Nutrition and mortality SMART survey,

IYCF survey and household dietary diversity score

Panna district,

Pawai block

Madhya Pradesh, India

Dates of the survey : 15^th – 21^st of September 2013

THANKS

Welhungerhilfe and Jansahas would like to thank Madhya Pradesh health and nutrition authorities, and the head of villages surveyed.

Welthungerhilfe and Jansahas would also like to thank all the surveyors for their work, their enthusiasm, the quality of their work and their flexibility, and also the drivers without whom this work could not have been done.

Welthungerhilfe and Jansahas would like to thank all the mothers, all the heads of household and all the children who have participated to this survey.

Rakesh Shrivastava (MSW,MPH), Project Coordinator from Welthungerhilfe and Harshal Jariwala(MSW-TISS,PGDHRM-MSU), Project Coordinator from Jansahas Social Development Society have significantly contributed in conducting this research study and its presentation of findings in form of reports, posters, summaries, brief notes at further various platforms. 

Mr. Vijay Rai, Program Manager, Welthungerhilfe and Ms. Heena Panday, Nutritionist, Welthungerhilfe had major support to conduct this research. 

CONTENTS

ABBREVIATIONS USED

LIST OF THE TABLES

LIST OF THE FIGURES

INTRODUCTION / CONTEXT

1.    OBJECTIVES

1.1    Main objectives

1.2    Specific objectives

2.    METHODOLOGY

2.1    Survey

2.2    Target population

2.3    Sampling

2.4    Children selection for the anthropometry questionnaire

2.5    Data collected

     2.5.1. Mortality questionnaire

     2.5.2. Anthropometry questionnaire

     2.5.3. IYCF questionnaire, children aged 0 to 23 months

     2.5.4. Household dietary diversity questionnaire

2.6 Indicators and cut off used

     2.6.1 Weight for Height indicator

     2.6.2 Mid Upper Arm Circumference (MUAC)

     2.6.3 Height for Age indicator

     2.6.4 Weight for Age

     2.6.5 Mortality rate

     2.6.6 IYCF indicators

     2.6.7Household Dietary Diversity Score (HDDS) 

2.7    Action taken in case of a malnourished child (included or not in the selected household)

2.8  Training, Supervision and survey

2.8.1   Training

2.8.2   Supervision

2.8.3   Survey

2.9 Ethic

2.10 Data analysis

3      ANTHROPOMETRIC RESULTS

3.1 Sample characteristics

3.2 Prevalence of acute malnutrition

3.3 MUAC distribution

3.5  Prevalence of stunting

3.5 Prevalence of underweight

3.6 Retrospective mortality

4      IYCF RESULTS

5      HOUSEHOLD DIETARY DIVERSITY RESULTS

6      DISCUSSION

7      CONCLUSION & RECOMMENDATIONS

ANNEXES

ABBREVIATIONS USED

CTC: Community based Therapeutic Care

ENA: Emergency Nutrition Assessment

FANTA: Food and Nutrition Technical Assistance Project

GAM: Global Acute Malnutrition

HDDS: Household Dietary Diversity Score

IDDS: Individual Dietary Diversity Score

IYCF: Infant and Young Child Feeding

MAM: Moderate Acute Malnutrition

MUAC: Mid Upper Arm Circumference

NFHS: National Family Health Survey

NIN: National Institute of Nutrition

NRC: Nutrition Rehabilitation Centre

PPS: Probability Proportional to Size

SAM: Severe Acute Malnutrition

SMART: Standardized Monitoring and Assessment of Relief and Transitions

UNICEF: United Nation Childrens’ Fund

WHO: World Health Organization

WWH: Welt Hunger Hilfe

LIST OF THE TABLES

Table 1: Description of the method used to form the sample of the SMART nutrition survey

Table 2: Weight for Height (W/H) values defining global acute malnutrition (GAM), moderate acute malnutrition (MAM) and severe acute malnutrition (SAM), with WHO 2006 standards

Table 3: Cut off values for MUAC measurement defining moderate acute malnutrition and severe acute malnutrition with WHO 2006 standards

Table 4: Cut off values for Height for Age (H/A) indicator, defining global chronic malnutrition, moderate chronic malnutrition and severe chronic malnutrition, with WHO 2006 standards

Table 5: Cut off values for Weight for Age (W/A) indicator, defining global, moderate and severe underweight, with WHO 2006 standards

Table 6: admission criteria for children in Madhya Pradesh

Table 7: Sampling characteristics

Table 8: Distribution of age and sex of sample

Table 9: Mean z-scores, Design Effects and excluded subjects

Table 10: Malnutrition rates, children aged 6-59 months and children aged 6-29 months, Weight for Height z-score, WHO 2006 reference population

Table 11: Prevalence of acute malnutrition by age, based on weight-for-height z-scores and/or oedemas

Table 12: Prevalence of acute malnutrition based on MUAC cut off's (and/or oedema) and by sex

Table 13: Prevalence of acute malnutrition by age, based on MUAC cut off's and/or oedemas

Table 14: Prevalence of stunting based on height-for-age z-scores and by sex

Table 15: Prevalence of underweight based on weight-for-age z-scores by sex

LIST OF THE FIGURES

Figure 1: Population age and sex pyramid

Figure 2: Distribution of Weight for Height (W/H) indicator, WHO 2006 standards

Figure 3: Food groups eaten by surveyed households in the last 24 hours

Figure 4: Household Dietary Diversity Score (HDDS) in household surveyed

INTRODUCTION / CONTEXT

Madhya Pradesh is the state with the highest level of child malnutrition rate in India.

The results of the National Family Health Survey NFHS-3 (2005-06) for Madhya Pradesh show that, among children under 5 years old, 50% of them were suffering from chronic malnutrition (stunting). Chronic malnutrition leads to growth, physical and intellectual development retardation for children, with a strong impact on the human capital. Moreover, 35% of the children were suffering from acute malnutrition (stunting). Acute severe malnutrition has immediate consequences on the child survival.

Madhya Pradesh is also the only state in India that falls into the “extremely alarming” category of the India State Hunger Index developed by Welthungerhilfe and IFPRI. The last few years have seen an increased focus on malnutrition in Madhya Pradesh by both government and non-governmental organizations.

Responding to the alarming situation of malnutrition in Madhya Pradesh, Welthungerhilfe in partnership with Jansahas have started the implementation of a pilot project to address the issue of malnutrition in the Pawai and Jhirniya blocks of Panna and Kargone districts.

The SMART nutrition and mortality survey has been carried out at the beginning of this project, in order to have baseline data.

The objectives were to evaluate:

-          the prevalence of acute and chronic malnutrition among children 6-59 months,

-          the crude death rate and the 0-5 years death rate ;

-          Infant and Young Child Feeding (IYCF) practices for children aged 0-23 months;

-          Household Dietary Diversity Intake (HDDI) at household level.

The SMART survey has taken place at the end of the rainy season, just after the hunger peak season (jly august).

Data collection for Pawaï block took place from the 15th of September until the 21^stof September.

The total population for the surveyed area (Pawaï block) was estimated 165983 (census data, 2011).

1.       OBJECTIVES

1.1   Mainobjective

To evaluate the nutritional status of children 6-59 months and the death rates, in Pawaï block (Panna district), in order to have a better understanding of the nutrition problematic.

1.2 Specific objectives

In Pawaï block :

1.      Determine the prevalence of Global Acute Malnutrition (GAM) and Severe Acute Malnutrition (SAM) for children aged 6 months to 59 months (WHO[1] 2006) ;

2. Bring out, or not, vulnerable groups regarding malnutrition (age, gender) ;
3. Estimate the prevalence for chronic acute malnutrition for children 6-59 months and 0-5 months ;
4. Estimate the crude death rate and the 0-5 years children’s death rate, for the retrospective period from the 20^th of June 2013 until the 18^th of September (90 days) ;
5. Evaluate the IYCF practices among children 0-23 months ;
6. Evaluate the Household Dietary Diversity Score.

1.      

2.     METHODOLOGY

Added to mortality and anthropometry questionnaires, two other questionnaires have been used during this survey:

a)      IYCF questionnaire : to evaluate breastfeeding practices and complementary feeding for children aged 0-23 months ;

b)      Household Dietary Diversity questionnaire: to evaluate the household’s access to food diversity (24h recall period).

2.1. Survey

Because of thesparse populationand lackofhousehold lists, cluster samplingwas conducted. It is a transversal nutrition cluster survey (2 degrees), with a data collection done with anthropometric measurements and questionnaires.

2.2. Target population

For the anthropometry, target population was children aged 6-59 months because there are the most vulnerable population regarding malnutrition.

For the mortality, target population was the entire household.

For the IYCF, target population was the children aged 0-23 months.

For Household Dietary Diversity Intake, target population was the entire household.

2.3. Sampling

Population data from the 2011 census have been used. Data were available at village level.

Sample size calculation has been done with ENA software, with the hypothesis presented in table 1.

The survey has been conducted using the SMART[2] methodology, a quick, standardized and simplified survey method with daily data entry, to improve the quality.

A total of 42 clusters, with 17 household each (including a non-responding rate of 7%) has been included in the sample. This population’s sample is enough to represent statistically the whole population in the area surveyed. Cluster selection has been done with the ENA[3] for SMART software, to ensure that each household had the same chance to be selected in all the different village of the area surveyed.

Some villages had several hamlets (falyas) sometimes quite far one from another. When these villages have been randomly selected, selection of the hamlet to survey has been done using the PPS method (Probability Proportional to Size).

When a big village had to be segmented, segmentation and random selection has been done using the PPS method.

Knowing that eight teams would be on the field, it has been calculated than 42/8=6 days of data collection will be necessary to survey all the villages.

Table 1: Description of the method used to form the sample of the SMART nutrition survey (SMART Nutrition and mortality survey, Panna district, Pawaï block, WHH & Jansahas, September 2013)

---

[1]World Health Organization

[2] Standardized Monitoring and Assessment of Relief and Transition

[3] Emergency Nutrition Assessment

Nutrition survey	Parameters	Calculation	Justification
	Total population	165983	Census 2011
	Population children <5	24067	Considering 14,5% of total population (2011 census, Madhya Pradesh)
	GAM prevalence estimated	29,0%	National Institute of Nutrition (NIN) 2011
	Precision desired	5,0	Precision desired according to the prevalence rate
	Design effect	1,5	NFHS
	Average HH size	6,0	NIN 2011
	Number of children less than 5 years per HH	0,87	6,0x0,145
	Non response rate %	7	NFHS 2005-06
	Sample size – number of children	517	Calculation done by ENA for SMART
	Sample size – number of HH	709	Calculation done by ENA for SMART with the data entered
	Number of households per cluster	17	Estimation of the number of HH that could be surveyed per day regarding the context
	Number of clusters to include	42	709 (total number of HH for the survey) / 17 = 42 clusters

 

Household selection

The second sampling’s degree was the selection of the households inside the clusters.

This selection has been done using the systematic random sampling with calculation of the sampling interval (most of the villages) or using the simple random sampling (if small village/hamlet where a Households’ list was available).

Definition for the household was “all the persons who are eating from the pot and who have one household’s chief”.

If several women were eating from the same pot, all these women were considered as members of the household and their children were included in the sample.

If households where parents and married children were living, each married child with his wife and or children was considered as one household. If one or both parents were depending on the married child he was part of the household.

If one household or one child was absent, he was not replaced by another household or child.

Segmentation :

If the cluster selected was with a spread population (falyas) or was a big village (important number of households):

1.      Population was divided in segments with 100 households maximum.

2.      One of these segments was randomly selected using the PPS method and the village was reduced to an area with no more than 100 household,easier to survey.

3.      The 17 households were then selected by the systematic random sampling or with the simple random sampling in case a list of the households was available.

2.4 Children selection for the anthropometry questionnaire

In the household, all the children from 0 to 59 months were surveyed for the anthropometry. Analysis has been done for children 6-59 months. Stunting has also been studied for children 0-5 months

Each absent child who was  filling the inclusion criteria was written on the anthropometry questionnaire.

The team came back to the household at the end of the day to take the measurements. If the child was still absent, this was notified and the child was not replaced.

If households members were absent, surveyors had to come back to the household at the end of the day. If the household was still absent, this was notified on the household selection sheet and this household was not replaced by another one.

If one child was hospitalized (hospital or NRC), he had to be measured at the end of day by the team, and if not possible, the team had to contact the centre to have the anthropometric data.

Disabled children were included in the survey, and disability was written on the questionnaire. All the measurements which could be taken were taken and written on the questionnaire. If the disability made not possible the measurements, these data were considered as missing. For children with a disability on the left arm, MUAC measurement was not taken and a note was written on the anthropometrc questionnaire.

2.5Data collected

2.5.1.   Mortality questionnaire(see annex)

Retrospective mortality survey has been done on a 3 months recall period before the day of the survey. The beginning of the recall period was the beginning of the moonsoon (20th of june) and the end of the recall period was the middle of the survey, the 18th of september, which was corresponding to a 90 days recall period.

Questionnaire used was the simplified questionnaire at cluster level (one line per household). It has been administrated in every surveyed household, to the head of household, his representant or to the mother of the children.

Following information was collected :

·         Total number of household’s members, present on the day of the survey (total and children less than 5 years).

·         Number of people who came during the recall period (total and children less than 5 years).

·         Number of people who left the household during the recall period (total and children less than 5 years.

·         Number of births during the recall period.

Number of deaths during the recall period (total and children less than 5 years who have lived more than 24h in the household).

2.5.2.   Anthropometry questionnaire (see annex)

It has been done for all the children in the household from 0 to 59 months.

Sex:

It was coded « M » for male and «  F » for female.

Age:

Age was written in months, except if the exact birthdate was available on official documents (ID card, birth certificate,…). When birthdate was not confirmed by official document, local event calendar was used (see annex) to estimate the age (in months). This calendar was made with seasonal events, as beginning or end of the moonsoon, harvesting periods, religious events, national or local events.

All the children from 0 to 59 months were included, that is to say children who were born between september 2008 and september 2009.

Age criteria was preferred to height criteria, as recommended by SMART metodology.

Weight:

Weight measurement was done with Salter scales with a 100 g precision. Children were weighted totally naked. Each day, before leaving for the field, teams were checking the scale with a 2 kg standard weight.

Height:

Height was measured with a locally made measuring board, in centimeters, at the nearest millimeter.

Children less than 87 cm were measured lying down and children more than 87 cm were measured standing up.

A wood stick of 110cm and marked at 87cm was used to see whether the child had to be measured lying down or standing up.

Diagnosis of oedemas:

Only bilateral oedemas were considered as nutritional oedemas.

A three seconds pressure was done on both feet with the thumbs. Oedemas were present if there were thumb’s prints on both feet. They were coded Y for yes, and N for no.

Mid Upper Arm Circumference (MUAC) :

MUAC was measured on the left arm with the MUAC tape, in the middle of the arm, according to the methodology. MUAC was measured in millimeters and at the nearest millimeter.

2.5.3    IYCF questionnaire, children aged 0 to 23 months (see annex)

IYCF questionnaire has been done only when there was a child 0-23 months in the selected household. In every household surveyed for anthropometry, one child 0-23 month was randomly selected among all the children 0-23 months present in the household and IYCF questionnaire was done for this child. If no child 0-23 month was present in the household, no IYCF was done.

IYCF questionnaire was based on the WHO recommandations for the breasfeeding part.

Data collected gave us information on :

-       Breastfeeding(introduction of breastfeeding, early initiation to breastfeeding, exclusive breastfeeding, breastfeeding practices, breastfeeding duration…)

-       Introduction of other food after birth (if yes, which one)

-       Weaning (reasons, way of weaning the child)

-       Consumption of liquid in the last 24h

-       Individual dietary diversity in the last 24h (IDDS, for children consuming complementary food)

IDDS questions were based on the FANTA[1] method.

Only pertinent questions were asked to the person in charge of the child : if the child has been breasfed, no question was ask about breasfeeding practices.

2.5.4.   Household dietary diversity questionnaire(see annex)

Household Dietary Diversity questionnaire has been done in one household upon two (9 first households). If one household selected was absent, the next one was surveyed, in order to have 9 questionnaires per cluster.

Dietary diversity gives qualitative information on food consumption.

Questions were based on the FANTA method, but at household level. This allowed to measure the household access to the different food groups.

Twelve food groups were defined in the questionnaire:

1.            Cereals

2.            Roots and tubers

3.            Vegetables

4.            Fruits

5.            Meats, poultry and offals

6.            Eggs

7.            Fish and seafood

8.            Pulses, legumes, seeds and nuts

9.            Milks and milk products

10.        Oils and greases

11.        Sweets and honey

12.        Misceallenous

The methodology uses these 12 groups for the analysis.

A score at household level has been calculated : Household Dietary Diversity Score (HDDS).

HDDS can facilitate the dietary changes measurement before and after anintervention (improvement) or after a disaster (worsening situation).

---

[1] Food and Nutrition Technical Assistance Project

2.6 Indicators and cut off used

2.6.1 Weight for Height indicator

For children, acute malnutrition rates are estimated from the Weight for Height values combined with presence of oedemas. Weight for Height indicator compares the weight of the measured child to a reference population for the same size. References values used are those from WHO 2006. Weight for Height indicator is expressed in z-score.

Table2 : Weight for Height (W/H) values defining global acute malnutrition (GAM), moderate acute malnutrition (MAM) and severe acute malnutrition (SAM), with WHO 2006 standards(SMART Nutrition and mortality survey, Panna district, Pawaï block, WHH & Jansahas, September 2013)

Acute malnutrition	Weight for Height indicator
Global	W/H<-2 z-score and/or bilateral oedemas
Moderate	-2≤W/H <-2 z-score
Severe	W/H<-3 z-score And/ or Bilateral oedemas

2.6.2 Mid Upper Arm Circumference (MUAC)

MUAC is used for a rapid screening and measures the mortality risk. This is also a secondary malnutrition indicator, because of the link between MUAC and muscular mass. MUAC measurement shows few variations for children aged 6-59 months and was analysed as a malnutrition indicator for children more than 6 months.

Table3 :Cut off values for MUAC measurement defining moderate acute malnutrition and severe acute malnutrition with WHO 2006 standards (SMART Nutrition and mortality survey, Panna district, Pawaï block, WHH & Jansahas, September 2013)

Severity levels	MUAC (mm)
Mortality risk	MUAC<115
Moderate acute malnutrition	115 ≤ MUAC< 125
At risk of malnutrition	125 ≤ MUAC< 135
Normal	135 ≤ MUAC

2.6.3 Height for Age indicator

Chronic malnutrition is defined by a low height for age (stunting). Height for Age indicator, which show the height of a child regarding to his age is a malnutrition’s long term effects measurement. This indicator compares the height of the child with the average height of a reference population for the same age.

Table 4 : Cut off values for Height for Age (H/A) indicator, defining global chronic malnutrition, moderate chronic malnutrition and severe chronic malnutrition, with WHO 2006 standards (SMART Nutrition and mortality survey, Panna district, Pawaï block, WHH & Jansahas, September 2013)

Chronic malnutrition	Height / Age indicator
Global	H/A< -2 z-score
Moderate	-3 z-score <H/A≤ -2 z-score
Severe	H/A< -3 z-score

2.6.4 Weight for Age

This indicator compare the weight of the child to the median weight of a reference population for the age. It allows to determine underweight for a given age. It informs us at the same time on chronic malnutrition and acute malnutrition.

Table 5 : Cut off values for Weight for Age (W/A) indicator, defining global, moderate and severe underweight, with WHO 2006 standards (SMART Nutrition and mortality survey, Panna district, Pawaï block, WHH & Jansahas, September 2013)

Underweight	Weight for Age indicator
Global	< -2 z-score
Moderate	<-2 z-score and ≥ -3 z-score
Severe	< -3 z-score

2.6.5    Mortality rate

Crude Mortality Rate (CMR) _10000/d = n / [(N / 10 000) x d]

n = total number of deaths in the surveyed households

N = total number of persons living in the surveyed households at the time of the survey

d = number of days of the considered recall period 

Mortality rate 0-5 years_10000/d= n / [(N / 10 000) x d]

n = total number of deaths for children 0-5 years in the surveyed households

N = total number of children living in the surveyed household at the time of the survey

d = number of days or the considered recall period

2.6.6   IYCF indicators

Basic indicators, WHO 2011

Early initiation of breastfeeding: Proportion of children born in the last 24 months, who were put to the breast within one hour of birth

Exclusive breastfeeding under 6 months: Proportion of infants 0-5.9 months of age who are

fed exclusively with breastmilk

Breastfeeding at the age of 1 year: Proportion of children 12 – 15.9 months of age who are fed

breastmilk

Introduction of solid, semi solid or soft food (complementary feeding) : Proportion of infants 6-8.9 months of age who receive solid, semi-solid or soft foods.

Minimum dietary diversity:Proportion of children 6-23.9 months of age who receive foods

from 4 or more food groups.

Optional indicators, WHO 2011

Children ever breastfed: Proportion of children born in the last 23.9 months who were ever breastfed

Continued breastfeeding at 2 years: Proportion of children 20 – 23.9 months of age who are

fed breastmilk

Other IYCF information collected:

Other data collected will give information about the introduction of other food given just after birth and which one, the way the child is breastfed and the duration of breastfeeding, the weaning (reason, way of weaning the child).

2.6.7    Household Dietary Diversity Score (HDDS)

HDDS gives the information on household economic access to food.

Twelve food groups are used in the HDDS:1 - Cereals ; 2 – Roots and tubers ; 3 Vegetables ; 4-Fruits ;  5-Meat, poultry and offals ; 6- Eggs ; 7- Fish nad seafood ; 8- Pulses, legumes, seeds and nuts ; 9-Milk and milk products ; 10-Oils and greases ; 11– Sweets and honey; 12- Misceallaneous.

HDDS includes food groups that need ressources at household level, like spices, sugar and sweets, oil and greases, and drinks.

HDDS per household is obtained counting the number of food groups eaten in the last 24 hours (24h recall).

HDDS evaluate the average number of food groups eaten by households.

Calculation of the HDDS = Sum, for all the households surveyed, of the number of food groups per household/ Total number of households

HDDS also allows to evaluate the percentage of households who are eating a specific food group.

2.7Action taken in case of a malnourished child (included or not in the selected household)

Children with admission criteria for Nutrition Rehabilitation Centre (NRC) have been referred to NRC with a double referral form (in order to keep their data and to check their admission).

Table 6: admission criteria for children in Madhya Pradesh(SMART Nutrition and mortality survey, Panna district, Pawaï block, WHH & Jansahas, September 2013)

	Children aged 6-59 months
SAM children :Admission in NRC	Bilateral oedemas with W/H≥ -3 zscore OrW/H< -3 Z-scores And / or MUAC < 115 mm
MAM children:  No nutrition structure for MAM children in Madhya Pradesh	W/H < -2 and ≥ -3 or MUAC ≥ 115 et < 125 mm And no bilateral oedemas

2.8Training, Supervision and survey

2.8.1 Training

Twenty four surveyors have been recruited by Jansahas. Surveyors and supervisors from Jansahas and Welthungerhilfe have received a 6 days training: 4 days theory and 2 days practical(standardisation test and field test).

Main topics were:

1.      Definition and objectives of a survey

2.      Nutrition, malnutrition and undernutrition

3.      Nutritional status and admission criteria (nutritional indicators, W/H table, local event calendar)

4.      Anthropometric measurements

5.      Sampling (sampling methods, household selection)

6.      Questionnaires (anthropometry, mortality, dietary diversity and IYCF)

Organization of the teams on the field has been explained.

Practical sessionincluded:

- standardization test with weight, height and MUAC measurements (results in annex).

- field test for one day (in a village who will not be surveyed during the survey) where the surveyors have practiced household selection, filling up of the questionnaires, anthropometric measurements.

Results for standardization test and field test have been used to make the final teams.

Each person trained has received the surveyor’s guideline summarizing the survey methodology and procedures on the field.

2.8.2 Supervision

During the survey, teams have been surveyed by the nutrition consultant, Welhungerhilfe and Jansahas managers.

Anthropometric and mortality data have been entered by the nutrition consultant day after day on the ENA software, in order to have a daily evaluation of the data quality for each team (with the flagged data on ENA screen and the plausibility check report about digit preferences) and to give them recommendations for the next day.

IYCF and Household Dietary Diversity data have been entered at the end of the survey.

2.8.3 Survey

Survey has taken place from the 15^th of September until the 21^stof September in Pawaï block, Panna district.

Four vehicules have been used (2 teams per vehicle). Communication with the teams has been done by mobile phone, when network was available.

Team were composed as follow:

• One team leader who was in charge of the methodology, the presentation of the objectives to the chief of the village and the heads of household, and who was in charge of checking the measurements and filling up the questionnaires. In case the team leader was a male, questionnaire for IYCF was done by the female of the team.

Two measurers in charge of anthropometric measurements and responsible for the anthropometric equipment.

When necessary, in the villages surveyed, one local guide was helping the team to define the borders of the village and to indicate the households who were part or not of the village/segment selected.

Surveyors’ list is available in annex.

2.9Ethic

Villages authorities have been contacted before the survey and they were asked the authorization to do the survey.

Free and informed consent for the participation to the survey has been asked to the head of household or his representant in case he was absent.

Same procedure has been followed with the mothers, before anthropometric measurements and for the questionnaires.

2.10Data analysis

Anthropometric and mortality data have been entered in ENA software, version june 2013.

Epi Info have been used for the Chi² tests.

Individual measurements have been compared to the OMS 2006 international reference standards.

Household dietary diversity and IYCF data have been entered in Excel.

1.     ANTHROPOMETRIC RESULTS

3.1 Sample characteristics

A total of 422 children aged 6-59 months have been surveyed for the anthropometry.

Data analysis has been done excluding the FLAG SMART, in order to exclude extreme data which are more likely to be mistakes.

Finally, a total of 407 children’s data have been used for the Weight/Height analysis, 411 for the Weight/Age analysis, 398 for the Height/Age analysis and 408 for the MUAC analysis.

Table 7: Sampling characteristics (SMART Nutrition and mortality survey, Panna district, Pawaï block, WHH & Jansahas, September 2013)

	n	Percentage (%)
Total number of children 6-59 months surveyed	422
Children aged 6 - 29 months	185	43,8
Children aged 30 - 59 months	237	56,2
Gender
Girls	213	50,5
Boys	209	49,5
Ratio (m/f)	1,0

Sex ratio is 1,0: both genders are equally represented, which validate the representativeness of the sample.

Table 8: Distribution of age and sex of sample(SMART Nutrition and mortality survey, Panna district, Pawaï block, WHH & Jansahas, September 2013)

	Boys		Girls		Total		Ratio
AGE (months)	n	%	n	%	n	%	Boy:girl
6-17	51	53,1	45	46,9	96	22,7	1,1
18-29	46	51,7	43	48,3	89	21,1	1,1
30-41	54	47,8	59	52,2	113	26,8	0,9
42-53	41	47,7	45	52,3	86	20,4	0,9
54-59	17	44,7	21	55,3	38	9,0	0,8
Total	209	49,5	213	50,5	422	100,0	1,0

Figure 1: Population age and sex pyramid(SMART Nutrition and mortality survey, Panna district, Pawaï block, WHH & Jansahas, September 2013)

Age interval 6-29 is under represented for both sex.

Exact birthday has been found for only 41% of the children.

Table 9: Mean z-scores, Design Effects and excluded subjects(SMART Nutrition and mortality survey, Panna district, Pawaï block, WHH & Jansahas, September 2013)

Indicator	n	Mean z-scores ± SD	Design Effect (z-score < -2)	z-scores not available*	z-scores out of range
Weight-for-Height	407	-1,32±0,92	1,00	8	7
Weight-for-Age	411	-2,13±1,05	1,93	7	4
Height-for-Age	398	-2,12±1,18	1,70	8	16

* contains for WHZ and WAZ the children with edema.

Design effect (WHO standards) for Weight for Height was equal to 1,0.

3.2 Prevalence of acute malnutrition

Figure 2: Distribution of Weight for Height (W/H) indicator, WHO 2006 standards(SMART Nutrition and mortality survey, Panna district, Pawaï block, WHH & Jansahas, September 2013)

The red curve (children from the sample) has shifted on the left comparing to the reference population curve (green curve), indicating that the population surveyed has more malnourished children than the reference population.

Kurtosis test for Weight for Height (0,05) is less than the absolute value of 0,2, then the distribution can be considered as normal.

Table 10: Malnutrition rates, children aged 6-59 months (n=468) and children aged 6-29 months (n=228), Weight for Height z-score, WHO 2006 reference population (SMART Nutrition and mortality survey, Panna district, Pawaï block, WHH & Jansahas, September 2013)

Weight for Height in z-score or bilateral oedemas	WHO 2006
Global Acute Malnutrition 6-59 months	23,6 % (19,8 - 27,8 95% C.I.)
Severe Acute Malnutrition 6-59 months	2,9 % (1,6 - 5,5 95% C.I.)
Global Acute Malnutrition 6-29 months	25,7 % (20,4 - 31,9 95% C.I.)
Severe Acute Malnutrition 6-29 months	2,2 % (0,9 - 5,6 95% C.I.)

Oedemas’ prevalence is equal to 0,0% (n=0).

GAM prevalence (23,6%, WHO 2006) show a critical situation in the Pawaï block, according to the WHO classification cut-off[1].

Table 11: Prevalence of acute malnutrition by age, based on weight-for-height z-scores and/or oedemas(SMART Nutrition and mortality survey, Panna district, Pawaï block, WHH & Jansahas, September 2013)

---

[1]Cut-off defined by WHO:

§   Acceptable situation : <5%

§   Precarious situation : 5 to 9%

§   Sérieuse situation : 9 to 14%

§   Critical situation : ≥ 15 %

		Severe wasting (<-3 z-score)		Moderate wasting (>= -3 and <-2 z-score )		Normal (> = -2 z score)		Oedema
Age (months)	Total N	n	%	n	%	n	%	n	%
6-17	93	2	2,2	22	23,7	69	74,2	0	0,0
18-29	86	2	2,3	20	23,3	64	74,4	0	0,0
30-41	106	5	4,7	10	9,4	91	85,8	0	0,0
42-53	85	3	3,5	21	24,7	61	71,8	0	0,0
54-59	37	0	0,0	11	29,7	26	70,3	0	0,0
Total	407	12	2,9	84	20,6	311	76,4	0	0,0

3.3 MUAC distribution

MUAC is a good indicator for the mortality risk linked with malnutrition.

Table 12: Prevalence of acute malnutrition based on MUAC cut off's (and/or oedemas) and by sex(SMART Nutrition and mortality survey, Panna district, Pawaï block, WHH & Jansahas, September 2013)

	All n = 408	Boys n = 202	Girls n = 206
Prevalence of global malnutrition (< 125 mm and/or oedema)	(38) 9,3 % (6,4 - 13,4 95% C.I.)	(13) 6,4 % (3,1 - 12,9 95% C.I.)	(25) 12,1 % (8,0 - 17,9 95% C.I.)
Prevalence of moderate malnutrition (< 125 mm and >= 115 mm, no oedema)	(28) 6,9 % (4,3 - 10,8 95% C.I.)	(11) 5,4 % (2,4 - 11,7 95% C.I.)	(17) 8,3 % (4,8 - 13,7 95% C.I.)
Prevalence of severe malnutrition (< 115 mm and/or oedema)	(10) 2,5 % (1,3 - 4,6 95% C.I.)	(2) 1,0 % (0,2 - 4,0 95% C.I.)	(8) 3,9 % (2,0 - 7,5 95% C.I.)

Table 13: Prevalence of acute malnutrition by age, based on MUAC cut off's and/or oedemas(SMART Nutrition and mortality survey, Panna district, Pawaï block, WHH & Jansahas, September 2013)

		Severe wasting (< 115 mm)		Moderate wasting (>= 115 mm and < 125 mm)		Normal (> = 125 mm )		Oedema
Age (months)	Total N	n	%	n	%	n	%	n	%
6-17	90	4	4,4	9	10,0	77	85,6	0	0,0
18-29	86	3	3,5	7	8,1	76	88,4	0	0,0
30-41	110	2	1,8	8	7,3	100	90,9	0	0,0
42-53	85	0	0,0	2	2,4	83	97,6	0	0,0
54-59	37	1	2,7	2	5,4	34	91,9	0	0,0
Total	408	10	2,5	28	6,9	370	90,7	0	0,0

3.4 Prevalence of stunting

               

Exact birthday has been found for only 41% of the children. Despite the systematic use of the local calendar of events, age has to be used with caution, especially regarding estimating of chronic malnutrition.

Table 14: Prevalence of stunting based on height-for-age z-scores and by sex(SMART Nutrition and mortality survey, Panna district, Pawaï block, WHH & Jansahas, September 2013)

	All n = 398	Boys n = 195	Girls n = 203
Prevalence of stunting (<-2 z-score)	(216) 54,3 % (47,7 - 60,7 95% C.I.)	(113) 57,9 % (48,4 - 67,0 95% C.I.)	(103) 50,7 % (43,1 - 58,3 95% C.I.)
Prevalence of moderate stunting (<-2 z-score and >=-3 z-score)	(125) 31,4 % (26,3 - 37,0 95% C.I.)	(68) 34,9 % (26,9 - 43,8 95% C.I.)	(57) 28,1 % (21,7 - 35,5 95% C.I.)
Prevalence of severe stunting (<-3 z-score)	(91) 22,9 % (18,6 - 27,8 95% C.I.)	(45) 23,1 % (17,3 - 30,1 95% C.I.)	(46) 22,7 % (17,3 - 29,0 95% C.I.)

3.5Prevalence of underweight

Table 15: Prevalence of underweight based on weight-for-age z-scores by sex(SMART Nutrition and mortality survey, Panna district, Pawaï block, WHH & Jansahas, September 2013)

	All n = 411	Boys n = 201	Girls n = 210
Prevalence of underweight (<-2 z-score)	(222) 54,0 % (47,1 - 60,8 95% C.I.)	(103) 51,2 % (43,1 - 59,4 95% C.I.)	(119) 56,7 % (47,9 - 65,0 95% C.I.)
Prevalence of moderate underweight (<-2 z-score and >=-3 z-score)	(138) 33,6 % (27,7 - 40,0 95% C.I.)	(62) 30,8 % (24,0 - 38,6 95% C.I.)	(76) 36,2 % (28,6 - 44,6 95% C.I.)
Prevalence of severe underweight (<-3 z-score)	(84) 20,4 % (16,4 - 25,2 95% C.I.)	(41) 20,4 % (16,0 - 25,7 95% C.I.)	(43) 20,5 % (14,7 - 27,7 95% C.I.)

3.6Retrospective mortality (90 days recall period)

§  Crude mortality rate (90 days recall period) : 0,63/10 000/day (IC 0,41-0,97)

§  Mortality rate children under 5 years old (90days recall period):0,96/10 000/day (IC : 0,36-2,55)

Both rates are above the alert cut-off[1].

---

[1]CMR=1 death/10 000 persons/day ; MR 0-5=2 deaths/10 000 persons/day

4.      IYCF RESULTS

Table 16 : IYCF results(SMART Nutrition and mortality survey, Panna district, Pawaï block, WHH & Jansahas, September 2013)

Indicator		n	%
Early breastfeeding (2 missing data, N=163)	First hour of life	47	28,8
	Less than 24 hours	75	46,0
	More than 24hours	39	23,9
	Do not know	0	0,0
Children who have received the colostrum (8 missing data, N=163)		87	53,4
Exclusive breastfeeding before the age of 6 months (children 0-5 months: N=45 children)	Children less than 6 months, who did not received other food in the 3 days after birth, did not eat any liquidi or solid/semi solid food in the last 24 hours	6	13,3
Introduction of food other than breastmilk in the 3 days following birth (N=163 children breastfed), children 0-23 months, 9 missing data	Yes	39	23,9
	No	114	69,9
Type of fluids given in the 3 days following birth, for children 0-23 months (N=39)	Water / Hot water	4	10,3
	Sugar water	0	0,0
	Tea	5	12,8
	Honey	1	3,6
	Jaggery water	4	10,3
	Ghutti	0	0,0
	Animal milk powder or fresh animal milk	13	33,3
	Infant formula	No data
	Do not know	1	3,6
	Other	0	0,0
Way of breastfeeding  (N=147  children breastfed on the day of the survey)	Day only	1	0,7
	Night only	0	0,0
	Day and night at specific time	33	22,4
	On demand	51	34,7
Breastfeeding at the age of 1 year (children  12-15 months: n=42; 3 missing data)		36	85,7
Breastfeeding at the age of 2 years (children 20-23 months: n=21)		18	85,7
Introduction of solid, semi solid or slack food timely (children 6-8 months, n=15, 3 missing data)	Yes	10	66,7
	No	2	13,3
IDDS (Individual dietary diversity), n=116 children 6-23 months	Children 6-23 months who has eaten food from at least 4 different food groups	58	50,0
Average breastfeeding time (n=5 children not breastfed anymore on the day of the survey, 1 missing data)		9,9 months
Way of weaning the child (number of children 12-23 months who are not breastfed anymore on the day of the survey n=10, 5 missing data	Progressively	2	20,0
	Abruptely	3	30,0
	DNK	0	0,0
Most frequent reasons for weaning, n=5 children not breastfed on the day of the survey, 1 missing data	New pregnancy	1	20,0
	Not anough milk anymore	1	20,0
	Sick child	1	20,0

5. HOUSEHOLD DIETARY DIVERSITY RESULTS

94,5% of surveyed households are eating at leat 4 food groups (equal or greater that 4 groups).

1.     DISCUSSION/CONCLUSION

The main objective of the SMART survey had to evaluate the nutritional status of children 6-59 months in Pawaï block, Panna district.

Data quality                                                                                       

Sex ratio, SD and design effect

§  Collected data show a balanced gender repartition as sex ratio is equal to 1,0, which is in the norm (between 0,8 and 1,2).

§  SD for Weight for Height (0,93) and for Height for Age (1,19) are between 0,8 and 1,2 (WHO 2006 reference), which indicate a normal distribution for the sample.

§  Skewness tests for Weight for Height (0,04) and Height for Age (0,05) give results between -0,2 and 0,2, then the distributions for these indicators can be considered as symmetrical.

§  Kurtosis test for Weight for Height (0,05) and for Height for Age (-0,46) is less than the absolute value of 0,2, then the distribution can be considered as normal.

§  Design effect is 1,0, less than the planned design effect (1,5) which indicate an homogenous population.

Absent households & reserve cluster

Data collection had been done in September, when households have started harvesting. It often happened that households were absents, working in their fields and coming back at the end of the day, when the teams have left.

Thus, the five reserve clusters has been surveyed, in order to reach the number of children required in the ENA planning phase.

Nutrition situation

Acute malnutrition

This survey has been done at the end of the hunger peak.

With a GAM rate of 23,6% (19,8 - 27,8 95% C.I.) with WHO standards, the situation in the surveyed area is critical, according to the WHO cut-off.

Moreover, it has to be underlined that prevalence represent an instant picture of the nutrition situation at the time of the survey. This indicator can not explain or predict the evolution of the situation.

Thus, it would be possible than in some areas surveyed, the nutrition situation become worth : because of important flooding during the rainy season, some crops (soja beans) have been spoiled.

Chronic malnutrition

Chronic malnutrition affects 54,3% of the children aged 6-59 months (47,7 - 60,7 95% C.I.), 22,9% of severe forms (18,6 - 27,8 95% C.I.) (WHO 2006).The global chronic malnutrition rate is above than the WHO alert cut off of 40%.

Results for children 0-5 months show that stunting probably starts during the pregnancy, as 34,7% (21,5-50,7 95% CI) of children 0-5 months are stunted with 12,2% (5,5-25,2 95% CI) of severe stunting.

Malnutrition and risk factors

·         6-29 months vs 30-59 months

X² testdid not show any significative difference between children aged 6-29 months and children 30-59 months regarding GAM and SAM (p<0,05 in both cases).

·         Girls vs Boys

X² test did not show any significative difference between grils and boys regarding GAM or SAM (p>0,05 in both cases).

There is no significant difference between girls and boys regarding stunting (global or severe) (p>0,05).

Infant and Young Child Feeding (IYCF)

According to WHO recommendations, the type of diet to ensure optimal health and proper growth of the child is exclusive breastfeeding for the first 6months of life, then a sufficient and safe complementary foods accompanying the breastfeeding up to 24 months or more.

In fact, breast milk no longer meets 60 to 80% of needs between 6 and 11 months and only 35 to 40% of needs between 12 and 23 months.The transition from breastfeeding(exclusive)to the consumption of family foods is a very delicate phase for the infant.During this period, many children suffer from malnutrition more or less associated with infectious episodes which contribute significantly to morbidity and mortality in children under 5 years.

Among children aged 0 to 23 months surveyed, 28,8% of them were early breastfed, in the first hour of life. Regarding the colostrum, 53,4 % of mothers reported that the child has received;23,9% of children received another liquid within three days of birth.

This, coupled with the fact that complementary feeding is often introduced before the age of six months, gives a rate of 13,3% of the children who are exclusively breastfed up to 6 months. It is possible that the exclusive breastfeeding rate is lower, because question regarding exclusive breastfeeding was asked in a way to avoid stereotypical responses, with other questions like introduction of other fluids in the days following birth, consumption of other foods or liquids within 24 hours preceding the survey. This does not give us information on the period from 3 days after birth child before and 24 hours before the survey day.

Breasfeeding is on demand for 34,7% of the children, and 85,7% of the children are breastfed until the age of two years.

Breastfeeding is generaly stopped abruptely (30,0% of the children); the three reasons were a new pregnancy, a sick child, or not enough milk anymore.

Note: for some indicators, data were very few (n<30).

Household dietary diversityscore (HDDS) and Individual Dietary Diversity Score 6-23 months (IDDS)

Results show that household have access to a food mainly based on cereals and pulses.

There is a big percentage of households eating milk products (67,9% of the households), but most often milk is taken in small quantity, like sugar, with the tea.

Most of the vegetable eaten (41,3%) at this season were cucumbers.

Regarding the HDDS, 94,5% of surveyed households are eating at leat 4 food groups (equal or greater that 4 groups).

For children 6-23 months, 50,0% of them are eating at least 4 food groups.

Although households surveyed for IYCF and those surveyed for dietary diversity/food security are not necessarily the same,it appears that the HDDS is significantly higher than IDDS (ԑ>1,96).

Dietary diversity that household scan access may not be fully exploited to provide a more diverse diet to children 6 to 23 months: the problem of food diversification for children 6-23 months may not lie necessarily on household's accessto foodbut also on infant feedingpractices.

RECOMMENDATIONS

·         To do a nutrition surveillance on a regular basis;

·         To implement nutrition education sessions in the community, on a regular basis, in order to sensitize the community on undernutrition, the causes and the treatment; 

·         To cure the moderate acute malnourished (MAM) children in order to prevent these children from severe acute malnutrition.

·         To implement IYCF sensitization sessions in the community;

To implement dietary diversity education sessions for the households to improve dietary diversity of their children.