23 mo 200 6.0

[3.4-10.4] (12)

5.5 [3.0-10.0]

(11)

0.5 [0.1-3.6]

(1)

24-59 mo 428 3.3

[2.0-5.4] (14)

3.0 [1.8-5.2]

(13)

0.2 [0.0-1.7]

(1)

The GAM rate based on WHZ was 4.1% (95% CI: 2.9-5.9) and SAM was 0.3% (95% CI: 0.1- 1.3).

An equal number of boys and girls had GAM, and by age group, GAM prevalence is higher in

the 6-23 months old than the 24-59 months old.

No child had bilateral edema.

Of the 26 children 6-59 months of age affected by GAM, the two who were SAM and the 12

who were MAM came from households that are recipients of the 4Ps (Table 13). Further

analysis revealed that the risk of children to become acutely malnourished in 4Ps households

is four times more than the risk of children from non-4Ps households.


Table 13: Prevalence of MAM and SAM in children 6-59 months of age, recipients of the

4Ps (WHZ, WHO 2006)

N

MAM (WHZ <-2 Z-score &>-3

Z-score, no edema)

SAM (WHZ<-3 Z-score and/or edema)

% [95% CI]

(n)

% [95% CI]

(n)

Children 6-59 months Recipients of the 4Ps

164 7.3

[3.0-11.6] (12)

1.2 [0.5-3.0]

(2)

The figure below shows the GAM rate of NNS 2011 and SMART Survey. Although this two

surveys are not directly comparable because of different age groups (0-60 months and 6-59

months, respectively) and other contextual considerations, the NNS 2011 is still used in the

absence of baseline data to show the changes in NS of these typhoon-affected areas before

and after the onset of the typhoon.

Figure 8. Global Acute Malnutrition by region (GAM) NNS 2011 and GAM, Moderate Acute

Malnutrition (MAM) and Severe Acute Malnutrition (SAM) SMART Survey 2014 - (WHZ,

WHO 2006)

According to the WHO classification, the results of the survey showed that the prevalence of

GAM in the Survey population is considered "acceptable" (not exceeding the 5% threshold).

Compared with children 0-60 months old in the 7th National Nutrition Survey (NNS) that was

conducted in 2011, GAM prevalence of the Survey population is lower than the 2011 national

prevalence of 7.3%, and the mean average of the 3 regions which is 6.3% (FNRI, 2012).

1.7 1.02.6

0.3

5.85.3

7.8

4.1

0

1

2

3

4

5

6

7

8

9

Region VINNS 2011

(0-60 months)

Region VIINNS 2011

(0-60 months)

Region VIIINNS 2011

(0-60 months)

SMARTFeb. - March 2014

(6-59 months)

SAM

MAM


Prevalence of Acute Malnutrition According to MUAC

According to the 2008 WHO and UNICEF Joint Statement on Child Growth Standards and the

identification of SAM in infants and children, a MUAC measure of less than 115mm in children

6 to 59 months old is recognized as severe acute malnutrition. MUAC less than 115mm

indicates a high elevated risk of death. The use of MUAC measure in children is simple as it is

a unisex measure not standardized by age.

Figure 9. Cumulative Distribution of MUAC by sex

Based on MUAC, the prevalence of GAM was 0.8% (95% CI: 0.3-1.9) and SAM 0.2% (95% CI:

0.1-1.2). GAM and SAM prevalence calculated by MUAC was much lower than the prevalence

obtained from Weight-for-Height Z-scores.

Table 14. Prevalence of acute malnutrition (global, moderate and severe) in children 6-59

months of age based on MUAC, by sex and by age group

N

GAM (MUAC <12.5 cm and/or edema)

MAM (MUAC<12.5 cm and >11.5 cm)

SAM (MUAC<11.5 cm and/or edema)

% [95% CI]

(n)

% [95% CI]

(n)

% [95% CI]

(n)

All 639 0.8

[0.3-1.9] (5)

0.6 [0.2-1.6]

(4)

0.2 [0.0-1.2]

(1)

Boys 328 0.6

[0.2-2.4] (2)

0.3 [0.0-2.2]

(1)

0.3 [0.0-2.2]

(1) Girls 311 1.0 1.0 0.0


N

GAM (MUAC <12.5 cm and/or edema)

MAM (MUAC<12.5 cm and >11.5 cm)

SAM (MUAC<11.5 cm and/or edema)

% [95% CI]

(n)

% [95% CI]

(n)

% [95% CI]

(n) [0.3-3.0]

(3) [0.3-3.0]

(3) [0.0-0.0]

6-23 mo 200 2.0

[0.8-5.1] (4)

1.5 [0.5-4.5]

(3)

0.5 [0.1-3.6]

(1)

24-59 mo 439 0.2

[0.0-1.7] (1)

0.2 [0.0-1.7]

(1)

0.0 [0.0-0.0]

WHZ is considered to be the base measure for global acute malnutrition, but it should be

clearly noted that there is no gold standard measure for acute malnutrition. MUAC is an

important measure of acute malnutrition that has a much closer relation to infant and child

mortality than Weight-for-Height. Hence, it is imperative to use both methods as

independent admission criteria of children enrollment in feeding program. Only about 20%

of children identified by either of the methods overlap with the other.

Prevalence of Chronic Malnutrition

Stunting indicates a failure to achieve one’s genetic potential for height. It usually reflects the

persistent, cumulative effects of poor nutrition and other deficits that often span across

several generation, which is caused by failure to receive adequate nutrition over a long period

of time and is also affected by recurrent and chronic illness. It is not sensitive to recent/short-

term changes in dietary intake.

Figure 10. Distribution of Height-for-Age in Z-score compared to WHO Standards (2006)


The table above shows that the distribution of Height-for-Age is shifted to the left when

compared to the Gaussian distribution. The mean of Height-for-Age in Z-score was -1.48 with

a Standard Deviation of 1.04. The SD should be between 0.8 and 1.2 to reflect data of weight

and height of good quality.

Stunting, or low Height-for-Age, generally occurs due to poor nutrition during pregnancy and

the first two years of life (the first 1000 days) and frequent infections. Its effects are largely

irreversible after 24 months of age. Stunting prevents children from reaching their full

physical and mental potential as it impairs brain development, which makes children less

capable in school and reduces productivity as they grow into adulthood.

Prevalence of stunting or chronic malnutrition is shown in Table 15. Nearly one third (30.6%)

of children were affected by stunting, and based on the WHO classification, the prevalence

is high as it exceeds 30%. Compared with children 0-60 months old in the NNS 2011, the

prevalence is close to the national prevalence of 33.6%.

Table 15. Prevalence of stunting or chronic malnutrition (global, moderate and severe) in

children 6-59 months of age by sex and by age group (HAZ, WHO 2006)

N

Stunting (HAZ <-2 Z-score)

Moderate Stunting (HAZ <-2 Z-score and >-3 Z-score)

Severe Stunting (HAZ<-3 Z-score)

% [95% CI]

(n)

% [95% CI]

(n)

% [95% CI]

(n)

All 628 30.6

[25.6-36.0] (192)

22.6 [18.7-27.1]

(142)

8.0 [5.8-10.8]

(50)

Boys 323 33.7

[27.6-40.5] (109)

26.6 [21.1-33.0]

(87)

7.1 [5.0-10.1]

(23)

Girls 305 27.2

[21.0-34.5] (83)

18.4 [13.8-24.0]

(56)

8.9 [5.8-13.2]

(27)

6-23 mo 198 27.3

[21.2-34.3] (54)

23.7 [18.0-30.7]

(47)

3.5 [1.6-7.7]

(7)

24-59 mo 431 32.3

[26.3-38.9] (139)

22.0 [17.6-27.3]

(95)

10.2 [7.3-14.1]

(44)


Figure 11. Stunting by region NNS 2011 and Stunting (Global, Moderate and Severe)

SMART Survey 2014 - (HAZ, WHO 2006)

Prevalence of Underweight

Weight-for-age is a composite index of height-for-age and weight-for-height. It takes into

account both acute and chronic malnutrition. While underweight or weight-for-age is used

for monitoring the Millennium Development Goals, it is no longer in use for monitoring

individual children as it cannot detect children who are stunted but of normal weight;

furthermore, it does not detect acute malnutrition that threatens children’s lives.

Figure 12. Distribution of Weight-for-Age in Z-score compared to WHO Standards (2006)


The above figure shows that the distribution of Weight-for-Age is shifted to the left when

compared to the Gaussian distribution. The mean of Weight-for-Age in Z-score was -1.17 with

a Standard Deviation of 1.02. The SD should be between 0.8 and 1.2 to reflect that the data

of weight and height is of good quality.

Prevalence of underweight is shown in Table 16.

Table 16. Prevalence of underweight (global, moderate and severe) in children 6-59 months

of age by sex and by age group (WAZ, WHO 2006)

N

Underweight (WAZ <-2 Z-score)

Moderate Underweight

(WAZ<-2 Z-score and >-3 Z-score)

Severe Underweight (WAZ <-3 Z-score)

% [95% CI]

(n)

% [95% CI]

(n)

% [95% CI]

(n)

All 632 20.7

[17.3-24.6] (131)

17.4 [14.3-21.0]

(110)

3.3 [2.1-5.3]

(21)

Boys 325 21.2

[16.4-27.0] (69)

18.5 [14.1-23.8]

(60)

2.8 [1.4-5.4]

(9)

Girls 307 20.2

[15.8-25.5] (62)

16.3 [12.6-20.9]

(50)

3.9 [2.1-7.1]

(12)

6-23 mo 198 17.2

[12.1-23.8] (34)

14.6 [10.0-21.0]

(29)

2.5 [1.1-5.7]

(5)

24-59 mo 434 22.4

[18.3-27.0] (97)

18.7 [15.0-23.0]

(81)

3.7 [2.4-5.7]

(16)

Figure 13 below shows the comparison of the prevalence of underweight of the Survey

population with the children 0-60 months old in the 2011 NNS. As shown, the prevalence is

higher than the national prevalence that is 20.2% but lower than the mean average of the 3

regions that is 23.7%.


Figure 13. Underweight by region NNS 2011 and Underweight (Global, Moderate and

Severe) SMART Survey 2014 - (WAZ, WHO 2006)

Regarding the prevalence of underweight, the level can be considered “high” by WHO cut-

offs for levels of public health significance (>20%).

3.4 Morbidity / Health Status

The survey found that of the 645 children, 4.3% were reported to have suffered from

diarrhea during the past 24 hours (Table 17), and 37.2%, or about 4 out of 10 surveyed

children, were reported to have suffered from the symptoms of acute respiratory infections

such as fever with whooping cough during the past 2 weeks (Table 18).

Table 17. Percentage of children 6-59 months with diarrhea during the past 24 hours

N

Diarrhea No Diarrhea Don’t know Total

n %

[95% CI] N % n % %

645 28 4.3

[2.3-6.4] 615 95.4 2 0.3 100.0

Table 18. Percentage of children 6-59 months with Acute Respiratory Infection (ARI) during

the past 2 weeks

N

ARI No ARI Don’t know Total

n %

[95% CI] N % n % %

645 240 37.2

[29.2-45.2] 404 62.6 1 0.2 100.0

5.3 4.2 5.6 3.3

23.921.6

25.7

20.7

0

5

10

15

20

25

30

Region VINNS 2011…

Region VIINNS 2011…

Region VIIINNS 2011…

SMARTFeb. - March 2014…

SevereUnderweight

ModerateUnderweight


3.5 Children Access to Programs/Services

Measles Vaccination

Immunization is an important public health intervention that protects children from illness

and disability. As part of the Expanded Program on Immunization (EPI), measles vaccination

is given to infants on the 9th month. A special mass campaign for measles was also done in

November 2013 in typhoon-affected areas. In the Survey, measles immunization was asked

among parents or caregivers of children 9-59 months old.

The results showed that based on vaccination card records, 47.7% of children were

vaccinated, either through the routine services or during the special massive measles

vaccination. On the other hand, 43.5% were reported vaccinated but without cards (Table

19). About 7% were not yet vaccinated and this group will be at risk to measles infection and

its consequences like undernutrition and deficiencies in vitamin A and zinc.

Table 19. Coverage of Measles vaccination in children 9-59 months of age

N

With

Card

Without

Card

Total Measles

Vaccination

No Measles

Vaccination DK

%

(n)

%

(n)

%

[95% CI]

(n)

%

(n)

%

(n)

606 47.7

(289)

43.5

(264)

91.2

[88.4-94.1]

(553)

6.8

(41)

2.0

(12)

Table 20 shows that 91.0% of children 9-59 months of age of 4Ps households were immunized

against measles. This result is very close to the coverage of 91.2% found for all children 9-59

months of age.

Table 20. Coverage of Measles vaccination in children 9-59 months of age, recipients of

the Four P’s

N

Total Measles

Vaccination

No Measles

Vaccination DK

%

[95% CI]

(n)

%

(n)

%

(n)

155

91.0

[84.5-97.4]

(141)

6.5

(10)

2.5

(4)


Vitamin A Supplementation

Provision of vitamin A supplementation every 6 months can help protect a child from death

and disease associated with vitamin A deficiency and is recognized as one of the most cost-

effective approaches to improve child survival. Improving the vitamin A status of deficient

children through supplementation enhances their resistance to disease and can significantly

reduce mortality. Vitamin A was distributed simultaneously with measles vaccinations last

November 2013. The proportion of all children aged 6-59 months who received vitamin A

(based solely on recall) in the last 6 months was 78.3% (Table 21). About 20.0% of the children

did not receive vitamin A supplement.

Table 21. Coverage of Vitamin A supplementation (VAS) during the last 6 months in children

6-59 months of age

N

With VAS Without VAS Don’t know

n %

[95% CI] n % N %

645 505 78.3

[73.3-83.3] 122 18.9 18 2.8

Table 22 shows that 84.2% (95% CI: 76.4-91.8) of children 6-59 months of age and who are

recipients of the 4P’s had received vitamin A in the last 6 months, which is higher than the

coverage of 78.3% found for all children 6-59 months of age.

Table 22. Coverage of Vitamin A supplementation (VAS) during the last 6 months in children 6-59 months of age, recipients of the 4P’s

N

With VAS Without VAS Don’t know

n %

[95% CI] n % N %

164 138 84.2

[76.4-91.8] 25 15.2 1 0.6

Deworming

Helminths or intestinal worms represent a serious public health problem in areas where

climate is tropical and inadequate sanitation and unhygienic conditions prevail. Helminths

cause significant malabsorption of vitamin A and aggravate malnutrition and anemia, which

eventually contributes to retarded growth and poor performance in school. Children under

five years old are extremely vulnerable to the deficiencies induced by worm infestations. This

puts deworming as critical for the reduction of child morbidity and mortality.


For these reasons, deworming is recommended for children 12-59 months old as children in

this age group are considered to be susceptible to disease. Deworming also helps enhance

the iron status of children which eventually helps children to exercise their intellectual ability

to the fullest. Deworming is only a short-term measure of reducing worm infestation and re-

infestation is frequent. Control measures through improved sanitation, hygiene and

deworming are needed to prevent infestation and re-infestation.

In the country, deworming of children 1-5 years old is conducted simultaneously with vitamin

A supplementation and measles vaccination in April and October 2013 during the

Garantisadong Pambata (GP) week. This program is a week long package of health services

for children 6-59 months old that aims to reduce morbidity and mortality. Provision of

deworming drugs was also done in November 2013 post-Haiyan, along with vitamin A

supplements and measles vaccination.

Questions on deworming were asked from mothers/caregivers of children aged 12-59

months. Table 23 shows that the proportion of children who received deworming in the last

6 months was 54.7% (95% CI: 48.1-61.3).

Table 23. Coverage of Deworming during the last 6 months in children 12-59 months of age

N

Deworming No Deworming Don’t know

n %

[95% CI] n % n %

574 314 54.7

[48.1-61.3] 249 43.4 11 1.9

The coverage of deworming is slightly higher for children who are recipients of the 4Ps with

a coverage of 60.4% (95% CI: 49.8-71.0).

Table 24. Coverage of Deworming during the last 6 months in children 12-59 months of age,

recipients of the 4Ps

N

Deworming No Deworming Don’t know

n %

[95% CI] n % N %

144 87 60.4

[49.8-71.0] 54 37.5 3 2.1


MUAC Screening Coverage

The survey found that 46.5% (95% CI: 37.5-55.5) of children aged 6-59 months have been

screened with MUAC measure since typhoon Haiyan (Table 25), and more than half were not

yet reached (Table 28). The high proportion of children who were not yet screened is rather

unexpected because the Survey was done 3 months after Typhoon Haiyan. In the PET (DOH,

2012) guidelines, after an emergency, like Haiyan, screening should be done to identify and

treat children who are severely malnourished (DOH, 2012 PET). In the guidelines, measuring

MUAC in children is recommended in the absence of the weighing scales and height board.

Table 25. Coverage of Screening with MUAC measure since Haiyan in children 6-59 months

of age

N

MUAC taken on child since

typhoon No Screening Don’t know

N %

[95% CI] n % n %

645 300 46.5

[37.5-55.5] 342 53.0 3 0.5

Maternal, Infant and Young Child Nutrition In 2012 the World Health Assembly (WHA 65.6) endorsed the Comprehensive Implementation Plan (CIP) for Maternal, Infant and Young Child Nutrition that calls for the achievement of six (6) global targets by the year 2025: 1. 40% reduction in children under five that are stunted6

2. 50% reduction in anaemia in women of reproductive age1

3. 30% reduction of low birth weight1

4. 0% increase in childhood overweight1

5. Increase in the rate of exclusive breastfeeding in the first 6 months to at least 50%1

6. Reduce and maintain childhood wasting to less than 5%1

6 Global Nutrition Targets as stated in the Comprehensive implementation plan for Maternal, Infant and Young Child Feeding.


3.6 Women Nutritional Status The second of the six targets is to reduce by 50% anemia in women of reproductive age. This

has been reaffirmed in the Regional Committee for the Western Pacific Meeting (2012),

where Member States adopted Resolution WPR/RC63.R2 on “Scaling up Nutrition in the

Western Pacific”, of which the Philippines is a member state.

Adequate nutrition is especially critical for women because inadequate nutrition causes

damage not only to women’s own health but also to their children and the development of

the next generation. MUAC was used to assess nutritional status of women of reproductive

age (15-49 years old). Of the 1292 women, 1077 non-pregnant and 70 pregnant women were

measured for MUAC.

Mid-Upper Arm Circumference (MUAC) Measurement for Women

Women are vulnerable to malnutrition because of their high nutritional requirements during

pregnancy and lactation.

MUAC was measured of a total of 1150 women 15 to 49 years of age; only 42 (3.6%) were found having a MUAC less than 210mm (Table 26). Table 26. Result of MUAC measurement for women of reproductive age (15-49 years old) with <210 mm

N n

Low MUAC (MUAC <210 mm)

% [95% CI]

All (15-49 years)

1150 42 3.6

[2.5-4.8]

Of the 70 of pregnant women of whose MUAC was measured, only one (1) was identified having a MUAC measurement less than 210 mm.

Figure 14. Number of women with MUAC Measurement less than 210 mm.

1

41

0

10

20

30

40

50

MUAC less 210 mm

Pregnant

Non_Pregnant


About 42% of women found with MUAC measurement below 210 mm, were adolescents

suggesting the importance for a strong maternal nutrition program and adolescent health

program.

Figure 15. Malnutrition among women of reproductive age

30% of the women with MUAC less than 210 mm had a child less than 24 months at the time

of the interview; lactating mothers require additional energy intake to prevent them from

becoming malnourished.

Prenatal Care

Many health problems in pregnant women can be prevented, detected and treated during

prenatal care visits with trained health workers. WHO recommends a minimum of four

prenatal visits, comprising of interventions such as tetanus toxoid vaccination, screening and

treatment for infections, and identification of warning signs during pregnancy.

Globally, during the period 2005–2012, over 50% of women received the recommended

minimum prenatal care. The Philippines adopted the global recommendations and 73%(95%

CI: 63%-83%) of pregnant women that were interviewed in the survey, had at least one

prenatal visit; while 23% did not or have not yet started their prenatal visits.

Iron and/or Folic Acid Supplementation

It is estimated that almost half of all pregnant women and one third of non-pregnant women

worldwide have anemia, a condition that significantly increases the risks to health for both

mothers and infants (World Health Report 2009).

Iron deficiency anemia is ranked globally as the third leading cause of disability adjusted life

years (DALYS) for women aged 15-44 years. From 1995 to 2011, the global prevalence of

anemia decreased from 33% (29–37) to 29% (24–35) in non-pregnant women, from 43% (39–

47) to 38% (34–43) in pregnant women, and from 47% (43–51) to 43% (38–47) in children.

42.5

15

32.5

10

0

20

40

60

80

100

MUAC less than 21 cm (%)

15 to 19 years

20-23 years

24 to 35 years

36 and above


These prevalence rates translate to 496 million non-pregnant women, 32 million pregnant

women, and 273 million children with anemia in 20117.

Iron deficiency is thought to be the most common cause of anaemia globally (50-60% of the

cases)8. Seventy five of the 78 pregnant women (96%), provided information on their current

use of Iron Folic Acid (IFA), which was: 50 or 66.6% (95% CI: 55.5-78) taking iron folic acid, 23

(30.6%) not taking IFA. This represents a steep increase from the 14% that were not taking

IFA in 2008 (Regional average of DHS 2008).

Table 27. Pregnant women consuming Iron and/or Folic Acid

N n

Pregnant Women Consuming IFA

%

[95% CI]

Pregnant 75 50 66.6

[55-78]

Figure 16. Iron and/or Folic Acid intake among pregnant women

3.7 Infant and Young Child Feeding Practices

Globally, suboptimal breastfeeding is causing 804,000 under-five deaths per year.9 More than

30 studies from around the world, in the developing and developed countries alike, have

7Stevens GA, et al. Global, regional, and national trends in haemoglobin concentration and prevalence of total and severe anaemia in children and pregnant and non-pregnant women for 1995—2011: a systematic analysis of population-representative data. Lancet Global Health 2013;1:e16-25

8Stevens GA, et al. Global, regional, and national trends in haemoglobin concentration and prevalence of total and severe anaemia in children and pregnant and non-pregnant women for 1995—2011: a systematic analysis of population-representative data. Lancet Global Health 2013;1:e16-25 9 The Lancet Maternal and Child Nutrition Series, June 2013.

66.6

30.3

3.1

0

20

40

60

80

100

Pregnant women taking IFA (%)

Yes

No

DK


shown that optimal breastfeeding dramatically reduces the risk of infants and young children

dying.10

The World Health Assembly and the Executive Board of the United Nations Children’s Fund

(UNICEF) endorsed The Global Strategy for Infant and Young Child Feeding in 2002. The global

IYCF strategy provides governments and other stakeholders with strategies and key

components to improve IYCF practices. These need to be adapted and contextualized to the

situation, conditions and cultural norms at the country level.

IYCF recommendations in the Philippines are aligned with the Global Strategy for Infant and

Young Child Feeding and include initiation of breastfeeding within the first hour of life,

exclusive breastfeeding for six months, and provision of appropriate, adequate and safe

complementary food at six months while continuing breastfeeding until two years and

beyond.

Overwhelming evidence shows that virtually all mothers can breastfeed. Lactation failure is

virtually unknown in societies where breastfeeding is highly valued, regarded as a natural

physiological function which is the only way to nourish an infant. In these circumstances,

families and societies strongly encourage and support breastfeeding. Women in these

societies are also less exposed to environments that undermine lactation. Studies show that

when mothers are given proper information and support, they successfully breastfeed.

A total of 265 children 0-23 months were included in the survey, 73 (27.5%) of which were

under the age of 6 months and 192 (72.5%) were 6 to 23 months old.

Initiation of breastfeeding in the first hour would prevent an additional 22% of newborn

deaths4. The survey revealed that 58% of children 0-23 months were initiated to breastfeeding

within 1 hour, while 26 were initiated after 1 hour (Table 28). This result is below the 66%

average early initiation rate for Region VI, VII and VIII recorded in the NNS 2011.

Table 28. Early Initiation of Breastfeeding (children 0-23 months old)

N

Proportion of children born in the last 24 months who were put to the breast within one hour of birth

n %

[95% CI]

265 155 58

[52-64]

10Edmond KM, Zandoh C, Guigley MA, Amgena-Etego S, Owusu-Agyei S, Kirkwood BR. Delayed breastfeeding initiaiton increases risk of neonatal mortality. Pediatrics 2006;177:380:386


Figure 17. Timing of initiation of breastfeeding in Haiyan affected areas

It is alarming that 14% of infants were never breastfed in the Haiyan affected areas, compared

with the 4% regional average recorded in the NNS 2011.

More than eight hundred thousand deaths are attributed to sub-optimal breastfeeding

annually worldwide. Exclusive breastfeeding for the first six months and continued

breastfeeding would prevent 13% of under-five deaths, primarily from infections resulting to

diarrhea, pneumonia and neonatal sepsis.

While it is not possible to provide conclusive information on the infant feeding practices

below six months of age because of the small sample size for this age group, the SMART survey

suggests that less than 50% of infants less than 6 months of age were exclusively breastfed in

the Haiyan affected areas. This is below the average of the 3 regions which is 57% reported in

NNS 2011.

At the same time, about 41% of infants less than 6 months of age were given infant formula

the day before the interview (95% CI: 31-53). This is higher than the 36% national average

recorded in DHS 2008.

Figure 18. Infant Formula Usage before and after the typhoon Haiyan

58

197

142

0

20

40

60

80

100

Initiation of Breastfeeding (%)

within 1st hour

1 to 23 hours

After 1 day

Never started BF

DK

3641

0

10

20

30

40

50

Infant Formula Usage (%) less than 6 months of age

DHS 2008

Smart 2014


About 46% of children 0-23 months were bottle-fed (Table 29), a substantial increase from

the 39% average of the 3 regions recorded in NNS 2011.

Table 29. Bottle feeding among children 0-23 months

N Proportion of children 0-23 months of age who are fed with a bottle

n %

[95% CI]

265 122 46%

[40-52]

Optimal complementary feeding practices are often discussed in terms of the most frequently

used indicators which focus on continued breastfeeding and timely introduction, frequency,

and variety of foods consumed. These do comprise the primary components of appropriate

complementary feeding, but secondary components of complementary feeding that are often

overlooked are responsive feeding and hygiene related to food preparation, handling, and

feeding.

It is important to emphasize that continued breastfeeding is part of optimal complementary

feeding. Breastfeeding should occur frequently and on demand until at least two years of age.

Around the age of six months, an infant’s need for energy and nutrients starts to exceed what

is provided by breast milk, and complementary foods are necessary to meet those needs. If

complementary foods are not introduced when a child has reached six months, or if they are

given inappropriate complementary foods, an infant’s growth may falter. Breastfed children

at 12–23 months of age receive on average 35% to 40% of total energy needs from breast

milk with the remaining 60% to 65% covered by complementary foods. Growth faltering is

most evident during this period, particularly between 6 and 12 months when foods of low

nutrient density begin to replace breast milk and rates of diarrhoeal illness caused by e.g.

food contamination are at their highest.

The SMART survey findings show that of the 40 children 6 to 8 months of age, only 36 (90%)

(Figure 19) were consuming solid/semi-solid and soft foods at the time of the survey (95% CI:

80%-100%) higher than the regional average of 83.7% (NNS, 2011); 30 (83.2%), of the 36 were

still breastfeeding while consuming solid/semi-solid foods. In addition, 66% of the children 6-

23 months were consuming iron rich foods (95% CI: 60-73), lower than the 74% national

average (DHS, 2008).


Figure 19. Complementary feeding practices

Unfortunately, in this survey, no data are available for specific complementary feeding

practices, namely: time of introduction, diversity and frequency. The GAM rate of 6.0%

recorded in 200 children 6-23 months of age being higher than the 4.1% overall GAM rate, is

suggesting that breastfeeding and complementary feeding practices after the age of six

months are far from optimal and appropriate.

4. Conclusion and Recommendations

According to the WHO classification, the results of the survey showed a level of Global Acute

Malnutrition considered "acceptable", not exceeding the 5% threshold. The prevalence of

GAM is lower than the 2011 NNS levels (5.8%, 5.3% and 7.8% for region 6, 7 and 8

respectively). Prevalence of GAM using Weight-for-Height in Z-score (WHZ) was 4.1% whereas

using MUAC, GAM was 0.8%.

The large discrepancy between these two measures merit further study. WHZ is considered

to be the base measure for Global Acute Malnutrition, but it should be clearly noted that it is

not by any means a “gold standard” measure for acute malnutrition. Following the release of

WHO Child Growth Standards, the relationship between Weight-for-Height and the risk of

dying was reassessed in existing epidemiological studies. The analysis showed that children

with a Weight-for-Height below -3 Z-score based on WHO growth standards have a high risk

of death exceeding 9-fold that of children with a Weight-for-Height above -1 Z-score. Similar

studies using MUAC as diagnostic criterion showed that the risk of dying is increased below

115mm. MUAC is an important measure of acute malnutrition that has a much closer relation

to infant and child mortality than WHZ. Hence it is imperative to use both methods as

90

66

0

20

40

60

80

100

6 to 8 months old Introduced tosold/semisolid foods (%)

6 to 23 months old consuming IronRich Foods (%)

Complementary Feeding Practices


independent criteria for enrollment of children in feeding programs. Only about 20% of

children identified by either of the methods overlaps with the other.

These results have significant programmatic implications as follows:

1. There is a need to recalibrate the targets for acute malnutrition in line with the findings;

2. The implementation of CMAM needs to be targeted to areas with high acute malnutrition,

where the situation is of concern (10% GAM and above); and,

3. There is an urgent need to identify and address bottlenecks in the implementation of

Infant and Young Child Feeding interventions.

Based on the WHO classification, the survey results, though lower than the 2011 NNS levels

(41%, 38.6% and 38.8% for Regions VI, VII and VIII respectively), show a level of chronic

malnutrition considered "high", exceeding the 30% threshold of WHO for public health

significance. This reflects the existence of long term undernutrition and highlights the need

to prioritize stunting prevention interventions. Programming for stunting prevention

interventions will require a more comprehensive and long-term approach (outside the

emergency context). It has been estimated that the prevalence of chronic malnutrition can

be reduced by about a third if effective large scale interventions are implemented (2008

Lancet series on Maternal and Child Undernutrition).

The most effective interventions in preventing stunting occur during the window of

opportunity, from the time of pregnancy until the end of the first two years of life of the child.

To achieve them, the Nutrition Cluster should aspire to contribute to efforts to:

1. Invest in the establishment of community, health and nutrition system workplaces and

public places for promoting, supporting and protecting exclusive breastfeeding for the

first six months of life and continued breastfeeding up to two years of age and beyond;

2. Support community-based programs to provide information and counseling on optimal

and appropriate complementary feeding practices;

3. Link with livelihood, food security and social welfare clusters and programs to ensure

increased access by vulnerable families to appropriate and safe diets;

4. Reduce infections by educating households on proper care and hygiene practices and

improving health seeking behavior for management of children’s infections;

5. Educate pregnant women about the importance of prenatal care and protect maternal

nutrition and health to prevent low birth weight babies;

6. Promote regular growth monitoring and include measurement of length/height (not just

weight) in nutrition programs;

7. Invest in a mass communication campaign for development based on preventive

activities: nutrition of pregnant women, promotion of exclusive breastfeeding,


complementary feeding and continued breastfeeding, good hygienic practices, the

production and consumption of available complementary foods; and

8. Continue distribution of micronutrient powders (MNPs) for children 6-23 months old after

the end of the emergency operation.

Regarding the prevalence of underweight, the level is still considered “high” by WHO standard

for public health significance (>20%). Investments should be made to allow immediate

procurement of equipment to measure all children length/height for timely and preventive

nutrition interventions

Efforts should be strengthened to improve coverage of vitamin A supplementation and

deworming (80% target) to reach and address key concerns including those on:

1. Raising awareness of mothers on micronutrient supplementation and deworming

campaigns;

2. Strengthening distribution channels of vitamin A and deworming supplies and monitoring

and evaluation of campaigns;

3. Planning the achievement of mass activities around supplementation and deworming at

least twice a year, through Garantisadong Pambata weeks (April and October).

The survey results suggest that breastfeeding practices are generally suboptimal and

inappropriate. While not enough data are available to be able to determine complementary

feeding practices, limited evidence suggests that infant formula and bottle feeding practices

continue to challenge the country’s promotion of exclusive breastfeeding. The government

should continue its vigilance in capacitating workers at the local level to step up support to

breastfeeding mothers and improve the quality of breastfeeding support systems.

Major efforts have been exerted prior to typhoon Haiyan, and renewed during the acute

phase of the emergency.

The findings of the survey call for all government and non-governmental partners to carefully

review the on-going efforts aimed at the promotion, protection and support of optimal and

appropriate infant and young child feeding practices. Among the immediate actions necessary

to reverse the negative trends, the following recommendations are made:

1. Organize IYCF community support systems through deployment of trained local health

and nutrition volunteers, NGOs, other civil society organizations;

a. design and implement an incentive scheme that would ensure sustainability and

quality of services


2. Ensure that high quality IYCF counseling and support services are integrated in the health

care delivery system;

a. Ensure supportive supervision, monitoring, documentation, tracking, dialogues and

frequent reviews;

3. Ensure that all hospitals and health facilities assisting deliveries comply with the Mother

Baby Friendly Hospital Initiative and provide IYCF counseling services;

a. Monitor and provide supportive supervision and technical assistance to hospitals and

health facilities, to ensure compliance and full implementation;

4. Integrate IYCF counseling and monitoring in the routine public health programs (prenatal,

EPI, post-partum) both health facility based and community efforts;

5. Strengthen the enforcement and accountability mechanisms for key legislations like the

Milk Code (EO51), the Enhanced Breastfeeding Promotion Act of 2009 (RA 10028) and the

Rooming-in Act, (RA 7600);

6. Develop and invest in a massive communication/for behavioral change campaign based

on the findings of a formative research.

a. Conduct a formative research to identify and surface gaps, issues, barriers that hinder

the success of IYCF interventions and activities

7. Monitor and track the progress being made in reaching pregnant women and mothers of

children 0-23 with IYCF counseling services and their improvement;

Finally, in order to monitor the effect of present and future interventions on trends of

malnutrition, it is recommended that a follow-up SMART survey be implemented in

September 2014 following the same methodology as the present investigation.


ANNEXES

Annex 1: List of Surveyed Barangays

CLUSTER # REGION PROVINCE MUNICIPALITY/CITY BARANGAY TEAM

LEADER

1 VI Aklan Balete Arcangel

Cory

2 VI Aklan Batan Palay

May

3 VI Aklan Kalibo New Buswang

Jel

4 VI Capiz Dao Centro

May

5 VI Capiz Dumalag Santa Cruz

Derich

6 VI Capiz Ivisan Santa Cruz

Jel

7 VI Capiz Jamindan Agbun-Od

Cory

8 VI Capiz Mambusao Pangpang-Sur

Jel

9 VI Capiz Panay Agbalo

Cory

10 VI Capiz Panitan Conciencia

Derich

11 VI Capiz Pontevedra Ilaya (Poblacion)

Cory

12 VI Capiz Pres. Roxas Quiajo

Jel

13 VI Capiz Roxas City Dayao

May

14 VI Capiz Roxas City Mongpong

Derich

15 VI Capiz Sapian Maninang

Derich

16 VI Capiz Sigma Mangoso

May

17 VI Iloilo Ajuy Pili

Jel

18 VI Iloilo Banate Carmelo

Derich

19 VI Iloilo Barotac Viejo Nueva Sevilla

Cory

20 VI Iloilo Bngawan Alabidhan

Cory

21 VI Iloilo Concepcion Plandico

May

22 VI Iloilo Estancia Poblacion Zone Ii

Derich



LEADER

23 VI Iloilo Lemery San Jose Moto

Jel

24 VI Iloilo Passi City Tagubang

Derich

25 VI Iloilo San Dionisio Agdalinan

Cory

26 VI Antique Barbaza Poblacion

May

27 VI Antique Culasi Caritan

Derich

28 VI Antique Pandan Bagumbayan

Cory

29 VII Cebu Bantayan Atop-Atop

May

30 VII Cebu Madridejos Kangwayan

Jel

31 VIII Leyte

Tacloban City 48 Derich

32 VIII Leyte

Tacloban City 71 Cory

33 VIII Leyte

Tacloban City 88 May

34 VIII Leyte

Tacloban City 109 Jel

35 VIII Leyte

Ormoc City Bgy.1 Poblacion Jel

36 VIII Leyte

Ormoc City Punta Derich

37 VIII Leyte

Ormoc City Cabingtan May

38 VIII Leyte

Ormoc City Juanton Cory

39 VIII Leyte

Merida Poblacion Derich

40 VIII Leyte

Dulag Cabacungan Cory

41 VIII Leyte

Palo San Joaquin Jel

42 VIII Leyte

Palo Campetik May

43 VIII Leyte

Abuyog Nebga Cory

44 VIII Leyte

Alangalang Milagrosa Jel

45 VIII Leyte

Kananga Lim-Ao Derich

46 VIII Leyte

Carigara Canal May



LEADER

47 VIII Leyte

Tanauan Cabuynan Derich

48 VIII Leyte

Albuera Poblacion Cory

49 VIII Eastern

Samar

Guiuan (Including Homonhom) Bgy. 9a

Jel

50 VIII Eastern

Samar Oras

Dao Cory

51 VIII Eastern

Samar Can-Avid

Mabuhay May

52 VIII Eastern

Samar Gen. Mc Arthur

Poblacion Bgy 5 Jel

53 VIII Eastern

Samar Taft

Mantang Derich

54 VIII Western

Samar Marabut

Santa Rita May

55 VI Capiz Cuartero

Balingasag Jel

56 VI Iloilo Sara

Ardemil Jel

57 VIII Leyte Tacloban City

65 May

58 VIII Leyte Tacloban City

94 May

59 VIII Leyte Palompon

Mazawalo Cory

60 VII Cebu Daanbantayan

Tapilon Derich


Annex 2: SMART Survey Questionnares


Annex 3: Data Dictionary

DATA DICTIONARY FOR ANTHROPOMETRY: CHILDREN 6-59 MONTHS OF AGE + WOMEN 15-49 YEARS OF AGE IDENTIFIERS

Header Variable

name Question Definition Special Instructions

SURVDATE Complete date of Interview written in (DD/MM/YYYY)

Record this information automatically, DO NOT ASK MOTHER.

ALWAYS FILLED IN AT THE TOP OF EACH QUESTIONNAIRE

CLUSTER Cluster Number (range 01-60)

Record this information automatically, DO NOT ASK MOTHER.


TEAM Team Number (range 1-4) Record this information automatically, DO NOT ASK MOTHER.


HH Household Number- determined after the random selection of households in the barangay

Household is defined as group of people living together, related or not, and sharing the same cooking pot. Record this information automatically, DO NOT ASK MOTHER.

ALWAYS FILLED IN AT THE TOP OF EACH QUESTIONNAIRE- Refer to List of Household Number.

SECTION G: FILL IN FOR ALL HOUSEHOLDS- EVEN THOSE WITHOUT CHILDREN Question number

Variable name

Question Definition Special Instructions

G0 Consent Consent given by the head of household/caregiver

Household head is a member of household that is regarded as such in the household, usually the father of children in the household. If HH head is out, we can ask the main caregiver of the children (if any).

An individual will be marked as ‘absent’ only after at least two re-visits to the household have been made. This column is to ensure that consent is asked and obtained; and that absent individuals are recorded and followed-up on.


Question number

Variable name


Valid values are: 1=Yes 2=No 3=Absent

G1 DISPLACE Since Yolanda, have you ever been displaced?

Displacement is defined as a situation where HH has moved/transferred from original housing unit to another place (evacuation centers, etc.) because of Yolanda. Valid values are:

1=Yes 2=No 8=Don’t know

Ask question even if household has no children or women. Important that respondent is only referring to displacement since the Typhoon Yolanda (November 8th, 2013) and not other typhoons. If answer is “yes”, ask questions G1A (time of displacement) and G2 (type of displacement).

G1A TIMEDISPL If YES, how recent was the displacement:

Valid values are: 1=Before the New Year 2=Since the New Year

Make sure respondent understands that it is only with regards to Typhoon Yolanda. ONLY if Question G1 received a “1=Yes”

G2 TYPEDISPL If YES, where are you currently living?

Valid values are: 1= Original Home 2= Evacuation Centers 3= Transitional Shelters 4 = Spontaneous settlements 5 = Host Family or Relative

Make sure respondent understands that it is only with regards to Typhoon Yolanda. ONLY if Question G1 received a “1=Yes” answer Evacuation centres (ECs): Pre-existing buildings established to accommodate displaced families since the onset of a disaster, e.g. schools, covered courts, gymnasiums, barangay halls, health centres and private buildings. Transitional sites: Sites established to temporarily host families facing dis-placement for more than a month and typically awaiting permanent relocation. Families are usually transferred from evacuation


Question number

Variable name


centres to these sites. Transitional sites host families in tents, bunkhouses or alternative transitional sites. Spontaneous settlements: Displaced family or families who live collectively outside of government designated evacuation centres or transitional sites. These families normally stay in open spaces in makeshift shelters on the roadside or near their homes and communities (CCCM 2013). Host family= living with a non-related household or relative

G3 FOURPS Are you a recipient of the Four P’s?

Valid values are: 1=Yes 2=No 8=Don’t know

Ask question even if household has no children or women.

SECTION C: ANTHROPOMETRY OF ALL CHILDREN AGED 6-59 MONTHS

Question number

Variable name


C1 ID Child number in the household

As many eligible children there are in the surveyed household

List all children 6-59 months in the household To be entered into ENA Software before leaving the Cluster.

C2 HH Household number

Copy number that was put in the identifier. DO NOT ASK MOTHER.

To be entered into ENA Software before leaving the Cluster.

C3 SEX Sex of child Valid values are: ‘f’ for female ‘m’ for male



Question number

Variable name


C4 BIRTHDAT Exact birth date of child (DD/MM/YYYY)

Valid values are: 59 months before the exact survey date up until 6 months prior to the first day of the survey. Children born before February 2009 will be excluded, and those born before August 2013 are included.

It is only recorded if an official age documentation is available or mother says birthdate without hesitation. The date cannot be after survey date. Leave blank if no exact age given. To be entered into ENA Software before leaving the Cluster.

C5 AGE (in months)

Age of child in completed months

Valid values: Ranges from 6 to 59.99 months (automatically created if birthdate is given)

Skip if birthdate is given. To be entered into ENA Software before leaving the Cluster.

C6 WEIGHT Weight of child in kg

Valid values: Ranges from 3 to 31 kg

To be entered into ENA Software before leaving the Cluster. If any entered value is outside of the set range, the value will turn pink in the Data Entry Anthropometry screen of ENA Software. Should double-check measurement on questionnaire or retake the weight of that child.

C7 HEIGHT Height/length of child in cm

Valid values: Ranges from 54 to 124 cm

To be entered into ENA Software before leaving the Cluster. If any entered value is outside of the set range, the value will turn pink in the Data Entry Anthropometry screen of ENA Software. Should double-check measurement on questionnaire or retake the height/length of that child.

C8 EDEMA Presence of bilateral oedema

Valid values are: ‘y’ for yes ‘n’ for no


C9 MUAC MUAC of child in mm

Valid values are: Ranges from 75 to 230 mm

To be entered into ENA Software before leaving the Cluster. ENA can only analyse MUAC values entered in mm.

C10 MEASURE If child’s height is measured opposite to protocol

L = if child is measured lying down instead of standing up H = if child is measured standing up instead of lying down

ONLY fill in if a child is measured opposite to protocol. To be entered into ENA Software before leaving the Cluster.


Question number

Variable name


C11 MEASLES Measles vaccination received

Valid values are: 1=Yes with card 2=Yes without card 3=No 5= Not Applicable 8=Don’t know

Ask this question to all children 9-59 months old. Make sure not to confuse with other vaccines.

C12 VITA Vitamin A supplementation received in the past six months


A vitamin A capsule should be shown to the caregiver to help in the recall if there is no documentation available. The same capsules as are used locally should be shown.

C13 DIARRHEA Diarrhea over 24 hours

Diarrhea is defined as 3 loose stools in a day. Valid values are:


Make sure to use the proper case definition of 3 or more loose or watery stools in a 24-hour period.

C14 ARI ARI in the past two weeks

Ask if child had fever, deep cough, with running nose. Valid values are:


Make sure to use the proper case definition of ARI, i.e., fever and coughing with sneezing. Any other symptoms given by the caregiver do not correspond to ARI.

C15 DEWORMING Deworming in the past 6 months

Valid values are: 1=Yes 2=No 5=Not Applicable 8=Don’t know

ONLY for children aged 1 year of older.


Question number

Variable name


C16 BNSVISIT MUAC taken since the typhoon


Make sure that the mother understands the type of measurement we are referring to (MUAC ONLY). It can be taken either by a BNS or a NGO staff member.

C17 ENROL Child enrolled in feeding programme

Valid values are: 1=Plumpy Doz 2=Health Center (OTP) 3=Hospital (ITP/SC) 4=None 8= Don’t know

Make sure that the mother understands the different types of programs by showing her the sample of Plumpy Doz.

C18 MNP Child is currently consuming MNP

Valid values are: 1=Yes 2=No

8=Don’t know

Make sure that the mother understands the use of MNP by showing her the sachet.

SECTION M: WOMEN AGED 15-49 YEARS WITH CHILD/CHILDREN UNDER 24 MONTHS, OR IS PREGNANT Question number

Variable name


M1 Woman ID Woman number in the household

As many eligible women there are in the surveyed household

Make sure woman is aged between 15-49 years old.

M2 Marital Status

What is your current marital status?

If living together without marriage granted by the church or legally, consider living together Valid values are:

1 = Single/never married. 2 = Currently married.

Make sure you enter the current value based on the woman’s answer.


Question number

Variable name


3 = Living together. 4 = Separated/divorced/annulled. 5 = Widowed.

M3 YEARS Age of woman Valid values are: 15 – 49 years.

Do not include any woman younger than 15 years old nor older than 49 years in the questionnaire.

M4 MUAC MUAC of woman Binary answer- valid value is: 1 = 210 mm or more

If MUAC is less than 210 mm, write the value in mm (000)

Take the measurement on the left arm and it is in mm. Due to the tools available for the survey, only these two options are available.

M5 PREG Pregnancy Status of Woman

Valid values are: 1= Yes 2= No 8=Don’t know

Ask only if woman has a child under 24 months or is pregnant If 2=NO, skip to M6.

M5A PRENATAL Prenatal care during pregnancy


Ask only if woman has a child under 24 months or is pregnant Answer only if M5 is 1=Yes.

M5B VISIT No. of visits Valid values are: 1= 1 visit 2= 2 visit 3=3 or more visits 8=Don’t know

Ask only if woman is having prenatal care during pregnancy. Answer only if M5A is 1=Yes.

M6 IRON Iron/folic acid supplements


Ask only if woman has a child under 24 months or is pregnant Show the capsule to the woman.


Annex 4: Lessons Learned and Recommendations

Stage Division Bottleneck Action Taken Recommendation

Hiring

Human

Resource

Late Hiring of Survey

Manager

Hiring process of key positions

(Survey Manager, Field

Supervisor, Administrator,

Logistics) should be done at

least 1 month before the

implementation of program

(Orientation, Training, Data

Collection)

NNC Regions VI, VII and VIII

were requested by ACF to

submit 3 CVs for

enumerators; did not

receive any feedback from

ACF.

It was explained to the NNC

Regional Nutrition Program

Coordinators (RNPC) the

possibility of ACF hiring of

enumerators.

Hiring process of enumerators

should be taken care of by one

focal person to avoid

communication gap.

Planning

Technical Insufficient time during the

planning phase (protocol,

sampling, logistics, human

resource)

Planning (survey protocol,

budget, sampling, human

resource, logistics) should be

done at least 2 weeks prior to

the data collection. Summary of

survey protocol including all key

details for implementation

should be finalized before the

training.


Planning

Technical: Sampling

Dummy tables should be prepared ahead of sampling to make sure that sample size help provide representative information

None Stratified sampling per indicator

Planning

Technical: Development and finalization of survey questionnaire

Not all questions were consistent with the national/international recommended questions Missing important questions on complementary feeding.

Discussed, suggested. For the next round ensure that all questions are consistent with pre-tested questions used in national surveys to allow comparison

Planning

Teams/HR Limited no. of surveyors Discussed and suggested to enlarge survey teams

Discuss the possibility for the next rounds to add surveyors to the existing teams. Explore the possibility to hire FNRI/NSO surveyors that have experience with NNS and DHS

Training

Technical One team leader was not

trained on the plausibility

check report.

Mentoring in lieu of training

was done during

supervision visit.

Prior to the training, team

leaders should be

identified/hired in order to

include him/her on the

orientation of use of the ENA

software.

Technical Short duration of the

training and insufficient

time for discussion of the

One day of the training must be

dedicated to the ENA software.

It must cover data entry, back


ENA software for all team

leaders

up, analysis of the plausibility

report and flagged data.

Training should be longer to

have enough time to discuss

other important

protocols/standards such as

random sampling (systematic

and simple) and the use of

different survey tools. This will

standardized understanding in

all of the teams.

Coordination

Technical Late sending of

communications to local

chief executives and

barangay captains

Communications were

done through phone and

with the help from NNC

regional office.

Letters should be sent to NNC

regional office at least two

weeks prior to the data

collection to give them time to

disseminate the letter to

respective local government

units. Develop and implement a

communication plan to inform

mayors/barangay captains of

selected areas about the date of

data collection and to request

households with children under

five years of age be available on

date of survey.


Data Collection/Implementation

Logistics Backpacks provided were

worn-out during first few

days of data collection

Called the attention of

logistics

Good quality backpacks should

be purchased for the next

survey.

Security risk in assigned

areas

Teams consulted the

Provincial Nutrition Action

Officer regarding security in

survey areas. Teams made

adjustments in survey

schedule and travel.

There should be an ocular visit

by the logistics on "critical"

areas prior to the data

collection. It is possible to

cancel one cluster or more due

to security issues. At least 25 to

26 clusters and 80% of the

sample size should be reached.

In addition, it is also possible to

visit the reserved clusters.

During the planning stage, areas

with access difficulties and with

security issues should be

identified to make necessary

adjustments in schedules and

itineraries.


Logistics

Vehicle cannot reach

mountainous areas with

rough roads

Team requested the

logistics to change vehicle.

Sports Utility Vehicle (SUV)

should be used by teams during

the data collection for the next

survey.

Qualifications and behaviour

(impolite) of the driver for

safety of the team.

Change of driver due to

misbehaviour/misconduct

Drivers should be background

checked and oriented relative to

their work. Drivers should have


good driving skills and

experience.

The team's vehicle had flat

tires in critical area.

The team had opted to

quickly look for vulcanizing

shop while waiting for

another vehicle to fetch the

team.

The driver and the team should

check the running condition of

the vehicle before leaving the

base.

Administrative Surveyors lack time to

encash cheque (salary) since

data collection was from

Mondays to Saturdays from

8:00-5:00 pm.

Teams allotted time for

enumerators to encash

cheque during weekdays

despite the busy schedule.

Option 1: Surveyors' salaries can

be on cash basis

Option 2: Surveyors to open

ATM account so that salaries

can be coursed through it

Administrative No budget proposal for field

survey at the start of the

survey

Budgetary requirement during

field work should be forecasted

before the actual

implementation of the project.

Human

Resource

One enumerator had URI

and needed to rest to

recover from the illness.

Assistant data encoder

replaced enumerator for 1

day. Orientation was given

the night before the field

work.

Train additional surveyors that

can be pulled out anytime as

need arises. For instance, with 4

teams at the end, train 5 teams

(15 persons). The final

enumerators will be selected

based on post-test results and

standardization test.



Technical Some barangay had no household list, no map, and no defined boundaries. Barangay officials and health

workers were also not

familiar with their areas.

The team mapped out the

barangay in order to

determine the size,

boundaries and number of

households. Team also

visited each household to

assign household

numbering for random

selection.

Early communication and

coordination prior to the data

collection in order for the

barangay to prepare necessary

documents for the survey.

Technical Protective parents refused

to have their child

measured. The child was

also sick (vomiting) at the

time of visit.

Team explained the

importance of getting the

measurement of the child

and its overall contribution

to the survey. The team

revisited the household the

following day but the

members were not around.

Include in the

protocol/guidelines on

anthropometric measurement

what to do if the child is sick. If

it is too difficult to take

measurements with a sick child,

data are considered as missing

(always write a section special

case in enumerator guide)

Technical Height of the child with

disability (cerebral palsy)

was not measured.

The team measured what

can be measured

Measurement of child with

physical disabilities should be

discussed thoroughly in the

training and included in the

protocol. A section for special

cases can be included in the

enumerator’s guide. Always

take into account children with


disabilities (no stigma). This

case may be recorded as

“missing data”.

Data Collection/ Implementation

Technical Older children refused to remove their clothes and to be measured.

Older siblings were also measured to reassure and to gain trust of the younger children.

Always ask the assistance of the parents/guardians.

Technical Almost all children in one survey area cried as the team approaches.

Households with crying and restless children were scheduled for a revisit to give time to calm the children.

Develop a strategy on how to manage crying and hysterical children. The protocol of measurements should be explained to mothers /caregivers/older children. Always start with older children/more confident as example. Have some candies ready. Be patient.



Technical Some households asked why

they were not included in

the survey. The use of the

term “random” was not

clear to some HH.

Before the departure of the

team for the field work, it

was explained to the

barangay officials and

volunteer workers the

objective of the survey and

the protocol being followed.

They were also involved in

the random sampling of

households for them to

understand the process and

help the team in explaining

to the community the

selection procedure of

households. Showing of the

random number table

facilitated easier

understanding of the

selection process.

Development of a simple and

easy to understand graphic

presentation that can be used in

the field for the next survey.

Also, prior to data collection,

ensure that barangay officials

and health workers clearly

understand the objective and

purpose, and the random

selection process. This will

facilitate smooth flow of the

data collection.

Technical In poor urban areas, some HH did not have enough space for measuring the height and weight of the children

Children were brought to a nearby HH for the measurement.

Plan for alternative nearest venue in the household to get children’s measurements.

Technical Majority of the HH were expecting relief goods from the survey team.

The team explained to the community that the survey

Always clearly explain to the household visited the objective of the survey and emphasize


is for assessment of NS and not a relief operation.

how important their contribution as respondents to the survey and future program planning.


Technical In urban areas, most of the randomly selected household were at work during the time of survey.

Households were revisited early the next day.

Better to schedule urban barangays during weekends to ensure availability of the respondents/HH members.

Technical Missing column for the date

of weighing/measurement

in anthropometric

questionnaires.

Date of

weighing/measurement

were noted in the

questionnaire

Add another column on date of

weighing/ measurement in the

questionnaire for the next

survey. This will give accurate

computation of age in months

at the time children are

measured.

Technical No field supervisor to

address administrative,

logistics and technical

concerns in the field.

One team leader was doing

the coordination and

addressing concerns with

regards to administrative,

logistics, and technical

issues in the field.

There should be at least two

field supervisors in the field to

oversee and address concerns

in the field. The best scheme for

the next survey is to have 2

supervisors for 4 teams and one

national survey coordinator.

Technical Different classification of nutritional status between Unisex CGS tables and ENA software

Team agreed to use the sex-based WHO-CGS reference tables to be consistent with the ENA software in

Tools to use in the survey should be finalized during the training and pre-tested before the actual survey to avoid inconsistencies.


classifying nutritional status of children.

Technical Improvised test weights used in calibration of weighing scale is too light (water bottle).

Teams agreed to use car “brick” or “jack” to calibrate/verify the weighing scales as also suggested during the monitoring visit.

Use standard test weight usually with weights 3kg, 5kg and 10kg in calibration of weighing scales for the next survey.


Technical Far-flung and hard to reach households

The team asked the help of barangay officials to meet these households in a more accessible location.

Households that are inaccessible may not be visited but make sure to that this is documented. No replacement of randomly selected household is allowed.

Technical No calibration/verification for height board and MUAC tape

Replacement of MUAC tapes when necessary

Standardization of height board using a piece of wood (1.10 m) and MUAC tapes with a PVC tube or a new one every day.

Technical No MUAC tapes for women The team used MUAC tape for children

Purchase of MUAC tapes for women for the next survey.

Technical Answer for Vitamin A and Deworming (Yes or No)

Add for Yes, with card and without card (like measles vaccination)

Technical No double data entry Each team validated data encoded by the encoder against the raw questionnaires and data encoded in ENA software.

Consider the double data entry for the next survey to facilitate validation and accuracy of dataset.


Data analysis

Technical

Carried out in short time by different agencies

-

In the next survey round, a part time statistician should be hired and added to the team to generate the necessary tables/data.

Report writing and finalization

Technical Carried out in a limited time by the survey team with the help of individual agencies

- In the next survey round, a part time writer should be hired and added to the team to generate the draft report for discussion among partners.


Annex 5: Data Quality Plausibility check for: PHI_201402_ACF_FINAL_updated_Mar20.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

Missing/Flagged data Incl % 0-2.5 >2.5-5.0 >5.0-7.5 >7.5

(% of in-range subjects) 0 5 10 20 0 (0.9 %)

Overall Sex ratio Incl p >0.1 >0.05 >0.001 <=0.001

(Significant chi square) 0 2 4 10 0 (p=0.408)

Overall Age distrib Incl p >0.1 >0.05 >0.001 <=0.001

(Significant chi square) 0 2 4 10 2 (p=0.077)

Dig pref score - weight Incl # 0-7 8-12 13-20 > 20

0 2 4 10 0 (5)

Dig pref score - height Incl # 0-7 8-12 13-20 > 20

0 2 4 10 0 (4)

Dig pref score - MUAC Incl # 0-7 8-12 13-20 > 20

0 2 4 10 0 (6)

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 2 6 20 0 (0.94)

Skewness WHZ Excl # <±0.2 <±0.4 <±0.6 >=±0.6

0 1 3 5 1 (0.22)

Kurtosis WHZ Excl # <±0.2 <±0.4 <±0.6 >=±0.6

0 1 3 5 1 (0.39)

Poisson dist WHZ-2 Excl p >0.05 >0.01 >0.001 <=0.001

0 1 3 5 0 (p=0.547)

Timing Excl Not determined yet

0 1 3 5

OVERALL SCORE WHZ = 0-9 10-14 15-24 >25 4 %

The overall score of this survey is 4 %, this is excellent.

There were no duplicate entries detected.

Missing data:

WEIGHT: Line=1/ID=1, Line=2/ID=1, Line=3/ID=1, Line=4/ID=1

HEIGHT: Line=1/ID=1, Line=2/ID=1, Line=3/ID=1, Line=4/ID=1, Line=229/ID=2,

Line=245/ID=1, Line=281/ID=1, Line=285/ID=2, Line=286/ID=1, Line=295/ID=1,

Line=312/ID=1


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 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=5/ID=1: WAZ (-4.173), Weight may be incorrect

Line=64/ID=2: HAZ (-4.583), Age may be incorrect




Line=505/ID=1: HAZ (1.900), Age may be incorrect

Line=532/ID=1: WAZ (2.258), Weight may be incorrect

Line=627/ID=1: WHZ (4.947), WAZ (3.061), Weight may be incorrect

Line=632/ID=1: WAZ (1.922), Weight may be incorrect





Line=645/ID=1: WHZ (3.746), HAZ (2.223), WAZ (3.862)

Percentage of values flagged with SMART flags:WHZ: 0.9 %, HAZ: 0.9 %, WAZ: 1.4 %

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 : #####################

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: 0.76 (The value should be around 0.85).

Statistical evaluation of sex and age ratios (using Chi squared statistic): Age cat. mo. boys girls total ratio boys/girls

-------------------------------------------------------------------------------------

6 to 17 12 62/77.3 (0.8) 67/72.4 (0.9) 129/149.7 (0.9) 0.93

18 to 29 12 78/75.3 (1.0) 72/70.6 (1.0) 150/145.9 (1.0) 1.08

30 to 41 12 84/73.0 (1.2) 72/68.4 (1.1) 156/141.4 (1.1) 1.17

42 to 53 12 78/71.9 (1.1) 76/67.3 (1.1) 154/139.2 (1.1) 1.03

54 to 59 6 31/35.5 (0.9) 25/33.3 (0.8) 56/68.8 (0.8) 1.24

-------------------------------------------------------------------------------------

6 to 59 54 333/322.5 (1.0) 312/322.5 (1.0) 1.07

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


Overall sex ratio: p-value = 0.408 (boys and girls equally represented)

Overall age distribution: p-value = 0.077 (as expected)

Overall age distribution for boys: p-value = 0.209 (as expected)

Overall age distribution for girls: p-value = 0.433 (as expected)

Overall sex/age distribution: p-value = 0.034 (significant difference)

Digit preference Weight:

Digit .0 : ################################

Digit .1 : ########################

Digit .2 : ####################################

Digit .3 : #############################

Digit .4 : ################################

Digit .5 : #################################

Digit .6 : ########################################

Digit .7 : ########################

Digit .8 : ####################################

Digit .9 : #################################

Digit preference score: 5 (0-7 excellent, 8-12 good, 13-20 acceptable and > 20 problematic)

p-value for chi2: 0.098

Digit preference Height:

Digit .0 : ##############################

Digit .1 : ##################################

Digit .2 : ########################################

Digit .3 : ################################

Digit .4 : ##############################

Digit .5 : ##############################

Digit .6 : ###############################

Digit .7 : ##################################

Digit .8 : ##########################

Digit .9 : ##############################


p-value for chi2: 0.607

Digit preference MUAC:

Digit .0 : ##################

Digit .1 : ################################

Digit .2 : #############################

Digit .3 : ####################################

Digit .4 : ######################################

Digit .5 : ##########################


Digit .6 : #############################

Digit .7 : ####################################

Digit .8 : ######################################

Digit .9 : ####################################


p-value for chi2: 0.008 (significant difference)

Evaluation of Standard deviation, Normal distribution, Skewness and Kurtosis using

the 3 exclusion (Flag) procedures . no exclusion exclusion from exclusion from

. reference mean observed mean

. (WHO flags) (SMART flags)

WHZ

Standard Deviation SD: 1.02 1.02 0.94

(The SD should be between 0.8 and 1.2)

Prevalence (< -2)

observed: 4.1% 4.1%

calculated with current SD: 5.8% 5.8%

calculated with a SD of 1: 5.5% 5.5%

HAZ



Prevalence (< -2)

observed: 30.9% 30.9% 30.6%

calculated with current SD: 32.0% 32.0% 30.9%

calculated with a SD of 1: 30.5% 30.5% 30.1%

WAZ



Prevalence (< -2)

observed: 20.6% 20.6% 20.7%

calculated with current SD: 21.4% 21.4% 20.8%

calculated with a SD of 1: 19.1% 19.1% 20.3%

Results for Shapiro-Wilk test for normally (Gaussian) distributed data:

WHZ p= 0.000 p= 0.000 p= 0.005

HAZ p= 0.039 p= 0.039 p= 0.008

WAZ p= 0.000 p= 0.000 p= 0.007

(If p < 0.05 then the data are not normally distributed. If p > 0.05 you can consider the data

normally distributed)

Skewness

WHZ 0.68 0.68 0.22

HAZ 0.14 0.14 0.23

WAZ 0.51 0.51 0.21

If the absolute 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 2.06 2.06 0.39

HAZ 0.39 0.39 -0.10

WAZ 1.06 1.06 0.09

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=0.97 (p=0.547)

WHZ < -3: ID=0.98 (p=0.512)

GAM: ID=0.97 (p=0.547)

SAM: ID=0.98 (p=0.512)

HAZ < -2: ID=2.51 (p=0.000)

HAZ < -3: ID=1.59 (p=0.003)

WAZ < -2: ID=1.87 (p=0.000)

WAZ < -3: ID=1.24 (p=0.100)

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: 1.40 (n=59, f=3) #########################

02: 1.03 (n=59, f=0) #########

03: 0.91 (n=58, f=0) #####

04: 1.11 (n=55, f=2) #############

05: 0.92 (n=56, f=0) #####

06: 0.89 (n=53, f=0) ####

07: 1.15 (n=53, f=1) ###############

08: 0.82 (n=44, f=0) #

09: 0.91 (n=41, f=0) #####

10: 1.07 (n=35, f=0) ###########

11: 0.81 (n=30, f=0)

12: 0.87 (n=22, f=0) ###

13: 0.92 (n=20, f=0) OOOOO

14: 0.87 (n=14, f=0) OOO

15: 0.84 (n=10, f=0) ~

16: 0.94 (n=07, f=0) ~~~~~~

17: 0.42 (n=05, f=0)

18: 1.03 (n=03, f=0) ~~~~~~~~~~

19: 1.41 (n=03, f=0) ~~~~~~~~~~~~~~~~~~~~~~~~~~

20: 1.25 (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 n = 135 173 171 166

Percentage of values flagged with SMART flags: WHZ: 3.0 1.2 0.6 6.3

HAZ: 3.0 0.6 1.8 5.7

WAZ: 2.2 0.6 0.6 4.9

Age ratio of 6-29 months to 30-59 months: 0.71 0.84 0.73 0.77

Sex ratio (male/female): 0.93 1.22 1.14 0.98

Digit preference Weight (%): .0 : 8 11 11 9

.1 : 7 9 6 7

.2 : 10 13 10 12

.3 : 11 8 9 9

.4 : 10 8 13 10

.5 : 9 10 11 11

.6 : 11 11 12 15

.7 : 7 7 10 7

.8 : 12 12 11 10

.9 : 14 11 7 10

DPS: 7 7 7 8

Digit preference score (0-7 excellent, 8-12 good, 13-20 acceptable and > 20 problematic)

Digit preference Height (%): .0 : 3 13 8 11

.1 : 8 7 13 14

.2 : 12 13 14 11

.3 : 7 10 13 9

.4 : 11 11 8 8

.5 : 11 9 8 9

.6 : 14 9 9 8

.7 : 11 13 12 8

.8 : 11 10 5 8

.9 : 11 4 11 13

DPS: 10 9 9 8


Digit preference MUAC (%): .0 : 4 10 4 5

.1 : 11 7 9 14

.2 : 7 8 13 7

.3 : 11 13 11 9

.4 : 13 14 9 12

.5 : 7 9 7 9

.6 : 10 7 12 8

.7 : 16 10 13 7

.8 : 10 11 15 12


.9 : 10 10 8 16

DPS: 11 7 11 11


Standard deviation of WHZ: SD 1.07 0.99 0.88 1.12

Prevalence (< -2) observed:

% 4.5 4.4

Prevalence (< -2) calculated with current SD:

% 5.6 7.4

Prevalence (< -2) calculated with a SD of 1:

% 4.4 5.2

Standard deviation of HAZ: SD 1.20 1.00 1.00 1.11

observed:

% 32.3 40.6 23.3

calculated with current SD:

% 31.0 41.4 24.9

calculated with a SD of 1:

% 27.6 41.4 22.5

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 10/15.1 (0.7) 17/16.2 (1.0) 27/31.3 (0.9) 0.59

18 to 29 12 19/14.7 (1.3) 10/15.8 (0.6) 29/30.5 (0.9) 1.90

30 to 41 12 10/14.3 (0.7) 17/15.3 (1.1) 27/29.6 (0.9) 0.59

42 to 53 12 19/14.0 (1.4) 20/15.1 (1.3) 39/29.1 (1.3) 0.95

54 to 59 6 7/6.9 (1.0) 6/7.5 (0.8) 13/14.4 (0.9) 1.17

-------------------------------------------------------------------------------------

6 to 59 54 65/67.5 (1.0) 70/67.5 (1.0) 0.93








-------------------------------------------------------------------------------------

6 to 17 12 23/22.0 (1.0) 17/18.1 (0.9) 40/40.1 (1.0) 1.35

18 to 29 12 25/21.5 (1.2) 14/17.6 (0.8) 39/39.1 (1.0) 1.79

30 to 41 12 27/20.8 (1.3) 23/17.1 (1.3) 50/37.9 (1.3) 1.17

42 to 53 12 12/20.5 (0.6) 16/16.8 (1.0) 28/37.3 (0.8) 0.75

54 to 59 6 8/10.1 (0.8) 8/8.3 (1.0) 16/18.5 (0.9) 1.00

-------------------------------------------------------------------------------------


6 to 59 54 95/86.5 (1.1) 78/86.5 (0.9) 1.22








-------------------------------------------------------------------------------------

6 to 17 12 16/21.1 (0.8) 14/18.6 (0.8) 30/39.7 (0.8) 1.14

18 to 29 12 17/20.6 (0.8) 25/18.1 (1.4) 42/38.7 (1.1) 0.68

30 to 41 12 28/20.0 (1.4) 15/17.5 (0.9) 43/37.5 (1.1) 1.87

42 to 53 12 24/19.6 (1.2) 18/17.3 (1.0) 42/36.9 (1.1) 1.33

54 to 59 6 6/9.7 (0.6) 8/8.5 (0.9) 14/18.3 (0.8) 0.75

-------------------------------------------------------------------------------------

6 to 59 54 91/85.5 (1.1) 80/85.5 (0.9) 1.14








-------------------------------------------------------------------------------------

6 to 17 12 13/19.0 (0.7) 19/19.5 (1.0) 32/38.5 (0.8) 0.68

18 to 29 12 17/18.5 (0.9) 23/19.0 (1.2) 40/37.6 (1.1) 0.74

30 to 41 12 19/18.0 (1.1) 17/18.4 (0.9) 36/36.4 (1.0) 1.12

42 to 53 12 23/17.7 (1.3) 22/18.1 (1.2) 45/35.8 (1.3) 1.05

54 to 59 6 10/8.8 (1.1) 3/9.0 (0.3) 13/17.7 (0.7) 3.33

-------------------------------------------------------------------------------------

6 to 59 54 82/83.0 (1.0) 84/83.0 (1.0) 0.98







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: 1.21 (n=15, f=0) #################

02: 1.02 (n=14, f=0) #########

03: 1.07 (n=15, f=0) ###########

04: 1.46 (n=14, f=2) ############################

05: 0.86 (n=15, f=0) ###

06: 1.14 (n=12, f=0) ##############

07: 0.77 (n=12, f=0)

08: 0.68 (n=06, f=0)

09: 0.96 (n=06, f=0) #######

10: 1.30 (n=06, f=0) #####################

11: 0.65 (n=05, f=0)

12: 0.71 (n=03, f=0)

13: 0.15 (n=02, f=0)

14: 0.15 (n=02, f=0)





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: 1.71 (n=15, f=1) ######################################

02: 0.77 (n=15, f=0)

03: 0.98 (n=15, f=0) ########

04: 0.83 (n=15, f=0) #

05: 0.93 (n=15, f=0) ######

06: 0.94 (n=15, f=0) ######

07: 0.76 (n=15, f=0)

08: 0.68 (n=13, f=0)

09: 0.81 (n=12, f=0) #

10: 1.01 (n=12, f=0) #########

11: 1.09 (n=10, f=0) ############

12: 0.70 (n=06, f=0)

13: 0.89 (n=05, f=0) OOOO

14: 0.72 (n=03, f=0)

15: 1.20 (n=03, f=0) ~~~~~~~~~~~~~~~~~

16: 2.09 (n=02, f=0) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~





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: 1.09 (n=15, f=0) ############

02: 0.79 (n=15, f=0)

03: 0.88 (n=14, f=0) ###

04: 0.64 (n=13, f=0)

05: 1.02 (n=14, f=0) #########

06: 0.90 (n=13, f=0) ####

07: 1.00 (n=13, f=0) ########

08: 0.91 (n=13, f=0) #####

09: 0.97 (n=11, f=0) #######

10: 1.04 (n=09, f=0) ##########

11: 0.43 (n=07, f=0)

12: 0.99 (n=06, f=0) ########

13: 0.70 (n=06, f=0)

14: 0.93 (n=05, f=0) OOOOOO

15: 0.58 (n=03, f=0)

16: 0.07 (n=03, f=0)

17: 0.36 (n=02, f=0)






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: 1.52 (n=14, f=2) ##############################

02: 1.36 (n=15, f=0) #######################

03: 0.72 (n=14, f=0)

04: 1.31 (n=13, f=0) #####################

05: 0.95 (n=12, f=0) ######

06: 0.58 (n=13, f=0)

07: 1.41 (n=13, f=1) ##########################

08: 0.85 (n=12, f=0) ##

09: 1.03 (n=12, f=0) ##########

10: 1.11 (n=08, f=0) #############

11: 0.81 (n=08, f=0)

12: 0.99 (n=07, f=0) ########

13: 0.84 (n=07, f=0) #

14: 1.22 (n=04, f=0) OOOOOOOOOOOOOOOOO

15: 1.03 (n=03, f=0) ~~~~~~~~~~




Nutrition Survey Using SMART Method · PPS Probability Proportion to Size PSU Primary Sampling Unit...

Documents

Transcript of Nutrition Survey Using SMART Method · PPS Probability Proportion to Size PSU Primary Sampling Unit...