Conference place – Addis Ababa By: Tesfaye Hailu (EPHI)

Effectiveness of Chickpea based Ready-to-use-Supplementary Feeds, Super Cereal, and Super Cereal PLUS for Management of

Moderate Acute Malnutrition (MAM)

A cluster randomize controlled trial involving Ethiopian children aged 6–59 months with MAM

Content

• Introduction

• Method

• Result and Discussion

• Conclusion

• Recommendation

• Acknowledgement

Introduction• Globally MAM affects a total of 11% of children

under the age of 5 years.

• In Ethiopia, overall 12.5% of under-five children arewasted and 9.7% are moderately wasted.

• Moderately wasted children are typically givenfortified corn-soy blended flour (CSB).

• However, CSB has been reported to be nutritionalinappropriate and inadequate in treating of acutemoderate malnutrition in children

Introduction

• In a study conducted in Ethiopia in 2012 byTsinuel, Mekitie, and Fessahaye, a recovery rateof 64.4% was recorded which was below theSPHERE standards of 75% for supplementaryfeeding programmes.

• RUSF on the other hand have been indicated foruse in supplementary feeding programs as atreatment for moderate acute malnutrition, andan alternative to fortified blended flour (CSB).

Objectives

General• To evaluate the effectiveness in terms of nutritional

outcomes and cost of Chickpea Based Ready-to-use-Supplementary Feeds (C-RUSF and CMS-RUSF), andSuper Cereal PLUS(CSB++) as compared to SuperCereal( CSB+) for the Management of MAM inEthiopia

Specific objectives• To determine the recovery rates of children with moderate acute

malnutrition (MAM)

• To determine the rate of weight gain of children with moderate acutemalnutrition ( MAM)

• To determine the time of recovery of children with moderate acutemalnutrition ( MAM)

• To assess programme performance rates for each intervention arm interms of rates of recovery, transfer, default, death and non-responsecomparing them against the SPHERE standards;

• To explore the knowledge, attitude and practices of caregivers for eachof the four products including preparation and sharing;

• Explore the beneficiaries’ and their caregivers’ perceptions of eachcommodity; and

• To determine and compare cost of the products in terms of cost perbeneficiary, product, storage and transport costs; length of treatment andtime required for recovery of children

Methodology• Cluster randomized controlled effectiveness trial• Amhara, Oromia, SNNPR and Tigray using the 4 study

products– High prevalence of MAM compared to other regions;– High number of OTP sites; and– Food insecurity hotspot 2 prioritization.

• From each woreda,– Two health centres were randomly selected based on

Woreda Health Office (WOHO) report on the incidence ofchild malnutrition.• For each of the selected Health Centres, four Health

Posts (clusters) were selected based on the number ofmalnourished children identified in the 3 months priorto the study.

Regions and Respective Zones and Woredas for the Study

Region Zone Selected Woreda Health CentersNumber of Health

Posts (Clusters)

Amhara

North Wollo Ambasel/Tenta Mariye&Fito 8

North Wollo Kutaber Kutaber&Kundi 8

North Wollo Sayint Adjibar&Euwa 8

OromiyaWest Arsi Hitossa Itaya&Welarge 8

West Arsi ZewyDugda Ogalcho&Kiyansho 8

SNNPR

Gedeo Wonago Mekonisa&Hasseharo 8

Gedeo DillaZuriya Tuticha&Uddo 12

Gedeo Kochore Beloya&Chelaliktu 8

Tigray

Southern Ofla Hashengie&Fala 8

Southern Endamohonie Meswaeti&Neksegie 8

Southern EmbaAlaje Bora &Silawa/Adishu 12

Total 4 11 22 96

Methodology

• Enrolment of the study participants was based on:

– Mid Upper Arm Circumference (MUAC) ofbetween greater than or equal to 11 cm and lessthan 12.5cm with no bilateral oedema;

• Therefore the primary indicator used foradmission and discharge was MUAC 11-12.5cm.

MethodologyInclusion• All children aged 6 to 59 months of age identified as moderately

malnourished based on MUAC screening measurement of (> 11 - <12) and/or WHZ of <-2 and > -3 without bipedal pitting oedema.

Exclusion• Children identified as severely malnourished based on MUAC

screening <11cm and/or bilateral oedema;• Children with any illness or clinical condition that would have

prevented them from safely ingesting the study food. Children weretherefore assessed upon enrolment for any complicated clinicalcondition such as oedema, malaria, ARI, vomiting, diarrhoea(defined as 3 or more watery stools in 24hours), and lack ofappetite that would require medical attention;

• Chronic debilitating illness including congenital or acquiredmalformations;

• Children in Outpatient Therapeutic Programme; and• Children already enrolled and participating in the TSF programme.

Sample Size

• Formula of calculating sample size was:

• n = [(Zα + Zβ)2 × {(p1 (1-p1) + (p2 (1-p2)}]/(p1 - p2)2

Where:

• n = sample size required in each group;

• p1 = proportion of subject recovered by SC/oil = 0.644;

• p2 = proportion of subject cured by RUSF = 0.724;

• p1-p2 = clinically significant difference = 0.08;

• Zα: Level of significance, for 5% which is 1.64; and

• Zβ: Power, for 80% which is 0.84

• 2,425 children aged 6 to 59months (606/ product)

Methodology

Overall Region Woreda Kebele

(Clusters)

Sample size for

children

2,425 606 152 25

Study sites 4 12 96

Formula and Nutritive Value for the Chickpea Based RUSF

Supplementary Feed Chickpea Chickpea + maize + soy

Proportion of Ingredients (%) per 100gms

Chickpeas 32.8 31.3

Maize - 16.1

Soy Flour - 8

Soy oil 9 16.9

Palmolein oil 19 7.8

Dry Skim Milk 20 7.9

Sugar 15 9.9

Premix 1.75 1.75

Nutritional Vale

Energy (Kcal) 521 507

Protein (g) 15 13.7

Fat (g) 32.3 31.3

Iron (g) 13.6 14.3

Vitamin A 2,061.2 2060.5

Cost per Metric Tonne

Cost per Metric Tonne (USD) 2,317.3 1,617.3

Composition of Super Cereal and Super

Cereal PLUS

IngredientsValues (%)

Super Cereal PLUS Super Cereal

Corn 58.24 78.24

Soybean (de-hulled) 20 20

Soybean oil 3 -

Dry Skimmed Milk 8 -

Sugar 9 -

Vitamin/mineral mix 0.2 0.2

Ca (H2PO4)2. Water 0.8 0.8

Potassium Chloride 0.76 0.76

Nutritional Composition of Super Cereal

and Super Cereal PLUS

ParameterValues

Super Cereal PLUS Super Cereal

Protein (%) MinVegetable source 13.5 14Animal source 2.5 -

Fat (%) Min 9 6Fibre (%) Max 2.9 5Moisture (%) Max 7 10Aflatoxin MAX 5ppb 20ppdBostwick mm/30 sec(450C)

100mm (50g/250g water) 90Kcal/100ml

porridge

55mm (40g/250g water) 60Kcal/100ml

porridge

Approximate cost perMetric Ton

$1,280 $1,657 (Including oil)

Amount of food required per study subject per product

Treatment

food

Amount per

Day

Amount per

week

Amount for 12

weeks per child

Total for 12 weeks for

the entire study

CMS-RUSF1 sachet

(100gms)

7 sachets

(100gms)

84 sachets

(100gms)

74,592 sachets

(100gms)

C-RUSF1 sachet

(100gms)

7 sachets

(100gms)

84 sachets

(100gms)

74,592 sachets

(100gms)

Super CerealSC (200gms)

and Oil (33mls)

SC (1.4kgs) and

Oil (231mls)

SC (18kgs) and

Oil (19.4lts)

14,919kgs

(1.5tonnes)SC

17.227lts Oil

Super Cereal-

PLUS200gms 1.5kgs 16.8kgs

15,984kgs (1.6tonnes)

Methodology

• All the required information from the individual child was recordedon the Individual questionnaire and subsequently TSF IndividualCard

• The variables assessed included: – Age: – Sex: – Weight:– Height/Length:– Mid-Upper Arm Circumference (MUAC): – Bilateral oedema:– Vaccination in children 6-59months– Child already enrolled in selective feeding programme– Vitamin A supplementation in last 6months– Morbidity in last 2weeks

Methodology

• MAM received the routine medications as perthe national protocol

– Anthelminthic (Albendazole/Mebendazole,

– Vitamin A and

– Routine immunization

• Dietary treatment, nutrition counseling,follow-up (weekly visit) and exit

Methodology

• Data Collection Methods– MUAC Screening Tool

– Referral slips

– Individual questionnaire

– TSF Individual treatment card

– TSF Ration card

– TSF Register

– Home visit tool

– Site Tally Sheet

– Monthly region reporting form

• Qualitative Data collection – FGD

Study Team size

National Regional Woreda Kebele

Study number 4 12 96

Study team

Principal

Investigator

1 - - -

Coordinator(s) - 10 - -

Supervisor(s) - - 12 -

Enumerators - - - 96

Statisticians 2

Total 3 10 12 96

Methodology

• Training and Pre-testing

• Consent

• Ethical Clearance - Scientific and Ethical Review Office (SERO) of EPHI

Result and Discussion

Variable Recovery rate by product type

SC+ SC Formula-1 Formula-2Recovery rate by sex

Male 42.9 37.2 31.1 34.3

Female 41.3 32.7 33.9 33.0

Recovery rate by age 6 - 11 Months 52.5 36.4 37.7 28.7

12 - 23 Months 40.2 36.9 34.3 32.6

24 - 35 Months 36.4% 33.3% 28.7% 32.6%

36 - 59 Months 41.2% 27.8% 28.1% 22.7%

Total 42.0% 34.7% 32.6% 33.5%

Table 1: distribution of recovered children by sex and age

Result and Discussion

Variable Type of products

SC+ SC Formula

1

Formula

2

Mean time of

recovery in week

7.87 8.02 7.76 7.65

95% CI Lower

bound

7.48 7.59 7.38 7.21

Upper

bound

8.25 8.45 8.15 7.69

Table 2: Mean time of recovery by product

Result and Discussion

Follow

up

Weeks

Product Type

SC+ SC Formula-1 Formula-2

N

Mean Wt.

gain/kg N

Mean Wt.

gain/kg N

Mean Wt.

gain/kg N

Mean Wt.

gain/kg

Week1 589 .20 554 .14 517 .14 552 .16

Week2 590 .32 542 .25 510 .24 536 .26

Week3 591 .40 543 .31 505 .33 534 .36

Week4 590 .51 540 .37 491 .42 524 .43

Week5 602 .56 528 .43 480 .49 508 .50

Week6 584 .63 540 .51 470 .56 512 .57

Week7 574 .68 523 .57 459 .64 486 .64

Week8 567 .75 510 .61 458 .68 464 .69

Week9 563 .79 492 .65 436 .75 472 .75

Week10 542 .85 479 .72 431 .80 457 .80

Week11 533 .88 485 .79 425 .84 434 .85

Week12 517 .95 485 .87 419 .89 434 .89

Table 3. Average change of anthropometric measurements (weight, height and MUAC) by products

Percentages of recovery by recovery time and product type

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

35.0%

40.0%

45.0%

1 2 3 4 5 6 7 8 9 10 11 12

CSB++

CSB+

Formula-1

Formula-2

Total

Outcomes % (95% CI) Product Type Total

SC+ SC Formula-1 Formula-2

Recovery N 618 580 554 579 2331

Percent (%) 42.1 34.7 32.7 33.5 35.9

95% CI [38.2,46.0] [30.9, 38.6] [28.9, 36.7] [29.8, 37.5] [33.9, 37.8]

Developme

nt of SAM

N 609 571 551 576 2307

Percent (%) 9.5 6.0 6.0 9.4 7.8

95% CI [7.4, 12.1] [4.3, 8.2] [4.3, 8.3] [7.2, 12.0] [6.7, 8.9]

Non-

Recovery

N 618 580 554 579 2331

Percent (%) 57.9 65.3 67.3 66.5 64.1

95% CI [54.0, 61.8] [61.4, 69.1] [63.3, 71.1] [62.5, 70.2] [62.2, 66.1]

Death N 2344 2344 2344 2344 2344

Percent (%) 0.0 0.0 0.0 0.0 0.0

Default N 621 582 557 584 2344

Percent (%) 1.3 1.7 4.7 4.5 3.0

95% CI [0.7, 2.6] [0.9, 3.2] [3.2, 6.8] [3.0, 6.5] [2.4, 3.8]

Program performance indicator by product type

33%

13%26%

28%

household Foodsecured

household Mild foodinsecure

householdModerately foodinsecure

household Severelyfood insecure

Household food security Status N=2010

Estimated cost per each arm

List of products Cohort

Product cost

of recovery

per child

Cost

Product 1 (SC+) 606 16.23 9835.09Product 2 (SC and

Oil) 606 10.50 6365.16Product 3

(Formula-1) 606 15.77 9559.36Product 4

(Formula-2) 606 12.22 7405.39

Limitation

• Withdrawal of Somali region due to unable to find enough eligible MAM children

• One of the inclusion criteria to select woreda was no TSF program currently exists in the area. The study woreda don’t have a system of implementing TSF program and behavioral change communication is require longer time than what was given during the study period (3 months).

• The study was done during the huger season. Hence, this condition increases the probability of sharing and not providing additional food for the study subjects.

Conclusion • The mean MUAC of the children ranges from 11.6-11.7 cm with an average

of 11.6 cm. The average weight of the children is 7.9 kg

• There is no significant difference between the mean weight, MUAC of thechildren across the products (P > 0.05).

• Statistically significant recovery rate (42%) observed in Super-cereal plus(SC+) than the other three products; SC, formula 1 and formula 2.

• There is no statistically significant difference on mean weight changeamong the products.

• The more vulnerable age group for moderately acute malnutrition is thehighest probability to react for treatment timely.

• Following the recommended ICER procedure for cost effectiveness, SC andSC+ were strongly dominated and eliminated along the step-wise costeffectiveness ratio calculation. Finally, the ICER for Formula 2 was very lowas compared to formula 1.

Recommendation

• The study revealed that, SC+ has statisticallysignificant higher rate of recovery comparedwith SC, formula 1 and 2. However, there is nosignificant difference among SC, formula 1 and2 in terms of recovery rate.

• Further study among the three products (SC, formula 1 and 2) under control setup like animal trial is recommended.

Acknowledgement

• WFP

• EPHI staff

• Study participants

• Data collectors and supervisors

Thank you