january 2019 issue 59 issn 1743-5080 (print) · south asia exceeds all other regions, yet the...

January 2019 Issue 59

Scaling up high impact nutritioninterventions in Nairobi

Improving inpatient SAM care in IndiaScaling up nutrition services in

north-eastern NigeriaMulti-sector nutrition programming

in Bangladesh

ISSN 1743-5080 (print)

Field Exchange issue 59, January 2019, www.ennonline.net/fex

Contents...............................................................

1 EditorialField Articles2 Scale up of high-impact nutrition interventions in the informal settlements of Nairobi, Kenya

9 Institutionalising quality of care in inpatient facilities for the management of severe acute malnutrition in India

43 Nutritional response in north-eastern Nigeria: Approaches to increase service availability in Borno and Yobe States

54 A multi-sector approach to improve nutrition: Experiences of the Nutrition at the Center project, Bangladesh

News6 Report of the Global Nutrition Cluster side event on Yemen and Sudan

6 UNICEF programme guidance on adolescents

7 The State of the Humanitarian System 2018 report

8 Wasting in South Asia: Consultation on building the evidence base on the policy and programme response

Research 13 Snapshots 13 Linear growth following complicated severe malnutrition

13 Use of MUAC by novel community platforms to detect, diagnose and treat severe acute malnutrition

14 Post-discharge follow-up of children treated for severe acute malnutrition

14 Post-discharge morbidities and mortalities among children with severe acute malnutrition who did not undergo nutrition rehabilitation

14 Cash-based intervention and risk of acute malnutrition among children in internally displaced persons camps in Somalia

15 Concurrent wasting and stunting among under-five children in Senegal

15 Improving screening for malnourished children at high risk of death

16 Risk factors for acutely malnourished infants aged under six months

16 Outcomes and risk factors for infants under six months old with severe acute malnutrition in Bangladesh

16 Short malnourished children and fat accumulation with food supplementation

17 A review of wet nursing experiences, motivations, facilitators and barriers

17 Gender dynamics of phone ownership and use in a Fresh Food Voucher scheme in Ethiopia

18 Barrier analysis of infant and young child feeding and maternal nutrition behaviours among IDPs in northern and southern Syria

21 Micronutrient powder distribution strategies to increase coverage and adherence among children aged six to 23 months as part of an IYCF strategy in Cambodia

24 Development of a maternal service package for mothers of children with severe acute malnutrition admitted to nutrition rehabilitation centres in India

29 The current state of evidence and thinking on wasting prevention

30 Can low-literate community health workers treat severe acute malnutrition? A study of simplified algorithm and tools in South Sudan

32 Is there a systematic bias in estimates of programme coverage returned by SQUEAC coverage assessments?

36 Infant and young child feeding in emergencies: An analysis of key factors of a strong response

37 Higher heights: A greater ambition for maternal and child nutrition in South Asia

40 Independent and combined effects of improved WASH and improved complementary feeding on child stunting and anaemia in rural Zimbabwe

41 Undersized Indian children: Nutrients-starved or hungry for development?

Evaluation47 Impact evaluation of WASH in nutrition intervention on morbidity and acute malnutrition in Niger

51 Nutrition-sensitive outcomes of a permaculture project in Nepal

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Shad

rak

Wer

ungu

A community health volunteer counselling a group of womenon appropriate complementary feeding, Nairobi, 2017

Farmers meet to implement theirpermaculture design skills andexchange labour, Surkhet, Nepal, 2016

Chris

Eva

ns

1

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Field Exchange issue 59, January 2019, www.ennonline.net/fex

EditorialDear readers,

This 59th issue of Field Exchange coversan interesting range of subjects. Com-mon themes include programme inte-gration, nutrition-sensitive programming

and piloting new programme designs. An articleby Kassim Lupao and Esther Mogusu describesefforts by Concern Worldwide and UNICEF toscale up and integrate high-impact nutritioninterventions (HINI) in health systems in urbanNairobi. While successful in many respects, keychallenges have included short-term fundingand weak systems into which HINI can be inte-grated. The authors identify a range of urban-specific challenges and learning.

A second article reports on UNICEF-supportedintegration of community-based managementof acute malnutrition (CMAM) into fixed healthfacilities in north east Nigeria during a humani-tarian crisis. The success of this programme isattributed to strong government leadership anddedicated funding, good communications, useof existing community platforms for communitymobilisation, and investment in on-the-job train-ing and supervision. The authors conclude thatit is possible to transition from an emergency toa more developmental approach using human-itarian funding and internal UNICEF resources.

An article from CARE Bangladesh describes asub-district multi-sector approach to addressingmalnutrition that uses existing structures to es-tablish community-based multi-sector coordi-nation platforms (committees). These helpedcoordinate a range of nutrition-specific and nu-trition-sensitive activities, including maternal,infant and young child nutrition (MIYCN); water,sanitation and hygiene (WASH); folic acid sup-plementation; vouchers for health service access;and women’s empowerment. The article claimssignificant impact on prevalence of wastingand stunting based on baseline and end-linecomparisons, so it should be interpreted cau-tiously. Nevertheless, the cost for the approachat US$1,000 per year per committee seems tosuggest a sustainable model.

Field Exchange 59 reports on three other inter-esting pilot studies. The first is an Action AgainstHunger programme in Rajasthan, India, whichprovides psychosocial counselling for mothersof severely malnourished children, as well asWASH and educative play. An evaluation foundthat the programme led to early identificationof malnourished children, an increase in meanweight gain of children, reduction in defaulterrates and greater post-discharge attendance.Although not expensive, the approach did

require significant staff capacity building. Thefindings were not statistically valid, so againcautious interpretation is needed. However, theexperience leads the authors to conclude thatthere is potential to improve severe acute mal-nutrition treatment outcomes through staff ca-pacity and infrastructure development to supportoperationalising government guidelines.

A second pilot programme in the Himalayas ofNepal involved building the capacity of farmersto undertake permaculture farming. Althoughthe programme had no explicit nutrition ob-jectives, an evaluation found higher yields andresulting income, as well as increases in dietarydiversity. The author concludes that there isstrong potential for positive nutrition impact. The final article on integration by CooperazioneInternazionale describes a pilot study in Nigertargeting WASH to households and communitiesof severely malnourished children enrolled inoutpatient therapeutic programmes. The studyfound no impact on programme performance,but some impact on non-response and diarrhoeaco-morbidity.

This edition also includes two articles lookingat infant and young child feeding (IYCF) andmaternal nutrition and health. The first is abarrier analysis conducted among internallydisplaced persons in camps and urban popula-tions in northern Syria, where IYCF practiceshave remained poor. The analysis concludesthat, in order to make progress on IYCF, betteraccess to maternal nutrition services and IYCFsupport is needed, as well as integration ofIYCF support into other sectors, including re-productive health, food security and agriculture. An article by Vani Sethi, Praveen Kumar andArjan De Wagt in India reports on the circum-stances of mothers of children admitted to fournutrition rehabilitation centres. They found a

quarter of mothers were stunted, 23% wereunderweight and more than a quarter wereoverweight/obese. Morbidity was common andfamily planning was low. These findings catalysedthe development of a maternal service packageintegrated within existing government services.This experience spotlights a missed opportunityto provide services for mothers of malnourishedchildren and the need to engage across thehealth and reproductive health sectors (includingfamily planning) in service delivery.

These articles reflect the considerable effort,from different corners of the world, to integrateand work with existing systems and servicesand involve multiple sectors in order to maximiseopportunities for impact. Capturing impact inoperational settings remains an ongoing chal-lenge, with the need to pay careful attention tothe quality of evidence on impact. This continuesto be a limiting factor to knowing if we aretruly moving in the right direction.

Finally, you will notice that we have a shorterprint edition than usual, with fewer articles andsnappier research summaries. This reflects ourNew Year’s resolution to deliver a more curated,lighter print edition and make the most of ouronline facility (www.ennonline.net/fex) to shareyour experiences quickly and widely. Furtherdevelopments are being planned through 2019to maximise our impact and effectiveness, in-formed by our readers and contributors. Theeditorial team welcomes feedback at any time(please see contacts below).

Happy digested reading,

Jeremy Shoham & Marie McGrath, Field Exchange Co-editors

Please send feedback to Chloe Angood, FieldExchange Sub-editor: [email protected]

WFP

/Tob

in Jo

nes

Cash based interventions in Somalia

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2Field Exchange issue 59, January 2019, www.ennonline.net/fex

Field Article .....................................................

By Kassim Lupao and Esther Mogusu

Kassim Lupao is a seniormanager for the Nairobiurban health andnutrition programme atConcern Worldwide. Hehas many years ofexperience in the

implementation of high-impact nutritioninterventions in urban contexts, is a trainer oftrainers in nutrition programming, and hashelped develop Nairobi county nutritionassessments, strategies and action plans.

Esther Mogusu is thecounty nutritioncoordinator for NairobiCity County. She has over15 years of experience inimplementing nutritionprogrammes in Kenya at

health facility, sub-county and county levels.She is a mentor and trainer of trainers innutrition programming.

The funding to support scale up of high-impactnutrition interventions in Nairobi was providedby UNICEF.

Location: Kenya What we know: Scale-up of high-impact nutrition interventions (HINI) isnecessary to comprehensively address malnutrition.

What this article adds: In Nairobi county wasting prevalence in urban informalsettlements is normal but caseload is high; stunting, micronutrient deficienciesand obesity are also prevalent. Concern Worldwide implemented a five-yearprogramme of support to government to scale up HINI through the health systemin Nairobi. Activities included strategic policy development, nutrition assessmentand analysis, health worker nutrition capacity development and development ofan urban early warning system and emergency nutrition response mechanism.This has contributed to improved coverage of HINIs and acute malnutritionprogramming and reduced prevalence in stunting, wasting and underweight.Challenges to HINI scale-up include limited and short-term funding, shortfalls incommunity nutrition capacity, insufficient government budgetary allocation andpredominance of private health care. This successful partnership will continueto 2022, prioritising stunting impact. More broadly, new thresholds and ways ofworking are needed for urban emergency response programming.

Background Nairobi informal settlements Over 60% of the urban population in Nairobilive in informal settlements on 5% of the land(Nairobi Urban Sector Profile, 2006). epoorest urban-dwellers spend up to threequarters of their income on staple foods alone(Oxfam, 2009) and regularly engage in negativecoping strategies, such as reducing the numberof meals consumed, decreasing food varietyand quality, and food scavenging. Residentsliving in informal settlements are vulnerableto rising prices as they are highly dependenton the market for their food and non-foodneeds. Informal settlements in Nairobi arecharacterised by inadequate access to potablewater and sanitation facilities, leading to in-creased risk of waterborne, food-borne andvector-borne diseases such as diarrhoea,cholera, typhoid and malaria. Children underfive years old attending informal daycare cen-

tres are particularly vulnerable, due to lack ofregulation and staff training and inadequatefood and play activities.

Research shows that only 39% of childrenbetween six months and two years living inurban slums of Nairobi receive an adequatediet, in terms of both quantity and diversity(Concern Worldwide and Welthungerhilfe,2009). Deficiencies are common in iron, zinc,vitamin A, vitamin C and protein, and mostcaregivers do not feed their children appro-priately during and aer illness (ConcernWorldwide, 2014). Although the prevalenceof acute malnutrition is relatively low inNairobi County, at 54,438 cases the burdenof acute malnutrition is the second highest inKenya due to population density (Kenya Foodand Nutrition Security Seasonal Assessment,2018). One in every three children is stunted(SMART survey, 2017) and, in terms ofabsolute numbers, Nairobi County has the

Scale up ofhigh-impactnutritioninterventionsin the informalsettlements ofNairobi, Kenya

Pete

r Cat

on

A family living in Mukuru, one ofNairobi's largest slums, Nairobi, 2017

3

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


highest caseload of stunted children in thecountry at 104,074 children (Kenya Food SecuritySteering Group, 2018). Substantial levels of over-weight/obesity have also been observed, demon-strating the double burden of malnutrition; inNairobi 43% of women are obese, compared to25% of women of reproductive age nationally(KNBS, 2009). Childhood obesity is also on therise, although the problem is much more prevalentamong adults at present.

Concern urban nutrition programme Concern Worldwide has partnered with thegovernment to support scale-up of HINI inNairobi since 2012. is support has focusedon strengthening county government healthsystems to support the delivery of quality healthand nutrition services to children under fiveyears old and pregnant and lactating women(PLW) and improving community-based servicesthrough participatory approaches to behaviourchange and innovations in emergency nutritionprogramming, including preparedness. Fundedby UNICEF, Concern Worldwide supported im-plementation of HINI in Nairobi’s informal set-tlements between 2012 and 2017 (US$2.5 million).

Package of support to scale upHINIPolicy developmentConcern Worldwide has played a critical role insupporting Nairobi City County to develop keystrategic documents, including the Urban Nu-trition Strategy (UNS), the first County NutritionAction Plan (CNAP) (2013-2017), the Countymaternal infant and young child nutrition (MIY-CN) Social Behaviour Change and Communi-cation Strategy (SBCC), and the County NutritionCapacity Development Framework. Developmentof the UNS has been critical in highlighting theunique nutrition challenges in an urban context.As a result, other donors are now showinginterest in funding urban nutrition projects,

such as the Korea International CooperationAgency (KOICA), which is supporting imple-mentation of a nutrition innovations projectcalled ‘Lishe Poa’ that aims to develop a highlynutritious, affordable and ready-to-eat productthat will be promoted in the informal settlementsas a replacement for popular but unhealthystreet foods. e CNAP guides and costs nutritioninterventions in the County and is used as a toolto advocate for government to increase allocationof resources for nutrition services. Such advocacyefforts have been successful; for 2018/2019 theCity County of Nairobi has allocated 10 millionKES (Kenyan Shillings) to the County Nutritionprogramme, compared to 2012 when nutritionservices were 100% supported by developmentpartners. e Nairobi County MIYCN SBCCstrategy was developed to operationalise theMIYCN policy to protect, promote and supportoptimal maternal, infant and young child feedingpractices and improve child survival, and to pro-vide strategic guidance for implementation offocused interventions as outlined in the Advocacy,Communication and Social Mobilization (ACSM)strategy (an advocacy strategy for nutrition in-terventions in Kenya).

Nutrition assessments and capacity-building Concern Worldwide has supported the CityCounty of Nairobi by conducting a series of nu-trition-related surveys and assessments to provideevidence for decision-making and policy direc-tion. Baseline and endline nutrition programmecoverage surveys (2012 and 2018) and SMARTsurveys (2014 and 2017) were conducted. A nu-trition bottleneck analysis and nutrition causalanalysis for Nairobi County were carried out in2017. Health management team members atboth county and sub-county levels were trainedto conduct coverage and SMART surveys andbottleneck and causal analyses, and were involvedin the whole survey process (training of enu-

merators, supervision of data collection, super-vision of data entry and analysis, report writingand dissemination of results).

Concern Worldwide also provided technicaland financial resources to carry out a nutritioncapacity-gap analysis in 2016 for Nairobi County.Health management team members at countyand sub-county levels were trained and supportedto conduct analysis around the pillars of thehealth system. Results (summarised in Box 1)were used as a basis for a county nutritioncapacity development framework, which pri-oritised key capacity-building activities (classroomtrainings, on the job training and mentoring)for the nutrition workforce, including all healthcadres involved in providing nutrition servicesthrough to community health volunteers andcommunity health assistants involved in deliveringnutrition services at community level. ConcernWorldwide supported training of the healthmanagement team as trainers in IMAM, MIYCNand the Baby Friendly Community Initiative(BFCI), and in how to provide on-the-job trainingand mentorship to healthcare workers. etraining was then cascaded by the health man-agement team. It is expected that such trainingwill be carried out with government support infuture as part of the increased budget allocationto nutrition services. Concern Worldwide alsofacilitated the development of Nutrition TechnicalForums (NTF) at county and sub-county levelsthat bring together stakeholders implementingboth nutrition-specific and nutrition-sensitiveinterventions in the county. ese complementthe work of existing multi-sector platforms,which exist to support a range of outcomes, notjust nutrition.

Emergency preparedness Concern Worldwide supported the City Countyof Nairobi to mitigate and respond quickly tothe impacts of slow-onset emergencies by es-

Field Article

Box 1 Nutrition capacity in the Nairobi County health system: Key findings (2016)

Nutrition service needs in Nairobi county arecovered by the County Integrated DevelopmentPlan (CIDP) and the County Health SectorStrategic and Investment Plan (CHSIP). NairobiCity County also has a County Nutrition ActionPlan (CNAP) that specifies the capacity of thehealth system to deliver on nutrition.

Coordination of nutrition services occurs througha county and sub-county Nutrition TechnicalForum (NTF), which brings together nutritionstakeholders across the county to discuss how thenutrition agenda can be taken forward. A multi-sector platform also exists at county-level thatbrings together nutrition; water, sanitation andhygiene (WASH); agriculture; social services andeducation sectors. Challenges described includedifficulties in the dissemination and application ofkey nutrition policy documents, such as theBreast Milk Substitutes (BMS) Act (2012), to healthfacility and community levels, resulting in poorimplementation and enforcement. This highlightsthe need to strengthen the multi-sector platform.

Private, faith-based and non-governmentalorganisation (NGO) owned hospitals constitute78% of Nairobi County’s health facilities andprovide the bulk of nutrition services in theinformal settlements (this is much higher than inrural areas in Kenya, where health services arepredominantly government-owned). Most HINIsare offered at health facility level, but very few areoffered at community level due to low coverageof community health units and limitations of thenutrition workforce in the community. Manyhealth workers (nutritionists and other cadres)have been trained in national guidelines on theIMAM, MIYCN and the Kenya Medical SuppliesAgencies (KEMSA) Logistics Management andInformation System (LMIS); however, new skills arenot always practiced and training available to thecommunity-level workforce is more limited.

Ten per cent of the county’s health budget isallocated to the reproductive, maternal, newborn,child and adolescent health (RMNCH) sub-

programme, within which nutrition is a deliveryunit. Around 75% of the RMNCH budgetallocation pays for health worker salaries, withlittle resources allocated to service development.

HINIs described within the CNAP are notadequately financed; therefore the county isheavily reliant on partners for HINIimplementation. Equipment to assess nutritionstatus is available at health-facility level,although not always in adequate supply.Nutrition supplies for managing acutemalnutrition and routine medications areavailable at health facilities, although occasionalstockouts are experienced due to debts owed toKEMSA; an overreliance on partners to procurenutrition supplies and commodities persists.

All nutrition-related data collection tools areavailable at health-facility level, with theexception of the Maternal and Child Health(MCH) handbook.

4

tablishing an Urban Early Warning and EarlyAction (UEWEA) system. e UEWEA systemwas born out of the Indicator Development andSurveillance for Urban Emergencies (IDSUE)project that Concern Worldwide successfullypiloted for five years in informal settlements ofNairobi. e system is based on five key indica-tors: percentage of households experiencingshocks; equivalised monthly income; numberof food baskets; percentage of households withat least one stable income-earner; and percentageof households with at least one child reportingdiarrhoea. e county is being supported bythe Kenya Red Cross Society to collect monthlysurveillance data from the informal settlementsto feed into the UEWEA system, aer which re-sults are summarised as ‘normal’, ‘alert’, ‘alarm’or ‘emergency’. resholds for each categorywere set based on data from the five-year pilotphase (see Table 1). Once the normal thresholdis surpassed, the county government mobilisesresources to respond to the situation; emergencyresponse actions are mounted when UEWEAindicators indicate an emergency situation, anexample of which is provided in Box 2.

Concern Worldwide also supported the countyto rollout IMAM surge using a phased approach.IMAM surge aims to build the resilience of thehealth system to better deliver services for treat-ment of acute malnutrition over time, particularlyduring periods of high demand, without un-dermining the capacity and accountability ofgovernment health actors. During the first phaseof implementation, 18 health facilities (at countyand sub-county levels) were targeted throughthe training of two county trainer of trainers(TOTs) and 24 sub-county TOTs, who then cas-caded training to healthcare workers in the 18identified health facilities. During implementationof IMAM surge, seven health facilities were ableto surpass their normal IMAM surge thresholdsand no health facilities hit emergency thresholds,indicating that the health system managed tohandle the situation. During surge periods,health facilities notified the health managementteam members and, depending on the IMAMsurge phase, scale-up surge actions were un-dertaken, such as mass screening of childrenunder the age of five in the health facility catch-ment areas, prepositioning of medical and nu-

trition supplies. and cancelling healthcare workers’leave. At the county and sub-county levels,health management team members monitoredimplementation of IMAM surge activities usingdashboards that were updated monthly, indicatingwhich health facilities had passed normal thresh-olds and needed IMAM surge actions.

Nutrition programme coverageand nutrition outcomes Findings of nutrition SMART surveys conductedin 2014 and 2017 indicate marked improvementsin the coverage of all HINI indicators over thisperiod, including deworming of children aged12-59 months (59.0% to 65.1% versus target of50%); zinc supplementation for managementof diarrhoea among children under five yearsold (29.0% to 73.0% versus target of 80%);vitamin A supplementation for children aged6-11 months (81.7% to 87.2%) and childrenaged 12-59 months (36.4% to 80.2%) (versustarget for age 6-59 months of 80%) and ironand folic acid supplementation (IFAS) (23.9%to 56.7% versus target of 80%) (Figure 1). Whiletargets were met for two HINI indicators (de-worming and vitamin A supplementation), zincsupplementation for management of diarrhoeaamong children under five years old fell slightlyshort of target coverage. Achievement of IFASfor pregnant women improved but fell far belowthe set target. is was mainly due to supplychallenges as IFAS was not included in the es-sential drug list; however, it is now and goingforward all pregnant women will be supplementedas per the schedule.

Coverage of the county’s outpatient therapeuticprogramme (OTP) improved from 39.2% in2012 to 53.5% in 2018, while the county’s sup-plementary feeding programme (SFP) coverageimproved from 36.4% in 2012 to 51.5% in 2018(Figure 2). Results of SMART surveys show thatstunting prevalence declined from 36.0% in2014 to 26.0% in 2017; wasting prevalence re-duced from 5.7% in 2014 to 4.6% in 2017; andprevalence of underweight reduced from 13.6%in 2014 to 11.4% in 2017 (see Figure 3).

Aside from some small-scale interventions,such as that of Afya Jijini, which conducted nu-trition-related trainings, Concern Worldwide’sprogramme of health systems strengthening was

Field Article

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Indicatorthresholds

% of HHexperiencingshocks

Equivalisedmonthly income

No. ofbaskets

% of HH with at leastone stable incomeearner

% of HH with at leastone child reportingdiarrhoea

Normal less than 10% more than 8000 85 or more more than 30% less than 15%

Alert 10% -15% 6500-8000 70-85 25% - 30% 15% - 20%

Alarm 15% - 20% 5000-6500 55-70 15% - 25% 20% - 25%

Emergency 20% lower than 5000 55 or less less than 15% more than 25%

Table 1 Thresholds for the Urban Early warning and Early Action system

Box 2 Case study of county-level response to the Urban Early warning and Early Action(UEWEA) system

A drop in the number of food baskets and a surgein diarrhoea cases were reported to emergencythresholds in three of the sentinel surveillancesites (Korogocho, Mukuru and Kibera) in February2017, May to June 2017, and November 2017. Inresponse Concern Worldwide liaised with healthmanagement team members at both county andsub-county levels to plan and mount thefollowing response activities: mass screening andreferral of children with acute malnutrition;distribution of water purification products fortreatment of water at both household level and atdesignated water points; treatment of active

diarrhoea cases with zinc and oral rehydrationsolution (ORS), including referral of severe casesfor further management at the nearby healthfacilities.

A total of 32,244 children under the age of fivewas screened; of these 131 were diagnosed withsevere acute malnutrition (SAM) and 849 werediagnosed with moderate acute malnutrition(MAM). All were referred for management.Without the UEWEA surveillance system, suchchildren would not have been identified andreferred for treatment.

Pete

r Cat

on

A mother and her eightmonth old daughterenrolled in the outpatienttherapeutic programmesupported by Concern,Nairobi, 2017

Coverage of HINI indicators Nairobi County 2014 and 2017Figure 1100%90%80%70%60%50%40%30%20%10%0%

Deworming 12-59m Zinc supplementation <5s VAS 6-11m VAS 12-59m IFAS

2014 2017

59%65% 73%

29%

81%87.2%

80.2%

23.9%

56.7%

36.4%

Source: Nutrition SMART survey 2014 and 2017

VAS - vitamin A supplementation IFAS - iron and folic acid supplementation

HH = households

Cove

rage

5

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


the only large-scale, systematic programme overthis period to work with the government toidentify gaps in the health system and develop aclear plan of response. Improved coverage of nu-trition programmes and nutrition outcomes overthis period can therefore reasonably be attributedto the government’s commitment and dedicatedactions to develop health system capacity andscale up key actions and the technical and financialsupport provided by Concern Worldwide.

ChallengesLimited and short-term funding: Not manydonors are familiar with urban contexts andinterest in funding urban programmes is low.Donor resources are traditionally channelledto areas with high prevalence of malnutrition,rather than considering caseload. Over theyears, Concern Worldwide has received short-team (yearly to two-yearly) funding to supportimplementation of HINIs in Nairobi County,which hampers scale-up across all health fa-cilities. Inconsistent funding has also led tointerruption of HINI services. A case in pointis the period between January to July 2016,when the Nairobi urban nutrition programmedid not have funding and services were tem-porarily suspended.

Inadequate budgetary allocation for nutritionby the county government: e nutrition pro-gramme in Nairobi has a costed nutrition actionplan. However, a large proportion of resourcesreceived by the County Department of Healthfrom the national treasury goes towards takingcare of recurrent expenditures, with few resourcesle for development purposes. In addition, dueto government bureaucracy, it is difficult forthe County Department of Health to access thefew resources for development that are available;hence the government relies heavily on partnersto implement HINI.

Insufficient nutrition workforce: e Countyhas a total of 130 health facilities implementingHINIs but only 52 nutritionists, out of which12 are serving in managerial positions at countyand sub-county levels. e remaining nutritionistsare distributed between county referral hospitals(level-four health facilities) and a few healthcentres (level-three health facilities). ere areno nutritionists working in any of the dispensaries(level-two health facilities) and at the community(level one). is situation has meant other cadresof health workers, especially nurses and clinicalofficers, have had to take up nutrition roles

while they are already overwhelmed with existingresponsibilities.

Low coverage of community health units (CHU):Based on the population in the county, Nairobishould have at least 868 Community HealthUnits (CHU); however, only 217 exist, of whichonly 121 are functional (the rest are completelynon-functioning), which hampers efforts to createdemand for nutrition services at communitylevel. In addition, low coverage of CHUs limitscapacity to follow up clients at the communitylevel, which may hamper recovery rates.

Private health care: e majority of health fa-cilities in the urban informal settlements areprivately owned, which makes health servicesexpensive for residents when they would other-wise be free for children under five years oldand PLW. ere is therefore need for public pri-vate partnerships (PPP) to ensure access toquality nutrition services, especially in the in-formal settlements.

Conclusions andrecommendationsScale up of HINI services in the informal settle-ments of Nairobi through existing governmenthealth structures and with technical supportfrom Concern Worldwide resulted in improve-ment in the nutrition status of children underfive years old. ere was also a notable improve-ment in the coverage of OTP and SFP services.is provides evidence that scale-up of HINIsthrough the strengthening of existing health sys-tems is a viable approach that could be appliedto other urban areas. While wasting prevalenceis acceptable, absolute numbers are usually veryhigh. is requires a new way of thinking; forexample, coming up with urban-specific thresh-olds (as per the UEWEA) to determine when tomount emergency nutrition interventions.

Heavy dependence on development partnersto fund HINI scale-up persists. A review of thekey strategic documents, including the UNSand the Kenya Nutrition Action Plan (KNAP),is needed to ensure that they capture the chal-lenges faced by the urban nutrition programmeand to direct budgetary allocations towards de-velopment to address these.

Scale-up of HINIs in Nairobi continues to facethe challenges of limited and short-term fundingthat is secured based on prevalence rather thancaseloads. Continued advocacy is needed withdonors to inform and change practice.

Field Article

Increased advocacy is needed to the gov-ernment and other key stakeholders to allocateresources for strengthening community-levelhealth and nutrition services, including recruit-ment of more nutritionists.

Advocacy is also needed to the governmentto embrace PPPs, whereby the government pro-vides drugs and additional healthcare workersin private healthcare facilities to enable poorresidents in the informal settlements to accesshealthcare services at an affordable rate.

Scale up of HINI services in Nairobi Countyhas continued beyond this first phase, fundedby a county government allocation of KES 10million and by UNICEF (to Concern Worldwide)to support this process up to 2022. Lookingahead, tackling childhood stunting is a highpriority for the Nairobi urban nutrition pro-gramme, since stunting is the predominant formof malnutrition in the County. Investment inthe government’s BFCI and promotion of MIYCNpractices are priorities to this end.

For more information, please contact KassimLupao at [email protected]

References Nairobi Urban Sector Profile, 2006.https://unhabitat.org/books/kenya-nairobi-urban-profile/

OXFAM, 2009, Urban Poverty and Vulnerability in Kenya.https://urbanhealthupdates.files.wordpress.com/2009/09/urban_poverty_and_vulnerability_in_kenya1.pdf

The Global Hunger Index report 2009, Concern Worldwideand Welthungerhilfe. The link between gender inequalityand hunger.

Process for Improving Child Feeding (ProPAN) (2014),Preliminary Report, Concern Worldwide. Socialdeterminants of child under-nutrition in urban informalsettlements in Kenya.

Kenya Food and Nutrition Security Seasonal Assessment,2018 available at http://www.nutritionhealth. or.ke /reports/seasonal-assessment-reports/situation-reports/#toggle-id-1

Nutrition Survey Conducted in The Slums of Nairobi County,May 2017 (SMART survey) available at http://www.nutritionhealth.or.ke/reports/smart-survey-reports/#toggle-id-2

Kenya Demographic and Health Survey 2008-09. KenyaNational Bureau of Statistics (KNBS), ICF Macro (2009).Calverton, Maryland: KNBS and ICF Macro. Available athttp://dhsprogram.com/pubs/pdf/fr229/fr229.pdf.

Food and Agriculture Organization (FAO) of the UnitedNations (2006). The double burden of malnutrition. Casestudies from six developing countries. FAO food andnutrition paper 84. Rome: FAO. Available atwww.fao.org/docrep/009/a0442e/a0442e03.htm

IMAM coverage Nairobi County2012 and 2018 Figure 2

60%50%40%30%20%10%0%

Years

OTP SFP

39.2% 36.4%

51.5%53.5%

Nutrition status of children under five years old, Nairobi County2014 and 2017 Figure 3

40%35%30%25%20%15%10%5%0%

Wasting Stunting Underweight

5.7% 4.6%

34%

26%

11.4%13.6%

2012 2018

Cove

rage

OTP - outpatient therapeutic care SFP - supplementary feeding programmes

Source: Nutrition coverage survey of 2012 and 2018 Source: Nutrition SMART survey 2014 and 2017

Lupao, K. and Mogusu, E. (2019) Scale up of high-impact nutrition interventions in the informalsettlements of Nairobi, Kenya. Field Exchange issue 59,January 2019. www.ennonline.net/fex

Cove

rage

6


News...........................................................................

Participants at the GNCthree-day annual meeting

in in Amman, Jordon

Glo

bal N

utrit

ion

Clus

ter

Aone-day side meeting convened by theGlobal Nutrition Cluster (GNC) andhosted by the UNICEF Middle East andNorth Africa (MENA) regional office was

held in Amman, Jordon on 21 October 2018 on theoccasion of the GNC three-day annual meeting.Emergency Nutrition Network (ENN) supportedcountry teams to develop background papers andpresentations and produced the meeting report.

The aims of the side meeting were to discuss the nu-trition situation in Sudan, propose immediate ac-tions to support the Government of Sudan (GoS) ina longer-term comprehensive nutrition approachand identify the operational implications of the ‘Callto Action’ to end malnutrition in Yemen1, and definethe support needed to operationalise the three-yearplan. The event brought together 76 participantsfrom Sudan and Yemen, including representatives ofgovernment, Nutrition Clusters, the Scaling Up Nu-trition (SUN) Movement, United Nations (UN) andnon-gonvernmental organisations (NGOs), as wellas key donors and GNC partners with an operationalpresence or interest in the countries.

Representatives from Sudan and Yemen presentedan analysis of the nutrition situation in each country,as well as drivers of the current high rates of under-nutrition, a description of existing delivery systemsand challenges and proposals for next steps. Similarchallenges exist in both countries. Both are highlydependent on humanitarian funding, with limitedaccess to longer-term, flexible funding tostrengthen government systems. Both countrieshave challenges coordinating between humanitar-ian and development partners and between the

Nutrition Cluster and the SUN Movement. The inte-gration of nutrition-specific interventions into gov-ernment health systems is also difficult in bothYemen (where many frontline staff receive no reg-ular salary) and Sudan (where supply chain man-agement and procurement services are managedby external partners). In both countries there hasbeen a lack of progress in scaling up multi-sectorprogrammes; data are out of date and not nation-ally representative (leaving partners working in adata vacuum); and there are difficulties in the coor-dination of nutrition programming at all levels.

Participants divided into working groups to exam-ine key issues and identify actions. For Yemen, par-ticipants divided into three groups – UN, NGO anddonor – to focus on the commitments made in theCall to Action and examine what has been tried,what has worked, what hasn’t and identify clearnext steps. For Sudan, groups were divided intothree groups by theme to identify ways to moveforward in each area: financing (immediate andlonger-term); multi-sector nutrition programming;and coordination. Group feedback to all was fol-lowed by plenary discussion.

Conclusions included the need for the humanitar-ian nutrition community to proactively engagewith development actors and vice versa. The GNCand the SUN Movement need to move quickly andclearly to develop guidance on strengthening co-ordination in both countries. Connections be-tween the SUN Movement Secretariat and GNC inGeneva must be re-established and developedfurther. Opportunities are needed to include

longer-term objectives and programming withinhumanitarian response plans (HRPs) in protractedcrises; the Cluster should examine the potential formulti-year HRPs and multi-year humanitarianstrategies with the inclusion of development indi-cators. Greater clarity is needed on transition trig-gers, agreed milestones and indicators fordeactivation of clusters in contexts such as Sudanthrough a comprehensive transition plan that en-sures gradual handover to government. Senior UNofficials must encourage donors to provide longer-term flexible funding in protracted crises.

Complex contractual arrangements that vary byUN agency frustrate joined-up severe acute mal-nutrition (SAM) and moderate acute malnutrition(MAM) programming. This challenge is being ex-amined by the agencies; discussion at this meetingreinforces the necessity and urgency to addressthis. The continuum of care (i.e. prevention preced-ing treatment) requires much more effort and in-vestment. This must translate into integrated,high-impact nutrition interventions (HINI),strengthened health systems and multi-sector ap-proaches. In general there is a need to improve thenarrative so that there is one common approachthat spans humanitarian and development objec-tives for nutrition.

A meeting report, background papers fromYemen and Sudan and presentations areavailable at: http://nutritioncluster.net/ what-we-do/events/

UNICEF programmeguidance on adolescents

UNICEF has released new guidance thatprovides an overview of its strategic di-rection to advance the quality of life ofadolescents. The guidance underscores

the importance of adequate nutrition for optimalgrowth and development in adolescent girls andboys. It emphasises the provision of nutritioncounselling and services, the promotion ofhealthy and diversified diets, including fortified

foods, and, where necessary, the use of micronu-trient supplements.

The guidance recognises schools as a critical plat-form to deliver nutrition interventions and im-prove nutrition literacy, while acknowledging thatschool-based interventions need to be supportedby community-based approaches and other de-livery models, especially to reach out-of-school

adolescents. Download the guidance from:https://bit.ly/2TlhTtm

To complement this overarching guidance,UNICEF will issue Programme Guidance on Ado-lescent Nutrition in early 2019 to detail the organ-isation’s approach for the operationalisation ofevidence-based interventions to address all formsof malnutrition among adolescent boys and girls.

1 Following commitments made at the United Nations GlobalAssembly (UNGA) in September 2018.

Report of the GlobalNutrition Clusterside event onYemen and Sudan

http://nutritioncluster.net/what-we-do/events/

http://nutritioncluster.net/what-we-do/events/

7

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


The State of the HumanitarianSystem 2018 report

The fourth ‘State of the Humanitarian Sys-tem’ report1, covering the three yearsfrom 2015 to 2017, is based on a reviewof over 200 evaluations of humanitarian

action, interviews with over 500 people and casestudies from five countries (Bangladesh, Kenya,Lebanon, Mali and Yemen). The report outlineshumanitarian needs, provides an overview of hu-manitarian funding and the current size andstructure of the humanitarian system, and pres-ents an assessment of the system’s performancein addressing humanitarian needs. Findingsdemonstrate that humanitarian needs continuedto increase in this period. An estimated 201 mil-lion people required international humanitarianassistance in 2017 alone, the highest number todate. The number of people forcibly displaced byconflict and violence also increased, reaching68.5 million in 2017.

A small number of complex crises received themajority of funding: half of all international hu-manitarian assistance went to just four crises(Syria, Yemen, South Sudan and Iraq). There wasalso a gradual shift in the geographic location ofrecipients, from sub-Saharan Africa to the MiddleEast. A small number of donor governments con-tributed the majority of international humanitar-ian assistance. Most donor funding went tomultilateral agencies, much of which was thenpassed on as grants to non-governmental organ-isations (NGOs). Among NGOs, funding wasconcentrated among large, international organ-isations; national and local NGOs only received0.4% of all international humanitarian assistancedirectly. Money for pooled funds reached a recordUS$1.3 billion in 2017, 53% higher than in 2014.

Cash transfers also grew to an estimated US$.8 bil-lion in 2016, a 40% increase on 2015.

Coverage: Despite increased funding, the hu-manitarian system still does not have sufficientresources to cover needs due to growing num-bers of people in need and potentially also in-creased ambition on the part of the humanitariansector. The 2015-2017 period saw a decline incoverage of humanitarian needs, particularlyamong internally displaced people (IDPs) outsidecamps; people and communities hostingrefugees; people in situations of conflict whereaccess is challenged; and large numbers of irreg-ular migrants.

Appropriateness: Evidence suggests that the hu-manitarian system’s relevance and appropriate-ness has improved since 2015. Humanitarian aidcomprises a basic package of life-saving assis-tance, which is seen as relevant in many situa-tions. However, some needs are often not met,including priority protection needs, needs be-yond the immediate response ‘package’ and thoseof the elderly and people with disabilities. Thereis evidence that multi-purpose cash grants can gosome way to increasing the relevance of aid.

Accountability and participation: There hasbeen limited progress in the accountability andparticipation of the humanitarian system. Themain challenge identified in the 2015 report –that existing feedback mechanisms do not influ-ence decision-making – has not been addressed.While there are number of initiatives and ap-proaches that show potential, they have not yetdelivered greater accountability or participation.Many interviewees are concerned that account-ability to affected populations (AAP) has becomea ‘box-ticking exercise’.

Effectiveness: There is evidence of improved ef-fectiveness of the humanitarian system, particu-larly in terms of meeting immediate life-savingneeds in ‘natural’ disasters, responding to suddenmovements of refugees and responding to foodinsecurity in complex emergencies. The system isstill not effective in meeting protection needsoverall.

Efficiency: There has been limited progress in ef-ficiency, particularly in terms of non-harmonisedreporting and ‘pass-through’ arrangements offunding. Increased work on early response hasprevented inefficient ‘peak-of-crisis’ responses insome areas and some improvements have beenmade in joint procurement and supply chainswithin the United Nations. Increased use of cashhas increased efficiency in many areas and thereis potential for the ‘Grand Bargain’ process to ad-dress several areas related to efficiency.

Coherence: There has been a decline in the levelof coherence of the humanitarian system. The in-creased integration of humanitarian action intodevelopment and stabilisation agendas hasmade coherence with humanitarian principlesmore difficult for operational agencies. There isalso evidence of declining respect for interna-tional humanitarian law (IHL) and laws concern-ing refugee.

Connectedness: There appears to be improvedconnectedness in the humanitarian system, facil-itated by changes in policy and increased fund-ing, which has led to closer connections betweenhumanitarian and development activities, oftenin the form of ‘resilience’ work. There is some evi-dence that this has been effective at protectingagainst future shocks where the work has beendone with governments, and where it addressesforeseeable ‘natural’ disasters, but less evidencein other circumstances. Donors are increasinglyinterested in fragile and refugee-hosting states,‘peace-building’ initiatives, and in developing fi-nancing in countries experiencing conflict andrefugee situations.

Complementarity: There is improved complemen-tarity in the humanitarian system. Relations withthe governments of crisis-affected states are im-proving in many cases, although there is still a ten-dency to push governments aside in rapid-onset,‘surge’ situations. Relations with governments areoften more difficult where the state is a party to in-ternal armed conflict in refugee contexts. There hasbeen significant activity at policy level in strength-ening the role of national and local NGOs in the in-ternational humanitarian system but, to date, thishas had limited effect on the ground.

Impact: There is little hard data measuring theimpact of humanitarian responses on wide pop-ulations or across time. Very few evaluations at-tempt to assess impact; in part because the shortfunding cycles of humanitarian action preventconsistent longitudinal research. There is also alack of baseline data against which to measureprogress. Overall, information on impact is scat-tered and largely anecdotal and does not allowany overall conclusion to be drawn.

The summary and full reports can be accessedfrom: https://sohs.alnap.org/

1 The State of the Humanitarian System is an independent study that compiles the latest statistics on the size, shape and scope of the humanitarian system and assesses overall performance and progress. Published every three years, it provides a sector-level mapping and assessment of international humanitarian assistance.

Bangladesh: What future for the Rohingya?Case study – The State of the HumanitarianSystem 2018 report

Euro

pean

Uni

on 2

018

News

8

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


News

Wasting in South Asia: Consultation onbuilding the evidence base on the policyand programme response

Aone-day consultation was held inNew York on 16 November 2018 toexamine the evidence on wasting inSouth Asia and guide the direction of

future collaborative efforts of the No WastedLives coalition in the region. The consultationwas organised by UNICEF with the followingobjectives: (1) to share the status of policy andprogramme action to care for severely wastedchildren in South Asia; and (2) to identify evi-dence gaps, research priorities and way forwardto build the evidence base to inform the policyand programme response in South Asia. Mem-bers of No Wasted Lives and the Council of Re-search and Technical Advice on AcuteMalnutrition (CORTASAM) and researchers andacademics were invited to join the consultation.There were 32 participants, including 13 partic-ipants who joined the meeting remotely.

In the morning, presentations examined thecontext of and response to wasting in SouthAsia, with a specific focus on India, which carriesabout 80% of the regional wasting burden andwhere the government is developing nationalguidelines on the community-based manage-ment of acute malnutrition (CMAM). In the af-ternoon, the participants discussed theimplications of the data and evidence pre-sented on the design of policies and pro-gramme to prevent and manage severewasting. The following conclusions were drawn.

First, the South Asia context has several uniquecharacteristics that require further explorationbecause they may warrant alternative ap-proaches to the care of wasted children. The‘very high’ prevalence of wasting (15.9%) inSouth Asia exceeds all other regions, yet thepost-neonatal mortality rate is relatively low.Contrary to countries in other regions, theprevalence of wasting is highest at birth inSouth Asia, which suggests that poor maternalnutrition is a key driver. A higher proportion ofwasted children in South Asia experience wast-ing for prolonged periods than in sub-SaharanAfrica. In India, severely wasted children re-spond lower and slower to treatment for rea-sons that are not fully understood. There arealso questions concerning the mortality risks ofsevere wasting and child survival benefits oftreatment in South Asian countries, which ap-pear to be lower than in sub-Saharan Africancountries. Nevertheless, the mortality risks arenot low enough to ignore, particularly in thefirst six months of life, and there are potentiallylong-term impacts of wasting on cognition andlearning.

Second, the draft of India’s national CMAMguidelines and the country level adaptationsbuild on the relatively strong community plat-forms for early case detection, community-based management and referral in India. Withthese guidelines, the government seeks sus-tainable solutions that focus on both the pre-vention and treatment of wasting, includingduring the first six months of life. The currentdraft of the guidelines supports the use ofweight-for-height (but not mid-upper arm cir-cumference) to identify wasted children anddoes not promote the use of ready to use ther-apeutic food to treat severely wasted children,even though these are supported by the WorldHealth Organization (WHO) recommendations.These approaches are likely to be adequate,provided the intervention provides a qualityproduct that complies with WHO specifications,and systems are in place to identify and refer se-verely wasted children with medical complica-tions for inpatient care. These guidelinesprovide an opportunity for a learning agendaincluding the cost-effectiveness of this alterna-tive model of care for severely wasted children.

Third, research in South Asia can contribute toglobal and regional efforts in optimising and in-novating care and treatment approaches forchildren with severe wasting. Areas of researchinclude identifying effective approaches to pre-vent and manage wasting in infants aged lessthan six months; modifications in the quantity,

duration and formulations of ready-to-use ther-apeutic food (RUTF) use in nutritional rehabili-tation; the use of home-based foods orhome-augmented foods to treat severe wast-ing; and transitioning from treatment foods tofamily diets. This research could facilitate thedevelopment of a greater range of treatmentoptions that are tailored to cultural preferencesand have the potential for greater coverage,quality and sustainability of care and treatmentfor severe wasting.

In moving forward, the participants identifiedthe need for a new narrative on wasting inSouth Asia (and globally) that considers pro-longed versus short episodes of severe wasting;that links wasting with stunting; and thatframes the functional consequences of wastingon cognition and learning as well as the mor-tality risks. CORTASAM offered its expertise bysupporting further exploration on wasting inSouth Asia through a sub-working group of theCORTASAM. Potential areas of focus for thisworking group include: further context-specificrefinement and expansion of priority evidencegaps in South Asia that complement the COR-TASAM research agenda; the design and/or re-view of protocols for secondary data analysisand implementation research; and the interpre-tation and dissemination of research findings.

For further information please contact HarrietTorlesse ([email protected])

UN

ICEF

; Ind

ia


By Meeta Mathur, Naveen Jain, ShivangiKaushik and Aakanksha Pandey

Meeta Mathur is Head of Programmesfor Action Against Hunger India. Shehas over 13 years of experience in thefield of nutrition and health, workingwith international non-governmentalorganisations, government and theprivate sector. Meeta is a qualified

nutritionist with a Masters’ degree in food and nutrition.

Naveen Jain is Mission Director of theNational Health Mission, Rajasthan andSecretary of the Medical Health andFamily Welfare Department. He haspioneered various e-initiatives,including innovative software for use inmalnutrition treatment centres in India

and has over 20 awards for piloting technical innovations inthe field of governance.

Shivangi Kaushik is a public healthprofessional currently working for ActionAgainst Hunger India. She has a masters’degree in public health and over sevenyears’ experience leading nutritionalsurveys for Action Against Hunger andtraining government and non-

government staff on community-based management of acutemalnutrition, infant and young child feeding, nutrition inemergencies programming and SMART survey methodology.

Aakanksha Pandey is the StateProgramme Manager for ActionAgainst Hunger Rajasthan. She hasmany years of experience working withthe government system, providingtechnical support in theimplementation, assessment, planning

and monitoring of health and nutrition programmes. Shecurrently supports the government of Rajasthan in thedevelopment and implementation of community-basedmanagement of acute malnutrition protocols.

The authors would like to thank Naveen Jain and his team atthe Medical Health and Family Welfare Department,Government of Rajasthan, Baran District, for their leadershipand dedicated collaboration in this programme.

Location: India What we know: Comprehensive management of children with severeacute malnutrition (SAM), including psychosocial support for thecaregiver and child, improves treatment outcomes.

What this article adds: A pilot study by Action Against Hunger (AAH)India was carried out in Baran, a tribal district of Rajasthan, north India,to improve implementation of government SAM guidelines. This involvedstaff training (n=32) in nine malnutrition treatment centres (MTCs); theappointment and training of counsellors in five MTCs to providecaregiver and SAM children support (one-to-one counselling and groupsessions); and improving water, sanitation and hygiene (WASH) andplay/education infrastructure facilities in four MTCs. In counsellor-supported MTCs (1,041 children, 2016-18), performance improved interms of earlier identification by and referral from the community,increase in mean weight gain, reduction in defaulting rates and greaterpost-discharge attendance at follow-up. The pilot study shows thepotential to improve SAM treatment outcomes through staff capacity andinfrastructure development to support operationalising governmentguidelines. Dedicated staff are necessary to deliver adequate psychosocialsupport and follow-up. Results have been presented to government anddiscussions are ongoing about the potential for scale-up.

Institutionalisingquality of care ininpatientfacilities for themanagement ofsevere acutemalnutrition inIndia

IntroductionIn Rajasthan, north India, 20.4%(1,810,670) of children under five yearsold are wasted and 7.3% (647,936) are se-verely wasted (National Family HealthSurvey (NFHS) 3, 2006). Wasting is par-ticularly prevalent in Baran, a largely tribaland one of the poorest districts of Ra-jasthan. High prevalence of severe acutemalnutrition (SAM) led the Governmentof Rajasthan to develop malnutrition treat-ment centres (MTCs) in the state, the firstof which was set up in Baran in 2006 andnow has the highest number of MTCs. Inthe Integrated Management of Acute Mal-nutrition (IMAM) programme childrenwith SAM are referred to an MTC for arequired inpatient stay of 14 days or moreto meet medical and weight-gain discharge

criteria (community-based managementfor uncomplicated cases is not available).However, most caregivers, especially moth-ers, find it challenging to be away fromhome for over two weeks; hence theyeither refuse to be admitted or leave thecentre abruptly against medical advice.Action Against Hunger (AAH) has beenworking in Baran since 2012, focusing onSAM management by early identificationand referral of SAM children to MTCs.Despite efforts by AAH and the govern-ment, the proportion of children com-pleting the treatment and rate of followup post-discharge is low (below 50% forthe first follow-up and progressively lessfor each subsequent follow-up). To addressthis, AAH carried out a qualitative as-sessment of MTC services in the district,

9

AA

H In

dia

An MTC counsellor engages mothersof SAM children in a massagesession, MTC Kelwara, 2017

Field Article ......................................................

10

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Field Articlewhich subsequently informed a series of pilotinterventions (training, building counsellingcapacity and infrastructure development) intarget MTCs. Experiences and findings, par-ticularly related to the counselling intervention,are shared in this article.

Quality assessment of MTCs At the end of 2015 a qualitative survey was un-dertaken to assess perceptions of and barriersto access of MTC services. In-depth interviewswere carried out with caregivers and frontlineworkers, while knowledge and skills of MTCstaff and quality of MTC infrastructures wereassessed. Key barriers identified were resistancefrom husbands and family members to mothersand children staying in the MTC, caregiversfeeling intimated by the hospital setting, andcaregivers finding it difficult to stay alone at theMTC without their husbands or family membersaccompanying them. Many caregivers also com-plained that they feel bored in the MTC as theyhave nothing to do. ese issues were aggravatedby inadequate living conditions at the centresand negative behaviour of some MTC staff,which damaged parental trust and perceptionsof quality of public health services. Findingswere shared with district and state-level officialsand possible solutions were discussed.

Quality improvement of MTCsBased on the assessment findings, a programmeof quality improvement was undertaken byAAH, with the full support of the district healthadministration in Baran, to improve the qualityof nine MTCs in the district. A three-prongedprogramme was devised to improve communitydemand and the quality of supply services inMTCs. is involved: (1) training of MTCstaff in inpatient management of SAM in allnine MTCs; (2) appointing and training ofMTC counsellors from local communities infive selected MTCs (those with the highestcaseloads and referrals) to provide caregiverswith quality care and supportive counselling;and (3) key infrastructure developments infour MTCs where existing infrastructure wasparticularly poor to improve the living conditionsfor caregivers and patients and the general en-vironment. Each of these improvements aredescribed in more detail below, particularlythe strengthened counselling capacity, whereperformance was subsequently measured inpilot MTCs.

Technical support to MTCs Since 2016 AAH Baran has conducted numerousorientation and refresher trainings for staff ofall nine MTCs to build their capacity and ac-quaint them with new developments in SAMmanagement protocols. Trainings are carriedout by AAH staff and a highly experiencedAAH paediatrician and include an initial three-day classroom training, followed by regular re-fresher training and ongoing supportive su-pervision. Information booklets (such as ‘Op-erational Guidelines of Malnutrition TreatmentCentres’), SAM management protocols and job

Box 1 Key roles and skills of MTC counsellors

To optimise their stay in the MTC, counsellorswork with caregivers to understand theirchallenges from a psychosocial perspective,provide emotional support and empower themto care for their own and their child’s nutritionand psychosocial needs. Tribal caregiverstravelling long distances to an unfamiliartreatment centre are usually overwhelmed andconfused on arrival. MTC counsellors are trainedto welcome new admissions, assign them a bedand introduce them to MTC staff and facilities.This helps caregivers settle into the surroundingsand establishes trust and confidence in thecounsellors. Soon after this, counsellorscomplete a ‘care practices evaluation’ with thecaregiver, using an adapted checklist of items toidentify areas of weakness in caregivingbehaviours. Based on this preliminary evaluation,counsellors prepare a plan for priorityinterventions with the caregiver to seek theiracceptance, consent and active participation.

Important skills and personal attributes of MTCcounsellors are: • Empathy: ability of the counsellor to comprehend the problems, experiences, thoughts and feelings of caregivers and offer supportive space for expression and self- disclosure that further enhances mutual trust and confidence. • Congruence and warmth: the ability to act in a way that enables caregivers to feel comfortable in the counselling relationship, receive encouragement to interact and attend to child’s health needs.• Respect: the ability to be impartial and non- judgmental about tribal caregivers, their family challenges and their social situations, so that mothers feel safe, comfortable and confident. • Unconditional positive regard: acceptance of the caregivers, irrespective of their caste, class and origin; irrespective of caregiver’s weaknesses, negativity and unfavourable conditions. • Active listening and assistance: a skill that lets caregivers speak for themselves and identify and accept problems and which empowers them to problem-solve on their own, leading to insight, new learning and sustained behaviour change.

Individually tailored, one-to-one psychosocialsupport sessions are then organised in responseto the plan. MTC counsellors also organise dailygroup education sessions for mothers onspecific topics, repeated every 14 days so thateach caregiver receives a complete package ofknowledge about maternal care and nutritionduring pregnancy; post-natal care and infant-feeding practices; immunisation; sanitation andhygiene practices; cooking demonstrations toimprove diet diversity; and the benefits of familyplanning and related services. These sessions aredelivered through games, videos andstorytelling via flip charts and caregivers areencouraged to share practical challenges theyface and ideas on how these can be tackled.

In addition, MTC counsellors teach caregiverssimple relaxation exercises, including yoga andbreathing exercises, with the aim of improvingthe caregiver’s mood and helping to reducemental stress and physical tension and managetheir emotions effectively. MTC counsellors alsoteach mothers to practice daily baby massage topromote a secure mother-child attachment andsupport the child’s physical and psychologicalrecovery and development. Visiting fathers arealso involved in massage sessions to promotepaternal affection and attachment with thebaby. Daily hour-long play sessions are alsoorganised to stimulate physical and cognitivedevelopment in admitted children andstrengthen the relationship between caregiverand child further.

On discharge MTC counsellors motivate andinform caregivers to complete four follow-upvisits (every 15 days) to track post-dischargerecovery of the child from SAM. Counsellorsinform the AAH community mobiliserresponsible for the village the child is from oftheir discharge and ensure that the childreceives a nutritionally balanced food basketwithin 24 hours to prevent a gap in the provisionof a nutrient-rich diet. During each follow-upvisit the counsellor revises key messages onmaternal and child nutrition with the caregiver,answers any queries they have, providesinformation to address any specific caregivingchallenges and motivates the caregiver on thebasis of progress achieved.

AA

H In

dia

An MTC counsellor engages SAM childrenin a play session, MTC Baran, 201

11


aids are provided to participants as referenceguides and monitoring tools.

Appointing MTC counsellors Improvements in maternal-child bonding andinteraction are the major intended child devel-opment outcomes during MTC admission; how-ever, this aspect of SAM management was absentfrom the MTCs prior to 2016. Governmentguidelines recommend the appointment of acounsellor in each MTC, but funding to supportthis position was not available. To fill this gap,AAH appointed and financed MTC counsellorsin five MTCs with a view to advocating forthese positions to be filled and paid for by thegovernment in future. Recruited counsellorshad prior experience of working with womenand children and received extensive training onfacility-based management of acute malnutrition,maternal and child care practices, early stimu-lation methods, psychosocial support, counselling

and communication skills. (See Box 1 for moredetails on counsellors’ skillsets and how theywork.) MTC counsellors also provided technicalguidance and on-the-job practical supervisorysupport and monitoring through qualified andexperienced AAH technical managers.

Improvements in facilityinfrastructure Adequate infrastructure supports staff to carryout SAM management protocols successfully,improves staff motivation and provides a healthyenvironment for admitted children and theircaregivers, which aids recovery from SAM andreinforces demand for and utilisation of servicesby target communities. e assessment identifiedthe critical missing infrastructure to be providedas follows:

Water, sanitation and hygiene(WASH) infrastructure WASH facilities are closely linked with the man-

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

agement of infections and disease. In publichealth services in India, budgets are largely allo-cated to treatment rather than preventive services.Shortfalls identified in four MTCs were non-functioning toilets, no wash basins, no access tosafe drinking water (in two MTCs), no availabilityof warm water and no place for washing (anddrying) clothes. In these MTCs (and where MTCcounsellors were provided), AAH facilitated thedevelopment of adequate WASH infrastructure,including repair of toilets, installation of geysersfor round-the-clock provision of hot water formedical use and patient care in winter, washbasins with soap dispensers, laundry facilitiesfor caregivers, and repair of water pipes andwater purifiers to ensure a regular supply of safedrinking water for parents and children. Sensi-tisation of MTC staff, beneficiaries and visitorswas also undertaken to ensure proper upkeepand maintenance of these services.

Kitchen utilitiesRefrigerators used to store therapeutic milk ineach of the MTCs were repaired. At ShahabadMTC caregivers were not offered any food fromthe facility, which created great distress amongcaregivers. AAH helped to build a kitchen shedand installed traditional mud stoves to givecaregivers a place to cook food of their choice.

Play facilities e quality improvement intervention aimedto ease the anxiety and stress of hospitalisationon admitted children by making the MTC en-vironment as child-friendly and unintimidatingas possible. A play area was designed in eachMTC and in hospital wards, equipped with avariety of safe toys and games to support physicalstimulation and child development. MTC coun-sellors support mothers in playing with theirchildren and are taught how to make home-made toys from locally available resources andsafe waste materials. Televisions were provided

MTC outcomes 2016 to 2018Figure 120151050-5

AdmissionMeanWHZ

DischargeMeanWHZ

Mean Los Mean GoWgms/kgs/day

Mean GoW% of body

weight

TotalLAMA %

2016 -3.9 -2.6 14.6 9.5 13 31%

2017 -3.7 -2.2 16.6 10.7 14 20%

2018 -2.9 -1.1 14.1 16.1 15 9%

WHZ = weight-for-height z-score LoS = length of stay

LAMA = left against medical advice GoW = weight gain

Field Article

AA

H In

dia Play area with slides developed

by AAH, MTC Baran, 2017

AA

H In

dia

An MTC counsellor engages SAM childrenin a play session, MTC Baran, 201


. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

12

Field Articleto MTCs to show educational videos related tochildcare practices and recreation.

Provision of attendant beds Patient beds are ordinarily shared by their care-giver. However, most children and caregiversare accompanied by the father or another relative.In the absence of residential facilities, the ac-companying adult would usually stay for a day,then leave without the child completing treatment,or would also share the patient’s bed, leading toovercrowding. In response AAH provided sidebeds with storage facilities in the MTCs toenable fathers to stay too.

ResultsMTC quality care outcomes andimpacts Since their appointment in 2016 MTC counsellorshave maintained records of SAM beneficiaries.Training on monitoring was also provided inthe four MTCs that did not receive counsellorsas part of the general SAM management training,although having a dedicated person to keeprecords in the MTC counsellors greatly strength-ened monitoring activities and the quality ofrecords. Data for five MTCs where counsellorswere appointed (Baran, Kishanganj, Relawan,Kelwada and Shahabad) were analysed (seeFigure 1). In summary:• MTC counsellors provided psychosocial support and counselling services to around 33,160 parents and caregivers between 2016 and 2018, including the person accompanying the child for the period of admission and visitors. • During the project period, 32 MTC personnel, including the MTC in-charge, food demonstrators, cook and auxiliary nurse midwife (ANM)/general nurse midwife (GNM) were trained on protocols on the inpatient management of complicated SAM and nutrition data-monitoring tools. All these staff were continuously provided on-the-job support through AAH-appointed MTC counsellors and the AAH Field Officer.• Treatment outcomes improved for the 1,041 children treated between 2016 and 2018 in the five counsellor-intervention MTCs. During the two week inpatient treatment of SAM children’s anthropometric growth indicators improved significantly: - Average mean weight-for-height z-score (WHZ) was higher on admission aer the intervention due to earlier identification and referral from the community because of increased confidence in MTC services. - Average mean difference in WHZ between admission and discharge dropped by 1.5 SD. - Mean weight gain increased by 6.6g/kg/day; i.e. by 2% of average body weight gain recorded between admission and discharge. - Reduction by 22% of cases where caregivers le MTCs against medical advice (LAMA), indicating improvement

in caregiver compliance to treatment protocol due to enhanced quality of care. is also ensures the child receives the required treatment, increasing the percentage of recovery and reducing length of recovery.• Counsellors reported positive outcomes in response to child development support provided to caregivers and children at each facility. is included both caregivers and children engaging and responding positively to baby massage, bathing, play sessions and educational video sessions.• Improvements in the overall management of MTCs and monitoring were observed.• Percentage of cases attending all four post- discharge follow-up sessions increased in all MTCs assessed (from 35% to 43% in Baran; 23% to 53% in Kelwada; 19% to 25% in Relawan; 28% to 37% in Shahabad; and 33% to 39% in Kishanghj). is clearly indicates the focus of MTC staff, including the counsellor on post-discharge follow-up. e increase in post-discharge follow-up increases the probability of the child being cured and reduces the recovery period.

Discussion and conclusionsNon-parametric tests were not performed onthe data to test their reliability. However, resultsseem to suggest that improved performance ofthe services and treatment outcomes found inthis pilot can be attributed to the MTC counsellorsand the work of other team members (MTCnurses, feeding demonstrators and cleaning staff),infrastructure improvements leading to improve-ments in WASH, play and education facilities ateach centre, and additional capacity-developmentactivities, such as provision of anthropometryequipment and protocols and training for medicalstaff. Results indicate the value of investing inboth human resource capacity and infrastructureof government MTCs to improve service deliveryand SAM treatment outcomes.

A formal assessment of non-counsellor MTCswas not undertaken. However, programmingexperience is that outcomes for non-counsellorMTCs remain less favourable, including follow-ups happening in under 50% of cases. isreflects a strong need to reinforce post-dischargefollow-ups and improve length of stay amongthe caregivers of SAM children admitted at eachfacility. While ACF has been building the ca-pacities of MTC staff in captivating caregiversduring their stay at MTCs, regular staff arealready burdened with record-keeping and day-to-day management of SAM cases, making itdifficult to effectively and efficiently deliver allcomponents of a comprehensive service. isexperience reinforces the necessity of securingstaff who can exclusively deliver on family coun-selling and related follow-up.

e financial investment per facility is not hugeand the potential return on investment high. Find-ings of the study are being used to advocate withthe government for the future appointment ofcounsellors in all MTCs and the upgrading ofMTC infrastructure. As the study was carried outin partnership with the government health depart-ment, the results have so far been received well.

e results of this pilot programme show thatthe addition of dedicated SAM staff at facilitylevel, in addition to existing medical and supportstaff and structured and thoughtful improvementsin infrastructural facilities, can significantly en-hance facility performance indicators and improvetribal communities’ confidence, trust and per-ceptions of treatment services.

For more information please contact MeetaMathur at [email protected] [email protected]

AA

H In

dia Intake interview by an MTC

counsellor, MTC Baran, 2016

Mathur, M., Jain, N., Kaushik, S., and Pandey, A. (2019)Institutionalising quality of care in inpatient facilitiesfor the management of severe acute malnutrition inIndia. Field Exchange issue 59, January 2019.www.ennonline.net/fex

13

Research .............................................................Snapshots


Linear growth following complicated severe malnutrition

Children presenting with severe acutemalnutrition (SAM) are oen also se-verely stunted. e purpose of thisstudy was to evaluate linear growth

and its determinants aer medically complicatedSAM. A secondary analysis was performed ofclinical trial data from HIV-uninfected Kenyanchildren aged 2–59 months hospitalised withSAM. e outcome of interest was change inheight/length-for-age z-score (HAZ) between en-rolment and 12 months later. Exposures of interestwere demographic, clinical, anthropometric char-acteristics and illness episodes during follow-up.

Among 1,169 children with HAZ values atmonth 12 (66% of those in the original trial),median (interquartile range) age 11 (7-17) months

and mean (standard deviation) HAZ −2.87 (1.6)at enrolment, there was no change in mean HAZbetween enrolment and month 12: −0.006Z (95%CI −0.07 to 0.05Z); 262 (23%) children experi-enced minimal HAZ change (within ±0.25 HAZ),472 (40%) lost >0.25 and 435 (37%) gained >0.25HAZ. Aer adjusting for regression to the mean,inpatient or outpatient episodes of diarrhoeaand inpatient severe pneumonia during follow-up were associated with HAZ loss. Childrenborn prematurely had greater linear growth thantheir term peers in this study, reaching a similarHAZ to other children aer 12 months of fol-low-up. Not being cared for by the child’s biologicalparent (in this study these children were mostlyin the care of children’s homes) was also associatedwith HAZ gain. Increases in mid-upper arm cir-

cumference (MUAC) and weight-for-age (WFA)were associated with HAZ gain and protectedagainst HAZ loss. Increase in weight-for-height(WFH) was not associated with HAZ gain butprotected against HAZ loss. No threshold ofweight gain preceding linear catch-up growthwas observed. e authors conclude that inter-ventions to improve dietary quality and preventillness over a longer period may provide oppor-tunities to improve linear growth.

1 Ngari MM, Iversen PO, Thitiri J, et al (2018) Linear growth following complicated severe malnutrition: 1-year follow-upcohort of Kenyan children.Archives of Disease in Childhood. Published online first: 28 September 2018. doi: 10.1136/archdischild-2018-315641

Research snapshot1

Use of MUAC by novel community platforms to detect, diagnoseand treat severe acute malnutrition

There is growing consensus that makingmid-upper arm circumference(MUAC) use more widely accessibleamong caregivers and community

health workers (CHWs) will help decentralisethe management of severe acute malnutrition(SAM) and increasing programme coverage,including the management of uncomplicatedSAM by CHWs. A systematic review was con-ducted of published and operational evidencesince 2000 describing the use of MUAC for de-tection and diagnosis of SAM in children aged6-59 months by caregivers and CHWs, and ofmanagement of uncomplicated SAM by CHWsoutside of formal healthcare settings. A total of1,072 records were screened. Of these, 43 wereselected for full-text screening and 22 werefound to meet the study eligibility criteria. Datawere extracted on study design, intervention,control and key findings, and operational lessonswere synthesised.

Findings show that caregivers can use MUACto detect SAM in their children with minimalrisk and with many potential benefits in termsof early case detection and coverage. ere isalso evidence that CHWs can correctly useMUAC for SAM detection and diagnosis and tohelp provide a high quality of care in thetreatment of uncomplicated SAM when training,supervision and motivation are adequate.

Important limitations of the review were thesmall number of published research studies andtheir narrow geographic scope, and most de-

scribed intensive, small-scale interventions.Findings are therefore not generalisable to pub-lic-sector healthcare systems.

e authors conclude that scaling up the useof MUAC by caregivers and CHWs to detectSAM in household and community settings is apromising step towards improving the coverageof SAM detection, diagnosis and treatment.Further research is needed on scalability, appli-

cability across a wider range of contexts andcoverage impact and cost, as well as on the pri-mary use of MUAC for SAM detection.

1 Bliss J, Lelijveld N, Briend A, Kerac M, Manary M, McGrath M et al (2018) Use of mid-upper arm circumference by novel community platforms to detect, diagnose, and treat severe acute malnutrition in children: a systematic review. Global health science and practice, 2018. Available at: www.ghspjournal.org/content/early/2018/09/05/GHSP-D-18-00105

Research snapshot1

WFP

/Rei

n Sk

ulle

rud

Mid-upper arm circumference(MUAC) being measured in

Bakasi camp , Borno state

14


. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Research snapshot1

1 Mondal P, Islam MM, Hossain MI, Huq S, Shahunja KM, Haque Alam MN, Ahmed T (2017) Post discharge morbidities and mortalities among children with severe acute malnutritionwho did not undergo nutrition rehabilitation. Adv Pediatr Res 4:15. doi:10.12715/apr.2017.4.15

Post-discharge morbidities and mortalities among children with severeacute malnutrition who did not undergo nutrition rehabilitation

Aprospective study evaluated post-dis-charge morbidities among childrenwith severe acute malnutrition (SAM),including diarrhoea and other acute

illnesses, who did not undergo the nutrition re-habilitation (NR) phase of SAM management atDhaka Hospital of International Centre for Di-arrhoeal Disease Research, Bangladesh (icddr,b).NR follows the initial ‘stabilisation phase’ ofWorld Health Organization (WHO)-recommend-ed SAM treatment and provides energy-densefood to rebuild body tissues and achieve catch-up growth. e probable causes of not undergoingNR were family and other unavoidable commit-ments. Follow-up was undertaken between May

and August 2014 of 90 children aged 6-59 monthsof both sexes suffering from SAM with associatedmorbidities who undertook the stabilisation phaseof management but not NR. ree follow-upschedules were planned at two-week intervals(the second at the follow-up unit and the firstand third over the phone). During the first fol-low-up, 37 of 70 (53%) reported various mor-bidities. Only seven children came for the secondfollow-up; all required hospitalisation for differentmorbidities. On third follow-up, 23 of 58 (40%)children reported morbidity. e odds of mor-bidities were 7.7 times higher (95% CI: 2.33–26.58, p<0.0001) among the children who camefrom a poor family (monthly income < USD127).

Children with SAM and diarrhoea bypassing NRfrequently suffered from different types of mor-bidities. e authors conclude that NR is an im-portant component of the management of SAMand nutrition programmes in Bangladesh shouldconsider including community-based managementof acute malnutrition (CMAM) for the completemanagement of SAM in young children.

Research Snapshots

1 O’Sullivan NP, Lelijveld N, Rutishauser-Perera A, Kerac M, James P (2018) Follow-up between 6 and 24 months after discharge from treatment for severe acute malnutrition in children aged 6-59 months: A systematic review. PLOS ONE 13(8): e0202053. https://doi.org/10.1371/journal.pone.0202053

Post-discharge follow-up of children treatedfor severe acute malnutrition

Severe acute malnutrition (SAM) is a majorglobal health problem affecting some 16.9million children under five years old. Littleis known about what happens to children

aged 6-24 months post treatment discharge. Toinvestigate further, a systematic review was carriedout on follow-up between six and 24 months aerdischarge from treatment for SAM in childrenaged 6-59 months. e literature search was carriedout between June and August 2017. Studies wereselected if they included children who experiencedan episode of SAM, received a therapeutic feedingintervention, were discharged as cured and pre-sented any outcome from follow-up between six

and 24 months later. In total 3,691 articles wereretrieved from the search, 55 full-texts werescreened and seven met the inclusion criteria.Loss to follow-up, mortality, relapse, morbidityand anthropometry were outcomes reported. Be-tween 0.0% and 45.1% of cohorts were lost to fol-low-up. Of those discharged as nutritionally cured,mortality ranged from 0.06% to 10.4%, at anaverage of 12 months post-discharge. Relapse wasinconsistently defined, measured and reported,ranging from 0% to 6.3%. Two studies reportedimproved weight-for-height z-scores, while threestudies that reported height-for-age z-scores foundeither limited or no improvement.

e authors conclude that there is a scarcityof studies that follow up children six to 24 monthspost-discharge from SAM treatment. e limiteddata that exists suggest that children may exhibitsustained vulnerability even aer achieving nu-tritional cure, including heightened mortalityand morbidity risk and persistent stunting.Prospective cohort studies assessing a wider rangeof outcomes in children post-SAM treatment area priority, as are intervention studies exploringhow to improve post-SAM outcomes and identifyhigh-risk children.

Research snapshot1

Cash-based intervention and risk of acute malnutrition amongchildren in internally displaced persons camps in Somalia Research snapshot1

Cash-based interventions (CBI) havebeen used in Somalia since 2011, acountry with one of the highest preva-lences of acute child malnutrition in

the world. A non-randomised cluster trial in in-ternally displaced person (IDP) camps in peri-urban Mogadishu was undertaken to understandwhether a CBI would reduce acute malnutritionand its risk factors. Ten IDP camps (clusters) wereselected for the CBI, comprising a monthly un-conditional cash transfer of US$84 for five months,one-off distribution of non-food item kits andprovision of free piped water. Ten adjacent clusterswere selected as controls. Primary outcomes weremean child dietary diversity score (CDDS) collectedfrom children aged 6-59 months and their primarycarers (155 intervention; 177 control) from ran-domly selected households (household cohort)and incidence of first episode of acute malnutrition,

defined as mid-upper arm circumference (MUAC)< 12.5cm and/or oedema, collected from an ex-haustive sample of children aged 6-59 months(759 intervention; 1,379 control) (child cohort).

In the household cohort, the CBI appearedto increase CDDS by 0.53 (95% CI 0.01; 1.05).In the child cohort, the acute malnutrition in-cidence rate (cases/100 child-months) was 0.77(95% CI 0.70; 1.21) and 0.92 (95% CI 0.53;1.14) in intervention and control arms respec-tively. The CBI did not appear to reduce therisk of acute malnutrition: unadjusted hazardratio 0.83 (95% CI 0.48; 1.42) and hazard ratioadjusted for age and sex 0.94 (95% CI 0.51;1.74). The CBI appeared to increase the monthlyhousehold expenditure by US$29.60 (95% CI3.51; 55.68), increase household food consump-tion score (FCS) by 14.8 (95% CI 4.83; 24.8),

and decrease the reduced coping strategies index(RCSI) by 11.6 (95% CI 17.5; 5.96).

e authors conclude that CBI appeared toimprove beneficiaries’ wealth and food securitybut did not appear to reduce acute malnutritionrisk in IDP camp children. Study limitationswere that the trial was not randomised, thehousehold cohort sample size was small, noother anthropometric measurements were takendue to insecurity in the field, and no foodmarket data was available to aid interpretationof results.

1 Grijalva-Eternod CS, Jelle M, Haghparast-Bidgoli H, Colbourn T,Golden K, King S, et al. (2018) A cash-based intervention and the risk of acute malnutrition in children aged 6–59 months living in internally displaced persons camps in Mogadishu, Somalia: A non-randomised cluster trial. PLoS Med 15(10): e1002684. https://doi.org/10.1371/journal.pmed.1002684

Concurrent wasting and stunting among under-five children in Senegal

15


Research Snapshots. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Research snapshot1

The study describes the patterns of con-current wasting and stunting (WaSt)among children age 6-59 months livingin the 1980s in Niakhar, a rural area

of Senegal under demographic surveillance.Wasting and stunting were defined by z scoreslower than −2 in weight for height (WHZ) andheight for age (HAZ) respectively. Both conditions

were found to be highly prevalent; wasting moreso before age 30 months, stunting more so aerage 30 months. As a result, concurrent WaStpeaked around age 18 months and its prevalence(6.2%) was primarily the product of the twoconditions, with an interaction term of 1.57 (p <10 −6). e interaction was due to the correlationbetween both conditions (more stunting if wasted,

more wasting if stunted). Before age 30 months,boys were more likely to be concurrently wastedand stunted than girls (relative risk = 1.61), butthe sex difference disappeared aer 30 monthsof age. e excess susceptibility of younger boyscould not be explained by muscle mass or fatmass measured by arm or muscle circumference,triceps, or subscapular skinfold. ConcurrentWaSt was a strong risk factor for child mortality(the combined effect of WaSt explained some51% of the total mortality), the effect of whichwas the product of the independent effect ofeach component, with no significant interaction.More research is needed on common factorsunderlying wasting and stunting, the potentialincreased susceptibility of one to the other(stunted children being more likely to becomewasted and wasted children being more likely tobecome stunted) and potential underlying factorswhich cause variations by age and sex. Whateverthe aetiology of WaSt, the authors argue that thecase deserves more attention due to its dramaticimpact on child survival.

Mic

hel G

aren

ne

1 Garenne M, Myatt M, Khara T, Dolan C, Briend A. Concurrent wasting and stunting among under-five children in Niakhar, Senegal. Matern Child Nutr. 2018; e12736. https://doi.org/10.1111/mcn.12736

Taking anthropometric measures in avillage near Niakhar, Senegal, 1983

Improving screening for malnourished children at high risk of death

The purpose of this study was to in-vestigate whether children with con-current wasting and stunting (WaSt)require therapeutic feeding and to

better understand whether multiple diagnosticcriteria are needed to identify children with ahigh risk of death and in need of treatment. Acommunity-based cohort study (1983-84, pre-dating community-based management of acutemalnutrition) in Niakhar, a rural area of theFatick region of central Senegal, followed 5,751children aged 6-59 months living in 30 villages.Each child was visited up to four times at six-month intervals and at each visit anthropometricmeasurements were taken. Survival was mon-itored using a demographic surveillance systemoperating in the study villages. e outcomeof interest was death within 183 days (i.e. sixmonths) of anthropometric assessment.

Results showed that the lowest weight-for-age z-score (WAZ) threshold that, incombination with mid-upper arm circum-ference (MUAC) <11.5mm, detected all deathsassociated with severe wasting or concurrentWaSt was WAZ <− 2·8. Performance for de-tecting deaths was best when only WAZ andMUAC were used. Additional criteria forwasting (weight for height z score (WHZ)<-3) or WaSt (WHZ<− 2·0 and HAZ<− 2·0)did not improve performance. Risk ratios

for near-term death in children identifiedusing WAZ and MUAC suggest that childrenidentified by WAZ <− 2·8 but with MUAC≥115mm may require lower-intensity treat-ment than children identified usingMUAC<115 mm.

Results show that a combination of MUACand WAZ detected all near-term deaths asso-ciated with severe anthropometric deficits,including concurrent WaSt. erapeutic feedingprogrammes (TFPs) may achieve higher impactif both WAZ and MUAC admission criteriaare used (MUAC<115mm or WAZ<− 2·8),although more work is required before thiscan be considered a general recommendation.e authors challenge the current programmingapproach that distinguishes acute and chronicmalnutrition treatment. e analysis shouldbe repeated in different contexts and small-scale field studies should investigate the intensityand duration of treatment required to treatchildren identified using WAZ and appropriatedischarge criteria. Operational research isneeded to link growth monitoring/growthmonitoring promotion and TFPs.

1 Myatt M, Khara T, Dolan C, Garenne M and Briend A. (2018) Improving screening for malnourished children at high risk of death: a study of children aged 6–59 months in rural Senegal. Public Health Nutr. 2018 Dec 3:1-10. doi:10.1017/S136898001800318X.

Research snapshot1

Taking anthropometric measures in avillage near Niakhar, Senegal, 1983

Mic

hel G

aren

ne

16


. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Research SnapshotsRisk factors for acutely malnourished infantsaged under six months

Acute malnutrition (wasting) in infantsaged under 6 months (< 6m) is oenneglected. Worldwide, some 8.5 mil-lion infants < 6m are affected, yet re-

cent World Health Organization (WHO) mal-nutrition guidelines highlight numerous evidencegaps on how to best manage them. To inform fu-ture research, policy and programming this studyaimed to identify risk factors associated withinfant < 6m wasting through a secondary dataanalysis of nationally representative, cross-sectionalDemographic and Health Surveys (DHS) con-ducted in the last 10 years. Wasted infants < 6m(weight-for-length z-score (WLZ) <-2) were com-pared to non-wasted infants (WLZ ≥-2). Simpleand adjusted (for infant age, sex, socio-economic

Research snapshot1

1 pending publication in peer reviewed journal

1 Munirul Islam M, Arafat Y, Connell N, et al. Severe malnutritionin infants aged <6 months—Outcomes and risk factors in Bangladesh: A prospective cohort study. Matern Child Nutr. 2019;15:e12642. https://doi.org/10.1111/mcn.12642

status) logistic regression was used to calculateodds of wasting associated with household-related,maternal-related and infant-related risk factors.

A total of 16,123 infants < 6m were analysedfrom 20 countries. Results showed that multiplerisk factors were statistically associated with wasting.ese included: poverty (odds ratio (OR) 1.22 (95%CI 1.01-1.48, p=0.04)); low maternal body massindex (BMI) (adjusted OR (aOR) 1.53(1.29-1.80,p<0.001); small infant size at birth (aOR 1.32(1.10-1.58, p<0.01)); delayed start of breastfeeding (aOR1.31(1.13-1.51, p<0.001)); infant having had a pre-lacteal feed (aOR 1.34(1.18-1.53, p<0.001)); recenthistory of diarrhoea (aOR 1.37(1.12-1.67,p<0.01));and disempowered mother (experiences violence,

does not make decisions about health issues, doesnot engage with health services such as antenatalcare, does not give birth in a health facility).‘Protective’ factors associated with significantly de-creased odds of infant < 6m wasting included:mother being educated (OR 0.64(0.54-0.76, p<0.001));mother in work (OR 0.82(0.72-0.94, p<0.01)); infantcurrently being breastfed (aOR 0.62(0.42-0.91,p=0.02)) and infant having been exclusively breastfed(aOR 0.84(0.73-0.97, p=0.02).

Infant < 6m wasting is a complex, multifactorialproblem associated with many risk factors. Manyof the risk factors identified in this study are bio-logically plausible and/or socially important andshould therefore be considered when assessing andmanaging infants < 6m. Although supporting breast-feeding is core to future interventions, this alone isunlikely to be sufficient; strategies should involvemultiple sectors beyond just health and nutrition.

Severe acute malnutrition (SAM) affectsaround four million infants under six monthsof age (u6m) worldwide, but evidence un-derpinning their care is limited and of low

quality. e purpose of this study was to identifyrisk factors for infant u6m SAM and describe theclinical and anthropometric outcomes of treatmentwith current management strategies to informfuture research and policy. A prospective cohortstudy was undertaken in Barisal district, Bangladesh,among one group of 77 infants u6m with SAM(weight‐for‐length z‐score (WLZ) <−3 and/orbipedal oedema) and a second group of 77 infantsu6m without SAM (WLZ ≥−2 to <+2, no oedema,mid‐upper‐arm circumference ≥125 mm). All wereenrolled at four to eight weeks of age and followedup at six months. Infants identified with SAM werereferred to existing services (inpatient treatment).

Despite referral, only 13 (17%) reported for in-patient care. At six months, 18 (23%) infants withSAM at the outset still had SAM and three (3.9%)had died. In the non‐SAM group, one child devel-oped SAM and none died. e SAM group hadsignificantly more stunting (62% vs. 15%), moresevere stunting (40% vs. 0%), and more underweight(68% vs.7%) compared to the non-SAM Group.

Risk factors associated with infant SAM in-cluded non‐exclusive breastfeeding at enrolmentand at follow-up, age at enrolment, years of ma-ternal education, access to electricity and mothernot satisfied with breastfeeding at enrolment.Mothers of the SAM infants were significantlylighter, shorter and had lower MUAC than non-SAM mothers; only MUAC showed a clinicallymarked as well as statistically significant difference

between the two groups. Maternal mental healthwas worse in mothers of SAM infants.

e authors conclude that current treatmentstrategies have limited practical effectiveness. emain reason for this is poor uptake of inpatientreferral. World Health Organization recommen-dations and other intervention strategies in out-patient‐focused care for malnourished, clinicallystable infants u6m need to be tested. Better casedefinitions of nutritionally at‐risk infants areneeded. Breastfeeding support is likely central tofuture treatment strategies but may be insufficientalone. Maternal factors should be consideredwhen evaluating potentially at-risk infants.

Outcomes and risk factors for infants under six months old withsevere acute malnutrition in Bangladesh Research snapshot1

Short malnourished children andfat accumulation with foodsupplementation

Supplementary feeding programmes(SFPs) commonly do not to measurethe mid-upper arm circumference(MUAC) of children whose length is

<67 cm. These shorter children are enrolled intreatment only if they meet weight-for-heightz-score (WHZ) criteria, which is less closelylinked to risk of mortality and misses childrenat high risk who would have been identified byMUAC. This is based on expert opinion thatthe supplementation of shorter children withlow MUAC and WHZ ≥−2 may increase therisk of excessive fat accumulation. The aim of

this study was to assess if shorter children gainmore fat than taller children when treated formoderate acute malnutrition (MAM) diagnosedby low MUAC alone.

In this observational study children were in-cluded aged 6 to 23 months with a MUACbetween 115 and 125 mm and a WHZ score of≥−2. Based on length at admission, childrenwere categorised as short if they were <67 cmand long if ≥67 cm. Linear mixed-effects modelswere used to assess body composition using deu-terium dilution and skinfold thickness. Aer 12

1 Fabiansen C, Phelan K.P.Q., Cichon B, et al. Short Malnourished Children and Fat Accumulation With Food Supplementation. Pediatrics. 2018;142(3):e20180679

Research snapshot1

weeks of supplementation there was no differencein change in fat mass index (−0.038 kg/m2, 95%confidence interval [CI]: −0.257 to 0.181, P =.74) or fat-free mass index (0.061 kg/m2, 95%CI: −0.150 to 0.271, P = .57) in short childrenversus long children. In absolute terms, the shortchildren gained both less fat-free mass (−230 g,95% CI: −355 to −106, P < .001) and fat mass(−97 g, 95% CI: −205 to 10, P = .076). ere wasno difference in changes in absolute subscapularand triceps skinfold thickness and z scores (P >0.5). Results demonstrate that short childrenwith low MUAC do not gain excessive fat duringsupplementation. On this basis, the authors rec-ommend that all children ≥ 6 months with lowMUAC be included in SFPs, regardless of length.

17

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Research Snapshots

Gender dynamics of phone ownership and use in a FreshFood Voucher scheme in Ethiopia

The World Food Programme (WFP)began its Fresh Food Voucher (FFV)programme in three woredas2 inAmhara Region, Ethiopia, in October

2017, to promote consumption of fruits, veg-etables and animal-source foods and therebyimprove dietary diversity. e programme isbased on restricted vouchers that are redeemedthrough mobile phone payments to fresh-foodvendors. Even though a preliminary study sug-gested otherwise, a few months into the pro-gramme beneficiary-monitoring data revealedthat some FFV clients did not own a phone orwere not accustomed to using one. is led tochallenges for some women in redeeming theirvouchers at vendors. In response, a study wasconducted to understand the dynamics of phoneownership, access to phones and social normsrelated to phone use and ownership in commu-

nities and households participating in the FFVprogramme. e study took place in Ayo Medaand Hamaro villages and Dire Roka town inHabru woreda, Amhara Region, in 2018.

A total of 75 women and 10 men were inter-viewed and 18 focus group discussions heldwith women’s associations, female participantsof the FFV programme and their spouses. Resultsshowed that most FFV clients in the two villages,the majority of whom were illiterate or educatedto primary level and engaged in farm labour,did not own a phone (72% and 76% of womeninterviewed in Ayo Meda and Hamaro respec-tively). A much higher proportion of femaleparticipants (64%) owned a phone in urbanDire Roka, where women were more likely tobe educated to secondary level and engaged ina broader range of livelihood activities. Most

women who did not own a phone borrowedtheir spouse’s device, while others borrowedphones from friends and family to redeem FFVvouchers. No beneficiaries had missed theirFFV entitlement due to the lack of a phone.

Discussion revealed minimal conflict asso-ciated with women owning and using mobilephones at community and household levels.Several advantages of providing phones to femaleFFV clients were identified, including empow-erment; increased autonomy and safety; newskills; and access to mobile payment platforms.From WFP’s perspective, benefits of femalephone ownership include the ability to reachfemale FFV participants for programme moni-toring and to provide health-promotion messages,and the reduced need for WFP monitors atmarket/distribution points.

Based on these results, WFP will continuewith plans to provide phones as part of the FFVscheme, provided that a risk assessment of allprogrammatic challenges related to the provisionof phones to clients (such as protocols for lostand stolen phones) is completed. In addition,male spouses of FFV clients must be engaged;supporting solar/electric charging facilities inFFV villages should be explored; and partnershipsshould be built with local women’s groups andassociations in support of the programme.

1 Akabwai P and Haaij H. (2018) Fresh food vouchers: gender assessment on phone ownership and utilization. Available from: https://www.ennonline.net/resources/ genderdynamicsfreshfoodvouchers

2 Woredas (districts) are the third-level administrative divisions of Ethiopia. They are further subdivided into a number of wards (kebele) or neighborhood associations, which are the smallest unit of local government in Ethiopia.

A review of wet nursing experiences,motivations, facilitators and barriers

Wet nursing, the practice of awoman breastfeeding a non-bi-ological child, is recommendedin the World Health Organiza-

tion/UNICEF Global Strategy on Infant andYoung Child Feeding (2006) in situations wherematernal breastfeeding is not possible. A literaturereview of open access evidence of medical,social/cultural and religious factors that supportor hinder wet nursing was undertaken by Savethe Children to understand how this recom-mendation can be applied.

Results show that wet nursing has been widelypracticed since 2000 years BCE and has histor-ically been widely and positively accepted bysociety, culture and religion, although it has in

some contexts been used as a form of slavery(ancient Rome and the southern United Statesduring the 18th and early 19th centuries). Since1500, wet nursing has mostly been practicedwithin close circles of families and friends toensure child survival. It has declined as apractice with increased use of artificial feedingand marketing by formula manufacturers.

Factors identified in the review that facilitatewet nursing include wide acceptance within so-ciety, culture and religion; knowledge of theimportance of breastmilk; when wet nurses andmothers/caregivers know each other; availabilityof milk-sharing websites; appropriate supportfrom health facilities and authorities; access tolactation consultants or nurses equipped to pro-

vide support for wet nursing; and breast-milkscreening. Factors identified that hinder wetnursing include availability and promotion ofartificial feeding; fear of disease transmission;practical limitations for wet nurses (such as costof travel); unwillingness to wet nurse outsideknown relationships; lack of facilities (milkbanks; milk storage; pasteurisation); and lackof protocols and support from health authorities.

A significant limiting factor of the reviewwas a paucity of studies or documented experi-ences on wet nursing in emergencies and lowand middle-income countries, and a consequentlack of guidelines on the operationalisation ofwet nursing in emergency contexts. is is acritical gap area to address to put global recom-mendations into practice.

1 Teshome, S. (2018) Wet-nursing: A review of wet nursing experiences, motivations and factors that helped or hinder wet nursing. Save the Children UK https://www.ennonline.net/wetnursingreview

Research snapshot1

Research snapshot1

WFP

/Mic

hael

Tew

elde

WFP Food Voucher Programme,Ethiopia, 2018

18


Research .............................................................

Barrier analysis of infant and young child feeding andmaternal nutrition behaviours among IDPs in northernand southern Syria By Shiromi Michelle Perera

Shiromi Michelle Perera is aTechnical Officer with theTechnical Unit atInternational Medical Corps(IMC), Washington DC.Shiromi has coordinated and

conducted barrier analysis assessments inSierra Leone, Jordan, Lebanon, Turkey,Afghanistan, South Sudan, Ethiopia and Sudan.

The author would like to acknowledge thesupport provided by Majd Alabd (Consultant),Bonnie Kittle (Consultant), SuzanneBrinkmann (International Medical Corps),Andi Kendle (International Medical Corps) and

Location: Syria What we know: Promoting and supporting optimal infant and young child feeding (IYCF) practices and maternal nutritionare essential interventions in crisis-affected populations.

What this article adds: A barrier analysis regarding several key infant feeding and maternal nutrition practices wascommissioned by UNICEF in Syria in 2017. It was born of little success in existing interventions to change prevalentpractices. Significant barriers and enabling factors to change were determined among ‘doers’ and ‘non-doers’ with regardto exclusive breastfeeding, minimum dietary diversity and an additional meal per day in pregnancy. ese included lackof correct knowledge and misconceptions; lack of access to markets and availability of diverse foods; inability to affordfood; stress; lack of support from critical family members; and lack of time to prepare meals. Improved access to IYCFand maternal nutrition services are needed, with support for mothers and pregnant women more effectively integratedinto other sectors, particularly food security, agriculture, livelihoods and reproductive health. Detailed recommendationsare informing current and future programming by Nutrition Cluster partners in specific districts and have widerrelevance in Syria. is experience reflects that barrier analysis is possible in an insecure/emergency contexts.

Background e Syrian crisis continues to be one of theworst humanitarian and protection crises ofour time, taking a significant toll on the lives ofthe Syrian people and impacting all basic needs.Over half of the population has been internallydisplaced, resulting in many families living incamps, informal settlements and collectivecentres throughout the country. In 2016 theWhole of Syria (WoS) Nutrition Sector responsereached 3.4 million children and pregnant andlactating women (PLW) affected by the crisiswith a range of preventative and therapeuticnutrition interventions (WoS, 2016). Includedin this were efforts by the Nutrition Clusterand its partners to promote and support optimalinfant and young child feeding (IYCF) practices,as well as maternal nutrition, as priority life-saving interventions. Nevertheless, a knowledge,

attitudes and practices (KAP) survey, conductedin February 2017, indicated that the prevalenceof certain IYCF behaviours was either low orlargely unchanged compared to the results ofnutrition surveys carried out before the response.ree IYCF behaviors in need of further inves-tigation were: exclusive breastfeeding (EBF)(30.9%); complementary feeding for minimumdietary diversity (CF-MDD) (57.3%); and eatingan extra meal during pregnancy (40.3%). Asco-lead of the Nutrition Cluster, UNICEF com-missioned a barrier analysis (BA) to determinethe reasons behind prevalent IYCF and maternalnutrition practices among internally displacedpeople (IDP) in camp and urban settings in theAleppo, Idlib and Dar’a Governorates in orderto better tailor partner programme activities.is article summarises findings and recom-mendations from this first-ever BA on feeding

practices in Syria, conducted in August 2017 innorthern and southern Syria.

MethodologyAn initial two-day training of trainers (TOT)was conducted in Gaziantep, Turkey, amongNutrition Cluster partner organisations fromnorth and south Syria. e training includedcontent on internationally recognised BA guide-lines (Kittle, 2013), questionnaire developmentand interviewing skills and the use of KoBo, afree, open-source tool for mobile data collection.Training was cascaded to 15 data collectors andtwo supervisors in the north (Physicians AcrossContinents and Human Appeal) and 10 datacollectors and two supervisors in the south(Syria Relief and Development).

Following this, three cross-sectional surveyswere carried out in camp IDP and urban IDP

Adelaide Challier (Save the Children). Theauthor also acknowledges the TurkeyNutrition Cluster and its partners, includingWigdan Madani (UNICEF), Mona Maman(Physicians Across Continents) and Dr SajaAbdullah (Whole of Syria Cluster Coordinator),as well as the assessment trainers, supervisorsand data collectors. Special thanks to thefollowing trainers: Kotham Saaty (PhysiciansAcross Continents), Anas Barbour (HumanAppeal) and Feras Ahmed (USSOM), AmerBasmaci (Consultant) and Marwa AlSubaih(Syria Relief and Development). The author isalso thankful to the communities andespecially the Syrian mothers who gave their

time to be part of this assessment. Specialthanks go to UNICEF for financial support toconduct this barrier analysis assessment.

This barrier analysis report was made possibleby the generous support of the Americanpeople through the United States Agency forInternational Development (USAID) in thetechnical support for this assessment andUNICEF for its implementation. The contentsare the responsibility of the Technical RapidResponse Team (Tech RRT) and the NutritionCluster and do not necessarily reflect theviews of UNICEF, USAID or the USGovernment.

19

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Researchlocations in north and south Syria, chosen ac-cording to the presence of operational nutritionprogrammes and security/accessibility. Purposivesamples of 90 Syrian mothers (45 ‘doers’ (thosewho practice the behaviour) and 45 ‘non-doers’(those who do not)) were selected for each be-havior. Groups included mothers of infants aged0-<6 months exclusively breastfeeding (or breast-feeding but not exclusively); mothers of childrenage 6-23 months feeding children meals con-taining foods from at least four of seven specifiedfood groups per day (or not); and pregnantmothers who consume an extra meal per dayduring pregnancy (or not).

Mothers were first screened and classified as‘doers’ or ‘non-doers’, aer which they wereasked questions according to their classificationto identify which of the 12 specified determinantsof behaviour change acted as barriers to theparticular behavior among ‘non-doers’ andwhich facilitated its adoption among ‘doers’.Data from closed-ended questions were collectedwith KoBo using mobile devices, which is anuncommon adaptation to the BA approach butworked well in this emergency context. Codingof qualitative responses was achieved throughan iterative group process with each team, usingvarious online applications, depending on con-nectivity. Codes were then tabulated and recordedfor analysis in a BA tabulation Excel spreadsheet.Findings were interpreted by the BA team andpresented at a results workshop of participatingpartners, and later with Nutrition Cluster partnersto help inform interpretation of results and rec-ommendations.

ResultsIn total, 551 mothers were interviewed in north(n=271) and south Syria (n=280). e northwas stratified into camp IDP and urban IDP lo-cations; specifically, Atmeh Camp in Idlib Gov-ernorate, Al’Mara District in Idlib Governorateand Jebel Saman District in Aleppo Governorate.e south was stratified into urban IDP locationsin Dar’a Governorate; specifically, Tafas andHrak Districts. In total, 11 determinants in thenorth and 5 determinants in the south werefound to be significant for EBF; 11 determinantsin the north and 8 determinants in the southfor CF-MD, and 11 determinants in the northand 9 determinants in the south for an extrameal during pregnancy.

Exclusive breastfeedingCommon barriers experienced by ‘non-doers’included maternal stress, perception that thebaby is unsatisfied, maternal anemia, physicalissues with breastfeeding for the mother (breastproblems) and baby (stomach problems, colic,teething) and lack of support from the husband.Mothers and mothers-in-law were described by‘non-doers’ as people who disapprove of EBF.Factors that facilitated EBF indicated by ‘doers’were knowledge of IYCF, family support, privatespaces to breastfeed, access to and consumptionof diverse foods by the mother in order toproduce milk, and having enough and continuousmilk. Barriers identified by ‘doers’ of particular

relevance were market-access issues and concernsrelated to breast problems (pain in breasts orinflammation in nipples).

While both ‘doer’ and ’non-doer’ mothersdemonstrated adequate knowledge about thepositive and negative consequences of EBF, theyhad several misconceptions, such as thinkingthat breastfeeding is a “waste of time”, the babyis le unsatisfied, breastfeeding changes breastshape and leads to maternal health problems(loss of weight, illness, loss of calcium, loss ofimmunity) and problems in the family. Additionalsignificant determinants were perceived access(‘doers’ and ’non-doers’ stated it was “somewhatdifficult” to get the support needed to EBF);perceived cues for action/reminders (‘non-doers’were more likely to say it was “somewhat difficult”to remember to give only breastmilk for the firstsix months); perceived risk of malnutrition anddiarrhoea; perceived severity of malnutritionand diarrhoea; perceived efficacy of EBF (‘doer’and ‘non-doers’ do not fully understand the re-lationship between EBF and malnutrition/ diar-rhoea); divine will (‘doers’ were more likely tosay that Allah may cause malnutrition or diarrhea);and culture (‘doers’ were more likely to say thatthere are cultural rules/taboos against EBF).

Minimum dietary diversity (MDD) incomplementary feedingSome of the barriers for ‘non-doers’ includednot enough food preparation time for mothersdue to work outside the home, the child not ac-cepting prepared food, the child being sick orhaving thyroid problems, lack of food diversity

Physicians across Continent data collectorinterviewing mother for the BarrierAnalysis, Northern Syria, 2017

Phys

icia

ns a

cros

s Co

ntin

ents

in markets and being unable to afford diversefoods. ‘Non-doers’ indicated that sisters andaunts disapprove of feeding a diverse diet tochildren. ‘Doers’ indicated several facilitatingfactors, such as support from husband and familymembers, access to markets, availability of foodsin the house, enough time to feed their child,the child loving/wanting food, having electricityto cook food, and receiving advice about com-plementary feeding. Stated barriers for ‘doers’ ofparticular relevance were interference by familymembers, distance to markets and lack of timedue to the mother working outside of the house.

Some lack of knowledge and misconceptionswere found, such as mothers perceiving that adiverse diet does not provide immunity andleads to children getting sick from food poisoningor intestinal complications. Additional significantdeterminants were perceived access (‘non-doers’indicated that it was “very difficult” to get foodfrom at least four of the seven food groups),perceived cues for action/reminders (‘non-doers’were more likely to say it was “somewhat difficult”to remember to include foods from at least fourof the seven food groups during meal prepara-tion), perceived risk of malnutrition, perceivedseverity (‘doers’ considered becoming malnour-ished as only “somewhat serious”), perceivedaction efficacy (‘non-doers’ did not fully under-stand the relationship between a diverse dietand malnutrition), divine will (‘non-doers’ weremore likely to say Allah causes malnutrition)and culture (‘doers’ were more likely to say thereare cultural rules/taboos against feeding theirchild a diverse diet).

20

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Research

ReferencesNutrition Cluster IYCF-E Operational Strategy 2017- 2020.

Whole of Syria (WoS) Nutrition Sector Bulletin, Issue 2July-December 2016.

Kittle, Bonnie. 2013. A Practical Guide to Conducting aBarrier Analysis. New York, NY: Helen Keller International.

Extra meal during pregnancySome of the barriers for ‘non-doers’ includedpregnancy-related sickness (vomiting, pressure,stomach pain), markets being far away, lack ofmoney to buy foods, lack of privacy, lack oftime to cook food, not receiving food basketsfrom non-governmental organisations and regulardisplacement. ‘Non-doers’ indicated that no onewould disapprove of eating an extra meal duringpregnancy. Facilitators for ‘doers’ included havinga supportive husband, availability of food inthe house and accessible markets, kitchen ap-

pliances to store and cook food, advice fromnutrition workers, provision of an NGO foodbasket, and the mother having an appetite andnot being stressed or sick. Barriers for ‘doers’ ofparticular relevance included lack of availabilityof food in the house and the mother being tootired or lacking an appetite to eat an extra meal.

Some lack of knowledge and misconceptionswere revealed, with mothers perceiving that anextra meal leads to weight gain, sickness, feelinglazy and increased blood pressure when eating

Behaviour Recommendation

General 1. Scale up IYCF and maternal nutrition programming (through existing mother support groups approach and care groups).2. Continuously assess needs and access to markets and services as each wave of displaced individuals resettle in programme areas to understand evolving needs.

Exclusivebreastfeeding

1. Conduct one-to-one counselling, skilled support and educational/support sessions (within health facilities, antenatal and post-natal clinics) to improve knowledge on EBF, address misconceptions and assess and address breastfeeding problems. 2. Refer mothers as needed for nutrition assessment and support (micronutrient supplementation, treatment for anaemia), food distribution and rations and psychosocial support to reduce stress. 3. Scale up community-level breastfeeding support through the integration of IYCF support activities into midwifery and reproductive health services and provision of baby-friendly spaces to give privacy to breastfeeding mothers. 4. Hold discussions with families (husbands, mothers and mothers-in-law) about how to support mothers to breastfeed exclusively. 5. Policy actions: reinforce and advocate for adherence to the International Code of Marketing of Breastmilk Substitutes (given prevalent large-scale distribution of infant formula); develop guidelines for physicians in supporting BF; advocate for the Integrated Management of Childhood Illnesses (IMCI) in all health facilities.

Minimum dietarydiversity incomplementaryfeeding

1. In one-to-one counselling and group sessions with mothers, include information on appropriate complementary feeding and dietary diversity, feeding sick children and appropriate WASH practices. 2. Conduct assessment of infrastructure at community and household levels (electricity, food storage, water access, etc). 3. Conduct food security assessments and interventions to determine access, food availability and diversity.4. Create community or home gardens and establish mobile markets to increase access and availability to diverse foods. 5. Provide cash/vouchers, especially in times of electricity outages, and expand food basket distribution, especially to vulnerable groups (households with infants under two years old and PLW).6. Provide community-level support through development and sharing of locally appropriate recipes; cooking demonstrations; meal planning and family education sessions on complementary feeding.7. Educate families during support group sessions, house visits and community outreach on the importance of supporting mothers to feed children a diverse diet.8. Hold group discussions with aunts and sisters to discuss the benefits of diverse diets and how they can be supportive towards mothers.

Consumption ofan extra meal inpregnancy

1. In one-to-one counselling and group sessions with mothers, include information on the importance of an extra meal per day in pregnancy, healthy weight gain, management of pregnancy-related symptoms and addressing cultural myths (such as food allergies, sickness and reactions to eating certain foods in pregnancy).2. Referral of pregnant mothers, as appropriate, for further nutrition assessment and support, blanket feeding, food distribution and rations and psychosocial support to reduce stress.3. Provide community-level support by ensuring rapid response to newly displaced mothers; set up mother and child friendly spaces; develop community gardens; establish mobile markets to increase access to fresh fruits and vegetables; establish cash/food voucher programme and income-generating activities for vulnerable families with PLW; and provide information in group education sessions on the importance of eating an extra meal in pregnancy.

Table 1 Summary of recommendations

certain foods. Additional significant determinantswere perceived access (‘non-doers’ were morelikely to indicate that it was “very difficult” toaccess what they need to eat an extra meal),perceived cues for action/reminders (‘non-doers’were more likely to say it was “very difficult” toremember to eat an extra meal), perceived riskthe baby will be born too weak and small, per-ceived severity (‘doers’ and ‘non-doers’ perceivedthe baby being born too weak and small as“very serious”), perceived action efficacy (‘non-doers’ did not fully understand the relationshipbetween eating an extra meal and giving birthto a healthy baby), divine will (‘non-doers’ weremore likely to say that Allah wants them to eatan extra meal) and culture (‘non-doers’ weremore likely to say there are cultural rules/taboosagainst eating an extra meal).

Discussion andrecommendationsis article reflects that barrier analysis is possiblein an insecure/emergency context. It appliedKoBo, which is not commonly used with BA,and adapted training and coding methods forremote application. Results show various deter-minants that create barriers to mothers properlypracticing the three assessed behaviours, includinglack of correct knowledge and misconceptions;lack of access to markets and availability ofdiverse foods; inability to afford food; stress;lack of support from critical family members;and lack of time to prepare meals. Results suggestthat improved access to IYCF and maternal nu-trition services are needed and that support formothers of infants and young children and PLWmust be more effectively integrated into othersectors, particularly food security, agriculture,livelihoods and reproductive health to ensurethat the multiple needs of this group are addressed.Recommendations, summarised in Table 1, buildon existing programme activities and plans. Al-though recommendations were tailored to specificdistricts, they will also likely benefit similar pro-gramming locations in northern and southernregions. Following this assessment, the NutritionCluster partners held a workshop to begin plan-ning how to move forward with these recom-mendations. e author developed a social be-havior change strategy to aid the Cluster in thedesign, implementation and monitoring andevaluation of the recommended activities.

e full report can be found at https://www.en-nonline.net//resources/barrieranalysiycfsyria

For more information, please email ShiromiPerera at [email protected]

Perera, S.M. (2019) Barrier analysis of infant and youngchild feeding and maternal nutrition behaviours amongIDPs in northern and southern Syria. Field Exchangeissue 59, January 2019. www.ennonline.net/fex

Micronutrient powder distribution strategies to increasecoverage and adherence among children aged six to 23months as part of an IYCF strategy in CambodiaBy Sarah Gibson, Hou Kroeun and Gary Mundy

Sarah Gibson is a publichealth nutritionist andcurrently works as ananalyst at the Children’sInvestment FundFoundation (CIFF). She

holds an MSc in nutrition for globalhealth from the London School ofHygiene and Tropical Medicine.

The project described in this article was implemented by Helen Keller International and funded by Irish Aid. The authorswould like to thank the Cambodian Ministry of Health for their collaboration and support of this project.

Hou Kroeun is the Deputy CountryDirector for Programmes at Helen KellerInternational Cambodia and theChairman of the Scaling Up NutritionCivil Society Alliance (SUN CSA) inCambodia. He holds a master’s degree in

public health and has 20 years of experience in the designand management of integrated household food productionand nutrition-related programmes in Cambodia.

Gary Mundy is currentlySenior RegionalEvaluation Adviser forthe British Council inEast Asia. He has spentthe last 15 years

supporting the evaluation and researchwork of a number of non-governmentalorganisations in Asia and Africa.

Location: Cambodia What we know: Micronutrient powder (MNP) distribution improves micronutrient statusand reduces anaemia and iron deficiency in children aged six to 23 months.

What this article adds: Helen Keller International (HKI) and the Ministry of Health (MoH)in Cambodia implemented an integrated infant and young child feeding (IYCF)/MNP pilotprogramme for children aged 6-23 months between 2012 and 2014. ree strategies wereused to distribute MNPs and provide MNP and IYCF counselling: community-basedvolunteers direct to households; collection from health centres (fixed); and a combinationof the two (mixed). Results of an endline cross-sectional survey and coverage data (final 11months) showed significantly higher average coverage for community sites compared tomixed and fixed, and for mixed compared to fixed. Programme adherence and differencesin caregiver knowledge were significantly higher for community compared to fixed sites; nosignificant differences were found between community and mixed. Community-baseddistribution has the potential to improve coverage of MNP programmes among childrenage 6 to 23 months in Cambodia: combining community and facility-based distributionmay be more feasible and cost-effective.

BackgroundMicronutrient deficiencies are a major publichealth challenge in Cambodia. Over half (56%)of children six to 59 months are anaemic, withnotably higher prevalence among infants agednine to 11 months (83%) (National Institute ofStatistics, 2014). ere are multiple causes ofanaemia, including deficiencies in iron and zinc,infection and genetic haemoglobin disordersand, for infants and young children, inadequatedietary intake and poor infant and young childfeeding (IYCF) practices (Balarajan et al, 2011).Only 30% of Cambodian children aged six to23 months receive a minimum acceptable dietand, particularly in rural households, comple-mentary foods tend to lack in quantity, diversityand animal-source foods (National Institute ofStatistics, 2014).

Micronutrient powders (MNPs) are a provenhome-fortification strategy recommended bythe World Health Organization (WHO) to im-prove micronutrient status and reduce anaemiaand iron deficiency in children aged 6 to 23

months (WHO, 2016). e efficacy of large-scale MNP programmes in reducing anaemiain this age group has been clearly demonstratedin various low- and middle-income countries,including Cambodia (WHO, 2016). Despite this,there is a lack of implementation research ex-ploring which distribution strategies are mosteffective in supporting sustained coverage andadherence to MNPs at scale.

Current global MNP programmes typicallyuse more than one channel of distribution; eitherfacility-based, community-based or market-based(Jefferds et al, 2011). Community-based distri-bution has been identified as one of the bestways to reach rural populations in resource-limited settings where existing coverage may bechronically low (Micronutrient Forum, 2008).Integrating MNPs into well-established com-munity-based services may provide an increasedopportunity for long-term sustainability, a criticalcomponent for successful programme scale-up.A recent review of nine global MNP interventionsthat delivered MNPs at no charge through the

national health sector, found that coverage wasgenerally higher when distribution took place atthe community level (Reerink et al, 2017).

Helen Keller International (HKI) worked withthe Ministry of Health (MoH) in Cambodia toimplement an integrated IYCF/MNP pilot pro-gramme among children aged 6-23 months ofage between 2012 and 2014, with the goal of re-ducing anaemia in this age group. is post hocanalysis looks at the different distribution strategiesemployed and their association with coverage,adherence and knowledge to help identify themost effective methods to improve delivery, cov-erage and use of MNPs in Cambodia.

MethodsPrior to the programme MNPs were availableat health centres (HCs), but household usagewas low. In response the programme employedthree strategies for distributing MNPs to eligiblehouseholds: (1) Community-based delivery: vil-lage health volunteers distributed MNPs directlyto households (five operational districts (ODs));(2) Fixed-site strategy: households within 10km of an HC collect MNPs from HC (1 OD);(3) Mixed strategy: Households within 5 km ofan HC collect MNP from HC staff and householdsfurther than 5 km received MNPs from villagehealth volunteers (VHVs) (1 OD).

Data were used from two sources: an endlinecross-sectional survey (n=800), conducted aer20 months of MNP distribution (no data werecollected on coverage or adherence at baselineas very few households (3%) had previouslyused MNP before) and monthly coverage datacollected over the final 11 months of MNP dis-tribution. e latter reported on monthly coverageestimates across the seven ODs, each coveringapproximately 50,000-100,000 people with fiveto 10 HCs (MoH, 2016).

A box containing 15 sachets of MNPs wasprovided each month to caregivers, along with

21

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Research

22

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Researchpreparation and feeding instructions. Both VHVsand HC staff were trained to provide standardisedmessages to participating families. A total of1,770 VHVs and 404 HC staff were trained todistribute and counsel on MNP use and improvedIYCF practices. Mass media spots and counsellingsessions provided educational messages with anemphasis on WHO-endorsed complementaryfeeding practices and continued breastfeedinguntil two years of age and beyond.

e Cambodian National Nutrition Pro-gramme (NNP) recommended that all childrenaged six to 23 months be allocated 15 sachets ofMNPs for consumption on a flexible scheduleevery month. Coverage was defined as the per-centage of children aged six to 23 months whoreceived 15 MNP sachets in the last month. Ad-herence was defined as the percentage of childrenaged six to 23 months who consumed all 15 sa-chets during that time period. In January 2015,aer the programme ended, the MoH began torecommend a revised dosage of 10 sachets permonth per child, so partial adherence was in-cluded in this analysis.

ResultsCoverage Table 1 summarises monthly coverage acrosseach of the three distribution strategies overthe final 11 months of distribution. Results

showed significantly higher average coverageamong the VHV distribution ODs, comparedwith the mixed distribution OD (73%, 64%;p=<.001), and compared with the OD usingfixed-site distribution (73%, 30%; p=<.001).ere was also significantly higher average cov-erage among the mixed distribution OD com-pared with the fixed-site distribution OD (64%,30%, p=<.001).

Between January and July 2014 the fixed-site strategy achieved 16% coverage (range: 12-23%; p=.03 for difference in coverage betweenJanuary and July). Aer observing higher coveragein the VHV and mixed distribution model, thefixed-site strategy changed to a mixed strategyand coverage increased to 55% over the lastfour months of the programme (range: 37-66%;p=<.001 for difference in coverage between Julyand November).

Adherence Table 2 summarises adherence estimates collectedat endline. ere was no significant differencein adherence, as measured by consumption of15 sachets in the past one month, betweenVHV-distribution ODs and the mixed-site OD(60%, 54%). ere were positive differences inadherence between VHV-distribution ODs andthe fixed-site OD (60%, 41%; p=.001), but nosignificant differences in adherence between the

mixed-site and fixed-site ODs (54%, 41%).

When measuring the consumption of 10 sa-chets in the past month, significantly more re-spondents in the mixed-site OD reported partialadherence, compared with the VHV-site ODs(86%, 70%; p=.002) and the fixed-site OD (86%,57%; p=<.001). Differences between the VHVsite and the fixed site were not significant. Aerdisaggregating the adherence data in the mixed-site OD, it was found that 90% of those who re-ceived MNPs from VHVs consumed at least 10sachets per month, compared to 79% amonghouseholds who collected MNPs from HCs.However, this finding was not statistically sig-nificant because of the small sample size of themixed-site OD (n=100).

Caregiver knowledge Table 3 summarises caregiver knowledge of fourpromoted benefits of MNP for young childrenat endline across the three distribution ap-proaches. ere were significant and positivedifferences in knowledge of three out of thefour promoted benefits between the VHV andfixed sites: promoting optimal growth (76%,54%; p=<.001), increasing immunity (35%, 19%;p=.004), and preventing iron deficiency (19%,7%; p=.01). However, there was no significantdifference in knowledge across respondents re-ceiving either VHV distributed MNPs or themixed model. e fixed-site group was signifi-cantly more likely not to know of any benefitsof MNPs compared to the VHV (p=<.001) andmixed groups (p=.02).

DiscussionMNP coverage was considerably higher in targetODs where VHVs were responsible for full orpartial distribution, compared to the OD usinga fixed-site strategy. Full community-based dis-tribution of MNPs by VHVs showed the highestlevel of coverage. Aer the fixed-site strategychanged to a mixed model for the final fourmonths of the programme, coverage rates in-creased rapidly and significantly. ese findingssuggest that community-based distribution ofMNPs using VHVs is more effective than anabsolute fixed-site distribution strategy at in-creasing coverage.

Many different factors can influence intakeadherence to MNPs, such as guidance to care-givers, perceived benefits of the product, sideeffects, administration frequency and distributionmodel (de Barros et al, 2016). In target ODswhere VHVs were responsible for total or partialdistribution, children were significantly morelikely to consume 15 or at least 10 sachets ofMNP each month compared to those in thefixed-site strategy. Similarly, when examiningmaternal/caregiver knowledge on the benefitsof MNPs, it was found that where VHVs wereresponsible for total or partial MNP distribution,caregivers demonstrated better knowledge thancaregivers who collected MNPs from HCs. Lowerrates of adherence and knowledge in the fixed-site OD suggest that VHVs may play a uniquerole in the community as an effective channelof communication linking households to MNPs.

VHV model Mixed Site Fixed Site P-Value

Month (2014) Total Coverage Total Coverage Total Coverage

n n (%) n n (%) n n (%)

Jan* 1753 934(53) 425 251(59) 319 38(12)

Feb 5478 3978(73) 498 379(76) 373 52(14)

March 3600 2427(67) 335 248(74) 251 30(12)

April 3917 2601(66) 364 231(63) 273 63(23)

May* 1472 938(64) 245 144(59) 184 32(18)

June* 1400 1059(76) 224 92(41) 168 22(13)

July 4217 3102(56) 392 224(57) 294 53(18)

Aug 4765 3617(76) 443 244(55) 332 123(37)1

Sept 3660 2856(78) 341 249(73) 255 138(54)

Oct 4575 3655(80) 426 268(63) 319 201(63)

Nov 3557 2852(80) 331 228(69) 248 164(66)

Total Children 38,394 28,019(73)3,a 4,024 2,558(64)4,b 3,016 916(30)5,c <0.0012

1 In August 2014 the fixed site OD changed to a mixed strategy.2 p-value comparing average coverage between the three groups (5% significance level).3 p=<0.001 comparing differences in coverage between Jan and Nov for the VHV strategy.4 p=0.005 comparing differences in coverage between Jan and Nov for the mixed strategy.5 p=0.03 comparing differences in coverage between Jan and July for the fixed strategy. After changing to a mixed strategy, p=<0.001 between July and Nov. letters (a,b,c): comparisons between each group. Row with different letters indicate statistically significant differences.

Table 1 MNP coverage (received 15 MNP sachets in the last month) from monitoringdata during the final 11 months of programme implementation

* In January, May and June 2014 MNPs were unavailable in three ODs (using the VHV distribution strategy) because of a disrupted supply, so there are no coverage data from those ODs, resulting in smaller sample sizes during those months. For other months, the total number of children that fell within the 6-23 months age range differed, which also resulted in variations in the number of households monitored each month.

23


. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Research

Endline=800 P-value

Outcome Group Freq(%)

Child consumed 15 MNP sachets each month (out of those who used MNPs) 0.006

VHV 218(60) a.

Mixed 49(54) a, b.

Fixed 32(41) b.

Child consumed at least 10 MNP sachets each month (out of those who used MNPs) <0.001

VHV 252(70) a.

Mixed 77(86) b.

Fixed 45(57) a.

p-value indicates differences between groups at endline (unadjusted analyses)letters (a,b): comparisonsbetween each group at endline. Two rows with different letters indicate differences at 5% significancelevel; rows that share a letter = no statistical significance.

Table 2 Adherence to MNP across the three distribution strategies at endline

ReferencesBalarajan Y, Ramakrishnan U, Ozaltin E, Shankar A,Subramanian S. Anaemia in low-income and middle-incomecountries. Lancet. 2011;378(9809):2123–35.

de Barros SF, Augusto Cardoso M, Cardoso MA. Adherenceto and acceptability of home fortification with vitamins andminerals in children aged 6 to 23 months: a systematicreview. BMC Public Health [Internet]. 2016;16(1):1–11.Available from: http://10.0.4.162/s12889-016-2978-0%0Ahttp://search.ebscohost.com/login.aspx?direct=true&db=a9h&AN=114454153&site=ehost-live&scope=site

Jefferds ME, Irizarry L, Timmer A, Tripp K. Unicef-cdc globalassessment of home fortification interventions 2011: Current

status, new directions, and implications for policy andprogrammatic guidance. Food Nutr Bull. 2013;34(4):434–43.

Micronutrient Forum Project UM and CB. Scaling UpMicronutrient Programs: What works and what needs morework? The 2008 Innocenti Process. Micronutr Forum. 2008.

Ministry of Health. Kingdom of Cambodia: HealthStrategic Plan 2016-2020 [Internet]. Phnom Penh,Cambodia; 2016. Available from: www.nationalplanningcycles.org/sites/default/files/planning_cycle_repository/cambodia/cambodia_nhpsp_2016-2020.pdf

National Institute of Statistics, Directorate General forHealth, ICF International. Cambodia Demographic and

Health Survey 2014. Phnom Penh, Cambodia andRockville, Maryland, USA; 2015.

Reerink I, Namaste SM, Poonawala A, Nyhus Dhillon C,Aburto N, Chaudhery D, et al. Experiences and lessonslearned for delivery of micronutrient powders interventions.Matern Child Nutr [Internet]. 2017;13(June):e12495.Available from: http://doi.wiley.com/10.1111/mcn.12495

WHO. WHO guideline: Use of multiple micronutrientpowders for point-of-use fortification of foods consumed byinfants and young children aged 6–23 months and childrenaged 2–12 years. World Health Organization. 2016.Licence: CC BY-NC-SA 3.0 IGO.

Conclusionsis analysis found that community-based dis-tribution using VHVs is more effective at max-imising coverage and promoting adherence toMNPs than fixed-site distribution from healthcentres. Although VHVs are trusted communitymembers who have been used for decades bynon-governmental organisations (NGOs) andgovernment to implement health interventions,they are not formally recognised as part of thehealthcare system of the MoH of Cambodia,nor are they permitted to deliver essential drugs.HKI has observed that VHVs operating in areaswhere NGOs are working tend to be well trained,financially supported, motivated and active.However, when NGOs phase out their support,VHV involvement and engagement oen declines.

Endline=800 P-value

Outcome Group Freq(%)

Promoting optimal growth 0.001

VHV 274(76) a.

Mixed 67(74) a.

Fixed 43(54) b.

Increasing immunity 0.01VHV 128(35) a.

Mixed 24(27) a, b.

Fixed 15(19) b.

Improving children's appetite 0.73

VHV 82(23) a.

Mixed 17(19) a.

Fixed 18(23) a.

Preventing iron deficiency anaemia 0.006

VHV 68(19) a.

Mixed 24(27) a.

Fixed 6(7) b.

Don't know <0.001

VHV 40(11) a.

Mixed 13(14) a.

Fixed 23(29) b.

p-value indicates differences between groups atendline (unadjusted analyses)letters (a,b):comparisons between each group at endline. Tworows with different letters indicate differences at5% significance level; rows that share a letter = nostatistical significance.

Table 3 Knowledge of benefits of MNPs amongcaregivers who had previously usedMNPs across the three distributionstrategies at endline

(As only one OD employed a fixed-site strategy and only oneOD employed the mixed strategy, sample sizes for these twostrategies were much lower than in the VHV strategy at endline(fixed n=100; mixed n=100; VHV n=600).

VHVs need consistent and cohesive centralisedsupport from government to ensure the sus-tainability of a large-scale MNP programme.

Large-scale MNP distribution depends onthe ability of governments to plan and budgeteffectively and their capacity to distribute productand monitor interventions. With time, countrieswill shi from external, donor-funded to gov-ernment-financed programmes, so long-termfinancing is an important consideration, par-ticularly if MNPs are to be distributed free ofcharge to the end-user. Although an absolutefixed-site distribution strategy is less burdensomeon resources, it can only be expected to reach aportion of the target population (approximatelyone quarter, based on this analysis). Communi-ty-based distribution using VHVs will be critical

to reach children aged 6 to 23 months in themost vulnerable villages across Cambodia. Amixed strategy of MNP distribution that includesboth community-based and facility-based dis-tribution may be a more cost-effective and com-prehensive model that can capitalise on existinginfrastructure and minimise the risk of over-burdening VHVs.

For more information, please contact SarahGibson at [email protected]

Gibson, S., Kroeun, H., and Mundy, G. (2019)Micronutrient powder distribution strategies toincrease coverage and adherence among children agedsix to 23 months as part of an IYCF strategy inCambodia. Field Exchange issue 59, January 2019.www.ennonline.net/fex

Mother preparing food foryoung child with micronutrient

powder, Cambodia, 2014

Hel

en K

elle

r Int

erna

tiona

l

Location: India

What we know: Maternal factors, such as poor nutrition in utero, anaemia andshort birth spacing, contribute toundernutrition in children.

What this article adds: e socio-economic, anthropometric, morbidity andbiochemical profile of 427 mothers of inpatient children with severe acutemalnutrition (SAM) admitted to four nutrition rehabilitation centres (NRCs)in India was assessed. Twenty-five per cent of mothers were low in stature(height <145 cm); 23% were underweight (body mass index (BMI) <18.5kg/m2); 5% were severely underweight (BMI <16); more than a quarter (27%)were overweight/obese (BMI ≥23 kg/m2); and 71% had moderate or severeanaemia. Of 33 mothers who were also pregnant, 41% had low mid-upper armcircumference (<23 cm). Diarrhoea, urinary tract infection and genitalinfection were common and use of family planning methods was low (23%).Findings informed the development of a maternal service package integratedwithin existing government services and according to nutrition/medical risk ofmothers. Further research is needed to test the effect of the protocol onmaternal nutrition outcomes and its operational feasibility at scale.

Development of a maternal service package formothers of children with severe acute malnutritionadmitted to nutrition rehabilitation centres in India By Vani Sethi, Praveen Kumar and Arjan De Wagt

Vani Sethi PhD is a publichealth nutritionist in theNutrition Division, UNICEFIndia, presently leading a girl’sand women’s nutritionprogramme linking nutri-

based livelihoods and empowerment andsupporting the Ministry of Health in maternalmalnutrition and anaemia programmes. She has17 years’ experience in several Indianprogramming contexts, including rural, tribal andurban slums.

Dr Praveen Kumar is apaediatrician and Professor ofPediatrics at Lady HardingeMedical College and KalawatiSaran Children’s Hospital inNew Delhi. He runs India’s

national centre of excellence on severe acutemalnutrition and supports the Ministry of Healthin the development of protocols for managementof children with severe acute malnutrition and HIVand capacity-building.

Arjan de Wagt is Chief of theNutrition Section, UNICEF Indiacountry office, New Delhi. Hehas over 25 years’ internationalexperience in several countries,including Nigeria, Zambia, New

York and Netherlands and has specialised in themanagement of severe acute malnutrition,emergency nutrition and nutrition in HIV.

The authors are grateful to Kalawati SaranChildren’s Hospital, Hindu Rao Hospital, BhagwanMahavir Hospital and Jawaharlal Nehru MedicalCollege for user of their premises in datacollection. The authors recognise and thank allpartners, consultants (Shikha Sayal, Swati Dograand Dr Neha Sareen) and advisors (Dr HPSSachdev, Dr Manju Puri and Dr Neena Bhatia) whohave supported the work highlighted in thisarticle at various stages, including the MaternalHealth Division, Ministry of Health and FamilyWelfare, Government of India, New Delhi. Thiswork was funded by UNCEF India.

Background In India 8% of children under five years oldsuffer from severe acute malnutrition (SAM)(NFHS 4, 2015-2016). Underlying causes ofSAM in children are directly and indirectlyrelated to maternal factors such as poor nu-trition in utero, short birth spacing and house-hold food insecurity. Currently 1,151 func-tional nutrition rehabilitation centres (NRCs)are set up by the Government of India (GoI)under the Ministry of Health and FamilyWelfare (MoHFW) to provide inpatient carefor children under five years old with SAMwith medical complications (~10% of burdenof SAM) (MoHFW, 2011). A typical NRChas 10 to 20 beds where mothers/ caregiversstay with admitted children for the durationof treatment (7-14 days) and attend daily

group and individual counselling sessions toprepare them for discharge.

Limited information exists on the nutri-tional profile of mothers of children withSAM in India. Facility and community-basedstudies with small sample sizes show thatmothers of children with SAM are oen un-derweight (33%-50% of mothers had a bodymass index (BMI) <18.5 kg/m2 in a study byBhandari et al, 2016), and anaemic (70%-88% in studies by Rai et al, 2015 and Nagab-hushan et al, 2017); however, more researchand national protocols are needed to supportthe effective screening and management ofmaternal malnutrition (including undernu-trition, overweight and obesity) at facilityand community levels.

24


ResearchMothers in interventionhospital receiving thehospital diet, 2018

© U

NIC

EF/2

018/

Das

25

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Research

crotoise (model 216) with 0.1 cm gradation andmid-upper arm circumference (MUAC) usingMUAC tape (procured from UNICEF supplydepartment) with a gradation of 0.1 cm. WorldHealth Organization (WHO) classifications wereused to classify short stature and thinness basedon BMI and anaemia (WHO, 1995, WHO, 2003,2011). South Asian classifications of overweight(≥23 kg/m2) and obesity (≥26 kg/m2) (WHO,2003) were also used. MUAC <23 cm and MUAC<19 cm were used as alternative cut-offs todefine malnutrition and SAM respectively (Mo-HFW, 2017). MUAC ≥30 cm was used to defineoverweight or obesity (Khadivzadeh, 2002). Ve-nous blood sample was collected from consentingmothers for haemoglobin estimation using ErbaSysmex KX-21 auto-analyser. Descriptive statisticswere generated using SPSS version 12.0.

Results are presented for all mothers, in threestratified categories: those who were pregnantand had a SAM child aged 0-59 months (n=33);non-pregnant mothers with a SAM child agedless than six months (n=102); and non-pregnantmothers with a SAM child aged 6-59 months(n=292). Chi-square test was used to comparethe nutritional status of mothers between thethree groups.

Phase two entailed the development of a ma-ternal service package. A nodal pre-testing centrewas identified (Kalawati Saran Children’s Hospital,paediatric ward NRC), a technical expert groupformed and GoI/WHO nutrition guidelines re-viewed. e findings of phase one were presentedto technical experts and GoI officials and a pro-tocol and counselling aids were developed andrefined. e final protocol was implemented aspart of routine service provision in the NRC bya nutrition counsellor post-graduate nutritionresearcher on a sample of 100 mothers over aperiod of six months (December 2017 to May2018). During this time changes were continu-ously made to the algorithm and counsellingaids for further contextualisation. e maternalservice package was then presented to the GoI.

ResultsPhase 1: Profiling thenutritional status of mothersGeneral characteristics A total of 427 mothers with a child admitted toan NRC participated in the study: 203 (47.5%)from Kalawati Saran Children’s Hospital, 144(33.7%) Hindu Rao Hospital, 64 (15%) JawaharlalNehru Medical College and 16 (3.8%) BhagwanMahavir Hospital. Of the 427 mothers, 102(24%) had an infant aged 0 to less than 6 monthsof age; 292 (68%) had a child aged 6 to 59months; and 33 (8%) were pregnant (as well ashaving an admitted child of 0-59 months ofage). e mean age of mothers was 18.9 years(range: 16-32 years). e majority of motherswere Hindus (65%), unemployed (89%) andhad more than four family members (89%).Forty per cent of mothers had never attendedschool and 23% had studied until higher sec-ondary school. Adoption of family planningmethods was low (23%).

Anthropometry Of the 427 mothers assessed, 25% were low instature (stunted, height <145 cm) and 44% ofmothers were malnourished either by BMI orMUAC. By BMI, 23% were underweight (BMI<18.5kg/m2), which is comparable to the nationalaverage of 22.9% (NFHS 4 2015-2016), and 5%were severely underweight (BMI <16 kg/m2).By MUAC, 41% were malnourished (MUAC<23cm) and 3.5% were severely malnourished(MUAC <19cm) (no national data to compare).

Overweight or obesity was documented in27% (by Asian cut-off, BMI ≥ 23 kg/m2) and17% (by WHO cut off of ≥25kg/m2), which isslightly lower than the national female averageof 20.6% by WHO cut-off (NFHS 4 2015-2016)).Obesity was documented in 13% by Asian cut-off of BMI ≥ 26 kg/m2 and 4% by WHO cut-offof ≥30kg/m2. According to MUAC ≥30 cmobesity was observed in 6% of mothers.

AnaemiaHaemoglobin concentration was measured in390 consenting mothers, of whom 377 (97%)were anaemic and 71% were moderately or se-

Nutritional Status Mothers of children with SAM

Pregnant motherswith child 0-59mN=33

Mothers withchild 0-<6mN=102

Mothers withchild 6 – 59m N=292

Pooled

N=427

p-value

n (%) n (%) n (%) n (%)

Height <145 cm 7 (21.2) 20 (19.6) 78 (26.9) 105 (24.6) 0.287

Body Mass Index (kg/m2)

N=99 N=290 N=389

<18.5 NA 10 (10.1) 81 (27.9) 91 (23.4) 0.001

<16 NA 4 (4.0) 16 (5.5) 20 (5.1) 0.001

≥ 30 NA 4 (4.0) 12 (4.1) 16 (4.1) 0.004

≥23 NA 36 (36.4) 70 (24.1) 106 (27.2) 0.021

≥25 NA 21 (21.2) 45 (15.5) 66 (17.0) 0.207

Mid Upper Arm Circumference (cm):

N=33 N=102 N=291 N=426

≥23 18 (54.6) 72 (70.6) 161 (55.3) 251 (58.9) 0.003

<23 15 (45.4) 30 (29.4) 130 (44.7) 175 (41.1) 0.024

≥22-<23.0 6(18.2) 16 (15.7) 36 (12.4) 58(13.6) 0.347

≥19 -<22.0 9 (27.3) 11(10.8) 82(28.2) 102 (23.9) 0.002

<19 0 3(2.9) 12(4.1) 15(3.5) 0.448

≥26 7 (21.2) 29 (28.4) 73 (25.1) 109 (25.5) 0.325

≥30 0 6 (5.9) 18 (6.2) 24 (5.6) 0.342

Malnourished by either BMI(<18.5kg/m2) or MUAC (<23 cm)

15 (45.4) 32 (31.7) 141 (48.2) 188 (44.0) 0.012

Combinations

<145 cms & BMI < 18.5 kg/m2 0.0 4 (3.9) 23 (7.8) 27 (6.3) 0.110

<145 cm & MUAC <23 cm 3 (9.0) 9 (8.8) 37 (12.6) 49 (11.4) 0.522

Anaemia N=31 N=88 N=271 N=390

Any Anaemia 18 (58.1) 88 (100.0) 271(100.0) 377 (96.7) 0.034

Mild 6 (19.4) 23 (26.1) 71 (26.2) 100 (25.6)

Moderate 11 (35.5) 30 (34.1) 114 (42.1) 155(39.7)

Severe 1 (3.2) 35 (39.8) 86 (31.7) 122 (31.3)NA: Not applicable *Pregnant women: 10-10.9 g/dl (mild anaemia), 7.0-9.9 g/dl (moderate anaemia) and <7 g/dl (severe anaemia) and for non-pregnant womenas 11-11.9 g/dl (mild anaemia), 8.0-10.9 g/dl (moderate anaemia) and <8 g/dl (severe anaemia)

Table 1 Nutritional status of mothers of children with severe acute malnutrition (N=427)

e aim of this study was to bridge thisknowledge gap by assessing the nutritional statusof mothers of children with SAM admitted tofour NRCs in Delhi and Aligarh, Uttar Pradesh,and, based on the results, develop a maternalservice package of interventions for mothers tobe integrated in routine NRC services.

Methodse study was conducted in two phases: inphase one the nutritional status of mothers wasprofiled and in phase two a maternal servicepackage was developed. e first phase was con-ducted between September 2016 and November2017 in four NRCs. ree NRCs were locatedin Delhi (Kalawati Saran Children’s Hospital,Hindu Rao Hospital and Bhagwan MahavirHospital) and one in Aligarh (Jawaharlal NehruMedical College and Hospital of Aligarh MuslimUniversity). In each NRC, once the SAM childhad been stabilised, all mothers were invited toenrol in the study and informed consent wasobtained. No mothers declined to take part.

Socio-demographic profile was recorded ona pretested proforma. Maternal weight was meas-ured using UNICEF SECA weighing scales(model 874) with at least 100 grams gradation.Height was recorded using UNICEF SECA mi-

26

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Research


verely anaemic. Morbidities such as diarrhoea(15%), urinary tract infection (UTI) and genitalinfection (20%), chronic cough (12%), Bitot’sspot (4%) and fluorosis (1.3%) were commonamong mothers who were stunted, underweightor overweight/obese. Cases of diarrhoea werehigher in mothers who were underweight com-pared to mothers who were overweight (5% vs1%) (Table 1).

Phase 2: Development of amaternal service packagee GoI offers a service package to mothersthrough various antenatal and postnatal care

platforms, targeted at pregnant mothers andthose with infants aged 0 to 6 months. eseinclude: i) physical examination (height andweight); ii) laboratory investigation (haemoglobinestimation and oral glucose tolerance test); iii)iron and folic acid (IFA) and calcium supple-mentation; iv) deworming; and v) counselling(on healthy eating and diet diversity, familyplanning, micronutrient supplementation, per-sonal hygiene, and food and recipe demonstra-tion). Additionally, there are protocols in placefor screening and management of adult severethinness and severe anaemia which are used intuberculosis wards and haematology departments.

e protocol developed for this study exam-ined these guidelines in the context of routineservices provided in an NRC setting with a viewto expanding the appropriate support to allmothers with a SAM child admitted to an NRC.Building on what exists, a protocol was developedwhereby mothers are screened using the criteriaoutlined in Table 2 and classified as: i) not atnutritional risk; ii) at some nutritional risk; oriii) at severe nutritional risk/medical risk. Nu-trition interventions for each group were devel-oped as follows:

1. Universal interventions for all mothers not at nutritional riski. Hospital diet As per the Operational Guidelines on FacilityBased Management of Children with SAM (Mo-HFW, 2011) and Janani Shishu SurakshaKaryakaram (JSSK) diet guidelines (MoHFW2018), the mother or the caregiver staying withthe child is provided with food by the NRC,funded by the State Program ImplementationPlans (PIPs) approved by the MoHFW.

ii. Micronutrient supplementation and deworming As per GoI guidelines, IFA and calcium supple-ments are provided to pregnant mothers fromthe second trimester onwards to six monthspost-partum if not already provided elsewhere(one IFA tablet daily (60 mg iron and 500 mcgfolic acid) and two calcium tablets (500mg)).As per GoI guidelines, mothers in the secondtrimester of pregnancy are provided with a

Assess Classify Supplement/Treat/Refer

Ask: • Age • History of illness• Sudden weight lossMeasure:• Weight• Height• Check/plot BMI (or

MUAC if over 20 weekspregnant)

• Blood pressureExamine:Presence of anysymptoms includingfever, cough, blood insputum, increasedurinary frequency/burning during urination,recurrent or prolongedillness, clinical sign ofgoitre or fluorosis orvitamin A deficiency (nightblindness, Bitot’s spot),oedema (Grade +,++),for post-partum womenTest:• Haemogram• Fasting blood glucose

Not at nutritional riskAll of the following:• Age ≥ 20 years• No signs of malnutrition (height ≥150 cm and BMI 18.5-22.9 kg/m2 if not pregnant/

or less than 20 weeks of pregnancy; if over 20 weeks pregnant use MUAC 23-25.9cm) • No anaemia (≥12g/dl non-pregnant and ≥11g/dl for pregnant)• No medical risk

1. Hospital diet as per national guidelines2. Mother-focused group counselling on 5 thematic areas (30 minutes) 3. Micronutrient supplementation (IFA, calcium) and deworming as appropriate, as per national guidelines4. If the mother is pregnant or lactating give catch-up diet (F100) 350ml once/day

At some nutritional riskAny of the following:• Young (age 18-20 years) • Short (height 145-<150 cms)• Thin (BMI 16.0-<18.5 Kg/m2 if not pregnant or less than 20 weeks of pregnancy or MUAC 19-22.9 cm if over 20 weeks pregnant)• Overweight/ pre-obese (23-24.9 kg/m2) if not pregnant or less than 20 weeks pregnant or MUAC 26-30cm if over 20 weeks pregnant)• Mild/moderate anaemia (haemoglobin 7-10.9g/dl (pregnant)/ 8-11.9 g/dl (non- pregnant)) And no medical risk

1-4 (as above) PLUS5. Bedside individual mother-focused need-based counselling for 15 minutes daily, based on the condition identified (in addition to 15 minutes child- focused bedside counselling)6. Recipe book and demonstration if mother is underweight and counselling if overweight.

At severe nutritional risk/medical riskAny of the following:• Mothers breastfeeding infants 0-6 months• Very young (age <18 years) • Very short (height <145 cms)• Severely thin (BMI <16.0kg/m2 if not pregnant or less than 20 weeks pregnant or MUAC <19cm if over 20 weeks pregnant)• Obese (BMI >25kg/m2 if not pregnant or less than 20 weeks pregnant or MUAC >30cm if over 20 weeks pregnant)• Severe anaemia (hemoglobin <8g/dl if not pregnant or <7g/dl if pregnant)• Presence of symptoms/ signs suggesting medical illnesses, including: - Fever with chills ≥3 days - Cough >2 weeks/blood in sputum - Urinary complaints - Recurrent or prolonged illness - Signs of goitre or Vitamin A deficiency or fluorosis - Raised blood pressure or raised fasting blood sugar

1-6 (as above)PLUS7. Provide catch-up diet (F100) 350ml two times/day if severely thin or breastfeeding infant <6 months and once per day if pregnant (and not severely thin)8. Therapeutic treatment according to government guidelines of severely anaemic 9. Link for confirmation of medical illness to appropriate ward/unit in hospital and subsequent medical care as per national guidelines 10. Counselling for obesity and recipe book for obese mothers

Table 2 Maternal services package for Nutrition Rehabilitation Centres

Group counselling session, 2018

© U

NIC

EF/2

018/

Das

27

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


single oral dose deworming tablet (400 mg ofalbendazole) if not already provided elsewhere.

iii.Group-based nutrition education and counsellingChild-focused group counselling is provided inmost NRCs daily for one hour. In the proposedNRC protocol, daily group counselling is extendedby 30 minutes to focus on the mother as well.Group counselling takes place five days perweek, covering one of five thematic areas perday (micronutrients and anaemia; diet diversity;personal hygiene and sanitation; breastfeeding,family planning; and non-communicable andcommunicable diseases, including TB, HIV,malaria, obesity, hypertension and diabetes).During Phase 2, group counselling aids weredeveloped to help staff deliver the counselling(pictorial flip book, game, food plate model,recipe book for underweight and obese andpocket cards for testing their knowledge).1

Pretesting showed that the counselling aids areeasy to carry around the wards and simple touse and understand, with emphasis on importantpoints. ey were also found to increase the in-terest and involvement of the mothers.

2. Additional interventions formothers at some nutritional risk In addition to the universal interventions, 15minutes of individual, mother-focused, bedsidecounselling is provided for mothers deemed tobe at some nutritional risk in addition to theroutine 15 minutes of child-focused bedsidecounselling. Counselling is tailored to the mother’sspecific nutritional risk. In addition, those moth-ers identified as having mild or moderate anaemiaare provided with a therapeutic dose of oralIFA, as per GoI guidelines.

A recipe book was developed for underweightmothers, which was explained through livedemonstration sessions based on four localrecipes (see Table 3). Pretesting showed thatmothers took a keen interest in these sessions,which were successful in increasing their in-volvement and practical knowledge.

3. Additional interventions formothers at severe nutritionalrisk/medical risk In addition to universal interventions and thosetargeted to mothers at some nutritional risk,extra support is provided to mothers deemedto be at severe nutritional or medical risk (seeTable 2 for criteria). Several options for an en-hanced energy-dense supplement were developedbut, aer pre-testing with mothers, F-1002 wasfound to be the most feasible as it is alreadyprovided to the SAM child aer stabilisation(for ‘catch-up’ growth) and therefore no extracooking and ingredients were required. F-100(350ml) is provided once per day to pregnantmothers and twice per day (700ml) to mothersbreastfeeding infants under six months of ageand severely thin mothers to cover estimatedadditional energy requirements. erapeutictreatment is provided to women with severeanaemia as per government guidelines. Manyof these mothers also have other signs of un-derlying illness (such as TB) and are referred torespective wards for treatment.

Obese mothers were provided with counsellingon physical exercise and given a recipe booktailored to obesity. Very young mothers weregiven additional counselling on family planningand very short mothers were counselled on dietand family planning.

Limitations of the studye present opportunistic study on a reasonablesample size was restricted to anthropometricmeasurements and anaemia, which precludeduse of robust biomarkers of nutritional profileas well as co-morbidities not routinely gatheredin NRCs. Furthermore, mothers of childrenwithout SAM were not assessed to determine ifthe maternal anthropometric profile was simplya reflection of all mothers visiting these facilities.is study helped inform the development of amaternal package, using a common-sense ap-proach based on available evidence and feasibilityand practicality in the given context. It was notpossible to test the effectiveness of the packageon nutrition outcomes of mothers.

Discussion Prevalence of malnutrition was high amongmothers of children admitted with complicatedSAM; one third of the mothers were malnour-ished, similar to that found in a multi-centretrial in India (Bhandari et al, 2016). e presentstudy also found a double burden of malnutritionin around one quarter of households, where thechild was acutely undernourished and the motherwas overweight or obese. Eight per cent ofmothers were pregnant and almost half weremalnourished, highlighting the need for inter-vention involving counselling and dietary sup-plementation as per GoI norms and, lookingahead, to address family planning. is is animportant preventive action as a mother being

Recipe Recipe Serving Size Nutrient composition

Cerealpulsepremix

Whole wheat flour - 20gBengal gram roasted - 50gMilk Powder - 20gPeanut - 40gJaggery - 25gSesame seeds - 20g

170g Energy - 412 Kcal Protein - 17.1gCalcium - 244mgIron - 5.7mg

DaliyaChikki

Daliya - 10gPeanuts - 60gSesame seeds - 15gJaggery - 70gCooking oil - 3g

140g Energy - 500 Kcal Protein - 14.2 gCalcium - 216 mgIron - 5.6 mg

PaushtikLadoo

Bajra - 10gBesan - 40gGreen gram whole - 5gPeanut - 40gJaggery - 40gGhee - 24ml

7 ladoos (1ladoo = 20g)

Energy - 500 Kcal Protein - 14.2 gCalcium - 94 mgIron - 4.8 mg

Energy-densebiscuit

Whole wheat flour - 30gBengal gram roasted - 55gSoya granules - 15gMilk Powder - 20gJaggery - 20gCooking oil - 20gSesame seeds - 10gBaking powder - ¼ tspSalt - a pinch

14 biscuits (1biscuit - 15g)

Energy - 333 Kcal Protein - 13 gCalcium - 135 mgIron - 4.1 mg

Composition and nutrient content of the recipes developedfor food demonstration using local ingredientsTable 3

1 Counselling aids are available on request from the authors.2 The composition of F-100 is cow’s milk/toned dairy milk

(treated buffalo milk) 315 ml, sugar 26.2g and vegetable oil 7g, which has 350 Kcal, 10.1 g of protein and 14.7 g of lactose.

Mother receiving individual counsellingsession by the nutrition counsellor, 2018

© U

NIC

EF/2

018/

Das

Research

28

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


ReferencesBhandari N, Mohan SB, Bose A, Iyengar SD, Taneja S,Mazumder S, Pricilla RA, Iyengar K, Sachdev HPS, MohanVR, Suhalka V, Yoshida S, Martines J, Bahl R. Efficacy ofthree feeding regimens for home-based management ofchildren with uncomplicated severe acute malnutrition: arandomised trial in India. BMJ Global Health 2016; 1:e000144.

Bhutta ZA. Early nutrition and adult outcomes: pieces of thepuzzle. Lancet 2013; 13: 60716-3.

Khadivzadeh T. Mid upper arm and calf circumferences asindicators of nutritional status in women of reproductive age.Eastern Mediterranean Health Journal 2002; 8: 612-618.

Ministry of Health and Family Welfare (MoHFW).Guidelines for Antenatal care and skilled attendance at Birthby ANMs, LHVs/SNs. Government of India. 2010.http://nhm.gov.in/images/pdf/programmes/maternalhealth/guidelines/sba_guidelines_for_skilled_attendance_at_birth.pdf, accessed 28 March 2018.

Ministry of Health and Family Welfare (MoHFW). Guidancedocument: Nutritional care and support for patients withTuberculosis In India. Government of India, New Delhi,India. 2017. https://tbcindia.gov.in/WriteReadData/Guidance%20Document%20-%20Nutritional%20Care%20%26%20Support%20for%20TB%20patients%20in%20India.pdf, accessed 22 April 2018.

Ministry of Health and Family Welfare, Government of India.(2011). Operational Guidelines on Facility Based Managementof Children with Severe Acute Malnutrition. Accessible from:http://nhm.gov.in/images/pdf /programmes/child-health/guidelines/operational_guidelines_on_fbmc_with_sam.pdf, accessed 18 March 2018.

Nagabhushan BM, Poornima J, Premalatha R. Study ofmaternal profile in children admitted to nutritionalrehabilitation centre at a tertiary hospital. InternationalJournal of Contemporary paediatrics 2017; 4: 1911-1913.

National Family Health Survey (NFHS 4) 2015-16.International Institute of Population Sciences, Mumbai,

India. 2016. www.rchiips.org/nfhs/nfhs4.shtml, accessed14 March 2018.

Rai R, Singh DK. Maternal Profile of Children with SevereAcute Malnutrition. Indian Pediatrics 2015; 52: 344.

World Health Organization (WHO). Physical status: the useand interpretation of anthropometry. Report of a WHOExpert Committee. WHO Technical Report Series 854.Geneva: World Health Organization. 1995.

World Health Organization (WHO). Adolescent nutrition: areview of the situation in selected South-East Asiancountries. New Delhi: Regional Office of South-East Asia,World Health Organization, World Health Organization2003; 3–29.

World Health Organization (WHO). Hemoglobinconcentrations for the diagnosis of anaemia and assessmentof severity. Vitamin and Mineral Nutrition InformationSystem. Geneva, (WHO/NMH/NHD/MNM/11.1). 2011.Available from: www.who.int/vmnis/indicators/Hemoglobin. pdf, accessed 28 September 2017.

Sethi, V., Kumar, P. and De Wagt, A. (2019) Developmentof a maternal service package for mothers of childrenwith severe acute malnutrition admitted to nutritionrehabilitation centres in India. Field Exchange issue 59,January 2019. www.ennonline.net/fex

stunted, wasted or underweight preconceptionor during pregnancy predisposes the unbornchild to undernutrition (Bhutta, 2013). epregnant mother’s stay in the NRC with theiradmitted child (and non-pregnant mothers)presents an important opportunity to improvethe mother’s own nutritional status and that ofher future children.

e capacity required to roll out the maternalservices package is minimal, given that the pro-gramme is operated through existing NRCs,which already have kitchen facilities and adequatehuman resources for cooking and nutritioncounselling. e IFA, calcium supplementationand deworming can be procured from thehospital supply, already funded by the GoI. Ad-

ditional costs will be incurred in the provisionof MUAC tapes and in the extension of the timeinput by existing nutrition counsellors for groupand bedside counselling, which we estimate ispossible within current capacity.

ConclusionIn conclusion, facility-based centres for treatmentof SAM in children present an opportunity tocapture the attention of mothers and target sup-port to them for a duration of at least one totwo weeks, to address the maternal doubleburden of malnutrition, anaemia and familyplanning practices. Co-morbidities are commonamong mothers who may be stunted and/orunderweight or overweight. More research isneeded in similar settings to corroborate this

evidence and urgent pilot testing is needed ofthe proposed service package in NRCs to deter-mine operational feasibility and effectivenesson nutrition outcomes.

Read the maternal malnutrition guideline here:https://www.ennonline.net/resources/mater-nalnutritionalcarenrc

For more information, please email Vani Sethiat [email protected]

Research

29

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


The current state ofevidence and thinkingon wasting preventionSummary of research1

BackgroundENN produced a report, through the MQSUN+programme, that synthesises existing evidenceand stakeholder opinion on what works toprevent wasting. is report is one output of amulti-phase scope of work, commissioned andfunded by the UK Department for InternationalDevelopment (DFID) on “adopting a strategic,evidence-based approach to wasting prevention”and follows a briefing paper produced by ENNtitled “e aetiology of wasting”.2 e currentreport set out to answer the following questions:What do we know about wasting prevention?What is the emerging evidence? What are theevidence gaps and key questions which cannotcurrently be answered? What new evidence willbe available in coming years?

MethodologyA detailed review of the evidence from bothpublished and grey literature and from semi-structured interviews with stakeholders was con-ducted between December 2017 and February2018. e prevention of wasting was consideredacross the main intervention contexts (humani-tarian and development, those with low and highlevels of wasting/low and high levels of stunting),along a ‘continuum’ of severity (moderate andsevere wasting and prevention of relapse). ereview looked at the evidence for so-called nu-trition-specific and nutrition-sensitive actions.

In total 235 studies and articles from the lit-erature search were included in the review, ofwhich 55 were randomised control trials, 23were protocols/ongoing studies and 17 weresystematic reviews. Fiy-seven grey literaturedocuments were reviewed and 46 experts/keystakeholders were interviewed.

ResultsHistorically there has been more of a focus onresearch into the prevention of stunting, whilethe prevention of wasting has been a more neg-lected research area. is is largely because thefocus on wasting has been on treatment. En-couragingly, the review found that the volumeof studies relating to wasting prevention has in-creased in recent years. However, the evidencebase for the prevention of wasting is both mixedand largely inconclusive.

1 ENN (2018) The current state of evidence and thinking on wasting prevention: MQSUN+ report.www.ennonline.net/ resources/wastingpreventionreport2018

2 ENN (2018) The aetiology of wasting: MQSUN+ report. www.ennonline.net/resources/aetiologyofwasting

e interventions with the largest body ofevidence include the use of supplementary foodproducts, cash-based interventions, behaviourchange to improve infant and young childfeeding, and where interventions are combined.ere is a lack of research on whether interven-tions targeted towards women and girls pre-conception and during pregnancy prevent wasting(see box 1 below).

e stakeholders consulted felt confident instating that acute periods of food insecurityand/or episodes of disease outbreak contributeto wasting, and that well designed early inter-ventions will have a preventive effect in such

contexts. Despite an inconsistent evidence base,a holistic approach based on the UNICEF Con-ceptual Framework and context-specific causalanalysis was advocated, through a range of in-terventions to tackle both the immediate andunderlying drivers of undernutrition. It was feltthat this should be complemented by an improvedunderstanding of the epidemiology and aetiologyof wasting to better identify and target childrenat highest risk.

Stakeholder opinion also suggested that muchless is known about the prevention of wasting innon-humanitarian contexts or in areas with per-sistently high levels of wasting. Key gaps in theunderstanding of the aetiology of wasting werehighlighted, including: kwashiorkor/ nutritionaloedema; the relationship/overlap between stuntingand wasting; differences relating to age and ge-ography; the role of interventions preconception;the relationship between maternal nutrition andhealth status and child nutrition status; the extentto which wasting in infants under six months ofage reflects non-nutritional factors (such as lowbirth weight); and mechanisms behind relapseaer successful treatment for wasting. Othergaps included the role of infection and gut mi-crobiota and the longer-term health and devel-opment impacts of childhood wasting.

BreastfeedingThere is very little evidence of the impact ofinterventions to promote exclusive and continuedbreastfeeding on the prevention of wasting.Results are modest or difficult to attribute to anincrease in breastfeeding alone.

Complementary feedingLittle evidence of the impact of complementaryfeeding interventions on preventing wasting wasfound. Some systematic reviews highlight thebenefits, but studies including clear wastingoutcomes are lacking.

Micronutrient supplementationOnly small impacts on wasting prevention wereobserved when using zinc supplements, despitesome high-quality systematic reviews and clinicaltrials in this intervention area.

Nutrition counselling and nutrition educationA number of randomised clinical trials (RCTs) inthis intervention area showed positive effects onthe prevention of wasting, although many still failto demonstrate a clear impact.

DewormingTreating children known to have worm infectionmay have some nutritional benefits for theindividual. However, despite some good qualitystudies, direct effects on preventing wasting havenot been proved.

Maternal education, women’s empowermentand genderResults predominantly from association studies andprogramme evaluations suggest that women’sempowerment interventions and education couldhave a positive impact on infant feeding and wasting.

HealthFew clear studies consider the relationship

Box 1 Summary of state of evidence by intervention area

between health interventions and prevention ofwasting. Only a few well designed studies found alimited impact on wasting reduction.

Water, sanitation and hygiene (WASH)There is currently very little evidence as to thedirect effect of WASH interventions in theprevention of wasting, with the exception of somestudies on sanitation.

Agriculture and other livelihoodsOnly limited impacts have been observed,suggesting that more needs to be done to linkincreased agricultural production with improvingchild nutritional status. Robust evidence relating tothe impact of livelihoods interventions on wastingprevention is also lacking.

General food distribution (GFD)Although a number of studies have shown animpact of GFD on wasting, it is difficult to attributethis to the intervention alone.

Cash transfers (CTs)There is a growing body of well designed trials thatare demonstrating a strong positive preventiveeffect of CTs on wasting.

Food supplementationThe highest number of studies was identified in thisintervention area and evidence is growing quickly:well designed RCTs and systematic reviews havedemonstrated the effect of food supplementationin preventing wasting, but questions around theircost-effectiveness and sustainability remain.

Combinations of interventionsThe review found that a combination ofinterventions may be more effective at preventivewasting than separately implementedinterventions, particularly when targeted to thesame population.

Research

30

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Research. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Can low-literate communityhealth workers treat severeacute malnutrition? A study ofsimplified algorithm and toolsin South Sudan Summary of research1

1 Link to full report here (treatment protocol available on pages 78-80): www.rescue.org/sites/default/files/document/ 3103/enablingtreatmentofsamincommunityfinalcompressed.pdf. and Van Boetzelaer E, Zhou A, Tesfai C, Kozuki N. Performance of low‐literate community health workers treating severe acute malnutrition in South Sudan. Matern Child Nutr. Pending publication in early 2019. https://onlinelibrary.wiley.com/journal/17408709.

IntroductionLow access to and coverage of acute malnutritiontreatment are persistent challenges due primarilyto distance from health services, high opportunitycosts to caregivers, insecurity and lack of aware-ness of the signs and symptoms of malnutrition(Bliss, Njenga, Stoltzfus & Pelletier, 2016; Puett& Guerrero, 2015; Rogers, Myatt, Woodhead,Guerrero & Alvarez, 2015). Community-baseddelivery has been shown to increase the timelyand effective treatment of childhood illnessesin low-resource contexts, such as through theintegrated community case management (iCCM)of childhood illness strategy. ICCM equips com-munity health workers (CHW) with training,simplified diagnostics, supervision and an un-interrupted supply of drugs to provide timelytreatment for uncomplicated pneumonia, diar-rhoea and malaria in the community. Commu-nity-based delivery models have also been testedto treat uncomplicated severe acute malnutrition(SAM). Recent studies from Pakistan, Malawiand Mali have shown positive outcomes in SAMtreatment delivered by CHWs when comparedto standard care at health facilities (Linnemanet al, 2007; Puett, Coates, Alderman & Sadler,2013). However, existing evidence is for literateCHW cadres only.

Recognising the burden of malnutrition tobe higher in areas with lower education, income

By Naoko Kozuki, Casie Tesfai, Annie Zhou and Elburg van Boetzelaer.The authors would like to acknowledge financial support from the Eleanor Crook Foundation.

Many stakeholders observed that there is an absence of a single organisationwith an overall leadership role for wasting prevention. e divisions and siloswhich characterise wasting were also highlighted as a concern.

e review identified numerous ongoing studies in the area of wasting pre-vention and some which are pending funding. Further evidence is anticipatedin the coming years relating both to aetiology/epidemiology of wasting andeffectiveness of various nutrition-specific and nutrition-sensitive interventions.Stakeholders stressed the importance of longer-term funding to facilitateresearch which encompasses the full 1,000 days window and a need for morelongitudinal studies, including consideration of where existing data can be used(e.g. retrospective cohort studies). e need to improve the evidence base ondrivers and effective approaches to wasting prevention through well designedprogramme monitoring and evaluation activities was also highlighted andseveral stakeholders supported the idea of a research prioritisation exercise.

and healthcare access, the International RescueCommittee (IRC) developed tools and a protocoladapted for low-literate CHWs to treat uncom-plicated SAM cases in their communities. A de-tailed description of the design process is availablein Field Exchange Issue 52 (Tesfai, Marron, Kim& Makura, 2016). e five resulting tools were:1) a patient register, 2) modified mid-upperarm circumference (MUAC) tape, 3) weightscale decal to identify the daily dosage of ready-to-use therapeutic food (RUTF), 4) weekly RUTFdosage calculator, and 5) pictorial counsellingcards (Figure 1). Following this, the IRC con-ducted a feasibility study to assess the ability oflow-literate community-based distributors (CBD,the CHW cadre in South Sudan) to adhere tothe treatment protocol using these tools. etreatment protocol followed South Sudan nationalguidelines for treatment of uncomplicated SAM,but MUAC was the only anthropometric toolused for admission, monitoring and discharge,and children were treated until fully recovered.Children were provided RUTF based on weightaccording to South Sudan national guidelines,but with half sachets rounded up. Treatmentoutcomes of children included in this study willbe available in a future publication.

MethodsA mixed-methods feasibility study was conductedto assess adherence of low-literate CBDs to a

simplified SAM treatment protocol, the outcomesof children treated in the community by theseCBDs, and the community acceptability of CBDsproviding SAM treatment. Sixty CBDs in AweilSouth County, Northern Bahr el Ghazal, SouthSudan were randomly selected to receive trainingon the simplified protocol and tools, 57 of whomcompleted the training. ose who met a pre-determined performance standard on a post-training assessment (n=44) were deployed totreat uncomplicated SAM in their communities.e checklist used to assess the performance isavailable in Annex 3 of the full report. BetweenMay and September 2017, 320 SAM childrenwere passively identified and enrolled, asked toreturn for weekly treatment, and followed untilthey reached a discharge outcome, with childrentreated to full recovery (two consecutive weekswith MUAC ≥12.5cm); 308 children had eligibleresults. CBD performance assessments wereconducted during bi-weekly supervisory visitsby research staff.

Community-based distributormeasuring child’s MUAC

Elbu

rg v

an B

oetz

elae

r / T

he IR

C

ENN’s continued work in this areaENN is continuing its focus on wasting prevention throughthe MQSUN+ mechanism by working with a team of expertsto carry out a research prioritisation exercise on wastingprevention. is work will conclude in July 2019. Informationon how you can participate in the exercise is available here:www.ennonline.net//chnriwastingprevention

ENN also continues to generate research publications andshort briefs about the links between wasting and stunting(WaSt). Information about the WaSt project can be foundhere:www.ennonline.net/ourwork/reviews/wastingstunting

31

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Research

ResultsIn a performance assessment immediately aer training, 91%of the CBDs passed the predetermined 80% performance scorecut-off and 49% of the CBDs had perfect scores. Researchofficers conducted 141 case management observations duringthe study period, resulting in a mean score of 89.9% (95% CI:86.4-96.0%). For each performance assessment completed, thefinal performance score of the CBD rose by 2.0% (95% CI: 0.3-3.7%). Treatment delivered by CBDs met SPHERE performanceindicators, even when looking at treatment outcomes to fullrecovery. In total 75% of children were discharged as fully re-covered, 15% defaulted from treatment, 9% were discharged asnon-respondent, and no child was reported to have died undertreatment. e median time of treatment to full recovery waseight weeks. All CBDs reported feeling proud of treatingchildren for SAM and some gained respect in the communityfor this new responsibility. Overall, caregivers trusted CBDs totreat their children, but some caregivers and community leadersalso expressed reservations that CBDs were keeping the RUTFfor themselves or providing treatment to select children only.

Discussione high adherence by CBDs to a simplified treatment protocolin this study and overall local acceptability of this service showpromise for deploying CBDs to improve access of acute malnutritiontreatment, regardless of their literacy levels, in remote communities.e upfront investment to design tools and protocol suited tothe skill set of CHWs in difficult contexts is invaluable in settingfrontline health workers up for success and assuring programmeeffectiveness. In the hard-to-reach areas of fragile contexts withlimited healthcare access, there is particular potential for the in-tegration of nutrition treatment into the community-basedservice delivery model of iCCM to better stem the infection-malnutrition cycle and more effectively reduce the incidence ofboth.

A challenge experienced in developing a low-literacy protocolwas how to monitor whether cases are stationary, regressing orprogressing slowly in treatment. To address this, smaller MUACcolour zones were created and a safeguard for referral aerfour consecutive weeks in one colour zone was put in place.Based on the larger-than-normal proportion of referrals fromthis study (37%), further exploration is needed to adjust thissafeguard.

ConclusionProper adaptations of tools and protocols can empower communityhealth worker cadres with no formal education to provide criticallifesaving health services successfully. ese results, combinedwith high recovery rates of the enrolled children, show greatpotential to increase effective coverage of acute malnutritiontreatment in fragile contexts. e IRC is currently leading a con-sortium of four other non-governmental organisations (ActionAgainst Hunger, Concern Worldwide, Malaria Consortium andSave the Children) to pilot versions of these tools adapted forother contexts to create a greater body of evidence behind CHW-delivery of acute malnutrition treatment.

For more information, please contact Casie Tesfai [email protected]

ReferencesBliss JR, Njenga M, Stoltzfus RJ & Pelletier DL. (2016).Stigma as a barrier to treatment for child acute malnutritionin Marsabit County, Kenya. Matern Child Nutr, 12(1), 125-138. doi:10.1111/mcn.12198

Linneman Z, Matilsky D, Ndekha M, Manary MJ, Maleta K& Manary MJ. (2007). A large-scale operational study ofhome-based therapy with ready-to-use therapeutic food inchildhood malnutrition in Malawi. Matern Child Nutr, 3(3),206-215. doi:10.1111/j.1740-8709.2007.00095.x

Puett C, Coates J, Alderman H and Sadler K. (2013). Qualityof care for severe acute malnutrition delivered by communityhealth workers in southern Bangladesh. Matern Child Nutr,9(1), 130-142. doi:10.1111/j.1740-8709.2012.00409.x

Puett C & Guerrero S. (2015). Barriers to access for severeacute malnutrition treatment servies in Pakistan andEthiopia: a comparative qualitative analysis. Public HealthNutr, 18(10), 1873-1882. doi:10.1017/S1368980014002444

Rogers E, Myatt M, Woodhead S, Guerrero S & Alvarez JL.

(2015). Coverage of community-based management ofsevere acute malnutrition programmes in twenty-onecountries, 2012-2013. PLoS One, 10(6), e0128666.doi:10.1371/journal.pone.0128666

Tesfai C, Marron B, Kim A & Makura I. (2016). Enabling low-literacy community health workers to treat uncomplicatedSAM as part of community case management: innovationand field tests. Retrieved from www.ennonline.net/fex/52/communityhealthworkerssam

Simplified tools developed by IRC Figure 1

32

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Research

Is there a systematic bias in estimates ofprogramme coverage returned by SQUEACcoverage assessments? By Mark Myatt and Ernest Gueverra

Mark Myatt is a consultant epidemiologist. His areasof expertise include infectious disease, nutrition andsurvey design. He worked with FANTA, ValidInternational, the US Centers for Disease Control andPrevention, Tufts University, Action Against HungerUK, Concern Worldwide and the Coverage

Monitoring Network as the lead developer of the SQUEAC coverageassessment method.

Location: GlobalWhat we know: Used since 2012, the Semi-Quantitative Evaluation of Access andCoverage (SQUEAC) coverage assessment method employs both qualitative andquantitative methods to identify key barriers to access and estimate coverage oftherapeutic feeding programmes (TFP) and, to a lesser extent, supplementary feedingprogrammes (SFP).

What this article adds: A recent article by Isanaka et al (2018) on SQUEACimplementation in Niger suggests that the analysis required is technically demanding andin part relies on subjective estimates of programme coverage. With typical operationalcapacities this will cause SQUEAC assessments to systematically overestimate coverage.is article investigates the risk of systematic bias by analysing a database of 304SQUEAC coverage assessment reports and data from 29 countries (2009-2017). ere is atendency for the ‘prior’ (an informed guess about programme coverage) to overestimatecoverage when the true coverage is low and underestimate coverage when the truecoverage is high. ere is an equal risk of the prior overestimating and underestimatingcoverage (i.e. no systematic bias). Problems were detected in 7.3% of the SQUEACassessments reviewed but this led to coverage estimates with poor precision in only 2.55%of assessments. e use of untrained staff and failure to use SQUEAC processes, methodsand tools correctly is likely to increase this risk. e authors conclude there is no evidenceof general and systematic overestimation of coverage using SQUEAC and that the risk ofthe method yielding estimates with poor precision is low. A key lesson from the Isanaka etal (2018) SQUEAC experience is the importance of using both properly trained staff andusing SQUEAC processes, methods and tools correctly.

BackgroundA recent article by Isanaka et al (2018) (sum-marised in Box 1) identifies a potentially seriousproblem with coverage estimates made usingthe semi-quantitative evaluation of access andcoverage (SQUEAC) method. Coverage estimatesmade by SQUEAC rely on condensing data col-lected from a variety of sources using a range ofmethods to make an informed guess about thelevel of coverage a programme is achieving.is informed guess is known as the prior. eprior is used to inform the design of a small-sample coverage survey. e prior is also com-bined with coverage survey data, known as thelikelihood, to provide an estimate of the coveragethat a programme is achieving using a widelyaccepted statistical technique known as conjugate

analysis. A problem with this approach is that avery poorly specified prior can result in a biasedestimate of coverage. A prior that is much higherthan the true coverage can lead to an upwardlybiased estimate of coverage. A prior that ismuch lower than the true coverage can lead to adownwardly biased estimate of coverage. esesituations are known as prior-likelihood conflicts.If a prior-likelihood conflict is detected, theresults of the conjugate analysis are discardedand a coverage estimate is made using the surveydata alone. is estimate will not be biased butmay lack precision (i.e. have a wide 95% credibleinterval), due to the small sample size used inthe coverage survey. e article by Isanaka et al(2018) identifies prior-likelihood conflicts as aweakness of the SQUEAC coverage assessment

method which leads the method to systematicallyoverestimate programme coverage.

e issue of prior-likelihood conflicts is notnew. It is covered at some length in the SQUEACtechnical reference (Myatt et al, 2012). A formaltest for prior-likelihood conflicts has been pro-vided by the BayesSQUEAC calculator since2013 (Myatt, 2013). e specific case of untrainedstaff producing an unrealistically optimistic andoverly strong prior, as reported in the article byIsanaka et al (2018), is presented as a case studyin the SQUEAC technical reference. e articleby Isanaka et al (2018) confirms the existenceof a problem that is frankly admitted, discussedand cautioned against in SQUEAC documenta-tion and training. is should not, however,prevent us from taking this criticism of theSQUEAC method very seriously. It is possiblethat there is a serious problem with the SQUEACmethod which is leading to a general and sys-tematic failure to identify programmes failingto meet coverage standards and leaving manyvulnerable children untreated. is issue is in-vestigated in this article.

MethodA database was created from SQUEAC coverageassessment reports and SQUEAC coverage as-sessment data provided by the Coverage Mon-itoring Network and VALID International. Re-ports and data for n = 304 SQUEAC coverageassessments from 29 countries undertaken be-tween 2009 and 2017 were available. Only datafrom SQUEAC coverage assessments whichcompleted a SQUEAC stage III coverage esti-mation survey (n = 274) are included in theanalysis reported here.

For each SQUEAC coverage assessment, themode of the prior was calculated as:

Ernest Guevarra leads Valid International’scoverage assessment and surveys team. He hasformal training as a physician and a publichealth practitioner. Most recently he hasworked in Sierra Leone, Niger, Sudan, Ethiopiaand Ghana. He is the lead developer of

SQUEAC, SLEAC and S3M coverage assessment methods at ValidInternational.

prior mode = αPrior-1

αPrior+βprior-2

was classified as ‘failing’ if a prior-likelihoodconflict was detected and the relative precisionof the coverage estimate made using the likelihooddata alone was worse than this gold standard.ResultsTable 1 presents a description of the study data-base. Figure 1 shows the distribution of thedifferences between the prior modes and likeli-hood modes. e median difference was −0.97%(IQR = −8.31%; +8.39%). e differences werenormally distributed (Shapiro-Wilk NormalityTest p = 0.6287) about a central value close tozero (mean = −0.67%, 95% CI = −2.25%; +0.90%).

Figure 2 shows the scatterplot of prior modesagainst likelihood modes. Prior modes and like-lihood modes were positively associated witheach other. Pearson’s correlation coefficient wasr = 0.73 (95% CI = 0.67; 0.78). is is verystrong evidence (p < 0.0001) against the nullhypothesis that prior modes and likelihoodmodes are not associated with each other. eslope of the regression line was β = 0.55 (95%CI = 0.49; 0.61).

Prior-likelihood conflicts were detected in20 (7.30%) of the 274 SQUEAC assessments. In10 SQUEAC assessments with prior-likelihoodconflicts the prior mode was below the likelihoodmode. In 10 SQUEAC assessments with prior-likelihood conflicts the prior mode was abovethe likelihood mode (see Figure 2). e relativeprecision of coverage estimates based on likeli-

The numerator (NLikelihood) and denominator(DLikelihood) for the likelihood mode were cal-culated for the principal coverage estimator(i.e. point, period, or single coverage) reportedin the SQUEAC coverage assessment report:

is was the estimator for which the originalprior was developed. Testing a prior intendedfor estimating point coverage for conflict with alikelihood estimate for single coverage would,for example, not be sensible or fair test andalmost always result in a prior-likelihood conflict.e analysis followed, therefore, the intentionsof the coverage assessment teams.

e relationship between the prior modesand the likelihood modes was explored by calcu-lating, plotting and summarising the difference:

prior mode - likelihood mode

and by plotting the prior modes against thelikelihood modes.

e strength of the linear association betweenthe prior modes and the likelihood modes wasassessed using the Pearson correlation coefficient.Ordinary least squares linear regression wasused to determine the slope of the line that bestdescribed the relationship between the priormodes and the likelihood modes.

For each SQUEAC coverage assessment, pri-or-likelihood conflicts were detected using atesting approach. Two-by-two tables were con-structed with cells:

and Fisher’s exact test of independence calculatedfor each of the constructed tables. Fisher’s exacttest was used to avoid issues with small samplesizes and very unequal distribution of datawithin tables giving rise to small expected num-bers, which would be problematic if approximatemethods such as the normal (z) test and thechi-squared test were used. A two-sided p-valueof p < 0.05 was taken as evidence of a prior-likelihood conflict.

e half-width of 95% confidence intervals(E) for the likelihood modes (p) were calculatedusing the normal approximation and applyinga finite population correction (i.e. because se-vere acute malnutrition is a rare condition) foreach SQUEAC coverage assessment for whicha prior-likelihood conflict was detected:

e population size used to calculate the fi-nite population correction (i.e. 600) was calcu-lated assuming an overall population of 100,000

33


. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Research

persons with 20% aged between 6 and 59months and a 3% prevalence of severe acutemalnutrition (SAM). ese are conservative as-sumptions. It was not necessary to use theprevalence of moderate acute malnutrition(MAM) to calculate the finite population cor-rection as no prior-likelihood conflicts werefound in assessments of supplementary feedingprogrammes (SFP).

e relative precision achieved using the likelihooddata alone was calculated as:

A coverage estimate of 50% (p) with a 95% con-fidence interval of ± 10% (E) has relative preci-sion:

e resulting relative precision was comparedto the relative precision that would have beenachieved by a standard Expanded Program ofImmunisation (EPI) ‘30 x 7’ coverage surveywith the same point estimate of the coverageproportion (p), a sample size of n = 210 and asurvey design effect of 2.0. is relative precisionwas used as the ‘gold standard’ for the precisionof methods assessing the coverage of childsurvival programmes. A SQUEAC assessment

‖αprior-1‖

NLikelihood

‖βprior-1‖

DLikelihood-NLikelihood

relative precsion = Ep

E=1.96 × ( )× ( )p (1-p)n

Population Size-DLikelihoodPopulation Size-1

relative precsion = = = 20%Ep

10%50%

The article by Isanaka et al. (2018) published inPopulation Health Metrics recognises thatassessing the coverage of nutrition programs ischallenging due to the low prevalence of diseaseand selective entry criteria. It also recognises thatSQUEAC is a "step forward in coverageassessment of therapeutic feeding programs"and can "simultaneously identify barriers toaccessing care and estimate program coverage".It notes, however, that "the validity of certainmethodological elements has been the subjectof debate". The methodological elements inquestion revolve about the use of a Bayesianconjugate analysis to improve the precision ofcoverage estimates made using small samplesizes. The concern is not that Bayesianapproaches are generally invalid but that theapproach is beyond the technical capacity ofstaff employed by NGOs, UNOs, and ministries ofhealth and that its use in the wrong hands willlead to (worst case) systematic overestimation ofcoverage and (best case) coverage estimateswith very poor precision.

Isanaka et al. (2018) investigate this issue bycomparing the results of a SQUEAC coverageassessment performed by untrained personsagainst a two-stage cluster sample survey with aspatially stratified first stage selectingcommunities and active and adaptive case-finding in the second stage. The comparisonmethod employed is very similar to the methodused by SQUEAC stage III likelihood surveys. The

Box 1 Summary of the article by Isanaka et al. (2018

only difference being that a larger sample size isused. This means that any substantial differencesfound between the SQUEAC results and thesurvey results will be due to the untrained staffdoing a poor job of specifying the prior used inthe Bayesian conjugate analysis.

The article reports that priors produced byuntrained staff and by untrained communitymembers led to upwardly biased coverageestimates. Point estimates of coverage madefrom the likelihood survey data alone weresimilar to those made from the larger two-stagecluster survey. This means that the problem iswith the Bayesian prior produced by untrainedstaff and untrained community members beingtoo optimistic and too strong (i.e. overly certain).The reported biases were, however, consistentlydetected using standard SQUEAC diagnosticmethods and SQUEAC software (i.e. plots andtests in BayesSQUEAC) for detecting prior-likelihood conflicts. Coverage estimates madeusing a prior produced by trained staff was inagreement with that made by the two-stagecluster sample.

The authors conclude that SQUEAC is technicallydemanding and should only be used when theappropriate technical capacity is available. Theyalso question the validity of the methods usedby SQUEAC to produce priors when they areused in capacity limited settings.

√ √

likelihood mode = NLikelihood

DLikelihood

34

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Research

hood data alone was worse than that whichwould have been achieved by the EPI-derivedgold-standard in seven of the 20 SQUEAC as-sessments with prior-likelihood conflicts. ismeans that seven (2.55%) of the 274 SQUEACassessments were classified as failing due to pri-or-likelihood conflicts and an inadequate samplesize for the likelihood data to estimate coveragewith useful precision.

DiscussionThe magnitude of Pearson’s correlation coeffi-cient (r = 0.73. 95% CI = 0.67; 0.78) might betypified as indicating only a moderately strongassociation. It is, however, unrealistic to expect astronger association between informed guesses

(i.e. prior modes) and estimates (i.e. likelihoodmodes) made using small sample likelihood data.e nature of the prior mode (i.e. an informedguess) and the likelihood mode (i.e. an estimatemade using a small sample size subject to con-siderable sampling variation) makes finding astronger association between them extremelyunlikely.

e distribution of differences between theprior mode was symmetrical about a centralvalue. is is not consistent with a systematicbias (in either direction) in prior modes. ereis a tendency for the prior to overestimate cov-erage when true coverage is low and to under-estimate coverage when true coverage is high.

Prior-likelihood conflicts followed this pattern.Prior-likelihood conflicts were detected in7.30% of the 274 SQUEAC assessments but ledto coverage estimates with poor precision inonly 2.55% of the 274 SQUEAC assessments.Prior-likelihood conflicts in which the priormode was below the likelihood mode wereequally as common as prior-likelihood conflictsin which the prior mode was above the likeli-hood mode. ese findings indicate that thereis no general and systematic failure in SQUEAC.

ere is an important lesson to be learnedfrom the Isanaka et al (2018) article. eSQUEAC assessment reported in the article wasnot done well. is is noted in the discussionsection of the article. ere is no evidence of theuse of standard SQUEAC tools and practicessuch as triangulation by source and method,sampling to redundancy, iteration, the barriers-boosters-questions (BBQ) tool, small studiesand surveys, mind-maps, and concept maps.e resolution of conflicting findings by furtherdata collection (iteration) is a key SQUEACprocess that was not used. e article states thatiteration was not done even when it was indi-cated. Finding a wide range of candidates for theprior mode, as is reported in the article, shouldhave forced a rethink and further data collection(iteration). e sources for the problematicprior modes were unorthodox. SQUEAC doesuse caregivers and community members toidentify and rank barriers to coverage, but theseinformants are never tasked with responsibilityfor building the prior. A weak or non-informa-tive prior should always be used with such awide range of candidates for the prior modewhen time and resources for iteration are notavailable.

A key but understated finding reported byIsanaka et al (2018) was that the prior developedby trained staff was unproblematic. e lesson

Item Description n

Number of records Total number of records in the study database 304

Excluded No stage III : Not required: 17

No stage III : Suspected patchy coverage 5

No stage III : Poor security / access 6

No stage III : Very low SAM prevalence 2

Included Number of records included in the analysis 274

Coverage type* Point coverage 199

Period coverage 70

Single coverage 5

Assessed programme** OTP 255

SFP 19

Table 1 The study database

Figure1 Distribution of the differences between the priormodes and likelihood modes in 274 SQUEAC stageIII coverage assessments

* Point coverage measures case-finding and recruitment; Period coverage measures case-finding, recruitment and retentionbut overestimates coverage; Single coverage adjusts period coverage removing bias by including an estimate of the numberof recovering cases in the community in the denominator.

** OTP = Outpatient Therapeutic Program treating cases of severe acute malnutrition; SFP = Supplementary Feeding Programtreating cases of moderate acute malnutrition.

Prior and likelihood modes are expressed as percentages.

The dashed vertical lines mark the position of the lower quartile, median, and upper quartileof the differences. Half of all SQUEAC coverage assessments fall between these two lines.

Prior mode - Likelihood mode

Figure 2 Scatterplot of prior modes against likelihood modesin 274 SQUEAC stage III coverage assessments

The solid line is the ordinary least squares regression line.

The dashed line is the line of equality (i.e. The line on which prior mode = likelihood mode).

Filled circles mark assessments with prior-likelihood conflicts.

35


Research. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ReferencesIsanaka S, Hedt-Gauthier BL, Grais RF, Allen BG. Estimating program coverage in thetreatment of severe acute malnutrition: a comparative analysis of the validity andoperational feasibility of two methods, Population Health Metrics, 2018,16:100,1-9 Fulltext available from:https://pophealthmetrics.biomedcentral.com/articles/10.1186/s12963-018-0167-3

Myatt M, Guevarra E, Fieschi L, Norris A, Guerrero S, Schofield L, Jones D, Emru E, Sadler K(2012) Semi-Quantitative Evaluation of Access and Coverage (SQUEAC) / Simplified LotQuality Assurance Evaluation of Access and Coverage (SLEAC) Technical Reference, Food andNutritional technical Assistance III Project (FANTA-III), FHI 360 / FANTA, Washington, DC,October 2012 www.fantaproject.org/ monitoring-and-evaluation/squeac-sleac

Myatt M (2013) BayesSQUEAC v3.00: A graphical calculator for Bayesian beta-binomialconjugate analysis of coverage proportions in CTC and CMAM programs using a bootstrapaggregating (bagging) estimator implementing sample size calculation and formal testingfor prior-likelihood conflicts, Brixton Health / Valid International, Oxford, UK, 2013.

Postscript By Sheila Isanaka, Rebecca F. Grais, and Ben G.S. Allen

We thank Mark Myatt and ErnestGueverra for adding this importantwork to the ongoing discussion sur-

rounding appropriate coverage methodolo-gies in the management of acutemalnutrition. A 2015 review of coveragemethodologies highlighted uncertainty in theuse of currently recommended methods forcoverage assessment and the need for morepeer-reviewed evidence to inform globalguidance (Epicentre, 2015). We are delightedthat our work (Isanaka et al 2018) may havemotivated additional consideration of theseimportant issues and hope that stakeholdersand policy makers continue to insist on high-quality, evidence-based experience to informnutrition programming.

The report by Myatt and Gueverra (2019)shows that coverage estimation using theSQUEAC methodology can yield biased esti-mations in either direction. For the first time,Myatt and Gueverra quantify the magnitudeof this bias, showing only a moderate corre-lation between the prior and likelihood esti-mates (Pearson correlation coefficient 95%confidence interval: 0.67 to 0.78). This new ev-idence is consistent with our findings andsupports our conclusion that conflicts be-tween prior and likelihood modes are possi-ble and can lead to biased and imprecisecoverage estimates. As discussed by Myattand Gueverra, the risk of such bias is lowwhen trained staff conduct a survey using ap-propriate SQUEAC methods. Our work sup-

ports this finding, as we similarly show noconflict when an external support team esti-mated the prior. Our experience, however,goes further than what is possible in the sec-ondary analysis of Myatt and Gueverra tosuggest that in resource-limited settings,where sufficient capacity and resources maynot be available and the correct methodol-ogy may not be faithfully executed, conflictand bias may be more common.

We wholeheartedly agree that coverage as-sessments should be done using appropriatemethods as outlined in the SQUEAC manual(Myatt et al. 2012). Our SQUEAC assessmentfollowed this guidance, including triangula-tion by source and method, use of the BBQtool, sampling to redundancy and a small sur-vey. We used booster and barrier weightingby caregivers of severely acutely malnour-ished (SAM) children as reported in othercontexts (Blanárová et al. 2016) as one com-ponent in developing the prior mode. Thevarious prior modes in our analyses werecombined to simulate different potential sce-narios, including situations where externalsupport is not available. This was done for thepedagogic purpose of the study, and as dis-cussed in the paper does not necessarily re-flect typical SQUEAC procedures.

We note that the database employed in theanalysis of Myatt and Gueverra is likely com-prised of surveys conducted by experiencedcoverage consultants (those provided, for ex-

ample, by the Coverage Monitoring Networkor Valid International) and therefore includeprior modes developed by dedicated consult-ants using gold standard methods that maybe less likely to conflict. The analyses furtherinclude data from supplementary feedingprogrammes and does not standardise calcu-lations of coverage estimate according to cur-rent guidance to use single coverage(Balegamire S, 2015), analytical choices whichmay influence the extent to which conflictsand bias were detected. Nonetheless, we wel-come the new evidence presented by Myattand Gueverra (2019) as an important step to-wards better understanding of the implica-tions of using the SQUEAC methodology forvalid coverage estimation.

Overall, we look forward to continued evi-dence-based and peer-reviewed discussionof appropriate coverage methodologies. Sev-eral methodologies are available to monitorprogramme coverage, and the appropriatestudy design should be selected in consider-ation of team capacity, resources and report-ing requirements. SQUEAC can be atechnically demanding method and requiresthe appropriate capacity to avoid the poten-tial for bias. As both we and Myatt and Guev-erra have shown, conflict and biasedcoverage estimation are possible and shouldbe considered in selecting the appropriatestudy design and allocating appropriate re-sources for assessment.

ReferencesBalegamire S, Siling K, Alvarez Moran JS, Guevarra E,Woodhead S, Norris A, Fieschi L, Paul Binns P, and MyattM, A single coverage estimator for use in SQUEAC, SLEAC,and other CMAM coverage assessments, 2015, FieldExchange 49 Full text available from:https://www.ennonline.net/fex/49/singlecoverage

Blanárová L, Woodhead S, and Myatt M, Communityweighting of barriers and boosters in Democratic Republicof Congo, 2016, Field Exchange 52 Full text available from:https://www.ennonline.net/fex/52/barriersandboostersdrc

Epicentre. Open review of coverage methodologies:comments, questions and ways forward. London:Coverage Monitoring Network; 2015.

Isanaka S, Hedt-Gauthier BL, Grais RF, Allen BG,Estimating program coverage in the treatment of severeacute malnutrition: a comparative analysis of the validityand operational feasibility of two methods, PopulationHealth Metrics, 2018,16:100,1-9 Full text available from:https://pophealthmetrics.biomedcentral.com/articles/10.1186/s12963-018-0167-3

Myatt M, Guevarra E, Fieschi L, Norris A, Guerrero S,Schofield L, Jones D, Emru E, Sadler K (2012) Semi-Quantitative Evaluation of Access and Coverage (SQUEAC)/ Simplified Lot Quality Assurance Evaluation of Access andCoverage (SLEAC) Technical Reference, Food andNutritional technical Assistance III Project (FANTA-III),FHI 360 / FANTA, Washington, DC, October 2012https://www.fantaproject.org/monitoring-and-evaluation/squeac-sleac

Myatt, M., and Gueverra, E. (2019) Is there a systematic bias in estimates of programmecoverage returned by SQUEAC coverage assessments? Field Exchange issue 59,January 2019. www.ennonline.net/fex

to be learned is that you risk bias when you do SQUEAC with untrainedstaff, use inappropriate sources and do not use key SQUEAC processes,methods and tools.

ConclusionPrior-likelihood conflicts can and do occur but seldom result in coverageestimates that lack useful precision. ey do not lead to a general and sys-tematic overestimation of coverage. e work of Isanaka et al (2018)demonstrates the importance of using trained staff and using SQUEACprocesses, methods and tools correctly.

36

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Research

Infant and young child feeding inemergencies: An analysis of keyfactors of a strong responseSummary of research1

Despite the evidence that appropriateand timely support of infant andyoung child feeding in emergencies(IYCF-E) saves lives, it is rarely pri-

oritised or adequately supported. is study bythe Save the Children Technical Rapid ResponseTeam (Tech RRT) explores factors that supportedor inhibited a strong IYCF-E response in threeemergencies: the El Niño drought in Ethiopia2015-2016; the earthquake in Nepal in 2015;and the ongoing Syrian crisis.

From 2015 to 2016 Ethiopia experienced itsworst drought in 50 years, leaving 10.2 millionpeople in need of emergency assistance by theend of 2015 and 420,000 children under fiveyears old requiring treatment for severe acutemalnutrition (SAM) in 2016. During the crisisa total of 1.1 million children aged 0-2 years oldand 600,000 pregnant and lactating women(PLW) were targeted with IYCF-E interventionsin 142 priority districts. e 2015 earthquakein Nepal led to a large humanitarian effort thatreached 3.7 million people in the 14 severely af-fected districts, including 404,000 children agedsix months to five years old and PLW in need ofnutrition support. In Syria, by 2017, the ongoingconflict had le 13,6 million people requiringhumanitarian assistance, including 4.9 millionpeople trapped in besieged and hard-to-reachareas and an estimated 2.5 million children agedsix months to five years and 1.9 million PLW;most infants among them were being fed infantformula, which presented particular challenges.e aim of this study was to identify commonfactors across these three different contexts thatsupported, and hindered, good IYCF-E response.

A panel of Save the Children Nutrition inEmergency Technical Advisors determined in-dicators defining strong IYCF-E programmingand potential key underlying factors, drawingfrom existing indicators, such as Sphere Standardsand Global Nutrition Cluster indicators. Sec-ondary data was collected and 13 interviewsconducted with key stakeholders from interna-tional non-governmental organisations (NGOs)(n=7), local NGOs (n=2), United Nations (UN)agencies (n=1) and the Nutrition Cluster (n=3)in October-November 2016 for Ethiopia, May-June 2017 for Syria and August-September 2017

for Nepal. Data was analysed and a score wasattributed to each country based on the strengthof the IYCF-E response out of a maximum pos-sible score of 16 (<8 weak; 8-11 medium and>12 strong). Ethiopia scored 9 (medium); Nepal11 (medium) and Northern Syria 13 (strong)2.

e first factor identified as supportive of agood IYCF-E response across the three casestudies was the activation of the NutritionCluster. In Northern Syria it was only aer theNutrition Cluster was activated that the responsegrew stronger; before this, nutrition was aworking group of the Health Cluster and theIYCF-E response was weak. e second factoridentified was the presence of a strong componenton IYCF-E as part of the national IYCF strategypre-crisis, which contributed to the strengthand timeliness of the response in Nepal. ird,commitment of government and national andinternational NGOs to IYCF-E were importantdrivers, particularly in Nepal, where strong gov-ernment commitment led to implementationof the International Code of Marketing of Breast-milk Substitutes and other important IYCF-Epolicies and strategies. In Northern Syria com-mitment of international NGOs enabled the ca-pacity-building of national NGOs and engenderedgreater commitment to IYCF-E. Assessment ofIYCF practices and needs was also important.is aspect was generally weak in the three casestudies, although recent improvement was seenin Syria, which may have helped to strengthenthe response. Another critical factor was thepresence of pre-crisis IYCF programmes andavailability of trained staff, important in the re-sponses in Ethiopia and, in particular, Nepal.However, transition from IYCF to IYCF-E, suchas the establishment of mother and baby areasto enable focused support to those affected, wasnot straightforward in either Ethiopia or Nepaland required advocacy among key stakeholdersand capacity-building. Pre-crisis IYCF pro-grammes and the availability of trained staffwere weak in Northern Syria, which may partlyexplain the slow IYCF-E response. Finally, linkingIYCF-E to health was also identified as key; inNorthern Syria significant funds for the IYCF-E programme were raised by embedding withinthe health programme.

On the basis of these findings the authorsput forward a set of recommendations for futureIYCF-E response, described in Box 1.

1. Include IYCF-E in pre-crisis IYCF and nutritionpolicies, strategies and preparedness plans.

2. Conduct IYCF-E assessments so that needsare clearly defined.

3. Activation of the Nutrition Cluster should notdepend solely on acute malnutritionprevalence but should also be justifiable on thegrounds of IYCF-E.

4. Secure and support IYCF-E champions (suchas the Ministry of Health, national andinternational NGOs and UN agencies) to raiseawareness about IYCF-E and advocate forimplementation of interventions when needed.

5. Strengthen pre-crisis IYCF interventions andcoordinate with pre-crisis IYCF staff andprogrammes. Consider their views, experienceand knowledge and orientate them oninterventions and mechanisms specific toemergencies, such as mother-baby areas andNutrition Cluster coordination, so that they arefully on board when additional IYCF-Eactivities are developed.

6. Prioritise awareness and capacity-building inIYCF-E by government and humanitarianagencies.

7. Reinforce links between IYCF-E and othersectors, especially health.

Box 1 Recommendations forIYCF-E responses

See Operational Guidance on infant and youngchild feeding in emergencies, v3, 2018 for moreguidance on preparedness and response.https://www.ennonline. net/operationalguid-ance-v3-2017

1 Technical rapid response team (Tech RRT) (2018). Infant and young child feeding in emergencies: an analysis of key factors of a strong response. www.ennonline.net/resource/ iycfanalysisstrongresponse

2 It should be noted that the score for Northern Syria was attributed based on the IYCF-E response as it was in 2017, which had improved considerably from the beginning of the crisis.

37

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Research

Higher heights: A greater ambition formaternal and child nutrition in South AsiaSummary of research1

Poor nutrition in early life threatensthe growth and development of chil-dren, which has a knock-on effect onthe sustainable development of nations.

is is particularly so in South Asia, where 40%of the world’s stunted children (59 million chil-dren) and 53% of the world’s wasted children(27 million children) live (UNICEF et al, 2018).Although the prevalence of child stunting isfalling in the region, the pace of progress is tooslow and most countries with available data arenot on track to meet stunting reduction targets.e UNICEF Regional Office for South Asiacommissioned a series of papers in 2016-2017to fill knowledge gaps in the current body ofevidence on stunting drivers, who is most affectedand effective programme approaches. isoverview paper summarises the evidence fromthese analyses and examines the implicationsfor the direction of future advocacy, policy andprogramme actions.

Child growth and developmentAnalysis of pooled national survey data fromBangladesh, India, Nepal and Pakistan showsthat stunted is concentrated among children ofhouseholds experiencing multiple forms of dep-rivation, including poor child diets, low levelsof maternal education and household poverty(Kirshna et al, 2018). Large inter‐country differ-ences were found in average rates of stuntingreduction, from 0.6 percentage points (pp) peryear in Pakistan, 1.3 pp in India, 2.9 pp inBangladesh and 4.1 pp in Nepal. Stunting hasdeclined across all wealth quintiles in all countries,but inequalities among wealth quintiles havepersisted and widened in Nepal and Pakistan.

A retrospective case series analysis (Aguyaoet al, 2018) examining the effectiveness of Pak-istan’s community-based management of acutemalnutrition (CMAM) programme for severelywasted children (most of whom were aged 6-23months) found that the programme was effectivein achieving high survival (99.6%) and recovery(87.8%) rates. Severely wasted or stunted childrenhad higher death and lower recovery rates com-pared to other children, suggesting that targetingchildren age 6-23 months old with multiple an-thropometric failure will increase impact.

As child survival improves in South Asia, thedevelopmental consequences of poor nutritionin early life become a more pressing concern

than mortality, particularly given the high numberof children with low cognitive and socio-emotionaltest scores in the region (McCoy et al, 2016).Pooled data from multiple indicator clustersurveys (MICS) in Bangladesh, Bhutan, Nepaland Pakistan found that stunted children wereat increased risk of sub-optimal learning/cognitiondevelopment at three to four years old, but foundno association between wasting and learning/cog-nition development (Kang et al, 2018), suggestingthat interventions effective in improving lineargrowth in the first years of life may improveearly childhood development.

Childhood anaemia is also associated withimpaired cognitive development and possiblymotor development. Studies from Nepal andPakistan (Harding et al, 2018) and Bhutan(Campbell et al, 2018) show that anaemia ismore likely in children with an anaemic motherand in infants (suggesting that mother’s anaemiastatus may affect that of her child), in stuntedchildren in all three countries and children withthin mothers in Pakistan (reflecting the contri-bution of dietary inadequacy before and duringpregnancy and in childhood).

Maternal nutrition and lowbirth weight (LBW)South Asia has the highest prevalence of LBW(26%) in the world (Lee et al, 2013), reflectingthe poor status of women’s nutrition in theregion. Goudet et al (2018) estimate that one in10 South Asian women of reproductive age havelow stature (<145 cm), one in five have lowbody mass index (BMI) (<18.5 kg/m2) andoverweight is rising rapidly, all of which arerisk factors for child stunting. Pooled nationalsurvey data from six South Asian countries(Harding et al, 2018) show that children withreported LBW are significantly more likely tobe wasted and severely wasted than non‐LBWchildren and LBW is a predictor of concurrentwasting and stunting. Anaemia in women ofreproductive age in Nepal and Pakistan is asso-ciated with thinness (BMI <18.5 kg/m2) andchildren under five years old are more likely tobe anaemic if their mother is anaemic. is andthe Campbell et al (2018) study in Bhutan showthat anaemia is concentrated in the most disad-vantaged women, including those from thepoorest households in Pakistan, women withoutschooling in Bhutan, and women lacking sani-tation facilities in Bhutan and Nepal. e preva-

lence of anaemia in pregnant women in Bhutanis lower than non‐pregnant, an atypical findingsuggesting that iron folic acid (IFA) supple-mentation during antenatal care is effectivelyprotecting pregnant women from anaemia.

A combination of nutrition-specific and nu-trition-sensitive actions are needed to addresswomen’s low-quality diets, poor access to health,and nutrition services and their causes. A sys-tematic review by Goudet et al (2018) identifiesbarriers to pregnant women receiving and con-suming IFA and calcium supplements at mater-nal-level (low women’s education level andknowledge), household-level (low husband’s ed-ucation level, support from husband and house-hold wealth) and health-facility level (late timingof first antenatal visit and low number of visits).Programme delivery platforms reaching pregnantwomen with supplements in their homes andcommunities, combined with information andcounselling, can improve access to services andconsumption of supplements.

BreastfeedingUsing pooled data from six South Asian countries,Harding et al (2018) find that children were lesslikely to be wasted if they were breastfed withinthe first hour of birth, were not given any pre-lacteal foods, and were exclusively breastfed.e rapid fall in the prevalence of wasting duringthe first few months of life in several SouthAsian countries suggests that early and exclusivebreastfeeding may help infants recover fromLBW. Focusing on Bhutan, Campbell et al(2018b) report that children under two yearsold are less likely to be overweight if they arecurrently breastfed.

Benedict et al (2018a) identify a steadyincrease in early initiation of breastfeeding,avoidance of pre-lacteal feeding and exclusivebreastfeeding in Bangladesh, India and Nepalover the last 25 years. Despite this, only abouthalf of children in these countries benefit fromearly initiation of breastfeeding and exclusivebreastfeeding, and rates of continued breast-feeding at two years have remained stagnant at

1 Torlesse H, Aguayo VM. Aiming higher for maternal and child nutrition in South Asia. MaternChild Nutr. 2018;14(S4): e12739. https://doi.org/10.1111/mcn.12739 within: Hall Moran V and Pérez-Escamilla R (eds) and Aguayo VM and Torlesse H (guest eds). Higher heights: a greater ambition formaternal and child nutrition in South Asia. Maternal and Child Nutrition, November 2018, Volume 14, Supplement 4.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

38

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Research

~70%. Progress in Afghanistan and Pakistanhas lagged behind other countries, with recentdeclines in breastfeeding practices in Afghanistanand in the early initiation of breastfeeding andavoidance of pre-lacteal feeding in Pakistan.

Using data from national surveys, Nguyenet al (2018) report that socio‐economic inequal-ities in the early initiation of breastfeeding andexclusive breastfeeding in India narrowed between2006 and 2016, a significant achievement giventhe rising economic equalities in the country.Improvements in breastfeeding in lower socioe-conomic quintiles appear to have been drivenby improved access to and use of health andnutrition services by mothers and children.

Multivariate analysis of national surveydatasets from South Asia’s five largest countries(Benedict et al 2018a and 2018b) reveals thatcommon predictors of delayed initiation ofbreastfeeding, pre-lacteal feeding and not beingexclusively breastfed include infant being bornby caesarean section, small size at birth andhome delivery (suggesting that these womenand infants need more breastfeeding support)and low women’s empowerment. A review of31 studies by Benedict et al (2018b) reports thatprogrammes to support breastfeeding are morelikely to be effective if they include multiple in-terventions (education and counselling, com-munity mobilisation, mass media and newbornhealth initiatives) in multiple intervention en-vironments (home, community and health fa-cility). Other important factors appear to be in-tervention coverage, timing relative to the ageof the child, frequency, duration and targeting.

Complementary foods andfeeding practices Poor complementary feeding practices are highlyprevalent in South Asia (UNICEF, 2016) andoen predict stunting and wasting in the firsttwo years of life. e likelihood of stunting is

higher in South Asia in infants aged 6-8 monthswho are not fed any complementary foods andin children aged 6-23 months whose diets donot meet minimum dietary diversity (MDD)(Kim et al, 2017). Likelihood of wasting in chil-dren age 6-23 months is higher if their diets donot meet MDD and of severe wasting if theirdiets do not meet minimum meal frequency(MMF) (Harding, Aguayo, and Webb, 2018).In India, not meeting MDD is also associatedwith concurrent wasting and stunting.

A review of South Asia national survey data(2006-2013) found that only 57% of infants aged6-8 months are fed any complementary foodsand, of the diets of children aged 6-23 months,only 48% meet MMF, 33% meet MDD and 21%meet minimum adequacy (sufficient number ofmeals, food groups, and breastmilk or milkfeeds) (Aguayo, 2017). Across all countries, MDDis consistently lower than MMF, indicating thatMDD is a greater problem. Considerable variationin feeding practices exists between countriesand only Sri Lanka and the Maldives have ratesthat exceed 50% for all these practices.

ree multivariate analyses of national surveydata in Afghanistan, Bangladesh and Nepal (Naet al 2018a, 2018b and 2018c) show that com-plementary feeding practices are more likely tobe sub-optimal among infants (6-11 months),first-born children, children whose mothers areyounger or less educated, and in communitieswith poor access to health and nutrition services.Cultural beliefs continue to be a barrier to rec-ommended feeding practices in Nepal, wherethe Dalit and minority ethnic and religiouscastes have poorer complementary feeding prac-tices than other population groups. Wealth quin-tile is associated with dietary diversity (DD) inAfghanistan, Bangladesh and Bhutan, suggestingavailability of affordable nutritious foods is acommon barrier to diverse diets in South Asia(Cambell et al, 2018b). Nguyen et al (2018)

show that the equity gaps in complementaryfeeding practices between socio‐economic statusquintiles in India narrowed between 2006 and2016, but practices remain poor across all groups.Paintal and Aguayo (2016) show that harmfulfeeding practices during childhood illness are awidespread concern in South Asia, in part drivenby inadequate and sometimes harmful advicefrom health workers. Information, educationand counselling delivered by a range ofwell‐trained primary healthcare workers andcommunity resource persons can improve thetimeliness, frequency, diversity and/or adequacyof complementary feeding, although impactmay be limited by lack of availability or afford-ability of nutritious foods (Aguayo, 2017).

Implications for futureadvocacy, policy andprogramme actions 1. e concurrence of child stunting with wasting and anaemia in South Asia requires governments to address all forms of malnutrition in an integrated manner across the life cycle. In the past there has been a tendency to addressdifferent forms of malnutrition in isolation andwith varying levels of intensity. However, theyoen affect the same children and share commonrisk factors (Khara & Dolan, 2014). Policies andprogrammes should move away from siloed ap-proaches and realign to address child malnutritionin all its forms.

2. Improving women’s and children’s diets is central to breaking the inter-generational cycle of malnutrition in South Asia.Complementary foods and feeding practices inSouth Asia remain unacceptably poor due toweaker policy on complementary feeding com-pared to breastfeeding and a lack of clarity ininterventions, approaches and coordination be-tween sectors on this issue. Children’s diets needmuch greater attention by all stakeholders con-cerned and strategies are needed to improveaccess to nutritious and affordable foods, coupledwith communication interventions for behaviourand social change. is series reaffirms the closeconnection between the nutritional status of amother and her children and the need to tackledietary drivers of poor women’s nutrition beforeand during pregnancy.

3. A coordinated, multi-system approach is needed to ensure families have all the inputs they need for children’s healthy growth. Coordination between the food, health, socialprotection, water, sanitation and hygiene (WASH)and education systems is needed and betweendifferent levels of government to combine actionsto improve the nutrition status of women andyoung children.

A blanket supplementary feedingprogamme that aims to prevent acutemalnutrition in children, Bangladesh

WFP

/Kau

ser H

aide

r

39

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


4. Deliberate actions are needed to address the disparities and inequalities in child growth and early life. Targeted efforts are needed to reach children, mothers andcommunities at greatest risk of malnutrition. Interventionsshould focus on children born small; children under two yearsof age; younger, less experienced mothers; and poorer households.At community level the focus should be on communities withhigher levels of anthropometric failure and lower access tomaternal and child health and nutrition services, using trainedhealthcare workers and community volunteers.

5. Continued attention is needed in all countries to gather, analyse and use data to assess progress and inform decisions.is includes the use of routine information systems and periodicsurveys to gather data on anthropometric indicators, feedingpractices and service coverage. Studies and implementationresearch are also needed to better understand the context-specificbarriers, enablers and pathways to improving access to servicesand adoption of recommended nutrition behaviours and practices.

Conclusions e South Asia region bears a disproportionate burden ofstunted children who experience worse health, cognition andlearning outcomes. ese children are concentrated in the mosteconomically disadvantaged households and oen experiencemultiple concurrent forms of nutrition deprivation. Moreattention is needed to improve the nutritional status of womenbefore and during pregnancy and the diets of infant and youngchildren in the first two years of life, while addressing underlyingdrivers. A coordinated, multi-system approach and actions totackle inequalities are needed.

ReferencesAguayo VM, Badgaiyan N, Qadir SS, Bugti AN, Alam MM,Nishtar N & Galvin M. (2018). Community management ofacute malnutrition (CMAM) programme in Pakistaneffectively treats children with uncomplicated severewasting. Maternal & Child Nutrition, 14(Suppl 4), e12623.https://doi.org/10.1111/mcn.12623

Aguayo VM. (2017). Complementary feeding practices forinfants and young children in South Asia. A review ofevidence for action post-2015. Maternal & Child Nutrition,13(Suppl 2), e12439. https://doi. org/10.1111/mcn.12439

Benedict RK, Craig HC, Torlesse H & Stoltzfus RJ. (2018a).Trends and predictors of optimal breastfeeding amongchildren 0-23 months, South Asia: Analysis of national surveydata. Maternal & Child Nutrition, 14(Suppl 4), e12698.https://doi.org/10.1111/mcn.12698

Benedict RK, Craig HC, Torlesse H & Stoltzfus RJ. (2018b).Effectiveness of programmes and interventions to supportoptimal breastfeeding among children 0-23 months, SouthAsia: A scoping review. Maternal & Child Nutrition,14(Suppl 4), e12697. https://doi.org/10.1111/mcn.12697

Campbell RK, Aguayo VM, Kang Y, Dzed L, Joshi V, Waid JL,…West Jr KP. (2018a). Epidemiology of anemia in children,adolescent girls and women age in Bhutan. Maternal &Child Nutrition, 14(Suppl 4), e12740.https://doi.org/10.1111/mcn.12740

Campbell RK, Aguayo VM, Kang Y, Dzed L, Joshi V, Waid J,…West Jr KP. (2018b). Infant and young child feedingpractices and nutritional status in Bhutan. Maternal & ChildNutrition, 14(Suppl 4), e12762.https://doi.org/10.1111/mcn.12762

Goudet S, Murira Z, Torlesse H, Hatchard J & Busch‐HallenJ. (2018). Effectiveness of programme approaches toimprove the coverage of maternal nutrition interventions inSouth Asia. Maternal & Child Nutrition, 14(Suppl 4),e12699. https://doi.org/10.1111/mcn.12699

Harding KL, Aguayo VM, Namirembe G & Webb P. (2018).Determinants of anemia among women and children inNepal and Pakistan: An analysis of recent national surveydata. Maternal & Child Nutrition, 14(Suppl 4), e12478.https://doi.org/10.1111/mcn.12478

Harding KL, Aguayo VM & Webb P. (2018). Birthweight andfeeding practices are associated with child growth outcomesin South Asia. Maternal & Child Nutrition, 14(Suppl 4),e12650. https://doi.org/10.1111/mcn.12650

Kang Y, Aguayo VM, Campbell RK, Dzed L, Joshi V, Waid JL,…West Jr KP. (2018). Nutritional status and risk factors forstunting in preschool children in Bhutan. Maternal & ChildNutrition, 14(Suppl 4), e12653.https://doi.org/10.1111/mcn.12653

Khara T & Dolan C. (2014). Technical briefing paper:Associations between wasting and stunting. Policy,programming, and research implications. EmergencyNutrition Network (ENN). Available athttps://www.ennonline.net/waststuntreview2014

Kim R, Mejia‐Guevara I, Corsi DJ, Aguayo VM &Subramanian SV. (2017). Relative importance of 13correlates of child stunting in South Asia: Insights fromnationally representative data from Afghanistan,Bangladesh, India, Nepal, and Pakistan. Social Science &Medicine, 187, 144–154.https://doi.org/10.1016/j.socscimed.2017.06.017

Krishna A, Mejía‐Guevara I, McGovern M, Aguayo VM &Subramanian SV. (2018). Trends in inequalities in childstunting in South Asia. Maternal & Child Nutrition, 2017.https://doi.org/10.1111/mcn.12517

Lee AC, Katz J, Blencowe H, Cousens S, Kozuki N, Vogel JP,… CHERG SGA‐Preterm Birth Working Group (2013).National and regional estimates of term and preterm babiesborn small for gestational age in 138 low-income andmiddle-income countries in 2010.

McCoy DC, Peet ED, Ezzati M, Danaei G, Black MM, SudfeldCR … Fink G. (2016). Early childhood developmental statusin low and middle-income countries: National, regional, andglobal prevalence estimates using predictive modeling. PLoSMedicine, 13, e1002034. https://doi.org/10.1371/journal.pmed.1002034

Na M, Aguayo VM, Arimond M, Dahal P, Lamichhane B,Pokharel R.… Stewart CP. (2018). Trends and predictors ofappropriate complementary feeding practices in Nepal: Ananalysis of national household survey data collectedbetween 2001 and 2004. Maternal & Child Nutrition,14(Suppl 4), e12564. https://doi.org/10.1111/mcn.12564

Na M, Aguayo VM Arimond M, Narayan A & Stewart CP.(2018). Stagnating trends in complementary feedingpractices in Bangladesh: An analysis of national surveys from2004-2014. Maternal & Child Nutrition, 14(Suppl 4),e12624. https://doi.org/10.1111/mcn.12624

Na M, Aguayo VM, Arimond M & Stewart CP. (2017). Riskfactors of poor complementary feeding practices in Pakistanichildren aged 6-23 months: A multilevel analysis of theDemographic and Health Survey 2012-2013. Maternal &Child Nutrition, 13(Suppl 2), e12463.https://doi.org/10.1111/mcn.12463

Nguyen PH, Avula R, Headey D, Tran LM, Ruel MT & MenonP. (2018). Progress and inequalities in infant and young childfeeding practices in India between 2006 and 2016. Maternal& Child Nutrition, 14(Suppl 4), e12663.https://doi.org/10.1111/mcn.12663

Paintal K & Aguayo VM. (2016). Feeding practices for infantsand young children during and after common illness.Evidence from South Asia. Maternal & Child Nutrition,12(1), 39–71. https://doi.org/10.1111/mcn.12222

UNICEF (2018). Child stunting, hidden hunger and humanCapital in South Asia: Implications for sustainabledevelopment post 2015. UNICEF: Kathmandu, Nepal.

Research

The provision of safe drinking water and sanitation are essentialcomponents for increasing the nutritional impact of school mealsat Shree Janajyoti Lower Secondary School, Dadeldhura, Nepal

WFP

/Lilu

KC

40

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Research

Independent and combined effects of improvedWASH and improved complementary feeding onchild stunting and anaemia in rural ZimbabweSummary of research1

Location: Zimbabwe What we know: Stunting and anaemia remain prevalent in children; plausible interventions have shown limited orinconsistent impact.

What this article adds: A cluster-randomised, community-based trial in two rural districts in Zimbabwe (2012-2015) tested the impact of a combined water, sanitation and hygiene (WASH) and infant and young child feeding(complementary feeding counselling and small-quantity, lipid-based nutrient supplement) intervention. Clusterswere randomly assigned to standard care, IYCF, WASH, or IYCF plus WASH. Primary outcomes were infantlength-for-age z-score (LAZ) and haemoglobin concentrations at 18 months of age. In total 5,280 pregnant womenwere enrolled and 3,686 children assessed. e IYCF intervention alone reduced the number of stunted childrenfrom 35% to 27% and the number of children with anaemia from 13.9% to 10.5%. e WASH intervention (aloneor combined with IYCF) had no effect on either primary outcome. Neither intervention reduced the prevalence ofdiarrhoea at 12 or 18 months. Given the drive to scale up nutrition-WASH interventions to address stunting,greater investment in the WASH sector to identify and deliver more efficacious interventions is urgently needed.

Child stunting reduces survival andimpairs neurodevelopment. Offspringof adults who were stunted as childrenare at increased risk of stunting. Tar-

geted interventions have largely failed to addressstunting. Childhood anaemia is also prevalentamong children under two years old in Africaand Asia and is a primary cause of cognitivedelay. Increasing dietary iron intake only reducesanaemia by 32-62% (SHINE, 2015). e UNICEFframework for undernutrition highlights inad-equate dietary intake and disease as the immediatecauses of child undernutrition and specifies thata multi-sector approach that addresses bothproximal and distal determinants is required.us, integration of improved infant diets withimproved water, sanitation and hygiene (WASH)is a logical approach, given the role of WASH inreducing morbidity, especially diarrhoea. In thisstudy the independent and combined effects ofimproved WASH and improved infant and youngchild feeding (IYCF) on stunting and anaemiawere tested in rural Zimbabwe (SHINE trial).

A cluster-randomised, community-based, 2× 2 factorial trial was carried out in two ruraldistricts in Zimbabwe. Clusters were defined asthe catchment area of between one and fourvillage health workers employed by the ZimbabweMinistry of Health and Child Care. Womenwere eligible for inclusion if they permanentlylived in clusters and were confirmed pregnant.Clusters were randomly assigned (1:1:1:1) tostandard of care (52 clusters), IYCF (20 g of asmall-quantity lipid-based nutrient supplement

(LNS) per day from age 6 to 18 months pluscomplementary feeding counselling; 53 clusters),WASH (construction of a ventilated improvedpit latrine, provision of two handwashing stations,liquid soap, chlorine, and play space plus hygienecounselling; 53 clusters), or IYCF plus WASH(53 clusters). A constrained randomisation tech-nique was used to achieve balance across thegroups for 14 variables related to geography,demography, water access and community-levelsanitation coverage. Masking of participantsand fieldworkers was not possible. e primaryoutcomes were infant length-for-age z-score(LAZ) and haemoglobin concentrations at 18months of age among children born to motherswho were HIV negative during pregnancy.ese outcomes were analysed in the inten-tion-to-treat population. e authors estimatedthe effects of the interventions by comparingthe two IYCF groups with the two non-IYCFgroups and the two WASH groups with thetwo non-WASH groups, except for outcomesthat had an important statistical interactionbetween the interventions.

Between 22 November 2012 and 27 March2015, 5,280 pregnant women were enrolledfrom 211 clusters. A total of 3,686 childrenborn to HIV-negative mothers were assessed atage 18 months (884 in the standard of caregroup from 52 clusters, 893 in the IYCF groupfrom 53 clusters, 918 in the WASH group from53 clusters, and 991 in the IYCF plus WASHgroup from 51 clusters). In the IYCF interventiongroups, the mean LAZ was 0·16 (95% CI 0·08–

0·23) higher and the mean haemoglobin con-centration was 2·03 g/L (1·28–2·79) higher thanthose in the non-IYCF intervention groups.e IYCF intervention reduced the number ofstunted children from 620 (35%) of 1,792 to514 (27%) of 1,879 and the number of childrenwith anaemia from 245 (13·9%) of 1,759 to 193(10·5%) of 1,845 and significantly increasedmean weight-for-age, weight-for-height andhead-circumference-for-age Z scores comparedwith the non-IYCF interventions. e WASHintervention had no effect on either primaryoutcome. Neither the IYCF nor the WASH in-tervention reduced the prevalence of diarrhoeaat 12 or 18 months of age.

e authors discuss the fact that, consistentwith decades of complementary feeding research,the IYCF interventions increased linear growthand haemoglobin concentrations, reduced stunt-ing by 21%, reduced anaemia by 24% and in-creased head circumference and ponderal growthcompared with the non-IYCF interventions. Al-though the effects of complementary feedingeducation could not be separated out from thoseof LNS, formative work shows that both com-ponents are important. In contrast, no benefitwas detected for the WASH intervention on any

1 Humphrey, Mbuya, Ntozini, Moulton, et al. (2019). Independent and combined effects of improved water, sanitation, and hygiene, and improved complementary feeding, on child stunting and anaemia in rural Zimbabwe: a cluster-randomised trial. Lancet Glob Health. 2019 Jan;7(1):e132-e147. doi: 10.1016/S2214-109X(18)30374-7. www.thelancet.com/journals/langlo/article/PIIS2214-109X(18)30374-7/fulltext

41

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Children under five years of age withbody dimensions <-2 standard devi-ations (SD) of World Health Organ-ization (WHO) Multicentre Growth

Reference Study (MGRS) charts are defined asundersized. However, the MGRS charts, designedas a global reference (for comparison), are beingused as a standard (target to be achieved). isaspiration should be balanced with realistic ex-pectations. e MGRS charts describe the growth

of children who are free from socioeconomic,environmental and biological constraints, andwhose care has followed recommended healthpractices and behaviours associated with healthyoutcomes. us, MGRS growth curves are ex-pected to provide a single international referencethat represents the best description of physio-logical growth for all children below five yearsof age; low anthropometric values in MGRSsubjects reflect biological variability or additional

unmeasured factors, rather than nutrition com-promise. Currently, only 5-10% of Indian childrenmeet the WHO MGRS eligibility criteria. Inthis context, a recent paper examines how nu-trition progress can be achieved in the Indianpopulation.

Undersized Indian children: Nutrients-starvedor hungry for development? Summary of research1

Location: India What we know: e WHO Multicentre Growth Reference Study (MGRS) growth curves provide a single internationalreference for population-level comparison for children under five years of age.

What this article adds: e MGRS charts describe the growth of children under ideal conditions; only 5-10% of Indianchildren fulfil the WHO MGRS eligibility criteria. e prevalence of undersized children is a proxy for overallsocioeconomic development, intergenerational factors, biological and environmental determinants, and nutrition.However, interventions to improve child undersize in India focus on food and nutrients-based interventions, despitelimited, unstainable benefits and some negative consequences. With the exception of water, sanitation and hygiene(WASH), other broader determinants are not actively targeted. Prevalence of stunting and underweight has declined inIndia (in tandem with overall national development), while wasting remains stagnant; this anthropometric pattern mayreflect nutrition transition. Undersize in Indian children reflects wider societal maldevelopment and inequity and a legacyof considerable intergenerational handicap, necessitating a greater catch-up period. Comprehensive (simultaneouslyaddressing all determinants), patient and equitable (prioritising the deprived) development is the key to progress.

1 Sachdev HPS. (2018) Undersized Indian children: nutrients-starved or hungry for development? Proc Indian Natn Sci Acad 84 No. 4 December 2018 pp. 867-875.

reported child health outcomes. is finding isinconsistent with a previous review on waterchlorination and handwashing promotion, whichwere estimated to reduce diarrhoea by around25% (Ejemot-Nwadiaro et al, 2015); most studiesin this review had very high intervention doses(daily to weekly contact between behaviour-change promoters and study participants), whichwas greater than the monthly contact deliveredby SHINE. us, adherence may not be sufficientto reduce diarrhoea when intervention dose isless frequent than monthly, even when behav-iour-change messages are based on extensiveformative research, delivered by highly trainedworkers and accompanied by free provision ofsoap and chlorine, as in SHINE.

Another important aspect may be that SHINEintervened at the household rather than com-munity level as it was reasoned that young chil-dren spend most of their time within their ownhousehold. Increased community sanitationcoverage, even in sparsely populated areas, maybe required to affect growth. Also, althoughthe SHINE WASH intervention considerablyreduced geophagia and consumption of chickenfaeces by maternal history, it did not preventthese behaviours (27% of WASH mothers still

ReferencesEjemot-Nwadiaro RI, Ehiri JE, Arikpo D, Meremikwu MM,Critchley JA. Hand washing promotion for preventingdiarrhoea. Cochrane Database Syst Rev 2015; 9: CD004265.

Luby SP, Rahman M, Arnold BF, et al. Effects of waterquality, sanitation, handwashing, and nutritionalinterventions on diarrhoea and child growth in ruralBangladesh: a cluster randomised controlled trial. LancetGlob Health 2018; 6: e302–15.

Null C, Stewart CP, Pickering AJ, et al. Effects of waterquality, sanitation, handwashing, and nutritionalinterventions on diarrhoea and child growth in rural Kenya:a cluster-randomised controlled trial. Lancet Glob Health2018; 6: e316–29.

SHINE Trial team. The Sanitation Hygiene Infant NutritionEfficacy (SHINE) trial: rationale, design, and methods. ClinInfect Dis 2015; 61 (suppl 7): S685–702.

reported they had observed geophagia at the12-month visit); more analysis of the data isneeded to find out why.

SHINE is the third trial in which a WASHintervention alone or in combination with anIYCF intervention had no effect on linear growth(Luby et al, 2018; Null et al, 2018). Althoughthese findings do not unequivocally prove thatan integrated WASH-nutrition approach willnever improve linear growth in any context,these trials included over 18,000 children inthree diverse settings with prevalent stuntingand poor environmental hygiene and infantdiets. e authors propose that this may be be-cause faecal ingestion does not reduce environ-mental enteric dysfunction (EED) (or preventionof EED does not improve linear growth); thatWASH interventions used were not sufficientlyeffective to facilitate linear growth or reduce di-arrhoea; or that the trials did not address inter-generational prenatal factors that could potentiallybe targeted by preconception dietary supple-mentation of mothers.

ere is a large movement to scale up inte-grated WASH-nutrition interventions for stuntingprevention. e SHINE trial provides high-level

evidence that elementary WASH interventionsdelivered at the household level in rural areasof low-income and middle-income countriesare unlikely to reduce stunting and might notreduce diarrhoea, and that implementation ofthese WASH interventions together with IYCFinterventions will not reduce stunting morethan implementation of IYCF alone. ese find-ings provide an urgent call for greater investmentin the WASH sector to identify and deliver moreefficacious interventions.

Research

42

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Considering the stringent MGRS eligibilitycriteria, the prevalence of undersized childrenat population level is a crude but convenientproxy for a blend of overall socioeconomic de-velopment and intergenerational constraint ofmaternal undersize. Nutrient deficits, particularlyin individuals, may comprise only one contributingfactor, yet food and nutrient subsidies or sup-plementation are typical and oen the onlystrategies used to address population undersize.While there are certain benefits in disaster andfamine situations, only some nutrient or food-based interventions, including those during preg-nancy, are evidenced to increase anthropometricindices or do so modestly (0.1-0.25 SD or 5%-10% deficit), with unsustainable gains. Undesirableconsequences include complacency among stake-holders that everything possible is being done;diversion of public finances and attention fromother crucial unattended determinants of un-dersize; dependence by the beneficiaries onpublic welfare; and risk of contributing to non-communicable diseases through long-term, food-based supplementation.

Other potential determinants of undersizeinclude water, sanitation and hygiene (WASH);nutrition counselling; maternal characteristics;

curative and preventive health care; maternal,household and community resources; literacy;income; women empowerment; safety nets; andgenes. ese are seldom visualised or activelytargeted for improving undersize, althoughefforts have begun to integrate WASH inter-ventions. Attempts to accelerate the pace ofprogress must consider inequity (the greaterthe proportion of the population that is deprived,the higher the burden and slower the improve-ment); comprehensive interventions; using ‘win-dows of opportunity’ (such as the first 1,000days and adolescence); sustainability of inter-ventions, ideally over generations; baselines(populations starting at lower levels will takelonger to reach targets); and intergenerationalhandicap (only a limited improvement is possiblewithin one generation).

In tandem with overall national development,time trends show a gradual decline in the inci-dence of undersize in children; this trend hashastened in the past decade, even among theunderprivileged. is is supported by intergen-erational comparison, over the past 30 years, ofanthropometric data of children and their parentsin the New Delhi Birth Cohort subjects frommiddle socioeconomic status who were not re-

cipients of food subsidies (Sinha et al, 2017). Incomparison to their parents, children were con-siderably taller (0-5 years 0.99 SD; 5-10 years1.17 SD) and heavier (0-5 years 0.77 SD; 5-10years 1.52 SD), while only those aged 5-10 yearswere broader (had a higher BMI; 1.03 SD). esteady decline in the prevalence of stunting andunderweight, with stagnant wasting levels, isobserved in neighbouring countries too andappears unrelated to specific determinants. eunderlying and proximate factors related tostunting and wasting are similar. A steady im-provement in stunting and underweight withstagnant wasting levels is therefore unlikely tobe related to non-improvement of specific de-terminants like food or nutrients, but is probablya reflection of biological thinness (thin-fat infantphenotype) of Indian populations (Yajnik et al,2003), or the pattern of anthropometric changein stunted populations undergoing nutritiontransition. In children under five years old, in-crease in length (or height) and breadth are twodistinct biological processes, which generallydo not occur simultaneously. us, rapid andsimultaneous declines in both stunting andwasting prevalence, as perceived in the SustainableDevelopment Goals, appear challenging, if notimpossible, to achieve.

Politicians, policy makers, other stakeholdersand the lay public must realise that there is nomagic solution to eliminate undersize in children,which reflects wider societal maldevelopmentand inequity. Focusing solely on nutrients andone or two additional determinants (for example,WASH interventions) will yield slow and disap-pointing results. Irrespective of the benefits onbody size, every ingredient of the developmentprocess deserves to be in place in its own right;for example, purchasing power; access to edu-cation, healthcare, water supply and sanitationservices; and nutritional security. A seeminglyunclear agenda can be more clearly structuredby uniting two to three key indicators each fromthe above key domains to monitor progress andfine-tune interventions.

e author concludes that undersized Indianchildren have a legacy of considerable intergen-erational handicap, necessitating a greater catch-up period even under the best circumstances.We therefore need to be patient and practical.ere has been a gradual improvement, whichhas sped up in the past decade. A predominantfocus on nutrients-based solutions will fail toaccelerate progress. Comprehensive and equitabledevelopment is the key to success.

ReferencesSinha S, Aggarwal AR, Osmond C, Fall CHD, Bhargava SKand Sachdev HS (2017). Intergenerational change inanthropometric indices and their predictors among childrenin New Delhi Birth Cohort. Indian Pediatr 54 185-192.

Yajnik CS, Fall CHD, Coyaji KJ, Hirve SS, Rao S, Barker DJ P,Joglekar C and Kellingray S (2003). Neonatalanthropometry: the thin-fat Indian baby. The Pune MaternalNutrition Study Int J Obes Relat Metab Disord 27 173-80.

ICDS centers where nutrition, hygeneand health care education is providedin Banswara, Rajasthan.

WFP

/Rei

n Sk

ulle

rud

43


By Sanjay Kumar Das, Dr Sule Meleh,Dr Umar Chiroma, Bulti Assaye andMaureen L Gallagher

Sanjay Kumar Das is a nutritionmanager for UNICEF Maidugurioffice for the north-easternNigeria response. He is a publichealth nutritionist with over 15years’ experience in nutrition

programme design, coordination, partnership andresource mobilisation and has worked in Somalia,Kenya, Ethiopia, Sudan, Yemen, Jordan, Pakistan,India and Nepal.

Dr Sule Meleh is ExecutiveDirector, Borno State PrimaryHealth Care DevelopmentAgency (BSPHCDA), where heoversees the overallmanagement of primary

healthcare services in Borno State, north-easternNigeria. He previously worked as a senior medicalofficer in paediatrics for 18 years before taking upa senior leadership position in public health eightyears ago.

Dr Umar Chiroma is DeputyDirector of Primary Healthcarefor Yobe State Primary HealthCare Management Board. He isa public health specialist,supervising and managing

nutrition response activities in Yobe State. He haseight years’ experience in public health, includingnutrition services.

Bulti Assaye is a nutrition officerfor UNICEF Abuja office and isresponsible for nutritionprogramme data analysis,information management andreporting.

Maureen L Gallagher is theChief of Nutrition for UNICEFAfghanistan. She waspreviously the NutritionSpecialist with UNICEF’sEmergency Response Team

based in UNICEF headquarters, when shesupported emergencies in Nigeria, Yemen andDemocratic Republic of Congo. She is a publichealth specialist with over 15 years’ experience innutrition programming in several countries inAfrica and Asia.

The authors thank Geoffrey Ijumba, Chief ofUNICEF Maiduguri field office, and Dr SimeonNanama, Nutrition Section Chief of UNICEF Nigeriafor reviewing this article and providing strategicguidance and support for the scale-up of thenutrition response in north-eastern Nigeria. Theauthors would also like to acknowledge the wholenutrition teams of Borno PHCDA, Yobe SPHCMB,UNICEF Maiduguri and Abuja office for theircommitment and hard work in the scale-up of thenutrition response in Borno and Yobe states. The findings, interpretations and conclusions inthis article are those of the authors. They do notnecessarily represent the views of UNICEF, itsexecutive directors, or the countries that theyrepresent and should not be attributed to them.

Location: NigeriaWhat we know: Six years of armed conflict in north-eastern Nigeria haveresulted in mass displacement, severe food insecurity and undernutrition, andhave destroyed livelihoods and further weakened fragile health systems.

What this article adds: In 2010 UNICEF supported integration of community-based management of acute malnutrition (CMAM) services (SAM treatment)into 75% of fixed health facilities in north-eastern Nigeria. In 2015 infant andyoung child feeding (IYCF) support and multiple-micronutrient powder (MNP)were added. Government has led scale-up of this integrated package; thenumber of centres has increased fourfold and the number of children treatedhas increased fivefold (2013-17). UNICEF has focused on supporting stategovernment to meet immediate needs of severely malnourished children whilestrengthening the government health system. A minimum nutritional servicespackage was finalised in early 2017. Multiple delivery platforms have been used(fixed health centres, IDP camp nutrition centres, community outreach, mobileservices) and community nutrition mobilisers have increased coverage,including in insecure areas. UNICEF has provided financial and technicalsupport, including supervision and monitoring, barrier analysis to poorperforming sites, and context-specific supply chain management with rolesdefined for multiple stakeholders. Financing sources for scale-up have beenmixed; some government, some UNICEF-dependent (especially in highhumanitarian need states). There has been some success and ongoing advocacyto transition from emergency to development funding sources. Key elementsthat enabled SAM scale up were strong government leadership with UNICEFsupport; good communication (especially use of WhatsApp groups); use ofexisting community platforms for community mobilization; and investment inon-the-job training and supportive supervision.

Nutritional response in north-eastern Nigeria: Approaches to increase service availabilityin Borno and Yobe States

BackgroundBefore the conflict the situation in Borno andYobe States was already fragile, with highlevels of poverty, underdevelopment, unem-ployment and inequality. Six years of armedconflict and clashes between government andnon-state armed groups (NSAGs) has resultedin millions of people displaced, thousands oflives lost and livelihoods destroyed in north-

eastern Nigeria. e prolonged crisis had ledto a deterioration in health and other basicservices, resulting in increased severe foodinsecurity and undernutrition. Outbreaks ofdiseases such as measles, cholera and diarrhoeahad also been reported in the area, furtheraggravating the situation.

More than half of internally displaced per-sons (IDPs) in north-east Nigeria fled their

Counselling caregivers on optimal IYCFpractices, Borno State, Nigeria, 2018

Field Article .....................................................

44

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Field Article

homes in 2014 and 2015 aer NSAGs seizedcontrol of a territory covering more than 30,000square kilometres, committing grave human rightsabuses against local populations. A government-led military campaign subsequently saw the gov-ernment regain control of the area. In 2016 theNigerian security forces recaptured the maintowns and villages of several local governmentareas (LGAs) and exposed the humanitarian needsof civilians there. Over 250,000 people living insecure ‘satellite’ camps at LGA capitals (includingBama, Dikwa, Monguno and Damboa) were re-ported in need of emergency assistance. Giventhe onset of the lean season and rainy season be-tween May and September, the near-term outlookwas also grim for the 180,000 food-insecure peoplein Maiduguri’s outer perimeter and 120,000 Maid-uguri camp residents.

Since 2016 there has been improvement inthe access of such communities to life-savingassistance; out of a total 27 LGAs in BornoState, 25 are currently accessible, compared tolimited access to 22 LGAs in 2016 (see Figure

1). However, humanitarian access in conflict-affected areas continues to be constrained byrestrictions on the movement of aid workersand civilians, especially in Borno State and tosome extent Yobe State, as well as ongoingviolence and a challenging physical environment(particularly during the rainy season). BornoState clearly remains the epicentre of the hu-manitarian crisis, with dozens of conflict incidentsreported each month. In the crisis affectingnorth-east Nigeria, nine out of ten displacedpersons come from Borno and the State alsohosts the majority (78%) of IDPs.

Nutrition situatione Cadre Harmonisé (CH) analysis1 of March2018 indicates that an estimated 3.7 millionpeople required emergency assistance betweenMarch and May 2018, with 3.9 million beingreported as food-insecure. For the projectedperiod of June to August 2018, these figureswere expected to increase to about 5.3 millionpeople in the absence of resilience-driven in-terventions and humanitarian assistance. Only

30% of health facilities are functional in BornoState, where malaria is endemic and choleraand other diseases affect the population regularlyand are oen life-threatening.

An estimated 943,000 children under fiveyears old across Borno and Yobe States are acutelymalnourished; 440,000 with severe acute malnu-trition (SAM) and 503,000 with moderate acutemalnutrition (MAM). Around 230,000 pregnantand lactating women (PLW) are estimated to beacutely malnourished. e nutrition situation inBorno and Yobe States declined rapidly due tothe crisis in 2015, peaked in 2017 and has stabilisedin 2018, with some exceptions in hard-to-reachareas (mainly in Damasak, Gubio, Kukawa andnorthern Yobe). Recent nutrition surveillance(by UNICEF, in collaboration with the NationalBureau of Statistics (NBS)) revealed GAM andSAM rates of 6.4% and 1.2% in Borno and 12.0%and 1.3% in Yobe respectively.

Nutrition response scale-up: Acombined approach e nutrition response in north-eastern Nigeriahas been driven by the strong political will andleadership of the Ministry of Health (MoH).While public services and institutions have beenoverstretched, and in some cases destroyed inaffected areas, services have continued to functionwhere possible. e emergency nutrition responsehas been integrated into the existing system,with active participation of key pre-existingstakeholders and therefore provides an exampleof humanitarian development programming inresponse to crisis.

UNICEF initiated support to community-based management of acute malnutrition(CMAM) programming in the north-east in2010 for identification and treatment of childrenwith SAM2. SAM management was integratedin about 75% of fixed health facilities (FHFs)and UNICEF provided training and onsite coach-ing to enhance the capacity of governmenthealth workers and supplied ready-to-use ther-apeutic food (RUTF) and basic equipment foridentification and treatment of SAM children.Infant and young child feeding (IYCF) supportand multiple micro-nutrient powder (MNP)supplementation activities were initiated in 2015aer the crisis began. ese were integratedinto service delivery platforms in health facilitiesalready delivering SAM treatment.

e push to scale up CMAM provision beganin 2015 (as described in Figure 2), when thenumber of fixed health facilities (FHFs) deliveringservices increased by almost 30%. Scale-up wasmainly focused on SAM children (admission

1 The Cadre Harmonisé is the equivalent of the Integrated Phase Classification (IPC) approach used in West Africa.

2 UNICEF started to use extended criteria (MUAC<125mm) in August 2017 to admit MAM children in outpatient therapeuticprogrammes and provided seven sachets of RUTF for one week in some hotspot areas as the nutrition situation was critical, based on nutrition surveillance findings. This has been stopped from 1 April 2018. World Food Programme is doing a blanket supplementary feeding programme (BSFP) for children aged 6-23 months and targeted supplementary feeding program (TSFP) for children aged 24-59 months in some areas.

Maps of Borno and Yobe state indicating availability of nutrition programmein LGA headquarters and inaccessible area in 2016Figure 1

Legend State boundry Other Areas Areas with Program in LGA HQ

In-accessible area

Timeline of major nutrition-related activities in north-eastern Nigeria Figure 2

45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Field Articlebased on MUAC <115mm or oedema, dischargeMUAC> 125 mm and no oedema), as reflectedin the national CMAM guideline that centredon SAM management. Between 2013 and 2017the number of sites delivering CMAM increasedfourfold (from 109 in 2013 to 461 in 2017) andthe number of children admitted for SAM treat-ment increased by over fivefold (from 30,288 in2013 to 181,170 in 2017) (see Figure 3). At theend of 2017 partners in the Nutrition Sectorstarted to work towards a minimum nutritionalservices package, which was finalised at the be-ginning of 2017. is established the premisethat all health facilities’ delivery services had toinclude a minimum package of CMAM, IYCFand MNP services (the latter provided to childrenaged 6-23 months not affected by SAM orMAM). is significantly increased the coverageof IYCF and MNP services (see Figure 4). eprocess was led by the Nutrition Sector, withUNICEF participating in the development ofan integrated nutrition training package anddata tools for health workers and communitynutrition mobilisers (CNMs).

e integrated nutrition package approachwas placed at the core of the scale-up strategyto increase availability of and access to nutritionservices throughout Borno and Yobe States.is was achieved through the use of multipleplatforms to deliver the minimum nutritionservices package, as summarised in Box 1. Effortswere made by sector partners to strengthen thecommunity-based aspects of the programmethrough the recruitment and use of CNMs, em-ployed to increase community mobilisation anddemand for services (Box 2). In the context ofconflict and insecurity, this approach has led tothe timely identification and treatment of SAMchildren, promotion of IYCF practices and ap-propriate use of MNP at household level. Workingwith local NGOs to provide CNMs has beenbeneficial, given their good community networkand access at community level with limited se-curity challenges. e new strategy achievedimpressive results in terms of increased numberof units delivering CMAM services and increasedadmissions (as illustrated in Figure 2).

Supervision and monitoring To address challenges of monitoring in areas ofdifficult access, UNICEF has a contract with alocal institution to deploy skilled personnel tobe based at LGA level. Considering the securitysituation, working through the local institution,personnel can move frequently to and withinthe LGA. e UNICEF Nutrition section de-ployed 31 senior nutrition LGA facilitators toprovide close monitoring and onsite coachingto nutrition staff at LGA and service deliverysite level. ey support coordination of nutritionactivities at LGA level and provision of onsitecoaching to nutrition service providers, as wellas collection and compilation of nutrition activityreports together with local government nutritionfocal persons. is initiative has been vital toimprove quality of services, avoid stock-outissues and strengthen timely reporting.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Box 1 Approaches implemented by partners to increase availability of nutritional services

Fixed health facilities (FHFs): FHFs are part ofthe government routine health system in hostcommunities before the crisis. Nutrition servicesare integrated into FHFs as a component of theprimary healthcare package. UNICEF hassupported state government to integratenutrition services (SAM treatment, IYCF supportand MNPs) into primary healthcare services in312 FHFs in Borno and Yobe States.

IDP camp facilities: These are set uptemporarily in IDP camps to provide basic healthand nutrition services to IDPs after they havearrived in a safe and secure location. IDP campfacilities were first established in 2015 inresponse to increased numbers of IDPs and thesubsequent set-up of the first IDP camp in BornoState. UNICEF has set up 14 nutrition servicesites in IDP camps for the provision of integratednutrition services. UNICEF mobilised healthworkers displaced to IDP camps to work in thehealth/nutrition clinics, since they were still paidby government, and offered financial top-ups tomotivate them to continue their work.

Community outreach: A community outreachteam comprises two skilled people (one CMAMprovider and one nutrition screener) for the

identification and treatment of SAM children indensely populated areas, mainly in hostcommunities. These teams were launched toreduce overcrowding in FHFs and provide dailynutrition services. Community outreach startedin central Borno LGAs (Maiduguri metropolitancity (MMC), Jere, Konduga and Mafa) in 2016 toaddress the increased IDP influx and increaseddemand in services. UNICEF, in collaborationwith state government, has set up 35 outreachsites from which community outreach teamsoperate.

Mobile teams: A mobile nutrition teamcomprises three skilled nutrition personnel (aCMAM provider, a nutrition counsellor and anutrition screener/recorder). The team movesfrom one location to another to cover betweenthree and five locations each week for theprovision of integrated nutrition services insettlements/ communities far away (>3 km) fromFHFs. This approach started in August 2017 toincrease the availability of nutrition services inremote communities. UNICEF, in collaborationwith state government, has so far deployed 49mobile nutrition teams, often to newlyaccessible areas where governmentinfrastructures are completely damaged.

Box 2 Community Nutrition Mobilisers (CNM)

CNMs are identified from the local community tostrengthen the community mobilisation andsensitisation component of the nutritionprogramme. Duties include: active house-to-house nutrition screening of all children underfive years old (700 to 1,000 children per CNM)and identification and referral of SAM children(ten days per month); follow up SAM children toensure completion of treatment and appropriateuse of RUTF and MNPs (four days per month);formation and facilitation of mother supportgroups (MSGs) for the protection, promotionand support of IYCF practices (five days permonth); and one-to-one support for motherswith breastfeeding problems (three days permonth). CMNs refer malnourished children(aged 6-23 months) or MAM children who arenot receiving supplementary food for MNP, while

SAM children are referred to receive appropriatetreatment (RUTF and routine medicine). CNMsare paid and are guided by a terms of reference(TOR) defined and agreed by sector partners. UNICEF has partnered with three local non-governmental organisations (NGOs) tostrengthen community mobilisation activitiesand ensure timely payment of CNMs. Localpartner NGOs provide coaching to andsupportive supervision of CNMs through 121ward-level nutrition supervisors (local NGOstaff). Supervisors conduct monthly reviewmeetings between CNMs and FHF nutrition staffto collect monthly reports from CNMs, providefeedback on data quality, discuss challenges andlessons learned, provide monthly incentives toCNMs and plan community mobilisationactivities for the subsequent month.

Trends in SAM admissions and scale-up of CMAM service delivery units (SDUs)in Borno and Yobe States 2013-2017Figure 3

20000018000016000014000012000010000080000600004000020000

0

Number of Admissions

2013 2014 2015 2016 2017Year

Number of sites delivering services Number of SAM children admitted for treatment

30,28848,384

71,325

146,298

181,170 500450400350300250200150100500

Number of Service Delivery Units

109 158 218 361 461

46

Field Article. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Field Exchange issue 59, January 2019, www.ennonline.net/fex

WhatsApp groups for nutrition were createdfor communication and regular reporting usingmobile coverage from neighbouring countries;coverage of networks within Nigeria was poor.As most of the LGA teams were new to the in-tegrated nutrition response, the platform wasalso used to discuss technical issues arising inthe field and provide remote technical support. In the beginning of 2017 achievement of nutritionindicators was very low relative to the humani-tarian response plan (HRP) target; for example,admission of SAM children reached only 24.4%of target by end of May 2018. e total statetarget was divided for each LGA andtarget/achievement ratios were monitored month-ly to identify areas with poor achievements andexplore the reasons. ose LGAs and FHFs wereprioritised to strengthen community mobilisationactivities and proper reporting and addressrelated challenges for better programme out-comes. is analysis sensitised the team andencouraged them to take adequate action to en-sure all children under five years old are screenedon a monthly basis and SAM children are ad-mitted to a nutrition programme.

Supply chain managementGiven the multiple stakeholders involved in thenutrition programme, and to overcome accessdifficulties, a context-specific supply managementapproach was used to mitigate any stock-outsat SDUs. Responsibilities of the different stake-holders were clearly defined, as follows: • UNICEF: Responsible for supply provision of outpatient therapeutic programme (OTP) sites of newly liberated areas and south Borno. e UNICEF nutrition staff work with LGA nutrition facilitators to compile RUTF and MNP requirements for the logistics team, who supply to SDUs every two weeks. • State government central Borno: Responsible for supply provision in FHFs and OTP sites in some LGAs in central Borno (MMC, Jere, Konduga and Mafa) due to state government access in these areas. e State Nutrition Officer (SNO) compiles RUTF and MNP requirements from each LGA, which are reviewed by UNICEF and then supplied by UNICEF Maiduguri logistics team to the government central medical store (CMS), from where they are sent to FHFs bi-weekly and outreach OTP

and who are key players in the prevention andtreatment of undernutrition. Working throughMSGs also strengthens social networks as well asenhancing mothers’ understanding of malnutrition.

Lessons learned andconclusionse programme demonstrates that it is possibleto increase access to treatment of SAM andother nutritional interventions in emergency-affected areas with strong government leadershipand UNICEF playing a supportive role in meetingthe immediate needs of SAM children whilststrengthening the government health system.Other key elements included the formationand utilisation of existing community platforms(identification of CNMs and MSGs), which wasvital for the timely detection and treatment ofSAM children and to improve programme cov-erage. Re-engaging displaced health workershas contributed to the observed achievement inscaling up the response programme and improvedquality of SAM treatment services and has pro-vided state governments with a cohort of qualifiedpersonnel who can be integrated into the healthsystem to ensure programme sustainabilityduring the transition period.

e effective coordination of partners anddetailed gap analysis by the sector helped toidentify gap areas for scale-up. is processplayed a vital role in the rapid expansion of theprogramme. Subsequent periodic data analysisand feedback helped to identify and addressongoing gaps and issues in a timely way. Acontext-specific service delivery approach andsupply chain management helped to addressstock-out issues and create effective programing.e use of WhatsApp allowed for timely re-porting and communication, even in areas withpoor network coverage.

Investing in strengthening on-the-job training,supportive supervision and feedback has thepotential to improve the quality of treatmentservices. Although the programme overall hasmet the minimum SPHERE standard (cure rate87.2%; defaulter rate 9.6%; death rate 0.6%; andnon-responder rate 2.6%), maintaining thequality of the programme across all sites whilescaling up remains a problem. Disaggregateddata reveal variations in quality across thedifferent geographic areas.

For more information, please contact SanjayKumar Das at [email protected]

sites weekly. Transport costs are covered by UNICEF. • International NGOs in Borno: RUTF, R75, F1003, MNP and ReSoMal4 requirements (and quarterly forecasted requirements) are sent monthly to UNICEF Maiduguri office by INGOs with programme cooperation agreements (PCAs) with UNICEF. Once requests are reviewed and released, INGO partners collect supplies from UNICEF Maiduguri warehouse and deliver to the OTP sites they support. • Yobe: Yobe state primary healthcare management board (YPHCMB) forecasts quarterly supply requirements and UNICEF delivers to CMS. INGOs submit requests for RUTF and MNP with their monthly nutrition programme report to the state nutrition officer, who releases supplies from CMS for delivery by INGOs to SDUs. UNICEF and the SNO visit sites periodically for monitoring. For UNICEF-supported sites, the LAG nutrition focal person col lates monthly SDU requirements and shares with the SNO, who arranges delivery in col laboration with UNICEF.

The humanitarian-development nexusUNICEF’s role in scale-up of services has focusedon supporting state government to meet the im-mediate needs of severely malnourished childrenwhile strengthening the government healthsystem. Strong advocacy is carried out with thegovernment to ensure that the provision of nu-trition services to the state population continuesto be the responsibility of state government. estrong leadership shown by the government andcontinuous support to governance systems ensuresthat services are sustainable in the long term.

e support and funding of scale-up of nu-tritional services is mixed. In some areas it isfunded by donors, while in others UNICEF in-ternal funding was used for scale-up where thesituation was critical and humanitarian assistanceneeds very high. In Borno and Yobe, governmenthas not allocated any funds for RUTF as mostof their funding is spent on security and safetyarrangements; in other states, government hasdirectly procured RUTF supplies. Financing forscale-up has been effected through state gov-ernment by recruitment of additional staff andcapacity building; their payment and opera-tionalisation has been undertaken by state gov-ernment with guidance and support of UNICEF.ere is ongoing advocacy to allocate 1% of theconsolidated revenue fund (CRF) for basichealthcare provision fund (BHCPF) as part ofAppropriation Act, 2018; nutrition is part ofthe BHCPF. Some states have already allocatedfunds, but in Borno and Yobe advocacy hasstarted now through the state committee onfood and nutrition (SCFN) in transitioning froman emergency to a development approach.

Community systems are also core to the nexusas the humanitarian response works throughCNMs, who are selected from their communities

3 F-100 and F-75 (also known as Formula 100 and Formula 75)are therapeutic milk products designed to treat severe malnutrition.

4 ReSoMal is used for the preparation of an oral rehydration solution used exclusively to treat severe acute malnutrition. It must be used under medical supervision in therapeutic feeding centres.

Beneficiaries reachedwith IYCF support andMNP supplementationin 2016 and 2017

Figure 4

700,000600,000500,000400,000300,000

200,000100,000

0

Year

146,011

2016 2017

Number of children/caregivers reached

137,962

593,123

507,168

Number of caregivers of children 0-23 months reached with IYCF messages

Number of children 6-23 months reached with MNP

Das, S.K., Meleh, S., Chiroma, U., Assaye, B andGallagher, M.L. (2019) Nutritional response in north-eastern Nigeria: Approaches to increase serviceavailability in Borno and Yobe States. Field Exchangeissue 59, January 2019. www.ennonline.net/fex

47


Evaluation ..........................................................

Impact evaluation of WASH in nutritionintervention onmorbidity andacute malnutritionin NigerBy SaïdouTamboura, Dr Moussa IssaLende and Lucia Pantella

Tamboura Saïdou has worked in nutrition andhealth for 14 years. He has worked in nutritionemergency responses in several countries andhas been the Nutrition Coordinator and FieldManager for COOPI (CooperazioneInternazionale) Niger since 2016.

Dr Moussa IssaLende is a consultant forCOOPI. He has over seven years’ experiencein planning, implementation, management,monitoring and evaluation of policies andprogrammes related to cash, food security,health and nutrition in situations ofemergency and development in Niger and

other countries.

Lucia Pantella is the Nutrition Adviser forCOOPI working mainly in West Africa andSahel Region. She is a registered nurse andholds a master’s degree in humanitarianprogramme management. She previouslyworked with Save the Children and Intersosin several humanitarian crises.

The authors acknowledge European Civil Protection andHumanitarian Operations (ECHO) for financing the projectdescribed here. COOPI has been receiving funding fromDirectorate General (DG) ECHO for the implementation ofnutrition projects in Niger since 2012. The authors would alsolike to acknowledge UNICEF for supporting WaSH in Nutritionin COOPI Niger.

Location: NigerWhat we know: There is increasing drive to integrate nutrition andwater, sanitation and hygiene (WASH) interventions, including thosetargeted at malnourished children.

What this article adds: An evaluation was carried out of a year-longintegrated water, sanitation and hygiene (WASH) in nutrition (WiN)programme in Tillabery, Niger in 2017 by CooperazioneInternazionale (COOPI). WASH facilities were improved in 11outpatient therapeutic centres (OTPs) and one stabilisation centreproviding acute malnutrition management services; 1470 WiN kitswere provided to households of all children with severe acutemalnutrition (SAM) to support personal hygiene, water treatmentand water storage until 30 days after discharge; and WASH educationsessions were provided in the same villages. Results showed nodifference in OTP performance indicators between intervention andcontrol areas with the exception of the non-response rate inintervention sites (4 to 2%). Improvements in hygiene practices andsignificantly lower diarrhea co-morbidity were found in interventionhouseholds receiving WiN kits and community awareness sessionscompared to controls. Co-morbidity (diarrhoea) was prevalent in allsites. In this context, scale up of WiN activities, strengtheneddiarrhoea treatment and village level water treatment, are warranted.

BackgroundMalnutrition is a chronic public health problem throughout Niger. Results of a2016 SMART survey show a global acute malnutrition (GAM) rate of 9.3% and asevere acute malnutrition (SAM) rate of 1.9% for the Tillabery region. For a pop-ulation of 322,381 people, Tillabery has 32 health centres with outpatient thera-peutic programmes (OTPs) with one primary healthcare facility per 10,074 people(compared to Sphere standards recommendations of 1/10,000), one District Hos-pital for the whole population (Sphere standards recommend 1/250,000), a doctorratio of 1/40,297 (compared to World Health Organization (WHO) recommen-dations of 1/10,000), a nurse ratio of 1/8,059 (WHO recommends 1/5,000) and amidwife ratio of 1/22,605 (WHO recommends 1/5,000).

Cooperazione Internazionale (COOPI) has been working in Niger since 2012to respond to the humanitarian needs of people affected by crisis, conflict and dis-placement. COOPI has operated a programme to improve the management ofSAM in Tillabery since 2012 in partnership with European Civil Protection andHumanitarian Aid Operations (ECHO), World Food Programme (WFP) and theFood and Agriculture Organization (FAO). Technical support has been providedto improve OTP and stabilisation centre (SC) performance indicators and qualityof care in OTPs. Since 2015 the programme has focused particularly on commu-nity mobilisation to raise awareness of malnutrition within the community andstrengthen prevention and early detection, with the support of families, in partic-ular mothers. GAM rates have not significantly reduced in Tillabery since thistime, however, achievements worthy of note include good and stable OTP and SCperformance indicators; improved quality of care in OTPs; increased communityparticipation; involvement of local communities in growth monitoring activitiesand heightened awareness of community leaders on the extent of the problem ofmalnutrition and their increased commitment to act.

WASH in Nutrition (WiN) intervention OverviewSince 2017 COOPI has integrated WASH into its nutrition programme through theWASH in Nut (WiN) strategy1. In 2016 COOPI conducted an assessment in 32 healthcentres, which showed that none of the surveyed facilities respected the 12 standards

Mother feeding her child withsoup after preparing the meal in

a group meeting in Thuong Xuan,Thanh Hoa, Vietnam, 2014

Beneficiaries of the WASHin Nutrition kit, Kandadji,Tillabéry, Niger, 2017

Tam

bour

a S

aïdo

u

1 Regional, Central and West Africa Group WiN Strategy. www.susana.org/_resources/documents/default/ 3-2297-7-1438183632.pdf

48

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Evaluation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


set by the WiN strategy. Hand washing devices(water point, soap, disinfectant) were absent inthe majority of the surveyed facilities. As a result,COOPI decided to strengthen the WASH com-ponent both at facility and community level. Atfacility level, WASH facilities were improved. Atcommunity level, a WiN kit was provided to allSAM children admitted to the community basedmanagement of acute malnutrition (CMAM)programme in 11 health centres in areas facing

recurrent drinking water supply problems duringthe rainy season, as well as to all children treatedas inpatients at the District SC.

Targeted health areas (Assani, BosseyBan-gou, Guru, Kandaji, Kofonou, Kurani, Mari,Sawani, Sona, eim and Wissili) were selectedin collaboration with the Regional Directorateof Hydraulic and Sanitation of Tillabery. ehouseholds of targeted children received theWiNkit to support personal hygiene, watertreatment and water storage for the entire familyfor the whole duration of SAM treatment plus30 days aer discharge. Depending on the typeof water consumed, two kits were distributed:SAM children living in communities with accessto river water (cloudy surface water) received akit consisting of ‘pur’ (a powdered mixture totreat contaminated water), a cup, a permeablefabric and soap; SAM children living in commu-nities with access to water from boreholes or

wells received ‘aquatab’, jerry cans and soap (Fig-ure 1). e quantity of pur, aquatab and soapdistributed was calculated according to numberof household members.

Full kits were distributed to the caregiver onadmission and at each weekly follow-up post-discharge the family received soap and eitherpur or aquatab. On the final follow-up the fam-ily received a one-month supply of soap and ei-ther pur or aquatab. Consumables (soap, purand aquatab) were provided by UNICEF andequipment (bucket, cup and jerry can) wereprovided by COOPI. In 2018, following a post-distribution survey (results below), COOPI alsoadded a kettle to the kits.

Each village from which a child was admittedto the programme received awareness raisingsessions on hygiene and sanitation practices,starting the week following the child’s admission.Sessions were run by community mobilisers andaimed to raise awareness among householdheads and the wider community on the impor-tance of hygiene and water quality for the treat-ment of SAM, the prevention of certain diseases(including diarrhoea) and to discourage the saleof kit items on local markets. Mobile teams alsoused village visits to determine the chlorine loadand drinking water pH for the household of eachbeneficiary. In total 70 households were sensi-tised once a week for at least four weeks in 288villages. MUAC-trained mothers (6,000 in total)were also used to reinforce positive WASH mes-sages within their villages.

Post distribution monitoringA post-distribution monitoring survey (PDM)was conducted from 8 to 15 February 2018 on arepresentative sample of beneficiary householdsrandomly selected to evaluate the impact of sen-sitisation activities on hygiene practices and useof WiN kits. Indicators assessed included carry-ing out hand washing in the critical moments ofthe day, the correct treatment and storage ofwater. e PDM also included a beneficiary sat-isfaction survey concerning the content of thekit and hygiene education sessions and soughtto identify mechanisms for sustainability of theproject using local means.

CMAM Indicators Interventions Controls P. value

Effectives(n)

Percentage(%)

Effectives(n)

Percentage(%)

2015 indicators

Cure rate 1132 96 1158 97 > 0,05

Death rate 2 0 1 0 > 0,05

Default rate 22 2 15 1 > 0,05

Non-responding rate 26 2 19 2 > 0,05

Referral rate 398 25 269 18 < 0.001

2016 indicators

Cure rate 765 92 1189 97 > 0,05

Death rate 3 0 3 0 > 0,05

Default rate 23 3 29 2 > 0,05


Referral rate 314 27 289 19 < 0.001

2017 indicators

Cure rate 726 91 795 92 > 0,05

Death rate 5 1 3 0 > 0,05

Default rate 51 6 63 7 > 0,05


Referral rate 342 30 187 18 < 0.001

Table 2 Indicators of malnutrition management in health centres in 2015-2017

Number of households Residual chlorine*

181 0.5 mg/l

5 1 mg /l

3 1.5 mg /l

2 2 mg/l

Table 1 Residual chlorine present inwater in beneficiary households

* World Health Organization standards specify that residualchlorine should be between 0,5 and 1 mg/l

Composition of the Wash in Nutrition (WiN) kits Figure 1

Number ofhousehold AQUATAB SOAP

2 14 13 21 24 28 35 35 36 42 47 49 4

Number ofhousehold AQUATAB SOAP

2 56 53 84 84 112 105 140 136 168 157 196 17

1 JERRYCANper household

Drinking water source(well, borehold)

Number ofhousehold PUR SOAP

2 14 13 21 24 28 35 35 36 42 47 49 4

Number ofhousehold PUR SOAP

2 56 53 84 84 112 105 140 136 168 157 196 17

Turbid water source(river, pond, lake)

1 Bucket, 1 Cup and1 fabric for filter

Admission During treatment Discharge

Sensitations of beneficiaries in use of aquatab kit, healthfacility of Bonseeybangou, Tillabéry, Niger, 2018

Tam

bour

a S

aïdo

u

49

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Evaluation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


e results of the PDM showed that, duringthe first year of implementation, 1470 WiN kitswere distributed through the 11 OTPs and oneSC against the1833 expected (80.2%). By theend of the first year all health centres were meet-ing WASH standards due to the installation of28 handwashing points and routine treatmentof water used for the ‘appetite’ test. e satisfac-tion survey showed that all beneficiaries sur-veyed (n=1098) were satisfied with their kits andthat households had used them. Householdhandwashing practices also improved due to theavailability of soap.

e quality of drinking water within house-holds was satisfactory, with chlorine levels in therecommended range in 95% of beneficiaryhouseholds (Table 1) and pH levels rangingfrom 6.8 to 7.2 (within the WHO standard rangeof 6.5 and 9) for 92% of households.

Evaluation of the WiNinterventionImpact of the WiN programme on SAM treat-ment was evaluated aer one year in 2017 interms of the performance criteria of targetedOTPs (average length of stay, relapse rate anddefaulting rate); morbidity and prevalence of

pling. e selection of health centres fromwhich mothers of SAM children were selectedin the intervention and control areas was deter-mined by a random draw. Data processing andanalysis was carried out using SPSS 21, Exceland Word soware.

ResultsSAM management performanceindicators e results of the impact evaluation showed thatall performance indicators exceeded the mini-mum standards set by the national protocolboth in intervention and control areas. isgood performance could be explained by thetechnical support provided by COOPI since2012 to build the capacity of health workers inOTPs across control and intervention areas. Re-sults show no difference in performance indica-tors between intervention and control areaswith the exception of the non-response rate,which reduced from 4% (n=37) to 2% (n=17)from 2016 to 2017 in intervention health centres(Table 2).

In addition, there was a significant differencein referral rates between intervention and con-trol areas. Since 2015 referral activities havebeen strengthened in the intervention areasthrough the scaling up of the MUAC motherstrategy, through which volunteer mothers aretrained in the detection of SAM children andtheir referral to the health centers. Since 2015more than 6,000 mothers in 400 villages (60%of whom are located in the intervention areas)have been trained and equipped with a MUACbracelet; this has led a positive impact in per-formance indicators, with increased programmecoverage (from 52.3% in 2015 to 59.9% of 2017),especially in the interventions areas.

Morbidities in SAM children e evaluation of morbidities among SAM chil-dren in both intervention and control house-holds showed that 28% suffered only from SAM,while 78% of children suffered from SAM asso-ciated with other diseases; 22% of acutely mal-nourished children haddiarrhoea. Diarrhoeaassociated with SAM was more prevalent amongchildren in the control health centres (82%)than in intervention health centres (66%)(p<0.02) (Figure 2). Diarrhoea affected girlsmore than boys in control households (58%)and intervention households (54%).

At the household level 59% of children incontrol households had diarrhoea during thetwo weeks preceding the survey, comparedto38% of children in intervention households,revealing a statistically significant difference (P<0.001) (Table 3). is result could be explainedby the reaching of minimum WASH standardsin intervention households due to the presenceof WiN kits and by the effectiveness of commu-nity education and participation of mothers inproject implementation. e main treatment fordiarrhoea in both intervention and controlhouseholds is based on the combination of oralrehydration solution (ORS) and zinc, dissemi-nated by health centres.

Diseases associated with SAM in intervention and control households Figure 2100%

80%

60%

40%

20%

0%

60%

Interventions Controls

11%23%

0%

82%

8% 7% 3%

SAM+Diarrhoea SAM+MalariaSAM+ARI (Acute Respiratory Infections) Others (anemia & Iymphadenitis)

Assessment ofdiarrhoea inhouseholds

Interventions Controls P. value

Effectives(n)

Percentage(%)

Effectives(n)

Percentage(%)

Diarrhoea 36 38 46 59 <0.001

No diarrhoea 55 57 30 38 <0.001

Do not know 5 5 2 3 > 0,05

Table 3 Assessment of diarrhoea in intervention and control households

Water Interventions Controls P. value

Effectives(n)

Percentage(%)

Effectives(n)

Percentage(%)

Unfit forhumanconsumption

No treatment 11 11 56 72 <0.001

Filtering cloth 7 7 5 6 <0.001

Let rest 2 2 1 1 >0.05

Drinkingwater

Filtering cloth &aquatab/pur

11 11 1 1 <0.001

Aquatab 33 34 9 12 <0.001

Pur 34 35 6 8 <0.001

Table 4 Drinking water

diarrheal diseases in SAM children; and hy-giene and sanitation knowledge, practices andattitudes in health centres, beneficiary house-holds and non-beneficiary households. epurpose of the evaluation was to contributeknowledge to improve implementation of WiNactivities in the health centres and at commu-nity level and to learn lessons and make recom-mendations for future COOPI nutrition-sensitive interventions.

e sample size was calculated using thesoware ENA 2011 for SMART using SAMprevalence in Tillabery region of 1.9%, precisionof 5%, a cluster effect of 1.5, the proportion ofchildren under five years old as 18% (by Re-gional Directorate of Public Health (RDPH ofNiger), average household size of five and non-respondent rate of 3%. e total size of the sam-ple to be surveyed was 88 mothers of SAMchildren out of 180 households drawn by ENAsoware. As this is a case control study in 22health areas, the number of children and house-holds to be surveyed was multiplied by two to176 children in360 households (half in 11 inter-vention areas and half in 11 control areas). eselection of health centres and health workersto be surveyed was based on exhaustive sam-

50


. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Evaluation

WASH practices e survey revealed that 80% of households inintervention areas had treated drinking waterwith aquatab or pure, compared to 21% ofhouseholds in control areas (Table 4). Ninety-seven percent of women and men in householdsin the intervention area compared to 96% ofwomen and men in control households used la-trines (no significant difference).

ere are 12 UNICEF WASH standards appli-cable to health centres, each of which was evalu-ated. All intervention health centres achieved theminimum standards set by the UNICEF WiNstrategy; 100% of patients and staff had access toa clean water point and handwashing device andeach health centre had a waste management sys-tem and an improved latrine.

Regarding the main sources of drinkingwater, intervention households mainly used well(35%), rivers (32%) and drilling (21%), and only12% used water from taps. In the control house-holds, the main source of water was taps (51%),followed by rivers (19%), drilling (18%) andwells (12%). is result confirms that the inter-ventions areas were selected among those com-munities which have faced more recurrentdrinking water supply problems, especially dur-ing the rainy season.

With respect to handwashing, 36% of inter-vention households used a handwashing devicewith soap or disinfectant compared to 12% ofcontrol households. Results showed that94% ofintervention households had knowledge aboutthe key moments of hand washing compared to82% of control households. e majority of inter-vention households washed hands at the idealtime (before and aer meals) (60%), compared to44% in control households. Reasons given forwashing hands in 31% of intervention house-holds were the prevention of faecal peril, diarrheaand parasites and to clean hands. e reasons

given for hand washing in 32% of control house-holds concerned the prevention of parasites.

Ninety-two percent of mothers in interven-tion households versus 78% mothers in controlused soap for hand washing. In control areas incases where soap was not used, ash was com-monly used to wash hands instead. e qualityof hand washing differed significantly betweenintervention and control households (p <0.05).is could be the result of availability of soapprovided throughout the project as well asawareness sessions on good hygiene practices athealth centres and in the community.

Concerning household sanitation, latrineswere used more in control households (71%)compared to intervention households (37%).Sixty-one per cent of mothers in interventionhouseholds compared to 55% of mothers incontrol used the latrines twice daily (no signifi-cant difference). Eighty-six per cent of interven-tion households kept the latrine clean comparedto 72% of control households. e good avail-ability of latrines in control areas was an initialadvantage for many households; however resultsshow that this was not synonymous with goodsanitation practices in the same households.is reflects the need to accompany access to la-trines with hygiene and sanitation education toimprove their use.

Study limitationsCMAM performance indicators (such as defaultrates and non-response rates) may have been in-fluenced not only by the WiN activities by otherexternal factors. For example, the referral ratein the intervention zones may have been af-fected by other community activities, such as themothers MUAC approach which has been par-ticularly promoted in these zones.

e PDM was conducted by COOPI projectstaff, which may have induced a bias in benefi-

ciary responses. Conversely the analysis of thedata was conducted by an external consultant.

e results of this study reflect the impact ofWiN activities aer one year of implementation.e WiN strategy should be systematically inte-grated into nutrition programmes in order tohave long lasting results in intervention com-munities.

Conclusion andrecommendationse evaluation of the WiN project indicates nodifference in performance indicators betweenintervention health centres with improvedWASH facilities and control health centres, butsignificant, positive impact on the non-responserate of acutely malnourished children in inter-vention compared to controls. Results at house-hold level show significant improvements inhygiene practices and significantly lower diar-rhoea co-morbidity in households receivingWiN kits and community education comparedto controls. is suggests that in order to reducemorbidity and mortality rates related to diar-rhoea, scale up of WiN activities is warrantedboth at community and household levels. Dis-tribution of the WiN kits should be accompa-nied by behaviour change communicationactivities, particularly around prevention of dis-ease through improved hygiene, treatment ofwater before consumption and use of householdlatrines.

Future WiN strategies should include thestrengthening of health worker skills to treat di-arrhoea and to promote and disseminate WiNawareness messages, for example during vacci-nation campaigns, to help its prevention. Fur-ther collaboration between governmentministries relating to nutrition and WASH at alllevels is also needed in order to carry out jointinterventions, such as the creation of watertreatment sites in villages with high prevalenceof SAM. All activities should be supported byeffective monitoring mechanisms in order tounderstand the effectiveness of interventionsand guide future programming.

Hand wash Interventions Controls P. value

Effectives(n)

Percentage(%)

Effectives(n)

Percentage(%)

With soap 90 92 61 78 > 0,05

Without soap 7 7 12 15 > 0,05

Other (ash) 1 1 5 6 > 0,05

Table 5 Household handwashing

Use of latrines Figure 3120

100

80

60

40

20

0

3


97

Only women Men & women

4

96

Source of drinking waterFigure 450%

40%

30%

20%

10%

0%

32%

Rivers Drilling Well Taps


19% 21% 18%

35%

12% 12%

51%

Tamboura, S., IssaLende, M. and Pantella, L. (2019)Impact evaluation of WASH in nutrition interventionon morbidity and acute malnutrition in Niger. FieldExchange issue 59, January 2019.www.ennonline.net/fex

51


Nutrition-sensitiveoutcomes of apermacultureproject in Nepal

By Anne-Marie Mayer

Anne Marie Mayer is a freelancenutritionist specialising in the linksbetween agriculture and nutrition.She holds a PhD in InternationalNutrition with Soil Science andEpidemiology from Cornell University.

Her current interests include multi-sector evaluations,nutritional quality of foods and links between nutritionand sustainable agriculture.

This study was commissioned by Himalayan PermacultureCentre. The author would like to thank Bhuwan Khadga,Director of the HPC, and the HPC team in Nepal for theirassistance throughout the implementation of the study.Thanks also go to Renu Shakya, research assistant andtranslator, and Chris Evans, HPC Technical Advisor, whosupported the planning and follow-up to the study.

Location: NepalWhat we know: Guidance is available for the design of nutrition-sensitiveagriculture projects but few have demonstrated nutrition impact.

What this article adds: The Himalayan Permaculture Centre (HPC) is agrassroots project that builds the capacity of farmers to implement permaculturefarming techniques with integrated food security, health, education andlivelihoods activities. Explicit nutrition objectives are not included. An evaluationof the HPC was carried out in April 2018 through focus group discussions andindividual interviews with HPC staff and beneficiaries. Respondents reporteduptake of a range of improved farming techniques and livelihoods activities thathave diversified agricultural systems, led to improved yields, increased householdincome, improved water, sanitation and hygiene practices, and reduced women’sworkload (leaving more time for household hygiene, child care and appropriateinfant and young child feeding). Households reported improved dietary diversity.There is strong potential for the HPC to impact on nutrition positively. Lessonslearnt from other nutrition-sensitive programmes can be applied to HPC to helpincrease nutrition impact, such as integrating behaviour change communication.Women’s participation in the programme, access to markets and nutrition andhealth education are key areas now being addressed.

Backgrounde nutrition situation is improving in Nepalbut remains a serious public health problem.e latest data shows that 36% of children underfive years old are chronically malnourished, asubstantial reduction from 57% in 1996 (Ministryof Health Nepal and New Era 2017). e Hi-malayan Permaculture Centre (HPC) is a grass-roots, non-governmental organisation set up in2010 by trained farmers from Surkhet districtin mid-west Nepal that aims to build sustainableagriculture and resilient domestic food andenergy security through rural development pro-grammes. It builds on experience gained overthe past 30 years in training and implementationof permaculture design and practice in remoteand challenging environments in Nepal.

e HPC works with 850 households in tworural districts in the west of Nepal, Humla andSurkhet, with an annual budget of approximately£84 (12,800 Nepalese rupees) per household.e HPC is nearing the end of its ‘BuildingLivelihoods for Household and Community Re-

silience’ (BLHCR) project (2015-2018), throughwhich work is organised around five areas: foodsecurity, health, education, livelihoods and ca-pacity-building. A range of over 45 techniquesis used to increase and diversify farm productivity;reduce cost of domestic activities (in terms oftime, labour and money); improve health throughbetter nutrition, hygiene and reduced work load;use and recycle local resources; and protect theenvironment. e four main components of theproject are: i) demonstration of techniques, ii)training of communities, iii) provision of resources,and iv) research on techniques and approaches.Activities are supported in the communitiesthrough capacity-building by HPC techniciansand ‘Barefoot Permaculture Consultants’ (BC)

– farmers who have been selected by communitiesfor their leadership, technical knowledge andsuccessful application of techniques.

e HPC is a multi-sector programme withactivities and planned outcomes that cross thedisciplines of agriculture, livelihoods, healthand the environment, among others. While im-provements in nutrition are not mentioned ex-plicitly in the objectives, many elements are inplace to support adequate food and nutritionsecurity (FNS) and the HPC offers a good casestudy to show the potential for permaculture tosupport FNS.

e case study here is not an evaluation ofthe BLHCR project, but a longer-term assessmentof the ongoing work of HPC in Surkhet, whereHPC works with 265 households. e study wascommissioned by HPC as impact assessmentshave not been carried out to date and there is in-terest in improving activities for nutrition impact.e study aims to assess the potential of theHPC to improve nutrition and to identify longer-term research and evaluation plans for the HPCand other similar multi-sector programmes.

MethodsAn initial literature review and guidance frompeers supported development of participatoryresearch tools. Question guides were developedfor focus group discussions (FGDs) and key in-formant interviews. Field research was conductedover a period of three weeks in April 2018 inSurkhet. FGDs were carried out with femalemembers of farmers’ groups (n=7), male membersof groups (n=6) and an adult literacy group(n=1). Individual interviews were carried outwith two HPC staff members, six BCs, one teacher,two women farmers and two local health workers.Findings were presented at a workshop of HPC

Permaculture may be defined as “The conscious design and maintenance of agriculturallyproductive systems which have the diversity, stability and resilience of natural ecosystems.It is the harmonious integration of the landscape with people, providing their food, energy,shelter and other material and non-material needs in a sustainable way.” Geoff Lawton,Permaculture Consultant, Designer and Teacher

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Evaluation

Focus group discussion withwomen during the evaluation,

Surkhet, Nepal, 2018

Ann

e-M

arie

May

er

52

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Evaluationtechnicians and BCs and a theory of change wasdeveloped. A final workshop with HPC was usedto discuss recommendations and next steps.

Findings Coverage and participation in training Training is open to all members of farmers’groups in the HPC villages. Participation intraining is skewed towards men, with 39% ofeligible men and 28% of eligible women attendingtechnical training between November 2017 andApril 2018. ere is a perception that healthand nutrition training is geared towards womenand livelihood and agriculture training is gearedtowards men. e majority of BCs are male, de-spite efforts to recruit and maintain women.

Adoption of new techniques Respondents reported many new techniques thatthey had implemented as a result of the HPC,including fruit nursery, beekeeping, cash crops,composting, tree planting, covering food, housecleaning and many others. New crops mentionedin particular were plums, pears, peaches, kiwis,almonds, oranges, mulberries, figs, bananas, fod-der trees, green leafy vegetables, garlic, onions,tomatoes and cash crops. HPC six-monthlyreports (available from www.himalayanperma-culture.com/reports/) give more detail on thetechniques adopted in each six-monthly period.

Building of resilience and productiveagriculture systems Challenges to farming highlighted were drought,pests, lack of market access and reduced rurallabour due to migration. e HPC has metthese challenges by introducing diverse, perennialagriculture systems with a rich mix of speciesin agro-forestry systems using a range of prop-agation techniques. Improved livestock breedshave been introduced, while stocks have beenreduced to allow natural forest regeneration ofhillsides and reduce labour demands. Compostingand water systems are managed to support in-creased production. A system of rice intensifi-cation (SRI) has been introduced that requiresless labour, seeds and weeding; those using SRIreported increases in yields (although there isno data available to verify this). Increased yieldswere also reported for corn wheat and buckwheatdue to succession planting, water provision andnew techniques such as intercropping. A widerange of fruits and vegetables has also beenadded that are faster growing with greater yields.

Improved livelihoods Livelihood improvement is a key aim of the HPCin order to stem outward migration and addresspoverty. HPC supports livelihoods through mi-cro-credit schemes, beekeeping, vegetable seedproduction, cotton growing and processing, de-veloping mills, growing and processing herbsand cash crops (with weaving and processing offruit and vegetables under development). Accessto markets, particularly for perishable foods, wasa limitation cited by communities in this research;activities to address this are planned by the HPC.

Income expenditure Respondents reported increased household incomedue to HPC livelihood activities, such as sale of

cash crops, honey, fruit and vegetables, fodderplants and seedlings, and money saved; for ex-ample, due to less need to purchase goods nowproduced at home, such as fruit and vegetablesand sugar (due to honey production). Respondentsreported spending additional income on schoolfees, soap, clothes, medicine, stationary, festivalsand foods (salt, sugar, pulses, oil, rice and noodles),some of which may have direct and indirect,short and long-term nutrition-related benefits(soap on improved hygiene, for example).

DietsMost FGD participants reported increased con-sumption of fruits, vegetables and pulses sincejoining the HPC. For some households con-sumption of animal milk has fallen due to labourshortage for livestock rearing; however, the mit-igation measures put in place (such as foddersupply, improved breeds and improved watersupply) appear to have minimised this effect.Persistent challenges for nutrition include lossof nutrients from over-cooking vegetables andover-processing rice, which are ingrained habits,and the tendency to use additional householdincome to purchase processed foods to replacetraditional, healthier foods. Cooking demon-strations, food processing, food combining anduse of sarbotum pito (a locally developed com-plementary food) are strategies that HPC hasused to address these problems, although wide-spread use of bought processed foods was re-ported by respondants and HPC staff. ere islittle data on local malnutrition prevalence tocompare with national data; however, respondentsremarked that health and nutrition had improved.Data is needed to verify these reports.

Gender empowermentHPC activities appear to have been effective inreducing the time and drudgery of women’s work;for example, through the application of specialtechnologies such as energy-efficient stoves. Re-spondents also reported time saved through im-proved sustainability of natural systems (improvedwater systems, soil fertility and reforestation andgrowing fodder trees close to the homestead); animpact not widely recognised in the debates onwomen’s time and nutrition. Female respondentsreported having more time to spend caring forchildren, attending to household cleanliness andhygiene, preparing special meals for children,engaging in income-generating activities or socialenterprises and attending meetings/socialising.Women’s energy expenditure on heavy agriculturalwork and fetching water and fodder is alsoreduced, which impacts on energy requirements.

Nutrition and health promotionHPC nutrition and health-promotion activitiesinclude nutrition training, cooking demonstra-tions and health fairs. e farmers’ manual in-cludes a chapter on diet and nutrition, wherenutritional properties of foods are describedand guidance is given on food processing, pro-duction of sarbotum pito and care of the sick ormalnourished child. Hygiene practices are alsowell covered in the training curriculum. Short-comings are the absence of sufficient guidanceon breastfeeding practices and absence of tar-geting of nutrition and health training to women

during the first 1,000 days. Several women re-ported that they had received health and nutritiontraining on topics such as infant and youngchild feeding (IYCF) and news menus, althoughfor some the training was between two and fouryears ago. Women also appreciated trainingthey had received on plant propagation, fruittree planting, salt lick for livestock, energy-effi-cient stoves and household hygiene practices.

Maternal care during the first 1,000daysAdequate maternal and child care is critical tothe prevention of malnutrition; the global rec-ommendations are for pregnant women to con-sume a healthy diet with adequate intake of en-ergy, protein, vitamins and minerals to meetmaternal and foetal needs (WHO, 2016). HPCmay contribute to improved maternal care byincreasing the availability of nutritious foodsand reducing women’s workload. Responsesfrom some women indicate a positive impact;for example, “I consume fish, meat, milk andcurd during pregnancy. Aer giving birth I restfor four months at home. During that time Ican feed my child four or five times per day.When I have to return to the fields this isreduced”. However, respondents from anothervillage reveal no culture of reducing workloador special attention to a woman’s diet duringpregnancy. It is not clear why there are differencesbetween villages. A prevailing serious problemwith anaemia among women was reported,caused by excessive bleeding (due to inappropriateuse of contraceptives) and poor diet.

IYCF practices Older women reported that IYCF practices haveimproved greatly since the start of HPC due totime-saving activities and the greater availabilityof diverse foods, which are important for IYCF.However, the health professionals interviewedestimate that half of women do not exclusivelybreastfeed to six months and use breastmilk sub-stitutes (BMS) such as buffalo milk, sarbotumpito and family foods. FGDs confirm this andrevealed challenges to exclusive breastfeeding forthe first six months, including perceived lack ofmilk, lack of time, close birth-spacing and earlymarriage. e uptake of sarbotum pito appearsto be mixed, with some mothers relying solelyon family foods (rice, dhal and vegetables) evenwhen ingredients for and time to prepare sarbotumpito, a valuable complementary food, are available.

Improved water and sanitation andhygiene practicese HPC has introduced many hygiene-promo-tion activities and easy-to-implement technologiesto help improve food safety and hygiene practices.e provision of piped water to households andsanitation is a key benefit for health. Time savedby women due to HPC-introduced practiceshas resulted in more time for light domesticwork, with reported benefits for house cleanlinessand washing clothes. Sanitation has also beensupported by the HPC. Toilets (either compostingor traditional) are now widely available.

Informed by the findings, a theory of change(ToC) was developed for outcomes of the HPC

53

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Field Exchange issue 59, January 2019, www.ennonline.net/fex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Evaluationcombination of interventions (Figure 1). eToC shows clearly how the different HPC ap-proaches interact and converge to build potentialfor improved nutrition. Women’s time is savedthrough a combination of activities, includingtechnologies, environmental improvement andsupply of basic services such as water and animalfodder. Food systems are diversified throughagroforestry, natural resource management andfruit and vegetable gardens. Income can be usedto improve nutrition or for purchase of highlyprocessed foods that are not nutrition-enhancing.Nutrition and health education can be used tomodify this effect.

LimitationsOther projects were being undertaken in thesame locations, so not all reported benefits cannecessarily be attributed toá HPC. Baseline dataand a comparison survey would help to assessattribution. It is possible that respondents over-emphasised the benefits of the project; FGDswere limited to community members withoutHPC staff present to mitigate against this risk.

Recommendationse strength of the HPC is the integration of itsdifferent activities for multiple purposes. isis inherent in permaculture designs which workto address the multiple challenges and needs ofcommunities. e following specific recommen-dations for HPC were formulated on the basisof the review:1. Create a specific objective and budget-line for nutrition in future projects to help prioritise and manage nutrition-related activities. 2. Collaborate with a research organisation to implement systematic, regular data collection across sectors, including baseline data in new villages. 3. Collect data against nutrition-specific indicators; e.g. dietary diversity score and sample household weekly food consumption

record collected in different seasons. 4. Improve women’s participation and recruitment and retention of female technical staff and BCs by broadening selection criteria from farming techniques and providing child creche facility and children’s meals during training sessions. 5. Promote nutrition training that emphasises the importance of home processing and cooking to men as well as women and include HPC staff, technicians and BCs. Target special training on IYCF and maternal care to pregnant and lactating women. Training materials on IYCF practices need updating, particularly the section on breastfeeding.6. Facilitate a training of trainers (TOT) for technicians and BCs in health and nutrition and build the capacity of health workers to support training delivery. 7. Develop a marketplace for nutrient-dense foods identified as inadequate in the local diet. 8. Provide training and support on people- centred topics for BCs, technicians and community members, such as time management, personal empowerment, confidence-building, report writing, public speaking and planning.

e HPC made several changes in the sixmonths following the review as a result of theserecommendations. e selection criteria for tech-nicians and BCs has been broadened to includewomen with gender-related training (such astraining in women’s health and gender advocacy),and older women have been recruited with broadexperience and who are no longer responsiblefor young children. Creches and children’s mealshave been introduced during training sessionsto enable women to attend. Some men have nowreceived basic training in nutrition and healthtopics and plans are in place to offer IYCF andmaternal care practices training to pregnantwomen, increase training on home processing

and cooking, and offer ‘people and permaculture’skills training. A part-time nurse has been re-cruited to assist with health and nutrition trainingin Humla district and HPC has now developedand delivered a women’s health TOT for the firsttime for nine women and nine men in Humla.

A research project has also since been carriedout to test the nutritional quality of rice grownusing the system for rice intensification versustraditional rice (data have not been yet analysed).Indicators have been developed to monitor allaspects of HPC and a research concept has beenwritten; funding is currently being sought.

ConclusionsNutrition has not been strongly emphasised inthe HPC approach; nevertheless many of the ac-tivities have strong potential to impact on nutritionpositively. Some of the lessons learnt from othernutrition-sensitive programmes have been drawninto guidelines that can be applied to HPC tohelp increase nutrition impact. For example, itis widely acknowledged that agriculture pro-grammes are more likely to show positive impacton nutrition where behaviour change commu-nication for nutrition is included (FAO 2013).

Using the permaculture design, integratedsystems that cross traditionally separate sectorscan be used to build health and nutrition ofcommunities in a socially, economically andenvironmentally sustainable way. With the mul-tifaceted challenges faced by rural Nepalesecommunities, a truly multisectoral programmeapproach is essential and HPC offers a modelfor similar contexts.

HPC staff and BCs are already working asambassadors for permaculture. e work carriedout by BCs in the 2015 earthquake response(where farmers from the HPC went to helpearthquake victims) is evidence of the value ofHPC’s work. Aer the earthquake the BCs weredeployed to eastern districts of Nepal to helprebuild communities using permaculture designsand techniques. Further opportunities for thistype of outreach to expand the reach of perma-culture in Nepal would be valuable. e sus-tainability of HPC relates to low costs and thecapacity that has been built for control of theproject by local people.

For more information please contact Anne-MarieMayer at [email protected]

ReferencesFAO (2013). Synthesis of guiding principles on agricultureprogramming for nutrition.

HPC (2018). Himalayan Permaculture Centre 6 monthlyreport April 2018. http://himalayanpermaculture.com/.

Ministry of Health Nepal and New Era (2017). Demographicand Health Survey 2016.

WHO (2016) WHO recommendations on antenatal care fora positive pregnancy experience.www.who.int/nutrition/publications/guidelines/antenatalcare-pregnancy-positive-experience/en/

Nutrition andHealth

Education

Nutrient-retaining Food processing

Livelihoodactivities

Agro-forestrySRI

Use of Energy-Efficient

Stoves

Genderempowerment

Naturalresource

management

Vegetablegardens

Improved water and sanitation

systems

Hou

seho

ld a

sset

s an

d liv

elih

oods

Trai

ning

and

pro

ject

sup

port

Nutritionknowledge

Income

Nutritional quality of

food

DiverseFood

System

FemaleTime

resources

Non-foodexpenditure

Quality ofdiets

Caringcapacity &practices

Health knowledge

Health careexpenditure

Nutrientintake

Healthstatus

Femaleenergy

expenditure

Childnutrition

outcomes

Mother’snutrition

outcomes

Figure 1 Theory of change for nutrition-sensitive outcomes in HPC

Mayer, A. (2019) Nutrition-sensitive outcomes of apermaculture project in Nepal. Field Exchange issue59, January 2019. www.ennonline.net/fex

By Jahangir Hossain, Nazneen Rahman,Mohammad Hafijul Islam, Md Hasanuzzaman,Khrist Roy and Dlorah Jenkins

Jahangir Hossain is a medical doctor with a masters’degree in public health, currently serving as Directorof Health Programmes for CARE Bangladesh. With 25years’ experience in the health, nutrition andpopulation sectors, he has led numerous innovativeand sustainable interventions to improve health and

nutrition in Bangladesh.

Nazneen Rahman is a public health professional withover 12 years’ experience in reproductive health,maternal, neonatal and child health, and nutrition.She currently serves as a Team Leader for the Healthunit of CARE Bangladesh and is overseeing theCollective Impact for Nutrition initiative.

Mohammad Hafijul Islam has worked in thedevelopment field for 22 years, in non-formalprimary education, safe motherhood, HIV/aids,migration and nutrition. He currently works withCARE Bangladesh as a Senior Technical Coordinator,Advocacy and Capacity Building for the Collective

Impact for Nutrition initiative.

Md Hasanuzzaman has worked in development forover eight years, including working on a community-based management of acute malnutritionprogramme in response to the Rohingya crisis inCox’s Bazaar, Bangladesh. He currently serves CAREBangladesh as the Technical Manager, Partnership

and Impact for the Collective Impact for Nutrition initiative.

Khrist Roy has 26 years of programme design,monitoring and evaluation experience in the healthand development sectors, focusing on maternal,newborn, child and adolescent health, and nutrition.He is currently a Technical Advisor for CARE USA’sFood and Nutrition Security unit.

Dlorah Jenkins is a research and communicationsspecialist with over seven years’ experience helpingnon-governmental, public and private sectororganisations use data to better understand theirimpact and communicate results. She currentlyworks with the Nutrition team at CARE USA to

support research and dissemination.

The authors acknowledge CARE USA and SALL Family Foundation forfunding the project described in the article.

Location: Bangladesh, Sunamgonj district What we know: A multi-sector approach is required to addressthe multifaceted challenges of undernutrition; however, muchremains unknown around process and methodologies related tomulti-sector activities, as well as how such activities translateinto actions on the ground.

What this article adds: Through the Nutrition at the Centerproject (N@C), CARE Bangladesh tested a sub-district, multi-sector approach to address malnutrition. One key strategy wasto use existing structures and resources to establish a model ofcommunity-based, multi-sector coordination platforms tosupport nutrition-sensitive and nutrition-specific interventions.As such, N@C brought together actors from the public andprivate sectors, non-governmental organisations and localgovernment at multiple sub-district levels. Results from theendline survey indicate that this multi-sector approach waseffective in driving substantial improvements in nutritionoutcomes.

BackgroundIn Bangladesh 36% of children aged under five years old are stunted(NIPORT, 2015). e burden of stunting is greatest in the Sylhet region,where the prevalence of under-five stunting is 49.6% (NIPORT, 2015). eGovernment of Bangladesh (GoB) has committed to achieving the SustainableDevelopment Goal of ending all forms of malnutrition by 2030; this com-mitment is made explicit in the country’s National Plan of Action forNutrition (NPAN). Given this political will and the heavy burden of mal-nutrition in Bangladesh, CARE Bangladesh has set its sights on implementingsolid, evidence-based programming in the country to improve maternaland child nutrition. To this end, from 2013-2017, the Nutrition at theCenter (N@C) project was implemented. is is an innovative project witha two-fold strategy: (1) to integrate nutrition within the existing communityhealth system; and (2) to strengthen multi-sector coordination around nu-trition, with the overall aim of reducing stunting in children aged 0-23months in project areas by 9%.

e GoB and development partners working in the country havecommitted to tackling the issue of malnutrition and to doing so via strength-ened multi-sector coordination. A multi-sector approach to nutrition isdefined in this context by coordinated action among multiple related

A multi-sectorapproach to improvenutrition: Experiencesof the Nutrition at theCenter project,Bangladesh

Pete

r Cat

on

54


Field Article ......................................................

A woman who received ANC services and counselling support,Pangaon Village, Derai sub-district, Bangladesh, 2015

CARE

Ban

glad

esh

55

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Field Article

national government departments, along withlocal governments and non-governmental or-ganisations (NGOs), to address both direct andunderlying causes of malnutrition. Implemen-tation of a multi-sector nutrition programme iscontext-specific and dependent on what pro-grammes and resources are already in place.e following examples illustrate multi-sectorcoordination under N@C:• Maternal and infant and young child feeding nutrition (MIYCN) education was integrated into existing platforms, such as women’s empowerment groups;• Water, sanitation and hygiene (WASH) education was included as part of infant and young child feeding (IYCF) counselling;• e health department launched school handwashing campaigns to complement school-based iron and folic acid (IFA) distribution programmes; and• e Women’s Affairs department provided vouchers to allow poor pregnant and lactating women (PLW) to access health and nutrition services.

Taken altogether, a multi-sector approach allowsus to address malnutrition from multiple anglesand levels of causation. However, there is a lackof operational knowledge of how to facilitatemulti-sector coordination processes, particularlyat sub-district levels. rough N@C, CAREBangladesh therefore sought to facilitate andevaluate methodologies and processes to oper-ationalise multi-sector coordination from sub-district to village levels and integrate nutritioninto community health systems. is article de-scribes the strategies, results and lessons learnedfrom these efforts.

Nutrition at Center (N@C)projecte project was established in the Derai andBishwambarpur sub-districts of the Sunamgonjdistrict of Sylhet and ran from May 2013 to De-cember 2017, with funding from the Sall FamilyFoundation and CARE USA. Derai and Bish-wamparbur were selected as the project sitesdue to their high concentration of poor andmarginalised populations and their remote lo-cation, leading them to be two of the most un-

derserved and low performing sub-districts inBangladesh. e area is also vulnerable to regularflash flooding and the land is waterlogged forhalf the year, making it unsuitable for growingcrops. ere were also opportunities in theselocations to work with other livelihood andhealth initiatives and the GoB encouraged part-ners to work in such remote communities todevise sustainable solutions using existing re-sources. N@C was designed in consultation withthe GoB, civil society organisations and UnitedNations and other development partners workingin the country.

e project aimed to integrate nutrition intothe existing community health system in theprogramme areas in order to support effectiveMIYCN. To this end, CARE trained 244 (nearlyall) government frontline health workers (FLWs)in the project area and their first and secondline supervisors on optimal IYCF practices.Training on supportive supervision, mentoringand monitoring (SSMM) was also provided forgovernment first and second-line health andfamily planning supervisors to build their capacityto effectively mentor and monitor the perform-ance of FLWs. Supervisors were introduced tothe concept of mentoring (or coaching) and of-fered competency-based mentoring training andpost-training follow-up to help integrate theirnewly acquired skills into routine supervision.

Prior to the programme, there was no growthmonitoring and promotion (GMP) training incommunity clinics in the area, primarily due tologistical problems and a lack of skilled healthcareproviders. N@C therefore advocated with theInstitute of Public Health and Nutrition (IPHN)to ensure supplies to support GMP, and in ad-dition arranged a one-day refresher training onGMP for community health care providers(CHCPs) based at community clinics (the lowestprimary healthcare structure in Bangladesh).Breastfeeding corners were established at 42community clinics with the help of managersfrom the Ministry of Health and Family Welfare(MoHFW) to create space for CHCPs, familywelfare assistants and health assistants to counselmothers on breastfeeding techniques and MIYCNand to provide privacy for mothers.

e second strand of the programme aimedto facilitate sub-district multi-sector coordinationplatforms using existing committees within thegovernment system at community, union andsub-district levels (figure 1). CARE facilitated aparticipatory process to develop terms of reference(TOR) for each of the committees, built the ca-pacity of committee members through awarenessand training sessions, and provided on-the-jobsupport to make the committees functional andeffective. While committees improved theireffectiveness specifically around nutrition, theskills and principles gained through the trainingscan also be applied to other priority program-ming. Additionally, bi-directional (upward anddownward) linkages between the coordinationcommittees were established.

Village level: Community SupportGroups e Community Clinic (CC) is a government-mandated structure at the village level. EachCC is supported by three Community SupportGroups (CSGs) that are responsible for the pro-motion of services offered by the clinic to com-munities in its catchment area. Each CSG com-prises approximately 17 representatives per 2,000community members and includes members ofadolescent/youth groups, women’s groups, moth-er-in-law groups, men’s groups, clubs and agri-cultural groups and businesses. CSG membersidentify PLW and children under the age of twoin their communities and refer them for utilisationof relevant services, such as antenatal care(ANC), postnatal care (PNC), immunisationsand GMP. Members then follow up with referredfamilies through household visits to ensure com-pliance with the advice received at the clinicvisit.

N@C mobilised 126 CSGs to promote nutri-tion and multi-sector activities. Frontline healthand family planning workers held sessions totrain 2,142 CSG members to build their capacityin nutrition counselling, referrals, linking familieswith both nutrition-specific and nutrition-sen-sitive services as needed, and strengthening co-ordination and accountability among serviceproviders. Coverage of services and supplieshave improved in the area in recent years dueto increased political commitment; thus thecommunity clinics are well positioned to handlereferrals.

As part of the capacity-building process,N@C organised training for CSG members ondeveloping action plans, engaging communitygroups and structures and committing to nu-trition-sensitive and/or nutrition-specific activ-ities. On completion of the training CSGs metmonthly to discuss successes and challengesand make pragmatic nutrition action plans. eproject facilitator worked with GoB supervisorsand staff to provide continued support to theCSGs and facilitate social mapping and imple-mentation of the nutrition action plans.

Women were proactively encouraged to par-ticipate in coordinating committees of the CSGs.As a result, of the 219 CSG members who joined

Figure 1 Structure and coverage of nutrition coordinating committees and platforms

Upazila Parishad(~200,000 pop.)

Sub-districtUpazila Nutrition

CoordinationCommittee

Union DevelopmentCoordination

Committee

Union Parishad(~30,000 pop.)

Community Clinic(~6,000 pop.)

Union

Village Community SupportGroup (~2,000 pop.)

LevelAnchoring institute

and populationcoverage

Multi-sectornutritionplatforms

56

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Field Article

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

the Union Parishad standing committee, almosthalf were women. Furthermore, 22 of 24 NutritionFocal Points (those responsible for overseeingand coordinating a school’s health and nutritionactivities) are women.

Union level: Union DevelopmentCoordination Committees e Union Council (UC) is the elected body inrural areas, mainly responsible for leading andcoordinating rural development. e Union De-velopment Coordination Committee (UDCC)coordinates all development and service-relatedUnion activities and is comprised of electedchairman-members, extension workers of de-partment line ministries and NGO representa-tives. Rather than creating a new coordinationplatform at Union level to promote a nutritionagenda, N@C worked with 14 UDCCs, encour-aging them to add nutrition as a priority totheir standing agenda, increase budget allocationto nutrition activities, and develop multi-sectorannual nutrition action plans based on local re-sources and in line with the ministries’ operationalplans. Nutrition plans included both nutrition-sensitive and nutrition-specific activities coveringhealth and nutrition; education; WASH; agri-culture and livestock; women’s and children’saffairs; and social safety net activities. Activitiesincluded distribution of IFA supplements toPLW and adolescent girls; sanitary napkin dis-tribution for adolescent girls; tiffin box (lunchbox) distribution for school children; tubewelland sanitary latrine installations; and distributionof fertilizer, seeds and agricultural equipment.Of the 14 UDCCs, N@C supported budgetpreparation for six to ensure allocation for nu-trition-related activities.

rough the UDCCs, N@C facilitated evi-dence-based local planning, recording and re-porting within community health systems and

the development of effective coordination, bestuse of local resources, and performance moni-toring in terms of improved nutrition withinthe catchment populations. For example, mem-bers of the UDCC now work together to reviewdata on services provided by government healthservice providers and NGOs to avoid duplicationof efforts and to develop a shared report onhealth and nutrition for the Union.

Sub-district level: Upazila NutritionCoordination Committee Two Upazila Nutrition Coordination Committees(UNCCs) were established (one in each sub-dis-trict) under the leadership of the Upazila Parishad(sub-district) elected chairman. e UNCC isthe key vehicle at the sub-district level for thestrengthening of multi-sector engagement forimproved nutrition. In project areas, the UNCCssuccessfully engaged key actors from the health,family planning, agriculture, women’s affairs, live-stock, WASH and education sectors to improveservice delivery and provide monitoring supportfor nutrition-sensitive and nutrition-specific in-terventions. Broader civil society engagementwas ensured through the active participation ofmajor development agencies operating in the up-azila and, most importantly, the voice of thecommunity was heard through the active partic-ipation of community members. Additionally,with the support of the UNCCs, budget allocationsfor nutrition activities increased in the UnionParishad and departments related to livestock,education and agriculture. e two UNCCs es-tablished by N@C are still in operation and theplatform is included in the GoB NPAN-2, withplans for replication in all sub-districts nationwide.

Service coverage andbeneficiariesTable 1 shows the number of training sessions

and participants/beneficiaries for select N@Cactivities between 2013 and 2017.

Baseline and endline surveymethodologyIn March 2014 a baseline survey was administeredto assess nutritional status and influencingfactors in Derai and Bishwambarpur (interventionsub-districts) and Itna and Nikli (control sub-districts). e survey evaluated household so-cioeconomic status, food security, maternal andchild anthropometrics, and household WASHconditions. e results of the baseline surveywere used to establish benchmarks to measureperformance of the N@C; specifically in termsof maternal and child nutritional outcomes.

e survey was based on a representativesample of non-pregnant women aged 15-49years who had lived in the sub-district for atleast six months and who had a biological childage 0-35 months living with her. Sample sizewas calculated based on the sample required todetermine a 9% decrease in stunting (n=1,998women and their children aged 0-35 months).If more than one eligible child resided in aselected household, one child was randomly se-lected for anthropometric measurement. Womenand children with any known or suspectedchronic or congenital disease were not eligiblefor inclusion.

A two-stage probability proportional-to-sizeprocedure was employed with sub-district unitsas the primary sampling units (PSU). e sec-ondary sampling units (SSU) were the householdswithin the selected sub-districts with eligiblewomen and children.

On arrival in each sampled area a communitymeeting was held (prearranged by CAREBangladesh) to inform community members ofthe purpose of the data collection and encouragecooperation. Starting from a prominent pointat the centre of a union/village, data collectorsvisited each household contiguously, interviewingone eligible woman. Where multiple eligiblewomen were present, one of them was randomlyselected for interview. If there were no eligiblewomen in the household, data collectors pro-ceeded to the next household. is procedurecontinued until the required number of mothersand children were surveyed.

Height/length and weight of children age 0-35 months and mothers were measured usingappropriate equipment. Blood samples werealso taken from every third/fourth child age 6-23 months and their non-pregnant mothers (theresults of which are still being analysed and willbe presented in future). All data were collectedusing ODK-based platforms for data collectionand uploaded to a central server. TAB and pro-grammes with necessary training were providedby CARE.

Identical sampling and survey procedureswere followed for the endline survey, administeredfrom March to April 2018. For the endline,1,809 mothers and their children aged 0-35months were surveyed.

Service/activity Total Participants/Beneficiaries

Stakeholder trainingMIYCN training (union to district level) 34 sessions 671 (target 671)

GMP training – 50 (target 50)

SSMM training – 39 (target 41)

Community Support Group trainingTraining for community support group members onleadership, management, and nutrition

35 sessions 672 (target 630)

Community and school-based activitiesComplementary feeding and cooking demonstrations 1,171 11,091

Mother & child gatherings at community clinics 1,407 18,336

Adolescent girls receiving IFA from school-based programme – 5,218

School-based health education sessions 451 6,485

Community handwashing demonstrations 119 2,452

Services for PLW 2015 (#) 2016 (#) 2017 (#)Counselling sessions for pregnant women 43,978 45,686 54,436

Counselling sessions for lactating women 101,970 110,271 116,200

ANC services 21,101 33,232 46,978

Pregnant women provided with IFA 30,850 27,654* 34,472

Lactating women provided with IFA 13,488 9,440* 13,479

Table 1 Coverage of N@C activities and numbers of beneficiaries

* Decline in IFA distribution from 2015 to 2016 was due to low supply at service centres. This was corrected the following the yearand IFA distribution to PLW increased again in 2017.

57

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Field Article

ResultsPreliminary results comparing baseline and endlinedata indicate some improvements in terms ofmaternal and child nutrition, IYCF practices, WASHand women’s empowerment in the N@C interventionarea. Further analysis of changes in the interventionversus the control area is required (forthcoming) todetermine with more certainty the extent to whichthese changes are attributable to N@C interventions.

Maternal health and nutritionAs shown in Figure 2, 59.1% of women in N@Careas met the standards for minimum dietary di-versity1 in the endline survey, an increase from23.6% at baseline. In N@C project areas, 90.5% ofwomen had at least one ANC visit, compared toonly 37.9% at baseline. e percentage of womenwho had at least four ANC visits also increasedfrom 22.5% at baseline to 61.4% at endline. Ironfolic acid (IFA) intake during last pregnancy rosefrom 48.3% to 86.1% in the N@C project areas. IFAintake for at least 90 days (the recommended mini-mum) also increased, from 38.1% to 67.1%. epercentage of women who received postnatal careincreased substantially, from 14.7% at baseline to49.7%.

e majority of infants (86.1%) in the N@Cproject areas were put to the breast within an hourof birth (74.8% baseline) (see Figure 4). e prevalenceof exclusive breastfeeding rose from 48.7% to 69.0%in the N@C intervention area. Timely complementaryfeeding rose from 82.1% to 90.3%. Minimum dietarydiversity and minimum acceptable diet increasedsubstantially, from 32.6% to 70.6%, and from 28.4%to 69.0% respectively.

WASHFigure 5 presents baseline and endline survey resultsfor select WASH indicators. In the interventionarea, most households had an improved water source(i.e., tubewell), which increased from 74.6% atbaseline to 98.6% at endline. e proportion ofhouseholds with an improved toilet facility also in-creased from 42.0% to 72.0%. In terms of hygienepractices, the percentage of mothers who reportedthat they washed their hands aer using the toiletincreased from 56.1% to 88.7%. However, not allhygiene practices were well adhered to: only aboutone third (33.8%) of mothers responded that theywashed their hands with soap before feeding theirchild; an increase from only 9% at baseline.

Women’s EmpowermentTo assess women’s empowerment, the survey collectedinformation from women who answered that theyare the primary decision-maker for the given topic.(see Figure 6). In general, the percentage of womenwho are the primary decision-makers generally in-creased. Involvement in economic decision-makingincreased from 54.3% to 87.1%. Most women arethe primary decision-maker in terms of decidinghow food is distributed within the household intimes of scarcity (85.3%). ere was a small increase,from a low baseline, in women’s decision making

1 Women’s Dietary Diversity is a proxy for food security; minimum dietary diversity is defined as having eaten food from five or more food groups (out of nine food groups) in the 24 hours preceding the survey.

Figure 2 Baseline and endline survey results: Maternal health and nutrition

37.9%Received PNC during last pregnancy

4+ ANC visits

Intake of IFA during last pregnancy

Intake of IFA for 90+ days during last pregnancy

Received ANC during last pregnancy

Minimum Dietary Diversity (5+ food groups)

22.5%61.4%

48.3%

90.5%

86.1%

67.1%38.1%

49.7%14.7%

23.6%59.1%

Figure 3 Baseline and endline survey results: IYCF practice

74.8%Timely Initiation of breastfeeding

Exclusive breastfeeding

Timely complementary feeding

Introduction of solid/semi-solid or soft food

Minimum dietary diversity

Minimum acceptable diet

48.7%69.0%

82.1%

86.1%

90.3%

89.3%82.5%

70.6%32.6%

28.4%69.0%

Baseline Endline

Baseline Endline

A CSG member educates pregnant and lactating women onthe importance of taking iron and folic acid, Soromangal

village of Derai sub-district, Bangladesh, 2017

CARE

Ban

glad

esh

ANC = antenatal clinic IFA = iron and folic acid PNC = postnatal care

Figure 4 Baseline and endline survey results: WASH

74.6%Main source of drinking water: Tubewell / Deep Tubewell

Toilet type: Pit latrine with slab/water

Handwashing with soap after toilet use

Handwashing with soap before feeding the child

42.%72.0%

56.1%88.7%

98.6%

Baseline Endline

9.1%33.8%

58

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Field Article

around her own health care (13.2%) and thatof her child (19.9%). Various platforms, suchas CSG meetings and mother/child gatheringsat community clinics, were mobilised to im-plement activities around women’s empow-erment, with a focus on increasing women’srole in household decision-making as well aspromoting men’s involvement in support ma-ternal and child nutrition. is particularresult pertaining to self-determination aroundhealth care indicates that further understandingneeded around how to best empower womento make decisions around their own healthand that of their children.

Impact: Children’s nutritionalstatusResults of the endline survey indicate a re-duction in stunting of just over 13 percentagepoints, although the prevalence of stuntingamong children in the N@C project area re-mains high at 33.4%. Underweight prevalenceamong children aged 0-35 months decreased

from 35.6% to 21.1%, while wasting also de-creased from 10% to 7.5%. Results for theanaemia analysis are forthcoming.

Lessons learnedLearnings from phase one suggest that devel-oping competency in mentoring is an ongoingprocess. • Specific TOR and guidelines are essential to clarify the structures and roles of multi-sector coordination committees• Identification and implementation of feasible and evidence-based actions encourage multi-sector committee members to add ambitious actions over time.• Regular progress-sharing and recognition are important to sustain the motivation of multi-sector committee members and hold them accountable for achieving nutrition goals.• Keep national-level policymakers informed and engaged by facilitating

field visits to observe the processes and achievements of sub-district multi-sector coordination committees.• Improvements in nutrition outcomes can be achieved through strengthening union and village-level platforms along with district and sub-district-level multi-sector platforms.• Special attention and specific efforts are required to maintain the quality of activities as the multi-sector nutrition coordination platforms are scaled up throughout the country in coming years.

In Phase 2 CARE envisions building a sustainablesystem of support that allows supervisors to improvetheir technical skills and expertise on differentMIYCN issues and to provide more mentoring toFLWs. Phase 2 is currently underway as the CollectiveImpact for Nutrition (CI4N) initiative. roughCI4N, CARE is collaborating with members of SUNto support the Bangladesh National Nutrition Council(BNCC) in operationalising multi-sector nutritionapproaches in all 64 districts of Bangladesh.

Dissemination and learning fornational scale upe results of N@C in Bangladesh have been sharedthrough a national-level dissemination workshop.CARE also facilitated multiple field visits for GoBhigh officials, donors and civil society organisations,including members of the Civil Society Alliance forScaling Up Nutrition (CSA for SUN), to allow theseimportant stakeholders to observe the multi-sectorapproaches on the ground in the Sunamgonj district.As a result, policy makers have begun to prioritiseand incorporate multi-sector nutrition committeesin the country’s second National Plan of Action forNutrition (NPAN-2).

Conclusione Government of Bangladesh has institutionalisedmulti-sector coordinating committees at differentlevels, incorporating these platforms in the NPAN-2. However, the benefits of multi-sector nutritioncommittees can be only achieved through effectiveand sustainable implementation. To this end, CAREBangladesh offers operational guidelines that arebuilt on our experience in effectively facilitatingthese sub-district platforms. Furthermore, at justUS$1000 per year, facilitating the sub-district coor-dinating committees is financially sustainable. Inorder to make these multi-sector committees fullyfunctional and effective, initial external facilitationsupports add critical value that should be consideredby government and development partners in nutritionprogramming in Bangladesh.

For more information, please contact JahangirHossain at [email protected]

ReferencesWHO (2014). Global nutrition targets 2025: policy brief series(WHO/NMH/NHD/14.2). Geneva: World Health Organization.

Unicef (2013). Improving child nutrition: the achievableimperative for global progress. New York: UNICEF.

National Institute of Population Research and Training(NIPORT), Mitra and Associates and ICF International. 2016.Bangladesh Demographic and Health Survey 2014. Dhaka,Bangladesh and Rockville, Maryland, USA: NIPORT, Mitra andAssociates and ICF International.

Figure 5 Baseline and endline survey results: Women’s Empowerment

70.9%How food is shared when there is not enough food (self)

Self-decision about own healthcare:

Self-decision about child’s healthcare:

Self-decision about spending own money:

Self-decision about spending husband’s money:

9.6%13.2%

12.4%19.9%

85.3%

Baseline Endline

54.3%87.1%

2.2%31.0%

Members of the development, health, and nutrition communities attend a sharing event whereCARE presented on the N@C multi-sectoral approach to nutrition, Dhaka, Bangaldesh, 2018

CARE

Ban

glad

esh

Figure 6 Baseline and endline survey results: Children’s nutritional status

47.2%Stunting (0 - 35 mos)

Underweight (0 - 35 mos)

Wasting (0 - 35 mos)

35.6%21.1%

10.0%

33.4%

7.5%

Baseline Endline

59

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Participants of theGlobal NutritionCluster annual meeting in Amman,Jordan, October 2018

Round table discussion atthe Global Nutrition Clusterannual meeting in Amman,

Jordan, October 2018

Participants of the Global Nutrition Cluster annual meeting in Amman, Jordan, October 2018

People in aid Share pictures of you reading wherever you are in the world - we will tweet them anda selection will be published in the print edition. Send to [email protected]

60


Meeting on 15th November 2018 in Niger for the development of Niger's firstnational nutrition planhttps://www.ennonline.net/mediahub/blog/nigersmultisectoralnutritionplan

Martha Mwangome (KEMRI/Wellcome TrustResearch Programme, Kenya) at the InternationalSymposium on Understanding the Double Burdenof Malnutrition for Effective Interventions, IAEA,Vienna, 10th December 2018

IYCF-E training for the Nutrition Technical Working Group in Gaza, OccupiedPaelesitinan Territory (oPt), supported by Save the Children, UNICEF, WFP and theNutrition Technical Working Group partners

Mar

y Co

rbet

t


Editorial teamJeremy Shoham Marie McGrath Chloe Angood

Office SupportClara RamsayJudith FitzgeraldMary MurrayPeter Tevret

Carmel Dolan, Emily Mates and Tanya Khara are Technical Directors.

Jeremy Shohamand Marie McGrathare Field ExchangeCo-Editors andTechnical Directors.

Orna O’ Reilly designsand produces all ofENN’s publications.

Azaria Morgan isSUN ProjectAssistant, basedin London.

Chloe Angood isField Exchangesub-editor.

Charulatha Banerjee isENNs RegionalKnowledgeManagement Specialistfor Asia, based in India.

Clara Ramsay is theENN’s Finance Assistant,based in Oxford.

Lillian Karanja-Odhi-ambo is ENN’s Re-gional KnowledgeManagement Spe-cialist for East and

Southern Africa, based in Kenya.

Ambarka Youssoufaneis ENNs RegionalKnowledgeManagementSpecialist for West

Africa, based in Senegal.

Peter Tevret isENN’s Senior Finance Manager, basedin Oxford.

About ENNENN is a UK registered charity, international in reach, focused on supporting populations at high risk ofmalnutrition. ENN aims to enhance the effectiveness of nutrition policy and programming by improvingknowledge, stimulating learning, building evidence, and providing support and encouragement topractitioners and decision-makers involved in nutrition and related interventions.

ENN is both a core team of experienced and academically able nutritionists and a wider network of nutritionpractitioners, academics and decision-makers who share their knowledge and experience and use ENN’sproducts to inform policies, guidance and programmes in the contexts where they work.

ENN implements activities according to three major workstreams:Workstream 1: Experience sharing, knowledge management and learning. This includes ENN’s coreproducts: , Nutrition Exchange and en-net, as well as embedded knowledge management within two keyglobal nutrition fora (the Scaling Up Nutrition Movement (SUN) and the Global Nutrition Cluster (GNC)).

Workstream 2: Information and evidence on under-researched nutrition issues. This comprises ENN’sresearch and review work on filling gaps in the evidence base for improved nutrition policy andprogramming.

Workstream 3: Discussion, cooperation and agreement. This includes a range of activities for discussingand building agreement and consensus on key nutrition issues. It includes ENN’s participation in andhosting of meetings, its activities as facilitator of the IFE Core Group and its participation in thedevelopment of training materials and guidance, including normative guidance.

ENNs activities are governed by a five year strategy (2016-2020), visit www.ennonline.net

Contributors for this issue:

Front coverWomen participating in the nationalprogramme for nutrition support toeducation (MDM) scheme, Odisha,Gajapati, India, 2014;WFP/Aditya Arya

Judith Fitzgerald,is the ENN OfficeManager basedin Oxford.

Mary Murray is Administrative Assistant at ENN,based in Oxford.

Natalie Sessions is ENN’sGlobal KnowledgeManagement Coordinator (SUNMovement).

The Team

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Aakanksha PandeyAnne-Marie MayerAnnie Zhou Arjan De Wagt Bulti Assaye Casie TesfaiDlorah JenkinsElburg van BoetzelaerErnest GueverraEsther MogusuGary MundyHou Kroeun Jahangir HossainKassim Lupao Khrist Roy Lucia PantellaMark Myatt Maureen L GallagherMd HasanuzzamanMeeta MathurMohammad HafijulIslamMoussa IssaLende Naoko KozukiNaveen JainNazneen RahmanPraveen Kumar SaïdouTambouraSanjay Kumar DasSarah GibsonShiromi Michelle PereraShivangi Kaushik Sule MelehUmar ChiromaVani Sethi

supported by:

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Thanks to all who contributed or helpedsource pictures for this issue.

DesignOrna O’Reilly/Big Cheese Design.com

WebsiteOxford Web Applications

The Emergency Nutrition Network (ENN) is a registered charity in the UK (charity registration no: 1115156)and a company limited by guarantee and not having a share capital in the UK (company registration no:4889844). Registered address: 32, Leopold Street, Oxford, OX4 1TW, UK. ENN Directors/ Trustees: MarieMcGrath, Jeremy Shoham, Bruce Laurence, Nigel Milway, Victoria Lack and Anna Taylor.

@theennonline

@ENNonline

ENN Online

ENN Online

ENN (EmergencyNutrition Network)

https://www.facebook.com/theennonline/

https://twitter.com/ennonline?lang=en

https://www.youtube.com/playlist?list=UU6fS6gY8WCWL8HRdT-L9hKg

https://soundcloud.com/enn-online

https://es.linkedin.com/company/emergency-nutrition-network

https://es.linkedin.com/company/emergency-nutrition-network

ENN32, Leopold Street, Oxford, OX4 1TW, UK

Tel: +44 (0)1 865 324996Fax: +44 (0)1 865 597669Email: [email protected]

www.ennonline.net

january 2019 issue 59 issn 1743-5080 (print) · south asia exceeds all other regions, yet the...

Documents